Content uploaded by Stephanie Serafim Carvalho
Author content
All content in this area was uploaded by Stephanie Serafim Carvalho on Apr 09, 2024
Content may be subject to copyright.
Mitochondrial
Medicine
Volkmar Weissig
Marvin Edeas Editors
Volume 2: Assessing Mitochondria
Second Edition
Methods in
Molecular Biology 2276
For over 35 years, biological scientists have come to rely on the research protocols and
methodologies in the critically acclaimed Methods in Molecular Biology series. The series was
the first to introduce the step-by-step protocols approach that has become the standard in all
biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by-
step fashion, opening with an introductory overview, a list of the materials and reagents
needed to complete the experiment, and followed by a detailed procedure that is supported
with a helpful notes section offering tips and tricks of the trade as well as troubleshooting
advice. These hallmark features were introduced by series editor Dr. John Walker and
constitute the key ingredient in each and every volume of the Methods in Molecular Biology
series. Tested and trusted, comprehensive and reliable, all protocols from the series are
indexed in PubMed.
Mitochondrial Medicine
Volume 2: Assessing Mitochondria
Second Edition
Edited by
Volkmar Weissig
Department of Pharmaceutical Sciences, Midwestern University, Glendale, AZ, USA
Marvin Edeas
Cochin Hospital, Cochin Institute, INSERM U1016, PARIS, France
Editors
Volkmar Weissig
Department of Pharmaceutical Sciences
Midwestern University
Glendale, AZ, USA
Marvin Edeas
Cochin Hospital
Cochin Institute, INSERM U1016
PARIS, France
ISSN 1064-3745 ISSN 1940-6029 (electronic)
Methods in Molecular Biology
ISBN 978-1-0716-1265-1 ISBN 978-1-0716-1266-8 (eBook)
https://doi.org/10.1007/978-1-0716-1266-8
©Springer Science+Business Media, LLC, part of Springer Nature 2021, Corrected Publication 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter
developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.
Preface
It is our distinct pleasure to present the second edition of MiMB Mitochondrial Medicine to
the ever increasing number of scientists and physicians who are as fascinated by this tiny
organelle as we are. We started working on the first edition in September 2014 and were able
to bring about one year later two volumes with a total of 70 chapters to the market. As of
today (July 2020), 195K downloads have been recorded for Volume I
1
and 90K downloads
for volume II
2
. In light of the rapidly growing and expanding field of Mitochondrial
Medicine, we readily accepted the invitation to compile a second edition, which we started
to work on in March 2019. This second edition as offered here involves a total of 88 chapters
with 45 of them written by new contributors who were not part of our first edition. The first
and second editions combined subsequently present work from 115 mitochondrial labora-
tories from around the globe. We therefore believe these five volumes combined to be the
most comprehensive source of know-how in the wide-ranging field of Mitochondrial
Medicine.
Dividing 87 chapters equally over three volumes proved to be a bit challenging. We
chose the subtitle Targeting Mitochondria for volume I, Assessing Mitochondria for volume
II, and Manipulating Mitochondria and Disease Specific Approaches for volume III while of
course being well aware of significant overlaps between these three areas of research. For
example, it is quite obvious that mitochondria are being targeted for the purpose of either
assessing them or to manipulate them. We therefore ask all authors not to be too critical
regarding the placement of their particular chapter. The reader we believe will anyway
choose to download a chapter of his/her interest quite independently of its placement in
one of the three volumes.
All chapters in these three volumes were written for graduate students, postdoctoral
associates, independent investigators in academia and industry as well as physicians by
leading experts in their particular field. We are extremely grateful to them for having
found the time to either update their chapter from the first edition or to write a new chapter.
We will not forget that for many if not all of our contributors the worldwide COVID-19
pandemic posed additional and unexpected hurdles towards finishing their manuscript in
due time. Thank you to all!
The idea for our original book proposal leading to the first edition of MiMB Mitochon-
drial Medicine originated in our efforts to organize a series of annual conferences on
Targeting Mitochondria (www.targeting-mitochondria.com), the tenth one of which mean-
while has taken place in November 2019 in Berlin, Germany. Due to the ongoing pandemic,
our 11th conference (October 29–30, 2020) will be a virtual one but we are sure it will not
be less exciting than all the previous editions.
Last but not least we would like to sincerely thank John Walker, the series editor of
Methods in Molecular Biology, for having invited us to compile this second edition and for his
1
https://link.springer.com/book/10.1007/978-1-4939-2257-4.
2
https://link.springer.com/book/10.1007/978-1-4939-2288-8.
v
unlimited guidance and help throughout the entire process. We also owe sincere thanks to
Patrick Marton, the Executive Editor of the Springer Protocol Series, for always having been
available in assisting us throughout the entire project.
Glendale, AZ, USA Volkmar Weissig
Paris, France Marvin Edeas
vi Preface
Contents
Preface . . ................................................................... v
Contributors................................................................. xi
1 Mitochondrial Dysfunction in Mitochondrial Medicine: Current
Limitations, Pitfalls, and Tomorrow . . ..................................... 1
Naig Gueguen, Guy Lenaers, Pascal Reynier,
Volkmar Weissig, and Marvin Edeas
2 Preparation of “Functional” Mitochondria: A Challenging Business . . . ........ 31
Stefan Lehr, Sonja Hartwig, and Jorg Kotzka
3 Isolation and Quality Control of Functional Mitochondria ................... 41
Sonja Hartwig, Jorg Kotzka, and Stefan Lehr
4 Purification of Functional Platelet Mitochondria
Using a Discontinuous Percoll Gradient ................................... 57
Jacob L. Le´ger, Nicolas Pichaud, and Luc H. Boudreau
5 Mechanical Permeabilization as a New Method for Assessment
of Mitochondrial Function in Insect Tissues . . .............................. 67
Alessandro Gaviraghi, Yan Aveiro, Stephanie S. Carvalho,
Rodiesley S. Rosa, Matheus P. Oliveira, and Marcus F. Oliveira
6 Analysis of Mitochondrial Retrograde Signaling in Yeast Model Systems ....... 87
Nicoletta Guaragnella, Mas
ˇaZ
ˇdralevic
´, Zdena Palkova
´,
and Sergio Giannattasio
7 Native Gel Electrophoresis and Immunoblotting to Analyze
Electron Transport Chain Complexes ..................................... 103
Gisela Beutner and George A. Porter Jr.
8 Measuring Mitochondrial Hydrogen Peroxide Levels and
Glutathione Redox Equilibrium in Drosophila Neuron Subtypes
Using Redox-Sensitive Fluorophores and 3D Imaging....................... 113
Lori M. Buhlman, Petros P. Keoseyan, Katherine L. Houlihan,
and Amber N. Juba
9 Assessment of Mitochondrial Cell Metabolism by Respiratory
Chain Electron Flow Assays . ............................................. 129
Flavia Radogna, De´borah Ge´rard, Mario Dicato, and Marc Diederich
10 Whole-Cell and Mitochondrial dNTP Pool Quantification
from Cells and Tissues................................................... 143
Juan C. Landoni, Liya Wang, and Anu Suomalainen
11 Single-Particle Tracking Method in Fluorescence Microscopy
to Monitor Bioenergetic Responses of Individual Mitochondria .............. 153
Camille Colin, Emmanuel Suraniti, Emma Abell, Audrey Se´mont,
Neso Sojic, Philippe Diolez, and Ste´phane Arbault
vii
12 Investigation of Mitochondrial ADP-Ribosylation
Via Immunofluorescence . . . ............................................. 165
Ann-Katrin Hopp and Michael O. Hottiger
13 Assessment of Mitochondrial Ca
2+
Uptake . . . .............................. 173
Andra
´s T. Deak, Claire Jean-Quartier, Alexander I. Bondarenko,
Lukas N. Groschner, Roland Malli, Wolfgang F. Graier,
and Markus Waldeck-Weiermair
14 Assessment of Mitochondrial Membrane Potential and NADH
Redox State in Acute Brain Slices ......................................... 193
Andrey Y. Vinokurov, Viktor V. Dremin, Gennadii A. Piavchenko,
Olga A. Stelmashchuk, Plamena R. Angelova, and Andrey Y. Abramov
15 Evaluation of Mitochondria Content and Function in Live
Cells by Multicolor Flow Cytometric Analysis .............................. 203
Hsiu-Han Fan, Tsung-Lin Tsai, Ivan L. Dzhagalov,
and Chia-Lin Hsu
16 Analysis of Mitochondrial Dysfunction During Cell Death ................... 215
Vladimir Gogvadze and Boris Zhivotovsky
17 Modified Blue Native Gel Approach for Analysis of Respiratory
Supercomplexes ........................................................ 227
Sergiy M. Nadtochiy, Megan Ngai, and Paul S. Brookes
18 Patch-Clamp Recording of the Activity of Ion Channels
in the Inner Mitochondrial Membrane .................................... 235
Piotr Bednarczyk, Rafał P. Kampa, Shur Gałecka,
Aleksandra Se˛k, Agnieszka Walewska, and Piotr Koprowski
19 Assessment of Mitochondrial Protein Glutathionylation
as Signaling for CO Pathway ............................................. 249
Ana S. Almeida, Cla
´udia Figueiredo-Pereira,
and Helena L. A. Vieira
20 3D Optical Cryo-Imaging Method: A Novel Approach
to Quantify Renal Mitochondrial Bioenergetics Dysfunction . . ............... 259
Shima Mehrvar, Amadou K. S. Camara, and Mahsa Ranji
21 Simultaneous Quantification of Mitochondrial ATP and ROS
Production Using ATP Energy Clamp Methodology . ....................... 271
Liping Yu, Brian D. Fink, and William I. Sivitz
22 High-Throughput Image Analysis of Lipid-Droplet-Bound
Mitochondria .......................................................... 285
Nathanael Miller, Dane Wolf, Nour Alsabeeh, Kiana Mahdaviani,
Mayuko Segawa, Marc Liesa, and Orian S. Shirihai
23 Cell Energy Budget Platform for Multiparametric
Assessment of Cell and Tissue Metabolism . . . .............................. 305
Dmitri B. Papkovsky and Alexander V. Zhdanov
24 Fluorescence-Based Assay For Measuring OMA1 Activity .................... 325
Julia Tobacyk and Lee Ann MacMillan-Crow
25 Studying Mitochondrial Network Formation by In Vivo
and In Vitro Reconstitution Assay . . . ..................................... 333
Wanqing Du, Xiangjun Di, and Qian Peter Su
viii Contents
26 Extraction of Functional Mitochondria Based on Membrane
Stiffness . . . ............................................................ 343
Md Habibur Rahman, Qinru Xiao, Shirui Zhao,
An-Chi Wei, and Yi-Ping Ho
27 A Protocol for Untargeted Metabolomic Analysis: From
Sample Preparation to Data Processing .................................... 357
Amanda L. Souza and Gary J. Patti
28 A Method for Analysis of Nitrotyrosine-Containing Proteins
by Immunoblotting Coupled with Mass Spectrometry ...................... 383
Matej Kohutiar and Adam Eckhardt
29 In Vivo Visualization and Quantification of Mitochondrial
Morphology in C. elegans ................................................ 397
R. de Boer, R. L. Smith, W. H. De Vos, E. M. M. Manders,
and H. van der Spek
30 Assessing Impact of Platinum Complexes on Mitochondrial
Functions . . ............................................................ 409
Suxing Jin and Xiaoyong Wang
31 In Silico Modeling of the Mitochondrial Pumping Complexes
with Markov State Models . . ............................................. 425
Roger Springett
32 Monitoring the Mitochondrial Presequence Import Pathway
In Living Mammalian Cells with a New Molecular Biosensor. . ............... 441
Maxime Jacoupy, Emeline Hamon-Keromen, and Olga Corti
Correction to: Measuring Mitochondrial Hydrogen Peroxide Levels
and Glutathione Redox Equilibrium in Drosophila Neuron Subtypes
Using Redox-Sensitive Fluorophores and 3D Imaging ........................... C1
Index . . . ................................................................... 453
Contents ix
Contributors
EMMA ABELL •Univ. Bordeaux, INSERM, Centre de recherche Cardio- Thoracique de
Bordeaux, U1045 & IHU Liryc, Electrophysiology and Heart Modeling Institute,
Fondation Bordeaux Universite´, Bordeaux, France; CHU de Bordeaux, Bordeaux, France
ANDREY Y. ABRAMOV •Orel State University, Orel, Russia; UCL Queen Square Institute of
Neurology, London, UK
ANA S. ALMEIDA •CEDOC, Faculdade de Cie
ˆncia Me´dicas/NOVA Medical School,
Universidade Nova de Lisboa, Lisboa, Portugal
NOUR ALSABEEH •Department of Physiology, Faculty of Medicine, Kuwait University,
Kuwait City, Kuwait
PLAMENA R. ANGELOVA •UCL Queen Square Institute of Neurology, London, UK
STE
´PHANE ARBAULT •NSysA group, Univ. Bordeaux, CNRS, INP Bordeaux, ISM, UMR
5255, Talence, France
YAN AVEIRO •Federal University of Rio de Janeiro, Institute of Medical Biochemistry
Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
PIOTR BEDNARCZYK •Department of Physics and Biophysics, Institute of Biology, Warsaw
University of Life Sciences—SGGW, Warsaw, Poland
GISELA BEUTNER •Department of Pediatrics-Division Cardiology, University of Rochester,
Rochester, NY, USA
ALEXANDER I. BONDARENKO •Gottfried Schatz Research Center for Cell Signaling,
Metabolism and Aging, Molecular Biology and Biochemistry, Medical University of Graz,
Graz, Austria
LUC H. BOUDREAU •Department of Chemistry and Biochemistry, Universite de Moncton,
Moncton, NB, Canada
PAUL S. BROOKES •Department of Anesthesiology, University of Rochester Medical Center,
Rochester, NY, USA
LORI M. BUHLMAN •Arizona College of Graduate Studies, Midwestern University, Glendale,
AZ, USA
AMADOU K. S. CAMARA •Department of Anesthesiology and Anesthesia Research, Medical
College of Wisconsin, Wauwatosa, WI, USA
STEPHANIE S. CARVALHO •Federal University of Rio de Janeiro, Institute of Medical
Biochemistry Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
CAMILLE COLIN •NSysA group, Univ. Bordeaux, CNRS, INP Bordeaux, ISM, UMR 5255,
Talence, France; Univ. Bordeaux, INSERM, Centre de recherche Cardio- Thoracique de
Bordeaux, U1045 & IHU Liryc, Electrophysiology and Heart Modeling Institute,
Fondation Bordeaux Universite´, Bordeaux, France; CHU de Bordeaux, Bordeaux, France
OLGA CORTI •Sorbonne Universite´, Institut du Cerveau (ICM), Inserm U1127, CNRS
UMR 7225, Paris, France
R. DE BOER •Molecular Biology & Microbial Food Safety, Swammerdam Institute for Life
Sciences (SILS), Faculty of Science (FNWI), University of Amsterdam, Amsterdam, The
Netherlands
W. H. DEVOS •Laboratory of Cell Biology and Histology, Department of Veterinary Sciences,
Antwerp University, Antwerp, Belgium; Cell Systems and Imaging Research Group,
Department of Molecular Biotechnology, Ghent University, Ghent, Belgium
xi
ANDRA
´ST. DEAK •Gottfried Schatz Research Center for Cell Signaling, Metabolism and
Aging, Molecular Biology and Biochemistry, Medical University of Graz, Graz, Austria
XIANGJUN DI•Institute for Biomedical Materials & Devices (IBMD), Faculty of Science,
University of Technology Sydney, Sydney, NSW, Australia
MARIO DICATO •Laboratoire de Biologie Mole´culaire et Cellulaire du Cancer, Ho
ˆpital
Kirchberg, Luxembourg, Luxembourg
MARC DIEDERICH •College of Pharmacy, Seoul National University, Seoul, South Korea
PHILIPPE DIOLEZ •Univ. Bordeaux, INSERM, Centre de recherche Cardio- Thoracique de
Bordeaux, U1045 & IHU Liryc, Electrophysiology and Heart Modeling Institute,
Fondation Bordeaux Universite´, Bordeaux, France; CHU de Bordeaux, Bordeaux, France
VIKTOR V. DREMIN •Orel State University, Orel, Russia; Aston University, Birmingham,
UK
WANQING DU•State Key Laboratory of Membrane Biology, Tsinghua University-Peking
University Joint Center for Life Sciences, School of Life Sciences, Tsinghua University,
Beijing, China
IVAN L. DZHAGALOV •Institute of Microbiology and Immunology, National Yang-Ming
University, Taipei, Taiwan
ADAM ECKHARDT •Department of Translational Metabolism, Institute of Physiology,
Academy of Sciences of the Czech Republic, Prague, Czech Republic
MARVIN EDEAS •Universite´ de Paris, INSERM U1016, Institut Cochin, CNRS UMR8104,
Paris, France; Laboratory of Excellence GR-Ex, Paris, France
HSIU-HAN FAN •Institute of Microbiology and Immunology, National Yang-Ming
University, Taipei, Taiwan
CLA
´UDIA FIGUEIREDO-PEREIRA •CEDOC, Faculdade de Cie
ˆncia Me´dicas/NOVA Medical
School, Universidade Nova de Lisboa, Lisboa, Portugal
BRIAN D. FINK •Division of Endocrinology and Metabolism, Department of Internal
Medicine, The University of Iowa, Iowa City, IA, USA
SHUR GAŁECKA •Laboratory of Intracellular Ion Channels, Nencki Institute of Experimental
Biology, Warsaw, Poland
ALESSANDRO GAVIRAGHI •Federal University of Rio de Janeiro, Institute of Medical
Biochemistry Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
DE
´BORAH GE
´RARD •Laboratoire de Biologie Mole´culaire et Cellulaire du Cancer, Ho
ˆpital
Kirchberg, Luxembourg, Luxembourg
SERGIO GIANNATTASIO •Institute of Biomembranes, Bioenergetics and Molecular
Biotechnologies, CNR, Bari, Italy
VLADIMIR GOGVADZE •Division of Toxicology, Institute of Environmental Medicine,
Karolinska Institutet, Stockholm, Sweden; Faculty of Medicine, MV Lomonosov Moscow
State University, Moscow, Russia
WOLFGANG F. GRAIER •Gottfried Schatz Research Center for Cell Signaling, Metabolism and
Aging, Molecular Biology and Biochemistry, Medical University of Graz, Graz, Austria
LUKAS N. GROSCHNER •Gottfried Schatz Research Center for Cell Signaling, Metabolism
and Aging, Molecular Biology and Biochemistry, Medical University of Graz, Graz,
Austria
NICOLETTA GUARAGNELLA •Institute of Biomembranes, Bioenergetics and Molecular
Biotechnologies, CNR, Bari, Italy; Department of Biosciences, Biotechnology and
Biopharmaceutics, University of Bari “A. Moro”, Bari, Italy
xii Contributors
NAIG GUEGUEN •UMR CNRS 6015-INSERM U1083, MitoVasc Institute, University of
Angers, Angers, France; Department of Biochemistry and Genetics, University Hospital of
Angers, Angers, France
EMELINE HAMON-KEROMEN •Sorbonne Universite´, Institut du Cerveau (ICM), Inserm
U1127, CNRS UMR 7225, Paris, France
SONJA HARTWIG •Institute for Clinical Biochemistry and Pathobiochemistry, German
Diabetes Center at the Heinrich-Heine-University Duesseldorf, Leibniz Center for Diabetes
Research, Duesseldorf, Germany; German Center for Diabetes Research (DZD e.V.),
Neuherberg, Germany
YI-PING HO•Department of Biomedical Engineering, The Chinese University of Hong
Kong, Hong Kong SAR, China; Centre for Novel Biomaterials, The Chinese University of
Hong Kong, Hong Kong SAR, China; Shun Hing Institute of Advanced Engineering, The
Chinese University of Hong Kong, Hong Kong SAR, China; The Ministry of Education Key
Laboratory of Regeneration Medicine, The Chinese University of Hong Kong, Hong Kong
SAR, China
ANN-KATRIN HOPP •Department of Molecular Mechanisms of Disease (DMMD), University
of Zurich, Zurich, Switzerland
MICHAEL O. HOTTIGER •Department of Molecular Mechanisms of Disease (DMMD),
University of Zurich, Zurich, Switzerland
KATHERINE L. HOULIHAN •Arizona College of Graduate Studies, Midwestern University,
Glendale, AZ, USA
CHIA-LIN HSU •Institute of Microbiology and Immunology, National Yang-Ming
University, Taipei, Taiwan
MAXIME JACOUPY •Sorbonne Universite´, Institut du Cerveau (ICM), Inserm U1127, CNRS
UMR 7225, Paris, France
CLAIRE JEAN-QUARTIER •Gottfried Schatz Research Center for Cell Signaling, Metabolism
and Aging, Molecular Biology and Biochemistry, Medical University of Graz, Graz,
Austria
SUXING JIN •State Key Laboratory of Coordination Chemistry, School of Chemistry and
Chemical Engineering, Nanjing University, Nanjing, P. R. China
AMBER N. JUBA •Arizona College of Graduate Studies, Midwestern University, Glendale,
AZ, USA
RAFAŁ P. KAMPA •Department of Physics and Biophysics, Institute of Biology, Warsaw
University of Life Sciences—SGGW, Warsaw, Poland; Laboratory of Intracellular Ion
Channels, Nencki Institute of Experimental Biology, Warsaw, Poland
PETROS P. KEOSEYAN •Arizona College of Osteopathic Medicine, Midwestern University,
Glendale, AZ, USA
MATEJ KOHUTIAR •Department of Medical Chemistry and Clinical Biochemistry, 2nd
Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague,
Czech Republic
PIOTR KOPROWSKI •Laboratory of Intracellular Ion Channels, Nencki Institute of
Experimental Biology, Warsaw, Poland
JORG KOTZKA •Institute for Clinical Biochemistry and Pathobiochemistry, German Diabetes
Center at the Heinrich-Heine-University Duesseldorf, Leibniz Center for Diabetes
Research, Duesseldorf, Germany; German Center for Diabetes Research (DZD e.V.),
Neuherberg, Germany
JUAN C. LANDONI •Research Programs Unit, Stem Cells and Metabolism, University of
Helsinki, Helsinki, Finland
Contributors xiii
JACOB L. LE
´GER •Department of Chemistry and Biochemistry, Universite de Moncton,
Moncton, NB, Canada
STEFAN LEHR •Institute for Clinical Biochemistry and Pathobiochemistry, German Diabetes
Center at the Heinrich-Heine-University Duesseldorf, Leibniz Center for Diabetes
Research, Duesseldorf, Germany; German Center for Diabetes Research (DZD e.V.),
Neuherberg, Germany
GUY LENAERS •UMR CNRS 6015-INSERM U1083, MitoVasc Institute, University of
Angers, Angers, France
MARC LIESA •Division of Endocrinology, Department of Medicine, David Geffen School of
Medicine at UCLA, Los Angeles, CA, USA
LEE ANN MACMILLAN-CROW •Department of Pharmacology and Toxicology, University of
Arkansas for Medical Sciences, Little Rock, AR, USA
KIANA MAHDAVIANI •Department of Medicine, Obesity Research Center, Boston University
School of Medicine, Boston, MA, USA
ROLAND MALLI •Gottfried Schatz Research Center for Cell Signaling, Metabolism and
Aging, Molecular Biology and Biochemistry, Medical University of Graz, Graz, Austria
E. M. M. MANDERS •van Leeuwenhoek Center for Advanced Microscopy, University of
Amsterdam, Amsterdam, The Netherlands
SHIMA MEHRVAR •University of Wisconsin-Milwaukee, Milwaukee, WI, USA
NATHANAEL MILLER •Division of Endocrinology, Department of Medicine, David Geffen
School of Medicine at UCLA, Los Angeles, CA, USA; Department of Medicine, Obesity
Research Center, Boston University School of Medicine, Boston, MA, USA
SERGIY M. NADTOCHIY •Department of Neuroscience, University of Rochester Medical
Center, Rochester, NY, USA; Department of Anesthesiology, University of Rochester
Medical Center, Rochester, NY, USA
MEGAN NGAI •Department of Anesthesiology, University of Rochester Medical Center,
Rochester, NY, USA
MARCUS F. OLIVEIRA •Federal University of Rio de Janeiro, Institute of Medical
Biochemistry Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
MATHEUS P. OLIVEIRA •Federal University of Rio de Janeiro, Institute of Medical
Biochemistry Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
ZDENA PALKOVA
´•Faculty of Science, Charles University, BIOCEV, Prague, Czech Republic
DMITRI B. PAPKOVSKY •School of Biochemistry and Cell Biology, University College Cork,
Cork, Ireland
GARY J. PATTI •Department of Chemistry, Washington University in St. Louis, Saint Louis,
MO, USA; Department of Medicine, Washington University in St. Louis, Saint Louis, MO,
USA
GENNADII A. PIAVCHENKO •Orel State University, Orel, Russia; I.M. Sechenov First Moscow
State Medical University (Sechenov University), Moscow, Russia
NICOLAS PICHAUD •Department of Chemistry and Biochemistry, Universite de Moncton,
Moncton, NB, Canada
GEORGE A. PORTER JR.•Departments of Pediatrics-Division Cardiology, Pharmacology
and Physiology, and Medicine (Aab Cardiovascular Research Institute), University of
Rochester, Rochester, NY, USA
FLAVIA RADOGNA •Laboratoire de Biologie Mole´culaire et Cellulaire du Cancer, Ho
ˆpital
Kirchberg, Luxembourg, Luxembourg
xiv Contributors
MDHABIBUR RAHMAN •Department of Biomedical Engineering, The Chinese University of
Hong Kong, Hong Kong SAR, China; Centre for Novel Biomaterials, The Chinese
University of Hong Kong, Hong Kong SAR, China
MAHSA RANJI •Biophotonics Lab, Florida Atlantic University, Boca Raton, FL, USA
PASCAL REYNIER •UMR CNRS 6015-INSERM U1083, MitoVasc Institute, University of
Angers, Angers, France; Department of Biochemistry and Genetics, University Hospital of
Angers, Angers, France
RODIESLEY S. ROSA •Federal University of Rio de Janeiro, Institute of Medical Biochemistry
Leopoldo de Meis, Rio De Janeiro, RJ, Brazil
MAYUKO SEGAWA •Division of Endocrinology, Department of Medicine, David Geffen School
of Medicine at UCLA, Los Angeles, CA, USA
ALEKSANDRA SE˛K •Laboratory of Intracellular Ion Channels, Nencki Institute of
Experimental Biology, Warsaw, Poland; Faculty of Chemistry, University of Warsaw,
Warsaw, Poland
AUDREY SE
´MONT •Univ. Bordeaux, INSERM, Centre de recherche Cardio- Thoracique de
Bordeaux, U1045 & IHU Liryc, Electrophysiology and Heart Modeling Institute,
Fondation Bordeaux Universite´, Bordeaux, France; CHU de Bordeaux, Bordeaux, France
ORIAN S. SHIRIHAI •Division of Endocrinology, Department of Medicine, David Geffen
School of Medicine at UCLA, Los Angeles, CA, USA
WILLIAM I. SIVITZ •Division of Endocrinology and Metabolism, Department of Internal
Medicine, The University of Iowa, Iowa City, IA, USA
R. L. SMITH •Academisch Medisch Centrum Universiteit van, Amsterdam, The Netherlands
NESO SOJIC •NSysA group, Univ. Bordeaux, CNRS, INP Bordeaux, ISM, UMR 5255,
Talence, France
AMANDA L. SOUZA •Life Science Mass Spectrometry Division, Thermo Fisher Scientific, San
Jose, CA, USA
ROGER SPRINGETT •CellSpex, Kent, UK
OLGA A. STELMASHCHUK •Orel State University, Orel, Russia
QIAN PETER SU•Institute for Biomedical Materials & Devices (IBMD), Faculty of Science,
University of Technology Sydney, Sydney, NSW, Australia; School of Biomedical
Engineering, Faculty of Engineering and IT, University of Technology Sydney, Sydney,
NSW, Australia
ANU SUOMALAINEN •Department of Neurology, University Hospital, Helsinki, Finland;
Neuroscience Center, University of Helsinki, Helsinki, Finland
EMMANUEL SURANITI •NSysA group, Univ. Bordeaux, CNRS, INP Bordeaux, ISM, UMR
5255, Talence, France
JULIA TOBACYK •Department of Pharmacology and Toxicology, University of Arkansas for
Medical Sciences, Little Rock, AR, USA
TSUNG-LIN TSAI •Institute of Microbiology and Immunology, National Yang-Ming
University, Taipei, Taiwan
H. VAN DER SPEK •Molecular Biology & Microbial Food Safety, Swammerdam Institute for
Life Sciences (SILS), Faculty of Science (FNWI), University of Amsterdam, Amsterdam,
The Netherlands
HELENA L. A. VIEIRA •CEDOC, Faculdade de Cie
ˆncia Me´dicas/NOVA Medical School,
Universidade Nova de Lisboa, Lisboa, Portugal; UCIBIO, Faculdade de Cie
ˆncias
e Tecnologia, Universidade Nova de Lisboa, Lisboa, Portugal; Instituto de Biologia
Experimental e Tecnol
ogica (iBET), Oeiras, Portugal
ANDREY Y. VINOKUROV •Orel State University, Orel, Russia
Contributors xv
MARKUS WALDECK-WEIERMAIR •Gottfried Schatz Research Center for Cell Signaling,
Metabolism and Aging, Molecular Biology and Biochemistry, Medical University of Graz,
Graz, Austria
AGNIESZKA WALEWSKA •Laboratory of Intracellular Ion Channels, Nencki Institute of
Experimental Biology, Warsaw, Poland
LIYA WANG •Department of Anatomy, Physiology and Biochemistry, Swedish University of
Agricultural Sciences, Uppsala, Sweden
XIAOYONG WANG •State Key Laboratory of Pharmaceutical Biotechnology, School of Life
Sciences, Nanjing University, Nanjing, P. R. China
AN-CHI WEI •Graduate Institute of Biomedical Electronics and Bioinformatics, National
Taiwan University, Taipei, Taiwan
VOLKMAR WEISSIG •Department of Pharmaceutical Sciences and Nanocenter of Excellence,
Midwestern University College of Pharmacy at Glendale, Glendale, AZ, USA
DANE WOLF •Division of Endocrinology, Department of Medicine, David Geffen School of
Medicine at UCLA, Los Angeles, CA, USA; Department of Medicine, Obesity Research
Center, Boston University School of Medicine, Boston, MA, USA
QINRU XIAO •Department of Biomedical Engineering, The Chinese University of Hong
Kong, Hong Kong SAR, China
LIPING YU•Division of Endocrinology and Metabolism, Department of Internal Medicine,
The University of Iowa, Iowa City, IA, USA
MAS
ˇAZ
ˇDRALEVIC
´•Faculty of Medicine, University of Montenegro, Podgorica, Montenegro
SHIRUI ZHAO •Department of Biomedical Engineering, The Chinese University of Hong
Kong, Hong Kong SAR, China; Shun Hing Institute of Advanced Engineering, The
Chinese University of Hong Kong, Hong Kong SAR, China
ALEXANDER V. ZHDANOV •School of Biochemistry and Cell Biology, University College Cork,
Cork, Ireland
BORIS ZHIVOTOVSKY •Division of Toxicology, Institute of Environmental Medicine,
Karolinska Institutet, Stockholm, Sweden; Faculty of Medicine, MV Lomonosov Moscow
State University, Moscow, Russia
xvi Contributors
Chapter 1
Mitochondrial Dysfunction in Mitochondrial Medicine:
Current Limitations, Pitfalls, and Tomorrow
Naig Gueguen, Guy Lenaers, Pascal Reynier, Volkmar Weissig,
and Marvin Edeas
Abstract
Until recently restricted to hereditary mitochondrial diseases, mitochondrial dysfunction is now recognized
as a key player and strategic factor in the pathophysiological of many human diseases, ranging from the
metabolism, vascular, cardiac, and neurodegenerative diseases to cancer. Because of their participation in a
myriad of cellular functions and signaling pathways, precisely identifying the cause of mitochondrial
“dysfunctions” can be challenging and requires robust and controlled techniques. Initially limited to the
analysis of the respiratory chain functioning, these analytical techniques now enlarge to the analyses of
mitochondrial and cellular metabolism, based on metabolomic approaches.
Here, we address the methods used to assay mitochondrial dysfunction, with a highlight on the
techniques used in diagnosis on tissues and cells derived from patients, the information they provide, and
their strength and weakness.
Targeting mitochondrial dysfunction by various strategies is a huge challenge, requires robust methods of
evaluation, and should be able to take into consideration the mitochondria dynamics and localization. The
future of mitochondrial medicine is strongly related to a perfect comprehension of its dysfunction.
Key words Mitochondrial dysfunctions, Mitochondria evaluation, Bioenergetics, Devices,
Metabolomics
1 Introduction
The contribution of mitochondrial dysfunctions in the onset and
development of many diseases has been widely studied. These
dysfunctions could interfere with many cellular, metabolic, and
homoeostatic functions [1]. It is therefore not surprising that,
beyond primary mitochondrial disorders, mitochondrial dysfunc-
tions now impact most areas of medical. Mitochondrial dysfunction
has been found in numerous common human diseases, including
metabolic diseases [2] such as obesity and diabetes, heart diseases
[3], neurodegenerative diseases, such as Parkinson’s, Alzheimer’s
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_1,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
1
[4], or Huntington’s [5], infectious diseases [6], inflammation [7],
and cancer [8].
Multiple reports have shown the impact of dysfunctional mito-
chondria on the immune response. One of the example is the
COVID-19 pandemic caused by the coronavirus (SARS-CoV-2).
Many recent studies revealed that human alveolar epithelial cells
with dysfunctional mitochondria displayed increased production of
pro-inflammatory cytokines which were found to be increased in
COVID-19 [9,10]. Recently, we proposed that not only the
intracellular mitochondria dysfunction is a consequence of
COVID-19 infection formation [6,11] but also the less explored
extracellular mitochondria (specifically platelets mitochondria) may
affect blood coagulation, clot, and thrombosis formation [10–
15]. These extracellular mitochondria may represent a crucial inter-
cellular mediators and may serve as strategic therapeutic targets in
COVID-19 pathogenesis [11].
Historically, mitochondrial diseases have been largely addressed
from the point of view of the defect of the respiratory chain and
therefore of a defect in energetic homeostasis. However, mitochon-
dria are now recognized to perform multiple additional cellular
functions beyond energy production. For example, electron trans-
fer chain (ETC) activity regulates the NADH/NAD+ redox state
through NADH oxidation by complex I, [16] which impacts to
hundreds of cellular reactions [17,18]. Mitochondria are the major
source of reactive oxygen species (ROS) [19], which not only
contribute to normal cell function but also are linked to increased
intracellular oxidative stress. ROS production regulates different
signaling pathways, from metabolic rewiring to inflammatory
response or cell survival [20,21]. Mitochondria also integrate
main cellular catabolic and anabolic pathways. They host not only
whole or partial components of several converging catabolic cellular
processes, i.e., glucose metabolism and the tricarboxylic acid (TCA)
cycle, and the β-oxidation but also several biosynthetic pathways,
such as folate and sulfur metabolism or heme biosynthesis. In
addition to generating reducing equivalents that feed the
OXPHOS system, the TCA cycle has numerous anabolic roles,
providing precursors for the lipids, proteins, carbohydrates, and
nucleotides biosynthesis [22]. TCA cycle metabolites also act as
signaling molecules, affecting critical cellular processes that con-
tribute to oncogenic transformation, such as nutrient signaling,
HIF1α-dependent metabolic reprogramming, or histone acetyla-
tion and demethylation by acetyl-CoA and α-ketoglutarate, respec-
tively [23]. Mitochondria are also involved in calcium homoeostasis
[24], stress response and quality control [25], and initiation of
caspase-dependent apoptosis [26].
Developing precise technologies to study mitochondrial physi-
ology is getting much more important as the prevalence of com-
mon and inherited mitochondrial disease increases. Targeting
2 Naig Gueguen et al.
mitochondrial dysfunction by various strategies is a huge challenge
and need a robust methods of evaluation [27].
Here, we address the techniques commonly used in clinical
diagnosis, the information that can be inferred from them, their
limitations, pits, and pitfalls.
2 Methods to Assess Mitochondrial Dysfunctions
2.1 Basis
in Bioenergetic
Principles of mitochondrial bioenergetics have been largely
reviewed over years [28]; therefore, only a rapid overview will be
given here. Mitochondria are surrounded by the outer (OMM) and
inner (IMM) membrane. The inner membrane is largely imper-
meant and constitutes the major barrier between the cytosol and
the mitochondrial matrix. IMM forms multiple cristae, which host
the ETC complexes and the enzymes involved in ATP synthesis, the
F0F1-ATP synthase, the adenine nucleotide translocator (ANT),
and the inorganic phosphate (Pi) transporter (PiC).
The ETC system is composed of multisubunits enzymes func-
tionally and physically linked together, the complex I (CI, NADH
ubiquinone reductase), complex II (CII, succinate ubiquinone
reductase), complex III (CIII, ubiquinol cytochrome c reductase),
and complex IV (CIV, cytochrome c oxidase), which transfers the
electron energetic potential from NADH/NAD+ (CI) and
FADH2/FAD+ (CII) to the electrochemical proton gradient
known as the protonmotive force (Δp). This process involved a
series of oxidoreductase reactions in which electrons flow sequen-
tially “downhill” along the ETC from a reduced to an oxidized
state, ending to molecular oxygen reduction to a water molecule, or
“respiration.” The release of free energy during electron transfer
drives the proton pumping across the IMM at complexes CI, CIII,
and CIV, in turn producing the Δp. Other oxidoreductases of the
ETC are unable of proton pumping, particularly the complex II,
the glycerol phosphate dehydrogenase, and the electron-
transferring flavoprotein quinone oxidoreductase (ETF) linked to
fatty acid β-oxidation. Thus, ten protons are extruded for each
electron pair transferred from NADH to oxygen, or six protons
for each electron pair transferred from FADH2 to O
2
, meaning that
the H+/O2 and ATP/O
2
stoichiometries differ according to the
initial substrates.
The fueling of reducing molecules, namely NADH and
FADH2, to the ETC is mainly ensured by the TCA cycle and the
β-oxidation pathways. According to the anaplerotic routes which
feed TCA or the relative contribution of TCA or β-oxidation, the
resulting NADH/FADH2 ratio differs, ultimately modifying the
efficiency of ATP synthesis. The substrate oxidation module con-
sists of all these reactions involved in substrate metabolisms and
electron transport, finally generating Δp.
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 3
The Δp is composed of the charge (Δψ
m
) and chemical (ΔpH)
components. The energy available in Δp drives the synthesis and
transport of ATP, as the protons return to the matrix through the
ATP synthase [29]. The ATP synthesis module includes, in addition
to the ATP synthetase, the ANT that exchange the ADP/ATP and
the PiC. The Δp is central to the physiological functions of mito-
chondria, connecting the substrate oxidation module to the ATP
synthesis. Under most circumstances, proton flux is tightly coupled
to OXPHOS. However, the Δp can also be dissipated by proton
leak [30], in which protons return into the matrix independently of
ATP synthesis, or be used for transport processes across the IMM of
metabolite, ions, or calcium [31].
The total proton entry and extrusion are exactly balanced
under steady-state conditions. Forward flux through ETC com-
plexes requires a thermodynamic disequilibrium, i.e., the free
energy available from electron transfer must be greater than that
required to pump protons against the Δp. The Δp variations thus
“dictate” the ETC activity: any decrease in the Δp is followed by an
increase in electron transfer and proton pumping. In other words,
the net flux of the proton current is reflected by the rate of mito-
chondrial oxygen consumption. Decrease in the Δp results from
either proton leak or ATP synthesis, which is regulated by the
cellular ATP demand.
Any alteration in one the above described processes leads to a
mitochondrial dysfunction. Subsequently, we shall describe the
conventional methods for dissecting these processes, and identify-
ing the primary site of impairment, starting with the methods used
in the laboratory for the biochemical diagnosis of mitochondrial
diseases. A complete set of these analyses will draw a picture of the
mitochondrial energy-generating system.
2.2 Analyses
of Maximal ETC
Activities
Evaluation of individual respiratory chain complex activities is a
routine biochemical approach for the diagnosis of mitochondrial
disorders. Assays to quantify CI, CII, CIII, CIV, and ATP synthase
enzymatic activities are performed either on tissue, mostly muscle
or liver biopsies, or on cells as primary fibroblasts or lymphocytes.
They require the preparation of tissue homogenates, after the
elimination of cell debris and nuclei, or cell lysates, from either
fresh or frozen sample. However, frozen samples are more com-
monly used because fresh samples must be processed immediately
after sampling.
Isolated activity of each complex can be analyzed by following
the oxidation/reduction of specific substrates or substrate analogs
by spectrophotometry. The spectrophotometric enzyme assays are:
lNADH:ubiquinone oxidoreductase (NUR, EC 7.1.1.2, NADH
oxidation followed at 340 nm) for CI.
4 Naig Gueguen et al.
lSuccinate:ubiquinone oxidoreductase (coupled to
2,6-Dichloroindophenol reduction followed at 600 nm) for
CII (SUR, EC 1.3.5.1).
lSuccinate:cytochrome c oxidoreductase (CII + CIII, cyto-
chrome c reduction at 550 nm).
lUbiquinol cytochrome c oxidoreductase for CIII (UCCR, EC
1.10.2.2, cytochrome c reduction at 550 nm).
lCytochrome coxidase for CV (COX, EC 1.9.3.1, cytochrome c
oxidation at 550 nm).
The measurement of complex V (oligomycin-sensitive ATPase)
is more challenging and requires mitochondrial-enriched fractions
and is often measured in cultured skin fibroblasts. Citrate synthase
activity is often used for normalization of respiratory chain complex
activities to mitochondrial mass. The activities normalized to CS
and/or the ratios between mitochondrial enzymes give a much
narrower range of normal values compared to activities expressed
with respect to sample protein content, since the mitochondrial
mass are highly variable among individuals and subjected to mito-
chondrial biogenesis regulation.
While the principles of the different protocols are similar, now-
adays, there is no universal assay for spectrophotometric quantifica-
tion of ETC enzyme activities. According to laboratories, assays
differ with respect to buffer conditions, reaction temperatures,
substrate concentrations, supplementation in Ca
2+
chelator
(EDTA), or addition of bovine serum albumin. However, in an
attempt to facilitate interlaboratory comparison of the results, some
reference centers for the diagnosis of mitochondrial diseases have
reevaluated and standardized their protocols [32].
Accurate enzymatic testing of tissue and cell samples can be
hampered by different pitfalls. A frequent cause for problems is
sample collection and handling. For optimal preservation of mito-
chondrial enzymes, samples must be immediately snap-frozen after
collection and stored at 80 C, without any thawing until analysis.
Other artifacts come from local anesthetic contamination of the
sample, particularly lidocaine [32,33], or, for studies on animals,
from euthanasia procedures using CO
2
. Furthermore, one of the
main difficulties in achieving reliable and reproducible dosages is
the quality of the reagents, with strong variability between reactive
references or even batches of reactive. Thus, these dosages require
the inclusion of quality controls in each series, in order to validate
the reaction mix. This quality is usually obtained from culture cells
or animal tissue [34,35], since the human tissues are limited.
In addition, assays that determine the amounts of OXPHOS
complexes, such as blue native (BN) gel electrophoresis followed by
Western blot analysis, should be performed to decipher an ETC
defect related to decreased catalytic activity or involving a lower
enzyme quantity.
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 5
2.3 Structural
Analyses: BN-PAGE
ETC enzymes are all multimeric and encoded by both the mtDNA
and nDNA, except for complex II that is fully encoded by the
nuclear genome. Their assembly into a functional complex requires
an intricate process assisted by chaperones [36]. Nowadays, the
mechanisms of assembly for each complex are almost completely
resolved [36,37]. These assembly processes require the striking
coordination of intra- and extramitochondrial transcription, trans-
lation, protein import into mitochondria, and protein folding and
incorporation into assembly intermediates, up to the final complex
assembly. Any disturbance in one of these processes would alter
complexes assembly and compromised ETC function and energy
production. Mutations that impair this assembly process are a
frequent cause of mitochondrial inherited diseases [38,39]. Com-
plex misassembly is not only associated with an increasing number
of mutations in ETC subunit genes or chaperones [40] but CI
misassembly has also been recently observed in neurodegenerative
diseases such as Alzheimer’s and Parkinson’s diseases [41], thus
broadening the spectrum of assembly defect-associated diseases,
and could be implied in the ageing process [42].
Finally, assembly of CI, III, and IV together leads to respiratory
super complexes, adding a layer of complexity in the structural
organization of the respiratory chain. The crystal structure of the
super complex has recently been elucidated [43]. Although the
pathways that lead to their formation and their function are still
not completely clear [37,44], they are now recognized as the final
functional ETC unit, or “respirasome.”
“Blue Native” polyacrylamide gel electrophoresis (BN-PAGE)
[45] is a relatively easy approach for analyzing the assembly and
abundance of OXPHOS complexes for the diagnosis of mitochon-
drial diseases [46,47]. The assembly profile provides information
for identifying disease-causing mutations and facilitates molecular
investigation by highlighting potentially involved subunits
[39,48].
BN-PAGE is performed on isolated mitochondria or enriched-
mitochondrial fractions from either tissues or cells. The protocols
are described in the literature [46,47,49]. However, it should be
stressed that results depend on the choice and concentration of the
detergent and on the choice of the antibodies. After mitochondrial
isolation, the mitochondrial membranes are solubilized using non-
ionic detergent. Because mitochondrial membranes have low cho-
lesterol but high cardiolipin contents, [50], digitonin is
preferentially used for super complexes analysis while
β-dodecylmaltoside should be used for isolated complex visualiza-
tion. However, both the concentration and the detergent to pro-
tein ratio must be carefully optimized to ensure that detergent
concentration is above critical micelle concentration but below
the detergent:protein ratio that would solubilize the complex-
bound cardiolipin, which are critical for their stability and activity.
6 Naig Gueguen et al.
In addition, after transfer of the electrophoretic gels for Western
blotting, the choice of the antibodies will determine whether the
quantification of the abundance of the fully assembled complex is
privileged or the detection of assembly intermediates. In principle,
any subunit can be targeted and would allow the detection of the
assembled complex, provided that the epitope is accessible for the
antibody. However, according to the location of the subunit within
the complex and the incorporation step during the assembly pro-
cess, the assembly intermediates which could be detected and their
number will vary (Fig. 1). Moreover, due to the relatively low
dynamic range of chemiluminescence detection, the high intensity
of the signal produced by the holoenzyme could blunt other bands
of lower abundance, as the assembly intermediates. To address this
issue, the detection of assembly factors that specifically bind inter-
mediates, but detach once the holoenzymes are fully assembled
should be preferred. In this case, only the assembly intermediates
will be visualized. An example of CI assembly analysis using anti-
bodies targeting assembly factors is illustrated in Fig. 1.
It is therefore highly recommended to master the assembly
pathway of the complex of interest and the location of the different
subunits and assembly factors within the complex structure/inter-
mediates before performing a BN-PAGE analysis.
However, a normal respiratory chain enzyme activity/assembly
does not exclude a functional impairment of the respiratory chain
functioning, which may not be detectable by enzymatic measure-
ments or assembly analyses.
2.4 Functional
Analyses:
Respiration Rates
Respiration studies are an efficient way to analyze ETC and meta-
bolic activities of cells or tissues. The last, irreversible, step of
electron transfer along the ETC is the transfer catalyzed by the
CIV of four electrons to a molecule of oxygen to generate two
molecules of water. The coupling between electron transport (oxi-
dation) and proton pumping within ETC and the tight coupling
between oxidation and phosphorylation through the Δp (51, 52)
mean that the mitochondrial respiration rate is an accurate measure
of the total ETC activity and mitochondrial ATP synthesis rate.
According to the experimental design, information can be gained
on multiple processes required for respiration, including substrate
transport into the mitochondria, reducing equivalent production
by TCA cycle or beta-oxidation and electron delivery to the respi-
ratory chain, activities of the different complexes, ATP synthesis,
proton leak, and mitochondrial metabolism. Beyond metabolic
analysis, O
2
consumption is now also used to analyze cytotoxicity.
2.4.1 Devices The rate of mitochondrial O
2
consumption could be determined by
a number of methods, the two main approaches are amperometric
O
2
sensors [51] and O
2
-dependent quenching of porphyrin-based
phosphors [52]. The amperometric approach, notably developed
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 7
Fig. 1 Detection of CI assembly intermediates by targeting the CI assembly factors. (a) Schematic represen-
tation of the modular assembly of CI The CI holoenzyme results from the sequential assembly of ten
intermediates (A.I.), assisted by at least 13 CI specific assembly factors. The Q earliest intermediate is
composed of the NDUFS2, NDUFS3, NDUFS7, NDUFS8, and NDUFA5, stabilized by two specific assembly
factors, namely NDUFAF3 and NDUFAF4 in a 170 kDa intermediate (1). The binding of Q module to the ND1
subunit and accessory subunits forms the 237 to 283 kDa (2) intermediates named Q/P
P-a,
which are stacked
to the inner membrane by NDUFAF5 and NDUFAF6. The P module is assembled in four distinct intermediates,
two distal, named P
D-a
and P
D-b
that contain ND4 and ND5 subunits, respectively, and two proximal, the P
P-a
and P
P-b
. The central P
P-b
intermediate (3) is the entry point for four of the seven MT-DNA-encoded subunit,
i.e., ND2, ND3, ND6, and ND4L, and is bound by six assembly factors, among which, NDUFAF1. Then, the
different modules progressively combine (4, 5), forming a Q/P intermediate stacked with the assembly factor
NDUFAF2 (6). In the last step of the process, the 160 kDa N module (7), constituted by the catalytic NDUFV1,
NDUFV2, NDUFS1, and the accessory NDUFA2 and NDUFS4 subunits, is added. Finally, once the CI assembly
is completed, all the assembly factors are released, leaving a functional holoenzyme of 980 kDa (8). Nuclear-
8 Naig Gueguen et al.
by Oroboros Instrument, has historically been the most common
for in vitro and in vivo investigations of mitochondrial respiration.
Nevertheless, phosphorescent probes are gaining interests since the
introduction of the XF Extracellular Flux Analyzer by Seahorse
Bioscience [53].
The O2K
®
Oxygraph The O2K
®
oxygraph developed by Oroboros measures oxygen
consumption by polarography with a Clark’s electrode. Briefly,
oxygen diffuses through a Teflon membrane, which is permeable
to uncharged gases, but not to water. A platinum/silver/KCl-
coupled electrode reduces oxygen and oxidizes silver, giving rise
to a current, which is proportional to oxygen concentration within
the physiological range of measurements. It operates in a closed
chamber of 2 ml, with a stirring magnet to homogenize oxygen and
biological material. The volume can be adjusted according to the
cells, tissues, or mitochondria concentrations to analyze. High-
resolution designs of Oroboros Instruments are optimized for
high sensitivity, precision, and minimal measurement interferences
(typically the detection limit is ~1 pmol/sec/ml and the quantifi-
cation limit is ~2.5 pmol/s/ml, on site evaluations).
Fig. 1 (continued) encoded subunit names were shortened by omitting the leading “NDUF.” Only the main
subunits are indicated. The numbers refer to the corresponding intermediates that can be detected by
BN-PAGE in panel b.(b)Study of CI assembly by BN-PAGE in two control fibroblasts (Ctr), one patient cell
line with a known assembly defect [48] and cells lacking MT-DNA (143B Rho
0
). (b1) First, CI assembly was
analyzed using “classical” antibodies targeting CI structural subunits located in different intermediates, i.e.,
NDUFS2 (Q, left panel), NDUFB6 (P
D
-a, middle panel), and NDUFS1 (N, right panel). In Ctr cells, anti-NDUFS2
antibody allowed the detection of the fully assembled holoenzyme (8), while Q/P A.I. (6) accumulated in the
NDUFS4 mutated cells. No CI holoenzyme or A.I. could be detected in Rho
0
.Further hybridization with anti-
NDUFB6 and NDUFS1 antibodies highlighted the presence of the Pp-b/P
D
-a at ~700 kDa (5) and the N module
(7), respectively. (b2) A.I. were then analyzed by targeting the assembly factor NDUFAF4 (Q), NDUFAF1 (Pp-b),
and NDUFAF2 (Q/P). The detection of NDUFAF1 revealed a main band at ~400 kDa which matched with the
described size for the Pp-b intermediate and two higher faint bands just below and above 800 kDa that
matched the Q/Pp and Q/P intermediates, respectively. As expected, these intermediates containing the
MT-DNA encoded subunits were not observed in Rho
0
cells. NDUFAF4 antibody detected the expected bands
for Q-containing intermediates, i.e., a main one for Q at 170 kDa (1) and fainter ones for Q/Pp-a (at ~240 kDa
and ~280 kDa, (2)), Q/Pp (~750 kDa, (4), and Q/P (~880 kDa, (6)). As expected, only the smallest Q
intermediates, constituted only of nDNA-encoded subunits, were detected in Rho
0
cells. The Pp-b intermedi-
ate was detected on the same membrane by hybridizing anti-NDUFAF1 antibody. Finally, additional hybridiza-
tion with NDUFAF2 antibody clearly highlighted the accumulation of the Q/P (6) that occurred in NDUFS4 cells
and highlighted different higher molecular weights up to ~1100 kDa for this intermediate. Thus, since they do
not bind the CI holoenzyme, whose intense signal can blunt the presence of A.I., the targeting of the assembly
factors allows a more sensitive detection of these intermediates. The two combined approaches allow the
identification of seven of the assembly intermediates in addition to the holoenzyme
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 9
This is allowed by (1) closed, air-tight reaction chambers;
(2) high sensitivity of the sensors with minimal noise/signal ratio;
and (3) low O
2
-permeable materials, which minimizes O
2
back-
diffusion and overestimations of respiration. In these systems,
instrumental background O
2
consumption (i.e., nonbiological O
2
flux, which constitutes potential sources of systematic error) can be
corrected for accurate determinations of O
2
consumption. The
device can be fully calibrated, both for 0 and 100% oxygen signals
and oxygen concentration within the respiratory medium accord-
ing to the actual PO
2
and O
2
solubility. Therefore, real-time abso-
lute quantitative measurements of oxygen concentration and fluxes
can be measured, and robust comparison of inter-assays achieved.
Normalization of fluxes to protein content or cell concentration
can be easily achieved by recovering the samples within the chamber
at the end of the experiment.
O2K
®
oxygraphs are a fully open system, allowing as many
injections of substrates or inhibitors as needed, and reoxygenation
during the experiment time-course if needed, thus enabling
extended substrate–uncoupler–inhibitor–titration protocols to be
applied. Each complex of the ETC can be studied independently,
on the same sample, using different substrate combinations.
The capability of the O2k
®
system has recently expanded
beyond respiration to permit simultaneous fluorescence-based
measurements, potentiometric measurements of H+ concentra-
tions (i.e., pH), of ΔΨm using triphenylphosphonium or fluores-
cent membrane potential probes, and Ca
2+
, as well as amperometric
measurement of nitric oxide.
There are however main limitations when considering the use
of Oroboros technology: First, the O2k
®
is not a high-throughput
technology, as only two samples can be performed at one time, and
it is not automated. Because analysis using substrate–uncoupler–
inhibitor–titration protocols take approximately one hour to com-
plete, large-scale analyses are time-consuming. Second, analysis
requires cell dissociation and suspension. As most cell culture–
based studies are conducted on adherent cells, this system shows
limits when the experimental objective is to assess cellular bioener-
getics in intact cultured cells. According to cell types, cell detach-
ment from the extracellular matrix could alter metabolism [54], in
any way, analyses must be performed immediately and achieved
rapidly after cell detachment. Third, despite the high sensitivity,
the relative high volume of the chambers (500 μl to 2 ml) imposes a
nonnegligible sample quantity. Typically, 10–40 μg of isolated
mitochondria, 1–2 mg permeabilized tissue or 2–6·10
6
cells,
according to cell types, are needed for one experiment.
10 Naig Gueguen et al.
The Seahorse
®
XF The Seahorse
®
XF Extracellular Flux Analyzer technology is based
on solid state sensor probes for detection of oxygen and H+ con-
centrations, residing 200 μm above the cell/sample monolayer on a
multi-well microplate format. Real-time measurements of oxygen
concentrations are made by isolating a small volume of about 2 μlof
medium above a monolayer of samples in a “transient microcham-
ber” within the microplate. The technology based on fluorescence
quenching is quite sensitive and has been optimized to avoid signal
drifting.
The main advantage of Seahorse
®
technology relies on its
multi-wells plate format and automated system, allowing high-
throughput analyses. Moreover, the small chamber volume is a
great benefit when limited material is available. Finally, this system
represented a technical breakthrough for the field of bioenergetics
by providing the first instrument with the capacity to measure
in vivo O
2
consumption on intact adherent cells in culture. The
XF Analyzer minimizes the potential limitation of oxygen diffusion
for adherent cells by transiently limiting the volume of solution in
which O
2
is measured to just above the monolayer of cultured cells.
Moreover, the capability of the system expands beyond respiration
and allows simultaneous measurements of H+ extrusion (i.e., pH),
as a reflect of glycolytic rates.
However, there are main limitations to consider when analyz-
ing mitochondrial respiration using the Seahorse
®
Analyzer. First,
as for many miniaturized setups, high-throughput methods, the
device loses in accuracy what it gains in rapidity. The biosensors are
calibrated for 100% oxygen, using the supplier solution, not in the
respiratory medium used. The calibration procedure does not con-
sider the real oxygen concentration within the medium (oxygen
solubility within the medium, real barometric pressure during the
experimental course). However, the temporary microenvironment
created over the cell layer is not completely sealed from atmospheric
O
2
; thus, as O
2
concentration declines with time, O
2
gradient
favors diffusion from the media to the micro-chambers. O
2
gradi-
ent could also favor back-diffusion from the plastic of the wells. As
the device back-diffusion of oxygen is not quantified and back-
ground correction is performed relative to the “background plate
noise” determined on different “control” wells (devoid of sample),
this could lead to an underestimation of cellular respiration rate.
Thus, this system allows quite quantitative determination of O
2
variation, but it did not allow absolute quantitative measurements of
O
2
concentration and fluxes. Second, inherent to the technologies
based on fluorescence quenching is also the quite high variability
between the different acquisitions (intra-assay variability (intra-
plate, inter-sensors) and more strikingly, inter-plates variability)
[55]. This limits the comparison between different experiments.
Third, the Seahorse
®
setup is not an open system, and only four
ports surrounding the sensor are available that must be loaded
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 11
before starting the analysis. This strongly limits the ETC sub-
strates/inhibitors titration protocols. It is essential to perform
robust optimization steps prior to any respiratory assay: the optimal
concentration of each injectable reagent must be checked for each
experimental condition. Care must be also taken not to overload
the wells, which can cause O
2
to quickly become depleted under
phosphorylating conditions, leading to hypoxia. Finally, this device
requires that samples adhere to the bottom of the well, which
corresponds to an advantage when working on adherent cells, but
could require coating of the samples in other conditions.
To resume, when accurate measurement of O
2
consumption
and high versatility is aimed, as for diagnosis purpose, high-
resolution O2K
®
respirometry system should be used. However,
if high-throughput dosages are needed as for the screening of
drugs, or when the sample quantities are limited, the Seahorse
®
offers a much higher throughput platform than the traditional
Clark electrode-based systems.
2.4.2 Permeabilized
Tissues and Cells
Within tissues or cells, mitochondria are not accessible for many
substrates and inhibitors, and the large catabolic processes that
must be considered when dissecting ETC activity and mitochon-
drial metabolism greatly complicated the understanding of the
results.
Therefore, Isolation of mitochondria through tissue/cell cul-
ture homogenization and differential centrifugations is routinely
used for assessment of mitochondrial respiration [56,57]. Isolated
mitochondria remain one of best approaches for studying mito-
chondrial bioenergetics free from the influence of cellular factors
like the cytoskeleton, endoplasmic reticulum, cellular ATPases,
together with a strict control on substrate supplies. It also allows
the studies of distinct subcellular mitochondria, such as subsarco-
lemmal and intermyofibrillar mitochondria of skeletal muscle,
which display different functional properties [58]. However, the
disadvantages of isolated mitochondria include:
1. The disruptions of mitochondrial structure [59], of mitochon-
drial network, mitochondria–endoplasmic reticulum and mito-
chondria–cytoskeleton interactions, which may impact
function [60–62].
2. The purification process also bias the mitochondrial composi-
tion, by selecting high-density mitochondria, while discarding
less dense ones during differential centrifugation steps [63].
3. The purification process requires relatively large sample sizes
(roughly 20 10
6
cells, or 100–150 mg wet weight of tissue)
to obtain relevant yields.
4. The loss of micro-compartmentalization and metabolic chan-
neling [64,65].
12 Naig Gueguen et al.
An alternative approach when the control of substrate supply is
needed is the use of permeabilized tissues, mostly permeabilized
muscle fibers, or cells using low concentration of plasma-selective
detergents [66]. Saponin and digitonin, a saponin derivate, are
mostly selective for cholesterol lipids, enriched in the plasma mem-
brane. Therefore, the use of saponin (50–100 μg/ml) for tissue
permeabilization or digitonin (10–30 μg/million cells) for cell
permeabilization leaves intact all intracellular structures, including
mitochondria.
Permeabilized cell and skinned fiber techniques have several
advantages for studies of mitochondrial function:
1. very small tissue samples are required;
2. all mitochondria can be investigated;
3. more importantly, mitochondria are studied in their natural
surroundings, as the mitochondrial network is maintained as
well as the potential interactions between mitochondria and
other subcellular structures and organelles.
However, the mitochondrial integrity after sample preparations
must be carefully controlled by checking the absence of any stimu-
latory effect of cytochrome c, witnessing the maintenance of the
mitochondrial membranes’ integrity.
The classical respiration rate experiments to determine mito-
chondrial bioenergetic function were defined by Chance and Wil-
liams [67]. Substrate is added in the respiratory medium (defined as
“state 2,” no ADP), followed by addition of ADP, allowing the
ATP synthase to function. This induces a drop in Δp and thus
accelerates the electron transport and proton pumping (“state
3”). On permeabilized fibers or cells, the presence of cellular
ATPase activity maintains a high ADP/ATP ratio through constant
ATP hydrolysis and subsequent ATP recycling by mitochondrial
synthesis. State 4o is achieved by adding oligomycin, the ATP
synthase inhibitor, to inhibit ATP synthesis, increasing back the
Δp resulting in slowing down the respiration rate. Although state
2 and state 4 respiration are experimentally quite equivalent, state
4 is more conveniently used when referring to respiration under
basal, non-phosphorylating conditions. Then, addition of a proto-
nophore, such as FCCP (carbonyl cyanide-p-trifluoromethoxyphe-
nylhydrazone), gives the uncoupled respiration (“state 3u”).
State 3u is controlled exclusively by substrate oxidation module
(including substrate uptake, activity of the NADH/FADH2 pro-
ducing pathway, e.g., TCA cycle, activity of the ETC complexes,
pool sizes of ubiquinone, and cytochrome c). Inhibition of any of
these processes will decrease the state 3u rate, but due to the
presence of controlled, unlimited substrates, state 3u mainly reflect
the maximal ETC capacity for a given substrate, even if the TCA
cycle activity could be sometimes rate-limiting.
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 13
State 3 (ADP) is controlled, depending on the tissue and con-
ditions, by both the ATP synthesis module (ANT, PiC, and ATP
synthase) and substrate oxidation. Indeed, in some mitochondria,
such as those from brown fat or liver, the ATP synthase activity is
limited and displays a striking control over the ETC activity, mean-
ing that the maximal ETC activity can only be reached in uncou-
pling conditions [68]. Comparing state 3u and state 3, when ADP
is added in saturating concentration, will thus provide insight into
the control exerted by the ATP synthesis module on the substrate
oxidation one.
State 4 is controlled predominantly by the proton leak and to a
small extent by the activity of substrate transport. In optimal con-
dition, mitochondria maintain a high Δp which restricts proton
pumping and thus electron transport leading to slowdown of the
respiration rate. An increased state 4o rate for a given substrate
would indicate altered proton leak.
Importantly, the specific subset of complexes and dehydro-
genases engaged by these assays depends on the substrates
provided. These respiratory states can be sequentially measured in
the presence of different combinations of mitochondrial substrates,
allowing multiple metabolic pathways or respiratory complexes to
be probed.
A typical titration protocol used in diagnosis pipelines, using
substrates of CI, CI + CII, CII, and IV, is as followed (Fig. 2):
First, state 2 (non-phosphorylating) respiration is initiated after
adding pyruvate and malate.
Second, the state 3 respirations are detailed for the different
complexes: the CI-linked maximal phosphorylating respiration is
stimulated by saturating ADP concentration (1.5 mM). Succinate
(10 mM) is added to measure the combined CI and CII-linked
respiration with convergent CI + II electron flow into the
Q-junction corresponding to the maximal stimulated phosphory-
lating respiration (OXPHOS capacity). Rotenone is then used to
inhibit CI activity and thus to obtain the maximal CII-linked
phosphorylating respiration.
Third, oligomycin is used to inhibit F0F1-ATP synthase and
state 4o measurement. FCCP is sequentially added to uncouple
mitochondria and measure the CII-linked respiration in state 3u.
Antimycin A (2 μg/ml) is used to inhibit complex III and check for
the non-mitochondrial oxidation. Finally, ascorbate + TMPD, the
artificial complex IV substrates, allow the measurement of maximal
COX-driven respiration rate, after inhibition of complex IV by
potassium cyanide and subtraction of this nonspecific oxidation.
The respiratory control ratio (RCR), defined as the quotient of
maximal state 3 to state 4 respirations, is often used as an index of
mitochondrial coupling of oxidation to phosphorylation for a given
substrate combination. The RCR indeed integrates the increase in
14 Naig Gueguen et al.
Fig. 2 Typical analyze traces of O2 consumption on permeabilized fibers. (a) Segmental analysis of ETC
function using sequential substrates injections. Mitochondrial oxygen consumption measurements were
performed at 37 C and atmospheric pressure using a high-resolution oxygraph (O2K, Oroboros Instrument,
Innsbruck, Austria). Respiration rates on permeabilized fibers using 50 μg/ml saponin (30 minutes, 4 C) were
measured in respiratory buffer (10 mM KH
2
PO
4
, 300 mM mannitol, 10 mM KCl, 5 mM MgCl
2
, 0.5 mM EGTA,
and 1 mg/ml serum albumin bovine, pH 7.2) using substrates of CI, CI + CII, and CII as followed: first, state
2 (non-phosphorylating) respiration was measured after adding 2.5 mM pyruvate and 5 mM malate. Then, the
CI-linked maximal phosphorylating respiration was stimulated by saturating ADP concentration (1.5 mM) and
3 mM NAD+, added to avoid TCA limitation by NAD+ availability. CI-linked maximal phosphorylating
respiration was further stimulated with 5 mM glutamate, to check for any limitation of substrate availability
by PDH activity. Succinate (10 mM) was then added to measure the combined CI and CII-linked respiration
with convergent CI + II electron flow into the Q-junction corresponding to the maximal stimulated phosphor-
ylating respiration (OXPHOS capacity). Rotenone (5 μM) was used to inhibit CI activity and thus to obtain the
maximal CII-linked respiration. Thirdly, oligomycin (F0F1-ATP synthase inhibitor, 4 μg/ml) and FCCP (carbonyl
cyanide-p-trifluoromethoxyphenylhydrazone, a mitochondrial uncoupler, 1 μM) were sequentially added to
ensure that the cells were fully permeabilized. Antimycin A addition (2 μg/ml) was used to check for the
non-mitochondrial oxidation. CIV maximal respiration was induced with 4 mM ascorbate and 0.3 mM TMPD,
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 15
respiration rate in response to Δp use for ATP synthesis and the low
respiration rate linked to proton leak in non-phosphorylating con-
ditions. RCR values depend on almost every OXPHOS functional
aspect and could therefore be a useful indicator of mitochondrial
dysfunction. However, RCR determinations are sensitive to experi-
mental inaccuracies; depending on the quality of sample prepara-
tions and accurate determination of background rate, state 4 rates
can be significantly over/underestimated. Moreover, there are no
absolute RCR values, as these ones are substrate- and tissue-
dependent. For example, substrates such as succinate or glycerol
phosphate translocate fewer protons per electron pair than NADH-
linked substrates, so the maintenance of the same Δp requires a
higher respiration rate. Furthermore, under identical substrate
conditions, different values may be observed for different tissues,
reflecting different substrate oxidation kinetics, endogenous pro-
ton leak, or phosphorylating capacities [68]. Thus, careful cautions
according to the experimental conditions should be recommended
when interpreting RCR values.
Permeabilized cells or fibers are also a useful model to assess the
maximal activity of the main substrate-providing pathway, i.e., TCA
cycle or beta-oxidation activities. Thus, different TCA cycle inter-
mediates or mitochondrial shuttles substrates (malate/aspartate,
glycerol phosphate shuttles) can be used to dissect specific regula-
tions within these pathways. Similarly, the comparison of maximal
respiration rates sustained in the presence of TCA cycle substrates
(malate, pyruvate, glutamate, and succinate for a fully operating
cycle) or in the presence of beta-oxidation substrates (short-chain
or long-chain fatty acids complexed with coenzyme A or
L-carnitine) allows determining the preferred metabolic pathways
according to pathophysiological conditions, treatments, or tissues.
Measurements of maximal respiratory chain complex activities,
assembly, and linked respirations are the cornerstone on which the
biochemical diagnosis of mitochondrial disorders is based. How-
ever, these can usefully be implemented by more integrative meth-
ods, approaching mitochondrial metabolism.
2.4.3 Intact Cells The use of isolated mitochondria or permeabilized cells has been
favored for years. Because the experimenter has control over con-
ditions, i.e., substrates availability and respiratory states, this
remains the method of choice to gain mechanistic insight into
respiration chain function and dysfunctions. However, in the last
Fig. 2 (continued) followed by CIV inhibition using 1 mM KCN and azide. (b) Analysis of O2 consumption linked
to the stimulation of β-oxidation. β-oxidation was stimulated using palmitoyl-L-carnitine, supplemented with
2.5 mM malate (PCM). Antimycin A addition (2 μg/ml) was used to check for the non-mitochondrial oxidation.
CIV maximal respiration was induced with 4 mM ascorbate and 0.3 mM TMPD, followed by CIV inhibition using
1 mM KCN and azide
16 Naig Gueguen et al.
decades, respiration analyses on intact cells became prominent in
the field. With whole cell models, intrinsic activity of respiratory
chain complexes is difficult to test and respiration rates somehow
difficult to interpret, since their activities are integrated with a
myriad of parameters. However, overriding these complexities is
the obvious higher physiological relevance. Using intact cells, mito-
chondria are present in a physiological environment with preserved
interactions with the rest of the cell. Therefore, this model is ideal
to assess metabolic pathways connected directly or indirectly to
mitochondrial activities.
A typical experiment using intact cells starts by the measure-
ment of respiratory rate in the respiratory medium, e.g., cell culture
medium, without any additions of inhibitors or uncoupler (Fig. 3).
This state is defined as the basal or routine respiration. Routine
respiration is the minimal rate of oxidative metabolism required to
Fig. 3 Typical analysis traces of O
2
consumption on intact cells. Mitochondrial oxygen consumption measure-
ments were performed at 37 C and atmospheric pressure using a high-resolution oxygraph (O2K, Oroboros
Instrument, Innsbruck, Austria). Respiration rates on primary fibroblast cells were measured in either DMEM-
F12 medium (3 g/l glucose, 0.3 mM pyruvate) (red traces) or in low-glucose medium (0.5 g/l glucose) (green
Traces). 4 10
6
cells were added in the oxygraphic chamber and the analysis started with routine respiration
(R) measurement, which is defined as respiration in medium without additional substrates or effectors (cell
endogenous respiration, corresponding to the cellular oxidative metabolism). Then, 1 mM glutamine was
added, to test for cell metabolic dependence to glutaminolysis. F0F1-ATP synthase was inhibited with
oligomycin (4 μg/ml), allowing the measurement of non-phosphorylating respiration (leak respiration). This
non-phosphorylating respiration (O) was subtracted from routine (R) one to calculate the cellular phosphor-
ylating respiration (R-O). This was followed by uncoupling of oxidative phosphorylation by stepwise titration of
FCCP (carbonyl cyanide-p-trifluoromethoxyphenylhydrazone) up to optimum concentrations allowing the
measurement of the maximal endogenous respiration (cellular oxidative capacity). The part of the maximal
capacity use for oxidative metabolism was calculated as R/F, and the part of the maximal capacity use for
oxidative ATP synthesis was calculated as (R-O)/F. Finally, respiration was inhibited by rotenone and antimycin
A (2.5 μM and 2 μg/ml, respectively) to check for non-mitochondrial oxidation
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 17
fulfill energy needs and support cell functions. The routine respira-
tion of most mammalian cells corresponds neither to the state
3 (unlimited availability of substrate and ADP) nor to the state 4o
(unlimited availability of substrate, but no ATP synthesis), but is
generally an intermediate state between these ones, according to
the ATP demand. Routine respiration is usually strongly controlled
by ATP turnover, but in many cultured mammalian cells, aerobic
glycolysis also contributes to the total ATP turnover [69], meaning
that the routine respiration is not equivalent to cell metabolic rate.
Routine respiration is further partly dependent on substrate oxida-
tion (including substrate uptake, TCA cycle, activity of the ETC
complexes...) and proton leak [70]. Therefore, routine respiration
could differ according to the substrate availability in the incubation
medium, to the oxidative ATP synthesis needs or in the presence of
any agents decreasing the Δp (uncouplers). For example, routine
respiration can be increased by decreasing glucose concentration
(Fig. 3). A modification in routine rate integrates any of these
changes and is therefore quite difficult to interpret, but the follow-
ing steps can help to elucidate the pathway involved.
Routine respiration measurement is followed by mitochondrial
ATP synthesis inhibition, through oligomycin addition. The result-
ing respiration is defined as leak respiration. Indeed, as for permea-
bilized cells, this respiration rate in the presence of oligomycin is
controlled predominantly not only by the proton leak but also, to a
smaller extend, by the activity of substrate or ions (e.g., Ca
2+
)
transports and substrate being oxidized. In intact cells, the sub-
strates oxidized by the ETC are not under the control of the
experimenter. Because the proton pumping is not equivalent
according to the substrate used, the respiration rate required to
maintain the Δp is also different. Thus, a modest change in leak
respiration rate may indicate either a change in proton leak or a
change in Δp caused by altered substrate oxidation or transport.
However, a large increase in the respiration rate strongly suggests
uncoupled mitochondria.
This leak respiration rate is subtracted from the preceding
routine state to estimate the in-situ phosphorylating respiration,
i.e., the respiration rate linked to ATP synthesis. However, the
proton leak is voltage-dependent, all the more important as the
ΔΨ is high. Since ATP synthase inhibition results in an increase in
ΔΨ, subtracting the oligomycin-insensitive respiration from the
routine one slightly overvalues the part of proton leak involved in
routine respiration and underestimates ATP synthesis. This remains
nevertheless one of the best approaches to estimate in situ phos-
phorylating respiration when absolute quantitative values are not
needed [71]. Also, calculating the part of the phosphorylating
versus leak respirations within the routine one further helps to
decipher which process explains a change in the routine respiration
18 Naig Gueguen et al.
rate. While an increase in phosphorylating respiration most likely
reflects the response of oxidative metabolism to increased energy
demand, a decrease in the phosphorylating respiration could illus-
trate a decrease in energy needs, a metabolic rewiring from oxida-
tive metabolism to “anaerobic” glycolytic one (the so-called
Warburg effect), or a blockage in the ATP synthesis module.
The next step involves uncoupler titration, e.g., FCCP titra-
tion, to obtain the maximal uncoupled respiration rate. As for
uncoupled respiration measured on permeabilized cells, this one is
control by the substrate oxidation pathways. However, there is a
marked difference here: while on permeabilized cell, the uncoupled
respiration is measured in the presence of unlimited availability of
defined substrate(s), meaning that uncoupled respiration is mainly
a reflect of maximal ETC capacity; on intact cells, the catabolic
pathways supplying ETC with NADH or FADH
2
is under complex
metabolic regulation and can be rate-limiting. Therefore, the
uncoupled respiration rate on intact cells reflects the maximal over-
all substrate oxidation capacity, from substrate uptake and catabo-
lism to ETC activity, achievable by cells under the assay condition.
A decrease in this maximal respiratory capacity is a reliable indicator
of mitochondrial dysfunction but, because of these complexities,
caution should be taken when interpreting which pathway is
responsible for this dysfunction. Comparing the maximal
uncoupled respiration obtained on permeabilized cells versus intact
cells provides further clues to determine whether a reduced ETC
capacity is involved or not.
CCCP and FCCP are proton shuttling compounds selectively
increasing the permeability of lipid membranes to protons. How-
ever, CCCP and FCCP also have high reactivity with thiol groups.
Thiol-combining agents uncouple OXPHOS at low concentrations
but inhibit respiration at high concentrations by the chemical mod-
ification of a small but significant number of mitochondrial thiol
groups [72]. Moreover, sustained perturbation of ΔΨm, which is
normally tightly controlled to ensure cell proliferation and survival,
triggers cellular stress responses, eventually leading to cell apoptosis
[72]. The optimal FCCP/CCCP concentration allowing the maxi-
mal respiration rate recording depends on cell types, medium com-
position (e.g., low glucose versus high glucose, fatty acids, or
not...), and mitochondria physiological state (e.g., any drug
which modifies the ΔΨm could change the optimal concentration).
Therefore, careful FCCP/CCCP titration must be performed for
each experimental condition to ensure the full achievement of
uncoupling yet limiting the drop in ΔΨm. When using a Seahorse
Analyzer, only two FCCP injections are allowed, therefore prelimi-
nary experiments must be performed before starting any complete
experiment. Unappropriated uncoupler doses are a common bias
observed when working on intact cells.
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 19
Finally, the experiment ends by the addition of ETC inhibitors,
rotenone, and antimycin A to determine residual oxygen consump-
tion. Residual oxidation (Rox) is the respiration due to oxidative
side reactions remaining after full inhibition of the electron transfer
pathway. Mitochondrial respiration is frequently corrected for Rox
as the non-mitochondrial respiration. However, Rox may be partly
related to ROS production which consumes O
2
and is increased
upon ETC inhibitor additions [73].
From these different respiratory parameters can be inferred
important ratios, particularly the spare respiratory capacity. The
spare respiratory capacity is the ratio of the routine to the
uncoupled respirations. With the cautions about the determination
of maximum rates described above, spare respiratory capacity indi-
cates the part of the ETC capacity that is used to sustain cell
function, i.e., how close to its bioenergetic limit cells are operating
[74]. While this ratio may be particularly informative, again, given
the complexity of the mechanisms involved, cautions must be taken
when interpreting it. For example, a decrease in spare capacity may
reflect an increase in cellular ATP requirements and a system
increasing its production, or a decrease in the substrate oxidation
capacity that limits uncoupled respiration, depending on whether
routine respiration increases or maximal respiration decreases. Sim-
ilarly, an increase in spare capacity may suggest either a decrease in
energy demand or a metabolic switch toward glycolysis, which then
mostly ensures the ATP production, although the capacity of the
ETC remains unchanged.
When working with intact cells, cell metabolism “determines”
which substrates mitochondria can used. However, the experi-
menter sets the extracellular conditions. The choice of the exact
respiratory medium composition, i.e., the substrates composition
and concentrations, the presence of hormones, growth factors,
cytokines, may determine the outcome of the analysis. This could
be a pitfall, but this also offers the opportunity to perform cell
metabolic phenotyping. High concentration of glucose is often
used in cell culture, but some cells prefer to oxidize other compo-
nents of the medium, such as fatty acids or amino acids, particularly
glutamine. For example, cancer cells undergo profound metabolic
rewiring for rapid growth. They not only could derive most of their
energy from glycolysis rather than OXPHOS but also could depend
on oxidative glutaminolysis or modified TCA cycle activity to sus-
tain their biosynthesis [75]. By modifying the medium composi-
tion, such as the glucose concentration (Fig. 3), or the presence of
glutamine, the experimenter can easily test which specific pathway is
preferentially used by cells to sustain their oxidative metabolism.
2.5 Mitochondrial
Membrane Potential
The ΔΨm is central to the bioenergetic processes. It not only
provides the driving force for ATP synthesis and dictates the ETC
activity as detailed above but also regulates metabolites transport
20 Naig Gueguen et al.
across the inner mitochondrial membrane [76], the mitochondrial
NADPH synthesis through transhydrogenase reaction [77], the
mitochondrial calcium buffering capacity [78], the ROS produc-
tion [79], and the cell apoptotic pathway [80].
Because of its apparent simplicity, fluorescent monitoring of
ΔΨm with “mitochondrial membrane potential indicators” is the
most common technique used for monitoring mitochondrial func-
tion at the single cell or even at the single-mitochondrial levels. A
number of cationic fluorescent probes have been developed for
assessing changes in mitochondrial membrane potential in cultured
cells using both microscopy and flow cytometry. This approach and
the advantage/limits of the available dyes have been extensively
reviewed elsewhere [81,82]. It should be stressed that any of
these cationic dyes, even at low concentration, inhibits the
OXPHOS activity to some extend, particularly the phosphorylating
respiration. Among these one, TMRM seems to display the lowest
side effects [82].
However, this simplicity is counterbalanced by the limited
information that can be inferred from this type of analysis. Indeed,
it is particularly difficult to calibrate the signal measured with these
probes [83–85]. Generally, studies are qualitative, indicating
whether mitochondria are “depolarized” or not, or, at best, relative
(semiquantitative); consequently, discrete variations are difficult to
detect using classical fluorescence approaches, while mitochondrial
ΔΨm is normally maintained by mitochondrial respiration and
therefore only discrete variations occur from state 4 to state 3 tran-
sitions. Moreover, because ΔΨm results from the balance between
ETC activity and the rate of back across into the matrix, its varia-
tions are difficult to interpret in absence of any information on
mitochondrial metabolic states (respiration measurement).
On isolated mitochondria or permeabilized cells, ΔΨm mea-
sures are typically performed using electrodes sensitive to potential
such as triphenylmethyl phosphonium cation (TPP+) [86]. Since
the development of technologies expanded beyond respiration to
allow multiplexing (Oroboros O2K), assessment of respiration
rates and ΔΨm (either fluorescence-based or potentiometric mea-
surements (TPP+)) can be achieved simultaneously. Measurement
of both proton fluxes and ΔΨm enables a full and quantitative
description of ETC functioning and is essential to detect whether
any change in ΔΨm is the primary or a secondary defect due to
ETC inhibition. It allows detecting subtle changes in proton leak
and coupling efficiency between mitochondrial preparations, exper-
imental conditions, or treatments. Because the proton leak is non-
ohmic, increasing disproportionately at high Δp[68], it is more
accurately determined by plotting the proton current (respiration
rate) to voltage (ΔΨm) over a wide range of potentials in the
absence of ATP synthesis [68]. This is achieved by progressively
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 21
inhibiting ETC activity, for example, by titrating rotenone with
mitochondria oxidizing NADH, and simultaneously assessing the
respiration rate and ΔΨm in state 4o condition.
Calculation of membrane potential is based on the Nernst
equation and requires an estimate of the matrix volume over
which the probe distributes. Thus, only the TPP+-based measure-
ments enable the calculation of the absolute quantitative values,
while the fluorescence-based technology is restricted to relative
measurements. Moreover, the absolute values of ΔΨm can be cal-
culated on isolated mitochondria, taking into account the “bind-
ing” factor of the TPP+ to membranes, while the absolute values of
ΔΨm cannot be reasonably calculated on permeabilized cells, due
to strongest unspecific binding [87].
Integrating the mechanistic studies of respiration rates and
ΔΨm on permeabilized cells and the metabolic studies of respira-
tion rates on intact cells provide a large picture of bioenergetic
dysfunctions. These approaches can be complemented by metabo-
lite dosages or more broadly by a cellular metabolomic analysis to
obtain a broader picture of the cellular metabolism and its
dysfunctions.
2.6 Metabolites
Dosages
and Metabolomics
2.6.1 Metabolite
Measurements
Defects in the mitochondrial energy production often lead to
higher lactate production due to reduced pyruvate utilization by
the mitochondria. Elevated lactate levels are easily detected in cell
culture medium or, in vivo, in blood, urine, and/or CSF. Indeed, in
cases of respiratory chain defect, the cytosolic, mitochondrial, or
both, redox states (NAD+/NADH) often increase due to decrease
in mitochondrial NADH oxidation and upregulation of glycolysis
to support ATP needs. These increased redox states can be detected
by the so-called “metabolic indicators,” i.e., an increase lactate/
pyruvate ratio shifted by the higher NADH/NAD cytosolic ratio,
while a shift in the mitochondrial redox state increases the ratio of
the 3-OH-butyrate to acetoacetate [88]. However, these features
are neither specific nor sensitive as a diagnostic test since other
defects in cellular metabolism could increase glycolytic rate and
lactate production.
Amino acid analysis can further reveal elevated alanine, as a
by-product of the transamination of pyruvate by alanine amino-
transferase, and confirm the hyperlactacidemia.
In addition, elevated levels of several other metabolites can be
observed upon ETC impairment. Particularly, increased levels of
TCA cycle intermediates, such as malate, succinate, 2-oxoglutarate,
and fumarate, are useful markers of mitochondrial metabolism
dysfunction [40,89].
2.6.2 Metabolomics As we enter the new era of omics technologies, targeted metabo-
lomic or global unbiased approaches can strikingly improve the
investigation of the multifaced mitochondrial functions.
22 Naig Gueguen et al.
Metabolomics is defined as an integrative approach consisting
in the comprehensive analysis of the metabolome comprising
thousands of small molecules present in biological samples. Mass
spectrometry (MS) coupled to liquid or gas chromatography is
among the major analytical tools used in metabolomics. It enables
comprehensive and systematic profiling of disease conditions and is
mostly used for identifying biomarkers or drug targets in mito-
chondrial diseases [90]. Metabolomics is now used to analyze on
a global scale the multitude of downstream effects of mitochondrial
dysfunction, including not only the consequences of energy defi-
ciency but also oxidative stress, NAD+/NADH redox imbalance,
and large metabolic rewiring [91,92]. For example, disturbed
cellular dNTP pools and one-carbon metabolism have been evi-
denced in diseases associated with mtDNA maintenance defects
[93,94]. Also, the metabolomic signature of Opa1 deficiency, a
gene involved in mitochondrial fusion, unexpectedly evidenced
aspartate and glutamate depletions, two main precursors of the
nucleotide synthesis pathways, which could partly explain the
mtDNA maintenance defect observed in OPA1 disease [92,95].
Thus, combining metabolomics and bioenergetics studies is a
promising strategy to improve our understanding of the pathologi-
cal mechanisms beyond the energetic deficiency, which is often
limited in explaining the clinical phenotypes observed in patients
[91,96], and to identify new routes for therapeutic solutions. This
was recently illustrated by reference [97] which demonstrated that
methionine supplementation induces an upregulation of electron
transport chain activity and respiration, related to an enhancement
of mitochondrial pyruvate uptake and TCA cycle activity using a
yeast model.
3 Discussion
Because of their complex properties and strategic role in cellular
metabolism, understanding mitochondrial functions and dysfunc-
tions in diseases remains a huge challenge. However, the recent
development of novel technologies allowing multiplexed and high-
throughput analyses contributed to improve our capacities for
subtle characterizations of mitochondrial dysfunctions.
Integrating the mechanistic studies of respiration rates and
ΔΨm on isolated mitochondria or permeabilized cells, and meta-
bolic studies using intact cells and metabolomic approaches, now
depicts a large overview of the cell metabolism and open a new area
of mitochondrial medicine. However, integration of different data
set, from the multidisciplinary approaches to create a more com-
plete representation of the pathophysiological mechanisms,
remains challenging and needs the development of robust integra-
tive bioinformatic tools, to predict perturbations and ultimately
translate our knowledge into improved patient care.
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 23
Until recent years, the main vision was that the energetic deficit
due to ineffective OXPHOS is the starting point for explaining the
pathophysiology of mitochondrial disease. However, this energetic
deficit alone often failed to explain the broad variability of affected
organs and clinical presentations. Considering the recent progress
resulting from the development of new technologies and omic
approaches, this old view now seems outdated. Thus, mitochondria
are now considered as master regulator of cellular homeostasis,
named mitohormesis; not only do they orchestrate the metabolic
stress response but also the ROS and proteostatic stress response,
among which the unfolded protein response and the integrated
stress response [98,99].
Nowadays, increasing evidences are accumulating showing that
these stress responses are main contributors to mitochondrial dis-
ease, rather than the OXPHOS defect by itself [100]. These recent
and unexpected developments open exciting perspectives for ther-
apeutic strategies of these disorders.
After all, mitochondria are no more considered as the inside-
cell powerhouse, since forms of extracellular mitochondria can be
found free (free Mitos), enclosed by a membrane as inside platelets
or vesicles, or as cell-free circulating mtDNA [101]. Recently, Al
Amir Dache et al. reported that blood contains intact cell-free full-
length mitochondrial DNA in dense and biologically stable struc-
tures over 0.22 μm in diameter and that these structures have
specific mitochondrial proteins, double membranes, and a mor-
phology resembling that of mitochondria [102]. More experimen-
tal studies suggest that mitochondria may be released and
transferred between cells [102,103]. These intriguing observations
are only starting to be characterized, while their functions remain
unknown. Whether they can elicit regenerative effects, induce para-
crine or endocrine, pro- or anti-inflammatory immune responses
[20,104] and more largely participate in the patho-mechanisms of
mitochondrial diseases are actually unsettled.
Furthermore, the biochemical diagnosis of mitochondrial dis-
ease, usually performed on muscle or liver biopsy and primary
fibroblasts cultured from skin biopsies, remains quite an invasive
approach. Determining whether circulating blood mitochondria
could be used to unravel mitochondrial dysfunction or be used as
a biomarker of disease will enhance our diagnosis tools. The poten-
tial role of these intriguing mitochondria or its spinoffs in blood of
patients remains to be elucidated in many pathologies [6,11,101].
The future of mitochondrial medicine is undoubtedly linked to
better comprehension of its dysfunction. All new investigational
drugs for the therapy of mitochondrial diseases have the potential
to markedly alleviate clinical symptoms, and none has the capacity
to actually cure a particular mitochondrial disease
permanently [27].
24 Naig Gueguen et al.
Acknowledgments
The authors thank the following institutions and patient associa-
tions: Universite
´d’Angers and CHU d’Angers, Fondation Mala-
dies Rares and UNADEV.
References
1. Burki F (2016) Mitochondrial evolution:
going, going, gone. Curr Biol 26:
R410–R412. https://doi.org/10.1016/j.
cub.2016.04.032
2. Smith RL, Soeters MR, Wu¨st RCI, Houtkoo-
per RH (2018) Metabolic flexibility as an
adaptation to energy resources and require-
ments in health and disease. Endocr Rev
39:489–517. https://doi.org/10.1210/er.
2017-00211
3. Veloso CD, Belew GD, Ferreira LL et al
(2019) A mitochondrial approach to cardio-
vascular risk and disease. Curr Pharm Des
25:3175–3194. https://doi.org/10.2174/
1389203720666190830163735
4. Onyango IG (2017) Mitochondria in the
pathophysiology of Alzheimer’s and Parkin-
son’s diseases. Front Biosci 22:4521.
https://doi.org/10.2741/4521
5. Jodeiri Farshbaf M, Ghaedi K (2017) Hun-
tington’s disease and mitochondria. Neuro-
tox Res 32:518–529. https://doi.org/10.
1007/s12640-017-9766-1
6. Edeas M, Saleh J, Peyssonnaux C (2020)
Iron: innocent bystander or vicious culprit in
COVID-19 pathogenesis? Int J Infect Dis
97:303–305. https://doi.org/10.1016/j.
ijid.2020.05.110
7. Rongvaux A (2018) Innate immunity and
tolerance toward mitochondria. Mitochon-
drion 41:14–20. https://doi.org/10.1016/
j.mito.2017.10.007
8. Vyas S, Zaganjor E, Haigis MC (2016) Mito-
chondria and cancer. Cell 166:555–566.
https://doi.org/10.1016/j.cell.2016.07.
002
9. Zhang D, Guo R, Lei L et al (2020) COVID-
19 infection induces readily detectable mor-
phological and inflammation-related pheno-
typic changes in peripheral blood monocytes,
the severity of which correlate with patient
outcome. medRxiv 2020.03.24.20042655.
doi:https://doi.org/10.1101/2020.03.24.
20042655
10. Zhou F, Yu T, Du R et al (2020) Clinical
course and risk factors for mortality of adult
inpatients with COVID-19 in Wuhan, China:
a retrospective cohort study. Lancet
395:1054–1062. https://doi.org/10.1016/
S0140-6736(20)30566-3
11. Saleh J, Peyssonnaux C, Singh KK, Edeas M
(2020) Mitochondria and microbiota dys-
function in COVID-19 pathogenesis. Mito-
chondrion 54:1–7. https://doi.org/10.
1016/j.mito.2020.06.008
12. Lodigiani C, Iapichino G, Carenzo L et al
(2020) Venous and arterial thromboembolic
complications in {COVID}-19 patients
admitted to an academic hospital in Milan,
Italy. Thromb Res 191:9–14. https://doi.
org/10.1016/j.thromres.2020.04.024
13. Giannis D, Ziogas IA, Gianni P (2020) Coag-
ulation disorders in coronavirus infected
patients: COVID-19, SARS-CoV-1, MERS-
CoV and lessons from the past. J Clin Virol
127:104362. https://doi.org/10.1016/j.jcv.
2020.104362
14. Zhang Y, Xiao M, Zhang S et al (2020) Coa-
gulopathy and antiphospholipid antibodies in
patients with Covid-19. NEJM 38:1–3
15. Oxley TJ, Mocco J, Majidi S et al (2020)
Large-vessel stroke as a presenting feature of
Covid-19 in the young. N Engl J Med 382:
e60. https://doi.org/10.1056/
NEJMc2009787
16. Chevrollier A, Cassereau J, Ferre
´Metal
(2012) Standardized mitochondrial analysis
gives new insights into mitochondrial dynam-
ics and OPA1 function. Int J Biochem Cell
Biol 44:980–988. https://doi.org/10.1016/
j.biocel.2012.03.006
17. Santidrian AF, Matsuno-Yagi A, Ritland M
et al (2013) Mitochondrial complex I activity
and NAD+/NADH balance regulate breast
cancer progression. J Clin Invest
123:1068–1081. https://doi.org/10.1172/
JCI64264
18. Canto
´C, Menzies KJ, Auwerx J (2015) NAD
+ metabolism and the control of energy
homeostasis: a balancing act between mito-
chondria and the nucleus. Cell Metab
22:31–53. https://doi.org/10.1016/j.cmet.
2015.05.023
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 25
19. Jez
ˇek P, Hlavata
´L (2005) Mitochondria in
homeostasis of reactive oxygen species in
cell, tissues, and organism. Int J Biochem
Cell Biol 37:2478–2503. https://doi.org/
10.1016/j.biocel.2005.05.013
20. Schieber M, Chandel NS (2014) ROS func-
tion in redox signaling and oxidative stress.
Curr Biol 24:R453–R462. https://doi.org/
10.1016/j.cub.2014.03.034
21. Finkel T (2011) Signal transduction by reac-
tive oxygen species. J Cell Biol 194:7–15.
https://doi.org/10.1083/jcb.201102095
22. Anderson NM, Mucka P, Kern JG, Feng H
(2018) The emerging role and targetability of
the TCA cycle in cancer metabolism. Protein
Cell 9:216–237. https://doi.org/10.1007/
s13238-017-0451-1
23. Martı
´nez-Reyes I, Chandel NS (2020) Mito-
chondrial TCA cycle metabolites control
physiology and disease. Nat Commun
11:102. https://doi.org/10.1038/s41467-
019-13668-3
24. Romero-Garcia S, Prado-Garcia H (2019)
Mitochondrial calcium: transport and modu-
lation of cellular processes in homeostasis and
cancer (Review). Int J Oncol. https://doi.
org/10.3892/ijo.2019.4696
25. Quiles JM, Gustafsson ÅB (2020) Mitochon-
drial quality control and cellular proteostasis:
two sides of the same coin. Front Physiol
11:515. https://doi.org/10.3389/fphys.
2020.00515
26. Wang C, Youle RJ (2009) The role of mito-
chondria in apoptosis. Annu Rev Genet
43:95–118. https://doi.org/10.1146/
annurev-genet-102108-134850
27. Weissig V (2020) Drug development for the
therapy of mitochondrial diseases. Trends
Mol Med 26:40–57. https://doi.org/10.
1016/j.molmed.2019.09.002
28. Bioenergetics 3 (Nicholls, D. G., and Fergu-
son, S. J., Academic Press, London, 2002).
Biochemistry 69, 818–819 (2004). doi:
https://doi.org/10.1023/B:BIRY.
0000040210.06512.a7
29. Hatefi Y (1985) The mitochondrial electron
transport and oxidative phosphorylation sys-
tem. Annu Rev Biochem 54:1015–1069.
https://doi.org/10.1146/annurev.bi.54.
070185.005055
30. Divakaruni AS, Brand MD (2011) The regu-
lation and physiology of mitochondrial pro-
ton leak. Physiology 26:192–205. https://
doi.org/10.1152/physiol.00046.2010
31. De Stefani D, Rizzuto R, Pozzan T (2016)
Enjoy the trip: calcium in mitochondria back
and forth. Annu Rev Biochem 85:161–192.
https://doi.org/10.1146/annurev-bio
chem-060614-034216
32. Hsieh VC, Krane EJ, Morgan PG (2017)
Mitochondrial disease and anesthesia. J
Inborn Errors Metab Screen
5:232640981770777. https://doi.org/10.
1177/2326409817707770
33. Haschke RH, Fink BR (1975) Lidocaine
effects on brain mitochondrial metabolism
in vitro. Anesthesiology 42:737–739.
https://doi.org/10.1097/00000542-
197506000-00018
34. Gellerich FN, Mayr JA, Reuter S et al (2004)
The problem of interlab variation in methods
for mitochondrial disease diagnosis: enzy-
matic measurement of respiratory chain com-
plexes. Mitochondrion 4:427–439. https://
doi.org/10.1016/j.mito.2004.07.007
35. Medja F, Allouche S, Frachon P et al (2009)
Development and implementation of standar-
dized respiratory chain spectrophotometric
assays for clinical diagnosis. Mitochondrion
9:331–339. https://doi.org/10.1016/j.
mito.2009.05.001
36. Signes A, Fernandez-Vizarra E (2018) Assem-
bly of mammalian oxidative phosphorylation
complexes I–V and supercomplexes. Essays
Biochem 62:255–270. https://doi.org/10.
1042/EBC20170098
37. Cogliati S, Lorenzi I, Rigoni G et al (2018)
Regulation of mitochondrial electron trans-
port chain assembly. J Mol Biol
430:4849–4873. https://doi.org/10.1016/
j.jmb.2018.09.016
38. Mimaki M, Wang X, McKenzie M et al (2012)
Understanding mitochondrial complex I
assembly in health and disease. Biochim Bio-
phys Acta Bioenerg 1817:851–862. https://
doi.org/10.1016/j.bbabio.2011.08.010
39. Desquiret-Dumas V, Leman G, Wetterwald C
et al (2019) Warburg-like effect is a hallmark
of complex I assembly defects. Biochim Bio-
phys Acta Mol basis Dis 1865:2475–2489.
https://doi.org/10.1016/j.bbadis.2019.05.
011
40. Wortmann S, Rodenburg RJT, Huizing M
et al (2006) Association of
3-methylglutaconic aciduria with sensori-
neural deafness, encephalopathy, and Leigh-
like syndrome (MEGDEL association) in four
patients with a disorder of the oxidative phos-
phorylation. Mol Genet Metab 88:47–52.
https://doi.org/10.1016/j.ymgme.2006.
01.013
41. Giachin G, Bouverot R, Acajjaoui S et al
(2016) Dynamics of human mitochondrial
complex I assembly: implications for
26 Naig Gueguen et al.
neurodegenerative diseases. Front Mol Biosci
3:43. https://doi.org/10.3389/fmolb.2016.
00043
42. Miwa S, Jow H, Baty K et al (2014) Low
abundance of the matrix arm of complex I in
mitochondria predicts longevity in mice. Nat
Commun 5:3837. https://doi.org/10.
1038/ncomms4837
43. Letts JA, Fiedorczuk K, Sazanov LA (2016)
The architecture of respiratory supercom-
plexes. Nature 537:644–648. https://doi.
org/10.1038/nature19774
44. Dudkina NV, Sunderhaus S, Boekema EJ,
Braun H-P (2008) The higher level of organi-
zation of the oxidative phosphorylation sys-
tem: mitochondrial supercomplexes. J
Bioenerg Biomembr 40:419–424. https://
doi.org/10.1007/s10863-008-9167-5
45. Sch€
agger H, von Jagow G (1991) Blue native
electrophoresis for isolation of membrane
protein complexes in enzymatically active
form. Anal Biochem 199:223–231. https://
doi.org/10.1016/0003-2697(91)90094-A
46. Calvaruso MA, Smeitink J, Nijtmans L (2008)
Electrophoresis techniques to investigate
defects in oxidative phosphorylation. Meth-
ods 46:281–287. https://doi.org/10.1016/
j.ymeth.2008.09.023
47. Leary SC (2012) Blue native polyacrylamide
gel electrophoresis: a powerful diagnostic tool
for the detection of assembly defects in the
enzyme complexes of oxidative phosphoryla-
tion. Methods Mol Biol 837:195–206.
https://doi.org/10.1007/978-1-61779-
504-6_13
48. Assouline Z, Jambou M, Rio M et al (2012) A
constant and similar assembly defect of mito-
chondrial respiratory chain complex I allows
rapid identification of {NDUFS}4 mutations
in patients with Leigh syndrome. Biochim
Biophys Acta Mol Basis Dis
1822:1062–1069. https://doi.org/10.
1016/j.bbadis.2012.01.013
49. Wittig I, Braun H-P, Sch€
agger H (2006) Blue
native PAGE. Nat Protoc 1:418–428.
https://doi.org/10.1038/nprot.2006.62
50. Mejia EM, Hatch GM (2016) Mitochondrial
phospholipids: role in mitochondrial func-
tion. J Bioenerg Biomembr 48:99–112.
https://doi.org/10.1007/s10863-015-
9601-4
51. Gnaiger E (2008) Polarographic oxygen sen-
sors, the oxygraph, and high-resolution respi-
rometry to assess mitochondrial function. In:
Dykens JA, Will Y (eds) Mitochondrial dys-
function in drug-induced toxicity. Wiley,
New York, pp 325–352. https://doi.org/10.
1002/9780470372531.ch12
52. Potter M, Lodge TA, Morten KJ (2018)
{CHAPTER} 8: monitoring of extracellular
and intracellular O2 on a time-resolved fluo-
rescence plate reader. In: Papkovsky DB,
Dmitriev RI (eds) Quenched-
phosphorescence detection of molecular oxy-
gen: applications in life sciences; RSC detec-
tion science series no. 11. The Royal Society
of Chemistry, London, pp 175–192
53. Simonnet H, Vigneron A, Pouysse
´gur J
(2014) Conventional techniques to monitor
mitochondrial oxygen consumption. Meth-
ods Enzymol 542:151–161. https://doi.
org/10.1016/B978-0-12-416618-9.00008-
X
54. Grassian AR, Coloff JL, Brugge JS (2011)
Extracellular matrix regulation of metabolism
and implications for tumorigenesis. Cold
Spring Harb Symp Quant Biol 76:313–324.
https://doi.org/10.1101/sqb.2011.76.
010967
55. Ye
´pez VA, Kremer LS, Iuso A et al (2018)
{OCR}-Stats: robust estimation and statistical
testing of mitochondrial respiration activities
using Seahorse {XF}. Analyzer 13:e0199938.
https://doi.org/10.1371/journal.pone.
0199938
56. Brand MD, Nicholls DG (2011) Assessing
mitochondrial dysfunction in cells. Biochem
J 435:297–312. https://doi.org/10.1042/
BJ20110162
57. Frezza C, Cipolat S, Scorrano L (2007)
Organelle isolation: functional mitochondria
from mouse liver, muscle and cultured filro-
blasts. Nat Protoc 2:287–295. https://doi.
org/10.1038/nprot.2006.478
58. Palmer JW, Tandler B, Hoppel CL (1977)
Biochemical properties of subsarcolemmal
and interfibrillar mitochondria isolated from
rat cardiac muscle. J Biol Chem
252:8731–8739
59. Picard M, Taivassalo T, Ritchie D et al (2009)
Mitochondrial structure and function are dis-
rupted by standard isolation methods. PLoS
One 6:e18317. https://doi.org/10.1371/
journal.pone.0018317
60. Zorzano A, Liesa M, Sebastia
´n D et al (2010)
Mitochondrial fusion proteins: dual regula-
tors of morphology and metabolism. Semin
Cell Dev Biol 21:566–574. https://doi.org/
10.1016/j.semcdb.2010.01.002
61. Milner DJ, Mavroidis M, Weisleder N, Cape-
tanaki Y (2000) Desmin cytoskeleton linked
to muscle mitochondrial distribution and
respiratory function. Semin Cell Dev Biol
150:1283–1298. https://doi.org/10.1083/
jcb.150.6.1283
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 27
62. Wai T, Langer T (2016) Mitochondrial
dynamics and metabolic regulation. Trends
Endocrinol Metab 27:105–117. https://doi.
org/10.1016/j.tem.2015.12.001
63. Piper HM, Sezer O, Schleyer M et al (1985)
Development of ischemia-induced damage in
defined mitochondrial subpopulations. J Mol
Cell Cardiol 17:885–896. https://doi.org/
10.1016/s0022-2828(85)80102-4
64. Affourtit C, Brand MD, Al Amir Dache Z et al
(2018) Understanding mitochondrial com-
plex I assembly in health and disease. Mito-
chondrion 9:818–819. https://doi.org/10.
1042/EBC20170098
65. Gueguen N, Lefaucheur L, Ecolan P et al
(2005) Ca 2+-activated myosin-ATPases, cre-
atine and adenylate kinases regulate mito-
chondrial function according to myofibre
type in rabbit. J Physiol 564:723–735.
https://doi.org/10.1113/jphysiol.2005.
083030
66. Kuznetsov AV, Veksler V, Gellerich FN et al
(2008) Analysis of mitochondrial function in
situ in permeabilized muscle fibers, tissues and
cells. Nat Protoc 3:965–976. https://doi.
org/10.1038/nprot.2008.61
67. Chance B, Williams GR (1955) Respiratory
enzymes in oxidative
phosphorylation. I. Kinetics of oxygen utiliza-
tion. J Biol Chem 217:383–393
68. Nicholls DG, Bernson VS (1977) Inter-
relationships between proton electrochemical
gradient, adenine-nucleotide phosphoryla-
tion potential and respiration, during
substrate-level and oxidative phosphorylation
by mitochondria from brown adipose tissue of
cold-adapted guinea-pigs. Eur J Biochem
75:601–612. https://doi.org/10.1111/j.
1432-1033.1977.tb11560.x
69. Gnaiger E, Kemp RB (1990) Anaerobic
metabolism in aerobic mammalian cells: infor-
mation from the ratio of calorimetric heat flux
and respirometric oxygen flux. Biochim Bio-
phys Acta Bioenerg 1016:328–332. https://
doi.org/10.1016/0005-2728(90)90164-Y
70. Brown GC, Lakin-Thomas PL, Brand MD
(1990) Control of respiration and oxidative
phosphorylation in isolated rat liver cells.
Eur J Biochem 192:355–362. https://doi.
org/10.1111/j.1432-1033.1990.tb19234.x
71. Affourtit C, Brand MD (2009) Chapter 23
measuring mitochondrial bioenergetics in
INS-1E insulinoma cell. Methods Enzymol
457:405–424
72. Kane MS, Paris A, Codron P et al (2018)
Current mechanistic insights into the CCCP-
induced cell survival response. Biochem Phar-
macol 148:100–110. https://doi.org/10.
1016/j.bcp.2017.12.018
73. Stepanova A, Konrad C, Manfredi G et al
(2019) The dependence of brain mitochon-
dria reactive oxygen species production on
oxygen level is linear, except when inhibited
by antimycin A. J Neurochem 148:731–745.
https://doi.org/10.1111/jnc.14654
74. Yadava N, Nicholls DG (2007) Spare respira-
tory capacity rather than oxidative stress reg-
ulates glutamate excitotoxicity after partial
respiratory inhibition of mitochondrial com-
plex I with rotenone. J Neurosci
27:7310–7317. https://doi.org/10.1523/
JNEUROSCI.0212-07.2007
75. Rodrı
´guez-Enrı
´quez S, Jua
´rez O, Rodrı
´guez-
Zavala JS, Moreno-Sa
´nchez R (2001) Multi-
site control of the Crabtree effect in ascites
hepatoma cells. Eur J Biochem
268:2512–2519. https://doi.org/10.1046/
j.1432-1327.2001.02140.x
76. Nicholls DG (2010) Mitochondrial ion cir-
cuits. Essays Biochem 47:25–35. https://
doi.org/10.1042/bse0470025
77. Zhang Q, Padayatti PS, Leung JH (2017)
Proton-translocating nicotinamide nucleotide
transhydrogenase: a structural perspective.
Front Physiol 8:1089. https://doi.org/10.
3389/fphys.2017.01089
78. Poburko D, Demaurex N (2012) Regulation
of the mitochondrial proton gradient by cyto-
solic Ca
2+
signals. Pflugers Arch 464:19–26.
https://doi.org/10.1007/s00424-012-
1106-y
79. Cadenas S (2018) Mitochondrial uncoupling,
ROS generation and cardioprotection. Bio-
chim Biophys Acta Bioenerg 1859:940–950.
https://doi.org/10.1016/j.bbabio.2018.05.
019
80. Mignotte B, Vayssiere JL (1998) Mitochon-
dria and apoptosis. Eur J Biochem 252:1–15.
https://doi.org/10.1046/j.1432-1327.
1998.2520001.x
81. Cottet-Rousselle C, Ronot X, Leverve X,
Mayol J-F (2011) Cytometric assessment of
mitochondria using fluorescent probes.
Cytom Part A 79A:405–425. https://doi.
org/10.1002/cyto.a.21061
82. Perry SW, Norman JP, Barbieri J et al (2011)
Mitochondrial membrane potential probes
and the proton gradient: a practical usage
guide. BioTechniques 50:98–115. https://
doi.org/10.2144/000113610
83. Ward MW, Rego AC, Frenguelli BG, Nicholls
DG (2000) Mitochondrial membrane poten-
tial and glutamate excitotoxicity in cultured
cerebellar granule cells. J Neurosci
20:7208–7219
84. Creed S, McKenzie M (2019) Measurement
of mitochondrial membrane potential with
28 Naig Gueguen et al.
the fluorescent dye tetramethylrhodamine
methyl ester (TMRM). Methods Mol Biol
1928:69–76
85. Rahn CA, Bombick DW, Doolittle DJ (1991)
Assessment of mitochondrial membrane
potential as an indicator of cytotoxicity. Fun-
dam Appl Toxicol 16:435–448. https://doi.
org/10.1016/0272-0590(91)90084-h
86. Moreno AJ, Santos DL, Magalha
˜es-Novais S,
Oliveira PJ (2015) Measuring mitochondrial
membrane potential with a
tetraphenylphosphonium-selective electrode.
Curr Protoc Toxicol 65:25.5.1–25.5.16.
https://doi.org/10.1002/0471140856.
tx2505s65
87. Renner-Sattler K, Fasching M, Gnaiger E
(2016) Mitochondrial Physiology Network
14.05(04):1–13, Technical note. https://
wiki.oroboros.at/index.php/MiPNet14.05_
TPP-mtMembranePotential
88. Trijbels JMF, Sengers RCA, Ruitenbeek W
et al (1988) Disorders of the mitochondrial
respiratory chain: clinical manifestations and
diagnostic approach. Eur J Pediatr
148:92–97. https://doi.org/10.1007/
BF00445910
89. Carrozzo R, Dionisi-Vici C, Steuerwald U
et al (2007) {SUCLA}2 mutations are asso-
ciated with mild methylmalonic aciduria,
Leigh-like encephalomyopathy, dystonia and
deafness. Brain 130:862–874. https://doi.
org/10.1093/brain/awl389
90. Chao De La Barca JM, Mirebeau-Prunier D,
Moal V et al (2015) Metabolome and mass
spectrometry: new biomedical analysis per-
spectives. Ann Biol Clin (Paris) 73:126–130.
https://doi.org/10.1684/abc.2014.1020
91. Esterhuizen K, van der Westhuizen FH, Louw
R (2017) Metabolomics of mitochondrial dis-
ease. Mitochondrion 35:97–110. https://doi.
org/10.1016/j.mito.2017.05.012
92. Kouassi Nzoughet J, Chao de la Barca JM,
Guehlouz K et al (2019) Nicotinamide defi-
ciency in primary open-angle glaucoma.
Investig Opthalmol Vis Sci 60:2509.
https://doi.org/10.1167/iovs.19-27099
93. Nikkanen J, Forsstro
¨m S, Euro L et al (2016)
Mitochondrial {DNA} replication defects dis-
turb cellular {dNTP} pools and remodel
one-carbon metabolism. Cell Metab
23:635–648. https://doi.org/10.1016/j.
cmet.2016.01.019
94. Khan NA, Nikkanen J, Yatsuga S et al (2017)
mTORC1 regulates mitochondrial integrated
stress response and mitochondrial myopathy
progression. Cell Metab 26:419–428.e5.
https://doi.org/10.1016/j.cmet.2017.07.
007
95. Bocca C, Kane MS, Veyrat-Durebex C et al
(2018) The metabolomic bioenergetic signa-
ture of opa1-disrupted mouse embryonic
fibroblasts highlights aspartate deficiency. Sci
Rep 8:11528. https://doi.org/10.1038/
s41598-018-29972-9
96. Rahman J, Rahman S (2018) Mitochondrial
medicine in the omics era. Lancet
391:2560–2574. https://doi.org/10.1016/
S0140-6736(18)30727-X
97. Tripodi F, Castoldi A, Nicastro R et al (2018)
Methionine supplementation stimulates mito-
chondrial respiration. Biochim Biophys Acta,
Mol Cell Res 1865:1901–1913. https://doi.
org/10.1016/j.bbamcr.2018.09.007
98. Andre
´asson C, Ott M, Bu¨ttner S (2019)
Mitochondria orchestrate proteostatic and
metabolic stress responses. EMBO Rep 20:
e47865. https://doi.org/10.15252/embr.
201947865
99. Yi H-S, Chang JY, Shong M (2018) The
mitochondrial unfolded protein response and
mitohormesis: a perspective on metabolic dis-
eases. J Mol Endocrinol 61:R91–R105.
https://doi.org/10.1530/JME-18-0005
100. Mick E, Titov DV, Skinner OS et al (2020)
Distinct mitochondrial defects trigger the
integrated stress response depending on the
metabolic state of the cell. elife 9:e49178.
https://doi.org/10.7554/eLife.49178
101. Miliotis S, Nicolalde B, Ortega M et al (2019)
Forms of extracellular mitochondria and their
impact in health. Mitochondrion 48:16–30.
https://doi.org/10.1016/j.mito.2019.02.
002
102. Al Amir Dache Z, Otandault A, Tanos R et al
(2020) Blood contains circulating cell-free
respiratory competent mitochondria. FASEB
J 34:3616–3630. https://doi.org/10.1096/
fj.201901917RR
103. Hayakawa K, Chan SJ, Mandeville ET et al
(2018) Protective effects of endothelial pro-
genitor cell-derived extracellular mitochon-
dria in brain endothelium. Stem Cells
36:1404–1410. https://doi.org/10.1002/
stem.2856
104. Caicedo A, Aponte PM, Cabrera F et al
(2017) Artificial mitochondria transfer: cur-
rent challenges, advances, and future applica-
tions. Stem Cells Int 2017:7610414. https://
doi.org/10.1155/2017/7610414
Mitochondrial Dysfunction in Mitochondrial Medicine: Current Limitations... 29
Chapter 2
Preparation of “Functional” Mitochondria: A Challenging
Business
Stefan Lehr, Sonja Hartwig, and Jorg Kotzka
Abstract
As the powerhouse of the cell, mitochondria, plays a crucial role in many aspects of life, whereby mitochon-
drial dysfunctions are associated with pathogenesis of many diseases, like neurodegenerative diseases,
obesity, cancer, and metabolic as well as cardiovascular disorders. Mitochondria analysis frequently starts
with isolation and enrichment procedures, which have become increasingly important in biomedical
research. Unfortunately, isolation procedures can easily cause changes in the structural integrity of mito-
chondria during in vitro handling having impact on their function. This carries the risk that conclusions
about isolated mitochondria may be drawn on the basis of experimental artifacts. Here we critically review a
commonly used isolation procedure for mitochondria utilizing differential (gradient) centrifugation and
depict major challenges to achieve “functional” mitochondria as basis for comprehensive physiological
studies.
Key words Isolation of mitochondria, Differential gradient centrifugation, Mitochondrial integrity
1 Introduction
Since their naming in 1898 by Carl Benda, the importance of
mitochondria mediating several fundamental cellular processes is
constantly growing. Due to the fact that mitochondria dysfunctions
are involved in the pathophysiology of a wide variety of human
diseases [1–3], dissecting mitochondria physiology is a main focus
of recent biomedical sciences. Approximately, one in 200 indivi-
duals bears a pathogenic mutation in mitochondrial genes [4],
affecting mitochondrial biogenesis and inheritance and therefore
containing the risk to develop severe diseases including neurode-
generative diseases, aging, obesity, cancer and metabolic as well as
cardiovascular disorders [5–9]. This underlines the importance to
characterize exact composition and function of mitochondria in
order to understand their role in cellular metabolism more pre-
cisely. Although, there have been made substantial improvements
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_2,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
31
to investigating mitochondria in living cells including Oroboros
O2k [10], Seahorse XF [11], and SNAP-Tag [12] technology,
reflecting the natural cellular and physiological context, isolation
of mitochondria is still required for certain issues. In this context, a
major task of all these associated studies is to preserve the pristine
mitochondrial structural integrity for analysis, which is indispens-
able linked to functionality of the organelles. Due to the complex
organelle morphology enclosed by two membranes, i.e., the outer
membrane with a large number of specialized proteins (porins)
enabling pass of molecules less than 5000 Da and the inner mem-
brane exhibiting a folding (cristae) to increase the surface area
containing the complex respiratory transport chain, being critical
for ATP production and last but not least the matrix harboring the
wide variety of enzymes for metabolic pathways, e.g., citric acid
cycle, mitochondrial ribosomes, tRNAs, and mitochondrial DNA,
the preparation of mitochondria approximating the in vivo situa-
tion is tremendously challenging.
2 Isolated Mitochondria: Addressing Composition and Function
In order to dissect mitochondrial composition as well as function in
detail and to allow direct manipulation of mitochondria by expo-
sure to specific substrates and inhibitors, the major part of con-
ducted studies utilizing isolated mitochondria [13,14] achieved
from diverse cellular and tissue sources. These isolation procedures
are most frequently based on fundamental work of George Pallades
group [15], done more than 60 years ago. They introduced a
differential centrifugation workflow enabling separation of almost
pure organelles with high yield, which have paved the way for such
revolutionary discoveries like the oxidative phosphorylation mech-
anism [16], discovery of mitochondrial DNA [17], or the descrip-
tion of the mitochondrial ultrastructure [18]. Although, today
diverse adapted methods are available to face almost all kinds of
sample sources and scientific questions [19–26], preparation of
functional mitochondria reflecting approximately the in vivo situa-
tion is still challenging.
3 First Step: The Inevitable
It is supposed that mitochondria in vivo develop complex tubular-
branched structures [27,28] undergoing complex remodeling of
their morphology (fusion and fission) in respond to cellular cues
[29], which are significantly different from the relatively homoge-
neous circular organelles occurring during standard isolation tech-
niques [14]. It has been commonly assumed that isolation of
mitochondria inevitably comes along with disruption of the native
32 Stefan Lehr et al.
mitochondrial morphology. Thereby, the mitochondria network is
disrupted and sealed again, which probably leads to a partial loss of
soluble mitochondrial proteins [30]. Up to now, the functional
consequences are largely unknown. In this context, studies com-
paring functionality of isolated mitochondria with mitochondria
within permeabilized cells, leaving the organelles in their native
surrounding [31], indicate impairments of mitochondrial function,
e.g., regarding mitochondrial respiration [30,32]. Accordingly, it
should be carefully considered that during data interpretation, the
common assumption isolated mitochondria preserve their com-
plete functionality and composition will not be valid in any case.
4 Isolating “Intact” Mitochondria: A Challenging Business
Despite the known limitations, proper isolated mitochondria keep
their compartment properties and provide a powerful tool for
in-depth analysis, especially, when comparing samples achieved
under identical conditions [14]. Accordingly, many scientific fields
comprising metabolite and protein transport up to dynamic remo-
deling of mitochondria as well as recent biomedical concerns
benefit from isolated, almost pure mitochondria.
The most frequently applied method for isolation is differential
centrifugation [13], whereby in a first step cell or tissue samples are
carefully homogenized in an appropriate buffer preventing damag-
ing of the organelles by mechanical forces, chemical reactions, or
osmosis. In order to separate components of different size and
density, e.g., cellular organelles, the homogenate is subjected to
repeated centrifugation consecutively increasing sedimentation
forces, enabling a rough fractionation of the cellular environment.
To achieve pure organelles, i.e., mitochondria, the last purification
step is an equilibrium density gradient centrifugation. Samples are
centrifuged at high g-forces in a buffer gradient, e.g., sucrose
gradient, focusing the target organelles in a concentration range
of comparable density (isopycnic point), resulting in high-purity
mitochondria.
Although most studies are carried out based on this general
isolation strategy, different sample sources and special scientific
requirements need specific adaptions of the used protocols
[14,24–26], in order to achieve optimal results regarding purity
as well as functionality. In the literature, a confusing diversity
regarding utilized sedimentation forces, buffer compositions for
homogenization, and gradient compositions are available, which
should be carefully reviewed before use. In the end, the choice of a
suitable separation protocol depends on the researcher’s require-
ments regarding organelle purity, yield, activity, and structural
integrity. In some cases, the most important factor is purity,
whereby activity, yield, and preparation time may be less important.
Preparation of “Functional” Mitochondria: A Challenging Business 33
In studies exploring cellular compartment metabolism, high activ-
ity of the organelles is often the most important requirement. For
high-throughput experiments in comprehensive studies, where
many samples are to be compared, it is important to shorten the
time for sample preparation.
From our point of view, the differential gradient centrifugation
approach, closely monitored by appropriate quality control meth-
ods, offers the most valuable compromise between applied efforts
(man power, expenses) and achievable yield, purity, activity, and
functional integrity. In order to achieve valid results, isolation of
“functional” mitochondria based on this methodology has some
basic requirements and should follow a general workflow (Fig. 1,
upper panel), illustrated and described in Chapter 2. This basic
protocol provides a useful starting point for the isolation of mito-
chondria by differential gradient centrifugation. It describes prepa-
ration in detail and guide through critical steps of the separation
method and how to control them regarding yield and organelle
functionality. In laboratory routine, frequently separation protocols
are utilized, which pass the last centrifugation step through a
density gradient, which requires an ultracentrifuge and some expe-
rience. Fractionation solely by different sedimentation forces is also
applied in commercial available isolation kits, but anyway results in
a decreased enrichment performance [14].
In addition to the discussed differential (gradient) centrifuga-
tion strategy, it is to mention that isolation of mitochondria using
integrated zone electrophoresis on a free-flow electrophoretic
device [14,33] represents a relevant alternative for preparing high
purity organelles, which are particular appropriate for comprehen-
sive proteomic studies. The major drawback of this method is that a
special instrument, i.e., a free-flow apparatus, is necessary and that
processing is time consuming and needs special expertise. Alterna-
tively, a combination of simple differential centrifugation with a
final purification of mitochondria utilizing anti-TOM22 magnetic
beads can be also used. This reproducible protocol has been
described to be suitable for various tissues yielding in the isolation
of highly pure mitochondria avoiding non-mitochondrial contam-
inations [34]. Both methods provide pure mitochondria fractions
with structural integrity, but coming along with much higher costs
than applied for differential gradient centrifugation methods.
5 Reproducibility and Quality Control: Guarantee for Successful Analysis
Many recent studies in the field of biomedical research addressing
potential mitochondrial dysfunction requires analysis properties in
a highly parallel fashion. In order to achieve valid results, therefore
it is crucial to utilize a standardized processing. In this context, it
would be advantageous to collect samples successively and store
34 Stefan Lehr et al.
Fig. 1 Impact of sample freezing on mitochondria structural integrity. Centrifugation-based isolation of
mitochondria is performed according to a general strategy consisting out of three major processing steps
shown in the upper panel. 1. Careful sample homogenization using a Potter, Douncer, or Ultra-Turrax device.
2. Consecutive differential centrifugation to separate cellular compartments according to size and density.
3. Increasing organelle purity due to equilibrium density gradient centrifugation, e.g., linear sucrose gradient,
which enable concentration of mitochondria in the gradient fraction of comparable density (isopycnic point).
To illustrate the importance of choosing appropriate sample material, the lower panels show a comparison of
isolated mitochondria from fresh and frozen material. Measurement of citrate synthase (CS) activity, an
enzyme in situ exclusively located in the mitochondrial matrix, allows monitoring structural integrity of
mitochondria during processing. CS release and therefore significant CS activity in the homogenate
Preparation of “Functional” Mitochondria: A Challenging Business 35
them in a freezer before use. Accordingly, one might think that
starting with frozen material for mitochondria isolation would
simplify the workflow significantly. But unfortunately, the most
striking prerequisite to isolate functional mitochondria is to use
fresh (not frozen) material. Investigating the impact of sample
freezing indicates dramatic consequences on mitochondria struc-
ture and functionality (Fig. 1middle and lower panel). Monitoring
activity of citrate synthase (CS) [35], an enzyme normally exclu-
sively located in the mitochondrial matrix, reveals that freezing
leads to a strong release of CS from the mitochondrial matrix into
the homogenate. In contrast to fresh material, where approximately
5% of total CS activity is found in the homogenate, more than 40%
of CS activity can be assigned to the homogenate if frozen material
is used. This eightfold increase indicates that the mitochondria
structure is significantly disrupted. Corresponding electron micros-
copy (EM) images confirm these observations and demonstrate the
loss of structural integrity due to a nearly complete destruction of
isolated mitochondria expected morphology (Fig. 1, lower panel).
Accordingly, these observations strongly suggest that frozen mate-
rial in any case is inappropriate for organelle isolation and therefore
to study mitochondria composition, function, or physiological
behavior. Another challenging point to enable reliable comparison
of a huge number of samples is the initial homogenization step. In
most laboratories, it is performed manually using Potter, Douncer,
or ULTRA TURRAX homogenizers, potentially introducing sig-
nificant variations. The utilized forces for homogenization are a
highly subjective parameter, which carefully have to be validated.
Due to the fact that organelle purity and functionality are the
striking prerequisites for any study addressing isolated mitochon-
dria, monitoring the whole isolation process is mandatory. Without
an appropriate quality control, regarding mitochondria functional-
ity and purity, no reasonable assessment of the organelle sample is
possible. In this context, it is to mention that Western blot analysis
of organelle specific proteins (e.g., mitochondria (anti-Tom20),
lysosomes (anti-Lamp-1), endoplasmic reticulum (anti-BiP/
GRP78), and peroxisomes (anti-catalase)), which is frequently
used as a standard method to examine product composition and
therefore the isolation success, only allow to detect relative
ä
Fig. 1 (continued) (extramitochondrial activity) suggest disruption of mitochondria structure. To assign the
total CS activity as control, CS activity of complete lysed mitochondria samples was set as 100%. Comparing
activity levels of CS, in the homogenate of fresh and frozen samples impressively, shows that initial freezing
destroys mitochondria structure and induce a release of CS from the mitochondrial matrix. This results in an
eightfold increase in CS activity in the homogenate. Accordingly, in case of frozen sample, nearly half of the
total CS activity is found outside of the mitochondria. In addition to that electron microscopy (EM) analysis of
gradient fractions shows a more or less complete loss of the typical mitochondrial cristae structure resulting in
blank membrane covers
36 Stefan Lehr et al.
distribution of the dedicated proteins but do not allow to assign
mitochondria functionality. A more meaningful evaluation, allow-
ing to determine mitochondria purity, can be achieved when addi-
tional biochemical assays are used. We recommend measuring
activity of marker enzymes specific for mitochondria as well as
contaminating cell compartments, i.e., succinate dehydrogenase
for mitochondria, glucose-6-phosphatase for endoplasmatic reticu-
lum [36], alkaline phosphatase for plasma membrane, acidic phos-
phatase for lysosomes [37], and catalase for peroxisomes
[38]. These control assays are easily applicable with standard labo-
ratory equipment and enable to calculate organelle distribution and
to estimate some functional aspects. These assays are also very
helpful during the establishing phase of the gradient centrifugation
protocols, in order to select the region within the gradient
corresponding to most pure and active mitochondria fraction. In
order to assess “functional” integrity, measuring JC-1 uptake [39]
or oxygen consumption, with the principle of a traditional Clark
electrode [40] as read out for ATP production, should be manda-
tory. Both methods are appropriate to monitor the status of mito-
chondrial membrane potential. If possible, additional investigations
of the isolated mitochondria by transmission electron microscopy
allow assessing structural integrity of the inner and outer mem-
branes as well as the mitochondrial matrix.
It is worth mentioning that not using such research tools
deprives the possibility of drawing well-founded conclusions
about the integrity of organelles and thus the functional status of
the underlying preparation. Accordingly, investigation of purity
and morphology–function relationship should be an inherent part
of any organelle fractionation procedures in order to avoid working
with inappropriate sample material.
6 Conclusion
Our recent knowledge suggest that available protocols fail to allow
isolation of native, functional mitochondria. This has to be consid-
ered, when planning and interpreting the experiments. Emerging
understanding of structure–function relationship and the effect of
morphology changes during isolation may help to improve isola-
tion methods or develop novel strategies for in-depth in situ analy-
sis. Nevertheless, proper isolated mitochondria may approximate
the “intact” status and still provide an indispensable tool to address
future challenges of mitochondrial participation in the pathophysi-
ology of diverse widespread diseases including neurodegenerative,
muscular, cardiovascular, metabolic disorders, and cancer.
Preparation of “Functional” Mitochondria: A Challenging Business 37
References
1. Picard M, Wallace DC, Burelle Y (2016) The
rise of mitochondria in medicine. Mitochon-
drion 30:105–116
2. Herst PM, Rowe MR, Carson GM, Berridge
MV (2017) Functional mitochondria in health
and disease. Front Endocrinol 8:296
3. Elliott HR, Samuels DC, Eden JA, Relton CL,
Chinnery PF (2008) Pathogenic mitochondrial
DNA mutations are common in the general
population. Am J Hum Genet 83
(1030):254–260
4. Schon EA, DiMauro S, Hirano M (2012)
Human mitochondrial DNA: roles of inherited
and somatic mutations. Nat Rev Genet
13:878–890
5. Cherry C, Thompson B, Saptarshi N, Wu J,
Hoh J (2016) A ‘mitochondria’ odyssey.
Trends Mol Med 22:391–403
6. Wallace DC (2018) Mitochondrial genetic
medicine. Nat Genet 50:1642–1649
7. Dai W, Jiang L (2019) Dysregulated mito-
chondrial dynamics and metabolism in obesity,
diabetes, and cancer. Front Endocrinol (Lau-
sanne) 10:570
8. Chan DC (2019) Mitochondrial dynamics and
its involvement in disease. Annu Rev Pathol
15:235–259
9. Piaceri I, Rinnoci V, Bagnoli S, Failli Y, Sorbi S
(2012) Mitochondria and Alzheimer’s disease.
J Neurol Sci 322:31–34
10. Gnaiger E, Steinlechner-Maran R, Me
´ndez G,
Eberl T, Margreiter R (1995) Control of mito-
chondrial and cellular respiration by oxygen. J
Bioenerg Biomembr 27:583–596
11. Divakaruni AS, Rogers GW, Murphy AN
(2014) Measuring mitochondrial function in
permeabilized cells using the Seahorse XF ana-
lyzer or a Clark-type oxygen electrode. Curr
Protoc Toxicol 60:1–16
12. Stephan T, Roesch A, Riedel D, Jakobs S
(2019) Live-cell STED nanoscopy of mito-
chondrial cristae. Sci Rep 9:12419
13. Frezza C, Cipolat S, Scorrano L (2007) Organ-
elle isolation: functional mitochondria from
mouse liver, muscle and cultured fibroblasts.
Nat Protoc 2:287–295
14. Hartwig S, Feckler C, Lehr S, Wallbrecht K,
Wolgast H, Mu¨ller-Wieland D, Kotzka J
(2009) A critical comparison between two clas-
sical and a kit-based method for mitochondria
isolation. J Proteome 11:3209–3214
15. Hogeboom GH, Schneider WC, Pallade GE
(1948) Cytochemical studies of mammalian
tissues. I. Isolation of intact mitochondria
from rat liver; some biochemical properties of
mitochondria and submicroscopic particulate
material. J Biol Chem 172:619–635
16. Mitchell P, Moyle J (1967) Chemiosmotic
hypothesis of oxidative phosphorylation.
Nature 213:137–139
17. Nass MM, Nass S (1963) Intramitochondrial
fibers with DNA characteristics. I. Fixation and
electron staining reactions. J Cell Biol
19:593–611
18. Palade GE (1952) The fine structure of mito-
chondria. Anat Rec 114:427–451
19. Graham JM (2001) Purification of a crude
mitochondrial fraction by density-gradient
centrifugation. Curr Protoc Cell Biol
Chapter 3:Unit 3.4
20. Gollihue JL, Patel SP, Mashburn C, Eldahan
KC, Sullivan PG, Rabchevsky AG (2017) Opti-
mization of mitochondrial isolation techniques
for intraspinal transplantation procedures. J
Neurosci Methods 287:1–12
21. Le
´ger JL, Jougleux JL, Savadogo F, Pichaud N,
Boudreau LH (2019) Rapid isolation and puri-
fication of functional platelet mitochondria
using a discontinuous Percoll gradient. Plate-
lets 5:1–7
22. Knebel B, Go
¨ddeke S, Hartwig S, Ho
¨rbelt T,
Fahlbusch P, Al-Hasani H, Jacob S,
Koellmer C, Nitzgen U, Schiller M, Lehr S,
Kotzka J (2018) Alteration of liver peroxisomal
and mitochondrial functionality in the NZO
mouse model of metabolic syndrome. Proteo-
mics Clin Appl 12(1)
23. Liu Y, He J, Ji S, Wang Q et al (2008) A
comparative study of early liver dysfunction in
senescence-accelerated mouse using mitochon-
drial proteomics approaches. Mol Cell Proteo-
mics 7:1737–1747
24. Stahl WL, Smith JC, Napolitano LM, Basford
RE (1963) Brain mitochondria: I. Isolation of
bovine brain mitochondria. J Cell Biol
19:293–307
25. Cannon B, Lindberg O (1979) Mitochondria
from brown adipose tissue: isolation and prop-
erties. Methods Enzymol 55:65–78
26. Mela L, Seitz S (1997) Isolation of mitochon-
dria with emphasis on heart mitochondria from
small amounts of tissue. Methods Enzymol
55:39–46
27. Ogata T, Yamasaki Y (1997) Ultra-high-reso-
lution scanning electron microscopy of mito-
chondria and sarcoplasmic reticulum
arrangement in human red, white, and inter-
mediate muscle fibres. Anat Rec 248:214–223
38 Stefan Lehr et al.
28. Fang H, Chen M, Ding Y, Shang W, Xu J et al
(2011) Imaging superoxide flash and
metabolism-coupled mitochondrial permeabil-
ity transition in living animals. Cell Res
21:1295–1304
29. Pernas L, Scorrano L (2016) Mito-morphosis:
mitochondrial fusion, fission, and cristae remo-
deling as key mediators of cellular function.
Annu Rev Physiol 78:505–531
30. Picard M, Taivassalo T, Ritchie D, Wright KJ,
Thomas MT et al (2011) Mitochondrial struc-
ture and function are disrupted by standard
isolation methods. PLoS One 6:e18317
31. Kuznetsov AV, Veksler V, Gellerich FN, Saks V,
Margreiter R, Kunz WS (2008) Analysis of
mitochondrial function in situ in permeabilized
muscle fibres, tissues and cells. Nat Protoc
3:965–976
32. Saks V, Guzun R, Timohhina N, Tepp K, Var-
ikmaa M et al (2010) Structure-function rela-
tionships in feedback regulation of energy
fluxes in vivo in health and disease: mitochon-
drial interactosome. Biochim Biophys Acta
1797:678–697
33. Islinger M, Wildgruber R, Vo
¨lkl A (2018) Pre-
parative free-flow electrophoresis, a versatile
technology complementing gradient
centrifugation in the isolation of highly purified
cell organelles. Electrophoresis 39:2288–2299
34. Franko A, Baris OR, Bergschneider E, von
Toerne C, Hauck SM et al (2013) Efficient
isolation of pure and functional mitochondria
from mouse tissues using automated tissue dis-
ruption and enrichment with anti-TOM22
magnetic beads. PLoS One 8:e82392
35. Schulman JD, Blass JP (1971) Measurement of
citrate synthase activity in human fibroblasts.
Clin Chim Acta 33:467–469
36. Pennington RJ (1961) Biochemistry of dystro-
phic muscle. Mitochondrial succinate-
tetrazolium reductase and adenosine tripho-
sphatase. Biochem J 80:649–654
37. Bergmeyer HU (1974) Methoden der enzyma-
tischen analyse. Verlag Chemie, Weinheim
38. Aebi H (1984) Catalase in vitro. Methods
Enzymol 105:121–126
39. Reers M, Smiley ST, Mottola-Hartshorn C,
Chen A, Lin M, Chen LB (1995) Mitochon-
drial membrane potential monitored by JC-1
dye. Methods Enzymol 260:406–417
40. Li Z, Graham BH (2012) Measurement of
mitochondrial oxygen consumption using a
Clark electrode. Methods Mol Biol 837:63–72
Preparation of “Functional” Mitochondria: A Challenging Business 39
Chapter 3
Isolation and Quality Control of Functional Mitochondria
Sonja Hartwig, Jorg Kotzka, and Stefan Lehr
Abstract
Even in times, when the study of mitochondria in their natural cellular context is becoming more and more
popular, some scientific questions still require the preparation of isolated mitochondria. Numerous proto-
cols are available being adapted for different cell or tissue types allowing isolation of “pure” mitochondria
trying to preserve their “structural and functional” integrity. In this chapter, we intend to provide a more
general framework introducing differential isopycnic density gradient centrifugation strategy with a special
focus sensitizing for the specific challenges coming along with this method and how to obtain “functional,”
enriched, “intact” mitochondria. Due to the fact that in any study dealing with these organelles standar-
dized processing is mandatory, here we describe a strategy addressing quality control of prepared intact
mitochondria. The quality control should be an integrated part of all isolation processes. The underlying
protocol should be seen as starting point and has to be carefully adjusted to cover different sample types
used for the diverse research questions.
Key words Sample pre-fractionation, Mitochondria enrichment, Isopycnic density gradient centrifu-
gation, Marker enzymes
1 Introduction
Mitochondria biology play a crucial role in many aspects of life,
including in pathophysiology of many diseases, like neurodegener-
ative diseases, obesity, cancer, and metabolic disorders [1,2]. In
order to dissect the specific role of mitochondria, various scientific
fields benefit from analysis of isolated, almost pure mitochondria.
Due to the fact that there is a strong relationship between mito-
chondria structure and functionality, there are strong demands to
preserve their “structural and functional” integrity during prepara-
tion. Today, numerous protocols are available enabling isolation of
“pure” mitochondria, including differential gradient centrifugation
[3,4], affinity purification with Anti-TOM22 magnetic beads [5],
or separation by free-flow electrophoresis [6]. In this context, the
differential gradient centrifugation strategy offers the most valuable
compromise between applied efforts (man power, expenses) and
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_3,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
41
achievable yield, purity, activity, and functional integrity. In order to
achieve high-quality mitochondria appropriate for further detailed
investigations moreover, it is mandatory to monitor functionality
and enrichment progress during the entire isolation process. If
available, controlling resulting preparations by electron micros-
copy, which gives an excellent impression of achieved quality, is
recommended. Here we describe the preparation of mitochondria
in detail and guide through critical steps of the separation method
and how to control them regarding yield and organelle
functionality.
2 Materials
2.1 Mitochondria
Fractionation
1. Potter S homogenizer with glass cylinders (15 ml) including
appropriate plunger (for breaking up the tissue in a gentle way).
2. Ground in glass douncer (loose fit, for gently manual
homogenization).
3. Centrifuges, corresponding rotors and tubes with capability for
11,000 g-force and 85,000 g-force, e.g., 70TI rotor and
swing-out rotor SW28 for the Optima XPN-80 (Beckmann).
4. Homogenization buffer: 225 mM mannitol, 75 mM sacchar-
ose, 10 mM Tris/HCl, 0.5 mM EGTA, pH 7.4, and 0.5 mM
DTT (see Note 1).
5. Resuspension buffer: 250 mM saccharose, 10 mM Tris/HCl,
0.5 mM EGTA, pH 7.4 and 0.5 mM DTT (see Note 1).
2.2 Linear
Saccharose Gradient
1. Gradient mixer.
2. Ultracentrifuge tubes (e.g., Beckmann tubes for SW28 rotor).
3. Three different saccharose solutions in resuspension buffer:
24% (w/w), 54% (w/w), and 57% (w/w) (see Note 2).
2.3 Marker Enzyme
Assays
1.5 and 2 ml reaction tubes with corresponding stand, incubator
for 37C, the usage of a Multipette would be a benefit, and a
photometer with corresponding solvent-resistant cuvette to mea-
sure at 405, 410, 490, and 815 nm are needed. For JC-1 assay a
spectrofluorometer with an excitation wavelength of 490 nm and
an emission wavelength of 590 nm is needed.
2.3.1 Succinate
Dehydrogenase
(SDH) Assay
1. INT solution:
2.5 mg/ml p-Iodonitrotetrazolium in 0.05 M Na dihydro-
gen phosphate pH 7.5.
2. Na succinate solution:
0.01 M Na succinate in 0.05 M Na dihydrogen phosphate
pH 7.5.
42 Sonja Hartwig et al.
3. Stop solution:
Ethyl acetate:ethanol:TCA 5:5:1 (v/v/w).
2.3.2 Acidic
Phosphatase Assay
1. Nitrophenylphosphate solution:
16 mM p-Nitrophenylphosphate in H
2
O.
2. Na acetate solution:
180 mM Na acetate pH 5.0 (adjustable with acetic acid).
3. Stop solution:
250 mM NaOH.
2.3.3 Basic
Phosphatase Assay
1. Nitrophenylphosphate solution:
16 mM p-Nitrophenylphosphate in H
2
O.
2. Na borate solution:
250 mM Na borate pH 9.8 (adjustable with NaOH).
3. Stop solution:
250 mM NaOH.
4. 1 M MgCl
2
.
2.3.4 Catalase Assay 1. Catalase sample buffer:
20 mM Tris/HCl pH 7.0, 1% BSA, 2% Triton X-100.
2. Assay buffer:
20 mM Tris/HCl pH 7.0, 1% BSA, 0.25% H
2
O
2
.
3. Titanyl solution:
Dissolve 22.5 mg titanium oxy sulfate sulfuric acid hydrate
in 100 ml 1 M sulfuric acid (see Note 3).
2.3.5
Glucose-6-Phospatase
(G-6-Pase) Assay
1. Saccharose buffer:
250 mM saccharose, 100 mM EDTA pH 7.2.
2. Cacodylate buffer:
100 mM cacodylic acid sodium salt adjusted to pH 6.5.
3. Glucose-6-phospate solution:
100 mM glucose-6-phospate (see Note 4).
4. Potassium dihydrogen phosphate solution:
200 μM potassium dihydrogen phosphate in H
2
O.
5. TCA solution:
8% trichloroacetic acid in H
2
O.
6. Fiske-Subbarow reagent:
Dissolve 0.75 g sodium sulfite in 5 ml H
2
O. Also dissolve
6.85 g sodium disulfite and 0.125 g Amino-2-hydroxy-
naphthalin-4 sulfonic acid in 50 ml H
2
O. Mix both solutions
and store it in a tightly capped amber bottle (see Note 5).
7. Ammonium heptamolybdate solution:
0.48% ammonium heptamolybdate in H
2
O (w/v).
Isolation and Quality Control of Functional Mitochondria 43
2.3.6 JC-1 Uptake Assay 1. 5x storage buffer:
50 mM HEPES, 1.25 M saccharose, 5 mM ATP, 0.4 mM
ADP, 25 mM sodium succinate, 10 mM K
2
HPO
4
, 5 mM DTT.
Weigh out all ingredients and dissolve it in H
2
O to 90% of
calculated end volume (e.g., for 100 ml add 90 ml). Adjust pH
to 7.5 with concentrated NaOH and fill up with H
2
O to 100%
of calculated volume. Pass the buffer through a 0.2 μm filter
and store 5–15 ml aliquots at 20 C.
2. 5JC-1 assay buffer:
100 mM MOPS, 550 mM KCl, 50 mM ATP, 50 mM
MgCl
2
, 50 mM sodium succinate, 5 mM EGTA. Weigh out
all ingredients and dissolve it in H
2
O 90% of calculated end
volume (e.g., for 50 ml add 45 ml). Adjust pH to 7.5 with
concentrated NaOH and fill up with H
2
O to 100% of calcu-
lated volume. Pass the buffer trough a 0.2 μm filter and store
2 to 5 ml aliquots at 20 C.
3. JC-1 stain:
Dissolve 25 μg JC-1 (5,50,6,60-tetrachloro-
1,10,3,30tetraethylbenzimidazol carbocyanine iodide) in 25 μl
DMSO (see Note 6) to obtain a solution with 1 μg/μl (result-
ing concentration is 1.53 mM).
3 Methods
Here we describe a protocol for reproducible mitochondria isola-
tion from mouse liver with isopycnic saccharose gradient centrifu-
gation. The general workflow and some typical electron microscopy
images from such a preparation are shown in Fig. 1. Table 1pro-
vides an example result for enzymatic quality check.
3.1 Subcellular
Fractionation
3.1.1 Preparation
of Linear Saccharose
Gradient
1. Pipette 4 ml 57% saccharose solution in a 36 ml ultracentrifuge
tube (for SW28 Beckmann rotor) as a pillow and place it in an
angular tube stand.
2. Build up the gradient mixer, assure that you can stir the solu-
tion in the first chamber and put a blunt cannula at the end of
the hose.
3. Fill 15 ml low 24% saccharose solution in the first chamber and
15 ml 54% saccharose solution in the second.
4. Position the blunt cannula in the angular positioned tube tight
above the 57% pillow and fix it with a clip.
5. Open the taps and let the saccharose solutions flow by gravity
into the tube (see Note 7).
6. Pull the blunt cannula carefully out, right the tube, and store it
at 4 C until use.
44 Sonja Hartwig et al.
3.1.2 Homogenization 1. Use a fresh liver tissue sample (obtained within 1 h of sacrifice)
kept on ice in homogenization buffer.
2. Wash mice liver twice with 2 volumes of homogenization
buffer and mince it into pieces with scissors.
3. Transfer liver pieces (in total about 1.5 g) to the 15 ml glass
cylinder and add tenfold (w/v) homogenization buffer.
4. Homogenize the liver pieces by 10 strokes with the ground in
glass douncer (see Note 8).
5. Take a 10% aliquot of the homogenate and store on ice (see
Note 9).
3.1.3 Differential
Centrifugation
1. Transfer homogenate to a 15 ml centrifugation tube and cen-
trifuge at 666 gat 4 C for 15 min to remove cell debris.
2. After centrifugation, transfer the supernatant to a 70TI centri-
fugation tube and fill up to 30 ml with homogenization buffer
and centrifuge at 11,000 g4C for 15 min.
3. Decant the supernatant and put it aside (mainly cytosolic
proteins).
Fig. 1 Mitochondria isolation workflow shows the consecutive steps from homogenization, differential
centrifugation up to the density gradient. Purification progress among the different processing steps is
visualized by electron microscopy
Isolation and Quality Control of Functional Mitochondria 45
Table 1
Example of monitoring the subcellular fractionation process
Mitochondria
Succinate dehydrogenase JC-1 uptake
Preperation step Total protein [mg] Total activity [μmol/min] Yield [%] Specific activity [μmol mg
1
min
1
] Accumulation factor Specific activity [Fluor./mg] Accumulation factor
Homogenate 243 74 12,649 5076 100 52.3 15.8 1 372 73 1
666 gsupernatant 152 56 5649 5692 39 33.9 31.3 0.7 308 79 0.8
11,000 gpellet 16 2 2113 836 17 134.2 52.8 2.6 601 91 1.7
Gradient fraction 5.3 1.6 865 420 7 164.1 51.3 3.2 687 97 1.9
Lysosomes Plasma membrane Endoplasmatic reticulum Peroxisomes
Acidic phosphatase Basic phosphatase Glucose-6-phosphatase Catalase
Preperation
step
Total activity
[μmol/min]
Yield
[%]
Specific activity
[μmol mg
1
min
1
]
Total activity
[μmol/min]
Yield
[%]
Specific activity
[μmol mg
1
min
1
]
Total activity
[μmol/min]
Yield
[%]
Specific activity
[μmol mg
1
min
1
]
Total activity
[μU/min]
Yield
[%]
Specific activity
[μUmg
1
min
1
]
Homogenate 464 161 100 1.96 0.6 11.95 3.4 100 0.05 0.01 1405 548 100 6.0 2.0 6195 2811 100 25.6 10.1
666 g
supernatant
249 59 56 1.73 0.4 6.06 1.7 51 0.04 0.03 627 159 47 4.4 1.0 4126 1610 72 27.9 8.8
11,000 g
pellet
56 12 13 3.54 0.6 0.51 0.4 4 0.03 0.03 128 60 9 7.8 2.9 517 323 9 31.9 18.1
Gradient
fraction
7.8 3.7 1.8 1.51 0.5 0.07 0.01 0.5 0.01 0.003 29 16 2 5.2 1.9 79 48 1.4 14.6 5.2
4. Wash the pellet with homogenization buffer by pipetting care-
fully across, to remove the fluffy layer (yellowish layer above a
darker tawny layer).
5. Resuspend the tawny pellet in 3–5 ml homogenization buffer,
take an aliquot, transfer the resuspended pellet to a new 70TI
tube, and add up to 30 ml with homogenization buffer and
centrifuge again at 11,000 gfor 15 min at 4 C.
6. Decant the supernatant and put it aside.
7. Resuspend the resulting tawny pellet in 2–3 ml resuspension
buffer and again take an aliquot.
3.1.4 Isopycnic Density
Gradient Centrifugation
1. Layer the resuspended pellet carefully onto a linear sucrose
gradient.
2. Carry out centrifugation in a SW28 swing-out bucket rotor at
85,000 gfor 60 min at 4 C without brakes (see Note 10).
3. After centrifugation, carefully remove the gradient tube from
the rotor bucket and the enriched mitochondria show up in the
middle of the tube as a light brown–yellowish ring.
4. Collect 2 ml aliquots by pipetting carefully from above in rotary
movements with a wide opening pipette tip (see Note 11).
5. Determine the mitochondrial activity from the aliquots by
measuring succinate dehydrogenase activity (see Subheading
3.3.1 SDH enzyme assay) and pool aliquots with highest activ-
ity (see Note 12).
6. Dilute pooled aliquots with fivefold volume (v/v) of resuspen-
sion buffer and transfer to 70TI tube follow a centrifugation at
11,000 gfor 15 min and 4 C.
7. Resuspend resulting final mitochondria pellet in 1–2 ml resus-
pension buffer.
8. This mitochondria fraction is ready for further experiments
including quality control assays.
3.2 Protein
Measurement
Protein measurements from all steps of the mitochondria prepara-
tion should be done. Therefore take 10–20 μl of each fraction/
aliquot and mix it up with the same amount of 1 M NaOH to
denature for protein measurement with standard Bradford assay.
3.3 Marker Enzyme
Assays
With all these enzyme assays, you can measure the protein activity
per ml and calculate the specific activity with the values from
protein content and document your mitochondria enrichment
and decrease of other organelles during preparation. For the differ-
ent enzyme activity assays, you need different dilutions of your
samples: succinate dehydrogenase delivers good results with a sam-
ple/aliquot dilution of 1:10 and 1:20, acidic phosphatase with 1:8
and 1:16, catalase 1:10, basic phosphatase pure and 1:2, and glu-
cose-6-phosphatase 1:10 (see Note 13).
Isolation and Quality Control of Functional Mitochondria 47
3.3.1 Mitochondria:
Succinate Dehydrogenase
(SDH) (See Ref. 7)
The succinate dehydrogenase catalyzes the oxidation from succi-
nate to fumarate under release of hydrogen. In vivo FAD would be
reduced to FADH
2
, but under these test assay conditions the
p-Iodonitrotetrazolium (INT), an artificial electron acceptor,
would be reduced to formazan which turns from colorless to
rusty red and can be measured at 490 nm.
1. Produce the INT solution: 100 μl for each sample/dilution is
needed.
2. Prepare 2 ml reaction tubes with 20 μl of diluted samples and
one only with resuspension buffer as a blank.
3. Add 300 μl Na succinate solution to each tube and incubate for
10–20 min at 37 C.
4. Add 100 μl INT solution and incubate 10 min at 37 C.
5. Stop the enzyme reaction by adding 1 ml stop solution.
6. Centrifuge the tubes for 2 min at 20,000 x g to get rid of
precipitates.
7. Measure the extinction of supernatants at 490 nm.
8. Calculate the activity per ml and the specific activity:
ΔEVEml½=VPml½
εmol ml=μmol cm½dcm½tmin½
¼μmol½
ml min½
ΔE: extinctions difference (measured value measured blank).
V
E
/V
P
: enzyme assay volume/added sample volume,
i.e., the entire dilution factor of the measured sample in the
assay.
ε
mol
: molar extinctions coefficient [ml/μmol cm] (for INT it
is 0.0134).
d: thickness of used cuvette [cm].
t: incubation time Na succinate solution with sample before
adding INT solution.
Calculate the specific activity [μmol/mg min] in your sample
by simple dividing through the protein concentration of the
measured sample.
3.3.2 Lysosomes: Acid
Phosphatase (See Ref. 8)
The acid phosphatase has a working optima at pH 4 to 5. In this
assay, the conversion from p-nitrophenylphosphate and H
2
Oto
p-nitrophenol is the enzymatic step and with a pH increase by
adding NaOH the “produced” nitrophenol shift to nitrophenolate
ions and the solution gets yellow color and extinction could be
measured at 410 nm.
1. Prepare an assay mix with 16 mM p-nitrophenylphosphate
solution and the 180 mM Na acetate solution (1:1, v/v).
48 Sonja Hartwig et al.
2. Prepare 1.5 ml reaction tubes and add 25 μl of recommended
sample dilutions and use resuspension buffer as blank.
3. Add 200 μl assay mix-solution to each tube and incubate for
20–30 min at 37 C.
4. Stop the enzyme reaction by adding 600 μl stop solution to
each tube.
5. Centrifuge the tubes for 2 min at 20,000 gto get rid of
precipitates.
6. Measure the extinction of supernatants at 410 nm.
7. Calculate the activity per ml and the specific activity:
ΔEVEml½=VPml½
εmol ml=μmol cm½dcm½tmin½
¼μmol½
ml min½
ΔE: extinctions difference (measured value measured blank),
V
E
/V
P
: enzyme assay volume/added sample volume,
i.e., the entire dilution factor of the measured sample in the
assay.
ε
mol
: molar extinctions coefficient [ml/μmol cm] here for
converted nitrophenol is 0.521.
d: thickness of used cuvette [cm].
t: incubation time.
Calculate the specific activity [μmol/mg min] in your sample
by simple dividing through the protein concentration of the
measured sample.
3.3.3 Plasma Membrane:
Basic Phosphatase
(See Ref. 8)
The basic phosphatase has a working optima at pH 9–10. In this
assay, the conversion from p-nitrophenylphosphate and H
2
Oto
p-nitrophenol is the enzymatic step and with a pH increase by
adding NaOH the “produced” nitrophenol shift to nitrophenolate
ions and the solution gets yellow color and extinction could be
measured at 410 nm.
1. Prepare an assay mix with nitrophenylphosphate solution and
the Na borate solution (1:1, v/v). Also add 1 M MgCl
2
to a
final concentration in the assay mix of 2 mM (i.e., 2 μl in 1 ml).
2. Prepare 1.5 ml reaction tubes with 25 μl of pure and diluted
samples and use resuspension buffer as blank.
3. Add 200 μl assay mix-solution to each tube and incubate for
20–30 min at 37 C.
4. Stop the enzyme reaction by adding 600 μl stop solution to
each tube.
5. Centrifuge the tubes for 2 min at 20,000 gto get rid of
precipitates.
6. Measure the extinction of supernatants at 410 nm.
Isolation and Quality Control of Functional Mitochondria 49
7. Calculate the activity per ml and the specific activity:
ΔEVEml½=VPml½
εmol ml=μmol cm½dcm½tmin½
¼μmol½
ml min½
ΔE: extinctions difference (measured value measured blank).
V
E
/V
P
: enzyme assay volume/added sample volume,
i.e., the entire dilution factor of the measured sample in the
assay.
ε
mol
: molar extinctions coefficient [ml/μmol cm] here for
converted nitrophenol is 0.521.
d: thickness of used cuvette [cm].
t: incubation time.
Calculate the specific activity [μmol/mg min] in your sample
by simple dividing through the protein concentration of the
measured sample.
3.3.4 Peroxisome:
Catalase (See Ref. 9)
The catalase converts hydrogen peroxide to water and hydrogen.
Titanoxid sulfate build up a yellow complex with hydrogen perox-
ide and the extinction of this complex could be measured at 405 nm
(see Note 14).
1. Prepare the titanyl solution fresh (see Note 3).
2. Test assay buffer by adding 1 ml titanyl solution to 500 μl assay
buffer, centrifuge for 5 min at 20,000 gand measure extinc-
tion from the supernatant at 405 nm. The value should be
between OD 0.5 and 0.6. If this is not the case, you have to
prepare the assay buffer fresh (see Note 15).
3. Mix 10 μl of the sample with 30 μl of the catalase sample buffer
and do all further steps on ice (see Note 16).
4. Add 500 μl from the tested assay buffer and stop the reaction
by adding 1 ml titanyl solution exactly after 1 min.
5. Centrifuge samples for 5 min at 20,000 g.
6. Measure the extinction at 405 nm against water and your
“positive” blank (see Note 17).
7. Define E
405nm
0.001 ¼0.001 U/ml and calculate the differ-
ence from extinction measured from sample to the “positive”
blank and also calculate the activity and specific activity for
catalase as follow:
ΔEVEml½=VPml½
εmol ml=Ucm½dcm½tmin½
¼U½
ml min½
ΔE: extinctions difference (measured “positive” blank
measured value).
50 Sonja Hartwig et al.
V
E
/V
P
: enzyme assay volume/added sample volume,
i.e., the complete dilution factor of the measured sample in the
assay.
ε
mol
: molar extinctions coefficient [ml/U cm] adaption
0.001 ¼1U(see Note 18).
d: thickness of used cuvette [cm].
t: incubation time before addition of titanyl solution.
Calculate the specific activity [U/mg min] in your sample by
simple dividing through the protein concentration of the measured
sample.
3.3.5 Endoplasmic
Reticulum:
Glucose-6-Phophatase
(See Ref. 10)
The glucose-6-phosphatase (G-6-Pase) dephosphosphorylate
G-6-P to glucose and free phosphate. Colorimetric measurement
of free phosphate content could be done by Fiske-Subbarow
method. In this assay, a blue complex is formed, when free phos-
phate is mixed with ammonium molybdate and 1-amino-2-naph-
thol-4-sulfonic acid, and could be measured at 815 nm. Important
for this enzyme assay is to measure next to a blank the sample
without or with substrate. To calculate the nmol/ml concentra-
tions, a standard curve with potassium hydrogen phosphate has to
be measured.
1. Prepare Fiske-Subbarow mix freshly by adding 3.75 ml Fiske-
Subbarow reagent and 10 ml perchloric acid (60%) to 86.25 ml
ammonium heptamolybdate solution.
2. For each sample, you have to measure two values, one without
substrate (endogenous P
i
) and one with substrate (G-6-P) for
enzymatic dephosphorylation combine in 2 ml tubes:
For endogenous value:
(a) 100 μl cacodylate buffer.
(b) 100 μl saccharose buffer.
(c) 100 μlH
2
O.
For enzymatic value:
(a) 100 μl cacodylate buffer.
(b) 100 μl saccharose buffer.
(c) 100 μl glucose-6-phosphate solution.
3. By adding 100 μl sample to each tube, the enzymatic reaction is
started.
4. Incubate all samples for 30 min at 37 C.
5. During incubation, prepare a standard curve with the potas-
sium hydrogen phosphate as replicates (0, 20, 50, 100, 150 to
200 nmol/ml). The volume of each standard sample is 1 ml (see
Note 19).
Isolation and Quality Control of Functional Mitochondria 51
6. Stop the enzymatic reaction (step 4) in the sample tubes by
adding 1.5 ml 8% TCA solution to each tube.
7. Centrifugate the sample for 10 min at 1000 gto get rid of
precipitates.
8. Transfer 1 ml of the resulting supernatant to a new 2 ml tube
and add 1 ml of the fresh prepared Fiske-Subbarow mix to each
tube and also to the samples of the standard curve.
9. Vortex all samples and incubate for 30 min at RT.
10. Measure the standard curve, basal, and enzymatic samples at
815 nm.
11. Generate a x/y-graph with values from standard curve, y-axis is
OD, and x-axis the known free phosphate (Pi) amount [nmol].
12. To calculate the produced Pi amount, subtract the endogenous
from the enzymatic OD, read off the Pi value from the standard
curve and calculate the enzymatic activity as follows:
Pi nmol=ml½VEml½=VPml½
tmin½ ¼nmol½
ml min½
V
E
/V
P
: enzyme assay volume/ added sample volume,
i.e., the complete dilution factor of the measured sample in the
assay.
t: incubation time before addition of titanyl solution.
3.3.6 Mitochondria
(Functional Integrity): JC-1
Uptake Assay (See Ref. 11)
Uptake measurement of the fluorescent carbocyanine dye (JC-1) is
a surrogate parameter for mitochondrial inner membrane integrity
because it is only possible if electrochemical proton gradient is
formed. Depending upon the transmembrane electric field, JC-1
will is taken up into mitochondrial matrix and if concentration
raised more than 1 mM a red-orange fluorescence will occur at
590 nm, due to aggregation of dye within the matrix.
1. Prepare 1buffer from storage and JC-1 assay buffer by 1:5
(v/v) dilution in H
2
O.
2. Prepare solution of the samples with storage buffer in a protein
concentration of 0.4 mg/ml.
3. From this prepare a dilution row where you apply in total
10, 20, 30, and 40 μg protein in a volume of 100 μl (filled up
with storage buffer) in a 2 ml reaction tube.
4. Add 1.9 ml 1JC-1 assay buffer to each sample of the
dilution row.
5. Add 2 μl JC-1 stain to the lid of the reaction tube.
6. Close the tube and vortex directly.
7. Incubate the samples 10 min at RT in the dark.
52 Sonja Hartwig et al.
8. Read the fluorescence of samples in a spectrofluorometer with
an excitation wavelength of 490 nm and an emission wave-
length of 590 nm.
9. Generate an x/y-graph with values from four samples of the
dilution row, x-axis is the amount of protein, and y-axis is the
fluorescence at 590 nm.
10. Calculate the fluorescence produced in the original sample per
mg protein:
ΔFL dil
VC¼FLU
mgP
FLU: fluorescence units.
mgP: milligram protein.
ΔFL: fluorescence (sample) fluorescence (blank).
dil: dilution factor to prepare 0.4 mg/ml sample dilution.
V: volume of the sample in [ml].
C: protein concentration [mg/ml].
4 Notes
General note: Especially in face of the complexity of the workflow
and the enormous biological variability, it appears that an extraordi
nary diligence in each step of the analysis is of great importance.
This begins with the first experimental step, i.e., sample collection
and preparation, which frequently is underestimated. In this con-
text, the close collaboration between the different scientific disci-
plines and the development of Standard Operation Procedures is of
particular importance.
1. First do the pH adjustment of buffers with 2 M NaCl and then
add required amount of DTT always fresh.
2. To bring high amounts of saccharose in solution quickly, warm
up the solutions a little bit (30–40 C).
3. The titanyl solution needs up to 30 min at RT to get dissolved
(clear solution) and is stable and usable at RT for 2 h.
4. Store the solution at 4 C, it is stable up to 4 weeks.
5. Store the solution in an amber bottle because it is light sensi-
tive. A benefit is to prepare the solution 2 days before use and
pass the solution through a 0.02 μm filter before storing. The
solution is perishable, if it gets yellowish, throw it away and
make a fresh one.
6. Use anhydrous solvents (i.e., in this case dimethyl sulfoxide
(DMSO)) to dissolve cyanine dyes like JC-1.
Isolation and Quality Control of Functional Mitochondria 53
7. Test the gradient linearity by aliquoting the gradient up to
20 fractions and measure the saccharose concentration with a
refractometer.
8. Make sure that the glass douncer reach the bottom of the
cylinder and that the solution and the equipment are kept on
ice. Prevent negative pressure while douncing by using loose fit
douncer.
9. If you take aliquots from each step of fractionation, you can
easily perform protein content measurement and marker
enzyme assays. The results show the mitochondria enrichment
and decrease of other organelles. For example, see Table 1.
10. The running down without brakes from here used high
g-forces that increase the standby time, therefore you can go
for lunch.
11. It is important to use a pipette tip with a wide opening to
collect the gradient suspension carefully from the top of the
surface.
12. The highest mitochondria content is located in a density of
about 42% saccharose. The refractometer measurement helps,
next to SDH activity assay, to identify the correct fraction.
13. For marker enzyme assays, different solutions are needed and
best use resuspensions buffer for all samples.
14. In this assay, the decrease of catalase substrate is the indicator
for enzyme activity. Adding H
2
O
2
to titanyl solution results in
yellow solution. If catalase is present, the H
2
O
2
amount
decreases and the solution get colorless. So, the decrease of
color represents enzymatic activity.
15. H
2
O
2
is light sensitive, therefore protect the buffer and assay
samples from light.
16. The catalase is one of the quickest known enzymes; therefore,
the exact compliance of incubation time and temperature of
buffers is mandatory.
17. The positive blank is the maximum amount of H
2
O
2
that could
be measured within the assay.
18. There is no molar extinction coefficient known, so to calculate
a relative activity adapt 0.001 OD difference ¼1 arbitrary unit.
19. This assay is very sensitive to free phosphate (Pi), so it is
mandatory to use always disposable, phosphate-free plastic
material because cleaning solution contains and leaves traces
of free phosphate and therefore influence or rather damage
your assay measurements.
54 Sonja Hartwig et al.
References
1. Nunnari J, Suomalainen A (2012) Mitochon-
dria: in sickness and in health. Cell
148:1145–1159
2. Cherry C, Thompson B, Saptarshi N, Wu J,
Hoh J (2016) A ‘Mitochondria’ Odyssey.
Trends Mol Med 22:391–403
3. Frezza C, Cipolat S, Scorrano L (2007) Organ-
elle isolation: functional mitochondria from
mouse liver, muscle and cultured fibroblasts.
Nat Protoc 2:287–295
4. Hartwig S, Feckler C, Lehr S, Wallbrecht K,
Wolgast H, Mu¨ller-Wieland D, Kotzka J
(2009) A critical comparison between two clas-
sical and a kit-based method for mitochondria
isolation. Proteomics 9(11):3209–3214
5. Franko A, Baris OR, Bergschneider E, von
Toerne C, Hauck SM et al (2013) Efficient
isolation of pure and functional mitochondria
from mouse tissues using automated tissue dis-
ruption and enrichment with anti-TOM22
magnetic beads. PLoS One 8:e82392
6. Islinger M, Wildgruber R, Vo
¨lkl A (2018) Pre-
parative free-flow electrophoresis, a versatile
technology complementing gradient centrifu-
gation in the isolation of highly purified cell
organelles. Electrophoresis 39:2288–2299
7. Pennington RJ (1961) Biochemistry of dystro-
phic muscle. Mitochondrial succinate-
tetrazolium reductase and adenosine tripho-
sphatase. Biochem J 80:649–654
8. Walter K, Schuett C (1974) In: Bergmeyer HU
(ed) Methods in enzymatic analysis, vol 3. Aca-
demic Press, Inc, New York, pp 860–864
9. Aebi H (1984) Catalase in vitro. Methods
Enzymol 105:121–126
10. Baginski ES, Foa PP, Zak B (1969) Determina-
tion of phosphate and phosphomonoesterases
in biologic materials. Am J Med Technol
35:475–486
11. Reers M, Smith TW, Chen LB (1991)
J-aggregate formation of a carbocyanine as a
quantitative fluorescent indicator of membrane
potential. Biochemistry 30:4480–4486
Isolation and Quality Control of Functional Mitochondria 55
Chapter 4
Purification of Functional Platelet Mitochondria
Using a Discontinuous Percoll Gradient
Jacob L. Le
´ger, Nicolas Pichaud, and Luc H. Boudreau
Abstract
The isolation of mitochondria is gaining importance in experimental and clinical laboratory settings. Of
interest, mitochondria and mitochondrial components (i.e.,circular mitochondrial DNA, N-formylated
peptides, cardiolipin) have been involved in several human inflammatory pathologies, such as cancer,
Alzheimer’s disease, Parkinson’s disease, and rheumatoid arthritis. While several mitochondrial isolation
methods have been previously published, these techniques are aimed at yielding mitochondria from cell
types other than platelets. In addition, little information is known on the number of platelet-derived
microvesicles that can contaminate the mitochondrial preparation or even the overall quality as well as
functional and structural integrity of mitochondria. Here we describe a purification method, using a
discontinuous Percoll gradient, yielding mitochondria of high purity and integrity from human platelets.
Key words Mitochondria isolation, Mitochondria membrane integrity, Percoll extraction method,
Platelet-derived microvesicles, Platelet-derived mitochondria
1 Introduction
Platelets are small mitochondria-containing anucleate cells (3–5 μm
in size) that patrol the vasculature to maintain the homeostasis by
preventing blood loss and promoting wound repair [1]. When
exposed to physiological agonists such as thrombin, collagen, aden-
osine diphosphate, or immune-complexes, platelets can release
small extension of their cytoplasm known as platelet-derived micro-
vesicles (0.1–1 μm in size) in the extracellular milieu. Interestingly,
the subpopulation of platelet-derived microvesicles is highly het-
erogeneous. In addition to releasing cell-derived microvesicles
upon their activation, platelets also shed fully functional mitochon-
dria in the extracellular milieu known as freeMitos [2]. Conse-
quently, platelet-derived mitochondria and mitochondrial
components are involved in the inflammatory response and con-
tribute to amplify this important cellular process.
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_4,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
57
While mitochondria are widely considered as the powerhouse
of the cell by actively participating in the transduction of energy via
the oxidative phosphorylation process (i.e., ATP generation) nec-
essary for most metabolic reactions [3], they also participate in the
immune response [2,4]. The mitochondrion possesses numerous
typical hallmarks of the bacteria such as unmethylated CpG motifs
circular DNA [5], N-formylated peptides [6], and unique phos-
pholipid cardiolipin embedded in their inner membrane [7]. When
released in the extracellular milieu, these bacteria-like components
can act as damage-associated molecular patterns (DAMPs) and
may trigger important inflammatory responses [2,5]. Therefore,
the isolation of these organelles and their derived components is
gaining importance in experimental and clinical laboratory settings
[8–10].
Given the important role of mitochondria in the inflammatory
response, it is essential to be able to characterize their physiology, as
well as their involvement in the immune response and intercellular
communication, by developing techniques allowing the rapid isola-
tion of mitochondria that will preserve their functional properties.
Our approach provides an ideal isolation method that contains fully
functional mitochondria with a minimal presence of platelet-
derived microvesicles.
2 Materials
Prepare all solutions using ultrapure water (prepared by purifying
deionized water with a sensitivity of 18 MΩ-cm at 25 C and a total
organic carbon content inferior to 5 ppb) and analytical grade
reagents. Reagents are prepared at room temperature and stored
at 4 C unless indicated otherwise.
2.1 Solutions
[11–13]
1. Acid citrate dextrose (ACD-A): Sodium citrate dihydrate
75 mM, citric acid 38 mM, dextrose 136 mM, pH 4.5. To
prepare 1000 mL of ACD-A, weigh 22.0 g of sodium citrate
dihydrate and dissolve in 850 mL of water. Then, add 7.3 g of
citric acid and 24.5 g of dextrose to the solution and mix.
Adjust the pH to 4.5 with HCl solution, then complete the
volume to 1000 mL with water. Filter the ACD-A solution
through a 0.2 μM sterile pore.
2. Acid citrate dextrose (ACD-B): Sodium citrate dihydrate
45 mM, citric acid 25 mM, dextrose 81 mM, pH 4.5. To
prepare 1000 mL of ACD-B, weigh 13.2 g of sodium citrate
dihydrate and dissolve in 850 mL of water. Then, add 4.8 g of
citric acid and 14.7 g of dextrose to the solution and mix.
58 Jacob L. Le
´ger et al.
Adjust the pH to 4.5 with HCl solution, then complete the
volume to 1000 mL with water. Filter the ACD-B solution
through a 0.2 μM sterile pore.
3. Ethylenediaminetetraacetic acid solution (EDTA): EDTA
0.5 M, pH 8.0. To prepare 1000 mL of EDTA solution,
weigh 186.1 g of EDTA and dissolve in 800 mL of water.
Adjust the pH to 8.0 by adding 20.0 g of NaOH to the
solution, then complete the volume to 1000 mL with water.
Filter the EDTA solution through a 0.2 μM sterile filter.
4. Sucrose solution: Sucrose solution 2.5 M. To prepare 1000 mL
of sucrose solution, weigh 855.0 g of sucrose and dissolve in
800 mL of water. Complete the volume to 1000 mL with water
and filter the sucrose solution through a 0.2 μM sterile filter.
5. Tris buffer (TB): Tris(hydroxymethyl)aminomethane 1 M,
pH 7.5. To prepare 50 mL of TB, dissolve 6.1 g of tris
(hydroxymethyl)aminomethane in 30 mL of water and adjust
the pH to 7.5 with HCl solution. Complete the volume to
50 mL with water and then filter the TB through a 0.2 μM
sterile syringe filter.
6. Isolation buffer (IB): Sucrose 0.2 M, tris(hydroxymethyl)-
aminomethane 11 mM, EDTA 1 mM, pH 7.5. To prepare
1000 mL of IB, mix 80 mL of sucrose 2.5 M and 2 mL of
EDTA 0.5 M with 718 mL of water. Add 1.33 g of tris
(hydroxymethyl)aminomethane and dissolve completely and
adjust pH to 7.5 with HCl solution. Complete the volume to
1000 mL and filter the IB through a 0.2 μM sterile filter.
7. Isolation buffer without EDTA (IB without EDTA): Sucrose
0.2 M, tris(hydroxymethyl)aminomethane 11 mM, pH 7.5. To
prepare 1000 mL of IB without EDTA, mix 80 mL of sucrose
2.5 M and 720 mL of water. Add 1.33 g of tris
(hydroxymethyl)aminomethane and dissolve completely and
adjust pH to 7.5 with HCl solution. Complete the volume to
1000 mL and filter the IB through a 0.2 μM sterile filter.
8. Tyrode’s buffer 7.4 (TY7.4): NaCl 134 mM, KCl 2.9 mM,
Na
2
HPO
4
0.34 mM, NaHCO
3
12 mM, HEPES 20 mM,
MgCl
2
1 mM, glucose 5 mM, bovine serum albumin (BSA)
0.5 mg/mL, pH 7.4. To prepare 1000 mL of TY7.4, dissolve
7.88 g NaCl, 0.22 g KCl, 0.048 g Na
2
HPO
4
, 1 g NaHCO
3
,
4.76 g HEPES, 0.9 g glucose, 0.5 g BSA, and 0.1 g MgCl
2
in
800 mL of water. Adjust the pH to 7.4 with HCl or NaOH
solutions and then complete the volume to 1000 mL with
water. Filter the TY7.4 through a 0.2 μM sterile filter.
9. Tyrode’s buffer 6.5 (TY6.5): NaCl 134 mM, KCl 2.9 mM,
Na
2
HPO
4
0.34 mM, NaHCO
3
12 mM, HEPES 20 mM,
MgCl
2
1 mM, glucose 5 mM, bovine serum albumin (BSA)
Purification of Platelet Mitochondria 59
0.5 mg/mL, pH 6.5. To prepare 50 mL of TY6.5, withdraw
50 mL of prepared TY7.4 and adjust the pH to 6.5 with HCl
solution. Filter the TY6.5 through a 0.2 μM sterile syringe
filter.
10. Protease inhibitor 10: 1 capsule of antiprotease (Roche) dis-
solved in 1 mL of IB.
11. Proteinase K: Proteinase K 5 mg/mL. Resuspend proteinase K
(Sigma-Aldrich) to 5 mg/mL in TB.
12. Percoll solution 90% (PS90): 10% (v/v) 2.5 M sucrose solution
to 90% (v/v) Percoll. To prepare 100 mL of Percoll solution
90%, mix 90 mL of Percoll Solution (GE Healthcare) with
10 mL of 2.5 M sucrose solution. Adjust the pH to 7.2 with
HCl and KOH solutions. Filter the solution through a 1.2 μM
Millipore filter. The solution is kept at 4 C for a week or at
20 C for a month. Refilter before use (see Note 1).
13. Percoll solution 15% (PS15): 16.7% (v/v) of PS90 to 83.3%
(v/v) of IB. To prepare 10 mL of Percoll solution 15%, mix
1.67 mL of PS90 with 8.33 mL of IB. The solution is kept on
ice until use for the discontinuous gradient.
2.2 Optional
Solutions [14]
1. Respiration buffer (MiR05): EGTA 0.5 mM, MgCl
2
·6H
2
O
3 mM, lactobionic acid 60 mM, taurine 20 mM, KH
2
PO
4
10 mM, HEPES 20 mM, D-sucrose 110 mM, BSA 1 g/L.
To prepare 1000 mL of MiR05, dissolve under sterile condi-
tions 0.19 g EGTA, 0.61 g MgCl
2
·6H
2
O, 2.5 g taurine, 1.36 g
KH
2
PO
4
, 4.77 g HEPES, 37.65 g D-sucrose, and 1 g BSA in
800 mL of water. Combine with 120 mL of 0.5 M lactobionic
acid solution (prepared in advance by dissolving 35.83 g lacto-
bionic acid in 100 mL of water, adjusting the pH to 7.0 with
KOH solution, and adjusting the volume to 200 mL with
water). Finally, adjust the pH to 7.1 with KOH solution and
complete the volume to 1000 mL. The solution is kept in
aliquots at 20 C in plastic vials for up to a year.
2.3 Percoll Gradient
Preparation [13,
15,16]
1. Discontinuous Percoll gradient: Fill a 1.5 mL microcentrifuge
tube with 500 μL of PS15 to prepare the 15% Percoll fraction
and store on ice until the 0% Percoll fraction is ready (crude
mitochondrial extract).
2.4 Blood Collection
Preparation
1. Blood collection: Obtain approximately 80 mL of blood from
consenting volunteers using 10 mL blood collection glass
tubes already containing 1 mL of ACD-A (see Note 2). These
tubes are prepared in advance under sterile conditions by add-
ing the ACD-A through a needle and syringe.
60 Jacob L. Le
´ger et al.
3 Methods
3.1 Platelet Isolation
[11,13]
1. Centrifuge blood in falcon tubes at 275 gfor 15 min (with-
out brake) at room temperature (RT).
2. Carefully withdraw the platelet-rich plasma (PRP) using a Pas-
teur pipette.
3. Add a volume of ACD-B to the PRP equivalent to 1/5 of the
collected PRP and add 10 mM of EDTA.
4. Centrifuge PRP at 400 gfor 2 min (without brake) at RT to
remove the contaminating erythrocytes and leucocytes.
5. Collect and centrifuge the supernatant at 1300 gfor 10 min
(without brake) at RT to pellet the platelets.
6. Gently resuspend the platelet pellet in 500 μL of TY6.5 by
doing some slow ups and downs with a P1000 micropipette.
7. Complete the volume to 40 mL with TY7.4.
8. Count total platelets in a 1:300 platelet dilution sample using a
hemocytometer.
9. Add 10 mM of EDTA and centrifuge platelets at 1300 gfor
10 min (without brake) at RT.
10. Slowly resuspend platelet pellet at 2 10
9
cells/mL in IB.
3.2 Crude
Mitochondrial
Extraction [13,17]
1. Platelets are incubated with 150 μg/mL of proteinase K and
mixed by inversion for 5 min at RT.
2. Homogenize the extract with a Wheaton overhead stirrer using
a Teflon homogenizer adapter for 30–40 passes with speed set
at 2.5–3.0 (see Note 6).
3. Add protease inhibitors cocktail (1) from the 10concen-
trated stock (optional step).
4. Transfer mitochondrial extract to 1.5 mL microcentrifuge
tubes.
5. Centrifuge the lysate at 1300 gfor 10 min at 4 C to remove
platelet debris.
6. Using a pipette, transfer the supernatant in a fresh tube and
centrifuge at 8000 gfor 10 min at 4 C to obtain the crude
mitochondrial extract.
7. Pool the pelleted crude mitochondrial pellets using 250 μLof
IB (see Note 3).
3.3 Mitochondria
Purification
1. Carefully layer the resuspended crude mitochondria on top of
the 15% Percoll prepared layer while avoiding the mixing of the
layers (see Note 4). To do so, use a P100 micropipette, and
carefully eject the crude mitochondria along the wall of the
microcentrifuge tube and gently touch the Percoll layer. Pull
Purification of Platelet Mitochondria 61
back the tip while keeping a small fluid bridge, slowly release
the sample, and allow it to sit on top of the Percoll layer (see
Note 7).
2. Centrifuge the sample at 21,000 gfor 8 min at 4 C using a
slow acceleration and deceleration.
3. Carefully remove the tube from the centrifuge and observe the
two white layers. The bottom contains mostly purified mito-
chondria while the top contains platelet membrane debris,
microparticles, and some mitochondria.
4. Using a P100 pipette, direct the tip to the bottom of the tube
to collect the mitochondria layer while being careful not to
collect the platelet membrane debris.
5. Resuspend the mitochondria in 1 mL of IB and centrifuge at
13,000 gfor 10 min to remove the remaining Percoll.
6. The obtained pellet consists of purified platelet mitochondria
(Fig. 1)[13].
7. The purified mitochondria are resuspended in a buffer of
choice and volume of choice depending on the downstream
application as suggested below (Table 1).
3.4 Mitochondrial
Yield by Flow
Cytometry
1. Resuspend the purified mitochondria in 500 μL of IB without
EDTA (see Note 5).
2. Take 1 μL of mitochondria, add 1 μL of MitoTracker™Deep
Red FM (1 μM) diluted according to the manufacturer’s pro-
tocol and 98 μL of PBS.
Fig. 1 Transmission electron microscopy visualization of the purified
mitochondria showing mitochondria (black arrowhead) and some remaining
platelet debris (white arrowhead)
62 Jacob L. Le
´ger et al.
3. Incubate in the dark for 15 min.
4. Add 500 μL of PBS.
5. Assess the total count of mitochondria using a flow cytometer
gating MitoTracker™Deep Red FM positive events
(Fig. 2)[13].
3.5 Mitochondrial
Yield by Bicinchoninic
Acid Assay (BCA)
1. Alternatively, resuspend the purified mitochondria in 500 μLof
IB buffer without EDTA.
2. Take 25 μL of mitochondria and add 25 μL of ice-cold NP40
lysis solution.
3. Keep on ice for 15 min while vortexing every 3 min.
4. Spin the mitochondrial sample at 10,000 gfor 10 min at 4 C
and collect proteins present in the supernatant and transfer to a
new microcentrifuge tube.
Table 1
Buffer selection for platelet mitochondria resuspension
Respiration buffer
Buffer with
EDTA Buffer without EDTA
Co-incubation
buffer
Resuspension
volume
2.2 mL 500 μL 500 μL 500 μL
Recipe MiR05 (see
Subheading 2.2,
item 1)
IB (see
Subheading
2.1,item 6)
IB without EDTA (see
Subheading 2.1,
item 7)
RPMI 1640 with
10% fetal bovine
serum
Fig. 2 Overlay dot plot flow cytometry gating strategy to obtain a total count of the mitochondrial population
Purification of Platelet Mitochondria 63
5. Proceed to determine the mitochondrial protein concentration
using a spectrophotometer following the procedures described
by the manufacturer of a micro BCA Protein Assay Kit (Pierce
Chemical).
4 Notes
1. Aliquoted Percoll solutions should be refiltered in order to
remove aggregated particles which can disrupt the cellular
fractionation.
2. Blood tubes should be carefully examined for blood clots as
they contain activated platelets. Activated platelets produce
microparticles, which are discarded in the isolation of platelet-
derived mitochondria.
3. Verify the pH of the solutions, especially the IB, as it can shift
into unfavorable range after a week of storage at 4 C.
4. Sudden movements while removing the tube of the centrifuge,
after the purification step on the Percoll gradient, can disturb
the mitochondrial layer on the gradient, rendering the purifi-
cation ineffective.
5. When resuspending the purified mitochondrial pellet, keep in
mind that in the presence of high concentrations of EDTA,
mitochondrial respiration measurements, protein assays mea-
surements, and biochemical measurements may be disturbed.
6. Note that the properties of the tissue homogenizer may vary
depending on the equipment used. Adjustments should be
made by the experimenter on the number of passes of the
Teflon in order to obtain a mostly foamy looking substance.
The foamy looking extract will then rest on ice and return to a
liquid state substance. Other extraction methods have yet to be
tested.
7. Separation issues may arise during the Percoll purification
steps, as the Percoll solution can change density depending
on the lot number. Additional adjustments could be necessary
as specific organelle density can vary between donors. Typically,
the mitochondria have a density ranging between 1.09 and
1.11 [18], while platelet membranes and microparticles are
situated at around 1.04 and 1.06 [19]. This allows the separa-
tion of the two when the appropriate density of the Percoll
solution is calculated. The following formula, obtained from
the GE Percoll manufacturer’s product sheet, can be used to
adjust the density of the 15% Percoll solution:
Vy¼Vi
ρiρðÞ
ρρy
64 Jacob L. Le
´ger et al.
where: V
y
: volume of IB in mL (8.33 mL).
V
i
: volume of PS90 in mL (1.67 mL).
ρ
i
: density of PS90 in mL (1.149 g/mL).
ρ
y
: density of IB in mL (~1.023 g/mL—can differ when
measured).
ρ: desired density of final 15% Percoll solution (>1.05 g/mL).
The density of PS15 is adjusted by increasing the sucrose
content of the IB used to dilute the PS90 to obtain the PS15.
This new density IB should not be used for manipulations other
than diluting the PS90 to PS15.
Acknowledgments
This work was funded by the Canadian Institutes of Health
Research (CIHR), New Brunswick Health Research Foundation
(NBHRF), New Brunswick Innovation Foundation (NBIF), Nat-
ural Sciences and Engineering Research Council (NSERC), and the
Universite
´de Moncton.
References
1. Davı
`G, Patrono C (2007) Platelet activation
and atherothrombosis. N Engl J Med 357
(24):2482–2494. https://doi.org/10.1056/
NEJMra071014
2. Boudreau LH, Duchez A-C, Cloutier N,
Soulet D, Martin N, Bollinger J, Pare
´A,
Rousseau M, Naika GS, Le
´vesque T et al
(2014) Platelets release mitochondria serving
as substrate for bactericidal group IIA-secreted
phospholipase A2 to promote inflammation.
Blood 124(14):2173–2183. https://doi.org/
10.1182/blood-2014-05-573543
3. Saraste M (1999) Oxidative phosphorylation at
the fin de Sie
`cle. Science 283
(5407):1488–1493
4. Zhang Q, Raoof M, Chen Y, Sumi Y, Sursal T,
Junger W, Brohi K, Itagaki K, Hauser CJ
(2010) Circulating mitochondrial DAMPs
cause inflammatory responses to injury. Nature
464(7285):104–107. https://doi.org/10.
1038/nature08780
5. Zhang L, Deng S, Zhao S, Ai Y, Zhang L,
Pan P, Su X, Tan H, Wu D (2016) Intra-
peritoneal administration of mitochondrial
DNA provokes acute lung injury and systemic
inflammation via toll-like receptor 9. Int J Mol
Sci 17(9):1425. https://doi.org/10.3390/
ijms17091425
6. Gray MW, Burger G, Lang BF (1999) Mito-
chondrial evolution. Science 283
(5407):1476–1481
7. Schlame M (2008) Cardiolipin synthesis for
the assembly of bacterial and mitochondrial
membranes. J Lipid Res 49(8):1607–1620.
https://doi.org/10.1194/jlr.R700018-
JLR200
8. Hayakawa K, Bruzzese M, Chou SH-Y,
Ning M, Ji X, Lo EH (2018) Extracellular
mitochondria for therapy and diagnosis in
acute central nervous system injury. JAMA
Neurol 75(1):119–122. https://doi.org/10.
1001/jamaneurol.2017.3475
9. Marcoux G, Duchez A-C, Rousseau M,
Le
´vesque T, Boudreau LH, Thibault L, Boilard
E (2017) Microparticle and mitochondrial
release during extended storage of different
types of platelet concentrates. Platelets 28
(3):272–280. https://doi.org/10.1080/
09537104.2016.1218455
10. Wilkins HM, Carl SM, Weber SG, Ramanujan
SA, Festoff BW, Linseman DA, Swerdlow RH
(2015) Mitochondrial lysates induce inflamma-
tion and Alzheimer’s disease-relevant changes
in microglial and neuronal cells. J Alzheimers
Dis 45(1):305–318. https://doi.org/10.
3233/JAD-142334
Purification of Platelet Mitochondria 65
11. Doucet MS, Jougleux J-L, Poirier SJ,
Cormier M, Le
´ger JL, Surette ME,
Pichaud N, Touaibia M, Boudreau LH
(2019) Identification of peracetylated querce-
tin as a selective 12-lipoxygenase pathway
inhibitor in human platelets. Mol Pharmacol
95(1):139–150. https://doi.org/10.1124/
mol.118.113480
12. Frezza C, Cipolat S, Scorrano L (2007) Organ-
elle isolation: functional mitochondria from
mouse liver, muscle and cultured filroblasts.
Nat Protoc 2(2):287–295. https://doi.org/
10.1038/nprot.2006.478
13. Le
´ger JL, Jougleux J-L, Savadogo F,
Pichaud N, Boudreau LH (2019) Rapid isola-
tion and purification of functional platelet
mitochondria using a discontinuous Percoll
gradient. Platelets 31(2):258–264. https://
doi.org/10.1080/09537104.2019.1609666
14. Gnaiger E, Kuznetsov AV (2002) Mitochon-
drial respiration at low levels of oxygen and
cytochrome c. Biochem Soc Trans 30
(2):252–258
15. Graham JM (2001) Purification of a crude
mitochondrial fraction by density-gradient
centrifugation. Curr Protoc Cell Biol
Chapter 3:Unit 3.4. https://doi.org/10.
1002/0471143030.cb0304s04
16. Kiss DS, Toth I, Jocsak G, Sterczer A, Bartha T,
Frenyo LV, Zsarnovszky A (2016) Preparation
of purified Perikaryal and Synaptosomal mito-
chondrial fractions from relatively small hypo-
thalamic brain samples. MethodsX 3:417–429.
https://doi.org/10.1016/j.mex.2016.05.004
17. Fukami MH, Salganicoff L (1973) Isolation
and properties of human platelet mitochon-
dria. Blood 42(6):913–918
18. Lo
´pez-Mediavilla C, Orfao A, San Miguel J,
Medina JM (1992) Developmental changes in
rat liver mitochondrial populations analyzed by
flow cytometry. Exp Cell Res 203(1):134–140.
https://doi.org/10.1016/0014-4827(92)
90048-D
19. Perret B, Chap HJ, Douste-Blazy L (1979)
Asymmetric distribution of arachidonic acid in
the plasma membrane of human platelets a
determination using purified phospholipases
and a rapid method for membrane isolation.
Biochim Biophys Acta Biomembr 556
(3):434–446. https://doi.org/10.1016/
0005-2736(79)90131-7
66 Jacob L. Le
´ger et al.
Chapter 5
Mechanical Permeabilization as a New Method
for Assessment of Mitochondrial Function in Insect Tissues
Alessandro Gaviraghi, Yan Aveiro, Stephanie S. Carvalho,
Rodiesley S. Rosa, Matheus P. Oliveira, and Marcus F. Oliveira
Abstract
Respirometry analysis is an effective technique to assess mitochondrial physiology. Insects are valuable
biochemical models to understand metabolism and human diseases. Insect flight muscle and brain have
been extensively used to explore mitochondrial function due to dissection feasibility and the low sample
effort to allow oxygen consumption measurements. However, adequate plasma membrane permeabiliza-
tion is required for substrates/modulators to reach mitochondria. Here, we describe a new method for
study of mitochondrial physiology in insect tissues based on mechanical permeabilization as a fast and
reliable method that do not require the use of detergents for chemical permeabilization of plasma
membrane, while preserves mitochondrial integrity.
Key words Metabolism, Bioenergetics, Animal models, Insect, Mitochondria, Respiration
1 Introduction
In recent years, insects are increasing their importance as organism
models for the study of human pathologies, evolutionary mechan-
isms, and drug discovery. This is due to the conservation between
insects and mammals of various key mechanisms such as signaling
pathways [1], energy metabolism [2–4], and organs structural and
functional components [5]. Furthermore, insects are cost-effective,
easy to rear, and have a relatively short-life cycle, which makes them
easy to use in aging studies. The main insect used as organism
model is the Drosophila melanogaster which has been used for
more than 100 years in biomedical science. Just to highlight the
importance of this organism, six Nobel prizes were awarded for
discoveries in which Drosophila was used as a study model [6–
11]. Indeed, our understanding of key mechanisms involved in
many physiological and pathological processes owes to the use of
Drosophila as an organism model [12–17].
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_5,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
67
The energy demand required for muscle contraction or physi-
ological function is provided by the production of ATP through
oxidative phosphorylation in mitochondria. This process occurs at
the inner mitochondrial membrane and involves protein complexes
which, through sequential redox reactions, generate a protonmo-
tive force (pmf) that is used by the ATP synthase complex for ATP
production. In fact, mitochondria play a fundamental role in cellu-
lar metabolism using the energy derived from oxidation of carbo-
hydrates, lipids, and amino acids to generate ATP. In this sense,
high-resolution respirometry is employed to assess mitochondrial
physiology and cellular metabolism. This technique provides pre-
cise and dynamic information on the metabolic flow (oxygen con-
sumption rate) concerning the transport of nutrients, oxidation
efficiency, and phosphorylation capacity (ATP production). High-
resolution respirometry represents a powerful tool for investigating
cellular bioenergetic state, identifying potential biomarkers, and
studying metabolic reprogramming in several diseases. Finally,
high-resolution respirometry can be used in the conditions when
amounts of cells or tissue are limiting.
In recent years, several approaches were developed and vali-
dated to study mitochondrial physiology using chemically permea-
bilized tissues and cells in different animal species [18–25]. These
methods allow bypassing the possible limitations due to the process
of isolation of mitochondria such as loss of mitochondrial morphol-
ogy [26], low recovery, and potential selection of distinct mito-
chondrial subpopulations [27]. The methods employed for cell or
tissue permeabilization are mainly based on the treatment of the
samples with either digitonin or saponin, two weak nonionic deter-
gents. These chemicals allow the selective permeabilization of the
plasma membrane by forming complexes with the cholesterol pres-
ent in the plasma membrane. This is due to the nonhomogeneous
distribution of the cholesterol in the different types of cellular
membranes. Indeed, the cholesterol content in the plasma mem-
brane of mammals is usually higher compared to organelles mem-
branes (>20% of weight mass for the plasma membrane vs. ~3% in
mitochondrial membranes) [28–31]. These chemical permeabiliza-
tion methods were developed for mammalians cells and subse-
quently used for other organisms, including insects. However, the
lipid composition of the plasma membrane is remarkably different
between mammals and insects. Particularly, the cholesterol content
in insect’s plasma membrane is ten times lower than their mamma-
lian counterparts, while the phosphatidylethanolamine/phosphati-
dylcholine ratio is four times higher [32,33]. These differences in
the lipid composition determine the plasma membrane properties,
and the cholesterol content plays a major role in membrane fluidity,
permeability, and hydrophobicity [29,31]. The very low content of
68 Alessandro Gaviraghi et al.
cholesterol in insect plasma membranes is due to the inability of this
class to synthesize cholesterol de novo, which is essentially obtained
through the diet to supply their physiological requirements
[34]. These unique insect membrane features have prompted us
to develop and validate new approaches for insect plasma mem-
brane permeabilization to allow studies of mitochondrial physiol-
ogy in these organisms [35,36]. In this chapter, we describe and
validate two reliable protocols to assess mitochondrial physiology in
situ using mechanically permeabilized tissues of the major arbovirus
vector Aedes aegypti mosquitoes and D. melanogaster fruit flies in
combination with high-resolution respirometry. Both protocols
were based on a previously published method designed for
A. aegypti flight muscle [36].
2 Materials
2.1 Solutions 1. Respiration buffer:120 mM KCl, 5 mM KH
2
PO
4
,3mM
HEPES, 1 mM EGTA, 1.5 mM MgCl
2
, and 0.2% fatty acid–
free bovine serum albumin, adjusted to pH 7.2. Prepare in
ultrapure double-distilled water, divide in aliquots, and store
at 20 C for several months.
2.2 Substrates 1. ADP: Prepare 0.5 M of ADP in ultrapure double-distilled
water. Neutralize to pH 7.2 with NaOH, check pH, divide in
0.2 mL aliquots, and store at 80 C.
2. Ascorbate: Prepare 0.8 M of ascorbate sodium salt in ultrapure
double-distilled water. To avoid autooxidation, the solution
should always be used fresh (see Note 1).
3. Cytochrome c: Prepare 2 mM of cytochrome cin ultrapure
double-distilled water. The solution can be stored as 0.2 mL
aliquots at 20 C(see Note 2).
4. G3P: Prepare 1 M of sn-glycerol 3-phosphate bis(cyclohexy-
lammonium) salt in ultrapure double-distilled water. G3P solu-
tion can be stored as 0.2 mL aliquots at 20 C(see Note 3).
5. Glutamate: Prepare 1 M of L-glutamic acid in ultrapure double-
distilled water. Neutralize to pH 7.2 with NaOH, check pH,
divide in 0.5 mL aliquots, and store at 20 C.
6. Malate: Prepare 0.5 M of L-malic acid in ultrapure double-
distilled water. Neutralize to pH 7.2 with NaOH, check pH,
divide in 0.5 mL aliquots, and store at 20 C.
7. Proline: Prepare 0.5 M of L-proline in ultrapure double-
distilled water. The solution can be stored as 0.5 mL aliquots
at 20 C.
Assessing Mitochondrial Function in Tissues 69
8. Pyruvate: Prepare 1 M of sodium pyruvate in ultrapure double-
distilled water; the solution should be used always fresh (see
Note 4).
9. Succinate: Prepare 0.5 M of sodium succinate dibasic in ultra-
pure double-distilled water. Succinate solution can be stored as
0.5 mL aliquots at 20 C.
10. TMPD: Prepare 0.2 M of N,N,N0,N0-Tetramethyl-p-phenyle-
nediamine dihydrochloride in ultrapure double-distilled
water. To avoid autooxidation, the solution should be fresh
(see Note 5).
2.3 Uncoupler 1. FCCP: Prepare 1 mM of carbonyl cyanide p-(trifluoro-meth-
oxy) phenyl-hydrazone in absolute ethanol. The stock solution
can be stored as 0.1 mL aliquots at 20 C(see Note 6).
2.4 Inhibitors 1. Antimycin A: Prepare a stock solution at 5 mM in absolute
ethanol, divide in 0.2 mL aliquots, and store at 20 C.
2. Carboxyatractyloside: Prepare 5 mM of carboxyatractyloside
potassium salt in ultrapure double-distilled water, divide in
0.2 mL aliquots, and store at 20 C.
3. KCN: Prepare 1 M of potassium cyanide in ultrapure double-
distilled water. The solution should be fresh (see Note 7).
4. Oligomycin: Prepare a stock solution at 4 mg/mL (5 mM) in
absolute ethanol, divide in 0.2 mL aliquots, and store at
20 C.
5. Rotenone: Prepare a stock solution at 1 mM in absolute etha-
nol, divide in 0.2 mL aliquots, and store at 20 C.
6. Sodium azide: Prepare a stock solution at 4 M in ultrapure
double-distilled water, divide in 0.5 mL aliquots, and store at
20 C. For a complete inhibition of cytochrome coxidase
(COX) activity, use 100 mM as final concentration.
3 Methods
3.1 Insects Aedes aegypti (Aedes-RIO strain [44]), obtained from F8–F10
eggs, were raised under a photoperiod of 12 h light/dark at
28 C and 70–80% relative humidity. Larvae were fed on a diet
consisting of commercial dog chow, and the adults were maintained
in a plastic cage in a 1:2 sex-ratio and fed ad libitum on cotton pads
soaked with 10% (w/v) sucrose solution. Females 1–4 days after the
emergence were utilized in the experiments.
Drosophila melanogaster (w
1118
strain) were raised under a
photoperiod of 12 h light/dark in a B.O.D incubator at 25 C
with a variation of 0.5 C, at normal levels of humidity (50% RH).
Flies were fed on a diet consisting of 1% agar, 2.7% yeast extract,
70 Alessandro Gaviraghi et al.
1.7% sucrose, 3.5% dextrose, and 8.3% cornmeal in a w/v rela-
tion with distilled water. After the culture medium is cooked,
1.1% propionic acid and 1.5% methylparaben v/v were added to
the hot solution. Culture media changes were done twice a week.
Males 1–6 days after hatching were utilized in the experiments.
3.2 Sample
Preparation
3.2.1 A. aegypti Head
Dissection (Fig. 1a–c)
1. Using an aspirator, collect the mosquitoes and transfer them to
a closed container (see Note 8).
2. Since the cold anesthetize the mosquitoes, put the container on
ice and wait 5–10 min until the mosquitoes stop moving.
3. Mosquitoes were then transferred to glass petri dishes on ice for
the dissection (see Note 9).
4. Use fine-tipped forceps to immobilize the mosquito by holding
the thorax gently against the glass petri dish.
5. Separate the head from the rest of the mosquito body severing
the membrane that joins the head to the thorax supported by
two laterals cervical sclerites [37] with a second fine-tipped
forceps.
Fig. 1 Simplified scheme employed for insect tissues dissection. A. aegypti head
dissection: (a) The mosquito was anesthetized by chilling on ice onto a precooled
petri glass dish. (b) Forceps were used to immobilize the mosquito and, using
another forceps, the head is separated by severing the membrane that joins the
head to the thorax. (c) The heads are grouped in a microcentrifuge tube in the ice
up to 15 units and then transferred to the respirometer chamber. D. melanoga-
ster thorax dissection: (d) Paws and wings were gently removed. (e) Immobilize
the fly using fine-tipped forceps and remove the head using a scalpel. (f) The
abdomen was also removed using a tweezers and a scalpel. (g) The two
thoraxes were immediately transferred to the respirometer chamber for
mechanical permeabilization
Assessing Mitochondrial Function in Tissues 71
6. Group 15 heads in a microcentrifuge tube on ice and keep
on cold.
7. Transfer the 15 heads to the Oroboros respirometer chamber
(Oxygraph-2 k, O2k Oroboros Instruments, Innsbruck, Aus-
tria) filled with 2 mL of “respiration buffer” described above.
3.2.2 D. melanogaster
Thorax Dissection
(Fig. 1d–g)
1. Flies (1–6 days old after hatching) were anesthetized for 8 s in
CO
2
and transferred to a CO
2
fly pad. Males were separated
and transferred in a vial immersed on ice for 1.5 min.
2. Flies were transferred to a precooled petri dish in the dorsal
position and dissected using scalpel and fine-tipped forceps.
3. Using fine-tipped forceps immobilize the fly by gently holding
the abdomen against the surface and remove the head by using
a scalpel.
4. Place the fly in lateral decubitus position and remove the abdo-
men from thorax by making a cut at the level of the
mediotergite.
5. Immediately transfer two thoraces to the Oroboros respirome-
ter chamber (Oxygraph-2 k, O2k Oroboros Instruments,
Innsbruck, Austria) filled with 2 mL of the “respiration buffer.”
3.3 Mechanical
Permeabilization
and Respirometry
Measurements
Protocols
3.3.1 A. aegypti Heads
1. Place the heads stored in microcentrifuge tubes into the Oro-
boros chambers by shedding the microcentrifuge tubes content
or transferring with forceps, if the samples get stick to the tube
bottom.
2. Before closing the chamber, insert a Hamilton syringe in the
stopper capillary (preferably 50 μL because the needle is larger)
to avoid the exit of the tissues. Then close the chamber.
3. The heads tend to cluster once in the chamber with respiration
buffer. Pull the stopper up and push down very carefully in
order to disaggregate them. Make sure that the heads’ cluster is
apart in small groups to facilitate the permeabilization and
avoid formation of air bubbles (see Note 10).
4. Mechanical permeabilization is carried out inside the respirom-
eter chamber by stirring at 750 rpm and by raising and lower-
ing the stopper until the solution appears turbid and the heads
are no longer observed (about 10 min).
5. Respirometry analysis is carried out at 27.5 C following the
protocol established by our group [36].
6. Start the substrate–uncoupler–inhibitor titration (SUIT) pro-
tocol by injecting 20 μL of a 1 M pyruvate (10 mM final
concentration) in each respirometer chamber and wait for sta-
bilization of oxygen consumption rates. Then add 40 μLofa
0.5 M Proline (10 mM final concentration) and wait for stabi-
lization of oxygen consumption rates (red trace, see Fig. 2).
72 Alessandro Gaviraghi et al.
7. Inject 4 μL of a 0.5 M ADP (1 mM final concentration) to
stimulate oxygen consumption linked to ATP synthesis (see
Note 11).
8. Make stepwise titrations of carbonyl cyanide
p-(trifluoromethoxy) phenylhydrazone (FCCP) from 0.2 to
1.5 μM (0.4–3 μL of a 1 mM solution) to induce maximum
uncoupled respiration.
Fig. 2 Simplified scheme of mechanical permeabilization and respirometry analyses for A. aegypti head. (a)
The heads were transferred to the respirometer chamber. (b) Mechanical permeabilization is performed by
magnetic stirring at 750 rpm raising and lowering the stopper of the respirometer for about 10 min. (c) The
permeabilization was obtained when the solution appears turbid and the heads are no longer observed. (d)
Start the substrate–uncoupler–inhibitor titration (SUIT) protocol. (e) Typical traces of oxygen consumption
rates (red line) and concentration (blue line) from mechanically permeabilized heads using 10 mM pyruvate
plus proline as substrate
Assessing Mitochondrial Function in Tissues 73
9. To determine the contribution of complex I on the electron
flow, add 1 μL of a 1 mM rotenone (a complex I inhibitor) in
order to reach a final concentration of 0.5 μM.
10. Add 1 μL of a 5 mM antimycin A (AA) (2.5 μM final concen-
tration), to inhibit complex III and to fully block oxygen
consumption linked to mitochondrial electron transfer system.
11. Raise the stopper using the Oroboros Stopper-Spacer to form
an air bubble in the chamber and inject an oxygen-enriched
gaseous mixture (70% O
2
and 30% N
2
mol/mol) using a
60 mL syringe to reach an oxygen concentration close to
500 nmol/mL (see Note 12).
12. Assess cytochrome coxidase (COX) activity by adding 5 μLofa
0.8 M ascorbate (2 mM final concentration) and 5 μLofa
0.2 M N,N,N0,N0-Tetramethyl-p-phenylenediamine dihy-
drochloride (0.5 mM final concentration) (TMPD), as an
electron-donor regenerating system (see Note 13).
13. Add 4 μL of 1 M KCN (2 mM final concentration) to inhibit
COX activity (see Note 14).
3.3.2 D. melanogaster
Thorax
1. Transfer the two thoraces to the Oroboros chambers using a
brush.
2. Before closing the chamber, insert a Hamilton syringe in the
stopper capillary (preferably 50 μL because the needle is larger)
to avoid the exit of the tissues. Then close the chamber. Make
sure to eliminate all air bubbles in the solution (see Note 10).
3. Permeabilize mechanically the thoraces by stirring at 900 rpm
for 7.5 min (other timepoints were tested and did not produce
satisfactory results).
4. Reduce the stirring speed to 750 rpm and start the respirome-
try analysis at 27.5 C as previously established by our group
[36](see Note 15).
5. After starting the analysis, raise the stopper using the Oroboros
Stopper-Spacer to form an air bubble in the chamber. Inject an
oxygen-enriched gaseous mixture (70% O
2
and 30% N
2
mol/-
mol) using a 60 mL syringe to reach an oxygen concentration
close to 500 nmol/mL (see Note 16).
6. Start the substrate–uncoupler–inhibitor titration (SUIT) pro-
tocol by injecting all substrates able to stimulate complex I, II,
and glycerol phosphate dehydrogenase (red trace, see Fig. 3)as
following: 20 μL of a 1 M pyruvate (10 mM final concentra-
tion), 20 μL of a 1 M glutamate (10 mM final concentration),
20 μL of a 0.5 M malate (5 mM final concentration), and wait
for stabilization of oxygen consumption rates. Then add 10 μL
of a 0.5 M succinate (5 mM final concentration), wait for
stabilization of oxygen consumption rates, and add 30 μLofa
1 M glycerol-3-phosphate (15 mM final concentration).
74 Alessandro Gaviraghi et al.
7. Inject 20 μL of a 0.5 M ADP (5 mM final concentration) to
stimulate oxygen consumption linked to ATP synthesis (see
Note 11).
8. Add 10 μL of a 2 mM cytochrome c(10 μM final concentra-
tion) (see Note 17).
9. Make stepwise titrations of FCCP from 0.2 to 1.5 μM
(0.4–3 μL of a 1 mM solution) to induce maximum uncoupled
respiration.
Fig. 3 Simplified scheme of mechanical permeabilization and respirometry analyses for D. melanogaster
thoraces. (a) Two thoraces were transferred to the respirometer chamber. (b) Mechanical permeabilization
was performed by magnetic stirring at 900 rpm for 7.5 min, then the stirring speed was reduced to 750 rpm.
(c) An oxygen-enriched gaseous mixture (70% O
2
and 30% N
2
mol/mol) was injected through the stopper of
each chamber using a 60 mL syringe to reach an oxygen concentration close to 500 nmol/mL. (d) Start the
substrate–uncoupler–inhibitor titration (SUIT) protocol. (e) Typical traces of oxygen consumption rates (red
line) and concentration (blue line) from mechanically permeabilized thoraces using all substrates able to
stimulate complex I, II, and glycerol phosphate dehydrogenase
Assessing Mitochondrial Function in Tissues 75
10. Add 1 μL of a 5 mM antimycin a (AA) (2.5 μM final concen-
tration) to inhibit complex III and to block oxygen consump-
tion linked to mitochondrial electron transfer system.
11. Raise the stopper using the Oroboros Stopper-Spacer to form
an air bubble in the chamber and inject an oxygen-enriched
gaseous mixture (70% O
2
and 30% N
2
mol/mol) using a
60 mL syringe to reach an oxygen concentration close to
500 nmol/mL (see Note 12).
12. Assess COX activity by adding 5 μL of a 0.8 M ascorbate
(2 mM final concentration) and 5 μL of a 0.2 M N,N,N0,N0-
Tetramethyl-p-phenylenediamine dihydrochloride (0.5 mM
final concentration) (TMPD), as an electron-donor regenerat-
ing system (see Note 13).
13. Add 4 μL of 1 M KCN (2 mM final concentration) to inhibit
COX activity (see Note 14).
3.4 Data
Interpretation
Starting from the oxygen consumption data obtained from the
respirometry analysis, it is possible to calculate different respiratory
states that characterize the mitochondrial metabolism (Fig. 4).
1. Leak: This metabolic state represents the oxygen consumption
induced only by the addition of the oxidizable substrates. The
respiratory rates of the leak state are essentially controlled by
the magnitude of proton leak through the inner mitochondrial
membrane. Higher the proton leak, higher the leak respiratory
Fig. 4 Schematic representation to calculate the mitochondrial metabolic states. Simplified scheme of the
typical traces of oxygen consumption (red line) and concentration (blue line) obtained from respirometry
analyses using the Oroboros system. Each mitochondrial metabolic state calculated is depicted in orange fonts
76 Alessandro Gaviraghi et al.
rates [38]. Leak is calculated by subtracting from the oxygen
consumption obtained after the addition of all the oxidizable
substrates from that after adding AA.
2. OXPHOS: This metabolic state represents the oxygen con-
sumption coupled to the ATP production and is obtained
after the addition of ADP. OXPHOS is calculated by subtract-
ing the substrates (leak) oxygen consumption rates from those
after ADP addition.
3. ETS (electron transfer system): This metabolic state reflects the
maximum oxygen consumption rates linked to mitochondrial
electron transport system (ETS) without the effect of proto-
nmotive force. To asses this, a proton ionophore
(or uncoupler) is added to provide the maximal respiratory
rates, in our case we used FCCP. The ETS state is calculated
by subtracting the oxygen consumption rates obtained after
FCCP from those obtained after adding AA.
4. Spare capacity: This metabolic state represents how much oxy-
gen consumption can increase in response to increased energy
demand. Thus, it represents the maximum capacity of the
electron transport system. Spare capacity is calculated by sub-
tracting the oxygen consumption rates obtained after FCCP
from those obtained after adding ADP.
5. ROX (residual oxygen consumption): This metabolic state
represents the oxygen consumption rates linked to cellular
processes other than respiration, such as oxygenase activities,
reactive oxygen species production. ROX is determined by
quantifying the oxygen consumption rates obtained after addi-
tion of AA.
6. Cytochrome coxidase (COX) activity: Represents the oxygen
consumption rates specifically due to COX activity and is cal-
culated by subtracting the oxygen consumption rates obtained
after addition of ascorbate and TMPD from those after KCN
(or sodium azide) addition.
7. Normalization of oxygen consumption data: This is a critical
aspect and several parameters have been used in the literature to
normalize respirometry data including tissue wet weight, pro-
tein content, citrate synthase activity, mitochondrial DNA, and
others. Each normalizer has its limitations, but we think that
using total protein content is a reliable solution. A group of
insects belonging to the same cohort are used to quantify the
protein content. For this purpose, the insects were dissected,
and tissues homogenized in a ground glass potter in hypotonic
buffer (25 mM potassium phosphate and 5 mM MgCl
2
,
pH 7.2). Subsequently, the solution was centrifuged at
1500 gfor 10 min, the supernatant was recovered, and the
protein content was determined by the Lowry method [39](see
Note 18).
Assessing Mitochondrial Function in Tissues 77
3.5 Protocols
Validation
3.5.1 A. aegypti Heads
The protocol designed here to assess respiratory rates in A. aegypti
heads was suitable for quantitative identification of mitochondrial
metabolic states (Fig. 5a,b) and represents an evolution of the
method developed for the flight muscle [36]. This was accom-
plished by performing few changes in the permeabilization proto-
col to make it optimal for the analysis of the oxygen consumption of
mosquito head. To validate the method described here, the follow-
ing set of experiments was designed to confirm the integrity of
inner mitochondrial membrane and OXPHOS coupling after
mechanical permeabilization.
1. To assess the OXPHOS coupling, add 1 μL of a 4 mg/mL
oligomycin (2 μg/mL final concentration), a powerful inhibi-
tor of F1Fo ATP synthase complex, after the addition of ADP.
AC
5
[pmol/s/µg protein]
4
3
O2 Consumption
[pmol/s/µg protein]
O2 Consumption
2
1
0
B
[pmol/s/µg protein]
4
3
O2 Consumption
2
1
0
Pyr+pro
Leak
OXPHOS
ETS
Spare capacity
ROX
compl IV
Leak
OXPHOS
ETS
Spare capacity
ROX
ADP
FCCP
Rot
AA
TMPD
KCN
PGM
Succinate
G3P
ADP
Cyto c
FCCP
AA
12
8
4
0
D
[pmol/s/µg protein]
O2 Consumption
12
8
4
0
HEAD
A. aegypti
THORAX
D. melanogaster
Fig. 5 Quantitative comparison of respiratory rates and mitochondrial metabolic states obtained in A. aegypti
and D. melanogaster samples. (a) and (c) Quantification of respiratory rates obtained in respirometry analyses
using Oroboros. (b) and (d) Mitochondrial metabolic states calculated from oxygen consumption data as
reported above. Data are expressed as mean O
2
consumption rates (pmoles/s/μg protein) standard error of
the mean (SEM) of at least four different experiments
78 Alessandro Gaviraghi et al.
2. To assess the integrity of the inner mitochondrial membrane,
inject 2 μL of a 5 mM carboxyatractyloside (CAT) (5 μM final
concentration), a highly selective and potent inhibitor of ade-
nine nucleotide translocator (ANT) after the ADP
addition [40].
Figure 6shows that respiratory rates in mechanically permea-
bilized heads provided by Pyr + Pro under phosphorylating condi-
tions (1 mM ADP) were reduced by 92% after oligomycin (Fig. 6a)
or CAT (Fig. 6b) treatment. This indicates that mitochondrial
inner membrane in mechanically permeabilized heads is intact,
conferring high OXPHOS coupling when using Pyr + Pro as
substrates.
3.5.2 D. melanogaster
Thorax
The protocol designed here to assess respiratory rates in
D. melanogaster flight muscle was suitable to quantitatively identify
distinct mitochondrial metabolic states (Fig. 5c, d).
250 4
3
2
1
0
200
150
O2 Concentration
[nmol/ml]
O2 Consumption
pmol O
2
/s/µg protein
100
50
0
05
Pyr Pro ADP
ADP
Oligo
Oligo AA
10
Time [min]
15 20
250
(b)
(a)
4
3
2
1
0
200
150
O2 Concentration
[nmol/ml]
O2 Consumption
pmol O
2
/s/µg protein
100
50
0
05
Pyr Pro
ADP
ADP CAT
CAT AA
10
Time [min]
15 20
Fig. 6 Intactness of mitochondrial inner membrane in mechanically permeabilized A. aegypti heads. Typical
traces of oxygen concentration (blue line) and consumption rates (red line) from mechanically permeabilized
heads in the presence of 5 μM carboxyatractyloside (CAT) (a)or2μg/mL oligomycin (oligo) (b). The strong
inhibition of respiratory rates after CAT and oligomycin injection demonstrates the intactness of the inner
mitochondrial membrane and the high degree of OXPHOS coupling. The addition of OXPHOS modulators
pyruvate (Pyr), proline (Pro), ADP, and antimycin A (AA) is represented with an arrow and their concentrations
were reported in the Methods section
Assessing Mitochondrial Function in Tissues 79
To validate the method described here, comparative respirom-
etry experiments were performed using a protocol established for
D. melanogaster [20,21,23,41], which use the combination of
chemical (using the detergent saponin) and mechanical permeabi-
lization of flight muscle (Fig. 7). The isolated thoraxes are chemi-
cally permeabilized in saponin (50 μg/mL), prepared in ice-cold
BIOPS (BIOPS: 2.77 mM CaK
2
EGTA, 7.23 mM K
2
EGTA,
5.77 mM Na
2
ATP, 6.56 mM MgCl
2
·6H
2
O, 20 mM taurine,
15 mM Na
2
phosphocreatine, 20 mM imidazole, 0.5 mM DTT,
50 mM MES hydrate), for 15 min in 80 rpm orbital shaker. The
thoraces were then washed for 2 times for 5 min each in 2 mL of
(a)
(b)
550
Pyr
Glu
Mal Succ G3P ADP
Cyt
c
FCCP AA
Asc
KCN
TMPD
O
2
O
2
440
330
220
110
0
0 5 10 15
5
4
3
2
1
0
Substrate
ADP
FCCP
Cyto c
20 25
Time [min]
30 35 40 45
18
15
12
9
6
3
0
O2 Concentration
[nmol/ml]
O2 Consumption
[pmol/s/µg protein]
O2 Consumption
pmol O
2
/s/µg protein
Fig. 7 Mechanical permeabilization provides respiratory rates undistinguishable from chemical permeabiliza-
tion. (a) Representative traces of oxygen consumption of mechanically permeabilized (red line) and chemically
(saponin) permeabilized (green line) D. melanogaster thoraces. Oxygen concentration levels are depicted as
blue traces. (b) Quantitative comparison of respiratory rates obtained using the two permeabilization methods
(white—mechanical, black—chemical). Oxygen consumption rates obtained using both methods were quite
similar, regardless the mitochondrial metabolic states. This indicates that mechanical permeabilization is
sufficient to allow free access for substrates and OXPHOS modulators in D. melanogaster thoraces, without
affecting mitochondrial structure and integrity. Data are expressed as mean O
2
consumption rates (pmoles/s/μ
g protein) standard error of the mean (SEM) of at least six different experiments. Comparisons between
groups were done by two-way ANOVA for repeated measurements and a posteriori Holm–Sidak post hoc test,
adjusted for multiple comparisons
80 Alessandro Gaviraghi et al.
ice-cold respiration buffer and then transferred to the Oroboros
chambers containing 2 mL of respiration buffer. The SUIT proto-
col used was the same as described in the Methods section.
3.6 Troubleshooting Figures 8and 9represent the two extreme cases of excessive and
insufficient permeabilization of the tissues, respectively.
550
440
330
220
Pyr
Glu
Mal
Succ G3P
ADP
Cyt c
Cyt c
110
0
20 25
8
6
4
2
0
Cytochrome c induced
oxygen consumption
Time [min]
O2 Concentration
[nmol/mL]
O2 Consumption
[pmol/s/µg protein]
Fig. 8 Representative traces of oxygen consumption rates associated to a correct (green line) and excessive
(red line) permeabilization. The green trace represents an optimal permeabilization condition (7.5 min at
900 rpm) where a large increase in oxygen consumption is observed after the addition of ADP while a slight
increase in oxygen consumption is obtained after the addition of cytochrome c. Instead, red trace represents
an excessive permeabilization condition purposely obtained by increasing the stirring time to 10 min at
900 rpm, where a large increase in oxygen consumption is observed after the addition of cytochrome c.A
significant increase in oxygen consumption after cytochrome cindicates the loss of mitochondrial integrity; in
fact the mitochondrial outer membrane is impermeable to cytochrome c. In addition, excessive permeabiliza-
tion (red trace) compromised inner mitochondrial membrane as ADP failed to increase respiratory rates as
should be expected
550
440
330
220
Pyr
Glu
Mal
Succ G3P
ADP Cyt c
Cyt cFCCP
FCCP
AA
AA
110
0
20 25 30 35 40
8
10
6
4
2
0
Time [min]
O2 Concentration
[nmol/ml]
O2 Concentration
[pmol/s/µg protein]
Fig. 9 Representative traces of oxygen consumption rate associated to a correct (green line) and insufficient
(orange line) permeabilization. The green trace represents an optimal permeabilization condition (7.5 min at
900 rpm) where a large increase in oxygen consumption is observed after the addition of ADP while a slight
increase in oxygen consumption is obtained after the addition of cytochrome c. Instead, orange trace
represents an insufficient permeabilization condition purposely obtained by reducing the stirring time to
3 min at 750 rpm, where only a slight increase in oxygen consumption was observed after the addition of
substrates, ADP, and FCCP
Assessing Mitochondrial Function in Tissues 81
4 Notes
1. Ascorbate is light sensitive, therefore protect the tubes with
aluminum foil.
2. Use only cytochrome cfrom equine heart.
3. Do not use the G3P racemic mixture because only the L
enantiomer is metabolized.
4. Pyruvate is an alpha keto acid which is unstable in aqueous
solution; therefore, the solution should be fresh.
5. TMPD is light sensitive, therefore protect the tubes with
aluminum foil.
6. FCCP is light sensitive, therefore the stock and working solu-
tions have to be kept in tubes protected with aluminum foil.
7. KCN is photosensitive and hygroscopic, therefore protect the
tubes with aluminum foil.
8. Be careful when handling with insects to avoid their escape.
9. Use a precooled petri dish to prevent contact with a warmer
surface that can awaken the mosquitoes.
10. The presence of air bubbles in the chamber during the analysis
causes alterations and instability in the detection of oxygen
consumption.
11. If a significant increase in oxygen consumption is not observed,
there may be problems with the permeabilization of the tissue.
12. COX activity is limited by low oxygen concentrations [42], so
an oxygen-enriched gas mixture is injected to avoid oxygen-
deprivation during measurements.
13. Ascorbate is added first to maintain TMPD in a reduced state.
14. TMPD autoxidizes in an oxygen and concentration-dependent
manner.
15. To validate the method described here, comparative respirom-
etry experiments were performed using a protocol established
for D. melanogaster studies [20,21,23,41], which use the
combination of chemical (using the detergent saponin) and
mechanical permeabilization of the tissue.
16. The oxygen tension is increased to avoid potential effects of
oxygen diffusion and electron transfer due to oxygen deficiency
during measurements [19]. Also, injections of oxygen-
enriched gaseous mixture were performed once the oxygen
concentration fell down below 150 nmol/mL into the
O2k-chamber.
17. This step allows the evaluation of the integrity of the outer
mitochondrial membrane. In fact, the external mitochondrial
82 Alessandro Gaviraghi et al.
membrane is impermeable to cytochrome c[43], so a signifi-
cant increase in oxygen consumption after its addition indicates
the loss of mitochondrial integrity.
18. The protein content cannot be measured at the end of the
respirometry analysis because the “respiration buffer” contains
albumin which, being a protein, distorts the dosage of the
protein content.
Acknowledgments
We would like to thank Mrs. Geane C. Braz for the excellent
technical assistance on maintenance of A. aegypti colony. This
study was financed in part by the Coordenac¸a
˜o de Aperfeic¸oamento
de Pessoal de Nı
´vel Superior—Brasil (CAPES)—Finance Code
001, by the Conselho Nacional de Desenvolvimento Cientifico
e Tecnolo
´gico (CNPq) [grant numbers 404153/2016-0 MFO,
and 483334/2013-8 AG], and the Fundac¸a
˜o Carlos Chagas
Filho de Amparo a
`Pesquisa do Estado do Rio de Janeiro (FAPERJ)
[grant numbers E-26/102.333/2013, E-26/203.043/2016, and
E-26/111.169/2011]. AG and MFO are CNPq fellows [grant
numbers 402409/2012-4 and 303044/2017-9] and from
PAPD-FAPERJ [grant number E-44/208702/2014]. The fun-
ders had no role in study design, data collection and analysis,
decision to publish, or preparation of the manuscript.
References
1. Stokes BA, Yadav S, Shokal U, Smith LC,
Eleftherianos I (2015) Bacterial and fungal pat-
tern recognition receptors in homologous
innate signaling pathways of insects and mam-
mals. Front Microbiol 6:19. https://doi.org/
10.3389/fmicb.2015.00019
2. Beenakkers AMT, Van der Horst DJ, Van Mar-
rewijk WJA (1984) Insect flight muscle metab-
olism. Insect Biochem 14:243–260. https://
doi.org/10.1016/0020-1790(84)90057-X
3. Rittschof CC, Schirmeier S (2018) Insect mod-
els of central nervous system energy metabo-
lism and its links to behavior. Glia
66:1160–1175. https://doi.org/10.1002/
glia.23235
4. Yang Y, Xu S, Xu J, Guo Y, Yang G (2014)
Adaptive evolution of mitochondrial energy
metabolism genes associated with increased
energy demand in flying insects. PLoS One 9:
e99120. https://doi.org/10.1371/journal.
pone.0099120
5. Baker KD, Thummel CS (2007) Diabetic lar-
vae and obese flies-emerging studies of
metabolism in Drosophila. Cell Metab
6:257–266. https://doi.org/10.1016/j.cmet.
2007.09.002
6. Morgan TH, Bridges CB (1916) Sex-linked
inheritance in Drosophila. Carnegie Institution
of Washington, Washington
7. Muller HJ (1928) The production of muta-
tions by X-rays. Proc Natl Acad Sci U S A
14:714–726. https://doi.org/10.1073/pnas.
14.9.714
8. The Nobel Prize in Physiology or Medicine
1995. In: NobelPrize.org.https://www.
nobelprize.org/prizes/medicine/1995/sum
mary/. Accessed 25 Nov 2019
9. Buck L, Axel R (1991) A novel multigene fam-
ily may encode odorant receptors: a molecular
basis for odor recognition. Cell 65:175–187.
https://doi.org/10.1016/0092-8674(91)
90418-x
10. Lemaitre B, Nicolas E, Michaut L, Reichhart
JM, Hoffmann JA (1996) The dorsoventral
regulatory gene cassette sp€
atzle/toll/cactus
controls the potent antifungal response in
Assessing Mitochondrial Function in Tissues 83
Drosophila adults. Cell 86:973–983. https://
doi.org/10.1016/s0092-8674(00)80172-5
11. The Nobel Prize in Physiology or Medicine
2017. In: NobelPrize.org.https://www.
nobelprize.org/prizes/medicine/2017/sum
mary/. Accessed 25 Nov 2019
12. Musselman LP, Fink JL, Narzinski K, Rama-
chandran PV, Hathiramani SS, Cagan RL, Bar-
anski TJ (2011) A high-sugar diet produces
obesity and insulin resistance in wild-type Dro-
sophila. Dis Model Mech 4:842–849. https://
doi.org/10.1242/dmm.007948
13. Ordonez DG, Lee MK, Feany MB (2018)
α-synuclein induces mitochondrial dysfunction
through spectrin and the actin cytoskeleton.
Neuron 97:108–124.e6. https://doi.org/10.
1016/j.neuron.2017.11.036
14. Capo F, Wilson A, Di Cara F (2019) The intes-
tine of Drosophila melanogaster: an emerging
versatile model system to study intestinal epi-
thelial homeostasis and host-microbial interac-
tions in humans. Microorganisms 7:E336.
https://doi.org/10.3390/
microorganisms7090336
15. Clark RI, Salazar A, Yamada R, Fitz-Gibbon S,
Morselli M, Alcaraz J, Rana A, Rera M,
Pellegrini M, Ja WW, Walker DW (2015) Dis-
tinct shifts in microbiota composition during
Drosophila aging impair intestinal function and
drive mortality. Cell Rep 12:1656–1667.
https://doi.org/10.1016/j.celrep.2015.08.
004
16. Lee K-S, Huh S, Lee S, Wu Z, Kim A-K, Kang
H-Y, Lu B (2018) Altered ER-mitochondria
contact impacts mitochondria calcium homeo-
stasis and contributes to neurodegeneration
in vivo in disease models. Proc Natl Acad Sci
U S A 115:E8844–E8853. https://doi.org/
10.1073/pnas.1721136115
17. McCommis KS, Hodges WT, Bricker DK,
Wisidagama DR, Compan V, Remedi MS,
Thummel CS, Finck BN (2016) An ancestral
role for the mitochondrial pyruvate carrier in
glucose-stimulated insulin secretion. Mol
Metab 5:602–614. https://doi.org/10.
1016/j.molmet.2016.06.016
18. Kuznetsov AV, Veksler V, Gellerich FN, Saks V,
Margreiter R, Kunz WS (2008) Analysis of
mitochondrial function in situ in permeabilized
muscle fibers, tissues and cells. Nat Protoc
3:965–976. https://doi.org/10.1038/nprot.
2008.61
19. Pesta D, Gnaiger E (2012) High-resolution
respirometry: OXPHOS protocols for human
cells and permeabilized fibers from small biop-
sies of human muscle. Methods Mol Biol
810:25–58. https://doi.org/10.1007/978-1-
61779-382-0_3
20. Pichaud N, Ballard JWO, Tanguay RM, Blier
PU (2011) Thermal sensitivity of mitochon-
drial functions in permeabilized muscle fibers
from two populations of Drosophila simulans
with divergent mitotypes. Am J Physiol Regul
Integr Comp Physiol 301:R48–R59. https://
doi.org/10.1152/ajpregu.00542.2010
21. Pichaud N, Ballard JWO, Tanguay RM, Blier
PU (2013) Mitochondrial haplotype diver-
gences affect specific temperature sensitivity of
mitochondrial respiration. J Bioenerg Bio-
membr 45:25–35. https://doi.org/10.1007/
s10863-012-9473-9
22. Saks VA, Veksler VI, Kuznetsov AV, Kay L,
Sikk P, Tiivel T, Tranqui L, Olivares J,
Winkler K, Wiedemann F, Kunz WS (1998)
Permeabilized cell and skinned fiber techniques
in studies of mitochondrial function in vivo.
Mol Cell Biochem 184:81–100
23. Simard CJ, Pelletier G, Boudreau LH, Hebert-
Chatelain E, Pichaud N (2018) Measurement
of mitochondrial oxygen consumption in per-
meabilized fibers of drosophila using minimal
amounts of tissue. J Vis Exp (134):57376.
https://doi.org/10.3791/57376
24. Veksler VI, Kuznetsov AV, Sharov VG, Kapelko
VI, Saks VA (1987) Mitochondrial respiratory
parameters in cardiac tissue: a novel method of
assessment by using saponin-skinned fibers.
Biochim Biophys Acta 892:191–196
25. Votion D-M, Gnaiger E, Lemieux H,
Mouithys-Mickalad A, Serteyn D (2012) Phys-
ical fitness and mitochondrial respiratory
capacity in horse skeletal muscle. PLoS One 7:
e34890. https://doi.org/10.1371/journal.
pone.0034890
26. Picard M, Taivassalo T, Ritchie D, Wright KJ,
Thomas MM, Romestaing C, Hepple RT
(2011) Mitochondrial structure and function
are disrupted by standard isolation methods.
PLoS One 6:e18317. https://doi.org/10.
1371/journal.pone.0018317
27. Piper HM, Sezer O, Schleyer M, Schwartz P,
Hu¨tter JF, Spieckermann PG (1985) Develop-
ment of ischemia-induced damage in defined
mitochondrial subpopulations. J Mol Cell Car-
diol 17:885–896
28. Jamieson GA, Robinson DM (2014) Mamma-
lian cell membranes: volume 2: the diversity of
membranes. Elsevier, London
29. Raffy S, Teissie
´J (1999) Control of lipid mem-
brane stability by cholesterol content. Biophys
J 76:2072–2080. https://doi.org/10.1016/
S0006-3495(99)77363-7
30. Korn ED (1969) Cell membranes: structure
and synthesis. Annu Rev Biochem
84 Alessandro Gaviraghi et al.
38:263–288. https://doi.org/10.1146/
annurev.bi.38.070169.001403
31. Subczynski WK, Pasenkiewicz-Gierula M,
Widomska J, Mainali L, Raguz M (2017)
High cholesterol/low cholesterol: effects in
biological membranes: a review. Cell Biochem
Biophys 75:369–385. https://doi.org/10.
1007/s12013-017-0792-7
32. Gimpl G, Klein U, Reil€
ander H, Fahrenholz F
(1995) Expression of the human oxytocin
receptor in baculovirus-infected insect cells:
high-affinity binding is induced by a
cholesterol-cyclodextrin complex. Biochemis-
try 34:13794–13801
33. Dawaliby R, Trubbia C, Delporte C, Noyon C,
Ruysschaert J-M, Van Antwerpen P, Govaerts
C (2016) Phosphatidylethanolamine is a key
regulator of membrane fluidity in eukaryotic
cells. J Biol Chem 291:3658–3667. https://
doi.org/10.1074/jbc.M115.706523
34. Krebs KC, Lan Q (2003) Isolation and expres-
sion of a sterol carrier protein-2 gene from the
yellow fever mosquito, Aedes aegypti. Insect
Mol Biol 12:51–60. https://doi.org/10.
1046/j.1365-2583.2003.00386.x
35. Teulier L, Weber J-M, Crevier J, Darveau C-A
(2016) Proline as a fuel for insect flight:
enhancing carbohydrate oxidation in hyme-
nopterans. Proc Biol Sci 283:20160333.
https://doi.org/10.1098/rspb.2016.0333
36. Gaviraghi A, Oliveira MF (2019) A method for
assessing mitochondrial physiology using
mechanically permeabilized flight muscle of
Aedes aegypti mosquitoes. Anal Biochem
576:33–41. https://doi.org/10.1016/j.ab.
2019.04.005
37. Snodgrass RE (1959) The anatomical life of
the mosquito. Smithson Misc Coll 139, 87pp
38. Nicholls DG, Ferguson SJ (2013) Bioenerget-
ics, 4th edn. Academic Press, Boston, pp i–iii
39. Lowry OH, Rosebrough NJ, Farr AL, Randall
RJ (1951) Protein measurement with the Folin
phenol reagent. J Biol Chem 193:265–275
40. Vignais PV, Vignais PM, Defaye G (1973)
Adenosine diphosphate translocation in mito-
chondria. Nature of the receptor site for car-
boxyatractyloside (gummiferin). Biochemistry
12:1508–1519
41. Rana A, Oliveira MP, Khamoui AV, Aparicio R,
Rera M, Rossiter HB, Walker DW (2017) Pro-
moting Drp1-mediated mitochondrial fission
in midlife prolongs healthy lifespan of Dro-
sophila melanogaster. Nat Commun 8:448.
https://doi.org/10.1038/s41467-017-
00525-4
42. Chandel NS, Budinger GR, Schumacker PT
(1996) Molecular oxygen modulates cyto-
chrome c oxidase function. J Biol Chem
271:18672–18677
43. Kuznetsov AV, Schneeberger S, Seiler R,
Brandacher G, Mark W, Steurer W, Saks V,
Usson Y, Margreiter R, Gnaiger E (2004)
Mitochondrial defects and heterogeneous
cytochrome c release after cardiac cold ischemia
and reperfusion. Am J Physiol Heart Circ
Physiol 286:H1633–H1641. https://doi.
org/10.1152/ajpheart.00701.2003
44. Santos CR, Rodovalho CM, Jablonka W, Mar-
tins AJ, Lima JBP, Dias LS, Silva-Neto MAC,
Atella GC (2020) Insecticide resistance, fitness
and susceptibility to Zika infection of an
interbred Aedes aegypti population from Rio
de Janeiro, Brazil. Parasit Vectors 13:293.
https://doi.org/10.1186/s13071-020-
04166-3.
Assessing Mitochondrial Function in Tissues 85
Chapter 6
Analysis of Mitochondrial Retrograde Signaling in Yeast
Model Systems
Nicoletta Guaragnella, Mas
ˇaZ
ˇdralevic
´, Zdena Palkova
´,
and Sergio Giannattasio
Abstract
Mitochondrial retrograde signaling is a mitochondria-to-nucleus communication pathway, conserved from
yeast to humans, by which dysfunctional mitochondria relay signals that lead to cell stress adaptation in
physiopathological conditions via changes in nuclear gene expression. The most comprehensive picture of
components and regulation of retrograde signaling has been obtained in Saccharomyces cerevisiae, where
retrograde-target gene expression is regulated by RTG genes. In this chapter, we describe methods to
measure mitochondrial retrograde pathway activation at the level of mRNA and protein products in yeast
model systems, including cell suspensions and microcolonies. In particular, we will focus on three major
procedures: mRNA levels of RTG-target genes, such as those encoding for peroxisomal citrate synthase
(CIT2), aconitase, and NAD
+
-specific isocitrate dehydrogenase subunit 1 by real-time PCR; expression
analysis of CIT2-gene protein product (Cit2p-GFP) by Western blot and fluorescence microscopy; the
phosphorylation status of transcriptional factor Rtg1/3p which controls RTG-target gene transcription.
Key words Mitochondrial retrograde pathway, Yeast, RTG genes, CIT2,ACO1,IDH1
1 Introduction
Cell homeostasis can be threatened by environmental stress and
nutrient availability. The type and level of the insult are strong
determinants of cell stress response, but the adaptive capacity of a
cell will ultimately determine its fate. Cells can respond to stress
either activating death pathways or using different pro-survival
strategies to prevent inappropriate or premature regulated cell
death (RCD) [1].
Mitochondria are at the crossroad of the complex regulatory
network integrating pro-life and pro-death signals, but while their
role in cell death processes is well established [2], how they are
implicated in cell stress response and adaptation is only now becom-
ing clear [3]. The best-characterized mechanism of cell response to
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_6,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
87
mitochondrial dysfunction is the RTG pathway in yeast Saccharo-
myces cerevisiae. RTG pathway activation leads to a reconfiguration
in the expression of a subset of nuclear genes enabling accommo-
dation to changes in the mitochondrial state, such as mitochondrial
DNA depletion. The molecular details of regulation of RTG-target
gene expression have been elucidated [4]. Mitochondrial dysfunc-
tion can also alter cytosolic proteostasis. Mitochondrial precursor
protein overaccumulation stress (mPOS), typically induced by pro-
tein misfolding in the mitochondrial inner membrane, can ulti-
mately cause cell death [4,5]. Specific retrograde transcriptional
changes are activated to mitigate mPOS stress and suppress mPOS-
induced cell death ([6] and ref. therein, [7]).
The prototypical retrograde-target gene is CIT2, which
encodes the peroxisomal isoform of citrate synthase active in the
glyoxylate cycle. CIT2 expression is largely increased in cells with
compromised mitochondrial functions, including those lacking
mitochondrial DNA (ρ
0
)[8]. Both basal and upregulated expres-
sion of CIT2 is mainly dependent on RTG genes, encoding regu-
latory proteins central to yeast retrograde signaling (Fig. 1) (for
refs. see [9]). Rtg1p and Rtg3p are basic helix-loop-helix/leucine
zipper (bHLH/Zip) transcription factors that interact as a hetero-
dimer to bind to target sites called R-boxes located in the promoter
region of RTG-target genes [10]. Rtg3p is a phosphoprotein,
localized with Rtg1p to the cytoplasm and phosphorylated on
multiple sites when the RTG pathway is off. Activation of the
pathway correlates with Rtg1/3p partial dephosphorylation and
translocation to the nucleus. The most proximal sensor to mito-
chondrial dysfunction is the positive regulator Rtg2p, a cytoplasmic
protein with an N-terminal ATP binding domain belonging to the
actin/Hsp70/sugar kinase superfamily [11,12]. Rtg2p regulates
Rtg1/3p localization through dynamic interaction with Mks1p, a
negative regulator of the RTG pathway. When bound to the func-
tionally redundant 14-3-3 proteins Bmh1p and Bmh2p, Mks1p
promotes the phosphorylation of Rtg3p, thus inhibiting the
nuclear translocation of Rtg1/3p [13]. A second group of
RTG-target genes (including ATO1 and ATO2) is activated in
differentiated yeast colonies in a cell subpopulation, characterized
by decreased mitochondrial respiration. This subpopulation differs
from others, in which the CIT2 gene is activated, suggesting the
presence of additional co-regulators that determine the target-
specific function of RTG regulators in colonies [14].
Expression of a third group of RTG-target genes is not under
stringent regulation by RTG genes. This group includes genes
encoding the first three enzymes of the TCA cycle, namely, mito-
chondrial citrate synthase (CIT1), aconitase (ACO1), and NAD
+
-
dependent isocitrate dehydrogenase (IDH1/2) which are regulated
by Hap2/3/4/5 transcriptional complex in cells with robust
88 Nicoletta Guaragnella et al.
respiratory activity but come increasingly under the control of the
RTG pathway in cells with compromised mitochondrial respiratory
function [9].
The mitochondrial retrograde pathway is evolutionary con-
served from yeast to mammals. Yet there are cell-specific differences
in factors involved in the propagation of retrograde signaling in
mammalian cells of various origins (for refs. see [15]). Although
several potential target genes of the mammalian mitochondrial
stress pathway have been reported in different cells, the complete
genetic footprint of mitochondrial retrograde stress response
remains unknown [16].
The retrograde response in yeast and related pathways in higher
organisms share the common adaptive function of supporting cell
survival. Indeed, activation of mitochondrial retrograde signaling
Rtg3p
Rtg1p
P
Nucleus
R box
CIT2
CIT2
Rtg3p
Rtg1p
P
P
P
Rtg2p
Mks1p
P
Bmh1/2p
Mks1p
P
P
P
Rtg3p
Rtg1p
P
PCD resistance
Life span extension
Tumorigenesis
Mitochondrial
dysfunction
+
Bmh1/2p
+
Rtg2p
Fig. 1 Regulation of RTG-dependent gene expression in response to
mitochondrial dysfunction.The critical regulatory step of the RTG pathway is
the dynamic interaction between Rtg2p and Mks1p. When the RTG pathway is
inactive, Mks1p is found in its hyperphosphorylated form and interacts with
Bmh1/2p to inhibit the nuclear translocation of Rtg1/3p, and hence the
expression of RTG-target genes (such as CIT2). When the RTG pathway is
activated, Mks1p becomes partially dephosphorylated and binds to Rtg2p,
which causes Mks1p inactivation, releasing the inhibition from Rtg1/3p
nuclear translocation and inducing the expression of RTG-target genes. RTG
signaling activation has been implicated in life span extension, PCD resistance,
and tumorigenesis (see text for details)
Mitochondrial RTG Pathway Detection in Yeast 89
has been shown to extend life span (for ref. see [17]) and to trigger
cell resistance to proapoptotic stimuli both in yeast [18] and mam-
mals ([16] and refs. therein). In addition, mitochondria have been
suggested to have a tumor suppressor function since mitochondrial
dysfunction provides a survival advantage to cancer cells
[19,20]. In this respect, mitochondrial retrograde signaling has
also been associated with changes favoring cellular reprogramming
towards tumorigenesis (for refs. see [16]).
Hence, characterization of the genes regulated by the retro-
grade pathway is central to understanding the role of mitochondrial
stress in ageing, cellular resistance to RCD and cancer progression.
The existence of conserved RCD pathways in yeast sharing
several biochemical and morphological features with mammalian
apoptosis has been established [21]. This, together with the possi-
bility of heterologous expression of human disease genes in yeast
[22,23], makes this organism a suitable experimental platform for
unveiling both the mechanisms of mitochondrial retrograde
response and their role in physiopathology [17,24–26].
Detection of RTG signaling activation in yeast has mainly relied
on Northern blotting analysis of readouts of certain RTG-target
genes such as CIT2 and D-lactate dehydrogenase (DLD3)
[27,28]. Alternatively, β-galactosidase activity assay may be per-
formed in yeast cells harboring an episomal CIT2 promoter
sequence fused to the lacZ gene of E. coli [13,29]. In addition to
microarray technology, which needs dedicated laboratory facilities
[30], real-time polymerase chain reaction (PCR) is a quantitative,
accurate, and fast technique which allows for the analysis of large
panels of selected gene transcripts [18,31]. Recently, the level of
CIT2-gene protein product (Cit2p) expression has also been used
to measure activation of the RTG pathway in yeast cells either
through immunoblot analysis or fluorescence microscopy analysis
of yeast cells expressing Cit2p fused to green fluorescent protein
(GFP) [14]. In this chapter, we will describe how to measure
mitochondrial RTG-dependent retrograde signaling at the level of
mRNA and protein products in yeast cell suspensions and micro-
colonies under different experimental settings.
In the first experimental setting, shift to extracellular pH 3.00 is
used to cause a transient RTG signaling activation in yeast cells in
the presence of coupled mitochondria [18]. This causes upregula-
tion of CIT2 mRNA while IDH1 and ACO1 mRNAs, whose
transcription mainly depends on Hap2/3/4/5 complex (see
above), remain unchanged (see Fig. 2b). The method used consists
of measuring both the expression of RTG-target genes by real time-
PCR and the phosphorylation state of Rtg3p by immunoblotting
analysis [32]. This procedure may be easily applied to detect RTG
pathway activation in different steady-state and time-dependent
conditions. In the second detection system, Cit2p-GFP levels are
measured by Western blot and fluorescence microscopy in yeast
microcolonies from wild-type and retrograde mutant cells.
90 Nicoletta Guaragnella et al.
2 Materials
2.1 Cells
and Growth Media
1. Saccharomyces cerevisiae strains: W303-1B [Matαleu2 trp1
ura3 ade2 his3] (WT), W303-1B rtg2::LEU2 cells (Δrtg2)
and W303-1B rtg3::LEU2 (Δrtg3) [18]. BY-Cit2p-GFP
[MATα, his3Δ1, leu2Δ0, lys2Δ0, ura3Δ0, CIT2-yEGFP-
kanMX]; BY-rtg1-Cit2p-GFP [MATα, his3Δ1, leu2Δ0,
lys2Δ0, ura3Δ0, rtg1::nat, CIT2-yEGFP-kanMX]; BY-rtg2-
Cit2p-GFP [MATα, his3Δ1, leu2Δ0, lys2Δ0, ura3Δ0, rtg2::
nat, CIT2-yEGFP-kanMX]; BY-rtg3-Cit2p-GFP [MATα,
his3Δ1, leu2Δ0, lys2Δ0, ura3Δ0, rtg3::nat, CIT2-yEGFP-
kanMX], BY-mks1-Cit2p-GFP [MATα, his3Δ1, leu2Δ0,
lys2Δ0, ura3Δ0, mks1::nat, CIT2-yEGFP-kanMX] [14].
CIT2
IDH1
ACO1
mRNA levels (a. u.)
Neutral pH pH 3.00
0
2
4
6
8
10
12
B
A
Ct
Log (V )
10 arbitrary
0,00
5,00
10,00
15,00
20,00
25,00
30,00
-2 -1 0 1 2
y = -3,2033Ct + 19,247
CIT2
2
R = 0,9831
y = -3,2183Ct + 18,166
ACT1
2
R = 0,9902
CIT2
ACT1
Fig. 2 Expression analysis of RTG-target genes. (a) Standard curve of Ct values
as a function of a log
10
of CIT2 and ACT1 mRNA V
arbitrary
, determined by linear
regression analysis. (b) mRNA levels of CIT2,IDH1, and ACO1 were measured by
real-time PCR in cells grown in neutral and pH 3.00 YPR medium (see text for
details). CIT2 mRNA levels, normalized with that of ACT1 mRNA, were reported in
arbitrary units (a.u.)
Mitochondrial RTG Pathway Detection in Yeast 91
2. YPR medium: 1% Yeast extract (DIFCO), 2% Bacto Peptone
(DIFCO), 2% raffinose. Solid YPD medium: 1% Yeast extract
(DIFCO), 2% Bacto Peptone (DIFCO), 2% glucose, 2% Bacto
Agar (DIFCO).
3. Acid YPR medium: 1% Yeast extract (DIFCO), 2% Bacto Pep-
tone (DIFCO), 2% raffinose, set to pH 3.00 with HCl.
4. GMA agar (1% yeast extract, 3% glycerol, 1% ethanol, 2% agar).
All media components were mixed in ultrapure water (prepared
by purifying deionized water to attain a sensitivity of 18 MΩcm at
25 C). Media were sterilized by autoclaving for 20 min at 121 C.
2.2 Buffers,
Reagents,
and Labware
2.2.1 Real-Time PCR
1. Presto mini RNA Yeast Kit (Geneaid) was used for RNA isola-
tion providing all the buffers required (sorbitol buffer for cell
wall digestion with zymolyase, RB buffer for cell lysis, W1
buffer for RNA binding, wash buffer for RNA washing steps,
and RNAse-free water for RNA elution).
2. Sterile tips and Eppendorf tubes (0.2 ml; 1.5 and 2.0 ml) only
for RNA use.
3. Zymolyase 20 T (from Arthrobacter luteus, Seikagaku Biobusi-
ness), store at 4 C.
4. β-mercaptoethanol, store at 4 C.
5. 70% ethanol solution.
6. QuantiTect Reverse Transcription Kit (Qiagen) was used for
cDNA synthesis providing all the buffers and reagents required
(gDNA wipeout buffer for genomic DNA digestion, RNAse-
free water, Quantiscript Reverse Transcriptase Buffer 5,
Quantiscript Reverse Transcriptase and RT Primer mix for
cDNA production).
7. QuantiTect SYBR Green PCR Kit (Qiagen) was used for abso-
lute quantification of cDNA targets (i.e., CIT2) providing all
the buffers and reagents required (QuantiTect SYBR Green
PCR Master Mix containing HotStarTaq DNA Polymerase,
QuantiTect SYBR Green PCR Buffer, dNTP mix and the fluo-
rescent dye SYBR Green I, RNAse-free water).
8. 96 sterile multi-well plates and films (MicroAmp Fast Optical
96-well Reaction Plate, Applied Biosystems).
2.2.2 Protein Extraction Rtg3 Phosphorylation Detection
1. Lysis buffer: 0.225 M NaOH/1% 2-mercaptoethanol, added
with protein inhibitors: 10 mM NaF, 1 mM Na
3
VO
4,
and
2 mM phenylmethylsulfonyl fluoride (PMSF) (see Note 1).
2. SDS PAGE sample buffer: 100 mM Tris–HCl (pH 6.8), 4%
sodium dodecyl sulphate (SDS), 20% glycerol, 100 mM dithio-
threitol (DTT), 0.002% bromophenol blue.
92 Nicoletta Guaragnella et al.
Cit2p-GFP Detection
3. Lysis buffer: 10 mM MES [2-(N-morpholino)ethanesulfonic
acid] buffer (pH 6), supplemented with Complete protease
inhibitor mixture (Roche Applied Science) and 1 mM AEBSF
[4-(2-aminoethyl)benzenesulfonyl fluoride, Sigma]; cell dis-
ruption by glass beads (acid washed 425–600 μm, Sigma
G8772).
4. SDS PAGE sample buffer: 50 mM Tris–HCl (pH 6.8), 10%
glycerol, 2% SDS, 0.01% bromophenol blue, 0.6% DTT.
2.2.3 SDS-
Polyacrylamide Gel
Electrophoresis
Rtg3 Phosphorylation Detection
1. Tris/Cl SDS pH 8.45: 3 M Tris–HCl, 0.3% SDS.
2. Stock acrylamide/bisacrylamide 30/0.8: Acrylamide 30 gr,
bisacrylamide 0.8 gr (for 100 ml). Filter with 0.4 μm filters
and store in a dark flask at 4 C. The solution is stable for 1 year
if kept at 4 C. Weigh the powder in a fume cupboard using
protective gloves and turn the ventilation on only after the
H
2
O has been added.
3. Cathode buffer (Top): Tris–Tricine–SDS Buffer, pH 8.25,
diluted 1:10.
4. Anode buffer (Bottom): 1 M Tris–HCl, pH 8.9.
5. Amersham Full-Range Rainbow Molecular Weight Marker
(GE Healthcare).
Cit2p-GFP Detection
6. 12% SDS-polyacrylamide gel. Mix ROTIPHORESE gel
30 (37,5:1) (Carl Roth GmbH+Co KG), #3029.1:0.5 M
Tris, pH 8.8, 0.4% SDS:ultrapure H
2
O¼4:2.5:3.5 (volume).
7. Running buffer: 25 mM Tris, 192 mM glycine, and 0.1% SDS,
pH 8.6.
8. Precision Plus Protein Kaleidoscope Prestained Protein Stan-
dards, BIO-RAD, #1610375.
2.2.4 Western Blotting Rtg3 Phosphorylation Detection
1. Transfer buffer: 9.93 mM 3-(Cyclohexylamino)-1-propanesul-
fonic acid (CAPS), 20% methanol, pH 11.
2. Polyvinylidene fluoride (PVDF) membranes (Millipore,
Immobilon-P 0.4 μm).
3. TBS buffer: 20 mM Tris, 150 mM NaCl, pH 7.6 (see Note 2).
4. Blocking buffer: 5% (w/v) nonfat dry milk in TBS, pH 7.6.
Cit2p-GFP Detection
5. Transfer buffer: 25 mM Tris, 192 mM glycine, 15% methanol.
Mitochondrial RTG Pathway Detection in Yeast 93
6. Polyvinylidene fluoride (PVDF) membranes (Millipore,
Immobilon-P 0.4 μm).
7. Blocking buffer: 1% casein in PBS (10 mM Na
2
HPO
4
·12 H
2
O,
154 mM NaCl, pH 7.4).
3 Methods
3.1 Analysis
of RTG-Dependent
Target Gene mRNAs
3.1.1 Cell Growth
and Low pH Shift
1. Preculture yeast cells in 5 ml of liquid YPR medium and incu-
bate on a rotary shaker at 150 rpm and 26 C for about 8 h.
Dilute aliquot of the culture in liquid YPR medium and grow
overnight (150 rpm and 26 C) up to exponential phase
(OD
600
about 0.7). Determine the concentration of cell sus-
pension using a cell-counting chamber (Neubauer Improved).
Collect cells (about 20 10
7
for each sample for RT PCR and
510
7
for protein extraction) by centrifugation at 3000 g
for 5 min. An aliquot of the culture (about 20 10
7
for each
sample for real-time PCR and 5 10
7
for protein extraction) is
shifted to pH 3.00 and then collected.
3.1.2 RNA Isolation
and Real-Time PCR
1. Resuspend the cells grown at neutral pH or shifted to pH 3.00
in sorbitol buffer containing zymolyase (5 mg/sample) and
incubate on a rotary shaker at 150 rpm and 30 C for 30 min
(see Note 3). Collect the spheroplasts by centrifugation and
proceed with RNA isolation (see Note 4). First lyse the cells in
300 μl RB buffer + β-mercaptoethanol, then transfer the lysate
to a 2 ml RB column to allow RNA binding, perform the
washing steps and elute RNA in 50 μl RNAse-free water.
2. Quantify RNA by UV spectrophotometric assay and evaluate
its purity (usually A
260
/A
280
2.0).
3. To avoid RNA degradation, cDNA synthesis must be per-
formed immediately. The QuantiTect Reverse Transcription
Kit allows synthesis of cDNA in two steps with a 20-min
procedure. Unless otherwise stated, RNA manipulation must
be performed on ice. First, to effectively remove contaminating
genomic DNA (gDNA), incubate isolated RNA (1 μg per
sample) at 42 C for 2 min with 2 μl of gDNA wipeout buffer
in a final volume of 14 μl in RNase-free water. Then place the
mixture on ice and use immediately for reverse transcription
reaction. Prepare the reverse transcription master mix (1 μl
Quantiscript Reverse Transcriptase, 4 μl Quantiscript RT
buffer 5, and 1 μl RT primer Mix) and add 14 μl template
RNA. Perform cDNA synthesis in a thermal cycler (Perkin
Elmer GeneAmp PCR System 2400) using the following pro-
gram: 42 C for 15 min, 95 C for 3 min.
4. Store cDNA at 20 C until use.
94 Nicoletta Guaragnella et al.
5. Create standard curves for each cDNA target to be analyzed
(CIT2,ACO1, IDH1) and a reference gene used for normali-
zation (ACT1). To this aim, prepare serial dilutions (1:1; 1:5;
1:25; 1:125; 1:625; 1:3125) of one of the cDNA sample
prepared (see Note 5). For each RTG-target and reference
gene, prepare a reaction mixture to analyze six cDNA dilutions
plus one blank (no cDNA) sample in triplicates, as follows: for
one sample add 0.1 μl specific primers (100 pmol/μl) (CIT2:
(F) 50-CGGTTATGGTCATGCTGTGCT-30and (R) 50-GGT
CCATGGCAAACTTACGCT -30;ACO1: (F) CATTTACCC
CCGATTTGGCT and (R) GGTACAAGAACCGATCAAAC
CG, IDH1; (F) TCGACAATGCCTCCATGCA and (R) AAAG
CAGCGCCAATGTTGC ; ACT1: (F) 50- CTTTGGCTCC
ATCTTCCATG -30and (R) 50- CACCAATCCAGACGGAG
TACTT-30), 10 μl2QuantiTect SYBR Green PCR Master
Mix and 8.8 μl RNAse-free water. Calculate the total volume of
the reaction mixture multiplying each volume by three times
the number of samples plus 1 (24 in this case). For each diluted
cDNA, add 3 μl cDNA to 57 μl reaction mixture and load 19 μl
per well in triplicate in a 96-multi-well plate. Run the following
real-time PCR program by using ABI Prism 7900HT System:
15 min 95 C; 40 cycles (15 s 95 C, 30 s 59 C, 30 s 72 C);
dissociation cycle. To obtain the standard curve, report the
mean Ct value for each diluted sample as a function of the
logarithm (log
10
) of the arbitrary volume (V
arbitrary
) values
corresponding to the dilution prepared (Fig. 2a). The linear
equations obtained will be used to calculate the amounts of
unknown samples.
6. For quantification of each RTG target (CIT2, IDH1, ACO1)
and reference gene (ACT1) set up a reaction mixture as
described at point 5. Calculate the total volume of the reac-
tion mixture multiplying the volume of each component by
three times the number of samples plus one blank sample plus
1. For each sample to be analyzed, add 3 μl cDNA to 57 μl
reaction mixture and load 19 μl per well in triplicate in a
96-multi-well plate. Run the real-time PCR program as previ-
ously described.
7. To determine the amount of each RTG-target gene mRNA,
calculate the log
10
(V
arbitrary
) using the respective standard
curve:
log 10 Varbitrary
¼CtYFG b
ðÞ
=a:
where YFG is Your Favorite Gene, ais the slope and bis the
intercept of the calibration curve.
Mitochondrial RTG Pathway Detection in Yeast 95
The normalized amount of an RTG-target gene is the ratio
between its V
arbitrary
value and that of ACT1 mRNA and is
expressed in the mRNA levels expressed in arbitrary units (a.u.)
(Fig. 2b).
3.2 Analysis
of CIT2p-GFP
3.2.1 Cell Lysis
1. Harvest cells (50–100 mg) from 3 to 4 days old microcolonies
of Cit2p-GFP producing strain (with genomic CIT2 gene
fused with the gene encoding GFP) on GMA. Harvested cells
can be stored at 70 C.
2. Resuspend harvested cells in 10 mM MES buffer, pH 6, sup-
plemented with Complete protease inhibitor mixture and dis-
rupt them using glass beads in FastPrep device.
3. Remove cell debris, centrifuge lysate for 3 min at 3000 gand
collect the supernatant.
4. Determine protein concentration in the supernatant using a
protein detection kit (Bio-Rad Protein Assay Dye Reagent
Concentrate # 5000006). Store the supernatants at 70 C.
3.2.2 Western Blot 1. After denaturation in Laemmli sample buffer, separate samples
by SDS-PAGE using 12% gels.
2. After transfer to a PVDF membrane (Immobilon-P, Millipore),
check the amount of loaded protein by Coomassie blue stain-
ing of each membrane.
3. Detect Cit2p-GFP by mouse monoclonal anti-GFP antibody,
horseradish peroxidase (HRP) conjugate (#sc-9996 HRP,
Santa Cruz) and visualize peroxidase signal with SuperSignal
West Pico (Pierce) on Super RX medical x-ray film (Fuji)
(Fig. 3).
3.2.3 Fluorescence
Microscopy
1. Grow cells expressing genomic CIT2-GFP (e.g., BY-Cit2p-
GFP and BY-mks1-Cit2p-GFP) under conditions in which
RTG pathway is activated (e.g., three-day-old microcolonies
on GMA) (see Note 6).
2. Harvest cells and assay Cit2p-GFP fluorescence in individual
cells by fluorescence microscopy (e.g., Leica DMR) using GFP
filter. In parallel, visualize all cells by differential interference
contrast (DIC) or bright-field microscopy (Fig. 4).
3.3 Analysis of Rtg3p
Phosphorylation
3.3.1 Total Yeast Protein
Extraction
1. Harvest 5 ml of cells grown at neutral pH or shifted to pH 3.00
by centrifugation for 5 min at 3000 g(see Note 7).
2. Resuspend the pellet in 150 μl of freshly prepared lysis buffer.
3. Incubate on ice for 10 min.
4. Add with the equal volume of trichloroacetic acid (TCA), to a
final concentration of 6.1%.
96 Nicoletta Guaragnella et al.
5. Incubate on ice for 10 min.
6. Centrifuge for 10 min at 10,000 gand 4 C.
7. Remove supernatant completely and resuspend in 150 μlof
SDS-PAGE sample buffer (see Note 8).
8. Heat for 10 min at 65 C. These samples can be stored at
20 C.
3.3.2 Western Blotting 1. Load equivalent amounts of total cellular protein extracts on
7.5% SDS-PAGE gels (Amersham Biosciences, Mighty Small II
mini vertical electrophoresis unit).
2. Activate PVDF membrane by soaking in methanol for 5 min,
rinse with distilled water, and equilibrate in transfer buffer at
least for 5 min. Cut away the stacking gel and equilibrate the
gel in transfer buffer; prepare the transfer stack in the following
order (bottom-up): four Whatman paper sheets saturated in
BY-Cit2p-GFP
BY-rtg1-Cit2p-GFP
BY-rtg2-Cit2p-GFP
BY-rtg3-Cit2p-GFP
BY-mks1-Cit2p-GFP
A
B
Cit2p-GFP
Fig. 3 RTG pathway activity in three-day-old microcolonies on GMA detected by
monitoring of Cit2p-GFP level. (a) Cit2p-GFP level in WT,rtg1,rtg2,rtg3, and
mks1 strains (Western blot). (b) Protein loading controls (PVDF membrane
stained by Coomassie blue)
Mitochondrial RTG Pathway Detection in Yeast 97
transfer buffer, the gel, activated PVDF membrane, four What-
man paper sheets saturated in transfer buffer; place the stack
onto semidry transfer units TE 70 (Amersham Biosciences) and
transfer proteins at 50 mA for 75 min (see Note 9).
3. Probe the membranes with polyclonal anti-Rtg3p antibody,
diluted in blocking buffer at 1:1000 (v/v) and monoclonal
anti-phosphoglycerate kinase (anti-Pgk1p) antibody, diluted
1:6000 (v/v) (Molecular Probes) (see Note 10).
4. Perform immunodetection with horseradish peroxidase-
conjugated anti-rabbit (for Rtg3p) or anti-mouse (for Pgk1p)
antibodies using chemiluminescence Western blotting reagents
(Amersham ECL Western Blotting Detection Reagent, GE
Healthcare Life Sciences). Immunofluorescent bands are visua-
lized with high-performance Amersham Hyperfilm™ECL
(GE Healthcare) (Fig. 5). RTG signaling is inactive in Δrtg2
cells and the Rtg3p hyper-phosphorylated state is shown by the
slower mobility of the immunoreactive band which is smeared
likely due to multiple phosphorylations. The faster mobility of
Rtg3p-immunoreactive band in WT cells shifted to pH 3.00
indicates RTG pathway activation [8].
Cit2p-GFP DIC
BY-Cit2p-GFPBY-mks1-Cit2p-GFP
Fig. 4 RTG pathway activity detected at single-cell level by microscopy. Micro-
colonies of BY-Cit2p-GFP and BY-mks1-Cit2p-GFP strains were grown 3 days on
GMA. DIC differential interference contrast
98 Nicoletta Guaragnella et al.
4 Notes
1. Since anti-Rtg3p antibody is not an anti-phosphoprotein anti-
body, it is highly recommended that set of specific protein
degradation inhibitors be added to the cell lysis buffer. In this
case, NaF, an inhibitor of serine/threonine and acidic phos-
phatases, Na
3
VO
4,
an inhibitor of tyrosine and alkaline phos-
phatases, and PMSF, an inhibitor of serine proteases, were
added.
2. It is critical to use TBS buffer without any detergent, other-
wise, you will have a very high background.
3. For cell wall digestion of 20 10
7
cells, 500 μl of sorbitol
buffer + zymolyase were used.
4. For RNA isolation, centrifugation temperature must be
between 20 C and 25 C.
5. Accurate pipetting is required when diluting template cDNA to
create the standard curve.
6. Strains with genomic CIT2-GFP have to be used for RTG
pathway detection via Cit2p-GFP.
7. To analyze several samples in different conditions, the amount
of proteins to be loaded onto the gel must be determined. This
can be done by measuring the OD
600
before processing each
sample and calculating the volume of each cell culture
corresponding to the initial OD600 value. Equal loading of
protein samples can also be confirmed by Ponceau staining of
membranes prior to incubation with antibody.
Rtg3p
Pgk1p
∆rtg3
WT
∆rtg2
WT pH 3.00
Fig. 5 Rtg3p phosphorylation state in WT and Δrtg2 cells in raffinose. Cell
protein extracts were prepared from WT and Δrtg2 cells grown at neutral pH and
from WT cells shifted to pH 3.00 and analyzed by immunoblot with anti-Rtg3p
and anti-Pgk1p antibodies. Anti-Pgk1p antibody was used as cytosolic marker
protein to normalize the quantity of proteins loaded. Cell extracts from Δrtg3
cells have been analyzed as a negative control
Mitochondrial RTG Pathway Detection in Yeast 99
8. If the solution turns yellow after resuspension in SDS-PAGE
buffer, add 5 μl Tris 1 M solution (untitrated) until the solution
becomes blue.
9. Proteins bind to the membrane as soon as contact occurs, so it
is important to place the gel correctly on the first try.
10. Primary antibody dilutions in blocking buffer can be stored at
20 C and used up to three times.
Acknowledgments
The authors thank Derek Wilkinson for proofreading of the manu-
script. Rtg3p antibody was kindly provided by Dr. Z. Liu, Univer-
sity of New Orleans, New Orleans (USA). This work was supported
by the Ministry of Science of Montenegro, Project “New methods
for risk stratification for the progression of cancer and Alzheimer‘s
disease in patients in Montenegro—DEMONSTRATE” to SG and
by Czech Science Foundation 19-09381S and by LQ1604 NPU II
(MEYS) to ZP; ZP research was performed in BIOCEV supported
by CZ.1.05/1.1.00/02.0109 provided by ERDF and MEYS.
References
1. Portt L, Norman G, Clapp C, Greenwood M,
Greenwood MT (2011) Anti-apoptosis and
cell survival: a review. Biochim Biophys Acta
1813(1):238–259
2. Wang C, Youle RJ (2009) The role of mito-
chondria in apoptosis. Annu Rev Genet
43:95–118
3. Quiros PM, Mottis A, Auwerx J (2016) Mito-
nuclear communication in homeostasis and
stress. Nat Rev Mol Cell Biol 17(4):213–226.
https://doi.org/10.1038/nrm.2016.23
4. Coyne LP, Chen XJ (2018) mPOS is a novel
mitochondrial trigger of cell death - implica-
tions for neurodegeneration. FEBS Lett 592
(5):759–775. https://doi.org/10.1002/
1873-3468.12894
5. Wang X, Chen XJ (2015) A cytosolic network
suppressing mitochondria-mediated proteo-
static stress and cell death. Nature 524
(7566):481–484. https://doi.org/10.1038/
nature14859
6. Guaragnella N, Coyne LP, Chen XJ, Giannat-
tasio S (2018) Mitochondria-cytosol-nucleus
crosstalk: learning from Saccharomyces cerevi-
siae. FEMS Yeast Res 18(8):foy088. https://
doi.org/10.1093/femsyr/foy088
7. Boos F, Kramer L, Groh C, Jung F,
Haberkant P, Stein F, Wollweber F,
Gackstatter A, Zoller E, van der Laan M,
Savitski MM, Benes V, Herrmann JM (2019)
Mitochondrial protein-induced stress triggers a
global adaptive transcriptional programme.
Nat Cell Biol 21(4):442–451. https://doi.
org/10.1038/s41556-019-0294-5
8. Liao XS, Small WC, Srere PA, Butow RA
(1991) Intramitochondrial functions regulate
nonmitochondrial citrate synthase (CIT2)
expression in Saccharomyces cerevisiae. Mol
Cell Biol 11(1):38–46
9. Liu Z, Butow RA (2006) Mitochondrial retro-
grade signaling. Annu Rev Genet 40:159–185.
https://doi.org/10.1146/annurev.genet.40.
110405.090613
10. Jia Y, Rothermel B, Thornton J, Butow RA
(1997) A basic helix-loop-helix-leucine zipper
transcription complex in yeast functions in a
signaling pathway from mitochondria to the
nucleus. Mol Cell Biol 17(3):1110–1117
11. Bork P, Sander C, Valencia A (1992) An
ATPase domain common to prokaryotic cell
cycle proteins, sugar kinases, actin, and hsp70
heat shock proteins. Proc Natl Acad Sci U S A
89(16):7290–7294
12. Koonin EV (1994) Yeast protein controlling
inter-organelle communication is related to
bacterial phosphatases containing the Hsp
70-type ATP-binding domain. Trends Bio-
chem Sci 19(4):156–157
100 Nicoletta Guaragnella et al.
13. Liu Z, Sekito T, Spirek M, Thornton J, Butow
RA (2003) Retrograde signaling is regulated by
the dynamic interaction between Rtg2p and
Mks1p. Mol Cell 12(2):401–411
14. Podholova K, Plocek V, Resetarova S,
Kucerova H, Hlavacek O, Vachova L, Palkova
Z (2016) Divergent branches of mitochondrial
signaling regulate specific genes and the viabil-
ity of specialized cell types of differentiated
yeast colonies. Oncotarget 7
(13):15299–15314. https://doi.org/10.
18632/oncotarget.8084
15. Butow RA, Avadhani NG (2004) Mitochon-
drial signaling: the retrograde response. Mol
Cell 14(1):1–15
16. Guha M, Avadhani NG (2013) Mitochondrial
retrograde signaling at the crossroads of tumor
bioenergetics, genetics and epigenetics. Mito-
chondrion 13(6):577–591. https://doi.org/
10.1016/j.mito.2013.08.007
17. Jazwinski SM, Kriete A (2012) The yeast retro-
grade response as a model of intracellular sig-
naling of mitochondrial dysfunction. Front
Physiol 3:139
18. Guaragnella N, Zdralevic M, Lattanzio P,
Marzulli D, Pracheil T, Liu Z, Passarella S,
Marra E, Giannattasio S (2013) Yeast growth
in raffinose results in resistance to acetic-acid
induced programmed cell death mostly due to
the activation of the mitochondrial retrograde
pathway. Biochim Biophys Acta 1833
(12):2765–2774. https://doi.org/10.1016/j.
bbamcr.2013.07.017
19. Cuezva JM, Ortega AD, Willers I, Sanchez-
Cenizo L, Aldea M, Sanchez-Arago M (2009)
The tumor suppressor function of mitochon-
dria: translation into the clinics. Biochim Bio-
phys Acta 1792(12):1145–1158. https://doi.
org/10.1016/j.bbadis.2009.01.006
20. Compton S, Kim C, Griner NB, Potluri P,
Scheffler IE, Sen S, Jerry DJ, Schneider S,
Yadava N (2011) Mitochondrial dysfunction
impairs tumor suppressor p53 expression/
function. J Biol Chem 286
(23):20297–20312. https://doi.org/10.
1074/jbc.M110.163063
21. Carmona-Gutierrez D, Bauer MA,
Zimmermann A, Aguilera A, Austriaco N,
Ayscough K, Balzan R, Bar-Nun S,
Barrientos A, Belenky P, Blondel M, Braun
RJ, Breitenbach M, Burhans WC, Buttner S,
Cavalieri D, Chang M, Cooper KF, Corte-
Real M, Costa V, Cullin C, Dawes I,
Dengjel J, Dickman MB, Eisenberg T,
Fahrenkrog B, Fasel N, Frohlich KU,
Gargouri A, Giannattasio S, Goffrini P, Gour-
lay CW, Grant CM, Greenwood MT,
Guaragnella N, Heger T, Heinisch J,
Herker E, Herrmann JM, Hofer S, Jimenez-
Ruiz A, Jungwirth H, Kainz K, Kontoyiannis
DP, Ludovico P, Manon S, Martegani E,
Mazzoni C, Megeney LA, Meisinger C,
Nielsen J, Nystrom T, Osiewacz HD, Outeiro
TF, Park HO, Pendl T, Petranovic D, Picot S,
Polcic P, Powers T, Ramsdale M,
Rinnerthaler M, Rockenfeller P,
Ruckenstuhl C, Schaffrath R, Segovia M,
Severin FF, Sharon A, Sigrist SJ, Sommer-
Ruck C, Sousa MJ, Thevelein JM,
Thevissen K, Titorenko V, Toledano MB,
Tuite M, Vogtle FN, Westermann B,
Winderickx J, Wissing S, Wolfl S, Zhang ZJ,
Zhao RY, Zhou B, Galluzzi L, Kroemer G,
Madeo F (2018) Guidelines and recommenda-
tions on yeast cell death nomenclature. Microb
Cell 5(1):4–31. https://doi.org/10.15698/
mic2018.01.607
22. Mager WH, Winderickx J (2005) Yeast as a
model for medical and medicinal research.
Trends Pharmacol Sci 26(5):265–273.
https://doi.org/10.1016/j.tips.2005.03.004
23. Greenwood MT, Ludovico P (2009) Expres-
sing and functional analysis of mammalian apo-
ptotic regulators in yeast. Cell Death Differ 17
(5):737–745
24. Giannattasio S, Guaragnella N, Arbini AA,
Moro L (2013) Stress-related mitochondrial
components and mitochondrial genome as tar-
gets of anticancer therapy. Chem Biol Drug
Des 81(1):102–112. https://doi.org/10.
1111/cbdd.12057
25. Guerra F, Arbini AA, Moro L (2017) Mito-
chondria and cancer chemoresistance. Biochim
Biophys Acta 1858(8):686–699. https://doi.
org/10.1016/j.bbabio.2017.01.012
26. Guaragnella N, Palermo V, Galli A, Moro L,
Mazzoni C, Giannattasio S (2014) The
expanding role of yeast in cancer research and
diagnosis: insights into the function of the
oncosuppressors p53 and BRCA1/2. FEMS
Yeast Res 14(1):2–16. https://doi.org/10.
1111/1567-1364.12094
27. Dilova I, Powers T (2006) Accounting for
strain-specific differences during RTG target
gene regulation in Saccharomyces cerevisiae.
FEMS Yeast Res 6(1):112–119. https://doi.
org/10.1111/j.1567-1364.2005.00008.x
28. Giannattasio S, Liu Z, Thornton J, Butow RA
(2005) Retrograde response to mitochondrial
dysfunction is separable from TOR1/2 regula-
tion of retrograde gene expression. J Biol
Chem 280(52):42528–42535. https://doi.
org/10.1074/jbc.M509187200
Mitochondrial RTG Pathway Detection in Yeast 101
29. Liao X, Butow RA (1993) RTG1 and RTG2:
two yeast genes required for a novel path of
communication from mitochondria to the
nucleus. Cell 72(1):61–71
30. Epstein CB, Waddle JA, Hale W, Dave V,
Thornton J, Macatee TL, Garner HR, Butow
RA (2001) Genome-wide responses to mito-
chondrial dysfunction. Mol Biol Cell 12
(2):297–308
31. Miceli MV, Jazwinski SM (2005) Common
and cell type-specific responses of human cells
to mitochondrial dysfunction. Exp Cell Res
302(2):270–280. https://doi.org/10.1016/j.
yexcr.2004.09.006
32. Sekito T, Thornton J, Butow RA (2000)
Mitochondria-to-nuclear signaling is regulated
by the subcellular localization of the transcrip-
tion factors Rtg1p and Rtg3p. Mol Biol Cell 11
(6):2103–2115
102 Nicoletta Guaragnella et al.
Chapter 7
Native Gel Electrophoresis and Immunoblotting to Analyze
Electron Transport Chain Complexes
Gisela Beutner and George A. Porter Jr.
Abstract
Native electrophoresis is a powerful tool to analyze the mitochondrial electron transport chain complexes
(Cx) I–V and their assembly into supercomplexes. Valuable information regarding the composition and
bioenergetic regulation in physiological and pathological conditions can be obtained. This chapter com-
pares different types of native electrophoresis to analyze mitochondrial supercomplexes.
Key words Native electrophoresis (blue native, colorless native, clear native, hybrid), Mitochondrial
supercomplexes
1 Introduction
The mitochondrial electron transport chain (ETC) transduces the
energy derived from the breakdown of various fuels into the bioe-
nergetic currency of the cell, ATP. The ETC is composed of five
massive protein complexes, which also assemble into supercom-
plexes called respirasomes (C-I, C-III, and C-IV) and synthasomes
(C-V) that increase the efficiency of electron transport and ATP
production. Various methods have been used for over 50 years to
measure ETC function, but these protocols provide only limited
information on the assembly of individual complexes and
supercomplexes.
Native electrophoresis is the tool of choice to identify and
characterize mitochondrial membrane-bound supercomplexes.
Several types of native electrophoresis are currently used, and the
main difference between them is the use of the anionic dye Coo-
massie G250 and detergent during sample preparation or the elec-
trophoresis itself. Most commonly used are blue native
(BN) electrophoresis [1] and high-resolution clear native (hrCN)
electrophoresis [2]. Colorless native (CLN; [3]) omits dye and
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_7,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
103
detergent during the electrophoresis and provides only a limited
resolution of protein complexes. However, a combination of BN
and CLN, called hybrid electrophoresis [4,5], is able to overcome
this limitation. All mentioned protocols of native electrophoresis
allow the separation of physiologically functional protein com-
plexes by their molecular weight, by their netcharge, or by a com-
bination thereof.
In BN and hybrid electrophoresis, Coomassie 250G binds to
the solubilized protein complexes and facilitates their migration
toward the anode. This allows relatively high resolution of the
bands of Coomassie-stained proteins that are visible as they migrate
through the gel. However, Coomassie 250G may interfere with
analytical methods following the native electrophoresis [2,6,7].
CLN electrophoresis is a variation of BN electrophoresis that
avoids adding Coomassie G250 to the cathode buffer and the
sample, while using the same technical conditions as for BN elec-
trophoresis [8]. Separation of protein complexes depends primarily
on the intrinsic charge and only secondary on the size of the protein
complexes. Consequently, this technique is limited to the separa-
tion of acidic proteins. However, adding mild neutral and anionic
detergents to the cathode buffer overcomes this limitation and
results in a high resolution of the separated protein complexes
(hrCN electrophoresis; [2]).
The hybrid method combines the advantages of the charge
given by Coomassie Blue G250 in BN electrophoresis with those
of the CLN electrophoresis: after all proteins have entered the
gradient gel, the blue cathode buffer is exchanged for a colorless
buffer. A mild charge shift due to the Coomassie Blue G250
binding to the protein complexes is maintained and therefore this
protocol allows all proteins to migrate through the gel.
After completion of native electrophoresis, a number of options
allow further analysis of the separated protein complexes. The
proteins in clear or colorless native gels may be transferred to
membranes for immunoblotting to visualize all proteins of interest
independent of their intrinsic enzymatic activity. In contrast, in-gel
assays (IGA; [7]) use the enzymatic activity of the ETC complex
within the gel to visualize the active protein complexes. Entire
lanes, parts of a lane, or single bands of a native gels can be used
for denaturing 2D electrophoresis and subsequent immunoblot-
ting. Single bands of a BN or hybrid PAGE may be further analyzed
by electroelution [6] or mass spectrometry for in-depth analysis of a
protein complex.
This chapter outlines the protocols of different types of native
electrophoresis and the transfer of these gels to perform immuno-
blotting. In-gel assays are described elsewhere in this book. The
reader is advised to consider the strengths and weaknesses of the
different techniques when designing experiments to test their par-
ticularly hypotheses.
104 Gisela Beutner and George A. Porter Jr.
2 Materials
2.1 Preparation
for Native
Electrophoresis (See
Note 1)
1. Anode buffer for BN/hrCN/hybrid electrophoresis: Dissolve
25 mM imidazole in H
2
O; pH 7.0 (7.0–7.5 for hybrid) at 4 C,
store at 4 C.
2. Cathode buffer for hrCN: Dissolve 7.5 mM imidazole and
50 mM tricine in H
2
O; pH to 7.0 at 4 C; add 0.5 g sodium
deoxycholate and 0.2 g dodecyl-maltoside per liter of buffer;
store at 4 C.
3. Cathode buffer for BN (see Note 2):A: Deep blue cathode
buffer: Dissolve 7.5 mM imidazole and 50 mM tricine in H
2
O;
pH to 7.0 at 4 C; add 0.02 (w/v) Coomassie G-250; B: Light
blue cathode buffer: Dissolve 7.5 mM imidazole and 50 mM
tricine in H
2
O; pH to 7.0 at 4 C, add 0.002% Coomassie
G250; Store buffers at 4 C.
4. Cathode buffer for hybrid (see Note 3): Buffer A: Dissolve
7.5 mM imidazole and 50 mM tricine in H
2
O; pH to
7.0–7.5 at 4 C; add 0.02% Coomassie G250. Buffer B: Same
as Buffer A, but without Coomassie G250.
5. Acrylamide/bisacrylamide (AAB) (see Note 4):To 48 g acryl-
amide and 1.5 g bisacrylamide, add H
2
O to a total of 100 mL
or alternatively use a premade acrylamide/bisacrylamide
solution.
6. Gel buffer for BN/hrCN/hybrid:Dissolve 75 mM imidazole
and 1.5 M aminocaproic acid in H
2
O; pH 7.0 as 4 C (pH 7.5
for hybrid).
7. Extraction buffer A:Dissolve 50 mM NaCl, 50 mM imidazole,
2 mM aminocaproic acid, 1 mM EDTA in H
2
O; pH 7.0 at
4C.
8. Extraction buffer B (see Note 5):Dissolve 150 mM sodium
acetate, 30 mM HEPES, 1 mM EDTA, 12% glycerol (w/v) in
H
2
O; pH 7.5; store at 4 C.
9. Loading buffer for hrCN:Dissolve 0.1% Ponceau S in 50%
glycerol.
10. Loading buffer for BN/hybrid:Add 5% Coomassie G250
(w/v) to 500 mM aminocaproic acid pH 7.0–7.5. Aliquot
and store at 20 C.
11. Detergents: Dissolve 1% and 10% dodecyl-maltoside or digito-
nin (w/v) in H
2
O. Make aliquots of 200 μL and store frozen.
12. Molecular weight marker (see Note 6):Prepare according to
the instructions of the supplier.
Native Electrophoresis 105
2.2 Preparation
for Western Blotting
1. Ponceau stain (500 mL): Prepare by adding 25 mL acetic acid
and 0.5 g Ponceau S to 475 mL H
2
O; store at room tempera-
ture, can be reused multiple times.
2. Tris-buffered saline (TBS): Prepare by dissolving 200 mM
NaCl, 25 mM Tris-Base, 2.7 mM KCl; pH to 7.5 and store at
room temperature.
3. TBS-Tween (TBST): Prepare by adding 0.5 mL/L Tween
20 to TBS; store at room temperature.
4. Milk solids/TBS: Prepare by dissolving 5 g milk solids in
100 mL TBS; store at 4 C and use within three days.
5. 3% BSA/TBS: Prepare by dissolving 3 g bovine serum albumin
(BSA, fraction V) in 100 mL TBS; store at 4 C and use within
three days or store in aliquots at 20 C.
3 Methods
3.1 Protocol Native
Electrophoresis
1. Prepare a 3–8% or 4–10% acrylamide/bisacrylamide (AAB)
gradients for hrCN, BN, and hybrid gels, respectively. Table 1
summarizes the quantities of buffer, AAB, H
2
O, glycerol,
ammonium persulfate (APS), and tetramethylethylenediamine
(TEMED) used for a mini-gel (size of the gel: 85 mm
wide 73 mm high 1.5 mm thick) or maxi-gel (size of the
gel: 160 mm wide 200 mm high 1.5 mm thick) without a
stacking gel. Assemblies of glass plates with native gels can be
stored refrigerated in a plastic bag with a few mL of 1gel
buffer. In a moist environment, the gels are stable for use for up
to a week.
Table 1
Quantities of ingredients needed to pour 1 mini or maxi PAGE. The volumes used in this table are
calculated for 1 mini-gel or 1 maxi-gel, 1.5 mm thick. The volume of AAB is based on a 40% stock
solution (37.5:1; 2.7% crosslinker). APS and TEMED are added after each column of the gradient
mixer is filled with the first 3 or 4 ingredients of the respective gel
3–8% (mini) 4–10% (maxi)
3% (light) 8% (heavy) 4% (light) 10% (heavy)
AAB (mL) 0.4 1.3 2.5 7.7
CN/BN buffer (mL) 1.6 1.6 8.5 8.5
H
2
O (mL) 2.7 1.4 14 6.3
Glycerol (g) 0.4 2.5
Volume (mL) 4.7 4.7 25 25
APS (μL) 30 30 65 65
TEMED (μL) 5 5 10 10
106 Gisela Beutner and George A. Porter Jr.
2. To pour the native gel, place the gradient mixer on an elevated
stirring plate to ensure that the gel-forming solutions will flow
by gravity into the prepared gel chamber. Fill the outflow
chamber of the gradient mixer with 4.7 mL/25 mL (mini/
maxi) of the heavy solution (see Note 7), and the other cham-
ber of the gradient mixer with 4.7 mL/25 mL (mini/maxi) of
the light solution.
3. Place a stir bar into the outflow chamber with the heavy solu-
tion and begin stirring using a stir bar speed that mixes quickly
but does not cause bubbling.
4. Quickly add APS and TEMED to each chamber to initiate
polymerization.
5. Open the stopcock between the two chambers of the gradient
mixer and allow mixing for a few seconds before opening the
outflow stopcock to pour the gel. Gravity will drain both
chambers equally, and mixing of the light AAB solution into
the heavy AAB solution will slowly decrease the acrylamide
density from the bottom to the top of the gel. Use the entire
content that is in the gradient mixer to pour the gel (see Note
8). When the gel has been poured, carefully mount the well
comb into the gel to avoid bubbles and mixing of the layers.
6. Wash the gradient mixer immediately with ethanol to rinse out
any gel, rinse with water and pour the second gel, if needed. Let
the gels polymerize—usually less than 20 min is needed for
mini-gels.
7. To run the gels, mount them into the electrode assembly clamp
and fill the center/upper chamber with cathode buffer
(Table 1) appropriate for the desired form of native electropho-
resis. After checking this chamber for leaks, pour the anode
buffer into the anode chamber below the gels and remove
bubbles under the gels.
8. Gently pull out well combs and wash wells with the cathode
buffer using a syringe or pipette before loading the gel with the
samples.
9. Run gels in the cold room (4 C) or completely packed in ice:
For CN mini PAGE, use 100 V for the first hour and 200 V
until finished, usually an additional 1–1.5 h. Alternatively, run
the mini-CN PAGE at 30–40 V overnight (see Notes 9 and
10). For BN maxi PAGE, use 100 V and run gels overnight
(about 18 h) (see Note 10).
10. For hybrid gels, change the cathode buffer to the dye-free
cathode buffer (Table 1) as soon as all proteins have entered
the separation gel.
11. For BN PAGE, change the deep blue cathode buffer after
about 1–2 h to the light blue cathode buffer.
Native Electrophoresis 107
3.2 Sample
Preparation
Membrane-bound mitochondrial supercomplexes must be
extracted from the inner mitochondrial membrane (see Note 11).
To preserve the integrity of mitochondrial supercomplexes, use
either freshly isolated mitochondria or samples, which have been
frozen and thawed for only one time as repeated freeze/thaw cycles
can disassemble protein complexes. Calculations/volumes below
are given for mini-gels (the well of a 10-well comb, 1.5 mm thick
gel holds up to 35–40 μL) and maxi-gels (the well of a 15-well
comb holds up to 200 μL).
1. Place desired amount (we use 10–50 μg protein for mini-gels
and 50–200 μg for maxi-gels) of isolated mitochondria or cell/
tissue homogenate in microtubes and centrifuge at 17,000 g
for 10–15 min at 4 C. This step removes some of the soluble
mitochondrial matrix and/or cytosolic proteins.
2. Aspirate and discard the supernatant and add extraction buffer
(Table 1) to amount desired to load onto the gel. Based on our
equipment, we use 30 μL for a mini-gel and 100 μL for a maxi-
gel, limiting the ratio protein/buffer to not more than 2 μg
protein/μL buffer. Resuspend the sediment gently on ice. If
desired, a general protease inhibitor mix might be added at this
point.
3. Add detergent. We use 2–4 μg lauryl maltoside/μg protein and
4–6 μg digitonin/μg protein to solubilize isolated mitochon-
dria. More detergent is needed if a cell or tissue homogenate is
used (see Note 12).
4. Incubate on ice for 10–30 min and gently mix at beginning and
occasionally during incubation by trituration and/or agitation
of tube.
5. Centrifuge at 17,000 gfor 10 min at 4 C to remove any
membrane and tissue fragments.
6. Transfer supernatant to a new tube. For hrCN samples, add
1μL hrCN loading buffer for every 10 μL sample volume. The
total volume of the sample should be approximately 40 μL for a
mini-gel well (130 μL for a maxi-gel). For BN and hybrid
samples, add Coomassie G250 to the samples so that the
ratio of dye:detergent is 1:4 (w/w) (see Note 13).
7. Load 30 and 120 μL of the samples into the wells of the mini-
or maxi-gel, respectively. The remaining 10 μL from each
sample is for a denaturing SDS gel to detect VDAC by immu-
noblotting as a loading control (seeNote 14).
8. Load a suitable molecular weight marker into a designated well
of the gel.
9. Run gels as outlined and transfer proteins onto membranes.
Wet- and semidry transfer conditions are possible. The transfer
conditions and timing depend on the available equipment (see
Note 15).
108 Gisela Beutner and George A. Porter Jr.
3.3 Immunoblotting 1. When the transfer is finished, place the membrane into Pon-
ceau S solution to visualize all transferred proteins. After
5–10 min, wash the excess Ponceau S off by rinsing the mem-
brane 3–4 times with deionized H
2
O. Label the position of the
marker proteins on the membrane with pencil and document
Ponceau S stained membrane (photograph or scan).
2. Wash membrane three times for 10 min with TBS with gentle
agitation to destain the transferred proteins.
3. Block membrane with milk solids/TBS for 1–2 h at room
temperature or overnight in a cold room with gentle agitation.
4. Wash membrane for three times 10 min with TBST with gentle
agitation.
5. Incubate with primary antibody overnight in the cold room
with gentle agitation (see Note 16).
6. Remove antibody and wash membrane for three times 10 min
with TBST with gentle agitation.
7. Incubate membrane with secondary antibody for 1–2 h at
room temperature with gentle agitation (see Note 17).
8. Wash membrane three times for 10 min at room temperature
with TBST with gentle agitation.
9. Proceed to detection of signal (see Note 18).
4 Notes
1. Important: All equipment used for native PAGES must be
clean of detergent. To ensure this, wash all equipment with
0.1 M hydrochloric acid, followed by extensive rinsing with
deionized H
2
O.
2. Coomassie G250 may interfere with the transfer of proteins as
it binds tightly to nitrocellulose membranes. Coomassie 250G
may also interfere with the ability of antibodies to detect the
target proteins [2,6,7].
3. The pH of protocols published for a hybrid of blue and clear/
colorless native electrophoresis varies from 7.0 to 7.5
[4,5]. However, a pH of 7.0 at 4 C is used in many protocols,
as it will facilitate the migration of negatively charged proteins.
4. Acrylamide/bisacrylamide: Acrylamide and bisacrylamide are
both carcinogens and neurotoxins. In its liquid forms, acrylam-
ide has a high potential of absorption through the skin and as a
powder acrylamide and bisacrylamide are easily inhaled. Several
suppliers now offer native gradient gels, which are commonly
1.0–1.5 mm thick and offered as mini-gel or analytical gel.
However, hand-casting a gradient gel has the advantage to
Native Electrophoresis 109
optimize the gel for the experimental needs. A disadvantage of
hand-casting a gel is the toxicity of AAB. Assemblies of glass
plates with CN or BN gels can be stored refrigerated wrapped
in plastic with a few mL of 1gel buffer to maintain moisture.
The gels are stable for use for up to a week.
5. Using extraction buffer B results in better separated supercom-
plexes and more defined bands of mitochondrial supercom-
plexes ( [4] and unpublished data). In this extraction buffer,
glycerol increases the stability of the solubilized protein com-
plexes and allows the storage of aliquots of solubilized mito-
chondrial membrane protein complexes at 80 C[4].
6. Native molecular weight marker range in size up to 1200 kD,
which is still below the size of observed supercomplexes. An
approximate estimation of the molecular weight of mitochon-
drial supercomplexes is possible using the considerations out-
lined in [9].
7. To prevent trapping bubbles in the connecting tube between
the two chambers of the gradient maker, add the heavy solution
into the outflow chamber, open the stopcock connection
between the heavy and light chamber gently and allow a drop
of solution to go through to the other side. This pushes air
bubbles from the connecting tube and stopcock. This cannot
be done if both sides have already been filled because the equal
pressure will prevent the bubble from moving through.
8. The entire content of the gradient mixer must be used for the
specified gradient. If necessary, quantities may be adjusted to
the available equipment.
9. The current during the run will be very low (<50 mA), so a
power supply that can handle these conditions is needed.
10. As the focus of this protocol is on mitochondrial supercom-
plexes with a high molecular weight, protein complexes with a
molecular weight of less than 100 kDa will run out of the mini-
gel using this protocol. Shorter running times for the electro-
phoresis can be used to retain low molecular weight complexes,
with the disadvantage of a poor resolution of the supercom-
plexes. Alternatively, protein complexes may be separated on
maxi-gels.
11. To preserve mitochondrial supercomplexes, it is important to
use either freshly isolated mitochondria or tissue homogenates,
which have been frozen and thawed for only one time.
Membrane-bound mitochondrial supercomplexes must be
extracted from the mitochondrial membranes with mild
anionic or zwitterionic detergents [2]. Dodecyl-maltoside
and digitonin are most commonly used.
110 Gisela Beutner and George A. Porter Jr.
12. To solubilize mitochondrial membranes, 2–10 μg detergent
per μg protein is needed. A series of concentrations of deter-
gent should be used to evaluate the best concentration for
experiment. Generally, more detergent is needed to solubilize
a tissue or cell homogenate. The final concentration of deter-
gent, which will solubilize the mitochondrial membranes with-
out compromising the integrity of supercomplexes, must be
determined experimentally and may vary within different lots
of a detergent.
13. The amount of Coomassie G250 depends on the amount of
detergent used and varies from 1 μg dye per 4–0 μg detergent
[10]. The dye binds to the detergent/detergent micelles and
will migrate in the running front ahead of the solubilized
protein complexes.
14. Sample preparation for native gels is a multistep process and
may involve centrifugations to remove debris or undissolved
tissue fragments. To account for protein loss during the sample
preparation, it is recommended to retain a small aliquot of each
sample (generally 10 μL), and denature it by the addition of,
for example, SDS-containing Laemmli sample buffer. These
samples can then be run on a denaturing SDS gel for the
detection of a suitable marker protein like the outer mitochon-
drial membrane protein VDAC (voltage-dependent anion
channel) as a loading control.
15. After completing the separation of the protein complexes by
hrCN or hybrid electrophoresis, they may be transferred onto
nitrocellulose or PVDF membranes for Western blotting
[4,11]. To minimize interference by Coomassie G250,
which binds tightly to the membranes, the transfer is done
preferably after hrCN or hybrid electrophoresis. An advantage
for using nitrocellulose membranes over PVDF membranes is
that the membrane may be stained with Ponceau S after the
transfer to assess protein loading and transfer efficiency, while
PVDF membranes yield better defined bands. Proteins can be
transferred by wet- or semidry transfer. Native transfer has
been described for wet-transfer [10], while denaturing transfer
protocols can be used for wet- [6,11] and semidry conditions.
To facilitate transfer, the protein complexes can be denatured
by placing the gel for 15–60 min into transfer buffer (for
example, 25 mM Tris and 200 mM glycine; pH 8.3 and add
0.0005 g/L SDS, and 200 mL/L methanol; stored at room
temperature). Adding β-mercaptoethanol or dithiothreitol will
prevent protein oxidation in the gel and will further facilitate
protein transfer. The transfer conditions and timing depend on
the equipment. The membranes are then suitable for Western
blots.
Native Electrophoresis 111
16. Dilute antibody (e.g., 1:2000 for anti-ATP5A, or 1:1000 for
NDUFB6) in BSA/TBS and incubate the membrane for 1–2 h
at room temperature or overnight in a cold room with gentle
rocking. Most antibodies will give a better signal on mem-
branes from native gels when incubated overnight.
17. Dilute the secondary antibody according to the instructions of
the supplier in milk solids/TBS. Dilutions may range from
1:5000 to 1:500,000. Secondary antibodies may be conju-
gated to fluorochromes, horseradish peroxidase, or alkaline
phosphatase for detection.
18. The detection of proteins is determined by the choice of the
secondary antibody and the available equipment. Image sys-
tems to capture the Western blot signal not only visualize the
signal but have also options to analyze it.
References
1. Schagger H, von Jagow G (1991) Blue native
electrophoresis for isolation of membrane pro-
tein complexes in enzymatically active form.
Anal Biochem 199(2):223–231
2. Wittig I, Karas M, Schagger H (2007) High
resolution clear native electrophoresis for
in-gel functional assays and fluorescence stud-
ies of membrane protein complexes. Mol Cell
Proteomics 6(7):1215–1225. https://doi.
org/10.1074/mcp.M700076-MCP200
3. Grandier-Vazeille X, Guerin M (1996) Separa-
tion by blue native and colorless native poly-
acrylamide gel electrophoresis of the oxidative
phosphorylation complexes of yeast mitochon-
dria solubilized by different detergents: specific
staining of the different complexes. Anal Bio-
chem 242(2):248–254. https://doi.org/10.
1006/abio.1996.0460
4. Cuillerier A, Burelle Y (2019) Hybrid clear/
blue native electrophoresis for the separation
and analysis of mitochondrial respiratory chain
supercomplexes. J Vis Exp (147). https://doi.
org/10.3791/59294
5. Newell C, Khan A, Sinasac D et al (2019)
Hybrid gel electrophoresis using skin fibro-
blasts to aid in diagnosing mitochondrial dis-
ease. Neurol Genet 5(3):e336. https://doi.
org/10.1212/NXG.0000000000000336
6. Beutner G, Porter GA Jr (2017) Analyzing
supercomplexes of the mitochondrial electron
transport chain with native electrophoresis,
in-gel assays, and electroelution. J Vis Exp
(124). https://doi.org/10.3791/55738
7. Wittig I, Carrozzo R, Santorelli FM et al
(2007) Functional assays in high-resolution
clear native gels to quantify mitochondrial
complexes in human biopsies and cell lines.
Electrophoresis 28(21):3811–3820. https://
doi.org/10.1002/elps.200700367
8. Wittig I, Schagger H (2009) Native electro-
phoretic techniques to identify protein-protein
interactions. Proteomics 9(23):5214–5223.
https://doi.org/10.1002/pmic.200900151
9. Wittig I, Beckhaus T, Wumaier Z et al (2010)
Mass estimation of native proteins by blue
native electrophoresis: principles and practical
hints. Mol Cell Proteomics 9(10):2149–2161.
https://doi.org/10.1074/mcp.M900526-
MCP200
10. Schagger H (2001) Blue-native gels to isolate
protein complexes from mitochondria. Meth-
ods Cell Biol 65:231–244
11. Beutner G, Alanzalon RE, Porter GA Jr (2017)
Cyclophilin D regulates the dynamic assembly
of mitochondrial ATP synthase into syntha-
somes. Sci Rep 7(1):14488. https://doi.org/
10.1038/s41598-017-14795-x
112 Gisela Beutner and George A. Porter Jr.
Chapter 8
Measuring Mitochondrial Hydrogen Peroxide Levels
and Glutathione Redox Equilibrium in Drosophila Neuron
Subtypes Using Redox-Sensitive Fluorophores and 3D
Imaging
Lori M. Buhlman, Petros P. Keoseyan, Katherine L. Houlihan,
and Amber N. Juba
Abstract
Disruptions in mitochondrial redox activity are implicated in maladies ranging from those in which cells
degenerate to those in which cell division is unregulated. This is not surprising given the pivotal role of
mitochondria as ATP producers, reactive oxygen species (ROS) generators, and gatekeepers of apoptosis.
While increased ROS are implicated in such a wide variety of disorders, pinpointing the cause of their
hyperproduction is challenging. Elevated levels of ROS can result from increases in their production and/or
decreases in their turnover. Disruptions in and/or hyperactivity of NADH-ubiquinone oxidoreductase or
ubiquinone-cytochrome c oxidoreductase can cause excessive ROS generation. Alternatively, if respiration
is functioning in a homeostatic manner, decreases in levels or activity of antioxidants like glutathione,
CuZn- and Mn-superoxide dismutase, and catalase could result in excessive ROS. Because of the diversity of
disorders in which oxidative damage occurs, the most effective therapeutic strategies may be those that
address the putatively diverse causes of increased ROS. Strategies for determining antioxidant activity
typically involve semiquantitative measurement of relative protein levels using immunochemistry and
mass spectrometry. These methods can be applied to a variety of samples, but they do not lend themselves
to detection of cell-specific analyses within tissue like brain.
Because we are interested in elucidating the cause of oxidative stress in selectively vulnerable brain
neurons, we have taken advantage of the easily manipulatable genetics and high fecundity of the fly.
Using a cell type-targeting approach, we have driven redox sensitive green fluorescent proteins (roGFP2)
into the mitochondria of tyrosine hydroxylase-producing (dopaminergic) neurons. In oxidizing conditions,
the fluorophore’s maximal excitation wavelength reversibly shifts. Therefore, the relative amount of
mitochondrial protein oxidation can be determined by taking the ratio of fluorescence excited with two
different lasers. In addition, these GFPs have been independently fused to human glutaredoxin-1 (mito-
roGFP2-Grx1) and yeast oxidant receptor peroxidase (mito-roGFP2-Orp1), facilitating measurements of
relative mitochondrial glutathione redox potential and H
2
O
2
levels, respectively. In order to obtain a more
comprehensive observation of redox states, we capture 3D images of roGFP2 excited by two different lasers.
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_8,
©Springer Science+Business Media, LLC, part of Springer Nature 2021, Corrected Publication 2022
The original version of this chapter was revised. The correction to this chapter is available at https://doi.org/
10.1007/978-1-0716-1266-8_33
113
Mito- and cytoplasmic-roGFP2-Grx1 and -Orp1 expression can be driven by hundreds of genetic drivers in
Drosophila, facilitating fixed or living whole organism or tissue- and cell-specific redox measurements.
Key words Drosophila, Redox, Mitochondria, Dopamine, Parkinson’s disease, 3D imaging, Glutathi-
one, Reactive oxygen species, Hydrogen peroxide, Antioxidant
1 Introduction
1.1 Redox Sensitive
Green Fluorescent
Proteins to Detect
Glutathione Redox
Equilibrium and H
2
O
2
Levels
Reduction-oxidation sensitive green fluorescent proteins (roGFPs)
have two surface-exposed cysteine residue substitutions that form a
structure-modifying disulfide bridge when oxidized [1]. In oxidiz-
ing conditions, the fluorophore’s maximal excitation wavelength
reversibly shifts from about 490 to 400 nm. Therefore, the relative
amount of protein oxidation can be determined by taking the ratio
of fluorescence emission when the fluorophore is excited with
405 and 488 nm lasers. Fusion of human glutaredoxin-1 (Grx1)
to roGFP2 increases roGFP2 sensitivity for detection of the gluta-
thione redox equilibrium component of oxidative stress [2]. Fusion
of yeast oxidant receptor peroxidase-1 (Orp1) to roGFP2 allows for
high-sensitivity detection of relative H
2
O
2
levels [3]. Oxidation
states of Grx1 and Orp1 come to equilibrium with roGFP oxida-
tion (Fig. 1), and roGFP2 oxidation is reversible, allowing for
dynamic live system imaging [4]. We utilize mito-roGFP2-Grx1
and mito-roGFP-Orp1 to capture GSH redox potential and relative
H
2
O
2
levels in mitochondria of fixed brain tissue [2,4].
Fig. 1 Mechanisms of mito-roGFP2 detection of glutathione redox equilibrium
and H
2
O
2
levels. When human glutaredoxin 1 (Grx1) is fused to roGFP, the
electron from oxidized glutathione disulfide (GS-SG) is rapidly and reversibly
transferred to roGFP by Grx1 due to their proximity (a). Similarly, an electron
from H
2
O
2
is transferred to roGFP by microbial oxidant receptor peroxidase
1 (Orp1) when Orp1 is fused to roGFP (b). Maximal excitation wavelength of
nonoxidized and oxidized roGFP is 490 and 400 nm, respectively. (Reprinted
from Cell Metabolism, 14/6, Simone C. Albrecht, Ana Gomes Barata, Jo¨rg
Großhans, Aurelio A. Teleman, Tobias P. Dick, In Vivo Mapping of Hydrogen
Peroxide and Oxidized Glutathione Reveals Chemical and Regional Specificity of
Redox Homeostasis, 819–829. Copyright 2011, with permission from Elsevier)
114 Lori M. Buhlman et al.
1.2 Directing
Cell-Specific Gene
Expression
in Drosophila
A relatively simple genome, genetic conservation, and high fecun-
dity make Drosophila an attractive in vivo model of human develop-
ment and disease. More than 75% of human disease-implicated
genes have Drosophila homologs, and the simpler Drosophila
genome offers less redundancy in signaling pathways [5]. Targeted
gene expression can be achieved by using yeast transcriptional
activator GAL4 to activate expression of transgenes when a geno-
mic enhancer sequence is inserted upstream of GAL4 (Fig. 2)
[6,7]. Drosophila stocks expressing GAL4 under the control of
hundreds of different genomic enhancers can be purchased from
the Bloomington Stock Center at the University of Indiana, Bloo-
mington (https://bdsc.indiana.edu/). Thousands of transgenes
under the control of GAL4 are also available from the Bloomington
Stock Center. Tyrosine hydroxylase (TH) is selectively expressed in
Drosophila dopaminergic cells; here we describe the use of a GAL4
construct in conjunction with a TH genomic enhancer to direct
expression of mito-roGFP2-Grx1 and mito-roGFP2-Orp1 into
dopaminergic neuron mitochondria.
1.3 Capturing
and Measuring
Oxidation Status
of Mito-roGFP2-Grx1
and Mito-roGFP-Orp1
in Drosophila
Dopaminergic Neuron
Mitochondria
To observed GSH redox equilibrium and relative H
2
O
2
levels, fly
brains are dissected, fixed, mounted on microscope slides, and
imaged with a confocal microscope. In order to define boundaries
for roGFP measurements, we immunolabel dopaminergic neurons
with an anti-TH antibody following fixation. We use Image-Pro
Premier 3D image processing software (Media Cybernetics, Rock-
ville, MD) to measure total volumes of mito-roGFP2 emissions
excited by 405 and 488 nm lasers. The ratio of the total volumes
of mito-roGFP2 excited by the 405–488 nm laser is reported for
1 region per brain.
Table 1
Confocal microscope settings used to capture z-stacks of nonoxidized roGFP, oxidized roGFP and the
Alexa Flour 594 nm-conjugated secondary antibody used to label tyrosine hydroxylase. The
sequences indicate the order in which z-stacks for each fluorophore were captured (i.e., nonoxidized
roGFP, followed by oxidized roGFP, and then Alexa Fluor 594). While the excitation lasers and
emission ranges indicated should be used for these fluorophores, the gain and offset can be
optimized to best capture these fluorophores depending on their intensities in different contexts.
Laser intensity can also be optimized; however, it should be kept as low as possible to avoid
excitation of nontargeted fluorophores. Optimized settings must remain unchanged when capturing
z-stacks for samples that will be compared to one another
Sequence Excitation laser (nm) Laser intensity (%) Emission (nm) Gain Offset
1 488 15 500 to 530 800 1
2 405 15 500 to 530 800 1
3 532 15 600 to 699 700 0
In vivo Mitochondrial Redox Reporters 115
2 Materials
Prepare and store all solutions at room temperature unless other-
wise stated in the method.
2.1 Drosophila
Stocks
1. TH-driven GAL4 transcription activator expression (Bloo-
mington Stock Center, Bloomington, IN. Item no, 8848):
Genotype: w[*]; +; TH-GAL4 (see Notes 1 and 2)
2. Expression of mito-roGFP2-Grx1 under the control of GAL4
(Bloomington Stock Center Item no. 67664):
Genotype: w
1118
; UAS mito-roGFP2-Grx1; + (see Note 3)
3. Expression of mito-roGFP2-Orp1 under the control of GAL4
(Bloomington Stock Center Item no. 67667):
Genotype: w
1118
; UAS mito-roGFP2-Orp1; +
Store all stocks at ambient temperature and light cycle.
Fig. 2 GAL4-UAS driven gene expression in Drosophila. Yeast transcription activator GAL4 binds to upstream
activating sequence (UAS) response element to initiate transcription of downstream genes of interest. GAL4
transcription is mediated by an upstream enhancer that determines expression. Crossing (female) flies
harboring a cell-specific enhancer upstream of GAL4 with (male) flies harboring UAS will yield an F1
generation that has cell-specific expression of any gene located downstream of UAS. In this chapter, we
use the tyrosine hydroxylase enhancer to drive expression of mito-roGFP2-Grx1 and mito-roGFP-Orp1.
(Reprinted from J. Vis Exp, 129/56174, Seth M Kelly, Alexandra Elchert, Michael Kahl, Dissection and
Immunofluorescent Staining of Mushroom Body Photoreceptor Neurons in Adult Drosophila melanogaster
Brains. Published online. Copyright 2017, with permission from MyJoVE Corporation [6])
116 Lori M. Buhlman et al.
2.2 Food Preparation 1. Methylparaben, Tegosept (antifungal agent), molecular grade
ethanol, propionic acid
2. Molasses, light corn syrup, NutriSoy
®
Flour, active dry yeast,
yellow cornmeal, Nutri-Fly
®
Drosophila Agar, Gelidium
3. Large rice cooker
4. Drosophila food pourer
5. Standard Drosophila vials, Flugs
®
to cap vials
6. Standard Drosophila vial racks
7. Dish cloth
8. Add Glad
®
Press and Seal
®
9. Lab refrigerator
2.3 Anesthetizing
Flies
1. Carbon Dioxide UN1013 Cylinder, gas regulator
2. CO
2
Bubbler Kit, T-fitting, Flypad, Flypad Frame, Clear Dro-
sophila polyurethane tubing 1/8 in, 3 mm (Genesee Scientific,
San Diego, CA)
3. Gas regulator, polypropylene Erlenmeyer flask
4. Polyvinyl chloride tubing 3/6 5/16, safety tip air blow gun
with hook, Spalding inflating needles
2.4 Drosophila Brain
Dissection
and Immunolabeling
1. 1phosphate buffer with Triton X-100 (PBT): Add 10 mL of
10phosphate-buffered saline (PBS), 100 μL Triton™X-100
to 90 mL of RO water.
2. 0.3% PBT: Add 50 mL of 10X PBS, 1.5 mL Triton™X-100 to
488.5 mL of RO water.
3. Fixing solution (3.7% formaldehyde): Add 50 μL of a 37%
formaldehyde solution to 450 μL1PBS and mix well in
each well of a glass spot plate. Prepare fresh and follow waste
disposal regulations.
4. Blocking solution (10% normal goat serum/NGS): Add
500 μL of normal goat serum to 4.5 mL of 1PBT. Prepare
fresh and use ~500 μL per glass well.
5. 2 μM NEM (n-ethylmaleimide): Prepare fresh in 1PBS,
about 20 μL per brain.
6. Orbital shaker (e.g., Benchmark BlotBoy)
7. Dissection plates and tools: Pyrex™glass spot plates, Nunc™
72 well MircoWell
®
plate, fine-tip transfer pipette. We recom-
mend using two Dumont #5 forceps to perform the
dissections.
8. Zeiss SteREO Discovery V8 dissecting microscope (Carl Zeiss
Microscopy, LLC., Thornwood, NY)
In vivo Mitochondrial Redox Reporters 117
9. Antibodies: Anti-tyrosine hydroxylase Rabbit #AB152MI
(MilliporeSigma, Burlington, MA), goat anti-rabbit Alexa
Fluor 594 (Thermo Fisher Scientific). Prepare fresh dilutions
for each experiment and avoid repeated freeze–thaw cycles.
2.5 Mounting
the Brains
on Microscope Slides
1. SuperFrost
®
microslides and micro cover glass (22 22 mm)
2. Molecular grade glycerol diluted to 70% in RO water, roughly
10 μL per 5 brains
3. ProLong™Diamond Antifade Mountant (Thermo Fisher Sci-
entific, Waltham, MA)
4. Sally Hansen
®
Hard as Nails
®
Hardener
5. Fine tip transfer pipette
6. Kimwipes
®
7. Microscope slide box
2.6 Dopaminergic
Neuron Cluster Image
Capture
1. Leica Microsystems™TCS SPE Confocal Microscope with
405 nm, 488 nm, and 532 nm excitation lasers (Leica Micro-
systems, Inc. Buffalo Grove, IL)
2. 63confocal microscope objective: ACS APO, oil, NA ¼1.3,
WD ¼0.16 mm
3. Optical table to reduce microscope vibration such as Kinetic
Systems Series 9100 Vibration Isolation Table
4. Compressed air tank with a CGA 346 regulator and connecting
hoses (for the table)
5. Leica Application Suite X (LASX) software (Leica
Microsystems)
6. Zeiss™Immersol™518F objective oil (Fisher Scientific)
7. Lens paper
2.7 Measuring roGFP
Fluorescence
Reporters
1. Image-Pro Premier 3D software version 9.3.3 or higher
(Media Cybernetics)
2. Image-Pro Premier 3D minimum computer system
requirements:
(a) Operating system: Windows 7, 8.1, or 10 64-bit
(b) Free disk space: Multi high-speed SATA hard disks or
SSDs; 8 GB free on installation drive, plus free space for
images (20+ GB)
(c) Processor: 2.8 GHz CPU Intel quad-core processor or
better
(d) RAM: 16GB or higher memory
(e) Graphics card: NVIDIA GeForce GTX cards with 4 GB
graphics memory and OpenGL 4.2 or higher
(f) Drives: DVD-ROM drive, if optional DVD purchased
118 Lori M. Buhlman et al.
(g) USB port: Required for hard licenses and offline license
activation
(h) Internet connection: Required for online services such as
automatic updates, support links, and access to video
tutorials
(i) Internet browser: Internet Explorer version 9 or higher
3 Methods
Each step should be carried out at room temperature unless other-
wise specified.
3.1 Standard Corn
Meal Molasses
Drosophila Food
Preparation
1. Prepare 10% methylparaben by adding 100 g of Tegosept, an
antifungal agent, to 1 L of molecular grade ethanol and
mix well.
2. Add 3600 mL of reverse osmosis (RO) water to a large rice
cooker and bring to a boil.
3. Add 490 mL of RO water to a 1 L graduated cylinder then
slowly add 247.1 mL of molasses, 247.1 mL of light corn syrup
and mix well.
4. Add 620 mL of RO water to a 1 L beaker. Add a stir bar
followed by 41.2 g of NutriSoy
®
Flour and mix well. Once
the soy flour is thoroughly mixed, add 82.4 g of active dry yeast
and mix until well combined.
5. Measure 411.8 g of yellow cornmeal and 41.2 g of Nutri-Fly
®
Drosophila Agar, Gelidium into a 2 L beaker, then add
1400 mL RO water and mix with a spoon.
6. Add in the following order to the boiling water in the rice
cooker: cornmeal/agar mixture, soy flour/agar, molasses/
corn syrup and continually stir with a spoon to avoid clumping
and burning.
7. Bring to a total volume of 7 L with RO water.
8. Allow food to boil while stirring for 10 min.
9. Turn off the rice cooker and allow food to cool to 70 C.
10. Add 30.9 mL of propionic acid and 70 mL of 10% methylpar-
aben and mix well.
11. Transfer food to a food pourer or pour by hand into standard
Drosophila vials in racks.
12. Cover with cloth and allow to sit for at least 4 h to overnight.
13. Add Glad
®
Press and Seal
®
to the top of each rack, flip, and
store at 4 C.
14. Allow food to warm to room temperature before transferring
flies. Use Flugs
®
to cap each vial.
In vivo Mitochondrial Redox Reporters 119
3.2 Drosophila
Stocks
and Maintenance
1. Maintain stocks of Drosophila expressing TH-GAL4, UAS--
mito-roGFP2-Grx1 or UAS-mito-roGFP2-Orp1 in plastic
vials with cotton Flugs
®
at ambient temperature and light.
Vials are ¼ to ½ full of standard cornmeal and molasses food
(see Subheading 2.2).
2. Transfer stock flies to a new food-containing vial one to two
times per month by rapidly removing the Flug
®
and inverting
the fly vial over the new one. Tap the flies into the new vial
and plug.
3.3 Crossing
Drosophila to Obtain
Progeny Expressing
Mito-roGFP2-Grx1 or
Mito-roGFP2-Orp1
in Dopaminergic Cells
1. Anesthetize flies by removing the Flug
®
from the vial and
rapidly placing the inverted vial onto a CO
2
pad. Place 10 to
15 virgin female flies (see Note 4) expressing TH-GAL4
(TH-driven GAL4 transcription activator) in a food-containing
vial with 5 to 10 males harboring either UAS-mito-roGFP2-
Grx1 or UAS-mito-roGFP2-Orp1 gene constructs (see Note 5).
2. Transfer flies to a new food-containing vial every 3–4 days;
maintain all vials for 20 days.
3. Collect progeny within 24 h of eclosion from pupa cases, place
them in a new food-containing vial, and transfer them every
3–4 days until the flies are of desired age (roGFP2 expression
can be detected on the day of eclosion and throughout life
span). To avoid collecting f2 generation progeny, which may
not express all of the desired transgenes, flies must not be
collected from vials in which parents were placed more than
20 days prior to collection date (see Note 6).
3.4 Drosophila Brain
Dissection
1. Remove the Flug
®
from the experimental fly vial and rapidly
flip it onto the CO
2
pad to anesthetize the flies. Using dissect-
ing forceps, pick up an anesthetized fly and place it on a
dissecting plate near a droplet of 2 μM NEM
(N-ethylmaleimide; see Note 7). Quickly decapitate the fly
and place the head into the droplet of NEM. Carefully dissect
the brain under a dissecting microscope by removing eye discs
and outer cuticle (see Note 8). Discard the body.
2. Place the brain in a glass spot plate well that contains 3.7%
formaldehyde for 15 min. To avoid over fixation, place each
newly dissected brain into a new well and set a timer for each.
3. Using forceps, transfer the brain from fixing solution to
another well containing 0.3% PBT. Up to five brains of the
same genotype can be transferred to a single well.
4. When all brains have been placed in 0.3% PBT, place the glass
well plate on an orbital shaker on the lowest setting for 5 min.
Repeat this washing step four times by using a fine-tip transfer
pipette to discard used solution and add fresh 0.3% PBT (see
Note 9). Brains remain in the wells during washes.
120 Lori M. Buhlman et al.
3.5 Tyrosine
Hydroxylase
Immunolabeling
1. Using a fine-tip transfer pipette, remove 0.3% PBT from each
well and add 500 μL blocking solution to brain-containing
wells. Keep the glass spot plate on the orbital shaker for at
least 30 min.
2. Prepare a fresh anti-TH primary antibody solution (1:100
primary antibody to blocking solution). Place 10 μL of the
primary antibody solution to each microtiter well.
3. Remove the glass spot plate from shaker and use forceps to
transfer brains from the glass wells to antibody-containing
microtiter wells. Each well can hold up to 5 brains. Be sure to
record which microtiter houses which genotype when using
multiple genotypes or conditions.
4. Place the microtiter plate in an empty pipette box with a moist
towel on the bottom to keep brains from drying and keep the
box at 4 C overnight.
5. Using forceps, transfer the brains from the microtiter plate to a
glass well containing 0.3% PBT. Using a fine-tip transfer pipette
to change solutions, wash the brains in 0.3% PBT four times for
5 min on the orbital shaker.
6. Remove 0.3% PBT and add 500 μL of blocking solution. Place
on shaker for at least 30 min.
7. While the brains are in the blocking phase, prepare a 1:200
solution of secondary antibody (goat anti-rabbit Alexa Fluor
594) to blocking solution. The secondary antibody is light
sensitive, so limit light exposure to solution (see Note 10).
8. At the end of the blocking phase, remove blocking solution and
add 400 μL of secondary antibody solution to each well. Cover
with aluminum foil and place on orbital shaker for 2 h.
9. Remove secondary antibody solution from each well and add
0.3% PBT. Repeat washing procedure four times for 5 min with
agitation. Limit light exposure.
3.6 Mounting
Drosophila Brains
on Microscope Slides
1. Label a microscope slide with genotype, antibody label, fly age,
treatment, and date mounted.
2. Using a fine-tip transfer pipette, place a small drop of 70%
glycerol onto the slide closer to the frosted left side. Drag the
glycerol droplet to the right along the slide with the transfer
pipette. The glycerol line created is a holding place for the
brains.
3. Use forceps to remove one brain at a time from the final wash
and place it in the left region of the glycerol line. This will
deposit a small amount of washing solution into the glycerol
(see Note 11). Carefully drag the brain to the right, keep it in
the upside-down orientation (see Notes 12 and 13). Repeat
In vivo Mitochondrial Redox Reporters 121
this for each brain so that the brains are aligned on the slide
(Fig. 3). Limit the space between the aligned brains so that they
can be easily located during imaging.
4. Use a Kimwipe
®
to remove excess glycerol from the slide—this
is easiest to do under the dissecting microscope. Ensure the
brains have not moved and are still in their correct orientation.
5. Place a small horizontal line of ProLong Antifade above the line
of brains. Carefully place a coverslip over the top of the
mounted brains. Allow the mounting media to spread for
1–2 min, keeping the slide covered to avoid light. Seal the
coverslip to the slide with clear nail polish. When dried, place
slide in a slide box and store at 20 C.
3.7 Capturing
Images of roGFP2s
in Dopaminergic
Neuron Mitochondria
1. Turn on computer and confocal microscope, and open LAS-X
Application. The following steps are executed in the “Acquisi-
tion” pane of the LAS X Application unless otherwise
indicated.
2. Turn on the 405 nm, 488 nm, and 532 nm excitation lasers.
3. Open three sequences and set the conditions as listed below (see
Note 14).
4. Set pinhole to 1 AU.
5. Set the image resolution to 512 512 pixels and speed to
600 Hz (see Note 15).
Fig. 3 Illustration of Drosophila brain mounting procedure. Step 1: Place a drop of
70% glycerol onto the left side of a microscope slide using a fine-tip transfer
pipette. Step 2: Using the fine-tip transfer pipette, drag the glycerol to the right,
forming a line. Steps 3 and 4: Place each brain in the left side of the glycerol and
drag it to the right using forceps. Brains should be in upside-down orientation if
imaging with an inverted microscope. Step 5: Continue adding brains in the
same fashion. Minimize distance between brains to facilitate location of the tiny
brains during image capture. Images are not to scale
122 Lori M. Buhlman et al.
6. Using the 63oil objective and the setting for Sequence
3, locate the brain and the region of interest (ROI) on the
confocal microscope (see Note 16) We recommend using
images in Fig. 1of Mao and Davis, 2013, as a guide in locating
dopaminergic neurons [8].
7. Use the TH antibody signal to define the ROI in the z-plane.
Click the “Live” button in the LAS X Application and manually
manipulate the zposition around your ROI. At the top of the
ROI, click the “Begin” button. Scroll to the bottom of the
ROI and click the “End” button.
8. When the ROI has been determined, change the settings to
capture a higher resolution image (1024 1024 pixels for
image resolution and 400 Hz for capture speed).
9. Set the z-stack step to 340 nm. The z-slice thickness should
remain constant for all samples; the number of slices per sample
will increase with larger ROI.
10. Select “Between Lines” under the “Sequential Scan” tab so
that each z-slice is excited by all lasers (sequences) before the
confocal moves to the next slice. This pattern continues until
the all z-slices in the ROI have been captured.
11. Click “Start” to take image—this can take several minutes
depending on the size of the ROI in the z-plane.
12. Save the image in .lif format with a descriptive title.
3.8 Image
Processing
1. Open Image-Pro Premier 3D Application and open .lif file.
2. Under “3D View” tab, click the “Reload” button to open the
“Load Image” window. The voxel sizes should read:
X¼0.114, Y¼0.114, Z¼0.340 (μm). These settings should
be constant among all samples as they are determined by image
capture settings.
3. In the same window, uncheck “Auto” and click “Reset” to
change all values to 1 to avoid subsampling and obtain the
most sensitive measurements.
4. Under the “3D View” tab, click the “Show VOI” and “Show
VOI handles” icons to allow adjustment of volume of interest
(VOI). Using the arrow tool from the “3D View” panel, grab
the green square handles on the sides of the image to adjust
the VOI.
5. Click the “Add Iso-Surface” icon in the right panel to initiate
data selection. This feature will add a pseudo “Iso-Surface” to
identify the fluorescence that will be measured.
(a) Select Channel 2 (405 nm, which was sequence 2 during
image capture) from the drop-down menu (see Note 17).
(b) Uncheck “Auto” and click “Reset” to change all values
to 1.
In vivo Mitochondrial Redox Reporters 123
(c) Select “Lo-Pass 3x3x3” from “Filter” drop-down menu
(see Note 18).
(d) Select “Auto Bright” from the “Threshold type” drop-
down menu.
(e) Uncheck “Execute Count.”
6. Repeat steps 5a to 5e for Channel 1.
7. For Channel 1 (Sequence 1) and Channel 2 (Sequence 2)
Volume Surfaces, set the minimum threshold (“From” win-
dow) to 50 and 95, respectively. The maximum threshold (“to”
window) can be determined automatically (see Note 19).
8. Under the “3D Measure” tab, click “Types” and select
“Volume: Volume.” Set the starting range to 0 and end range
to 1 10
308
. Remove any unwanted measurement parameters.
9. In the “3D Measure” tab, make sure the “Ranges” icon is
checked.
10. Click “Count” for Channel 1 and 2 Volume Surfaces.
11. In order to count mitochondrial objects within the TH-labeled
neurons, use the arrow tool under the “3D View” tab to select
and delete all iso-surfaces located outside the TH antibody-
stained (blue) areas. When an object is selected, press the delete
button on the keyboard to delete the object.
12. Under the “3D Measure” tab, click “Data Table,” then click
on “Microsoft Excel” icon to export the data to an Excel
spreadsheet.
13. Repeat steps 11 and 12 for Channel 1 (488 nm).
3.9 Data Analysis 1. Calculate the sum of all object volumes for each channel.
2. The relative H
2
O
2
levels or GSH redox equilibrium for each
sample is determined by reporting the ratio of the sum of
emission volumes excited by the 405 nm laser (Channel 2) to
the sum of volumes excited by the 488 nm laser.
4 Notes
1. Drosophila chromosomes are indicated by their mutations or
transgene symbols. When reporting Drosophila genotypes,
three of the four chromosomes are included, and they are
separated by a semicolon. Heterozygous chromosomes are
separated by “/.” If a chromosome has no known mutations
or transgenes, it is indicated with “+.” For example, the
TH-GAL4 stock genotype is written as follows: w[*]; +;
TH-GAL4.
124 Lori M. Buhlman et al.
2. The w[*] mutation occurs in the gene conferring wild-type red
eye color, and it is located on the first chromosome. Flies with
this mutation have white eyes. TH-GAL4 constructs “rescue”
eye color in w[*] mutants by conferring peach eye color.
3. w
1118
is another mutation in the gene conferring wild-type red
eye color, and flies with this mutation have white eyes. Both
UAS mito-roGFP2 constructs “rescue” eye color in w
1118
mutants by conferring peach eye color.
4. Newly eclosed female Drosophila are not fertile until 8 h after
eclosion at 25 C and after 16 h at 18 C. To collect virgin
females at room temperature, discard females of unknown age
from a vial in which pupa have formed, and collect newly
eclosed females that are anesthetized on a CO
2
pad within
7 h to ensure that they are unfertilized. For sexing and pheno-
typing images and information, we recommend the Atlas of
Drosophila Morphology: Wild-type and Classical Mutants,by
Chyb and Gompel [9].
5. Crossing strategies are set up as follows:
Genotype of female parents XGenotype of male parents.
Genotype of progeny.
Crossing strategies used to generate expression of mito-
roGFP2 reporters:
w[*] ; + ; TH-GAL4 Xw
1118
; UAS mito-roGFP2-Grx1 ; +
w[*]/w
1118
; UAS-mito-roGFP2-Grx1/+ ; TH-Gal4/+
w[*] ; + ; TH-GAL4 Xw
1118
; UAS mito-roGFP2-Orp1 ; +
w[*]/w
1118
; UAS-mito-roGFP2-Orp1/+ ; TH-Gal4/+
6. Pupae are visible as soon as 4 days after parent flies are placed in
a vial. Adult flies begin to elcose from pupa cases about 10 days
after parents are placed in a vial. Newly eclosed adults will
fertilize one another, so parent cross vials must be discarded
within 20 days after parents are placed in a food vial to avoid f2
generation contamination.
7. NEM is used to prevent additional oxidation of roGFP2 during
dissection [4]. Brains can be dissected in 1 PBS droplets
when redox sensitive reporters are not used.
8. We found it helpful to begin dissection in one drop of 2 μM
NEM. After crude dissection has been completed, move the
brain to a new droplet to remove any remaining tissue or cuticle
for a cleaner dissection (see https://www.youtube.com/watch?
v¼j4rVa7JCzdg for a useful instructional video [10]).
9. The washing step is easiest to do under a dissecting microscope
with a fine-tip transfer pipette—be careful not to aspirate or
damage the brain when removing previous washing solution.
In vivo Mitochondrial Redox Reporters 125
Additionally, when adding new solution back into a well,
ensure the brains are in solution before placing back on shaker.
If they stick to the glass above liquid level, they will become dry
and unusable. Keep a cover over the wells to ensure no debris
or dust falls into the wells when brains are on the orbital shaker.
10. To reduce light contamination, use foil to place over any tubes
or containers that house the secondary antibody. Additionally,
place a light-reducing cover (e.g., foil covered pipette tip box
top) over the glass well during the secondary staining step.
11. We recommend discarding the final wash solution from one
well at a time, just prior to mounting. This will prevent brains
from drying and reduce the amount of solution left on the
forceps in order to limit the amount of liquid on slide.
12. Placing the brains in the left end of the glycerol line and then
dragging them along the small line of glycerol limits the
amount of excess solution and decreases brain displacement
when top coverslip is added.
13. When brains are mounted upside down on the microscope
slide, they will be in the upright orientation during imaging
on an inverted microscope.
14. These are the settings we found to be optimized for our con-
ditions. Laser intensities, gain, pinhole, and z-stack may vary
depending on your conditions, experiment, and/or equip-
ment. Keep conditions consistent across all samples to reduce
analysis variability.
15. Use lower resolution settings to reduce photo bleaching while
locating the sample and defining region of interest.
16. Light at 532 nm photo bleaches the sample more slowly than
that at 405 or 488 nm. Begin by finding one brain on slide and
then follow the line back to the closest to the label. It may be
easier to locate the brain using a 20objective or with white
light. Limit light exposure to avoid photo bleaching.
17. Image analysis is initiated with Channel 2 because trachea have
autofluorescence under the 405 nm wavelength laser. Because
trachea can often run through the sample and make image
analysis impossible, we recommend beginning here to deter-
mine sample quality.
18. We have found that using this filter speeds up processing time
without sacrificing measurement specificity.
19. Minimum threshold values for image analysis should remain
consistent between samples, but values may need to be
adjusted depending on the equipment and fluorophore
expression.
126 Lori M. Buhlman et al.
Acknowledgments
This work was supported with intramural funds provided by the
Biomedical Sciences program at Midwestern University. We thank
Giulia Bertolin at the Institute of Genetics and Development of
Rennes for critical feedback on the manuscript.
References
1. Hanson GT, Aggeler R, Oglesbee D,
Cannon M, Capaldi RA, Tsien RY, Remington
SJ (2004) Investigating mitochondrial redox
potential with redox-sensitive green fluores-
cent protein indicators. J Biol Chem 279
(13):13044–13053. https://doi.org/10.
1074/jbc.M312846200
2. Gutscher M, Pauleau A-L, Marty L, Brach T,
Wabnitz GH, Samstag Y, Meyer AJ, Dick TP
(2008) Real-time imaging of the intracellular
glutathione redox potential. Nat Methods 5
(6):553
3. Gutscher M, Sobotta MC, Wabnitz GH,
Ballikaya S, Meyer AJ, Samstag Y, Dick TP
(2009) Proximity-based protein thiol oxida-
tion by H
2
O
2
-scavenging peroxidases. J Biol
Chem 284(46):31532–31540
4. Albrecht SC, Barata AG, Grosshans J, Teleman
AA, Dick TP (2011) In vivo mapping of hydro-
gen peroxide and oxidized glutathione reveals
chemical and regional specificity of redox
homeostasis. Cell Metab 14(6):819–829.
https://doi.org/10.1016/j.cmet.2011.10.
010
5. Pandey UB, Nichols CD (2011) Human dis-
ease models in Drosophila melanogaster and
the role of the fly in therapeutic drug discovery.
Pharmacol Rev 63(2):411–436. https://doi.
org/10.1124/pr.110.003293
6. Kelly SM, Elchert A, Kahl M (2017) Dissection
and Immunofluorescent staining of mushroom
body and photoreceptor neurons in adult Dro-
sophila melanogaster brains. J Vis Exp
(129):56174. https://doi.org/10.3791/
56174
7. Brand AH, Perrimon N (1993) Targeted gene
expression as a means of altering cell fates and
generating dominant phenotypes. Develop-
ment 118(2):401–415
8. Mao Z, Davis RL (2009) Eight different types
of dopaminergic neurons innervate the Dro-
sophila mushroom body neuropil: anatomical
and physiological heterogeneity. Front Neural
Circuits 3:5–5. https://doi.org/10.3389/
neuro.04.005.2009
9. Chyb S, Gompel N (2013) Atlas of Drosophila
morphology: wild-type and classical mutants.
Academic Press, London
10. Wu JS, Luo L (2006) A protocol for dissecting
Drosophila melanogaster brains for live imag-
ing or immunostaining. Nat Protoc 1
(4):2110–2115. https://doi.org/10.1038/
nprot.2006.336
In vivo Mitochondrial Redox Reporters 127
Chapter 9
Assessment of Mitochondrial Cell Metabolism by
Respiratory Chain Electron Flow Assays
Flavia Radogna, De
´borah Ge
´rard, Mario Dicato, and Marc Diederich
Abstract
Cellular energy metabolism is regulated by complex metabolic pathways. Although anaerobic glycolysis was
reported as a primary source of energy in cancer leading to a high rate of lactate production, current
evidence shows that the main energy source supporting cancer cell metabolism relies on mitochondrial
metabolism. Mitochondria are the key organelle maintaining optimal cellular energy levels. MitoPlate™S-1
provides a highly reproducible bioenergetics tool to analyze the electron flow rate in live cells. Measuring
the rates of electron flow into and through the electron transport chain using different NADH and FADH
2
-
producing metabolic substrates enables the assessment of mitochondrial functionality. MitoPlate™S-1 are
96-well microplates pre-coated with different substrates used as probes to examine the activity of mito-
chondrial metabolic pathways based on a colorimetric assay. A comparative metabolic analysis between cell
lines or primary cells allows to establish a specific metabolic profile and to detect possible alterations of the
mitochondrial function of a tumor cell. Moreover, the direct measurements of electron flux triggered by
metabolic pathway activation could highlight targets for potential drug candidates.
Key words Cancer metabolism, Electron transport chain, Mitochondrial respiration, Tricarboxylic
acid cycle, Bioinformatics
1 Introduction
Mitochondria supply cells with ATP by consuming a large quantity
of oxygen in the electron transport chain (ETC) through an elec-
trochemical gradient generated across the five complexes of the
oxidative phosphorylation system (OXPHOS). The main mito-
chondrial oxidative metabolisms involved in energy production
are pyruvate oxidation, the citric acid cycle (also known as the
tricarboxylic acid (TCA) cycle or the Krebs cycle), OXPHOS, and
β-oxidation of fatty acids [1]. The TCA cycle is a pivotal point in
mitochondrial respiration where different dietary metabolic path-
ways converge, leading to oxidative phosphorylation in cells to
satisfy their bioenergetic requirements, synthesize macromolecules,
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_9,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
129
and maintain redox balance [2]. This cycle, through a series of
biochemical reactions occurring in the mitochondrial matrix by
oxidizing fuel sources, produces intermediates such as building
blocks for macromolecule synthesis as well as energy and electron
acceptors used in downstream ETC cellular processes [3] (Fig. 1).
Deregulated TCA cycle function is implicated in several pathologi-
cal processes [4,5]. Mitochondria represent the major site of
reactive oxygen species production [6,7] generating oxidative
stress that may lead to mitochondrial dysfunction and cell death
in several pathologies such as aging [8], cardiac disease [9], diabetes
[10], neurological disorders [11], and cancer [12,13]. In particu-
lar, metabolic reprogramming contributes to tumor transformation
and progression. Thus, bioenergetic changes and metabolic repro-
gramming in tumors represent a new target for cancer therapy
[14,15].
The accumulation of genetic and epigenetic alterations [16]in
cancer cells, particularly those affecting oncogenes and tumor sup-
pressor genes that regulate the cellular metabolism, results in
uncontrolled cell proliferation and metabolic stress [17]. As an
example, we recently investigated the interplay between p53 and
SIRT1 as a target for future therapeutic intervention [18]. More-
over, the common Warburg metabolism was observed in neuro-
blastoma, resulting high glucose uptake, a high rate of lactate
production, and a low rate of oxygen consumption [19]. Evidence
has demonstrated that the amplification status of the N-MYC
oncogene is associated with energy metabolism regulation through
the regulation of genes involved in glycolysis as well as the use of
additional fuel sources for glutamine and fatty acid metabolism
[19,20]. Therefore, tumor cells are able to uncouple glycolysis
from the TCA cycle by using supplementary fuel substrates for fatty
acid metabolism by regulating mitochondrial transporter expres-
sion or the activity of specific dehydrogenase enzymes [2–21].
Whereas an increasing number of compounds are under inves-
tigation as potential modulators of cell metabolism, we investigated
the potential of natural compounds as modulators of cancer cell
metabolism [15]. Natural compounds constitute an important
source of anticancer molecules with chemopreventive and thera-
peutic properties but these compounds are also capable of meta-
bolic reprogramming to target oncogenic metabolic aberrations
[22]. Of particular interest, for example, is the emerging interplay
between cell metabolism and cell death eventually offering targeted
therapeutic options via modulation of hexokinases and glyceralde-
hyde-3-phosphate dehydrogenase [23].
Several approaches are available to measure cell energetics
through mitochondrial respiration, a physiological level or after
compound-mediated reprogramming of the metabolic function.
The perturbed metabolism associated with a specific pathway is
not always related to the energy supply due to full or partial
130 Flavia Radogna et al.
Isocitrate
Citrate
α-Ketoglutarate
Succinate
Fumarate
Malate
Oxaloacetate
Acetyl-CoA
TCA
cycle NADH
NAD+
NADH
NAD+
FAD H2
FAD +
NADH2
NAD+
Succinyl-CoA
GDP
GTP
FH
SDH
MDH
IDH2/3
OGDH
CO2
CO2
Fatty acid
SLC25A10/DIC
SLC25A1/CTP SLC25A11/OGC
MPC1,MPC2/MPC
Pyruvate
NADH
NAD+
CO2
PDH
SLC25A1/CTP
Transporter genes and proteins:
CII
CI Q
cycle
CIII
Cyt C
CIV
e-
e-
FAD H2
NADH
O2H2O
Redox
Dye
MC
ADP ATP
SLC25A20/CAC
SLC25A10/DIC
SLC25A1/CTP
SLC25A11/OGC
Fig. 1 Transporter genes/proteins, enzymes, and biochemical reactions driving mitochondrial electron flow in
the mitochondrial matrix. Pyruvate, the end product of glycolysis, is transported by mitochondrial pyruvate
carriers (MPC1/2) into the mitochondria, where it is processed to produce acetyl-CoA. Through a series of
redox reactions, acetyl-CoA is processed to produce high-energy electrons, which are transported through the
ETC by nicotinamide adenine dinucleotide (NAD
+
) and flavin adenine dinucleotide (FAD) to produce NADH and
FADH
2
. The electron flow rates through the ETC from metabolic substrates provide NADH (red; e.g., L-malate,
α-ketoglutarate, D-isocitrate, L-glutamate, D-β-hydroxy-butyrate) to complex I or FADH
2
(red; e.g., succinate,
glycerol-PO
4
) to complex II. The electrons from complexes I (CI) and CII are used to pump protons across the
mitochondrial cristae and are transferred to the distal portion of the ETC where an oxidizing agent, tetrazolium
redox dye (MC), acts as a terminal electron acceptor that turns purple upon reduction. SLC25A20 (solute
carrier family 25 member 20)/CAC (carnitine-acylcarnitine translocase): The SLC25A20 gene encodes CAC,
which mediates the transport of acylcarnitines of different lengths across the mitochondrial inner membrane
from the cytosol to the mitochondrial matrix for their oxidation by the mitochondrial fatty acid oxidation
pathway. SLC25A1 (solute carrier family 25 member 1)/CTP (citrate, isocitrate, and malate transporter
carrier): The tricarboxylate carrier protein, also known as TCP, is a protein encoded by the SLC25A1 gene
in humans. SLC25A10 (solute carrier family 25 Member 1)/DIC (dicarboxylate carrier): DIC, encoded by
SLA2510, exchanges dicarboxylates including malate, malonate, and succinate in exchange for phosphate,
sulfate, sulfite, or thiosulfate across the inner mitochondrial membrane, thereby providing substrates for
metabolic processes, including the Krebs cycle and fatty acid synthesis. SLC25A11 (Solute Carrier Family
25 Member 11)/OGC (oxoglutarate/malate carrier): The SLC25A11 gene encodes OGC, which transports
2-oxoglutarate across the inner mitochondrial membranes in an electroneutral exchange for malate or other
dicarboxylic acids.AH aconitate hydratase or aconitase, Ccomplex, CS citrate synthase, Cyt C cytochrome C,
FH fumarate hydratase, IDH isocitrate dehydrogenase, MDH malate dehydrogenase, OGDH oxoglutarate
dehydrogenase or α-ketoglutarate dehydrogenase, PDH pyruvate dehydrogenase, SDH succinate
dehydrogenase
Quantification of Cellular Energy Metabolism 131
compensation by other metabolic pathways. Here, we describe a
new bioenergetics tool to assess the rates of electron flow in live
cells with MitoPlate™S-1 (Fig. 2) using cytoplasmic and mito-
chondrial substrates to measure differential use of metabolic path-
ways. This technology is based on a colorimetric analysis of cell
energetics by analyzing electron flux through the mitochondrial
respiratory chain using a 96-well microplate. “Feeding” the cells
with 31 different substrates allows the researcher to selectively
measure electron flux through specific metabolic pathways, thus
measuring mitochondrial energy production. Each substrate fol-
lows different pathways using different transporters and dehydro-
genases to produce nicotinamide adenine dinucleotide (NAD)H or
flavin adenine dinucleotide (FAD)H
2
[21]. The electron flow rate
Cytoplasmic substrates
TCA cycle substrates
Other mitochondrial substrates
1st set 2nd set 3rd set
124
5 6 7 8
910 11 12
13 14 15 16
17 18 19 20
21 22 23 24
25 26 27 28
29 30 31 32
3124
5 6 7 8
910 11 12
13 14 15 16
17 18 19 20
21 22 23 24
25 26 27 28
29 30 31 32
3124
5 6 7 8
910 11 12
13 14 15 16
17 18 19 20
21 22 23 24
25 26 27 28
29 30 31 32
3
123 456789101112
MitoPlateTM with 31 substrates:
1: No substrate
3: α-D-Glucose
3: Glycogen
4: D-Glucose-1-PO4
5: D-Glucose-6-PO4
6: D-gluconate-6-PO4
7: D,L-α-Glycerol-PO4
8: L-Lactic Acid
9: Pyruvic Acid
10: Citric Acid
11: D,L Isocitric Acid
12: cis-Aconitic Acid
13: α-Keto-Glutaric Acid
14: Succinic Acid
15: Fumaric Acid
16: L-Malic Acid
17: α-Keto-Butyric Acid
18: D,L-β-Hyroxy-Butyric Acid
19: L-Glutamic Acid
20: L-Glutamine
21: Alanyl-Glutamine
22: L-Serine
23: L-Ornithine
24: Tryptamine
25: 100 μM L-Malic Acid
26: Acetyl L-Carnitine + 100 μM L-Malic Acid
27: Octanoyl-L-Carnitine + 100 μM L-Malic Acid
28: Palmitoyl-D,L-Carnitine + 100 μM L-Malic Acid
29: Pyruvic Acid + 100 μM L-Malic Acid
30: γ-Amino-Butyric Acid + 100 μM L-Malic Acid
31: α-Keto-Isocaproic Acid + 100 μM L-Malic Acid
32: L-Leucine + 100 μM L-Malic Acid
Fig. 2 The microplate layout for MitoPlate™S-1. Wells are pre-coated with triplicate sets of 31 different
substrates of the cytoplasm (row A-B), TCA cycle (row C-D), and other mitochondrial substrates (row E-H). The
plate can be used to run three assay samples or one sample in triplicate
132 Flavia Radogna et al.
through the ETC originating from metabolic substrates provides
NADH (e.g., L-malate, α-ketoglutarate, D-isocitrate, L-glutamate,
and D-β-hydroxy-butyrate) to complex I [24] or FADH
2
(e.g.,
succinate and glycerol-PO
4
) to complex II [6]. The electrons
from complexes I and II are used to pump protons across the
mitochondrial cristae [25] and are transferred to the distal portion
of the ETC where an oxidizing agent, tetrazolium redox dye (MC),
acts as a terminal electron acceptor that measures reductase activity
by turning purple upon reduction (Fig. 1). The assay provides a
simple, reproducible, and sensitive tool to study mitochondrial
function and metabolism via a cell suspension. In comparison to
other techniques such as mass spectrometry, this technique ana-
lyzes cell metabolism by preserving the mitochondrial membrane.
The mitochondrial membrane is a barrier that selects and transports
nutrients from the cytoplasm to the mitochondrial matrix and has
different transporters and exporters; thus, completely different
metabolic pathways from those of the whole-cell metabolism are
attained. MitoPlate™S-1 represents a new approach based on
cellular energy metabolism that is able to assess the functionality
of substrate transport, dehydrogenases, or the ETC on the mito-
chondrial organelles of permeabilized cells, which may enable the
identification of specific mitochondrial targets for potential drug
candidates.
Here, we investigate the metabolic profile of two N-MYC–
non-amplified SH-SY5Y and SK-N-AS cell lines versus N-MYC–
amplified BE-M17 neuroblastoma cells to identify a metabolic
fingerprinting cellular profile.
Neuroblastoma is a pediatric tumor that derives from the neu-
ral crest and develops into the sympathetic ganglia, paraganglia, and
adrenal gland. Amplification of N-MYC is a genetic marker of
high-stage neuroblastoma tumor with poor prognosis [26]. The
proto-oncogene N-MYC controls several transcriptional response
pathways including cell growth, proliferation, metabolism, and
cellular instability. BE-M17 neuroblastic-cell type (N-Type) cell
line derived from bone marrow of 2-year-old male represents the
more aggressive form with amplified N-MYC compared to
SH-SY5Y (N-Type) with neuron-like characteristic derived from
bone marrow of 4-year-old female and SK-N-AS stromal cells
(S-type) derived from bone marrow metastasis of a 6-year-old
female [27].
2 Materials
2.1 Equipment 1. MitoPlates™S-1 are 96-well microplates pre-coated with
31 substrates, which are dried on the bottom of each well
(BIOLOG, Inc., USA).
Quantification of Cellular Energy Metabolism 133
2. Microplate reader for kinetic reading using an optical density
(OD) read at 590 nm (OD590).
3. Microscope.
4. Multipipette, 300 μL.
5. Centrifuge with a rotor for 50 mL Falcon tubes.
6. CO
2
incubator.
7. 75 cm
2
flasks for cell culture.
2.2 Cells
and Reagents
1. Cell suspension.
2. Saponin with 20–35% sapogenin content or other permeabiliz-
ing agents.
3. Trypan blue.
4. BIOLOG Mitochondrial Assay Solution (MAS; BIOLOG,
Inc., USA), which is osmotically optimized to preserve the
physical structure of the cells following permeabilization.
5. BIOLOG Redox Dye Mix MC (BIOLOG, Inc., USA), which
is used to measure the electron flow to the distal end of the
ETC, where the tetrazolium redox dye (MC) acts as a terminal
electron acceptor that turns purple upon reduction.
3 Methods
The MitoPlate™S-1 assay allows a quantitative assessment of the
activity of the main metabolic pathways using mitochondrial sub-
strates as probes. The microplate layout is characterized by wells
that are pre-coated with 31 different substrates from the cytoplasm
(row A-B), the TCA cycle (row C-D), and other mitochondrial
substrates (row E-H) including an internal control (L-malate at
100 μM in wells G1, G5, G9) as well as one control without
substrate (wells A1, A5, A9), for a total of 32 wells which are
repeated in triplicates. Thus, one plate can be used for three samples
or one sample in triplicate (Fig. 2). Once absorbed by the cells, the
substrates are metabolized and converted by different pathways and
dehydrogenases into an electron flow to provide NADH or FADH
2
to the ETC to oxidize a tetrazolium redox dye (MC) that turns
purple (Fig. 1). The characterization of cellular metabolic profiles
using this approach could be useful in detecting possible trans-
porter or dehydrogenase gene mutations or deregulated mitochon-
drial functions.
As an example, we compared the metabolic profile of N-MYC–
amplified BE-M17 neuroblastoma cells with cell lines devoid of
N-MYC amplification (SH-SY5Y and SK-N-AS cells) to identify a
differential metabolic profile and detect possible alterations of the
mitochondrial function in these neuroblastoma cell models.
134 Flavia Radogna et al.
3.1 Optimization
of Saponin
Concentration and Cell
Density
1. For each cell type, it is essential to optimize the saponin concen-
tration required for adequate membrane permeabilization
before initiating the assay. A suitable starting range of concen-
trations is between 30 and 100 μg/mL, and an estimation of cell
permeabilization using a trypan blue exclusion test is recom-
mended. The percentage of positive trypan blue staining, repre-
senting permeabilized cells, should be approximatively 40–60%.
2. Then, a preliminary MitoPlate™S-1 analysis with different
saponin concentrations should be tested to analyze the rate of
color formation (see Note 1). Elevated saponin concentrations
might nevertheless damage the mitochondrial membrane,
causing the loss of mitochondria-associated electron flow.
3. For each cell model, it is also recommended to optimize the
number of cells per well; 30,000 cells per well should be a good
starting point for solid tumor cell lines. Nevertheless, this
number might be adjusted depending on the cell type and the
extent of color formation.
3.2 Electron Flux
Assay Mix Preparation
1. In advance, prepare a 20 mg/mL saponin stock solution in
sterile water.
2. Prepare an assay mix by combining 2 mL of 2BIOLOG
MAS, 1.33 mL of 6Redox Dye MC, 20 μL of 20 mg/mL
saponin stock solution that correspond to a final concentration
of 100 μg/mL of saponin for the neuroblastoma cells, we used
here as an example (see Note 2), and 650 μL of sterile water to
reach a total volume of 4 mL (see Note 3).
3.3 Cell Suspension 1. Grow neuroblastoma cells (SH-SY5Y, SK-N-AS, and BE-M17)
in RPMI1640 medium supplemented with 10% (v/v) fetal
bovine serum and 1% (v/v) antibiotics/antimycotics in a
75 cm
2
flask at low passage numbers.
2. Wash the cells in 1phosphate-buffered solution, harvest by
adding 5 mL trypsin solution, incubate at 37 C for 4–5 min,
and resuspend in 10 mL of complete medium to neutralize
trypsin.
3. Count the cell number and determine cell viability with a
trypan blue exclusion test. The cells should maintain >95%
viability.
4. Centrifuge the cells in a 50 mL Falcon tube at 340 rcf for 5 min
to obtain a pellet of 3 10
6
cells (see Note 4).
5. Aspirate the supernatant and resuspend the cells in 3 mL of
BIOLOG MAS (1) to reach a final concentration of 1 10
6
cells/mL to reach 30,000 cells/well.
6. To remove clumps, filter the cell suspension through a
70-micron nylon filter or slowly pipette up and down while
avoiding bubbles.
Quantification of Cellular Energy Metabolism 135
3.4 Step-by-Step
Instructions
1. Pipet 30 μL of the assay mix per well into all wells using a
multichannel pipettor and incubate at 37 C and 5% CO
2
for
1 h to allow substrates to fully dissolve (Fig. 3a).
2. During this time, detach the cells and resuspend them in 1
BIOLOG MAS to reach a final concentration of 1 10
6
cells/
mL (Fig. 3b).
3. In all wells, add 30 μL of the cell suspension (3 10
4
cells) per
well using a multichannel pipettor (see Note 5).
4. Load the MitoPlate™S-1 into a microplate reader with auto-
mated temperature-controlled incubation (37 C) to perform a
kinetic reading of the rate of purple color formation at a
wavelength of 590 nm (Fig. 3c).
Assay mix:
BIOLOG MAS
Redox Dye MC
Saponin
1 hr of incubation at 37°C
to allow substrates to dissolve
Kinetic analysis at 37°C
on a microplate reader
using OD590 nm
Harvest cells and resupend
in BIOLOG MAS
(3 × 106 cells/mL)
30 μL / well
30 μL / well
1st sample 2nd sample 3rd sample
a)
b)
c)
Fig. 3 MitoPlate™S-1 assay steps. (a)STEP 1: Prepare an assay mix by combining 2 mL of 2BIOLOG MAS,
1.33 mL of 6Redox Dye MC, 20 μL of a 20 mg/mL saponin stock solution (see Note 4) and 650 μL of sterile
water to reach a total volume of 4 mL. Pipet 30 μL per well of the assay mix using a multichannel pipettor into
all wells and incubate at 37 C for 1 h to allow substrates to fully dissolve. (b)STEP 2: Detach the cells and
resuspend them in 1BIOLOG MAS to reach a final concentration of 1 10
6
cells/mL. In all wells, add 30 μL
of the cell suspension (3 10
4
cells/well) using a multichannel pipettor. (c)STEP 3: Load the MitoPlate™S-1
into a microplate reader, which provides automated temperature-controlled incubation, to perform kinetic
reading of the rate of color formation using OD590 nm
136 Flavia Radogna et al.
3.5 Analysis
of the Results
The kinetic reading is conducted for 4–6 h maximum; then, the
data can be collected when the purple color formation reached a
plateau (Fig. 4a). The initial slope between 30 min and 2 h of
MitoPlate™S-1 incubation is analyzed to measure the electron
flow rate into the ETC for each metabolic substrate (Fig. 4b).
First, the absorbance from the “No substrate” control is subtracted
from that of the other 31 substrates to normalize against the
background signal. In addition, the well that is coated with
100 μM L-malate is used to normalize wells G2-H4 (G6-G8 and
G10-H12) containing specific mitochondrial substrates requiring
L-malate to be transported and metabolized by the cells. Then, the
initial rate is calculated as the slope derived from a linear regression
fitted to the linear portion of kinetic curves. Finally, the mean and
the standard error of the mean (SEM) of the initial rates are
calculated per substrate and per condition. Interesting substrates
are defined as outliers of the linear regression model by calculating
Cook’s distance. A substrate having a Cook’s distance higher than
4/n, where nis the number of observations (substrates), is consid-
ered an “interesting” substrate. In addition, the regression line
equation is derived from fitting a linear regression to the substrates
without interesting ones. The whole analysis can be performed
using our in-house R script written with R 3.6.0 [28] and RStudio
[29]. This script as well as the raw data set used for the present
example with neuroblastoma cell lines is available as a Mendeley
repository at the following doi: https://doi.org/10.17632/
b9mprfdvmv.1
3.6 Interpretation
of the Results
The scatterplot of the obtained average slopes in BE-M17 and
SH-SY5Y cells confirmed that SH-SY5Y cells present an increase
in mitochondrial respiration when using succinate, isocitrate, fuma-
rate, and malate as substrates compared to BE-M17 cells (Fig. 5a).
Overall, BE-M17 cells have a 33% decrease in its mitochondrial
activity compared to SH-SY5Y cells. The scatterplot of the mean
initial rate of SK-N-AS cells versus BE-M17 cells shows an increase
in electron flux when SK-N-AS cells use fumarate, succinate, iso-
citrate, and cis-aconitate as substrates compared to BE-M17 cells
(Fig. 5b). Overall, BE-M17 cells have a 46% decrease in its mito-
chondrial activity compared to SK-N-AS cells. The scatterplot of
the mean initial rate of SH-SY5Y versus SK-N-AS cells shows an
increase in electron flux when SK-N-AS cells use succinate, malate,
and cis-aconitate as substrates as compared to SH-SY5Y cells
(Fig. 5c). Overall, SY-SY5Y cells have a 32% decrease in its mito-
chondrial activity compared to SK-N-AS cells. Such an investiga-
tion could hint at possible mutations or differential gene expression
levels that alter the mitochondrial function in neuroblastoma. Our
results are in line with other studies, which detected a succinate
dehydrogenase (SDH) mutation in neuroblastoma tumors [30],
leading to an accumulation of the oncometabolite succinate
Quantification of Cellular Energy Metabolism 137
SH-SY5Y SK-N-AS BE-M17 SH-SY5Y SK-N-AS BE-M17
121110986543217
A
B
C
D
E
F
G
H
b
a
Fig. 4 Example of data processing for MitoPlate™S-1 assay. (a) left panel: Profile of an 18 h kinetic analysis
of MitoPlate™S-1 with three different neuroblastoma cell lines. (a) right panel: The slope between 30 min
and 2 h of MitoPlate™S-1 incubation was analyzed to measure the electron flow rate into the ETC for each
metabolic substrate. Kinetic graphs of color versus time for all wells were generated by SoftMax Pro 7.1
software. (b) The kinetics of the different substrates were normalized to their respective controls and are
shown for three different neuroblastoma cell lines. One representative analysis of three independent experi-
ments is shown
138 Flavia Radogna et al.
a) b)
D,L-Isocitric Acid
Fumaric Acid
L-Malic Acid
Succinic Acid
y=5.5 10
−60.67 x
0.000
0.001
0.002
0.003
0.004
0.005
0.000 0.001 0.002 0.003 0.004 0.005
SH-SY5Y cell line
BE-M17 cell line
Outliers_SHvsBE aa
No Yes
Change in electron flux between BE-M17 and SH-SY5Y cells
cis-Aconitic Acid
D,L-Isocitric Acid
Fumaric Acid
Succinic Acid
y= 1.7 10
−
5
0.54 x
0.000
0.001
0.002
0.003
0.004
0.005
0.000 0.001 0.002 0.003 0.004 0.005
SK-N-AS cell line
BE-M17 cell line
Outliers_SKvsBE
aa
No Yes
Change in electron flux between BE-M17 and SK-N-AS cells
c)
cis-Aconitic Acid
L-Malic Acid
Succinic Acid
y=3.5 10
−
5
0.68 x
0.000
0.001
0.002
0.003
0.004
0.005
0.000 0.001 0.002 0.003 0.004 0.005
SK-N-AS cell line
SH-SY5Y cellline
Outliers_SHvsSK
aa
No Yes
Change in electron flux between SH-SY5Y and SK-N-AS cells
Fig. 5 Example of data determined from MitoPlate™S-1 analysis. (a) The scatterplot of the initial mean rate of
SH-SY5Y cells versus BE-M17 cells shows an increase in electron flux in SY-SY5Y cells when using isocitrate,
succinate, fumarate, and malate as substrates compared to BE-M17 cells. The brown dashed line represents a
line with a slope of 1, while the grey dashed line represents the linear regression model and the 95%
confidence interval. Data points represent the mean of three independent biological replicates SEM. (b) The
scatterplot of the initial rate mean of SK-N-AS cells versus BE-M17 cells shows an increase in electron flux in
SK-N-AS cells when using fumarate, succinate, isocitrate, and cis-aconitate as substrates compared to
BE-M17 cells. The brown dashed line represents a line with a slope one of 1, while the grey dashed line
represents the linear regression model and the 95% confidence interval. Data points represent the mean of
three independent biological replicates SEM. (c) The scatterplot of the initial rate mean of SK-N-AS cells
versus SH-SY5Y cells shows an increase in electron flux in SK-N-AS cells when using cis-aconitate, succinate,
and malate as substrates compared to SH-SY5Y cells. The brown dashed line represents a line with a slope
one of 1, while the grey dashed line represents the linear regression model and the 95% confidence interval.
Data points represent the mean of three independent biological replicates SEM
Quantification of Cellular Energy Metabolism 139
known to promote tumor progression [31]. Moreover, the high
impact of SK-N-AS cells in mitochondrial activity respect to other
neuroblastoma cell lines could be linked to the presence of more
mitochondria [13].
4 Notes
1. BIOLOG suggests saponin from Sigma (SAE0073) presenting
a high sapogenin content (20–35%). Other permeabilizing
agents may be substituted for saponin, such as digitonin or
cholesterol-sequestering toxins, but these must be validated
before use.
2. Tissue analysis requires purified mitochondria; therefore, we
omitted saponin from the assay mix.
3. If using a multichannel pipettor and a reagent reservoir, 4 mL
of the mix are required per plate to fill tips accurately.
4. BIOLOG MAS is 2; thus, it should be diluted in distilled
sterile water to produce a 1stock solution before it is added
to cells.
5. If using a multichannel pipettor and a reagent reservoir, at least
2 mL of cells are required per 1 sample plate to fill tips
accurately.
Acknowledgments
The authors thank M Scheckenburger and S Chateauvieux for
helpful discussions, comments, and editions. FR thanks Te
´le
´vie
Luxembourg. DG thanks the “Recherche Cancer et Sang” founda-
tion. This work was supported by the “Recherche Cancer et Sang”
Foundation, “Recherches Scientifiques Luxembourg,” “Een Haerz
fir kriibskrank Kanner” Association, and Te
´le
´vie; MD received sup-
port from the Research Institute of Pharmaceutical Sciences, Col-
lege of Pharmacy, Seoul National University; NRF grants
019R1A2C1009231 and 2011-0030001 (Tumor Microenviron-
ment Global Core Research Center, GCRC); Brain Korea (BK21)
PLUS and the Creative-Pioneering Researchers Program at SNU
[Funding number: 370C-20160062].
References
1. Corbet C, Feron O (2017) Cancer cell metab-
olism and mitochondria: nutrient plasticity for
TCA cycle fueling. Biochim Biophys Acta Rev
Cancer 1868:7–15
2. Akram M (2014) Citric acid cycle and role of its
intermediates in metabolism. Cell Biochem
Biophys 68:475–478
3. Speijer D (2011) Oxygen radicals shaping evo-
lution: why fatty acid catabolism leads to per-
oxisomes while neurons do without it: FADH
(2)/NADH flux ratios determining mitochon-
drial radical formation were crucial for the
eukaryotic invention of peroxisomes and
140 Flavia Radogna et al.
catabolic tissue differentiation. BioEssays
33:88–94
4. Stanley IA, Ribeiro SM, Gimenez-Cassina A
et al (2014) Changing appetites: the adaptive
advantages of fuel choice. Trends Cell Biol
24:118–127
5. Sharma R, Ramanathan A (2020) The aging
metabolome- biomarkers to hub metabolites.
Proteomics 20(5-6):e1800407
6. Acin-Perez R, Carrascoso I, Baixauli F et al
(2014) ROS-triggered phosphorylation of
complex II by Fgr kinase regulates cellular
adaptation to fuel use. Cell Metab
19:1020–1033
7. Murphy MP (2009) How mitochondria pro-
duce reactive oxygen species. Biochem J
417:1–13
8. Stefanatos R, Sanz A (2018) The role of mito-
chondrial ROS in the aging brain. FEBS Lett
592:743–758
9. van Opbergen CJM, den Braven L, Delmar M
et al (2019) Mitochondrial dysfunction as sub-
strate for arrhythmogenic cardiomyopathy: a
search for new disease mechanisms. Front
Physiol 10:1496
10. Masschelin PM, Cox AR, Chernis N et al
(2019) The impact of oxidative stress on adi-
pose tissue energy balance. Front Physiol
10:1638
11. Rangaraju V, Lewis TL Jr, Hirabayashi Y et al
(2019) Pleiotropic mitochondria: the influence
of mitochondria on neuronal development and
disease. J Neurosci 39:8200–8208
12. Perillo B, Di Donato M, Pezone A et al (2020)
ROS in cancer therapy: the bright side of the
moon. Exp Mol Med 52(2):192–203
13. Radogna F, Cerella C, Gaigneaux A et al
(2016) Cell type-dependent ROS and mito-
phagy response leads to apoptosis or necropto-
sis in neuroblastoma. Oncogene
35:3839–3853
14. Anderson NM, Mucka P, Kern JG et al (2018)
The emerging role and targetability of the TCA
cycle in cancer metabolism. Protein Cell
9:216–237
15. Cerella C, Radogna F, Dicato M et al (2013)
Natural compounds as regulators of the cancer
cell metabolism. Int J Cell Biol 2013:639401
16. Schnekenburger M, Florean C, Dicato M et al
(2016) Epigenetic alterations as a universal fea-
ture of cancer hallmarks and a promising target
for personalized treatments. Curr Top Med
Chem 16:745–776
17. Hanahan D, Weinberg RA (2011) Hallmarks
of cancer: the next generation. Cell
144:646–674
18. Gonfloni S, Iannizzotto V, Maiani E et al
(2014) P53 and Sirt1: routes of metabolism
and genome stability. Biochem Pharmacol
92:149–156
19. Aminzadeh S, Vidali S, Sperl W et al (2015)
Energy metabolism in neuroblastoma and
Wilms tumor. Transl Pediatr 4:20–32
20. Xiao D, Ren P, Su H et al (2015) Myc pro-
motes glutaminolysis in human neuroblastoma
through direct activation of glutaminase
2. Oncotarget 6:40655–40666
21. Adeva-Andany MM, Carneiro-Freire N, Seco-
Filgueira M et al (2019) Mitochondrial beta-
oxidation of saturated fatty acids in humans.
Mitochondrion 46:73–90
22. Cerella C, Gaigneaux A, Dicato M et al (2015)
Antagonistic role of natural compounds in
mTOR-mediated metabolic reprogramming.
Cancer Lett 356:251–262
23. Cerella C, Dicato M, Diederich M (2014)
Modulatory roles of glycolytic enzymes in cell
death. Biochem Pharmacol 92:22–30
24. Parey K, Wirth C, Vonck J et al (2020) Respi-
ratory complex I - structure, mechanism and
evolution. Curr Opin Struct Biol 63:1–9
25. Lapuente-Brun E, Moreno-Loshuertos R,
Acin-Perez R et al (2013) Supercomplex
assembly determines electron flux in the mito-
chondrial electron transport chain. Science
340:1567–1570
26. He S, Liu Z, Oh DY et al (2013) MYCN and
the epigenome. Front Oncol 3:1
27. Harenza JL, Diamond MA, Adams RN et al
(2017) Transcriptomic profiling of 39 com-
monly-used neuroblastoma cell lines. Sci Data
4:170033
28. R Development Core Team (2010) R: a lan-
guage and environment for statistical
computing
29. RStudio Team (2015) RStudio: integrated
development for R
30. Rapizzi E, Ercolino T, Fucci R et al (2014)
Succinate dehydrogenase subunit B mutations
modify human neuroblastoma cell metabolism
and proliferation. Horm Cancer 5:174–184
31. Selak MA, Armour SM, MacKenzie ED et al
(2005) Succinate links TCA cycle dysfunction
to oncogenesis by inhibiting HIF-alpha prolyl
hydroxylase. Cancer Cell 7:77–85
Quantification of Cellular Energy Metabolism 141
Chapter 10
Whole-Cell and Mitochondrial dNTP Pool Quantification
from Cells and Tissues
Juan C. Landoni, Liya Wang, and Anu Suomalainen
Abstract
Deoxynucleoside 50-triphosphates (dNTPs) are the molecular building blocks for DNA synthesis, and their
balanced concentration in the cell is fundamental for health. dNTP imbalance can lead to genomic
instability and other metabolic disturbances, resulting in devastating mitochondrial diseases.
The accurate and efficient measurement of dNTPs from different biological samples and cellular com-
partments is vital to understand the mechanisms behind these diseases and develop and scrutinize their
possible treatments. This chapter describes an update on the most recent development of the traditional
radiolabeled polymerase extension method and its adaptation for the measurement of whole-cell and
mitochondrial dNTP pools from cultured cells and tissue samples. The solid-phase reaction setting enables
an increase in efficiency, accuracy, and measurement scale.
Key words dNTP, Nucleotide pools, mtDNA, Mitochondrial DNA depletion syndrome, Solid-phase
detection
1 Introduction
Deoxynucleoside 50-triphosphates (dNTPs) are the metabolites
that serve as building blocks for the synthesis of DNA in all living
cells. In eukaryotes, dNTP metabolism is a highly regulated net-
work of intermingled pathways, composed of the mostly cytosolic
de novo biosynthesis pathway, dominating in proliferating cells,
and the parallel cytosolic and mitochondrial salvage pathways, typi-
cal for postmitotic cells [1]. The concentration and balance of the
four canonical dNTPs are crucial for healthy cellular metabolism
and genetic stability. dNTP imbalance is known to alter DNA
replication fidelity and repair, cell cycle progression, oncogenesis,
apoptosis, and other key cellular processes [2].
Even though every DNA-containing cell in the body requires
dNTPs for DNA replication and maintenance, genetic defects in
the dNTP biosynthetic enzymes cause severe human diseases with
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_10,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
143
remarkable and unexplained tissue specificity and manifestation
diversity (reviewed by [3–6]). These multisystemic human disor-
ders, known as mitochondrial DNA depletion syndromes, are char-
acterized by the quantitative reduction of mitochondrial DNA copy
number in affected organs, and the molecular mechanisms behind
their pathogenesis remain poorly understood.
The important interplay between dNTP metabolism and mito-
chondrial homeostasis is further underscored by genetic defects
affecting mitochondrial DNA maintenance, which can secondarily
affect dNTP pool balance [7,8] and even consequently disturb
nuclear DNA stability [9].
Recent breakthroughs in the treatment of mitochondrial DNA
depletion syndrome have been reported. Several
nucleotide-modulation possibilities exist for thymidine phosphory-
lase deficiency (reviewed by [10]), and more recently significant
improvements have been reported in fatal thymidine kinase 2 defi-
ciency by deoxynucleoside supplementation [11–13]. Measurement
of dNTPs is crucial for the follow-up and mechanistic characteriza-
tion of these therapeutic interventions.
We present an updated state-of-the-art method for dNTP con-
centration measurement. The method takes advantage of the natu-
rally evolved dNTP specificity of DNA polymerase enzymes, which
proportionally incorporates the measured dNTPs together with
radioactively labeled nucleotides onto designed oligonucleotides,
from which radioactivity can be measured and the dNTP concen-
trations quantified. The method is based on the previously pub-
lished radioactive polymerase assay [14], which is further developed
and improved [15], and on ribonucleotide discrimination [16] and
allows for mitochondrial dNTP pool measurement [17]. This chap-
ter is an expansion on the most recent update of the method [18],
which adapted the measurement to a solid-phase setting (Fig. 1),
allowing for automation, improving efficiency, and circumventing
laborious and hazardous steps.
2 Materials
2.1 General
Equipment
1. Speed vacuum concentrator preferably with cooling trap.
2. Automated cell counter.
3. Centrifuge.
4. Scintillation vials and beta counter (MicroBeta
2
from
PerkinElmer).
5. Microplate washer.
2.2 Whole-Cell dNTP
Isolation from Cells
1. 60% methanol in water at 20 C.
2. 1Trypsin-EDTA.
3. Phosphate-buffered saline (PBS).
144 Juan C. Landoni et al.
2.3 Whole-Cell dNTP
Isolation from Tissues
1. 100% methanol at 20 C.
2. Tissue homogenization buffer: 0.2 mM EGTA, 10 mM Tris–
HCl pH 7.5, 0.5% bovine serum albumin (see Note 1).
3. Polytron tissue homogenizer or other suitable tissue
homogenizer.
2.4 Mitochondrial
dNTP Isolation
1. 60% methanol in water at 20 C.
2. Teflon-glass homogenizer.
3. Mitochondrial isolation buffer: 2 mM EGTA, 5 mM Tris–HCl
pH 7.4, 320 mM sucrose) (see Note 1).
2.5 Binding,
Polymerization,
and Detection
1. HPLC-purified biotinylated [B] oligonucleotides (50-30):
dATP-oligo: [B] AAATAAATAAATAAATAAATGGACAGAG
TATGTGTCTGTG.
dTTP-oligo: [B] TTATTATTATTATTATTAGGACAGAG
TATGTGTCTGTG.
dCTP-oligo: [B] TTTGTTTGTTTGTTTGTTTGGGACA
GAGTATGTGTCTGTG.
dGTP-oligo: [B] TTTCTTTCTTTCTTTCTTTCGGACA
GAGTATGTGTCTGTG (see Note 2).
2. Streptavidin-coated 96-well plate (BioBind Streptavidin Strip
Assembled Solid by Thermo Scientific).
3. Binding solution: 0.1% TWEEN
®
20 (Amresco) in PBS.
4. Universal primer (50-30): CCTGTCTCATACACAGACAC (see
Note 2).
Fig. 1 Graphic depiction of measurement reaction, using sample dCTP measurement as an example.
S¼streptavidin; B ¼biotin, dATP*/A* ¼tritium-labeled adenosine; NNNNNN ¼primer binding sequence.
(I) Affinity capture of dCTP-oligo onto streptavidin-coated plate. (II) Polymerization reaction: primer binding,
followed by the proportional incorporation of C bases from the sample and radiolabeled A bases from the
excess
3
H-dATP*. (III) Alkaline denaturation of DNA strand by sodium hydroxide and release of the newly
synthesized radiolabeled strand for scintillation counting
Whole-Cell and Mitochondrial dNTP Pool Quantification 145
5. 50 mM NaOH solution (see Note 3).
6. Thermostable DNA polymerase and 10optimized buffer (see
Note 4).
7. 0.5 M dithiothreitol (DTT) (see Note 5).
8. dNTP mix stock, 40 mM dNTPs (10 mM each) (see Note 6).
9. [8-
3
H(N)]- Deoxyadenosine 50-triphosphate Tetrasodium Salt
in 1:1 ethanol:water mixture (PerkinElmer).
10. [Methyl-
3
H]- Deoxythymidine 50-Triphosphate Tetrasodium
Salt in 1:1 ethanol:water mixture (PerkinElmer).
11. TENT solution: 40 mM Tris–HCl, 1 mM EDTA, 50 mM
NaCl, 0.1% TWEEN
®
20, pH 8.0–8.8.
12. Ultima Gold™liquid scintillation cocktail for aqueous and
nonaqueous samples (PerkinElmer).
3 Methods
3.1 Whole-Cell dNTP
Isolation from Cells
1. Culture at least 10
6
cells, varying the amounts according to cell
size and type (see Note 7).
2. Wash cells thoroughly with PBS and detach with trypsin-
EDTA.
3. Count cell number (for future normalization) and then centri-
fuge the cells at 250 g, wash with PBS once more. Snap
freeze and store the pellet at 80 C (max. 2–3 of days) or
continue to step 4.
4. Add 1 ml of cold 60% methanol and mix thoroughly.
5. Incubate >1hat80 Cor>2hat20 C.
6. Centrifuge the lysate at 4 C maximum speed (around
20,000 g) for 15 min.
7. Heat the samples containing pellet and supernatant in a hot
plate or boiling water for 3 min, followed by cooling down in
ice water bath and repeat the centrifugation step (see Note 8).
8. Carefully transfer supernatant to a new tube avoiding debris
from the pellet, and desiccate in speed vacuum until no liquid is
visible (see Note 9). Store solid extract at 80 C, optimally for
less than 3 days.
3.2 Whole-Cell dNTP
Isolation from Tissues
1. All samples and solutions should be kept on ice.
2. Collect ~100 mg of tissue (e.g., murine gastrocnemius mus-
cle), recording the exact weight (for normalization purposes).
Snap freeze and store the tissue at 80 C overnight, or con-
tinue with fresh tissue (see Note 10).
146 Juan C. Landoni et al.
3. Add 1 ml of tissue homogenization buffer into the tube and
homogenize using the polytron until no tissue debris is visible.
Cool down the sample in an ice water bath if necessary, during
homogenization.
4. Transfer homogenate into centrifuge tubes and spin at
2000 g10 min at 4 C to remove fibrous tissue and other
cell debris.
5. Split the supernatant into two separate centrifuge tubes.
6. Add cold 100% methanol to obtain a final 60% concentration,
usually 400 μl of supernatant and 600 μl of methanol in
each tube.
7. Continue with incubation as in Subheading 3.1,step 5.
3.3 Mitochondrial
dNTP Isolation
1. For cells culture samples, collect 20–60 million cells, depend-
ing on cell size and proliferative rate (see Note 7).Wash cells
thoroughly with PBS, detach with trypsin-EDTA and centri-
fuge the cells at 250 g, wash again with PBS and re-pellet.
Continue to step 4.
2. If using tissue, collect ~150 mg of soft tissue such as liver.
3. All steps on ice onwards.
4. Resuspend/mix sample with mitochondrial isolation buffer
and transfer to the Teflon-glass homogenizer.
5. Stroke 10 times and transfer homogenate into fresh
centrifuge tube.
6. Pellet nuclei and other particles by 3 min 2000 gcentrifuga-
tion at 4 C. Keep supernatant on ice and re-homogenize the
pellet with 500 μl mitochondrial isolation buffer.
7. Repeat re-homogenization.
8. Combine supernatants and pellet mitochondria at 12,000 g
for 10 min at 4 C.
9. Discard supernatant and resuspend the pellet with isolation
buffer.
10. Save a small aliquot (~40 μl) for protein quantitation, and then
repeat mitochondrial pelleting (12,000 gfor 10 min at
4C).
11. Snap freeze and store the pellet at 80 C overnight, or
continue with dNTP extraction as described in Subheading
3.1,step 4.
3.4 Affinity Capture
of Oligonucleotides
(Fig. 1I)
1. Design the location of each sample in the plate, considering
four reaction wells per replicate of the sample and each reaction
with duplicates. In addition, include wells for a 6-sample stan-
dard curve (24 wells).
Whole-Cell and Mitochondrial dNTP Pool Quantification 147
2. Create and distribute the master mix (0.25 μM of oligonucleo-
tide in binding solution (0.1% TWEEN20 in PBS)). This
equals 2.5 μlof5μM oligonucleotide and 47.5 μl binding
solution per well. You will need four different mater mixes,
one per measured nucleotide.
3. Incubate the plate at 37 C>1.5 h on a shaking heater or
incubator.
4. Discard the solution from the wells and wash thoroughly using
an automated plate washer: 4washes with 200 μl TENT
solution. Tap plate against a drying paper to ensure the wells
are dry.
3.5 Preparation
of Standard Series
and Sample Dilutions
1. Perform every step on ice.
2. Carefully prepare a dilution series from the commercial dNTP
mixture, optimally starting from 1 μM stock obtaining the
quantitative standard dilutions: 80 nM, 40 nM, 20 nM,
10 nM, 5 nM. Include also a water blank.
3. Dissolve the solid extract into 50–100 μl water, depending on
the sample size and expected dNTP amount, vortex, and leave
on ice for 10 min. Then quickly prepare working dilutions of
each sample at different concentrations (e.g., 5and 10), to
ensure replicability and linearity of the result. Store the sample
leftover back to 80 C rapidly (see Note 11). The working
dilutions should be at least 50 μl in volume to suffice for the
four reactions needed.
3.6 Polymerase
Reaction and Detection
(Fig. 1II-III)
1. Calculate the molar concentration of the tritium-labeled
dNTPs (
3
H-dNTP*), and desiccate the required amount to
obtain a 0.75 μM concentration in a final master mix per
reaction type, with reaction volume of 50 μl.
3
H-dATP* is
used for dTTP, dCTP, and dGTP quantitation, while
3
H-dTTP* is used for dATP.
2. To the tube containing the desiccated
3
H-dNTP*, add the rest
of the components in the master mix, as detailed below
(Table 1). Prepare fresh before reaction and keep on ice.
3. Load the reaction mixture to each well. Ensure to match the
bound oligonucleotide to the dNTP* present in the mix
(
3
H-dTTP* to Oligo-dATP and
3
H-dATP* to other oligos).
4. Incubate the plate on a flotation device in a 55 C water bath
for 1 h.
5. Discard contents of the wells and perform washes as in step
3.4.4.
6. Add 60 μl of 50 mM NaOH and incubate for >3 min at room
temperature to release freshly synthesized DNA (Fig. 1III).
148 Juan C. Landoni et al.
7. Transfer the DNA-containing NaOH solution into scintillation
vials, add 3 ml of liquid scintillation cocktail, and measure
radioactivity for 1 min in the beta counter.
3.7 Data Analysis 1. Export the counts per minute results, and analyze each reaction
type (dATP, dTTP, dCTP, and dGTP) separately.
2. Subtract the background (the water blank count) from all raw
data of each reaction.
3. Generate a standard curve by using linear regression method
allowing the curve crossing the origin using data from standard
samples. dNTP concentration on the x-axis and counts per
minute on the y-axis (see Note 12).
4. Calculate the dNTP concentration in each sample by using the
standard curve. Exclude samples outside of the linear range of
the measured standards, and repeat by adapting sample dilu-
tion (see Note 11).
5. Normalize the final dNTP concentration by using dilution
factors and sample volumes, and against cell number, tissue
mass, or mitochondrial protein amount to obtain the final
quantitative result (Eq. 1).
Calculation of dNTP concentration from counts per
minute
dNTP concentration nmol=xðÞ¼
Counts per minute
Standard slope nM1
ðÞ
Dilution coefficient Dissolution volume LðÞ
Normalization factor x¼cell number, mg tissue, or protein concentrationðÞ
ð1Þ
6. Analyze the results from each replica and present the final result
as mean SD.
Table 1
Polymerase reaction mixture composition
Volume (μl) Concentration
10polymerase buffer 5 1
0.5 M DTT 0.5 5 mM
15 μM radiolabeled dNTP (calculated and desiccated) 0.75 μM
5μM primer 2.5 0.25 μM
Thermostable polymerase (calculated from polymerase) 0.025 U/μl
Sample 12
H
2
O29
Total volume 50
Whole-Cell and Mitochondrial dNTP Pool Quantification 149
4 Notes
1. The tissue homogenization buffer and mitochondrial isolation
buffer can be stored in aliquots at 20 C for about a year.
2. The oligonucleotides should be HPLC-purified. When receiv-
ing stocks, they can be diluted into 5 μM concentration and
stored at 20 C in small aliquots to avoid freeze–thawing.
3. 50 mM NaOH solution should be made fresh every 4–6 weeks.
4. The polymerase used in the method setup was DyNAzyme II
DNA Polymerase by Thermo Fisher Scientific and its 10
commercial optimized buffer.
5. 0.5 M DTT should be stored in small aliquots at 20 C and
only used once after thawing due to instability of the com-
pound. A precipitate will be visible in the cold solution but it
will dissolve when fully thawed.
6. The dNTP mix can be diluted up to 1 μM aliquots and stored at
20 C, to simplify future serial dilutions. This is easily done by
a two-step 1:100 serial dilution (10 μl of dNTP mix into 990 μl
of water, and 10 μl of the resulting solution into additional
990 μl of water). Avoid repeated freezing and thawing.
7. dNTP pool concentrations vary dramatically between different
cell types and cell cycle stage. Furthermore, confluency and
proliferative state should also be considered when comparing
cell lines to one another.
8. The boiling step will denature and precipitate leftover proteins
in solution. If the first pellet is too large, the supernatant can be
transferred to a new tube before boiling and second
centrifugation. N.B: the temperature increase might make the
centrifuge tube caps pop open, which could lead to loss of
sample and hazard to the researcher, ensure tube caps are
held strongly and work in a fume hood.
9. The desiccation might take several hours and the samples and
apparatus might warm-up. If the temperature rises significantly
above room temperature, cool the samples in a freezer and
continue the drying once cold.
10. Longer storage of tissue, even deep frozen, shows measurable
signs of dNTP decay. If necessary, store sample as dried dNTP
extract and keep the same storage conditions consistent within
the experiment to avoid batch effects.
11. Due to the instability of dNTPs, the sample handling should be
done on ice and as rapidly as possible, avoiding warm tempera-
tures and freeze–thaw cycles. There is measurable decay of
dNTPs even after overnight 80 C storage of dissolutions,
150 Juan C. Landoni et al.
so if remeasurement is necessary, repeat multiple samples and
compare in a batch-specific manner.
12. The standard linear regression should have an R
2
value very
close to 1, commonly ~0.99.
References
1. Wang L (2016) Mitochondrial purine and
pyrimidine metabolism and beyond. Nucleo-
sides Nucleotides Nucleic Acids 35:578–594
2. Mathews CK (2014) Deoxyribonucleotides as
genetic and metabolic regulators. FASEB J
28:3832–3840
3. Viscomi C, Zeviani M (2017) MtDNA-
maintenance defects: syndromes and genes. J
Inherit Metab Dis 40:587–599
4. Gorman GS, Chinnery PF, DiMauro S et al
(2016) Mitochondrial diseases. Nat Rev Dis
Primers 2:16080
5. Nunnari J, Suomalainen A (2012) Mitochon-
dria: in sickness and in health. Cell
148:1145–1159
6. Suomalainen A, Isohanni P (2010) Mitochon-
drial DNA depletion syndromes – many genes,
common mechanisms. Neuromuscul Disord
20:429–437
7. Nikkanen J, Forsstro
¨m S, Euro L et al (2016)
Mitochondrial DNA replication defects disturb
cellular dNTP pools and remodel one-carbon
metabolism. Cell Metab 23:635–648
8. Dalla Rosa I, Ca
´mara Y, Durigon R et al (2016)
MPV17 loss causes deoxynucleotide insuffi-
ciency and slow DNA replication in mitochon-
dria. PLoS Genet 12
9. H€
am€
al€
ainen RH, Landoni JC, Ahlqvist KJ et al
(2019) Defects in mtDNA replication chal-
lenge nuclear genome stability through nucle-
otide depletion and provide a unifying
mechanism for mouse progerias. Nat Metab
1:958–965
10. Yadak R, Sillevis Smitt P, van Gisbergen MW
et al (2017) Mitochondrial neurogastrointest-
inal encephalomyopathy caused by thymidine
phosphorylase enzyme deficiency: from patho-
genesis to emerging therapeutic options. Front
Cell Neurosci 11:31
11. Garone C, Garcia-Diaz B, Emmanuele V et al
(2014) Deoxypyrimidine monophosphate
bypass therapy for thymidine kinase 2 defi-
ciency. EMBO Mol Med 6:1016–1027
12. Lopez-Gomez C, Levy RJ, Sanchez-Quintero
MJ et al (2017) Deoxycytidine and deoxythy-
midine treatment for thymidine kinase 2 defi-
ciency. Ann Neurol 81:641–652
13. Domı
´nguez-Gonza
´lez C, Madruga-Garrido M,
Mavillard F et al (2019) Deoxynucleoside ther-
apy for thymidine kinase 2–deficient myopathy.
Ann Neurol 86:293–303
14. Solter AW, Handschumacher RE (1969) A
rapid quantitative determination of deoxyribo-
nucleoside triphosphates based on the enzy-
matic synthesis of DNA. Biochim Biophys
Acta 174:585–590
15. Sherman PA, Fyfe JA (1989) Enzymatic assay
for deoxyribonucleoside triphosphates using
synthetic oligonucleotides as template primers.
Anal Biochem 180:222–226
16. Ferraro P, Franzolin E, Pontarin G et al (2010)
Quantitation of cellular deoxynucleoside tri-
phosphates. Nucleic Acids Res 38
17. Martı
´R, Dorado B, Hirano M (2012) Mea-
surement of mitochondrial dNTP pools.
http://www.scopus.com/inward/record.url?
eid¼2-s2.0-84856364625&partnerID¼40&
md5¼f3c1af28bae6e1c780027884b84be24f
18. Landoni JC, Wang L, Suomalainen A (2018)
Quantitative solid-phase assay to measure
deoxynucleoside triphosphate pools. Biol
Methods Protoc 3:bpy011
Whole-Cell and Mitochondrial dNTP Pool Quantification 151
Chapter 11
Single-Particle Tracking Method in Fluorescence
Microscopy to Monitor Bioenergetic Responses
of Individual Mitochondria
Camille Colin, Emmanuel Suraniti, Emma Abell, Audrey Se
´mont,
Neso Sojic, Philippe Diolez, and Ste
´phane Arbault
Abstract
The spectroscopic methods commonly used to study mitochondria bioenergetics do not show the diversity
of responses within a population of mitochondria (isolated or in a cell), and/or cannot measure individual
dynamics. New methodological developments are necessary in order to improve quantitative and kinetic
resolutions and eventually gain further insights on individual mitochondrial responses, such as studying
activities of the mitochondrial permeability transition pore (mPTP). The work reported herein is devoted to
study responses of single mitochondria within a large population after isolation from cardiomyocytes.
Mitochondria were preloaded with a commonly used membrane potential sensitive dye (TMRM), they
are then deposited on a plasma-treated glass coverslip and subsequently energized or inhibited by additions
of usual bioenergetics effectors. Responses were analyzed by fluorescence microscopy over few thousands of
mitochondria simultaneously with a single organelle resolution. We report an automatic method to analyze
each image of time-lapse stacks based on the TrackMate-ImageJ plug-in and specially made Python scripts.
Images are processed to eliminate defects of illumination inhomogeneity, improving by at least two orders
of magnitude the signal/noise ratio. This method enables us to follow the track of each mitochondrion
within the observed field and monitor its fluorescence changes, with a time resolution of 400 ms, uninter-
rupted over the course of the experiment. Such methodological improvement is a prerequisite to further
study the role of mPTP in single mitochondria during calcium transient loading.
Key words Mitochondria, Fluorescence microscopy, Membrane potential, Bioenergetics, Single
organelle, Single particle tracking, Fiji software, TrackMate
1 Introduction
The monitoring by microscopy of mitochondria, when isolated or
when forming a network in cells, constitutes a major approach to
resolve their organization, functioning, and activity under physio-
logical [1] or pathological [2,3] situations. Owing to the recent
developments of microscopy techniques, these are increasingly used
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_11,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
153
since they allow to follow-up responses in living cells or monitor
resolved kinetics on minute sample quantities under a noninvasive
manner. NADH and FAD autofluorescence can be used to monitor
endogenously the mitochondrial bioenergetics when modulations
of the respiratory chain are used. A large variety of fluorescent dyes
can also be found to assess mitochondrial functions (membrane
potential, ROS, pH...), their number continuously increasing in
literature [4]. The most representative family is the one dedicated
to monitor ΔΨ with high sensitivity and kinetic resolution, which
includes dyes quoted as JC1, TMRE, or TMRM. [5–8]
Microscopy approaches are used first for imaging and to a lower
extent to resolve metabolic responses such as depolarization waves
or transients by mitochondria under bioenergetic variations. There
are now multiple reports at the level of single mitochondrion
resolution from experiments in cells, notably in cardiomyocytes
thanks to their organization, [9,10] or for isolated mitochondria
deposited on various supports (glass coverslip, modified or not,
PDMS, etc.) [11–13]. However, only few remarkable events at
single mitochondria are usually reported and shown in communi-
cations, while the response of populations, networks with an indi-
vidual resolution would help in deciphering mechanisms and
discriminating the existence of subpopulations (heteroplasmy, met-
abolic variations) [14]. To do so, automated image processing and
data analysis protocols still need to be developed to promote large-
scale studies and statistics from individual mitochondria responses
in microscopy; this is the aim of the present methodological devel-
opment. We studied responses to bioenergetics effectors of mito-
chondria isolated from rat cardiomyocytes. After depositing on a
glass coverslip, their membrane potential evolutions were moni-
tored simultaneously from thousands of entities with a single mito-
chondrion resolution. This was made possible with microscopy
image corrections and improvements (Fiji software functions)
along with individual follow-up of objects over time and distance
(TrackMate plug-in) and additional personalized scripts to generate
matrices of data. All developed protocols are available free of charge
from the authors upon request.
2 Materials
2.1 Solutions 1. Extraction buffer: Saccharose 300 mM, Tris–HCl 10 mM,
EGTA 50 mM, pH adjusted to7.2 with NaOH 0.5 M.
2. Digestion buffer: Saccharose 300 mM, Tris–HCl 10 mM,
EGTA 50 mM, protease 0.1 mg.ml
1
(from Streptomyces gri-
seus, Type XIV, ref: P5147), pH 7.2.
154 Camille Colin et al.
3. Homogenization buffer: Saccharose 300 mM, Tris–HCl
10 mM, EGTA 50 mM, BSA 0.2% (W/v), pH adjusted to
7.2 with NaOH 0.5 M.
4. Respiration buffer: Saccharose 300 mM, KCl 100 mM, EGTA
1 mM, MgCl2 20 mM, KH
2
PO
4
10 mM, BSA 0.2% (W/v),
pH adjusted to 7.2 with NaOH 0.5 M.
5. Glutamate, malate, succinate (ref: G1626, M9138, S2378,
respectively) stock solutions are prepared at 500 mM in respi-
ration buffer.
6. ADP (Adenosine diphosphate, ref A2754) stock solution is
prepared at 100 mM in respiration buffer.
7. Rotenone (ref R8875) stock solution is prepared at 1 mM in
DMSO (see Note 1).
8. Oligomycin (ref 75351) stock solution is prepared at 1 mM in
methanol (see Note 1).
9. C-ATR (carboxy-atractyloside potassium salt, ref C4992) stock
solution is prepared at 5 mM in respiration buffer (see Note 1).
10. CCCP (Carbonyl cyanide 3-chlorophenylhydrazone, ref
C2759) stock solution is prepared at 1 mM in acetone (see
Note 1).
2.2 Fluorescence
Imaging
1. Experiments were performed with an inverted epifluorescence
microscope from Leica©(DMI 6000 B model) equipped with
a40objective (dry; NA: 0.75; Leica HC PL APO), and a
camera from Hamamatsu©(ORCA-Flash4.0).
2. Images were collected via MetaMorph (Molecular Devices©)
and analyzed with Fiji and ImageJ (NIH free supply) software.
3 Methods
3.1 Rat Heart
Extraction
1. Mitochondria are extracted from the heart of Wistar male rats
(from Janvier Labs, France) according to the following proce-
dure (see Note 2).
2. The rat is anesthetized by inhalation of isoflurane (4%) in an
induction chamber for 2 min.
3. Once anesthesia observed, the rat is weighted and injected
subcutaneously with 0.1 mL of heparin at 5000 U.mL
1
.
4. The rat is placed back in the induction chamber (4% isoflurane)
for further 3 min and euthanized by cervical dislocation.
5. The heart is rapidly extracted from the thorax and placed in
cold extraction buffer, to stop contractions (see Note 2).
6. The heart is rinsed with extraction buffer, weighed, and
unwanted tissue is removed.
Single-Particle Tracking 155
3.2 Extraction
of Cardiac
Mitochondria
1. Ventricular tissue is excised and thoroughly minced with sharp
scissors.
2. Minced tissue is transferred to a beaker containing 30 mL of
digestion buffer and placed on a magnetic stirrer for 7 min.
3. The solution is transferred to a Teflon-glass homogenizer and
remaining digestion buffer is used to remove any tissue left in
beaker. The partially digested tissue is homogenized for 3 min,
to break up any remaining tissue.
4. The homogenate is centrifuged at 7500 gfor 7 min.
5. The supernatant is discarded, and the pellet resuspended in
30 mL of homogenization buffer, homogenized for 2 min
and transferred to a new centrifuge tube.
6. The solution is centrifuged at 680 gfor 10 min. The super-
natant is then filtered with a nylon filter and transferred into a
new centrifuge tube.
7. The solution is centrifuged at 7000 gfor 10 min. The
supernatant is discarded, and the pellet resuspended in 40 μL
homogenization buffer.
8. Mitochondrial protein quantification is performed with a clas-
sic Bradford assay, the resulting optical density is measured at
595 nm with a spectrophotometer (see Note 3).
3.3 Coverslip
Preparation
1. Coverslips (12 mm diameter) are purchased from Fischer Sci-
entific (Carolina Science & Math manufacturer; ref.
NC9537307).
2. Coverslips are treated just before use with a low pressure-
oxygen plasma generator from Harrick Plasma©(ref. plasma
cleaner) at 300 mTorr, 100% O
2
, for 10 min (see Note 4).
3.4 Experimental
Procedure
for Mitochondria
Imaging
1. Before the experiment, a diluted solution of mitochondria is
prepared at 0.1 mg.mL
1
in respiration buffer (see Note 5),
supplemented with 10 nM of TMRM (Tetramethylrhodamine
methyl ester perchlorate, purchased from Sigma, ref T5428).
2. After 10 min of incubation, 600 μL of the solution is deposited
on the coverslip previously mounted on the microscope stage
(see Note 6).
3. In bright field mode, the focus is on the top surface of the
coverslip, and a 20-min interval is allowed for mitochondria to
sediment (see Note 7).
3.5 Imaging
of Mitochondria
1. The typical sequence for mitochondria imaging is the follow-
ing: images are captured every 10 s for 20 min; exposure times
are 40 ms to detect either TMRM fluorescence in mitochondria
or for additional observations in bright field. Fluorescence
images are obtained with a fibered light source (Leica, ref
156 Camille Colin et al.
EL6000) set at its minimum power, while white light images
were obtained with a fibered LED source (CoolLED, ref
PE100) connected to the microscope.
2. Observations are achieved using a 40objective (dry;
N.A. 0.75), allowing large field views (332.8 322.8 μm
area) and single mitochondria measurements.
3. N2.1-type filter (Leica) is used for TMRM detection (excita-
tion: 515–560 nm, emission, long-pass >590 nm), and A-type
filter (Leica) is used for endogenous NADH detection (excita-
tion: 340–380 nm, emission: long-pass >425 nm).
4. The following sequence of solutions is used to induce the
different bioenergetic stationary states in mitochondria:
(1) addition of respiratory substrates; glutamate plus malate
at 5 mM each, or succinate at 5 mM plus rotenone at 2 μM, or a
combination of glutamate, malate, and succinate, all at 5 mM;
(2) addition of ADP at 1 mM; (3) addition of oligomycin at
5μM or C-ATR at 5 μM; and (4) addition of CCCP at 0.5 μM.
5. Ten images are taken for each step of the sequence (100 s),
note that 10 images are taken before adding the first substrate.
3.6 Treatment
of Images with “Fiji”
1. Before any data analysis, all images are stacked (image/stack/
images to stack) in a file (Stack 1; Fig. 1a) and duplicated (Stack
2; image/duplicate). A profile of the intensity variation versus a
linear ROI (region of interest) can be drawn to check for the
signal/noise ratio of the object detection (Fig. 1b).
2. On Stack 2, a Gaussian-type filter is applied with a factor of
60 to correct a shading effect (process/filters/Gaussian Blur)
due to the illumination inhomogeneity of the lamp on the
epifluorescence microscope.
3. The obtained Stack 2 (Fig. 1c) is subtracted from the original
Stack 1 (image/process/image calculator/subtract) to gener-
ate a corrected Stack 3 (Fig. 1d).
4. A median filter with a factor of 2 is applied to Stack 3 to
eliminate noise resulting from the subtraction (image/pro-
cess/filters/median) and enhance the signal/noise ratio. This
produces a final Stack 4 (Fig. 1e) used for further analyses. The
result of the procedure on the image of a single mitochondrion
is shown in Fig. 1f to compare with the original one in Fig. 1b.
3.7 Mitochondria
Follow-Up
with “TrackMate”
1. On Stack 4, a Fiji software plug-in called “TrackMate” [15]is
applied (plug-in/Tracking/TrackMate). This plug-in enables
to track the path of each individual mitochondrion over dis-
tance and time, according to two main steps.
2. The first step is to detect on each image, all possible objects that
can be classified as mitochondria. To do so, the “DoG
Single-Particle Tracking 157
Fig. 1 Protocol for the treatment of microscopy images with “Fiji” software. (a) Raw image of mitochondria
after 20 min. Sedimentation on a glass coverslip. Mitochondria were incubated with 10 nM TMRM membrane
potential dye and energized by the addition of glutamate/malate/succinate at 5 mM. 10 images (every 10 s,
40 ms exposure time) are used to create the Stack 1. (b) Example of a fluorescence intensity profile for a
single mitochondrion of the image at this stage of the treatment. (c) A Gaussian-type filter is applied (factor 60)
to correct the shading effect due to the illumination inhomogeneity on the whole field, resulting images lead to
Stack 2. (d) Stack 2 is subtracted from Stack 1 to generate corrected images in Stack 3. (e) Noise in the
images of Stack 3 is corrected via a median filter (factor 2) and generates the final Stack 4. (f) Fluorescence
intensity profile for a single mitochondrion, the same as in (b), after the full image processing
158 Camille Colin et al.
detector” method is used with an “Estimated blob diameter”
of 10 pixels, equivalent to 1.625 μm, and an intensity threshold
between 1 and 5 AU. This method uses a difference of Gauss-
ian fits to detect particles and is optimal for spots of small size
(see Note 8).
3. The second step is aimed at identifying the same object over
time; for this purpose, the “simple LAP Tracker” method is
used (see Note 9). According to their identification, provided as
a number assigned by TrackMate (see Note 10), a follow-up of
the displacement of individual mitochondria is possible
(Fig. 2a), though some boundaries need to be given: “Linking
max distance” of 15 pixels, or 2.4375 μm; a “Gap-closing max
distance” of 15 pixels, or 2.4375 μm; and a “Gap-closing max
Fig. 2 Follow-up with the TrackMate plug-in of “Fiji” software, of mitochondria mobility over time on the
surface of a glass coverslip. (a) Display of all trajectories (named tracks) from individual mitochondria during a
sequence of 20 images (200 s); each object is identified and shown here with a colored trajectory. (b) Shortcut
of the whole image to display two mitochondria with very different behaviors along four images taken
sequentially. (c) Graph of the fluorescence variations due to displacements for the two mitochondria shown
in b. Data treatments were applied on the images of the sequence in Fig. 1
Single-Particle Tracking 159
frame gap” of 50 frames. Two different types of displacements
of mitochondria on the surface along an experiment are com-
pared in Fig. 2b.
4. Finally, the function “Analysis” is used to generate “Spots in
track statistics” datasheets allowing further data analyses of
displacements, intensity variations, etc.
3.8 Python Scripts
for Quantitative
Analyses
1. Specialized informatics scripts were written with “Python” 4.0,
“Spyder” environment, and “Pandas, NumPy, and SciPy”
libraries, to automatically process the analysis of data. These
scripts are available free of charge upon request from the
authors.
2. At the beginning of each script, two filters are used: (1) A filter
of movement to discard mobile mitochondria (>3.25 μm dis-
placement), which can distort the analysis and overall the sta-
tistics (example in Fig. 2c); (2) A time filter, to eliminate
mitochondria that are not present enough (<10 images,
100 s) along the whole sequence.
3. The first script was designed to calculate for every image the
mean intensity of all individual mitochondria (TMRM fluores-
cence) in order to apprehend the global variations of the sam-
ple (Fig. 3a).
4. To do so, “Spots in tracks statistics” is read with the function
“read_csv” and data are grouped for each image with the
function “.groupby (‘FRAME’).” After, the mean intensity is
calculated (“numpy.mean”) for each image. Other parameters
can be implemented as the number of mitochondria detected
(“len”) and the standard deviation of means (“numpy.std”).
5. A second script was designed to create a graphical representa-
tion of the fluorescence intensity variation over time for each
identified mitochondrion (“TRACK_ID”) (Fig. 3b).
6. To do so, “Spots in tracks statistics” are read and the data for all
images are grouped for each unique identification number of all
detected objects with the function “groupby (‘TRACK_ID’)”.
Then, the fluorescence intensity evolution over time for each
mitochondrion is displayed graphically and saved as a “.png”
file with the identification number as title.
4 Notes
1. This is a hazardous material; so, protective clothing must be
worn during preparation and manipulation.
2. It is necessary to maintain cold on ice all solutions and instru-
ments before starting and during the procedure of rat heart
extraction.
160 Camille Colin et al.
Fig. 3 Use of the image treatment protocol described in Figs. 1and 2and of the
specially developed scripts to monitor individual mitochondria responses within
a population. (a) Calculation of the mean fluorescence intensity (mean SEM) of
all the mitochondria in each image to display the variations along a sequence of
bioenergetics effectors. (b) 100 individual mitochondrial responses overlapped
which evidence the discrepancy of responses in a sample. The variations of the
mitochondrial membrane potential were monitored by fluorescence microscopy
with TMRM (10 nM) for 20 min during the following sequence: 20 min sedimen-
tation and immobilization on the glass surface; addition at 100 s of respiratory
substrates (glutamate, malate, succinate at 5 mM each); addition at 200 s of
ADP at 1 mM; addition at about 300 s of ATR at 5 μM; final addition at 400 s of
CCCP at 0.5 μM
Single-Particle Tracking 161
3. For the Bradford assay, there is a 10-min incubation period
before reading the optical density at 595 nm.
4. The treatment of the glass surface by an oxygen plasma (100%
O
2
) clearly improves its hydrophilicity. It is observed that the
plasma encourages mitochondria to be immobile on the surface
for longer times (at least 50 min), which is a mandatory condi-
tion for the quantitative monitoring of their fluorescence dur-
ing different steps, particularly when solutions (activators or
inhibitors) are injected into the surrounding area.
5. Please note that the concentration of mitochondria is an impor-
tant parameter: we observe for concentrations higher than
0.2 mg.mL
1
that the number of mitochondria on the cover-
slip is too high and hinders single-particle tracking. On the
contrary, for a concentration below 0.05 mg.mL
1
, mitochon-
dria are not viable, a sudden loss of TMRM fluorescence is
often observed.
6. The coverslip has to be immobile throughout an experiment
because of the 3D position-precision requested for particle
tracking; vacuum grease can be used to help sticking to the
support on the microscope platform.
7. 20x objective can also be used to help finding the right focus,
but is not suitable to monitor the fluorescence changes. On the
other hand, 63or higher magnification objectives can be used
to improve resolution but the number of mitochondria simul-
taneously observed may be too low to draw significant
conclusions.
8. “TrackMate” can also give an estimation of the object diameter
inside the region of interest. To be accurate, it is important that
the “Estimated blob diameter” parameter value is superior to
the size of mitochondria. Two distinct sizes of mitochondria
were detected in our sample, around 0.8 μm and 1.4 μm, which
may correspond to the two functional populations described
for cardiomyocytes [16].
9. The “simple LAP Tracker” method used is designed for parti-
cles with no fusion or fission behavior, such as in our experi-
mental case. “TrackMate” offers other methods to track
particles depending on the behavior; manual tracking is also
possible.
10. “TrackMate” plug-in assigns a unique identification number to
every mitochondrion, which facilitates the manipulation and
analyses of results for large sets of data. This number is also
used by our specialized scripts.
162 Camille Colin et al.
Acknowledgments
The project was financially supported by the ANR (“Agence Natio-
nale pour la Recherche,” project MITOCARD nANR-17-CE11-
0041), the University of Bordeaux, the CNRS (“Centre National
de la Recherche Scientifique“) and the INSERM (“Institut
National de la Sante
´Et de la Recherche Me
´dicale”). CC acknowl-
edges the University of Bordeaux (Interdisciplinary-
pluridisciplinary PhD funding program 2016) for his PhD fellow-
ship. AS acknowledges IHU-LIRYC (Pessac, France) for her PhD
fellowship. This study received financial support from the French
Government as part of the “Investments of the Future” program
managed by the National Research Agency (ANR), Grant reference
ANR-10- IAHU-04.
References
1. De Giorgi F, Lartigue L, Ichas F (2000) Elec-
trical coupling and plasticity of the mitochon-
drial network. Cell Calcium 28:365–370
2. Creed S, McKenzie M (2019) Measurement of
mitochondrial membrane potential with the
fluorescent dye tetramethylrhodamine methyl
Ester (TMRM): methods and protocols. Meth-
ods Mol Biol (Clifton, N.J.) 1928:69–76
3. Quintanilla RA, Matthews-Roberson TA,
Dolan PJ, Johnson GVW (2009) Caspase-
cleaved tau expression induces mitochondrial
dysfunction in immortalized cortical neurons:
implications for the pathogenesis of Alzheimer
disease. J Biol Chem 284:18754–18766
4. Jiang N, Fan J, Xu F, Peng X, Mu H, Wang J,
Xiong X (2015) Ratiometric fluorescence
imaging of cellular polarity: decrease in mito-
chondrial polarity in cancer cells. Angew Chem
Int Ed 54:2510–2514
5. Perry SW, Norman JP, Barbieri J, Brown EB,
Gelbard HA (2011) Mitochondrial membrane
potential probes and the proton gradient: a prac-
tical usage guide. BioTechniques 50:98–115
6. Scaduto RC, Grotyohann LW (1999) Mea-
surement of mitochondrial membrane poten-
tial using fluorescent Rhodamine derivatives.
Biophys J 76:469–477
7. Figueira TR, Melo DR, Vercesi AE, Castilho
RF (2012) Safranine as a fluorescent probe for
the evaluation of mitochondrial membrane
potential in isolated organelles and permeabi-
lized cells. Methods Mol Biol 810:103–117
8. Rottenberg H, Wu S (1998) Quantitative assay
by £ow cytometry of the mitochondrial mem-
brane potential in intact cells. Biochim Biophys
Acta 1404:393–404
9. Lu X, Kwong JQ, Molkentin JD, Bers DM
(2016) Individual cardiac mitochondria
undergo rare transient permeability transition
pore openings. Circ Res 118:834–841
10. Wei-LaPierre L, Dirksen RT (2019) Isolating a
reverse-mode ATP synthase–dependent mech-
anism of mitoflash activation. J Gen Physiol
151:708–713
11. Suraniti E, Vajrala VS, Goudeau B, Bottari SP,
Rigoulet M, Devin A, Sojic N, Arbault S
(2013) Monitoring metabolic responses of sin-
gle mitochondria within poly(dimethylsilox-
ane) wells: study of their endogenous reduced
nicotinamide adenine dinucleotide evolution.
Anal Chem 85:5146–5152
12. Vajrala VS, Suraniti E, Goudeau B, Sojic N,
Arbault S (2015) Optical microwell arrays for
large-scale studies of single mitochondria meta-
bolic responses. Methods Mol Biol 1264:47–58
13. Vajrala VS, Suraniti E, Rigoulet M, Devin A,
Sojic N, Arbault S (2016) PDMS microwells
for multi-parametric monitoring of single
mitochondria on a large scale: a study of their
individual membrane potential and endoge-
nous NADH. Integr Biol 8:836–843
14. Youle RJ, van der Bliek AM (2012) Mitochon-
drial fission, fusion, and stress. Science
337:1062–1065
15. Tinevez J-Y, Perry N, Schindelin J, Hoopes
GM, Reynolds GD, Laplantine E, Bednarek
SY, Shorte SL, Eliceiri KW (2017) TrackMate:
an open and extensible platform for single-
particle tracking. Methods 115:80–90
16. Hollander JM, Thapa D, Shepherd DL (2014)
Physiological and structural differences in spa-
tially distinct subpopulations of cardiac mito-
chondria: influence of cardiac pathologies. Am
J Physiol Heart Circ Physiol 307:H1–H14
Single-Particle Tracking 163
Chapter 12
Investigation of Mitochondrial ADP-Ribosylation Via
Immunofluorescence
Ann-Katrin Hopp and Michael O. Hottiger
Abstract
ADP-ribosylation is a posttranslational protein modification, involved in various cellular processes, ranging
from DNA-damage repair to apoptosis. While its function has been studied amply with respect to genotoxic
stress-associated nuclear ADP-ribosylation, the functional relevance of mitochondrial ADP-ribosylation
remains so far poorly studied. This is mainly attributed to the absence of powerful techniques able to detect
the modification. However, the usage of recently developed anti-ADP-ribose–specific antibodies allows
now to investigate mitochondrial ADP-ribosylation under physiological and pathophysiological conditions.
In the below method, we describe in detail how to efficiently detect and quantify mitochondrial
ADP-ribosylation via immunofluorescence.
Key words NAD
+
, Mitochondria, ADP-ribosylation, Posttranslational modifications, MacroD1,
Immunofluorescence, Antibodies
1 Introduction
Mitochondria represent not only the cell’s main metabolic hub but
are also key signaling organelles, crucial for controlling cell func-
tions and fate [1,2]. Because of their vital role, cells have evolved
various mechanisms to control mitochondrial biogenesis, bioener-
getics, and dynamics [3]. Posttranslational modifications (PTMs)
of proteins were underappreciated for decades, but have recently
emerged as promising new regulators of mitochondrial function
[4–6]. In fact, due to their central role in cell metabolism, mito-
chondria harbor high concentrations of various small metabolites,
such as ATP, acetyl-CoA, and NAD
+
, that are important cofactors
for the establishment or removal of numerous PTMs, including
phosphorylation or acetylation. While the existence of PTMs such
as phosphorylation, acetylation, or O-GlcNAcylation is to date well
accepted in mitochondria, mitochondrial ADP-ribosylation was
only recently uncovered [7]. ADP-ribosylation is an evolutionary
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_12,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
165
highly conserved PTM that consists in the transfer of ADP-ribose
moieties from NAD
+
onto specific amino acid residues of target
proteins (i.e., mono-ADP-ribosylation or MARylation), as well on
ADP-ribose itself (i.e., poly-ADP-ribosylation or PARylation)
[8]. The modification is mainly catalyzed by ADP-ribosyl-trans-
ferases (ARTs) and reversed by ADP-ribosylhydrolases (ARHs)
[9,10]. In addition to ARTs, some Sirtuins (SIRTs) were reported
to possess ART activity, as well [11,12]. Although mitochondrial
ADP-ribosylation and the existence of mitochondrially localized
ARTs and ARHs have already been postulated more than 30 years
ago [13], the detection of ADP-ribosylated mitochondrial proteins
remained challenging until recent advances in the field of proteo-
mics that allowed to identify various mitochondrial
ADP-ribosylated proteins [14,15]. While in mitochondria-rich
tissues, such as skeletal muscle or heart, roughly 20% of all identi-
fied ADP-ribosylated proteins localize to mitochondria, the frac-
tion of ADP-ribosylated mitochondrial proteins identified from
other tissues or cultured cells is rather low, suggesting that addi-
tional techniques might be required to further dissect the full
mitochondrial ADP-ribosylome. Moreover, also the ART or SIRT
responsible for catalyzing ADP-ribosylation in mitochondria has
not yet unambiguously been identified (reviewed in [16]). The
only ADP-ribose-metabolizing enzyme that has been found to
localize to mitochondria and to demodify ADP-ribosylated pro-
teins by several methods is the macro domain-containing mono-
ADP-ribosylhydrolase MacroD1 [17,18].
Despite the abovementioned technical advances in proteomics,
the detection of mitochondrial ADP-ribosylation, especially in cell
culture, remains challenging. To increase our current knowledge
about the regulation and function of mitochondrial ADP-ribosyla-
tion, it is key to improve and develop techniques, allowing to
identify and thus study ADP-ribosylation in mitochondria. These
tools would not only provide a better understanding of mitochon-
drial ADP-ribosylation under physiological and pathophysiological
conditions but also facilitate the identification of proteins and
cellular processes regulating mitochondrial ADP-ribosylation.
Because of the tight interplay between NAD
+
homeostasis and its
availability with ADP-ribosylation, a better understanding of mito-
chondrial ADP-ribosylation will also increase our current under-
standing of how metabolic circuits are regulated at the
posttranslational level.
The interaction between antibodies and their specific targets
are to date among the most specific and strongest biological inter-
actions. The generation of ADP-ribosylation–specific antibodies
has therefore the potential to greatly improve the detection of
ADP-ribosylation in general, and also mitochondrial ADP-ribosy-
lation. Unfortunately, the synthesis of ADP-ribosylated proteins or
peptides, which is required for the successful generation of
166 Ann-Katrin Hopp and Michael O. Hottiger
antibodies, is chemically demanding [19]. While amine-linked
ADP-ribose modified proteins were used to generate polyclonal
antibodies that recognize MARylated proteins, we have developed
an anti-ADP-ribose antibody that recognizes MARylated proteins
as well as PARylated ARTD1 by Western and dot blot [20]. Using
this anti-ADP-ribose antibody in immunofluorescence (IF), we
observed a strong extranuclear signal in untreated U2OS cells
that was not present with anti-PAR-Abs [7]. Here we describe a
detailed protocol how to identify and quantify mitochondrial ADP-
ribosylation via immunofluorescence at the single-cell level.
2 Materials
1. Rotenone stock solution (10 mM): Dissolve 39.44 mg Rote-
none in 10 mL deionized water (dH
2
O). Prepare aliquots and
store at 20 C.
2. PBS (1): Dilute 10PBS (1.37 M NaCl, 27 mM KCl,
100 mM Na
2
HPO
4
, and 18 mM KH
2
PO
4
in dH
2
O) 1:10 in
dH
2
O.
3. Fixation solution: 4% formaldehyde (FA) in PBS. To prepare a
200 mL stock solution, dilute 22.2 mL of FA (36%) in
177.8 mL PBS. Store at 4 C.
4. Permeabilization solution: 0.2% Triton X-100 in PBS. To pre-
pare a 100 mL solution, dilute 200μL Triton X-100 in PBS (see
Note 1). Store at 4 C(see Note 2).
5. Blocking solution: 0.1% Triton X-100 and 2% bovine serum
albumin (BSA; Sigma Aldrich, A9418) in PBS. To prepare a
100 mL stock, weigh 2 g BSA, dissolve it in 100 mL PBS, and
add 100μL of Triton X-100 (see Note 1). Sterile filter the
solution and store it at 4 C. The buffer can be kept for several
weeks (see Note 2).
6. Antibodies:
(a) Primary antibody: rabbit anti-ADPR [20].
(b) Secondary antibody: goat anti-rabbit.
7. DAPI solution: Dilute DAPI 1:10,000 in PBS.
8. Mounting medium: e.g., Mowiol.
9. Glass coverslips, autoclaved.
10. Microscopy glass slides.
11. Whatman paper.
12. Parafilm.
13. Plastic box as wet chamber.
Immunofluorescence Analysis of Mitochondrial ADP-Ribosylation 167
3 Methods
3.1 Cell Seeding 1. Seed cells (e.g., U2OS, NIH3T3, or HEK) on top of auto-
claved glass coverslips that have been placed inside the desired/
required cell culture dish (e.g., fourteen 12 mm coverslips into
a 6-cm dish). If different conditions/treatments are required,
cells can be seeded in a 24-well plate containing one glass
coverslip per well.
2. After performing the experiment (siRNA knockdown, plasmid
transfection, inhibitor/compound treatment), remove the
medium and wash cells once with PBS.
3.2 Treatment
of Cells to Increase
Mitochondrial
ADP-Ribosylation
1. To increase mitochondrial ADP-ribosylyation a 30-min treat-
ment with Rotenone (3 μM) a respiratory chain inhibitor can
be used (i.e. as positive control).
3.3 Fixation
and Permeabilization
1. Remove the PBS solution from the coverslips, add fixation
solution, and incubate cells for 15 min at RT. For a 24-well
plate approximately 250–300μL is sufficient to fully cover the
cells.
2. Replace the fixation solution with permeabilization solution
and incubate cells for another 10 min at RT. Again,
250–300μL is sufficient to cover cells seeded in a 24-well plate.
3. After this step, the permeabilization solution can be replaced
with PBS and cells can be stored for several weeks at 4 C until
further usage.
3.4 Blocking
and Immunostaining
1. Remove the PBS solution from the coverslips, add blocking
solution on top of the cells, and incubate for at least 30 min
at RT.
2. In the meantime, prepare a wet chamber for the antibody
incubation (Fig. 1):
(a) Line the bottom of a plastic box evenly with Whatman
paper.
(b) Humidify the paper with water.
(c) Add a layer of parafilm on top of the Whatman paper. The
parafilm should be slightly smaller as compared to the
Whatman paper.
3. Dilute the anti-ADPR antibody 1:500 in blocking solution and
incubate cells in presence of the primary antibody for 1 h at RT
or overnight at 4 C(see Note 3). If required, other primary
antibodies, raised in different species, can be co-incubated
together with the anti-ADPR antibody (see Note 4). To stain
for mitochondria, antibodies specific, e.g., to COX-IV (Abcam,
168 Ann-Katrin Hopp and Michael O. Hottiger
33985) or ATP5a (Abcam, 14748) can be used (Fig. 2). A
co-staining is especially advisable, if the ADP-ribosylation sig-
nal needs to be quantified. To economize antibody, 20μL
staining mix (antibody in blocking solution) is prepared per
coverslip and placed on top of the parafilm as separate drops.
Subsequently, each coverslip is placed on top of a separate drop,
with the cells facing the staining mix (see Note 5).
4. After the incubation, transfer each coverslip back into its origi-
nal plate/well for washing (coverslips have to be turned around
again) and wash 2–3 times with 250–500μL PBS (coverslips
have to be fully covered with liquid).
5. Dilute the secondary antibody 1:250 in blocking solution,
replace the PBS from cells with the blocking solution contain-
ing the secondary antibody, and incubate it for at least 1 h at
RT. Comparably as for the primary antibody, additional sec-
ondary antibodies can be added in a scaled-down volume of
20μL per coverslip, if desired.
Fig. 1 Scheme of the wet chamber: Wet chamber used to prevent cells and
antibody solution from drying out. The bottom of a plastic box is lined out evenly
with Whatman paper, which is subsequently wetted with water and covered with
a piece of parafilm
Fig. 2 Mitochondrial ADP-ribosylation can be detected via IF: Co-staining for ADP-ribosylation (ADPR, red)
together with the mitochondrial-localized ATP5a (green)
Immunofluorescence Analysis of Mitochondrial ADP-Ribosylation 169
6. After incubation, transfer coverslips back to their original
plate/well if required and wash cells 2–3 times with
250–500μL PBS (coverslips have to be fully covered with
liquid).
3.5 DAPI Staining
and Mounting
1. To visualize single nuclei (and thereby single cells), cells are
stained with 250–300μL DAPI solution per 24 well for
20 min RT.
2. Wash cells once with 250–300μL PBS.
3. Prior to mounting, add an additional wash step by dipping each
coverslip into double-distilled water and subsequently remov-
ing the excess of liquid by quickly drying the coverslip on top of
tissue paper (cells should not be facing the paper).
4. In the meantime, clean glass slides with ethanol if required, and
place drops of approximately 5–6μL of mounting solution
(e.g., Mowiol) per coverslip on top of the glass slide (up to
ten 12 mm coverslips can be mounted per slide).
5. Place one coverslip on top of each drop, with the cells facing
the mounting medium (coverslips have to be turned around
again).
6. Remove any excess of mounting medium from the glass slides
and let them dry for the required amount of time (at least some
hours in the case of Mowiol).
3.6 Analysis/
Quantification
of the Immuno-
fluorescence Signal
1. Mitochondrial ADP-ribosylation signals can be quantified at a
single-cell level with, e.g., the following software: FIJI, Cell-
Profiler (free software), or ScanR (Olympus).
2. Because of the cellular heterogeneity of the ADP-ribosylation
signal, it is highly advisable to create a mitochondrial mask
based on another staining (e.g., COX-IV or ATP5a) and to
access the ADP-ribosylation signal intensity within this mask.
4 Notes
1. As Triton X-100 is quite viscous, it is advisable to use the
P1000 for pipetting, or, alternatively, to cut the tip of the
pipette.
2. Before usage, check solutions for potential contaminations
(especially blocking solution, as it contains BSA).
3. Although both incubations work, staining quality is increased
after overnight incubation at 4 C.
4. Under unstressed conditions and due to the low ADPR in
other compartments, the used anti-ADPR antibody recognizes
mainly mitochondrial ADPR (see Fig. 2).
170 Ann-Katrin Hopp and Michael O. Hottiger
5. To facilitate the transfer of each coverslip from the 24-well plate
to the wet chamber and back, tweezers and needles can
be used.
Acknowledgments
The authors would like to thank Tobias Suter (University of Zur-
ich) for providing editorial assistance and critical input during
manuscript writing. Work on ADP-ribosylation in the laboratory
of M.O.H is supported by the Kanton of Zurich and the Swiss
National Science Foundation (SNF 31003A_176177).
References
1. Spinelli JB, Haigis MC (2018) The multiface-
ted contributions of mitochondria to cellular
metabolism. Nat Cell Biol 20(7):745–754
2. Chandel NS (2015) Evolution of mitochondria
as signaling organelles. Cell Metab 22
(2):204–206
3. Quiro
´s PM, Mottis A, Auwerx J (2016) Mito-
nuclear communication in homeostasis and
stress. Nat Rev Mol Cell Biol 17(4):213–226
4. Yang W, Nagasawa K, Mu¨nch C et al (2016)
Mitochondrial Sirtuin network reveals dynamic
SIRT3-dependent Deacetylation in response to
membrane depolarization. Cell 167
(4):985–1000
5. Kruse R, Højlund K (2017) Mitochondrial
phosphoproteomics of mammalian tissues.
Mitochondrion 33:45–57
6. Tan EP, McGreal SR, Graw S et al (2017)
Sustained O-GlcNAcylation reprograms mito-
chondrial function to regulate energy metabo-
lism. J Biol Chem 292(36):14940–14962
7. Hopp A-K, Teloni F, Gondrand C, et al Mito-
chondrial ADP-ribosylation controls nuclear
ARTD1-induced PARylation and PARP inhib-
itor efficiency, accepted in Mol Cell
8. Hottiger MO (2015) Nuclear
ADP-Ribosylation and its role in chromatin
plasticity, cell differentiation, and epigenetics.
Annu Rev Biochem 84:227–263
9. Hottiger MO (2015) SnapShot:
ADP-ribosylation signaling. Mol Cell 58
(6):1134–1134
10. Lu¨scher B, Bu¨ tepage M, Eckei L et al (2018)
ADP-Ribosylation, a multifaceted posttransla-
tional modification involved in the control of
cell physiology in health and disease. Chem Rev
118(3):1092–1136
11. Haigis MC, Mostoslavsky R, Haigis KM et al
(2006) SIRT4 inhibits glutamate
dehydrogenase and opposes the effects of calo-
rie restriction in pancreatic βcells. Cell 126
(5):941–945
12. Liszt G, Ford E, Kurtev M, Guarente L (2005)
Mouse Sir2 homolog SIRT6 is a nuclear
ADP-ribosyltransferase. J Biol Chem 280
(22):21313–21320
13. Kun E, Zimber PH, Chang AC et al (1975)
Macromolecular enzymatic product of NAD+
in liver mitochondria. Proc Natl Acad Sci 72
(4):1436–1440
14. Leutert M, Menzel S, Braren R et al (2018)
Proteomic characterization of the heart and
skeletal muscle reveals widespread arginine
ADP-Ribosylation by the ARTC1 Ectoen-
zyme. Cell Rep 24(7):1916–1929
15. Valm AM, Cohen S, Legant WR et al (2017)
Applying systems-level spectral imaging and
analysis to reveal the organelle interactome.
Nature 546(7656):162–167
16. Hopp A-K, Gru¨ter P, Hottiger MO (2019)
Regulation of glucose metabolism by NAD+
and ADP-Ribosylation. Cells 8(8):890
17. Agnew T, Munnur D, Crawford K et al (2018)
MacroD1 is a promiscuous ADP-ribosyl hydro-
lase localized to mitochondria. Front Micro-
biol 9:20
18. Neuvonen M, Ahola T (2009) Differential
activities of cellular and viral macro domain
proteins in binding of ADP-ribose metabolites.
J Mol Biol 385(1):212–225
19. Voorneveld J, Rack JGM, Ahel I et al (2018)
Synthetic α- and β-Ser-ADP-ribosylated pep-
tides reveal α-Ser-ADPr as the native epimer.
Org Lett 20(13):4140–4143
20. Xu Y, Zhou P, Cheng S et al (2019) A bacterial
effector reveals the V-ATPase-ATG16L1 axis
that initiates xenophagy. Cell 178(3):552–566
Immunofluorescence Analysis of Mitochondrial ADP-Ribosylation 171
Chapter 13
Assessment of Mitochondrial Ca
2+
Uptake
Andra
´s T. Deak, Claire Jean-Quartier, Alexander I. Bondarenko,
Lukas N. Groschner, Roland Malli, Wolfgang F. Graier,
and Markus Waldeck-Weiermair
Abstract
Mitochondrial Ca
2+
uptake regulates mitochondrial function and contributes to cell signaling. Accordingly,
quantifying mitochondrial Ca
2+
signals and elaborating the mechanisms that accomplish mitochondrial
Ca
2+
uptake are essential to gain our understanding of cell biology. Here, we describe the benefits and
drawbacks of various established old and new techniques to assess dynamic changes of mitochondrial Ca
2+
concentration ([Ca
2+
]
mito
) in a wide range of applications.
Key words Mitochondrial Ca
2+
uptake, Calcium Green, Fura-2, Rhod-2, FRET , Oxidative phos-
phorylation, Mitochondrial membrane potential, Mitoplast, Patch-clamp recording, Ca
2+
Imaging
1 Introduction
Ca
2+
transfer into the mitochondrial matrix is linked to numerous
important cellular functions including the regulation of energy
metabolism [1], mitochondrial ROS production [2], and cell sur-
vival and death [3]. The significance and complexity of mitochon-
drial Ca
2+
sequestration are highlighted by several recent studies
describing key components of the mitochondrial Ca
2+
uptake
machinery [4–12]. In order to study and to visualize the complex
process of mitochondrial Ca
2+
handling, various chemical and
genetically encoded Ca
2+
sensors, as well as electrophysiological
techniques have been developed and utilized over the past decades.
Here, we introduce some common methods, indicators applied on
isolated mitochondria or permeabilized as well as intact cells in
order to assess and quantify mitochondrial Ca
2+
uptake dynamics.
Chemical indicators exhibit altered fluorescent properties when
bound to Ca
2+
and are usually loaded into permeabilized or intact
cells by passive incubation. The most popular chemical Ca
2+
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_13,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
173
indicators used to assess mitochondrial Ca
2+
uptake are Rhod-2 and
Calcium Green. Rhod-2 accumulates predominantly within the
mitochondrial compartment and responds with increased fluores-
cence upon [Ca
2+
]
mito
elevation [13]. On the other hand,
Calcium Green is applied as an indirect reporter of mitochondrial
Ca
2+
uptake by measuring the decline of the extramitochondrial
Ca
2+
concentration upon an activation of mitochondrial Ca
2+
uniport [14].
Genetically encoded mitochondrial Ca
2+
probes are protein-
based fluorescent indicators targeted into mitochondrial matrix
(MM) via a specific mitochondrial signal sequence [15], and they
are typically incorporated into cells by gene transfer techniques.
These probes are usually divided in two classes: the single fluores-
cent protein (FP) and the Fo
¨rster resonance energy transfer
(FRET)-based Ca
2+
sensors. Single FP sensors including the peri-
cams [16], the GCamPs [17], and the recently developed GECOs
[18] consist of a circularly permuted FP (cpFP) flanked by a geneti-
cally modified Ca
2+
binding domain, the calmodulin (CaM) and its
interacting peptide (M13) from the myosin light-chain kinase. A
common feature of these sensors is the ability of their integrated
cpFPs to modulate their spectral properties in response to changes
of Ca
2+
concentration based on the interaction of CaM with the
M13 peptide.
The other large group of fluorescent Ca
2+
reporters is the
FRET-based cameleons, which consist of two different fluorescent
proteins possessing overlapping spectral properties. The “proto-
type” of such sensors contains the previously mentioned CaM and
M13 domains, which were inserted in tandem between the cyan
and the yellow fluorescent proteins (CFP/YFP) [19]. Cameleons in
aCa
2+
-bound state undergo a conformational change, whereupon
the donor CFP gets in close proximity to the acceptor YFP yielding
an enhanced energy transfer between the two fluorophores. Came-
leons are thus ratiometric indicators: increase of FRET is coupled
with the decrease of CFP fluorescence. Since the introduction of
the first cameleon in 1997, several derivates of this Ca
2+
sensor with
improved Ca
2+
sensitivities [20], higher FRET efficiencies,
increased pH stability, and appropriate mitochondrial targeting
have been developed [21,22]. However, none of these
mitochondrial-targeted sensors were suitable for a co-imaging
with the most popular cytosolic Ca
2+
indicator, Fura-2, because of
a significant overlap in the excitation and emission spectra [23]. To
resolve this problem, a novel mitochondrial-targeted, red-shifted
FRET-based sensor referred to as mtD1GO-Cam was recently
constructed. This genetically encoded Ca
2+
probe consists of a
green and orange fluorescent protein and was successfully used to
simultaneously measure [Ca
2+
]
cyto
and [Ca
2+
]
mito
in a ratiometric
manner within same individual cells [12,24].
174 Andra
´s T. Deak et al.
Although optical methods are most often used to investigate
mitochondrial Ca
2+
uptake, this approach bears some limitations.
Specifically, it does not allow controlling key parameters which
determine driving flux of Ca
2+
uptake, such as the mitochondrial
membrane potential and Ca
2+
concentration gradient. Additionally,
optical methods reflect rather the dynamical changes of Ca
2+
con-
centration in a specific cellular compartment, which is always a
balance between Ca
2+
influx, Ca
2+
extrusion, and Ca
2+
buffering
capacity, but not the Ca
2+
flux itself. Patch-clamp approach cir-
cumvents the aforementioned drawbacks associated with optical
methods and represents a direct method of assessing transmito-
chondrial ion fluxes and determining channel conductances
[10,25].
Considering the close interdependency of mitochondrial Ca
2+
uptake and metabolism [1], it is possible to indirectly assess mito-
chondrial Ca
2+
signals by measuring cellular O
2
consumption rates
(OCRs). In many cases, the analysis of metabolic activity is also an
essential tool in estimating mitochondrial Ca
2+
handling. The
availability of analytical instruments for measuring OCRs has led
to broad applications of this technique in the field of mitochon-
drial Ca
2+
signaling [5,26]. To this end, we also describe how to
use the Seahorse XF96 Extracellular Flux Analyzer in order
to measure cellular O
2
consumption in dependence of mitochon-
drial Ca
2+
.
2 Materials
Notably, all chemicals and reagents used are suitable for most
adherent mammalian cell lines, but need to be adjusted and opti-
mized for every cell type of choice. All buffers should be prepared
using deionized water and analytical grade reagents in order to
avoid possible ionic contaminants and at room temperature to
obtain best solubility options, unless otherwise indicated.
2.1 Chemical
Fluorescent Indicator
Components
1. Intracellular assay buffer: 110 mM KCl, 0.5 mM KH
2
PO
4
,
1 mM MgCl
2
, 0.03 mM HEPES, 5 mM succinate, 10 mM
glucose; pH adjusted with KOH to 7.4 (see Notes 1 and 2).
For storage, freeze aliquots at 20 C.
2. Permeabilization reagent: Prepare 0.1% or 1% stock solution of
digitonin by dissolving 1–10 mg in 1 ml of H
2
O
(0.81–8.1 mM). Heat the mixture up to 95 C until a clear
solution is obtained. Let it cool down to RT before use
(Table 1) and store at 4 C. The optimal final concentration
of digitonin depends on cell-type or sample material and has to
be adjusted accordingly.
Measuring Mitochondrial Calcium 175
3. Chemical sensors: Prepare stock solutions with DMSO and
freeze aliquots corresponding to manufacturer’s protocol
(Table 2). Accordingly, on the day of experiment, prepare
working dilutions using deionized water or a buffer of choice.
4. Further chemicals for measurement: Prepare 1 mM and 10 mM
CaCl
2
stock solutions in deionized water and apply 1–10 μM
Ca
2+
pulses into the cuvette (Fig. 1). Optionally prepare work-
ing stocks of inhibitors such as ruthenium red, cyclopiazonic
acid, or FCCP (Table 1).
Table 1
Description of established mitochondrial-targeted genetically encoded Ca
2+
Compound Source Molecular weight Solvent Stock (mM) Working solution/s (μM)
Histamine Sigma 184.07 H
2
O 100 100, 10, 1
ATP Sigma 507.20 H
2
O 100 100, 10, 1
Carbachol Sigma 182.65 H
2
O 100 100, 10, 1
CGP37157 Abcam 324.22 DMSO 100 10, 20
BHQ Sigma 222.33 DMSO 100 15
Cyclopiazonic acid Abcam 336.39 H
2
O 10 2–10
Thapsigargin Abcam 650.76 DMSO 1 1
Rotenone Sigma 394 DMSO 100 50
Ruthenium red Sigma 790.39 H
2
O1 10
Digitonin Sigma 1229.31 H
2
O 0.81–8.1 10–100
Oligomycin Abcam 786.78 DMSO 10 2
Antimycin Sigma 534.645 EtOH
abs
100 10
Ionomycin Abcam 709.01 DMSO 10 1–10
FCCP Abcam 255.97 EtOH
abs
10 0.2–4
mtRP mitochonrial targeted ratiometric pericam, cp circularly permutated, ECFP/CGFP enhanced CFP/GFP, mKO2
monomeric Kusabira-Orange 2
Table 2
Chemical sensors in stock and working concentrations including corresponding fluorimeter settings
Sensor Stock (mM) Final (μM) Kd (μM) Excitation λ(nm) Emission λ(nm)
Calcium Green 5 N 1 (DMSO) 2 14 506 10 532 10
Fura-5 K, 1 (DMSO) 1 0.2 340/380 5, 510 20,
Fura-2-AM 1 (DMSO) 3.3 0.2 340/380 10 510 10
Rhod-2-AM 1 (DMSO) 5 0.6 550 10 590 10
176 Andra
´s T. Deak et al.
2.2 Components
for the Use
of Genetically Encoded
Ca
2+
Indicators
1. Vector: All plasmids encoding for mitochondrial-targeted
GECIs (mtRP, mtD3cpv, mtD1GO-Cam, and mito-GEM-
GECO1) (Table 1) were constructed in a pcDNA3 vector
(Invitrogen, Carlsbad, CA, USA). This vector contains a
human cytomegalovirus immediate early (CMV) promoter
for high-level expression in a wide range of mammalian cells
(see Note 3).
2. Transformation: Many E.coli strains are suitable for transfor-
mation of these plasmids including TOP10, TOP10F´,
DH5α-T, JM101, or XL1-Blue. Most strains are available as
chemically competent or electrocompetent cells. However, the
plasmids can be transformed in any method of choice.
3. Antibiotic resistance: Select transformants on Lysogeny Broth
(LB) plates containing 2% Agar Agar and 50–100 μM/ml
ampicillin. For bacterial overnight culture, shake picked colo-
nies in a shaking incubator at 37 C using 100–250 ml LB
(10 g NaCl, 5 g yeast extract, 10 g tryptone/l) with
50–100 μM/ml ampicillin.
4. Plasmid isolation: Plasmid can be isolated with any plasmid
isolation kit. We recommend using the PureYield Plasmid
Maxiprep System (Promega, Mannheim, Germany).
Fig. 1 Indirect assessment of mitochondrial Ca
2+
uptake by Calcium Green 5 N.
Digitonin-treated HeLa cells (10
^8
) were exposed to repetitive pulses
(100–200 s/pulse) of 50 μMCa
2+
added exogenously to the bath, which also
contained Calcium Green 5 N. The decline of each Ca
2+
signal indicates the
magnitude of mitochondrial Ca
2+
uptake
Measuring Mitochondrial Calcium 177
5. Transfection medium: Dulbecco’s Modified Eagle’s
Medium—low glucose (Sigma Chemicals, St. Louis, MO,
USA) is suitable for most mammalian cell types (see Note 4).
6. Transfection reagent: We recommend using TransFast™(Pro-
mega, Mannheim, Germany) for transient transfection of the
sensor plasmid (see Note 5).
7. Creation of stable cell lines: Use G418 (Geneticin
®
) antibiotic
resistance of the vector for mammalian cell line (see Note 6).
8. Storage buffer (SB): 138 mM NaCl, 5 mM KCl, 2 mM CaCl
2
,
1 mM MgCl
2
, 10 mM HEPES, 2.6 mM NaHCO
3
, 0.44 mM
KH
2
PO
4
, 10 mM D-glucose supplemented with 0.1% vita-
mins, 0.2% essential amino acids, 1% penicillin/streptomycin,
and 1% fungizone (PAA Laboratories, Linz, Austria); pH
adjusted to 7.4 with NaOH (see Notes 7 and 8).
9. Experimental buffer (EB): 138 mM NaCl, 5 mM KCl, 2 mM
CaCl
2
, 1 mM MgCl
2
, 10 mM D-glucose, and 10 mM HEPES;
pH adjusted to 7.4 with NaOH (see Note 8).
10. Stock solution and working solution of compounds: Add ago-
nist(s), mediator(s), or any other compound to EB in a con-
centration of at least 1:1000 (Table 1) to mobilize Ca
2+
in a
distinct cell line.
11. Perfusion system: For real-time measurements, it is best to use
a perfusion chamber allowing a continuous perfusion of cells
with EB compound(s).
12. Digital wide-field imaging system (see [12,24,27,28]):
Co-imaging of Fura-2 and mtD1GO-Cam can be performed
on an inverted (digital) wide-field microscope using a 40oil
immersion objective (see Note 9). Use a high-speed polychro-
mator system to allow a fast switching of excitation wave-
lengths. Select suitable excitation filters according to the
sensor(s) need (Tables 2and 3), e.g., use an E500spuv and a
495dcxru (Chroma Technology Corp, Rockingham Vermont,
USA) for simultaneous excitation of Fura-2 and mtD1GO-
Cam. Collect emission light with proper dichroic filters
(Tables 2and 3); for FRET- and GECO-based indicators, use
either a motorized filter wheel or a beam splitter, both
equipped with appropriate dichroic filters. Images are best
recorded with a thermoelectric cooled charged-coupled
device (CCD) camera. For data acquisition, device
controlling, or post-acquisition image analysis, we recom-
mend to use an up-to-date software of MetaFluor, VisiView
(Universal Imaging, Visitron), or the live acquisition software
(TILL Photonics).
178 Andra
´s T. Deak et al.
2.3 Components
for Mitoplast Isolation
and Patch-Clamp
Recordings
1. Mitochondrial storage buffer: 10 mM HEPES, 250 mM
sucrose, 1 mM ATP, 0.08 mM ADP, 5 mM succinate, 2 mM
KH
2
PO
4
, 1 mM DTT; pH adjusted to 7.4 with KOH.
2. Hypotonic solution: 5 mM HEPES, 5 mM sucrose, 1 mM
EGTA; pH adjusted to 7.2 with KOH.
3. Hypertonic solution: 750 mM KCl, 80 mM HEPES, 1 mM
EGTA; pH adjusted to 7.2 with KOH.
4. Isolation of mitochondria: There are various protocols and kits
available, mostly based on cell disruption and homogenization
followed by differential centrifugation, such as the mitochon-
dria isolation kit for cultured cells (Thermo Scientific 89874,
USA). Suspend mitochondrial fraction in mitochondrial stor-
age buffer.
5. Micropipettes: For mitoplast patch-clamp recordings, micro-
pipettes are pulled from borosilicate glass capillaries and fire
polished [10]. When filled with KCl-containing solution, the
pipettes typically have a resistance of 8–12 MOhms.
6. Microscope: For mitoplast visualization, we used a Zeiss Axio-
vert 135 M microscope equipped for phase contrast with a 63
objective.
7. Standard bath solution: 150 mM KCl, 1 mM EGTA, 10 mM
HEPES, pH adjusted to 7.2 with KOH.
8. Pipette solution for single channel recordings: 105 mM CaCl
2
,
10 mM HEPES supplemented with 0.01 mM CsA, 0.01 mM
CGP37187; pH adjusted to 7.2 with Ca(OH)
2
.
Table 3
Concentration and solvent of relevant compounds in stock and working solution
Sensor [Reference]
Incorporated
Fluorophore(s)
Excitation λ
(nm)
c
Emission λ
(nm)
c
Mitochondrial
Targeting
a
mtRP [16] Single FP: cpYFP 430 10 (Ca
2+
)
480 5(H
+
)
b
535 10 COX4
mtD3cpV [20] FRET:
ECFP (donor)
cpVenus (acceptor)
430 10 480 5
535 10
(2) COX8
mtD1GO-CaM [24] FRET: cpEGFP(donor)
mKO2 (acceptor)
480 5 510 5
560 10
(2) COX8
Mito-GEM-GECO1
[18]
Single FP: cpGFP
d
377 50 447 40
520 35
(2) COX8
a
N-terminal targeting sequence obtained from the mitochondrial cytochrome C oxidase subunit (COX) 4/8
b
mtRP has two excitation maxima: a Ca
2+
-dependent and a H
+
-dependent, thus, can measure [Ca
2+
]
mito
and [H
+
]
mito
simultaneously
c
Aminoacid substitutions relative to GCamP3 are as follows: L60P/K69E/N77Y/D86G/N98I/K119I/L173Q/
T223S/N302S/R377P/K380Q/S404G/E430V
d
Values refer to major excitation/emission peak optimal bandwith
Measuring Mitochondrial Calcium 179
9. Pipette solution for whole-mitoplast recordings: 120 mM Cs
methanesulfonate, 30 mM CsCl, 1 mM EGTA, 110 mM
sucrose, 10 mM HEPES; pH adjusted to 7.2 with CsOH.
10. Ca
2+
-free bath solution: 130 mM Trizma HCl, 50 mM Trizma
base, 1 mM EGTA, 1 mM EDTA, 10 mM HEPES; pH 7.2.
11. Ca
2+
-containing bath solution: 130 mM Trizma HCl, 50 mM
Trizma base, 2–3 mM CaCl
2
, 1 mM EDTA, 10 mM HEPES;
pH 7.2. For I
Ca
recording, we add 1 mM EGTA instead of
2–3 mM CaCl
2
. For recording monovalent cationic current
carried by Na
+
, use 150 mM NaCl instead of Trizma and the
same Cs-based pipette solution and voltage protocol.
2.4 Components
for Measuring Cellular
O
2
Consumption Rates
1. Instruments: XF96 Extracellular Flux Analyzer from Seahorse
Bioscience including the controller; 37 C non-CO
2
incubator
(or XF Prep Station); XF96 4-port FluxPak or FluxPak mini (see
Note 10); pH meter; phase contrast microscope; pipettes
including a multichannel pipette and matching tips.
2. Solution: XF Calibrant Solution (Seahorse Bioscience Part
#100840-000).
3. Assay medium: Non-buffered DMEM (see Note 11).
4. Substrates: D-glucose and sodium pyruvate.
5. Reagents: Oligomycin A, carbonyl cyanide
4-(trifluoromethoxy)phenylhydrazone (FCCP), antimycin A
(see Note 12).
3 Methods
Carry out all procedures at room temperature unless indicated
otherwise.
3.1 Measuring
Indirect Ca
2+
Uptake
Via Calcium Green
1. For each measurement, use 10
6
cells/ml or isolated mitochon-
dria (see Note 13) with a protein content of 2 mg/ml (see
Note 14). Wash cells or isolated mitochondria in EGTA-
containing buffer before starting with procedures (see Note 15).
2. Resuspend cells in 2 ml assay buffer and add fresh 25 μl of the
0.1–1% digitonin stock (final concentration of 10–100 μM)
before measurement or resuspend isolated mitochondria in
2 ml assay buffer.
3. Transfer to 2 ml fluorimeter cuvette (see Note 16) including a
stirrer for mixing.
4. Load protocol on the fluorimeter and insert the cuvette.
Recommended settings for Hitachi fluorimeter are given in
Table 2. Blank the cuvette before adding the dye.
180 Andra
´s T. Deak et al.
5. Start measurement: Upon stablebaseline,addafirstpulseof6μl
10 mM CaCl
2
(¼30 μM) to overcome the EGTA (in buffer). Add
further pulses of 50 μMCaCl
2
(Fig. 1)(see Notes 17 and 18).
6. Repeat measurement using a new sample and add various
inhibitors as described above before blanking the cuvette.
3.2 Measuring Direct
Mitochondrial Ca
2+
Uptake Via Fura-2 or
Rhod-2
1. Grow cells on coverslips to ~70% confluency or use isolated
mitochondria >2 mg/ml in isolation buffer.
2. Incubate cells in growth medium or isolated mitochondria in
mitochondrial storage buffer with 3.3 μM Fura-2-AM for
60 min or with 5 μM Rhod-2-AM for 30 min at room temper-
ature in the dark (see Note 20).
3. After the incubation period, change growth medium to assay
buffer or pellet the sample, wash with EGTA containing buffer
(see Note 15), and let isolated mitochondria settle down onto
the microscopy slides in the assay buffer for at least 20 min (see
Note 19).
4. Transfer slide to the microscope and proceed with the measure-
ment by real-time monitoring Fura-2 or Rhod-2 using an
imaging system as described above with the excitation and
emission filter settings displayed in (Table 2).
3.3 Transfection
of Genetically Encoded
Ca
2+
Indicators
1. Grow adherent mammalian cells in its optimum culture
medium in a humidified incubator (37 C, 5% CO
2
, 95% air)
on perfusion chamber slides to 60–80% confluency.
2. For transient transfection of ~5 10
5
cells, mix 2 μgofa
plasmid encoding a mitochondrial targeted GECI and an appro-
priate amount of transfection reagent with 1 ml of serum- and
antibiotic-free transfection medium. Incubate cells in the incu-
bator for 16–20h and change back to complete culture medium.
Experiments can be performed 24–72 h after transfection.
3.4 Real-Time
Recordings of [Ca
2
+
]
cyto
and [Ca
2+
]
mito
Herein, we will describe the simultaneous measurement of Fura-
2 and mtD1GO-Cam in detail. For assessing mitochondrial Ca
2+
using chemical fluorescent indicators (Fura-2, Rhod-2) or other
mitochondrial-targeted genetically encoded Ca
2+
indicators (mtRP,
mtD3cpv, mito-GEM-GECO1), use appropriate filter settings as
described in Tables 2and 3.
1. Load mtD1GO-Cam transfected cells with Fura-2-AM in a
concentration of 3.3 μM dissolved in storage buffer for at
least 20 min (see Note 20).
2. Stop Fura-2-AM loading by washing the cells twice with stor-
age buffer.
3. Keep cells in storagebuffer prior to measurements (see Note 21).
4. Put a drop of immersion oil on top of the objective and place
the perfusion chamber slide with the cells upside onto the
droplet.
Measuring Mitochondrial Calcium 181
5. Connect the chamber with the perfusion system and start the
perfusion.
6. Set the cells in focus by turning the z-tuner of the microscope
table or use the autofocus of the system in the white
light mode.
7. Use the ocular for searching cells that have a high and good
targeted expression of the mitochondrial cameleon using an
excitation wavelength at ~480 nm and emission at ~510 nm
(cpEGFP, see Notes 22 and 23).
8. For simultaneous illumination of Fura-2 and mtD1GO-Cam,
use the time-lapse function of the imaging software in a triple
wavelength mode. To gain better fluorescence sensitivity, use
binning 2 or higher. Expose excitation of Fura-2 at 340 nm and
380 nm for 150 ms or 50 ms, respectively (see Note 24), and
collect emitted light at 510 nm. The mitochondrial sensor gets
excited for 400 ms (see Note 25) at 477 nm and emits at
510 nm (GFP, donor fluorescence) and 560 nm (FRET, accep-
tor fluorescence), respectively. Accordingly, the two sensors get
recorded in the time-lapse mode within 600 ms or less (see
Notes 24 and 25) by alternative exposes of the three excitation
wavelengths without any fluorescence interference (Fig. 2).
0 1 2 3 4 5 6
20
30
40
50
60
70
80
90
100
110
120
130
F340
F380
Histamine
Time(min)
Fintensity (a.U.)
0 1 2 3 4 5 6
200
250
300
350
400
0.5
0.6
0.7
0.8
0.9
1.0
FRETraw
GFPraw
Ratioraw
R0
Histamine
Time(min)
Fintensity (a.U.)
Ratio (F560/F510)
0 1 2 3 4 5 6
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0.96
0.98
1.00
1.02
1.04
1.06
1.08
1.10
1.12
1.14
1.16
Fura2
Histamine
mtD1GO-Cam
Time(min)
Ratio (F340/F380)
Ratio (F560/F510)/R0)
cba
Fig. 2 Simultaneous measurement of [Ca
2+
]
cyto
and [Ca
2+
]
mito
in the same individual cells. Original traces of
(a) cytosolic and (b) mitochondrial Ca
2+
signals and (c) their respective correlation over time upon cell
stimulation with 100 μM histamine in intact mtD1GO-Cam expressing HeLa cells loaded with Fura-2. (a) Raw
traces of Fura-2 signals at 340 nm (grey) and at 380 nm (blue) excitation.Inset image shows the cytosolic
accumulation of Fura-2 (b) Raw traces of GFP signal (GFP
raw
, green) at 510 nm excitation and respective FRET
signal (FRET
raw
, orange) at 560 nm excitation were plotted on the left y-axis. Red curve (Ratio
raw
) indicates the
FRET ratio computed from the raw traces (F
FRET
/F
GFP)
was plotted on the right y-axis. Black curve represents
photobleaching function (R
0
) assessed with a one phase exponential decay function. Inset image shows
mitochondrial targeting of mtD1GO-Cam. (c) Fura-2 (black, left y-axis) and mtD1GO-CaM signals (red, right
y-axis) were calculated from the raw traces shown in (a) and (b), respectively.Inset image is an overlay of
previous insets. The scale bar is 10 μM
182 Andra
´s T. Deak et al.
9. Use an appropriate experimental design for cell stimulation via
the perfusion system (e.g., 100 μM histamine in EB).
10. Analyze data of the recorded ratios from the two sensors sepa-
rately to verify the spatiotemporal correlation of [Ca
2+
]
cyto
and
[Ca
2+
]
mito
(see Note 26).
3.5 Mitoplast
Patch-Clamping
Recording
1. Mitochondria from cultured cells (e.g., HeLa) are freshly
isolated by differential centrifugation steps (see Note 27)as
previously described [10,25].
2. Mitoplast formation: Incubate isolated mitochondria kept on
ice, in four volumes of hypotonic solution for 10 min. This
results in mitochondria swelling and rupture of the outer
membrane. Add 1 volume of hypertonic solution to equilibrate
the tonicity (see Note 28).
3. For patch-clamp recordings, place 20–40 μl of mitoplast sus-
pension to the recording chamber (depending on the size of
mitochondrial pellet) and allow mitoplasts to settle down for
10 min prior to experimentation.
4. To form Gigaohm contact, position the pipette tip on the
chosen mitoplast away from the “cap” region, which represents
the attached remnants of the outer membrane. Press the
pipette against the mitoplast and apply negative pressure to
the pipette interior. When mitoplast-attached configuration is
reached, single channel openings can be detected during volt-
age ramps.
5. Before obtaining whole-mitoplast configuration, capacitance
transients are compensated and negative pressure is further
applied until the patch is ruptured.
6. Alternatively, voltage steps of 300–600 mV and 10–20 ms
duration may be applied.
7. Successful access to the matrix is accompanied by reappearance
of capacitance transients. Mitoplast capacitance measured with
Membrane Test tool of Clampex is around 1 pF. If membrane
rupture is accompanied by a leak current, new mitoplast should
be chosen and the procedure should be repeated again with a
new pipette.
8. Following a successful membrane rupture, experiments are
continued in the same way as they are done with cells.
9. For recordings of Ca
2+
currents in mitoplast-attached and
whole-mitoplast configuration, use respective pipette solutions
described above. Signals obtained are sampled at 10 kHz and
filtered at 1 kHz.
10. For recordings of Ca
2+
currents in whole-mitoplast configura-
tion, exchange Ca
2+
-free bath solution for Ca
2+
-containing
bath solution by bath perfusion. We typically apply voltage
Measuring Mitochondrial Calcium 183
ramps from 160 to +60 mV to record whole-mitoplast Ca
2+
currents (Fig. 3).
11. For recording whole-mitoplast monovalent cationic current
carried by Na
+
, use 150 mM NaCl instead of Trizma and the
same Cs-based pipette solution and voltage protocol.
12. Collect data using the Clampex software of pClamp (Molecular
Devices, Sunnyvale, CA, USA). Signals obtained are sampled at
5 kHz and filtered at 1 kHz.
3.6 Assessing Ca
2
+
-Dependent Changes
in Mitochondrial
Metabolism
1. Harvest, resuspend, and dilute cells in standard growth
medium. Typical cell seeding numbers vary from 5000 to
100,000 cells per well depending on the cell type, basal meta-
bolic activity, proliferation rate, cell size, and the time of plating
and must be determined empirically.
2. Gently seed 100 μl of cell suspension per well in XF96 Polysty-
rene Cell Culture Microplates (see Note 29). Wells A1, A12,
H1, and H12 are to be left blank for background correction.
3. After seeding, place the microplate with the cells in an incuba-
tor at 37 C, gassed with 5% CO
2
for 12 h to guarantee
optimal adherence (see Note 30).
4. Fill each well of the utility plate with 200 μl of XF calibrant
solution using a multichannel pipette and lower the XF96
sensor cartridge onto the plate, fully submerging the biosen-
sors (see Note 31).
5. Seal both the cartridge and plate, covered by the lid, using
parafilm in order to minimize evaporation and incubate at
37 C in a non-CO
2
incubator.
Fig. 3 Patch-clamp recording of transmitochondrial Ca
2+
flux. (a) Time course of the whole-mitoplast current
development at 155 mV before and after addition of 3 mM Ca
2+
followed by addition of 10 μM RuR. (b)
Corresponding Ca
2+
current responses to voltage ramps from 160 to 50 mV before and after addition of
10 μM RuR in the presence of 3 mM Ca
2+
184 Andra
´s T. Deak et al.
6. Switch on the instrument and open the XF software at least
12 h prior to the assay, in order to allow the system to stabilize
at 37 C. Table 4summarizes a commonly used protocol
template that can be programmed and modified using the
Assay Wizard of the Seahorse XF96 Software.
7. On the day of the assay, observe cells under the microscope to
assure both sufficient viability and confluency (see Note 29).
8. Pre-warm non-buffered assay medium containing the desired
amount of D-glucose and sodium pyruvate (typically 5.5 and
1 mM, respectively) to 37 C and adjust pH to 7.4
using NaOH.
Table 4
Commonly used protocol template programmed using the Assay Wizard of
the Seahorse XF96 Software
Protocol start
1. Calibrate probes.
2. Equilibrate.
3. Loop 3 times.
4. Mix for 3 min 0 s.
5. Measure for 3 min 0 s.
6. Loop end.
7. Inject port A.
8. Loop 3 times.
9. Mix for 3 min 0 s.
10. Measure for 3 min 0 s.
11. Loop end.
12. Inject port B.
13. Loop 3 times.
14. Mix for 3 min 0 s.
15. Measure for 3 min 0 s.
16. Loop end.
17. Inject port C.
18. Loop 3 times.
19. Mix for 3 min 0 s.
20. Measure for 3 min 0 s.
21. Loop end.
Program end
Measuring Mitochondrial Calcium 185
9. Carefully remove the growth medium from the microplate
using a multichannel pipette, making sure that ~20 μlof
media remains at the bottom of the well at all times. Wash
cells two times before replenishing the well with a final volume
of 150 μl of assay medium.
10. Incubate the cell plate in a non-CO
2
incubator at 37 C until
ready for use.
11. Prepare compound solutions for injection using assay medium
and reconstituted reagents (see Notes 32 and 33).
12. Open the appropriate assay template and start the assay using
the XF96 software.
13. Insert the sensor cartridge and the utility plate (without the lid)
into the instrument for calibration.
14. Upon completion of the automated calibration process, replace
the utility plate with the microplate containing the cells and
click “continue” to start the actual measurement (Fig. 4).
4 Notes
1. Add 8.2 g KCl, 0.114 g KH
2
PO
4
, 0.203 g MgCl
2
, 0.766 g
HEPES, 1.351 g succinate, and 1.982 g glucose. Dissolve in
800 ml H
2
O while stirring for 5 min. Titrate KOH to adjust
Fig. 4 Mitochondrial respiration assessed by the XF96 Extracellular Flux Analyzer
from Seahorse Bioscience: an indirect way to determine mitochondrial Ca
2+
uptake. O
2
consumption rates (OCRs) of HeLa cells stably expressing control
shRNA or shRNA targeting the mitochondrial Ca
2+
uniporter (MCU shRNA). Cells
were treated with 1 μM oligomycin, 500 nM FCCP, and 2.5 μM antimycin A to
assess basal, coupled, maximal, and residual OCRs. Data represent means
(n30)
186 Andra
´s T. Deak et al.
pH, then add 0.0114 g EGTA and adjust pH again. While
adjusting pH, use low concentrations of acid or base at the
end of titrations in order to avoid sudden drops or rises above
or below the required pH. Continue stirring for 5 min and
check the pH again.
2. For dissolving EGTA properly, adjust pH first to slightly basic
conditions. During addition of the salt, pH will shift to lower
values. After complete dissolution adjust pH again.
3. The vector contains an episomal replication site for cell lines
that are latently infected with SV40 or express the SV40 large T
antigen (e.g., COS-1, COS-7).
4. For transfection, use the culture medium of the cell line of
choice without supplements like FCS, antibiotics, and
antimycotics.
5. Transfection reagent and method depend on cell type and
equipment of the lab.
6. The pcDNA3.1 () vector contains a neomycin resistance gene
for creating stable mammalian host cell of choice. Transfect the
cell line with the sensor plasmid using TransFast™transfection
reagent and transfection medium supplemented with an appro-
priate concentration of neomycin or G418 (Geneticin
®
) and
feed the cells with selective medium every day. Stable cell
colonies can be easily identified on a fluorescence microscope
within 3–4 days after addition of G418. Pick and expand
colonies in 96- or 48-well plates.
7. Storage buffer can be stored after sterile filtration in aliquots
(of e.g., 30 ml) at 4 C for at least 3 months.
8. SB and EB are suitable for most non-excitable cell lines (e.g.,
HeLa, HUVEC, HEK293, COS), but need to be adjusted to
distinct cell type and/or protocol of measurement.
9. Use low fluorescent immersion oil like a Cargille Immersion
Oil Type HF or Type LDF (Optoteam, Vienna, Austria).
10. A FluxPak comprises sensor cartridges containing the fluores-
cent biosensor as well as the injection ports, utility plates,
loading guides, and XF calibrant solution.
11. Non-buffered assay medium is usually based on the formula-
tion of Dulbecco’s Modified Eagle Medium including
L-glutamine but does not contain any buffering agent (i.e.,
sodium bicarbonate). Such medium can be obtained from, for
example, Seahorse Bioscience or Sigma Aldrich. For measure-
ments under Ca
2+
-free conditions, Ca
2+
-free assay medium has
to be custom made or replaced by assay solution (6).
12. Dissolve reagents in DMSO at stock concentrations of
5–10 mM and store at 20 C.
Measuring Mitochondrial Calcium 187
13. Try to use fresh isolated mitochondria, since frozen mitochon-
dria display impaired membrane integrity. Furthermore, opti-
mal number of strokes for dounce homogenization varies
between cell types and homogenizer before impairing the
integrity of mitochondrial membranes. Additional to up- and
down-movements, rotation speed can increase homogeniza-
tion outcome accordingly. Keep all samples during isolation
on ice, precool centrifuges, all glassware, flasks, Eppis, and
tools used during isolation process. Cell samples are handled
at RT.
14. Optimal number of cells for each measurement can vary
between cell types. Suitable protein content of isolated mito-
chondria additionally depends on mitochondrial integrity after
isolation. If no or weak mitochondrial Ca
2+
uptake is observed,
we recommend using higher amounts of sample material per
measurement.
15. Make sure to get rid of any excess Ca
2+
. In case of harvesting
cells, wash away any leftover medium with PBS (at low speed
according to cell line ~700 g). In case of isolated mitochon-
dria, samples are usually prepared in EGTA/EDTA or other
Ca
2+
chelating agent containing buffers (at high speed for
isolated organelles >10,000 g).
16. Fluorimeter cuvettes may vary for volume content. Try to stick
to equal concentrations (compare Material section), if shifting
down to lower volumes. Furthermore, make sure that the
sample is mixed continuously during the measurement; other-
wise both, the sensor and isolated mitochondria or cells, will
settle down negatively affecting the readout.
17. Further pulses of various Ca
2+
concentrations may be tried out.
The optimal concentration is dependent on the individual cell
line, number of mitochondria or cells. Calcium Green 5 N is
best used within a concentration of 10–100 μM which also
covers the range of MCU. For smaller concentration, one may
change to a different sensor with a suitable Kd (Table 1).
18. Observation: The signal goes up upon Ca
2+
addition and down
when taken up by mitochondria. This can be repeated depend-
ing on mitochondrial uptake capacity and cell number (cave:
always normalize to same cell number within samples to over-
come laborious calculations afterwards). After a certain addi-
tion of Ca
2+
, the signal will not completely fall down to the last
baseline followed by a steady increase of the signal. This is due
to permeability pore opening and a release of excess calcium
over a particular threshold known as induction of mitochon-
drial apoptosis.
19. When using a perfusion system on isolated mitochondria, care-
fully use slow perfusion <0.5 ml/min, since mitochondria only
188 Andra
´s T. Deak et al.
settle down and would not stick to the microscopy slides as
adherent cells would do. Same procedures should be applied
on nonadherent cells.
20. Depending on cell type, loading time is prolongable and load-
ing concentration of Fura-2-AM is variable between 1 and
10 μM. Keep Fura-2-AM loading solution and Fura-2 loaded
cells in dark area at room temperature.
21. Cells are storable in storage buffer for at least 12 h at room
temperature.
22. The cpEGFP of the sensor is usually brighter and therefore
better visible. Alternatively, the OFP can be excited at ~550 nm
with an emission at ~565 nm.
23. No interfering emission light from cytosolic Fura-2 should be
visible in this channel.
24. Minimum exposure times for the excitation of Fura-2 are
below 20 ms for both wavelengths. Basal ratio levels are around
0.4 and reach above 1 upon stimulation.
25. Minimum exposure time of mtD1GO-Cam is below 100 ms.
Basal ratio levels are around 0.8 and reach more than 1 upon
stimulation.
26. Usually mitochondrial Ca
2+
uptake is delayed to cytosolic Ca
2+
rises.
27. Some laboratories successfully use French press method
[29,30], which is more advantageous and ensures more gentle
mechanical isolation of mitoplasts [31].
28. Mitoplasts (mitochondria devoid of outer membrane) have
larger size than mitochondria. When isolated from HeLa
cells, mitoplasts are typically around 5 μm in diameter.
29. Even seeding is especially important in order to minimize well-
to-well variation. Ideally, cells should reach a confluency of 90%
by the time of the experiment.
30. Nonadherent cells can be immobilized using Corning
®
Cell-
Tak™cell and tissue adhesive (Corning, Product #354240)
according to the protocol supplied by Seahorse Bioscience
(10). Immobilization of cells might also become necessary
when measuring under Ca
2+
-free conditions as most cell adhe-
sion molecules are Ca
2+
-dependent and the mixing steps dur-
ing the assay do impose considerable strain on cell adhesion.
31. The XF96 sensor cartridge must be hydrated for 16 up to 48 h
prior to the start of the assay.
32. Four channels (designated as port A-D) located on the sensor
cartridge allow the addition of substrates/inhibitors to the
wells during the measurement in order to resolve different
functional states of respiration. Here we depict the use of
Measuring Mitochondrial Calcium 189
three reagents, oligomycin A (an inhibitor of the ATP
synthase), FCCP (a protonophore/uncoupling agent), and
antimycin A (an inhibitor of complex III) to evaluate basal,
coupled, maximal, and residual OCRs. The optimal concentra-
tions of these compounds have to be determined in titration
experiments preceding the actual assay. Typical working con-
centrations are 1–2 μM oligomycin A, 0.1–2.0 μM FCCP, and
2μM antimycin A. In the course of the assay, 25 μl of com-
pound solution per channel will be injected sequentially. Based
on a starting volume of 150 μl per well, 7,8, and 9stock
solutions of the final working concentration will be required
for wells A, B, and C, respectively.
33. To ensure complete injection, all 96 channels of one port must
be evenly loaded using a multichannel pipette. We do not
recommend the use of the supplied loading guide as this is
prone to cause air bubbles.
References
1. Jouaville LS, Pinton P, Bastianutto C, Rutter
GA, Rizzuto R (1999) Regulation of mito-
chondrial ATP synthesis by calcium: evidence
for a long-term metabolic priming. Proc Natl
Acad Sci U S A 96(24):13807–13812
2. Brookes PS, Yoon Y, Robotham JL, Anders
MW, Sheu SS (2004) Calcium, ATP, and
ROS: a mitochondrial love-hate triangle. Am
J Physiol Cell Physiol 287(4):C817–C833
3. Giorgi C, Romagnoli A, Pinton P, Rizzuto R
(2008) Ca2+ signaling, mitochondria and cell
death. Curr Mol Med 8(2):119–130
4. De Stefani D, Raffaello A, Teardo E, Szabo
`I,
Rizzuto R (2011) A forty-kilodalton protein of
the inner membrane is the mitochondrial cal-
cium uniporter. Nature 476(7360):336–340
5. Mallilankaraman K, Ca
´rdenas C, Doonan PJ,
Chandramoorthy HC, Irrinki KM, Golena
´rT
et al (2012) MCUR1 is an essential component
of mitochondrial Ca2+ uptake that regulates
cellular metabolism. Nat Cell Biol 14
(12):1336–1343
6. Baughman JM, Perocchi F, Girgis HS,
Plovanich M, Belcher-Timme CA, Sancak Y
et al (2011) Integrative genomics identifies
MCU as an essential component of the mito-
chondrial calcium uniporter. Nature 476
(7360):341–345
7. Perocchi F, Gohil VM, Girgis HS, Bao XR,
McCombs JE, Palmer AE et al (2010)
MICU1 encodes a mitochondrial EF hand pro-
tein required for Ca(2+) uptake. Nature 467
(7313):291–296
8. Jiang D, Zhao L, Clapham DE (2009)
Genome-wide RNAi screen identifies Letm1
as a mitochondrial Ca2+/H+ antiporter. Sci-
ence 326(5949):144–147
9. Zecchini E, Siviero R, Giorgi C, Rizzuto R,
Pinton P (2007) Mitochondrial calcium signal-
ling: message of life and death. Ital J Biochem
56(4):235–242
10. Jean-Quartier C, Bondarenko AI, Alam MR,
Trenker M, Waldeck-Weiermair M, Malli R
et al (2012) Studying mitochondrial Ca(2+)
uptake - a revisit. Mol Cell Endocrinol 353
(1–2):114–127
11. Sancak Y, Markhard AL, Kitami T, Kova
´cs-
Bogda
´n E, Kamer KJ, Udeshi ND et al
(2013) EMRE is an essential component of
the mitochondrial calcium uniporter complex.
Science 342(6164):1379–1382
12. Waldeck-Weiermair M, Deak AT, Groschner
LN, Alam MR, Jean-Quartier C, Malli R et al
(2013) Molecularly distinct routes of mito-
chondrial Ca2+ uptake are activated depending
on the activity of the sarco/endoplasmic retic-
ulum Ca2+ ATPase (SERCA). J Biol Chem
288(21):15367–15379
13. Bright GR, Fisher GW, Rogowska J, Taylor DL
(1989) Fluorescence ratio imaging microscopy.
Methods Cell Biol 30:157–192
14. Eberhard M, Erne P (1991) Calcium binding
to fluorescent calcium indicators: calcium
green, calcium orange and calcium crimson.
Biochem Biophys Res Commun 180
(1):209–215
190 Andra
´s T. Deak et al.
15. Rizzuto R, Simpson AW, Brini M, Pozzan T
(1992) Rapid changes of mitochondrial Ca2+
revealed by specifically targeted recombinant
aequorin. Nature 358(6384):325–327
16. Nagai T, Sawano A, Park ES, Miyawaki A
(2001) Circularly permuted green fluorescent
proteins engineered to sense Ca2+. Proc Natl
Acad Sci U S A 98(6):3197–3202
17. Nakai J, Ohkura M, Imoto K (2001) A high
signal-to-noise Ca(2+) probe composed of a
single green fluorescent protein. Nat Biotech-
nol 19(2):137–141
18. Zhao Y, Araki S, Wu J, Teramoto T, Chang YF,
Nakano M et al (2011) An expanded palette of
genetically encoded Ca
2+
indicators. Science
333(6051):1888–1891
19. Miyawaki A, Llopis J, Heim R, McCaffery JM,
Adams JA, Ikura M et al (1997) Fluorescent
indicators for Ca2+ based on green fluorescent
proteins and calmodulin. Nature 388
(6645):882–887
20. Palmer AE, Jin C, Reed JC, Tsien RY (2004)
Bcl-2-mediated alterations in endoplasmic
reticulum Ca2+ analyzed with an improved
genetically encoded fluorescent sensor. Proc
Natl Acad Sci U S A 101(50):17404–17409
21. Palmer AE, Giacomello M, Kortemme T, Hires
SA, Lev-Ram V, Baker D et al (2006) Ca2+
indicators based on computationally rede-
signed calmodulin-peptide pairs. Chem Biol
13(5):521–530
22. McCombs JE, Palmer AE (2008) Measuring
calcium dynamics in living cells with genetically
encodable calcium indicators. Methods 46
(3):152–159
23. Carlson HJ, Campbell RE (2009) Genetically
encoded FRET-based biosensors for multipa-
rameter fluorescence imaging. Curr Opin Bio-
technol 20(1):19–27
24. Waldeck-Weiermair M, Alam MR, Khan MJ,
Deak AT, Vishnu N, Karsten F et al (2012)
Spatiotemporal correlations between cytosolic
and mitochondrial Ca(2+) signals using a novel
red-shifted mitochondrial targeted Cameleon.
PLoS One 7(9):e45917
25. Bondarenko AI, Jean-Quartier C, Malli R, Gra-
ier WF (2013) Characterization of distinct
single-channel properties of Ca
2+
inward cur-
rents in mitochondria. Pflugers Arch 465
(7):997–1010
26. Wiederkehr A, Szanda G, Akhmedov D,
Mataki C, Heizmann CW, Schoonjans K et al
(2011) Mitochondrial matrix calcium is an acti-
vating signal for hormone secretion. Cell
Metab 13(5):601–611
27. Vishnu N, Jadoon Khan M, Karsten F, Grosch-
ner LN, Waldeck-Weiermair M, Rost R et al
(2014) ATP increases within the lumen of the
endoplasmic reticulum upon intracellular Ca2
+-release. Mol Biol Cell 25(3):368–379
28. Nakano M, Imamura H, Nagai T, Noji H
(2011) Ca
2+
regulation of mitochondrial ATP
synthesis visualized at the single cell level. ACS
Chem Biol 6(7):709–715
29. Fieni F, Lee SB, Jan YN, Kirichok Y (2012)
Activity of the mitochondrial calcium uniporter
varies greatly between tissues. Nat Commun
3:1317
30. Fedorenko A, Lishko PV, Kirichok Y (2012)
Mechanism of fatty-acid-dependent UCP1
uncoupling in brown fat mitochondria. Cell
151(2):400–413
31. Decker GL, Greenawalt JW (1977) Ultrastruc-
tural and biochemical studies of mitoplasts and
outer membranes derived from French-pressed
mitochondria. Advances in mitochondrial sub-
fractionation. J Ultrastruct Res 59(1):44–56
Measuring Mitochondrial Calcium 191
Chapter 14
Assessment of Mitochondrial Membrane Potential and
NADH Redox State in Acute Brain Slices
Andrey Y. Vinokurov, Viktor V. Dremin, Gennadii A. Piavchenko,
Olga A. Stelmashchuk, Plamena R. Angelova, and Andrey Y. Abramov
Abstract
Brain is one of the most energy-demanding organs. Energy in the form of ATP is produced in brain cells
predominantly in oxidative phosphorylation coupled to mitochondrial respiration. Any alteration of the
mitochondrial metabolism or prolonged ischemic or anoxic conditions can lead to serious neurological
conditions, including neurodegenerative disorders. Assessment of mitochondrial metabolism is important
for understanding physiological and pathological processes in the brain. Bioenergetics in central nervous
system is dependent on multiple parameters including neuron–glia interactions and considering this, in vivo
or ex vivo,the measurements of mitochondrial metabolism should also be complimenting the experiments
on isolated mitochondria or cell cultures. To assess the mitochondrial function, there are several key
bioenergetic parameters which indicate mitochondrial health. One of the major characteristics of mito-
chondria is the mitochondrial membrane potential (ΔΨm) which is used as a proton motive force for ATP
production and generated by activity of the electron transport chain. Major donor of electrons for the
mitochondrial respiratory chain is NADH. Here we demonstrate how to measure mitochondrial NADH/
NAD(P)H autofluorescence and ΔΨm in acute brain slices in a time-dependent manner and provide
information for the identification of NADH redox index, mitochondrial NADH pool, and the rate of
NADH production in the Krebs cycle. Additionally, non-mitochondrial NADH/NADPH autofluores-
cence can signify the level of activity of the pentose phosphate pathway.
Key words Mitochondria, NADH, Mitochondrial membrane potential, Acute brain slices
1 Introduction
Functioning of the brain is one of the most energy-dependent
processes in our organism. Maintenance of the resting membrane
potential and ion homeostasis in neurons is regulated by utilization
of ATP, which is directly dependent on the consumption of oxygen
[1]. Oxygen is transported to the cells of CNS as a complex with
Hb in the form of oxyhemoglobin (HbO
2
) and is passing through
the blood–brain barrier through the basement membranes of the
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_14,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
193
microcirculatory vessels (brain capillaries). Oxygen is then used by
the cells for the execution of the biochemical process of oxidative
phosphorylation that is coupled with the functioning of electron
transport chain (ETC) in mitochondria. The donors of electrons
for complex I and complex II from the electron transport chain are
NADH and FADH
2
(flavoprotein) which are produced in the
tricarboxylic acid cycle (TCA cycle or Krebs cycle). NADH and
the oxidized form of FADH
2
—FAD—are fluorescent, and this
helps to assess them in live cells and tissues [2–4]. However,
NADH fluorescence cannot be separated from NADPH and it is
also produced in the cytosol. Thus, to assess the autofluorescence of
NADH in mitochondria, uncouplers in combination with inhibi-
tors of the ETC should be used [5].
Electron transport chain of mitochondria generates the mito-
chondrial membrane potential (ΔΨm), which is used as a proton
motive force to produce ATP in the F0-F1-ATP synthase. Any
change in mitochondrial metabolism reflects over to the ΔΨm
and considering this, measurement of the mitochondrial membrane
potential can be considered as an indicator of mitochondrial
“health” [6].
Although mitochondrial function could be assessed in isolated
mitochondria or in live cells in cell cultures, possibilities need to be
developed for this measurement in in vivo or ex vivo models. This is
specifically important for the brain where bioenergetics is complex
and dependent on the multiple processes including neuron–glia
interaction [7,8].
Mitochondrial metabolism is vitally important in central nervous
system and any alteration in function of mitochondria leads to the
development of pathology, i.e., neurodegeneration [9–13] and other
neurological conditions, including ataxias and epilepsy [14–16].
Here, we describe our methods for measurement of mitochon-
drial membrane potential and mitochondrial NADH and FAD in
acute brain slices using fluorescent imaging. This method helps to
visualize and measure the dynamics of brain metabolic changes and
biochemical processes that occur in the living brain cells.
2 Materials
All manipulations must be carried out in accordance to the ethical
standards for humane treatment of animals. Follow all waste dis-
posal regulations when disposing waste materials.
2.1 Equipment 1. Large scissors or guillotine.
2. Vibratome.
3. Fluorescent imaging setup (see Note 1).
194 Andrey Y. Vinokurov et al.
2.2 Reagents for
NADH Determination
1. HBSS with 10 mM HEPES (two types: with pH 7.4/RT and
with pH 7.4/4 C) (see Notes 2 and 3).
2. 100 mM FCCP stock solution in DMSO (close the tube with
foil and store at 20 C) (see Note 4).
3. 100 μM FCCP working solution in HBSS (pH 7.4/RT).
4. 100 mM NaCN working solution in double-distilled water.
Use freshly prepared solution (see Note 5).
2.3 Reagents for
Mitochondrial
Membrane Potential
Determination
1. HBSS with 10 mM HEPES (two types: with pH 7.4/RT and
with pH 7.4/4 C).
2. 1 mM of rhodamine 123 stock solution in DMSO (should be
stored in a dark place (cover the tube with tin foil) and store at
20 C).
3. 100 mM FCCP stock solution in DMSO (cover the tube with
tin foil and store at 20 C).
4. 100 μM FCCP working solution in HBSS (pH 7.4/RT).
5. 1 μM solution of rhodamine 123 in HBSS (pH 7.4/RT).
3 Methods
NADH possesses intrinsic fluorescence with an absorption maxi-
mum at 360 nm and an emission maximum at 460 nm. The
fluorescence of the slices is determined by the physiological content
of NADH, all over the cell (in the cytosol and in cellular compart-
ments). When protonophore (FCCP) is added to the slice under
measurement, cell respiration reaches its maximum due to the
ability of mitochondria to correct for the electrochemical gradient
and, accordingly, for the mitochondrial membrane potential. At the
same time, the NADH fluorescence intensity reaches a minimum,
determined by the reduced form of coenzyme content exclusively
in the cytosol, while in mitochondria, complete oxidation of
NADH occurs with the formation of nonfluorescent NAD+
(Fig. 1). With the subsequent addition of cyanides, blocking the
complex IV of the respiratory chain, and the oxidation of NADH
(Fig. 1), the fluorescence level reaches the maximum level deter-
mined by the reduced form of coenzyme content in the cytosol and
in mitochondria.
NADH is an electron donor for complex I, and therefore
NADH levels are inversely correlated with respiratory chain activity.
To measure the redox index, FCCP (1 μM) is used to maximize
respiration and therefore minimize the NADH pool, then NaCN
(1 mM) is added to block mitochondrial respiration and thus
maximize the NADH pool. Initial autofluorescence is calculating
as a percentage of this range (Fig. 1). Besides, the total mitochon-
drial pool of NADH can take as an indicator of the presence of
substrate for complex I.
Assessment of Mitochondrial Function 195
As soon as the baseline autofluorescence levels are obtained
using data collection (2 min), the maximum possible signal stimu-
lating respiration is achieved (defined as the reaction to the addition
of 1 μM FCCP (the uncoupler)) and the maximum signal reduced.
At the same time, respiration is completely inhibited as a reaction to
the addition of 1 mM NaCN.
Fig. 1 Typical NADH change curves in healthy acute brain slices: calculation of
(a) NADH pool and the level of non-mitochondrial NADH/NAD(P)H and (b) NADH
redox index and NADH production rate
196 Andrey Y. Vinokurov et al.
Finally, “redox indices of NADH” are generated by expressing
the baseline levels of NADH, as a percentage of the difference
between the most oxidized and maximally reduced signals. The
NADH redox indices along the NADH pool provide information
on the activity of the electron transport chain, as well as any
disturbances in the function of the mitochondrial respiratory com-
plexes. Using this method, it is also possible to assess the levels of
NAD(P)H through an autofluorescence analysis of NADH. This
allows the activity of the pentose phosphate pathway (PPP) to be
evaluated, as a place where glucose can be alternatively oxidized.
The NADH pool is calculated by subtracting the lowest fluo-
rescence value (after addition of FCCP) from the highest (after
addition of NaCN) (Fig. 1a). The NAD(P)H value is calculated
by subtracting the background from the minimum fluorescence (see
Note 6). The NADH redox index is calculated after the recorded
trace is normalized from 0% to 100% (Fig. 1b). The minimum
NADH autofluorescence (after FCCP) is 0%, and the maximum
autofluorescence (NaCN) is 100%. The NADH redox index is
represented as the basal autofluorescence before any inhibitors are
added and expressed as a percentage from the 0% to 100% scale
(48% in Fig. 1b). The NADH growth rate (slope, Fig. 1b) is a direct
reflection of the effectiveness of the tricarboxylic acid cycle (TCA
cycle), as NaCN blocks all mitochondrial respiration. The NADH
growth rate is then calculated using linear fit to determine the slope
of the fluorescence spectrum after NaCN administration. An exam-
ple of the calculation is shown in Fig. 1b [1].
The method of mitochondrial membrane potential determina-
tion is based on the ability of a lipophilic fluorescent cation rhoda-
mine 123 to accumulate in mitochondria. The degree of
accumulation is determined by the value of the mitochondrial
membrane potential—the higher it is, the higher the ratio of the
concentration of the cation in the mitochondrial interior to its
concentration in the cytoplasm. Examined brain slices are kept in
a rhodamine 123 solution for a certain time and after that the
measurement of the fluorescence (excitation—505 nm, emis-
sion >540 nm) level is done in the following mode: first 3 min
(basic fluorescence level), FCCP introduction, and subsequent
measurement to the moment of fluorescence level stabilization. In
the experiment, an increase in the level of fluorescence should occur
because of the effect of concentration quenching after the
subsequent introduction of FCCP (Fig. 2), a decrease in the mito-
chondrial potential occurs, and the subsequent exit of the cation
into the cytoplasm. The degree of increase in fluorescence is an
indicator of the level of membrane mitochondrial potential and can
be used to compare different types of brain.
Assessment of Mitochondrial Function 197
3.1 Acute Brain
Slices Preparation
Procedure
1. Cervical dislocation of the animal.
2. Separation of the head from the trunk in the cervical region of
spinal cord by using large scissors or guillotine.
3. Dissection of the scalp along the sagittal line and opening of the
cranium (generally starting in the areas of the nasal, temporal,
and parietal bones).
4. Quick extraction of the brain with cerebellum, followed by
periodic washing by cold HBSS (pH 7.4/4 C). It is important
to avoid surface drying of the sample.
5. Performing a sagittal section of the brain in an ice-cold glass
petri dish.
6. Excision of the brain regions of interest with subsequent prep-
aration of slices with a thickness of 100–200 μm. Acute brain
slices were cut on a vibratome, according to a standard brain
tissue cutting procedure [17,18], in ice-cold HBSS (pH 7.4/
4C).
3.2 NADH
Determination
Procedure
1. Prepared slices were maintained in the well of a glass slide, in
150 μl HBSS (pH 7.4/RT) for a minimum of 1 h before the
measurement.
2. Then the level of NADH fluorescence is determined in a time-
dependent manner:
(a) Record the basic level: 0–180 s.
(b) Addition of 5 μl of FCCP.
Fig. 2 Measurement of rhodamine 123 fluorescence in acute brain slices with
complete depolarization of mitochondria upon application of FCCP
198 Andrey Y. Vinokurov et al.
(c) Record of fluorescence level: 181–360 s.
(d) Addition of 5 μl of NaCN.
(e) Record of fluorescence level: 361–540 s.
3.3 Mitochondrial
Membrane Potential
Determination
Procedure
1. One slice is transferred to a working solution of rhodamine 123
and incubated for 15 min.
2. The slice is washed using a large amount of HBSS (pH 7.4/
RT), placed in clean HBSS (pH 7.4/RT) and incubated for
5 min. Then the washing procedure is repeated two more
times.
3. The prepared slice is placed in the well of a glass slide, to which
150 μl of HBSS (pH ¼7.4/RT) is added.
4. Then the level of fluorescence is determined at the settings of
the laboratory setup closest to the fluorescence parameters of
rhodamine 123:
(a) Record of basic level: 0–180 s.
(b) Addition of 5 μl of FCCP working solution.
(c) Record of fluorescence level before reaching a plateau.
4 Notes
1. Measurements of ΔΨm and NADH level in brain slices can be
done using various systems including confocal microscopy but
can be dome in less sophisticated systems. Here, we suggest an
easy-to-implement system for measuring endogenous NADH
and NAD(P)H content.
The optical scheme is a standard scheme of reflected light
microscopy. In this scheme, the microscope objective works
both as a condenser and as an image-forming system. The key
element for such optical schemes is a vertical illuminator. The
main function of a vertical illuminator is to form a collimated
light beam, direct it to the rear aperture of the lens, and then to
the surface of the sample. The main component of a vertical
illuminator in fluorescent studies is a dichroic mirror. This
mirror deflects the exciting radiation coming from the hori-
zontal illuminator by 90to the vertical optical image-forming
system. In addition, the dichroic mirror, located at an angle of
45to the optical axes of the illuminator and the image-
forming channel, passes fluorescent radiation coming out of
the lens (see Fig. 3).
2. Acute brain slices are placed on glass coverslips just before the
experiment. While imaging, tissue should be buffered using
HEPES-buffered salt solution (HBSS medium).
Assessment of Mitochondrial Function 199
3. This buffer allows the imaging of cells or slices in a static
chamber avoiding the need for continuous CO
2
-equilibrated
buffering.
4. FCCP is a lipid-soluble weak acid used as a mitochondrial
uncoupling agent. FCCP is negatively charged allowing the
anions to diffuse freely through nonpolar media, such as phos-
pholipid membranes. It abolishes the obligatory linkage
between the respiratory chain and the oxidative phosphoryla-
tion system which occurs in intact mitochondria.
Fig. 3 Scheme of the experimental setup for measurement of endogenous
fluorescence in acute brain slices. Excitation radiation (9) from an optical fiber
passes through a collimator (8) and an extinction band filter (7) to cut out a
narrow excitation band, and further through a dichroic mirror (4) and a planar
apochromatic lens (5) is directed to the study area (6). In the image-forming
channel, the back-reflected radiation from the source is filtered by a dichroic
mirror (4) and an emission filter (3), and the fluorescent radiation through a long-
focus lens (2) is registered by a highly sensitive cooled CCD camera (1). The field
view of the system is a rectangular area of about 1 mm
2
with a resolution of
about 2 μm
200 Andrey Y. Vinokurov et al.
5. NaCN is a respiratory chain inhibitor and it blocks respiration
in the presence of either ADP or uncouplers such as FCCP. It
specifically blocks the cytochrome oxidase (complex IV) and
prevents both coupled and uncoupled respirations despite the
presence of substrates, including NADH, and succinate.
6. The registered NADH autofluorescence upon addition of
FCCP represents both the non-mitochondrial NADH auto-
fluorescence and the NAD(P)H autofluorescence. To ensure
that only NAD(P)H autofluorescence is recorded and ana-
lyzed, background fluorescence is deducted from the total
fluorescence output.
Acknowledgments
This study was supported by the Russian Federation Government
grant No. 075-15-2019-1877. VD kindly acknowledges the per-
sonal support from the European Union’s Horizon 2020 research
and innovation program under the Marie Sklodowska-Curie grant
agreement No. 839888.
References
1. Abramov AY, Angelova PR (2019) Mitochon-
drial dysfunction and energy deprivation in the
mechanism of neurodegeneration. Turkish J
Biochem 44(6):723–729
2. Chance B (1954) Spectrophotometry of intra-
cellular respiratory pigments. Science 120
(3124):767–775
3. Chance B, Schoener B, Oshino R et al (1979)
Oxidation-reduction ratio studies of mito-
chondria in freeze-trapped samples. NADH
and flavoprotein fluorescence signals. J Biol
Chem 254(11):4764–4771
4. Chance B, Thorell B (1959) Fluorescence mea-
surements of mitochondrial pyridine nucleo-
tide in aerobiosis and anaerobiosis. Nature
184:931–934
5. Bartolome F, Abramov AY (2015) Measure-
ment of mitochondrial NADH and FAD auto-
fluorescence in live cells. Methods Mol Biol
1264:263–270
6. Zamzami N, Marchetti P, Castedo M et al
(1995) Sequential reduction of mitochondrial
transmembrane potential and generation of
reactive oxygen species in early programmed
cell death. J Exp Med 182(2):367–377
7. Gasser UE, Hatten ME (1990) Neuron-glia
interactions of rat hippocampal cells in vitro:
glial-guided neuronal migration and neuronal
regulation of glial differentiation. J Neurosci
10(4):1276–1285
8. Angelova PR, Abramov AY (2014) Interaction
of neurons and astrocytes underlies the mecha-
nism of Abeta-induced neurotoxicity. Biochem
Soc Trans 42(5):1286–1290
9. Angelova PR, Barilani M, Lovejoy C et al
(2018) Mitochondrial dysfunction in parkinso-
nian mesenchymal stem cells impairs differenti-
ation. Redox Biol 14:474–484
10. Arber C, Angelova PR, Wiethoff S et al (2017)
iPSC-derived neuronal models of PANK2-
associated neurodegeneration reveal mito-
chondrial dysfunction contributing to early dis-
ease. PLoS One 12(9):e0184104
11. Kinghorn KJ, Castillo-Quan JI, Bartolome F
et al (2015) Loss of PLA2G6 leads to elevated
mitochondrial lipid peroxidation and mito-
chondrial dysfunction. Brain 138
(Pt 7):1801–1816
12. Ludtmann MHR, Angelova PR, Horrocks MH
et al (2018) α-Synuclein oligomers interact
with ATP synthase and open the permeability
transition pore in Parkinson’s disease. Nat
Commun 9(1):2293
13. Tufi R, Gandhi S, de Castro IP et al (2014)
Enhancing nucleotide metabolism protects
against mitochondrial dysfunction and
Assessment of Mitochondrial Function 201
neurodegeneration in a PINK1 model of Par-
kinson’s disease. Nat Cell Biol 16(2):157–166
14. Abeti R, Parkinson MH, Hargreaves IP et al
(2016) Mitochondrial energy imbalance and
lipid peroxidation cause cell death in Frie-
dreich’s ataxia. Cell Death Dis 7:e2237
15. Kovac S, Domijan AM, Walker MC et al (2012)
Prolonged seizure activity impairs mitochon-
drial bioenergetics and induces cell death.
J Cell Sci 125(Pt 7):1796–1806
16. Kovac S, Preza E, Houlden H et al (2019)
Impaired bioenergetics in mutant
mitochondrial DNA determines cell fate dur-
ing seizure-like activity. Mol Neurobiol 56
(1):321–334
17. Angelova P, Muller W (2006) Oxidative mod-
ulation of the transient potassium current IA by
intracellular arachidonic acid in rat CA1 pyra-
midal neurons. Eur J Neurosci 23
(9):2375–2384
18. Egorov AV, Angelova PR, Heinemann U et al
(2003) Ca
2+
independent muscarinic excita-
tion of rat medial entorhinal cortex layer V
neurons. Eur J Neurosci 18(12):3343–3351
202 Andrey Y. Vinokurov et al.
Chapter 15
Evaluation of Mitochondria Content and Function in Live
Cells by Multicolor Flow Cytometric Analysis
Hsiu-Han Fan, Tsung-Lin Tsai, Ivan L. Dzhagalov, and Chia-Lin Hsu
Abstract
To evaluate how a cell responds to the external stimuli, treatment, or alteration of the microenvironment,
the quantity and quality of mitochondria are commonly used as readouts. However, it is challenging to
apply mitochondrial analysis to the samples that are composed of mixed cell populations originating from
tissues or when multiple cell populations are of interest, using methods such as Western blot, electron
microscopy, or extracellular flux analysis.
Flow cytometry is a technique allowing the detection of individual cell status and its identity simulta-
neously when used in combination with surface markers. Here we describe how to combine mitochondria-
specific dyes or the dyes targeting the superoxide produced by mitochondria with surface marker staining to
measure the mitochondrial content and activity in live cells by flow cytometry. This method can be applied
to all types of cells in suspension and is particularly useful for analysis of samples composed of heteroge-
neous cell populations.
Key words Flow cytometry, FACS, Mitochondria, Quantification, Reactive oxygen species
1 Introduction
Mitochondria is a double-membrane-bound organelle that plays a
critical role in the generation of energy in most eukaryotic organ-
isms [1]. Recently a surging number of studies aim to uncover the
additional involvement of mitochondria in cellular functions, e.g.,
differentiation or effector actions. Assays such as Western blot,
PCR-based measurement of mitochondrial DNA (mtDNA)
copy numbers, electron and immunofluorescent microscopy, or
extracellular flux analysis [2–4] are commonly applied to quantify
the mitochondria amount or function. Western blot is the standard
technique to evaluate the expression and modification of the target
protein in the purified mitochondria. With the isolation of both
mitochondrial DNA (mtDNA) and nuclear DNA, the quantitative
PCR-based assay has been established to measure mtDNA copy
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_15,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
203
number, which is a critical component of overall mitochondrial
health. Electron and immunofluorescent microscopy are powerful
tools to visualize the mitochondria morphology and spatial distri-
bution in the cells. The newly developed extracellular flux analysis is
specialized to measure the oxygen consumption rate and quantify
mitochondrial respiration in living cells [5]. Take the study on
mitochondrial protein, dynamin-related protein 1 (Drp1), as an
example, by combining the Western blot and electron microscopy
techniques, it is observed that the nutrient deprivation induces the
dephosphorylation of Drp1 Serine 616 and 637 which subse-
quently leads to mitochondrial tabulation and elongation
[6]. Numerous recent studies use extracellular flux analysis to eval-
uate the mitochondria activity via measuring the oxygen consump-
tion rate—the rate of reserve capacity/spare respiratory capacity
(SRC) is an indicator of mitochondria status which meaning the
higher SRC it detects, the more ATP is supplied by mitochondria
oxidative phosphorylation [7].
Although these abovementioned methods have all been instru-
mental in understanding the mitochondria biology, they are limited
to analyze the homogenous cell population only. Compared to
these classical methods, flow cytometry is a sensitive and time-
efficient technique. In combination with surface marker staining,
this method allows the detection of individual cell status from the
heterogeneous cell population. Multiple mitochondria-specific and
mitochondrial superoxide-detection fluorescent probes have been
developed, enabling these probes to use in conjunction with surface
markers. The application of these mitochondria-specific probes in
flow cytometry overcomes the cell homogeneity requirement in
classical methods and shortens the processing time, and, most
importantly, provides broader applicability.
The measurement of mitochondria mass or membrane poten-
tial has been essential to determine the status of intracellular ion
homeostasis, energy metabolism, or as an indicator for cell stress/
survival in eukaryotic cells [8,9]. MitoTracker™Green FM has
been used as a measurement of mitochondrial mass—non-fluores-
cent in aqueous solutions; it accumulates in the lipid environment
of mitochondria and becomes fluorescent regardless of mitochon-
drial membrane potential [10]. It can be visualized under a micro-
scope or quantified by flow cytometric analysis in the intact cells
without the isolation of mitochondria. Based on its unique chemi-
cal properties, one can assume that the dye’s fluorescent intensity
correlates with the mitochondria mass in the cells.
Because mitochondria inner membrane is negatively charged,
mitochondria-specific dyes are often cationic lipophilic dyes, e.g.,
JC-1 dye [11], rhodamine 123 [12], and tetramethylrhodamine
[13], as well as thiol-reactive chloromethyl groups, including Mito-
Tracker™Red and MitoTracker™Orange [14]. They are posi-
tively charged compounds that can be transported passively across
204 Hsiu-Han Fan et al.
the mitochondria membranes and accumulated within the mito-
chondria. When conjugated with different fluorescent chemicals,
these membrane potential-dependent dyes are useful indicators of
the mitochondria activity. Mitochondria activity can also be
reflected by oxidative stress, one of the most important indexes of
the cell metabolic status. By measuring the level of reactive oxygen
species (ROS) produced [15], one can evaluate the active effector
response and the metabolic change of the cell [16]. Cellular ROS
level has multiple implications: high oxidative stress in the cell could
lead to the activation of oxidative stress responding pathway or cell
death [17], while it is also a sign of active inflammatory response in
the macrophages [15,18]. However, ROS are extremely versatile,
making it a difficult parameter to measure. Although chromatogra-
phy, mass spectrometry, or electrochemical sensors are capable of
sensitive and accurate ROS detection, these methods require
specialized equipment and the support of a well-established core
facility. Several assays [19] have recently been developed to detect
the intracellular cell ROS, including the fluorescence-dependent
and chemiluminescence-derived methods. These techniques rely
on the cell-permeable chemicals that react directly to the ROS
and generating radical intermediates, which give rise to fluores-
cence, e.g., dihydroethidium (DHE) stain has been used to detect
mitochondrial superoxide. Combining with flow cytometry, one
now can apply cell-permeable fluorescent probes like CellROX™
for intracellular reactive oxygen species or MitoSOX™for mito-
chondrial superoxide [20] to perform analysis of ROS on complex
cell populations.
The current protocol is a demonstration of how to combine
surface marker staining and mitochondria- or ROS-specific dyes to
measure the mitochondrial activities of the cells of interest within
the heterogeneous cell population. It is worth noting that different
staining procedures often are necessary to optimize the efficiency of
different organelle-specific dye detection. We applied this tech-
nique to examine the mitochondria mass, membrane potential,
total ROS, and mitochondrial superoxide in tissue macrophages
(Mϕs). Depending on its biological functions, each organ forms its
unique microenvironment. Even cells from the same lineage may
behave distinctively when residing in different microenvironments
or under stress. Immune cells, for example, maintain a versatile
metabolic program adapting to their habitats and to satisfy the
biosynthetic needs upon encountering antigens. By establishing
an assay that spontaneously detects the cell identity and mitochon-
drial activities, it allows us to compare the immune cell status of
different organs and, more importantly, the function and physio-
logical role of the immune cells in homeostatic or disease settings
[21]. We found that compared to splenic Mϕs (Fig. 1), the perito-
neal cavity Mϕs harbored more mitochondria (Fig. 2a) and had
higher mitochondrial activities (Fig. 2b). Moreover, the elevated
Measuring Mitochondria Content and Function by FACS 205
mitochondria activity in peritoneal cavity Mϕs was also reflected in
their ROS production capacity (Fig. 3). Together, these data sug-
gested that tissue MΦs can adjust their metabolic profile according
to the resideing microenvironment.
2 Material
2.1 Preparation
of the Single-Cell
Suspension
1. Dulbecco’s Modified Eagle Medium (DMEM).
2. Serum-free DMEM: DMEM, supplemented with 44 mM
NaHCO
3
, 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesul-
fonic acid (HEPES), 100 U/mL penicillin, 100 mg/mL strep-
tomycin, 2 mM L-glutamine, 1 mM sodium pyruvate, 1%
MEM nonessential amino acids.
3. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl,
8mMNa
2
HPO
4
, and 2 mM KH
2
PO
4
.
4. 5 mL syringe.
5. 24G needle.
6. 15 mL centrifuge tube.
7. 6 cm petri dish.
8. Ammonium–Chloride–Potassium (ACK) lysing buffer:
155 mM NH
4
Cl, 10 mM KHCO
3
, and 0.1 mM Na
2
EDTA.
9. Nylon mesh with pore size 75 μm.
Fig. 1 Gating strategy of peritoneal and splenic macrophages. Cells were pre-gated on FSC/SSC and PI
to
obtain live singlets. Total peritoneal cells (a) and splenocytes (b) from 5.5 weeks old mice were stained with
F4/80 and CD11b. Macrophages were defined as F4/80
+
CD11b
+
population. The results are representative of
three independent experiments
206 Hsiu-Han Fan et al.
2.2 Detection
of Mitochondrial Mass,
Activity, and ROS (See
Notes 1 and 2)
1. MitoTracker Green FM, stock solution 1 mM in DMSO.
2. MitoTracker Red FM, stock solution 1 mM in DMSO.
3. CellROX™Green (Thermo Fisher Scientific), stock solution
2.5 mM in DMSO.
4. MitoSOX™(Thermo Fisher Scientific), stock solution 5 mM
in DMSO.
5. Round-bottom FACS tube.
2.3 Surface Marker
Staining
1. 2.4G2 hybridoma (ATCC
®
HB-197™) supernatant.
2. FACS buffer (1PBS supplemented with 2% fetal bovine
serum (FBS) and 1 mM EDTA).
3. PE-Cy7 Anti-mouse F4/80 (BioLegend, Cat#123114).
4. BV421 Anti-mouse F4/80 (BioLegend, Cat#123137).
Fig. 2 Quantification of mitochondria mass and membrane potential in peritoneal and splenic macrophages.
Cells were stained with mitochondria-specific dye for 15 min at 37 C, followed by surface marker staining.
The fluorescent signal of stained cells was acquired by the flow cytometer and analyzed. (a) Mitochondria
mass of peritoneal and splenic macrophages was quantified by MitoTracker™Green staining of peritoneal
and splenic macrophages. (b) Mitochondria membrane potential of peritoneal and splenic macrophages was
evaluated by MitoTracker™Red staining. The green and red represent mitochondria-specific staining, and the
gray line shows fluorescence minus one (FMO). The results are representative of one experiment with n¼3.
The statistics of mitochondrial mass (c) and membrane potential (d) were performed by calculating the
ΔMFI ¼MFI (Mitochondria staining)—MFI (FMO)
Measuring Mitochondria Content and Function by FACS 207
5. APC Anti-mouse CD11b (BioLegend, Cat#101212).
6. PE-Cy7 Anti-mouse CD11b (BioLegend, Cat#101215).
7. Propidium iodide solution.
8. DAPI.
3 Methods
3.1 Tissue Harvest
and Generation
of the Single-Cell
Suspension
1. Euthanize the mouse by an approved method such as CO
2
asphyxiation in a transparent acrylic chamber. Rinse the target
area with 75% ethanol.
2. To harvest the peritoneal cells, intraperitoneally inject 5 mL
serum-free DMEM medium with a 24G needle. Gently
Fig. 3 Measurement of cellular and mitochondrial ROS in peritoneal and splenic macrophages. Cells were
stained with surface marker staining, followed by reactive oxygen species-specific dye for 15 min at 37 C.
The fluorescent signal of stained cells was acquired by the flow cytometer and analyzed. (a) The level of total
cellular ROS in peritoneal and splenic macrophage was evaluated by CellROX staining. (b) MitoSOX™staining
of peritoneal and splenic macrophages measures the mitochondrial ROS level. The blue and purple represent
mitochondria staining, and the gray one shows FMO. The results are representative of one experiment with
n¼3. The statistical analysis was done by calculating the ΔMFI ¼MFI (reactive oxygen species staining)—
MFI (FMO) for cellular (c) and mitochondrial (d) ROS
208 Hsiu-Han Fan et al.
massage the peritoneal cavity, and harvest the peritoneal cell-
containing lavage as much as possible with the syringe. Transfer
the lavage to a 15 mL centrifuge tube and leave on ice until
assay. This is the peritoneal single-cell suspension.
3. Open the peritoneal cavity and locate the spleen. The spleen is
at the left upper quadrant of the abdomen, carefully remove the
surrounding connective tissue and harvest the spleen.
4. Dissociate the spleen by pressing gently with a syringe plunger
in a 6 cm dish containing 5 mL ice-cold serum-free DMEM.
Transfer the single-cell suspension to a 15 mL centrifuge tube,
wash the 6 cm dish with an additional 2 mL serum-free
DMEM, and pool the cell suspension together. This is the
splenocyte single-cell suspension.
5. Centrifuge the cell suspension for 5 min at 450 g,4C and
discard the supernatant. Resuspend the cell pellet with 2 mL
ACK lysing buffer for 2 min at room temperature (RT) to lyse
the red blood cells. At the end of the reaction, add 13 mL of
PBS to the cell to neutralize the ACK lysing buffer.
6. Pellet the cells again by centrifuging for 5 min at 450 g,4C
and resuspend peritoneal cells in 1 mL, splenocytes in 3 mL
serum-free DMEM.
7. Filter cell suspension through nylon mesh to remove any
clumps to obtain the single-cell suspension. Enumerate the
cell number.
3.2 Quantification
of Mitochondria Mass
and Membrane
Potential
1. Freshly prepare the mitochondria-specific dye working solution
by mixing 0.1 μL MitoTracker stock solution to 1 mL ice-cold
serum-free DMEM (see Note 3).
2. Add 1 10
6
peritoneal cells or 2 10
6
splenocytes to round-
bottom FACS tubes, and pellet the cells by centrifuging for
5 min at 450 g,4C(see Note 4). Discard the supernatant.
3. Resuspend the cells in 100 μL of mitochondria-specific dye
working solution and incubate for 15 min in 5% CO
2
incubator
at 37 C(see Note 5).
4. At the end of the incubation time, add 1 mL ice-cold FACS
buffer to stop the reaction, and centrifuge for 5 min at 450 g,
4C. Discard the supernatant. Cells are now ready to proceed
with surface marker staining.
3.3 Surface Marker
Staining
1. Resuspend the cells in 100 μL of 2.4G2 hybridoma supernatant
and incubate on ice for 10 min to block Fc receptors. Wash the
cells by adding 1 mL of ice-cold FACS buffer, and centrifuge
for 5 min at 450 g,4C. Discard the supernatant.
Measuring Mitochondria Content and Function by FACS 209
2. Resuspend the cells with 100 μL of FACS buffer containing the
pre-titrated fluorescence-conjugated antibodies and incubate
the mixture on ice for 20 min. Avoid light exposure during
the incubation period.
3. Wash the cells by adding 1 mL of ice-cold FACS buffer, and
centrifuge for 5 min at 450 g. Discard the supernatant.
4. Resuspend the pellet in 350 μL of FACS buffer containing
1μg/mL propidium iodide (PI) and immediately analyze the
samples on the flow cytometer.
3.4 Reactive Oxygen
Species Detection
1. To measure cellular ROS, the samples should be stained with
surface markers first (Subheading 3.3,steps 1–3), followed by
reactive oxygen species detection procedures (see Note 6).
2. To make the CellROX working solution, add 0.4 μL CellROX
stock to 200 μL of warm serum-free DMEM. MitoSOX work-
ing solution is made of 0.2 μL MitoSOX stock in 200 μL warm
serum-free DMEM. Both working solutions should be freshly
prepared. Resuspend the cell pellet from Subheading 3.3 in
working solutions and incubate at 37 C for 30 min. For the
control (fluorescence minus one, FMO), add 200 μL warm
serum-free DMEM only. Avoid light exposure during the incu-
bation period.
3. Add 1 mL ice-cold FACS buffer to each sample and centrifuge
for 5 min at 450 g4C. Discard the supernatant.
4. Resuspend the pellet in 100 μL ice-cold FACS buffer contain-
ing 0.1 μL DAPI stock solution and immediately analyze the
samples on the flow cytometer and perform data analysis (see
Note 7).
3.5 Data Analysis 1. Once the data are collected, use the analytical software of
choice to create a dot plot and gate on the populations of
interest. FSC-A vs. SSC-A is used to identify cell populations,
followed by FSC-A vs. FSC-H for singlet gating, FSC-A vs. PI
to gate on live cells. Cell surface markers are used to mark the
population of interests. Here, for example, we used
F4/80 vs. CD11b for macrophages gating (Fig. 1).
2. To visualize the fluorescence of mitochondria-specific dye in
the population of interests, choose the “Histogram” function
(Figs. 2a, b and 3a, b). Calculate the mean fluorescence inten-
sity (MFI) for each population of interest.
3. To perform statistical analysis, generate delta MFI (ΔMFI) for
each cell population by calculating MFI (organelle-specific
dyes)—MFI (FMO) (Figs. 2c, d, and 3c, d).
210 Hsiu-Han Fan et al.
4 Note
1. These organelle-specific dyes are very sensitive to freeze–thaw
cycles as well as light and air exposure. Aliquot the stock
solution in a small volume to avoid repeated freeze–thaw and
light exposure and to minimize the dye from constant air
exposure.
2. All of these mitochondria-specific dyes have cytotoxicity at high
concentrations, a higher staining concentration than that used
in the current protocol is not recommended.
3. To maintain good cell viability throughout the procedures, it is
recommended to use a serum-free culture medium during the
organelle-specific dyes staining step.
4. We recommend careful titration and optimization of the stain-
ing procedures for the cell population of interests. Once the
conditions are set, fix the cell-to-dye ratio to maintain the
consistent staining intensity.
5. To efficiently label the mitochondria-specific dyes, the reaction
has to been performed at 37 C. We recommend proceeding
with surface marker staining at 4 C upon the completion of
the mitochondrial-dye staining to achieve the best staining
results.
6. Due to the sensitivity of ROS-specific dyes and the versatile
nature of ROS, perform the surface marker staining first, fol-
lowed by the ROS-specific dye staining. Upon the completion
of the ROS-specific dye staining procedure, analyze the sample
immediately.
7. Before harvesting the sample, filter cell suspension through a
cell strainer to avoid cell clumps.
Acknowledgments
We would like to thank Dr. Chin-Wen Wei for the initial set up for
this experimental system and Yu-Ting Hsieh for critically reading
the manuscript. This work was supported by grants from Ministry
of Science and Technology, Taiwan (MOST 107-2320-B-010-020,
MOST 108-2628-B-010-005 to C.-L. H.; 107-2320-B-010 -016
-MY3, 106-2320-B-010 -026 -MY3 to I. L. D.) and Cancer Pro-
gression Research Center, National Yang-Ming University from
The Featured Areas Research Center Program within the frame-
work of the Higher Education Sprout Project by the Ministry of
Education (MOE) in Taiwan.
Measuring Mitochondria Content and Function by FACS 211
References
1. Spinelli JB, Haigis MC (2018) The multiface-
ted contributions of mitochondria to cellular
metabolism. Nat Cell Biol 20(7):745–754.
https://doi.org/10.1038/s41556-018-0124-
1
2. Bass JJ, Wilkinson DJ, Rankin D, Phillips BE,
Szewczyk NJ, Smith K, Atherton PJ (2017) An
overview of technical considerations for West-
ern blotting applications to physiological
research. Scand J Med Sci Sports 27(1):4–25.
https://doi.org/10.1111/sms.12702
3. Mumcuoglu EU, Hassanpour R, Tasel SF,
Perkins G, Martone ME, Gurcan MN (2012)
Computerized detection and segmentation of
mitochondria on electron microscope images. J
Microsc 246(3):248–265. https://doi.org/
10.1111/j.1365-2818.2012.03614.x
4. Pelletier M, Billingham LK, Ramaswamy M,
Siegel RM (2014) Chapter Seven—Extracellu-
lar flux analysis to monitor glycolytic rates and
mitochondrial oxygen consumption. Methods
Enzymol 542:125–149. https://doi.org/10.
1016/B978-0-12-416618-9.00007-8
5. Plitzko B, Loesgen S (2018) Measurement of
oxygen consumption rate (OCR) and extracel-
lular acidification rate (ECAR) in culture cells
for assessment of the energy metabolism.
Bio-protocol 8(10):e2850. https://doi.org/
10.21769/BioProtoc.2850
6. Rambold AS, Kostelecky B, Elia N, Lippincott-
Schwartz J (2011) Tubular network formation
protects mitochondria from autophagosomal
degradation during nutrient starvation. Proc
Natl Acad Sci U S A 108(25):10190–10195.
https://doi.org/10.1073/pnas.1107402108
7. Divakaruni AS, Paradyse A, Ferrick DA, Mur-
phy AN, Jastroch M (2014) Chapter Sixteen -
Analysis and interpretation of microplate-based
oxygen consumption and pH data. Methods
Enzymol 547:309–354. https://doi.org/10.
1016/B978-0-12-801415-8.00016-3
8. Van Blerkom J (2011) Mitochondrial function
in the human oocyte and embryo and their role
in developmental competence. Mitochondrion
11(5):797–813. https://doi.org/10.1016/j.
mito.2010.09.012
9. Tait SWG, Green DR (2013) Mitochondrial
regulation of cell death. Cold Spring Harb Per-
spect Biol 5(9):a008706. https://doi.org/10.
1101/cshperspect.a008706
10. Agnello M, Morici G, Rinaldi AM (2008) A
method for measuring mitochondrial mass and
activity. Cytotechnology 56(3):145–149.
https://doi.org/10.1007/s10616-008-9143-2
11. Poot M, Zhang YZ, Kr€
amer JA, Wells KS,
Jones LJ, Hanzel DK, Lugade AG, Singer VL,
Haugland RP (1996) Analysis of mitochondrial
morphology and function with novel fixable
fluorescent stains. J Histochem Cytochem 44
(12):1363–1372. https://doi.org/10.1177/
44.12.8985128
12. Chen LB (1988) Fluorescent Labeling of
Mitochondria. Methods Cell Biol
29:103–123. https://doi.org/10.1016/
S0091-679X(08)60190-9
13. Heiskanen KM, Bhat MB, Wang H-W, Ma J,
Nieminen A-L (1999) Mitochondrial depolari-
zation accompanies cytochrome C release dur-
ing apoptosis in PC6 cells. J Biol Chem 274
(9):5654–5658. https://doi.org/10.1074/
jbc.274.9.5654
14. Wiederschain GY (2011) The molecular probes
handbook. A guide to fluorescent probes and
labeling technologies. Biochem Mosc 76
(11):1276–1276. https://doi.org/10.1134/
S0006297911110101
15. Chouchani ET, Pell VR, Gaude E,
Aksentijevic
´D, Sundier SY, Robb EL,
Logan A, Nadtochiy SM, Ord ENJ, Smith
AC, Eyassu F, Shirley R, Hu C-H, Dare AJ,
James AM, Rogatti S, Hartley RC, Eaton S,
Costa ASH, Brookes PS, Davidson SM,
Duchen MR, Saeb-Parsy K, Shattock MJ,
Robinson AJ, Work LM, Frezza C, Krieg T,
Murphy MP (2014) Ischaemic accumulation
of succinate controls reperfusion injury
through mitochondrial ROS. Nature 515
(7527):431–435. https://doi.org/10.1038/
nature13909
16. Chen X, Song M, Zhang B, Zhang Y (2016)
Reactive oxygen species regulate T cell immune
response in the tumor microenvironment. Oxi-
dative Med Cell Longev 2016:10. https://doi.
org/10.1155/2016/1580967
17. Volpe CMO, Villar-Delfino PH, dos Anjos
PMF, Nogueira-Machado JA (2018) Cellular
death, reactive oxygen species (ROS) and dia-
betic complications. Cell Death Dis 9(2):119.
https://doi.org/10.1038/s41419-017-0135-
z
18. Tannahill GM, Curtis AM, Adamik J, Palsson-
McDermott EM, McGettrick AF, Goel G,
Frezza C, Bernard NJ, Kelly B, Foley NH,
Zheng L, Gardet A, Tong Z, Jany SS, Corr
SC, Haneklaus M, Caffrey BE, Pierce K,
Walmsley S, Beasley FC, Cummins E, Nizet V,
Whyte M, Taylor CT, Lin H, Masters SL,
Gottlieb E, Kelly VP, Clish C, Auron PE,
212 Hsiu-Han Fan et al.
Xavier RJ, O’Neill LAJ (2013) Succinate is an
inflammatory signal that induces IL-1β
through HIF-1α. Nature 496
(7444):238–242. https://doi.org/10.1038/
nature11986
19. Zhang Y, Dai M, Yuan Z (2018) Methods for
the detection of reactive oxygen species. Anal
Methods 10(38):4625–4638. https://doi.
org/10.1039/C8AY01339J
20. Zhou R, Yazdi AS, Menu P, Tschopp J (2010)
A role for mitochondria in NLRP3
inflammasome activation. Nature 469:221.
https://doi.org/10.1038/nature09663
21. Cassina P, Cassina A, Pehar M, Castellanos R,
Gandelman M, de Leon A, Robinson KM,
Mason RP, Beckman JS, Barbeito L, Radi R
(2008) Mitochondrial dysfunction in
SOD1G93A-bearing astrocytes promotes
motor neuron degeneration: prevention by
mitochondrial-targeted antioxidants. J Neu-
rosci Off J Soc Neurosci 28(16):4115–4122.
https://doi.org/10.1523/JNEUROSCI.
5308-07.2008
Measuring Mitochondria Content and Function by FACS 213
Chapter 16
Analysis of Mitochondrial Dysfunction During Cell Death
Vladimir Gogvadze and Boris Zhivotovsky
Abstract
Mitochondria play a key role in various modes of cell death. Analysis of mitochondrial dysfunction and the
release of proteins from the intermembrane space of mitochondria represent essential tools in cell death
investigation. Here we describe how to evaluate release of intermembrane space proteins during apoptosis,
alterations in the mitochondrial membrane potential, and oxygen consumption in apoptotic cells.
Key words Mitochondria, Cell death, Respiration, Membrane potential, Reactive oxygen species
1 Introduction
Investigation of various forms of cell death has become an impor-
tant area of biomedical research. Recently, several cell death mod-
alities in addition to necrosis and apoptosis have been described and
characterized based on morphological and biochemical criteria
[1]. The interaction between different forms of cell death is com-
plicated and still a matter of debate. Mitochondria play a crucial role
in the execution of various modes of cell death, although the precise
mechanisms of their involvement are still unclear.
Currently, it is widely accepted that mitochondria are impor-
tant participants in the regulation of apoptosis, an evolutionarily
conserved and genetically regulated process of critical importance
for embryonic development and maintenance of tissue homeostasis
in the adult organism [2]. Apoptosis is also involved in the sponta-
neous elimination of potentially malignant cells and therapeutically
induced tumor regression, whereas defects in the apoptosis pro-
gram may contribute to tumor progression and resistance to treat-
ment [3]. The release of different proapoptotic proteins from the
mitochondrial intermembrane space has been observed during the
early stages of apoptotic cell death [4]. Among these proteins is a
component of the mitochondrial respiratory chain, cytochrome c.
Once in the cytosol, cytochrome cinteracts with its adaptor
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_16,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
215
molecule, Apaf-1, resulting in the recruitment, processing, and
activation of pro-caspase-9, a member of the caspase family of
cysteine proteases, in the presence of dATP or ATP [5]. Caspase-
9, in turn, cleaves and activates pro-caspase-3 and -7; these effector
caspases are responsible for the cleavage of various proteins leading
to biochemical and morphological features characteristic of
apoptosis [6].
The mechanisms regulating cytochrome crelease remain partly
obscure. However, two distinct models for cytochrome crelease
have emerged, and these can be distinguished based on whether
Ca
2+
is required or not for the event. In one instance, mitochon-
drial Ca
2+
overload causes opening of a nonspecific pore in the
inner mitochondrial membrane, with subsequent swelling and rup-
ture of the outer membrane followed by the release of cytochrome c
and other intermembrane space proteins [7]. The Ca
2+
-indepen-
dent model asserts that the release occurs without changes in
mitochondrial volume or dissipation of the mitochondrial mem-
brane potential. This mechanism involves specific pores in the outer
mitochondrial membrane that are formed by certain proapoptotic
members of the Bcl-2 family of proteins, such as Bax or Bak
[8]. Truncated Bid, generated by caspase-8 and other proteases,
induces a conformational change in Bax that allows this protein to
insert into the outer membrane, oligomerize, and mediate cyto-
chrome crelease. In addition, Bax can modulate cytochrome c
release by facilitation of opening of the permeability transition
pore [9].
Permeabilization of the outer membrane can be stimulated by
reactive oxygen species (ROS), including those produced by mito-
chondria. Thus, the assessment of ROS production, especially
superoxide radical, can provide the information regarding the
mechanisms of mitochondrial permeabilization.
The following protocols represent basic tools widely used in
estimating the release of intermembrane space proteins during
apoptosis, assessment of the mitochondrial membrane potential in
apoptotic cells, and analysis of mitochondrial oxygen consumption.
2 Materials
2.1 Assessment
of Cytochrome
cRelease from
the Mitochondria
of Apoptotic Cells
In order to analyze the release of certain proteins from mitochon-
dria during apoptosis, the cellular plasma membrane should be
disrupted and the cytosolic fraction separated from the membrane
material, including mitochondria. This can be achieved by incuba-
tion of cells in a solution containing digitonin, a steroid glycoside
from Digitalis purpurea, which selectively permeabilizes the plasma
membrane leaving the outer mitochondrial membrane intact.
Plasma membrane permeabilization occurs due to the interaction
of digitonin with cholesterol. The molecular weight of digitonin is
216 Vladimir Gogvadze and Boris Zhivotovsky
about three times that of cholesterol; thus, digitonin binding to
cholesterol permeabilizes the plasma membrane by disrupting the
packing of lipids. At digitonin concentrations between 10 and
100 μg/ml, cholesterol-rich plasma membranes are permeabilized,
whereas those of intracellular organelles are not [10].
1. Cells of interest in culture (e.g., Jurkat cells, U 937, HeLa).
2. Apoptotic stimuli (e.g., etoposide, cisplatin, staurosporine).
3. Phosphate-buffered saline (PBS).
4. 1 M Tris: Dissolve 12.1 g of Tris in 100 ml of distilled water;
adjust pH to 7.4 with HCl.
5. 1 M MgCl
2
6H
2
O: Dissolve 203.3 mg of MgCl
2
6H
2
Oin
1 ml of distilled water.
6. 0.5 M EGTA: Dissolve 38.1 g of EGTA in 80 ml of distilled
water, adjust pH to 7.4 using KOH, bring the solution to
100 ml, and store at 4 C.
7. 0.2% digitonin: Dissolve 10 mg of digitonin in 5 ml of distilled
water, aliquot, and store at 20 C.
8. 10 ml of fractionation buffer: Weigh 0.11 g of KCl, transfer
into 15 ml tubes, add 50 μl of 1 M Tris, 10 μl of 1 M MgCl
2
,
50 μl of 0.5 M EGTA, 0.5 ml of 0.2% digitonin solution, and
make up to 10 ml with distilled water. Store at 4 C for
4–5 days.
9. Low-speed centrifuge.
10. Eppendorf centrifuge.
11. Reagents and equipment for electrophoresis in polyacrylamide
gel and subsequent Western blotting.
2.2 Assessment
of the Mitochondrial
Membrane Potential
Alterations
in Apoptosis
Alteration of the mitochondrial membrane potential is one of the
first responses of cells to any insult. The mitochondrial membrane
potential, which drives oxidative phosphorylation [11] and mito-
chondrial calcium uptake [12], is generated by the electron trans-
porting chain. When electron transport ceases, for example, during
ischemia, the inner membrane potential can be built up at the
expense of ATP, hydrolyzed by the mitochondrial ATP synthase.
The relationship between mitochondrial depolarization and
apoptosis remains controversial. Depending on cell death stimuli,
some investigators consider a decrease in the mitochondrial mem-
brane potential, an early irreversible signal for apoptosis [13], while
others describe it as a late event [14].
1. Cells of interest (e.g., Jurkat cells, U 937, HeLa).
2. RPMI-1640 medium supplemented with 5% (v/v) heat-
inactivated fetal bovine serum, 2 mM L-glutamine, penicillin
(100 U/ml), and streptomycin (100 μg/ml).
Mitochondrial Alterations in Apoptosis 217
3. 25 mM TMRE stock solution: Dissolve 12.8 mg of tetra-
methylrhodamine methyl ester (TMRE; Molecular Probes) in
1 ml ethanol; store according to the manufacturer’s
instructions.
4. HEPES buffer: 10 mM HEPES, 150 mM NaCl, 5 mM KCl,
1 mM MgCl
2
6H
2
O, adjust pH to 7.4 with NaOH; store up
to 2–3 days at 4 C.
5. 10 mM carbonyl cyanide 3-chlorophenylhydrazone (CCCP):
Dissolve 4.1 mg of CCCP in 2 ml of ethanol, and store at
20 C. Dilute the stock solution to 1 mM by adding 50 μlof
10 mM CCCP to 450 μl of ethanol.
6. Flow cytometer (e.g., FACS; Becton Dickinson).
2.3 Assessment
of Oxygen
Consumption in Intact
Apoptotic Cells
In many instances, apoptosis-inducing agents can directly affect
mitochondria; thus, assessment of vital functions of mitochondria,
such as respiration, provides important information concerning
involvement of mitochondria in cell death process. Analysis of
oxygen consumption can be performed using intact cells as well as
cells with digitonin-permeabilized plasma membrane. In the latter
case, the mitochondria can be supplied with
1. Cells of interest (Jurkat, U 937, HeLa).
2. Medium, in which cells were growing.
3. 10 mM CCCP (see above).
4. Oxygraph (Hansatech Instruments), or any other Clark-type
oxygen electrode connected to computer or chart recorder.
5. Hamilton-type syringes (10 and 25 μl).
2.4 Assessment
of Mitochondrial
Respiration
in Apoptotic Cells
with Digitonin-
Permeabilized Plasma
Membrane
1. Cells of interest.
2. 1 M Tris: Dissolve 12.1 g of Tris in 100 ml of distilled water;
adjust pH to 7.4 with HCl.
3. 1 M MgCl
2
6H
2
O: Dissolve 203.3 mg of MgCl
2
6H
2
Oin
1 ml of distilled water.
4. 0.5 M KH
2
PO
4
: Dissolve 340.2 mg of KH
2
PO
4
in 5 ml of
distilled water, adjust pH to 7.4, aliquot, and keep frozen.
5. 0.2% digitonin: Dissolve 10 mg of digitonin in 5 ml of distilled
water, aliquot, and store at 20 C.
6. 10 ml of respiration buffer: Weigh 0.11 g of KCl, transfer into
15 ml tube, add 50 μl of 1 M Tris, 100 μl of 0.5 M KH
2
PO
4
,
10 μl of 1 M MgCl
2
, 0.5 ml of 0.2% digitonin, and make up to
10 ml with distilled water. Store at 4 C for 4–5 days.
7. Sodium succinate (0.5 M): Dissolve 135 mg of sodium succi-
nate in 1 ml of distilled water, aliquot, and store frozen at
20 C up to 3 months.
218 Vladimir Gogvadze and Boris Zhivotovsky
8. Sodium pyruvate (0.5 M): Dissolve 135 mg of sodium pyruvate
in 1 ml of distilled water, aliquot, and store frozen at 20 Cup
to 3 months.
9. Malate (0.5 M): Dissolve 134.1 mg of malic acid in 5 ml of
distilled water, adjust pH to 7.4 with KOH, aliquot, and store
frozen at 20 C up to 3 months.
10. CCCP (10 mM): Dissolve 4.1 mg CCCP in 2 ml of ethanol,
and store at 20 C. Dilute the stock solution to 1 mM by
adding 50 μl of 10 mM CCCP in 450 μl of ethanol.
11. Rotenone (2.5 mM): Dissolve 1 mg rotenone in 1 ml ethanol,
store at 20 C up to 3 months.
12. Malonate (0.5 M): Dissolve 260.15 mg of malonic acid in 5 ml
of distilled water, adjust pH with KOH, aliquot, and store
frozen at 20 C up to 6 months.
13. Ascorbate (0.5 M): Dissolve 440 mg of ascorbic acid in 5 ml of
distilled water, adjust pH to 7.4 with KOH, aliquot, and store
at 20 C up to 3 months.
14. Tetramethyl phenylenediamine (TMPD) (0.03 mM): Dissolve
4.9 mg of TMPD in 1 ml of ethanol, store at 20 Cupto
3 months.
15. Oxygraph (Hansatech Instruments), or any other Clark-type
electrode connected to computer or chart recorder.
16. Hamilton-type syringes (10 and 25 μl).
2.5 Assessment
of Mitochondrial
Production
of Superoxide Radical
1. Cells of interest in culture (e.g., Jurkat cells, U 937, HeLa).
2. Apoptotic stimuli (e.g., etoposide, cisplatin, staurosporine).
3. Phosphate-buffered saline (PBS).
4. Dissolve the contents of one vial of MitoSOX™(50 μg) in
13 μl of dimethyl sulfoxide (DMSO) to make a 5 mM Mito-
SOX™reagent stock solution, or in 26 μl of DMSO to make a
2.5 mM MitoSOX™solution.
5. Keep MitoSOX™Red reagent desiccated and protected from
light at –20 C.
MitoSOX™Red mitochondrial superoxide indicator has exci-
tation/emission maxima of approximately 510/580 nm.
3 Methods
3.1 Evaluation
of Cytochrome
cRelease from
the Mitochondria
of Apoptotic Cells
1. Incubate cells with and without apoptotic stimuli (type, con-
centration, and incubation time determined by cell type).
2. Harvest cells using trypsin, transfer into 15 ml tubes, count
cells, and spin down at 200 gfor 5 min.
3. Gently remove the supernatant, without touching the pellet.
Mitochondrial Alterations in Apoptosis 219
4. Resuspend cells at concentration of 1 10
6
cells in 100 μlof
the fractionation buffer and transfer samples into Eppendorf
tubes.
5. Take aliquots (2–3 μl) for protein determination.
6. Incubate samples at room temperature for 10–15 min.
7. In the end of the incubation, vortex cells briefly and spin
samples down using Eppendorf centrifuge for 5 min at
10,000 g.
8. Gently, without touching the pellet, transfer supernatants
(approx. 95 μl) from each sample into new Eppendorf tubes
(see Note 1).
9. Add 95 μl of the fractionation buffer in each tube and resus-
pend the pellets.
10. Mix supernatant and pellet fractions with Laemmli buffer.
Detection of proteins of interest is performed using polyacryl-
amide gel electrophoresis with subsequent blotting and probing
with specific antibodies.
A typical distribution of cytochrome cbetween mitochondria
and cytosol in apoptotic cells is shown in Fig. 1.
3.2 Assessment
of the Mitochondrial
Membrane Potential
1. Prepare an aliquot of 0.3 10
6
cells in 300 μl of RPMI-1640
medium.
2. Dilute TMRE stock solution 1:1.000 with HEPES buffer (for a
concentration of 25 μM).
3. Add an aliquot of the diluted (25 μM) TMRE to cells for a final
concentration of 25 nM.
4. Incubate cells with TMRE 20 min at 37 C.
5. Further dilute the 25 μM TMRE 1:1.000 with HEPES buffer
for a final concentration of 25 nM. Centrifuge cells for 5 min at
200 gat room temperature, and resuspend in fresh HEPES
buffer containing 25 nM TMRE.
Cyt ccytosol
Cyt cmitochondria
GAPDH
Fig. 1 Release of cytochrome cfrom mitochondria in neuroblastoma Tet21N
cells during apoptosis induced by alpha-tocopheryl succinate (α-TOS). GAPDH
was used as loading control
220 Vladimir Gogvadze and Boris Zhivotovsky
6. Analyze membrane potential by flow cytometry according to
the manufacturer’s instructions for the instrument used (see
Note 2).
7. Use protonophore CCCP (5 μM) to dissipate the mitochon-
drial membrane potential completely as a positive control.
3.3 Assessment
of Oxygen
Consumption in Intact
Apoptotic Cells
1. Calibrate the oxygen electrode according to the manufacturer
protocol.
2. Add 0.3 ml of the medium to the oxygen electrode chamber
under conditions of constant stirring. Set the rate of stirring
10–15 rpm.
3. Harvest cells (two to four million per measurement, depending
on the cell type) and spin them down.
4. Remove the supernatant and start the program. Take 50–60 μl
of the medium from the oxygen electrode chamber for resus-
pending the cell pellet, transfer medium with cells back to the
chamber, and close the chamber with the plunger. The plunger
has a stoppered central precision bore allowing additions to be
made to the reaction mixture using a standard Hamilton-type
syringe. Expel all air bubbles through the bore in the plunger
(slight twisting of the plunger helps to gather the bubbles).
The level of oxygen in the chamber will start decreasing as
mitochondria consume oxygen.
5. After 3–4 min, add 10 μM CCCP through the bore in the
plunger using Hamilton-type syringe.
6. After adding CCCP, the rate of respiration will increase as
CCCP lowers the mitochondrial potential that stimulates oxy-
gen consumption.
7. Express the rate of respiration as the amount of oxygen con-
sumed by one million of cells in 1 min.
The protocol allows measuring basal respiration, and respira-
tion stimulated by CCCP (highest activity of the respiratory chain),
which is not suppressed by the mitochondrial membrane potential
or controlled by the activity of ATP synthase.
3.4 Assessment
of Mitochondrial
Respiration
in Apoptotic Cells
with Digitonin-
Permeabilized Plasma
Membrane
Permeabilization of the plasma membrane allows measurement of
mitochondrial activity in situ, without isolation of these organelles
and the accompanying potential risk of mitochondrial damage.
Selective disruption of the plasma membrane by digitonin
makes mitochondria accessible to substrates for various respiratory
complexes (see Note 3). The concentration of digitonin must be
chosen carefully and usually should not exceed 0.01% (w/v), since
higher concentrations might affect the outer mitochondrial mem-
brane (see Note 4).
Mitochondrial Alterations in Apoptosis 221
1. Calibrate the oxygen electrode according to the manufacturer
protocol.
2. Add 0.3 ml of the respiration buffer to the oxygen electrode
chamber under conditions of constant stirring. Set the rate of
stirring 20–25 rpm.
3. Harvest cells (two to four million per measurement, depending
on the type of the cells), count them, and spin down in 15 ml
tubes.
4. Remove the supernatant and start the program. Take 50–60 μl
of the respiration buffer from the oxygen electrode chamber
for resuspending the cell pellet, transfer the buffer with cells
back to the chamber, and close the chamber with the plunger.
Expel all air bubbles through the bore in the plunger (slight
twisting of the electrode helps to gather the bubbles at the
slot). The level of oxygen in the chamber will start decreasing as
mitochondria consume oxygen.
5. Analysis of the activity of the respiratory complexes should be
performed after uncoupling oxidation and phosphorylation by
adding the protonophore CCCP (10 μM final concentrations)
to get maximum rates of respiration. After 2–3 min, add succi-
nate, a substrate of Complex II (10 mM final concentration);
addition of substrate will stimulate respiration (see Note 5).
After 3–4 min, add malonate (10 mM final concentration). The
respiration will slow down as succinate dehydrogenase is
inhibited.
6. After 3–4 min, add pyruvate and malate, substrates for Com-
plex I (10 mM final concentration each) (see Note 6). Respira-
tion will be stimulated. After 3–4 min, add rotenone, an
inhibitor of Complex I (2.5 μM final concentration). This will
slow down oxygen consumption.
7. After 3–4 min, add 5 mM ascorbate and 0.3 mM (TMPD), an
artificial electron donor for cytochrome c. Respiration will be
stimulated again.
This approach allows the assessment of the Complexes I, II,
and IV of the mitochondrial respiratory chain. A typical curve of
mitochondrial oxygen consumption by permeabilized cells is
shown in Fig. 2.
3.5 Assessment
of Mitochondrial
Superoxide Radical
Production
1. Incubate cells with and without apoptotic stimuli (type, con-
centration, and incubation time determined by the cell type).
2. Harvest cells using trypsin, transfer into 15 ml tubes, count
cells, and spin down at 200 gfor 5 min.
3. Gently remove the supernatant, without touching the pellet.
4. Resuspend cells at concentration 1 10
6
cells/ml and aliquot
in Eppendorf tubes (0.5 ml of suspension per tube).
222 Vladimir Gogvadze and Boris Zhivotovsky
5. Add 1 μl of MitoSOX™Red to each tube; for the background
fluorescence, add 1 μl DMSO (see Note 7).
6. Incubate samples at 37 C for approx. 20 min. This incubation
time can be increased to 40 min if necessary.
7. Add 0.5 ml medium, centrifuge the samples, and discard the
supernatant.
8. Add 0.5 ml of prewarmed medium to the pellets and resuspend
the cells.
9. Transfer the cell suspension to the flow cytometer measure-
ment tube, mix, and record the fluorescence according to the
manufacturer’s instructions (Fig. 3).
4 Notes
1. Separation of the mitochondria from the supernatant should be
done thoroughly. Aliquots of supernatant should be taken
without disturbing the pellet.
2. The analysis of the membrane potential should be done shortly
after staining of the cells. TMRE is only accumulated by mito-
chondria with high membrane potential, and any delay in
analysis may negatively affect the functional state of the mito-
chondria and therefore cause leakage of the dye.
250 cells
CCCP
succinate
malonatepyruvate+
malate
TMPD
asc
rotenone
200
150
O2 (nmol/ml)
100
50
0
0 200 400 600 800 1000
Time (sec)
1200 1400 1600
Fig. 2 Assessment of mitochondrial respiration in digitonin-permeabilized cells
Mitochondrial Alterations in Apoptosis 223
3. The concentration of digitonin can be determined experimen-
tally by staining permeabilized cells with Trypan blue, a vital
stain used to selectively color cells with damaged plasma mem-
brane. The lowest concentration causing permeabilization of
90–95% cells should be used in the experiment.
4. Digitonin may precipitate after cooling; shake it vigorously
before adding to the buffer.
5. Experiment can be started with analysis of Complex I activity
instead of Complex II.
6. Pyruvate and malate can be mixed before the experiment and
added together.
7. The working concentration of MitoSOX™can vary depending
on the type of cells.
Acknowledgments
The work was supported by the Russian Science Foundation (grant
19-14-00122). The work in the authors’ laboratory is supported by
Swedish and the Stockholm Cancer Societies.
References
1. Galluzzi L, Vitale I, Aaronson SA et al (2018)
Molecular mechanisms of cell death: recom-
mendations of the nomenclature committee
on cell death 2018. Cell Death Differ 25
(3):486–541
2. Kerr JF, Wyllie AH, Currie AR (1972) Apopto-
sis: a basic biological phenomenon with wide-
ranging implications in tissue kinetics. Br J
Cancer 26(4):239–257
3. Lowe SW, Lin AW (2000) Apoptosis in cancer.
Carcinogenesis 21(3):485–495
4. Gogvadze V, Orrenius S, Zhivotovsky B
(2006) Multiple pathways of cytochrome
crelease from mitochondria in apoptosis. Bio-
chim Biophys Acta 1757(5–6):639–647
5. Zou H, Li Y, Liu X, Wang X (1999) An APAF-
1.Cytochrome cmultimeric complex is a
control etoposide
4,8% 17,2%
102
2
0 0.5 1 1.5
Count %
22.5
103102
27
0 0.5 1 1.5
Count %
22.5
103
Fig. 3 Superoxide radical production in control and etoposide-treated
neuroblastoma Tet21N cells
224 Vladimir Gogvadze and Boris Zhivotovsky
functional apoptosome that activates
procaspase-9. J Biol Chem 274
(17):11549–11556
6. Robertson JD, Orrenius S, Zhivotovsky B
(2000) Review: nuclear events in apoptosis. J
Struct Biol 129(2–3):346–358
7. Crompton M (1999) The mitochondrial per-
meability transition pore and its role in cell
death. Biochem J 341(Pt 2):233–249
8. Martinou JC, Green DR (2001) Breaking the
mitochondrial barrier. Nat Rev Mol Cell Biol 2
(1):63–67
9. Gogvadze V, Robertson JD, Zhivotovsky B,
Orrenius S (2001) Cytochrome crelease occurs
via Ca
2+
-dependent and Ca
2+
- independent
mechanisms that are regulated by Bax. J Biol
Chem 276(22):19066–19071
10. Mooney RA (1988) Use of digitonin permea-
bilized adipocytes for cAMP studies. Methods
Enzymol 159:193–202
11. Mitchell P, Moyle J (1967) Chemiosmotic
hypothesis of oxidative phosphorylation.
Nature 213(5072):137–139
12. Carafoli E, Crompton M (1978) The regula-
tion of intracellular calcium by mitochondria.
Ann N Y Acad Sci 307:269–284
13. Zamzami N, Susin SA, Marchetti P, Hirsch T,
Gomez-Monterrey I, Castedo M, Kroemer G
(1996) Mitochondrial control of nuclear apo-
ptosis. J Exp Med 183(4):1533–1544
14. Bossy-Wetzel E, Newmeyer DD, Green DR
(1998) Mitochondrial cytochrome crelease in
apoptosis occurs upstream of DEVD-specific
caspase activation and independently of mito-
chondrial transmembrane depolarization.
EMBO J 17(1):37–49
Mitochondrial Alterations in Apoptosis 225
Chapter 17
Modified Blue Native Gel Approach for Analysis
of Respiratory Supercomplexes
Sergiy M. Nadtochiy, Megan Ngai, and Paul S. Brookes
Abstract
In mitochondrial oxidative phosphorylation (Ox-Phos), individual electron transport chain complexes are
thought to assemble into supramolecular entities termed supercomplexes (SCs). The technique of blue
native (BN) gel electrophoresis has emerged as the method of choice for analyzing SCs. However, the
process of sample extraction for BN gel analysis is somewhat tedious and introduces the possibility for
experimental artifacts. Here we outline a streamlined method that eliminates a centrifugation step and
provides a more representative sampling of a population of mitochondria on the final gel. Using this
method, we show that SC composition does not appear to change dynamically with altered mitochondrial
function.
Key words Mitochondria, Supercomplexes, Blue-native, Clear-native, Permeability transition pore,
Respiration
1 Introduction
In mitochondrial oxidative phosphorylation (Ox-Phos), individual
electron transport chain complexes are thought to assemble into
supramolecular entities termed supercomplexes (SCs) [1]. The tech-
niques of blue native (BN) and clear native (CN) gel electrophore-
sis have emerged as methods of choice for analyzing SCs and have
provided evidence that SC assembly may be altered under condi-
tions such as ROS generation [2] and swelling associated with
opening of the mitochondrial permeability transition (PT) pore
[3,4]. However, the functional role of SCs has been questioned
[5], and some technical challenges associated with their quantita-
tive assessment have also been highlighted [6]. As such, it remains
unclear whether SCs are regulated on a dynamic time scale of
seconds to minutes, with changing mitochondrial function.
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_17,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
227
Current native gel methods for SC analysis necessitate replace-
ment of experimental incubation media with gel-compatible media,
often accomplished via centrifugation of mitochondria [1]. How-
ever, many perturbations that are applied to mitochondria (e.g., PT
pore opening) are known to alter mitochondrial density [7], thus
raising the possibility that differences seen on BN gels may arise
from centrifugal selection of mitochondrial subpopulations. There-
fore, SCs visualized on BN gels may not represent the full mito-
chondrial population present in the original incubations.
To address this potentially important experimental artifact, our
objective herein was to develop an improved BN gel method, using
a compromise buffer system compatible with both mitochondrial
function and sample preparation for BN gels. By eliminating a
centrifugation step, the proposed new method ensures the entire
mitochondrial population is sampled and also realizes significant
time savings.
To validate the method, we exposed mitochondria to condi-
tions of Ca
2+
overload that induce PT pore opening and large
amplitude swelling. Example results (Fig. 1) indicate that the
method yields good resolution of SCs, with similar BN gel band
patterns as those observed by existing methods. Based on our
observations using this new method, the levels of SCs do not
appear to change under conditions of PT pore opening. This
observation contrasts with the idea that SC assembly state is
dynamic in response to (and perhaps even regulates) mitochondrial
function [5,6]. Our results also suggest that some previously
reported differences in SC assembly should be reassessed, to ensure
they do not arise from differential sampling of dense vs. buoyant
mitochondria by existing extraction methods.
2 Materials (See Note 1)
The preparation of BN gels is addressed extensively in the accom-
panying article from Buetner and Porter in this volume [8]. For
convenience, recipes for conventional BN gel and buffer systems
are shown below, but this section will mostly focus on core differ-
ences between existing BN gel extraction buffers and those devel-
oped for this new method.
2.1 General
Solutions for Running
BN Gels
1. BN gel resolving buffer: 75 mM imidazole, 1.5 M aminoca-
proic acid. Dissolve 0.555 g imidazole plus 9.839 g aminoca-
proic acid in 50 ml water, then adjust pH to 7.0 at 4 C. Store
refrigerated. This solution is a 3buffer, so when preparing
gels, it should contribute one-third of the final volume along-
side other gel components (acrylamide, etc.).
228 Sergiy M. Nadtochiy et al.
2. BN gel loading buffer: 50 mM aminocaproic acid, 5% w/v
Coomassie blue. Dissolve 0.066 g aminocaproic acid in 10 ml
water, then pH to 7.0 at 4 C. Add 0.5 g Coomassie Brilliant
Blue-G and mix to form a suspension, then recheck pH. Store
refrigerated.
3. BN gel anode buffer: 25 mM imidazole. Dissolve 1.702 g
imidazole in 1 L water, then adjust pH to 7.0 at 4 C. Store
refrigerated.
4. BN gel cathode buffer: 7.5 mM imidazole, 50 mM tricine.
Dissolve 0.511 g imidazole plus 8.958 g tricine in 1 L water,
then adjust pH to 7.0 at 4 C. Store refrigerated.
5. Digitonin: Prepare a solution of 20% (w/v) by dissolving 0.2 g
digitonin in 1 ml water (see Note 2).
Fig. 1 Blue Native Supercomplexes and Mitochondrial PT Pore Opening. Isolated heart mitochondria [9] were
incubated at 37 C, in conditions to induce PT pore opening and swelling, with succinate (5 mM) and rotenone
(5 μM) present. Incubations took place either in respiration buffer at 0.5 mg/ml (Subheading 2.2, item 2)orin
compromise buffer at 2 mg/ml (Subheading 2.2, item 3). Pore opening was initiated by addition of 100 μM
CaCl
2
to the incubation, and experiments were stopped after 10 min. (a) Blue native gel from control and Ca
2+
treated mitochondria, processed according to existing BN gel methods involving centrifugation (Table 1left
column). Left image shows BN gel immediately following electrophoresis (note: blue color comes from the BN
gel itself, not post-gel staining with Coomassie Blue). Right panel shows results of an in-gel Cx-V assay
(Subheading 3.3), with white precipitate bands corresponding to active Cx-V in various assembly states. (b)
Blue native gel from control and Ca
2+
treated mitochondria, processed according to the novel BN gel method
(Table 1right column). Gel and Cx-V in-gel assay are as in panel b
Modified Blue-Native Gel Extraction 229
2.2 Compromise
Buffer
(CB) for Incubations
and BN Gel Extraction
The recipe for existing BN gel extraction buffer is given below. In
addition, a recipe for typical mitochondrial respiration buffer is
shown. Finally, a recipe for our new “compromise buffer” (hereaf-
ter referred to as CB) is given (see Note 3).
1. Existing BN gel extraction buffer: 50 mM NaCl, 40 mM imid-
azole, 2 mM aminocaproic acid, 1 mM EDTA. Dissolve
0.146 g NaCl, 0.170 g imidazole, 0.013 g aminocaproic acid,
and 0.019 g EDTA in 50 ml water. Then adjust pH to 7.0 at
4C. Store refrigerated.
2. Mitochondrial incubation buffer: 120 mM KCl, 10 mM
HEPES, 25 mM sucrose, 1 mM EGTA, 5 mM KH
2
PO
4
,
5 mM MgCl
2
. Dissolve 4.47 g KCl, 1.19 g HEPES, 4.28 g
sucrose, 0.19 g EGTA, 0.034 g KH
2
PO
4
, and 0.508 g MgCl
2
in 50 ml water, then adjust pH to 7.3 at 37 C. Store frozen in
5 ml aliquots.
3. Compromise incubation/extraction buffer (CB): 80 mM KCl,
2 mM HEPES, 40 mM imidazole, 10 mM sucrose, 5 mM
aminocaproic acid, 1 mM EGTA, 2 mM KH
2
PO
4
,2mM
MgCl
2
. Dissolve 2.98 g KCl, 0.238 g HEPES, 0.17 g imidaz-
ole, 1.712 g sucrose, 0.0325 g aminocaproic acid, 0.19 g
EGTA, 0.014 g KH
2
PO
4
, and 0.203 g MgCl
2
in 50 ml
water, then adjust pH to 7.2 at 37 C. Store frozen in 5 ml
aliquots.
Table 1
Existing and modified sample extraction protocols for BN electrophoresis
Typical native gel method Modified method
1 0.5 mg mito in 1 ml incubation (0.5 mg/ml) 0.5 mg mito in 250 μl CB (2 mg/ml)
#
2 Centrifuge 14,000 g, 10 min to recover mito pellet,
discard s/n
#
#
3 Resuspend pellet in 25 μl extraction buffer, add 10 μl
20% digitonin (5.7% final), mix, rest on ice 20 min
Add 10 μl 20% digitonin (0.8% final), mix,
rest on ice 5 min (see Notes 7 and 8)
##
3 Centrifuge 14,000 g, 5 min to remove insoluble
material
Centrifuge 14,000 g, 5 min to remove
insoluble material
##
5 Add 4 μl loading buffer, mix Add 30 μl loading buffer, mix
##
6 Load 10 μl sample/well (128 μg), run gel Load 40 μl sample/well (69 μg), run gel
(see Note 9)
230 Sergiy M. Nadtochiy et al.
2.3 Solutions
for In-Gel Complex
V Assay
For the assignment of identity to SCs on gels, an in-gel activity
assay can be performed for one of the respiratory complexes. For
the assay of Cx-V, a stock solution (Cx-V buffer A) is prepared as
below. During the assay, a separate aliquot of this buffer is modified
by addition of Mg
2+
, ATP, and lead nitrate, to yield Cx-V buffer B.
1. Cx-V buffer A: 35 mM Tris, 270 mM glycine. Dissolve 0.424 g
Tris plus 2.027 g glycine in 100 ml water, then adjust pH to
8.3 at 25 C. Store refrigerated.
2. Cx-V buffer B: 35 mM Tris, 270 mM glycine, 135 mM
MgSO
4
, 6.5 mM Pb(NO
3
)
2
, 7.8 mM ATP. To a 14 ml aliquot
of buffer A, add 0.023 g MgSO
4
, 0.03 g Pb(NO
3
)
2
, and 0.06 g
ATP (disodium salt). Mix and then recheck pH to 8.3 at 25 C.
Do not store (see Notes 4 and 5).
3 Methods
Typical starting material for BN gel analysis of SCs is isolated
mitochondria. The isolation of mitochondria from a variety of
tissues by differential centrifugation is covered extensively else-
where in this volume and series. In brief, for validation of this
method, we isolated mouse cardiac mitochondria, as previously
described [9](see Note 6).
3.1 Mitochondrial
Incubation and Sample
Extraction for BN Gel
Analysis
Table 1shows a comparison of the existing BN gel extraction and
preparation methods and our streamlined method using compro-
mise buffer (CB). Critical differences between the methods are as
follows: First, instead of incubating mitochondria for experiments
in one media, then pelleting and extracting in a different media, a
single media (compromise buffer, CB) is used for the whole proce-
dure. Second, instead of stopping experimental incubations by
centrifuging mitochondria, incubations are stopped by direct addi-
tion of detergent (digitonin) to the experiment. Third, due to the
lack of a centrifugation step to concentrate mitochondria, it is
necessary to perform mitochondrial incubations at higher than
normal concentration (2 mg/ml protein vs. typical 0.5 mg/ml).
Lastly, the overall volumes for extraction are larger, so the final
extract is more dilute. As such, it is necessary to load a greater
volume onto the BN gel (40 μl vs. the usual 10 μl). Nevertheless,
the new method still achieves protein loading in amounts that
enable successful visualization of SCs on the final BN gels.
3.2 BN Gel
Electrophoresis
The preparation and running of BN gels is addressed extensively in
the accompanying article from Beutner and Porter in this volume
[8]. For convenience, recipes for BN gel and buffer systems are
given in Subheading 2.1 above. Typical BN gel methods use a 3–8%
polyacrylamide gradient gel, prepared from a 37.5/1 acrylamide/
Modified Blue-Native Gel Extraction 231
bisacrylamide stock solution. Gels should be run at 40 V, 4 C (e.g.,
in a cold room) for 18 h, using standard mini-gel apparatus (e.g.,
Bio-Rad mini-Protean 3).
3.3 Cx-V
In-Gel Assay
1. While gel is running, prepare Cx-V assay solution A, as in
Subheading 2.3.
2. Remove gel from glass plates and place in a clean plastic box on
a shaker platform for 2 h. in 14 ml buffer A (see Note 10).
3. After 1.5 h, prepare Cx-V assay buffer B, using a separate
aliquot of buffer A (see Notes 3 and 4).
4. Transfer the gel to buffer B, on the shaker platform, then
observe the formation of white precipitate bands for up to
2 h. The reaction can be stopped by washing the gel extensively
in water (see Notes 10 and 11).
5. If desired, the gel can be fixed in a solution of 50% methanol
prior to imaging or storage.
4 Notes
1. All chemicals should be of the highest grade available. Good
quality deionized water (18 MΩ) should be used for all
solutions.
2. Digitonin solutions must be freshly prepared immediately prior
to use, from fresh digitonin powder (stored at minus 20 C).
Aged powders or stock solutions will yield vastly inferior sepa-
ration on final BN gels.
3. An important control during development of this method was
to determine that the compromise buffer (CB) would indeed
support mitochondrial function. Experiments to measure res-
piration using a Clark-type oxygen electrode (not shown) indi-
cated that for mitochondria at 2 mg protein/ml, respiring on
succinate as a respiratory substrate, a respiratory control ratio
(state 3/state 2) of 5.0, was typically obtained.
4. The MgSO
4
, Pb(NO
3
)
2
, and ATP should all be weighed from
fresh powders and added to the 14 ml aliquot of buffer “A”
immediately before use. The pH of the solution should be
rechecked, and corrected using NaOH if necessary, before
use. If solution B sits unused for >30 min, discard and
remake it.
5. Lead nitrate is toxic. Appropriate personal protective equip-
ment should be worn. Leftover solutions should be disposed
via appropriate hazardous waste channels.
232 Sergiy M. Nadtochiy et al.
6. Mice were of the C57BL/6J strain, both sexes, and were
maintained in an AAALAC accredited facility according to the
NIH Guide for the Care and Use of Laboratory Animals (2011
update), with food and water available ad libitum and all pro-
cedures approved by local committee.
7. Following cessation of the experiment and establishment of an
extraction (step 3 in Table 1), samples should be thoroughly
mixed, but not too aggressively. We find the use of vortex mixer
results in fewer SCs visible in the final gels, and as such we
recommend to mix the incubation tubes gently by inversion
once or twice during the 5 min. Rest on ice. We also find that
5 min. on ice is sufficient for optimal extraction, and the exist-
ing 20 min. Incubation is unnecessarily long.
8. Although the final concentration of digitonin is lower in the
new method (0.8% vs. 5.7%), the ratio of digitonin to protein is
similar, and this level of detergent still results in successful
resolution of SCs (see results).
9. Comparison of gel loading volumes shows that the existing BN
method loads ~128 μg protein per well in a 10 μl volume,
whereas the new method loads about half the protein in four
times the volume. Nevertheless, 40 μl is well within the volume
range that can be accommodated by common gel apparatus
(e.g., 10-well comb, 1.5-mm thick Bio-Rad mini-gel).
10. Plastic staining boxes used for gel handling, Western blotting,
etc., are often contaminated with residues that can interfere
with development of bands in the Cx-V in-gel assay. For this
reason, we recommend to retain a dedicated box or set of boxes
for exclusive use for this assay, not to be used for other
purposes.
11. If desired, a separate gel can be used to perform the Cx-V assay
in the presence of the Cx-V inhibitor oligomycin (1 μg/ml final
concentration). However, since oligomycin is hydrophobic,
care should be taken to not contaminate plastic containers, so
that residual oligomycin does not interfere with future assays.
Acknowledgments
Work in the lab of PSB is funded by a grant from the US National
Institutes of Health (R01-HL071158). We thank George Porter
and Gisela Beutner for critical discussions during the execution of
these studies.
Modified Blue-Native Gel Extraction 233
References
1. Sch€
agger H, Pfeiffer K (2000) Supercomplexes
in the respiratory chains of yeast and mammalian
mitochondria. EMBO J 19:1777–1783
2. Maranzana E, Barbero G, Falasca A, Lenaz G,
Genova M (2013) Mitochondrial respiratory
supercomplex association limits production of
reactive oxygen species from complex
I. Antioxid Redox Signal 19:1469–1480
3. Beutner G, Alanzalon R, Porter G (2017)
Cyclophilin D regulates the dynamic assembly
of mitochondrial ATP synthase into syntha-
somes. Sci Rep 7:14488
4. Alavian K, Beutner G, Lazrove E, Sacchetti S,
Park H, Licznerski P, Li H, Nabili P,
Hockensmith K, Graham M, Porter G, Jonas E
(2014) An uncoupling channel within the
c-subunit ring of the F1FO ATP synthase is the
mitochondrial permeability transition pore. Proc
Natl Acad Sci U S A 111:10580–10585
5. Fedor JG, Hirst J (2018) Mitochondrial Super-
complexes do not enhance catalysis by quinone
channeling. Cell Metab 28:525–531
6. Jang S, Javadov S (2018) Current challenges in
elucidating respiratory supercomplexes in mito-
chondria: methodological obstacles. Front
Physiol 9:238
7. Hackenbrock CR (1966) Ultrastructural bases
for metabolically linked mechanical activity in
mitochondria. I. Reversible ultrastructural
changes with change in metabolic steady state
in isolated liver mitochondria. J Cell Biol
30:269–297
8. Buetner G, Porter G (2020) Native gel electro-
phoresis and Immunoblotting to analyze elec-
tron transport chain complexes
9. Smith CO, Wang YT, Nadtochiy SM, Miller JH,
Jonas EA, Dirksen RT, Nehrke K, Brookes PS
(2018) Cardiac metabolic effects of K
Na
1.2
channel deletion and evidence for its mitochon-
drial localization. FASEB J 32:6135–6149
234 Sergiy M. Nadtochiy et al.
Chapter 18
Patch-Clamp Recording of the Activity of Ion Channels
in the Inner Mitochondrial Membrane
Piotr Bednarczyk, Rafał P. Kampa, Shur Gałecka, Aleksandra Se˛k,
Agnieszka Walewska, and Piotr Koprowski
Abstract
Mitochondria are intracellular organelles, which play a crucial role in the generation of ATP. Mitochondria
are surrounded by a double membrane, consisting of a smooth outer membrane (OMM) and a markedly
folded inner mitochondrial membrane (IMM). Mitochondrion that has been stripped of its outer mem-
brane, leaving the inner membrane intact is called mitoplast. There is a number of different transport
proteins located in the inner mitochondrial membrane including ion channels that mediate fluxes of
potassium, calcium, and chloride ions. These channels regulate the mitochondrial membrane potential,
respiration, and production of reactive oxygen species. The stability of mitoplasts offers the possibility of
measuring the activity of ion channels from IMM using the patch-clamp technique. Electrophysiological
measurements of currents through ion channels in the IMM permit discovery of unique properties of these
channels with the aim of new specific pharmacological therapies. In this chapter, we describe the isolation of
mitochondria, preparation of mitoplast for patch-clamp recordings and single-mitoplast PCR experiments,
which can be helpful in mastering the technique of recording the activity of mitochondrial ion channels.
Key words Mitochondria, Mitoplast, Patch-clamp technique, Inner mitochondrial membrane, Ion
channel, PCR
1 Introduction
Electrophysiological measurements are one of the biophysical
methods used to explore the electrical activity of cells and investi-
gate related molecular and cellular processes [1]. The patch-clamp
technique historically originated from the work of Alan Hodgkin
and Andrew Huxley, describing the recording of macroscopic cur-
rents by voltage-clamp experiments [2]. It allowed the determina-
tion of the physiological significance of ion flow through
membrane channels and laid the foundation for the development
of electrophysiological techniques. The work of Hodgkin and Hux-
ley was awarded the Nobel Prize in 1963. Next, this technique was
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_18,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
235
improved by Neher and Sakmann [3] for which they also received
the Nobel Prize in Physiology or Medicine in 1991 [4].
In their technique, which we call nowadays “patch-clamp“one
pipette sets up voltage and records ionic currents flowing through
single channels residing in small membrane patch. The patch-clamp
technique still remains one of the top methods for measuring
single-molecule dynamics. Thanks to this technique, it is possible
to record from small membrane objects to which other techniques
are not adapted. It was used to discover new types of ion channels
and their importance for the proper functioning of the cells
[4]. Continuous improvement of the technique allowed for more
accurate monitoring of the electrical function of cellular mem-
branes [5]. By this technique, the activity of ion channels residing
in the plasma membrane and those found in intracellular structures
of cells of many types of tissues may be recorded. Different types of
ion channels that can be studied by the patch-clamp technique are
in fact present in multiple membrane compartments, e.g., large-
conductance calcium-activated potassium (BK
Ca
) channel is present
in the plasma membrane [6] but also membranes of mitochondria
[7,8], lysosomes [9,10], and nucleus [11]. At present, there are
reports of successful use of the patch-clamp technique in studies of
ion channels present in mitochondria from endothelial cells [8],
cardiomyocytes [12], astrocytoma [13], skin fibroblasts [14], ker-
atinocytes [15], and even from the Drosophila melanogaster [16].
The use of the patch-clamp technique has significantly contrib-
uted to the development of mitochondrial science. It allowed the
discovery of the presence of ion channels in mitochondria and
comparison with their counterparts in the cell membrane, like in
the case of BK
Ca
[17]. Channel activities discovered in mitochon-
dria with the help of the patch-clamp technique include in addition
to mitoBK
Ca
, for instance, these of mitochondrial ATP-inhibited
potassium channel (mitoK
ATP
)[14,18], mitochondrial megachan-
nel also known as permeability transition pore (PTP) [19,20],
mito-TASK3 [21], and others like voltage-gated potassium chan-
nels [22,23] or intermediate-conductance Ca
2+
-regulated potas-
sium channel [24]. The patch-clamp technique is also uniquely
suited to study the mechanosensitivity of channels including
mitoBK
Ca
[25]. Activation of mitochondrial potassium channels is
associated with cytoprotection, and for this reason, they constitute
important pharmacological targets [26,27]. Due to the direct
nature of measurements of single-channel activity using the patch-
clamp method, one can readily observe changes in the current flow
after the application of potential activators or inhibitors of the
tested channel. Thanks to this type of research, new ion channel
activators could be discovered or cytoprotective activity of known
substances based on the activation of mitochondrial potassium
channels explained, e.g., flavonoids of natural origin—
naringenin [28].
236 Piotr Bednarczyk et al.
This protocol describes the methodology, which is used for the
patch-clamp experiments on mitoplasts along with practical tips
and descriptions of the implementation of its individual stages.
We described the methods for (1) the isolation of fresh mitochon-
dria from cell cultures, (2) formation of mitoplast, (3) patch-clamp
recordings on mitoplasts, (4) single-mitoplast PCR (Fig. 1). This
protocol is ultimately optimized for recording mitoBK
Ca
channels
from human cell lines. For recording other types of channels,
protocol has to be optimized—solutions might be modified to
include the appropriate concentration of required modulators,
e.g., higher concentration of Ca
2+
ions to record PTP activity [19].
Mitochondria for patch-clamp recordings could be isolated not
only from cell cultures but also from animal [23] or even plant
tissues [29].
Fig. 1 Schematic representation of the preparation of mitochondria, mitoplasts, and the patch-clamp
experiments on the inner mitochondrial membrane. Mitoplasts are obtained by osmotic shock resulting in
rupture of the outer mitochondrial membrane. Mitoplast patch-clamp experiments are carried out in the
inside-out mode with the perfusion system. The matrix side of a mitoplast is exposed to externally added
substances, and the ion channel activity from the inner mitochondrial membrane is recorded
Patch-Clamp Technique for the Inner Mitochondrial Membrane 237
The mitoplast could be formed either directly in the recording
chamber or preformed and snap frozen in liquid nitrogen. How-
ever, it should be noted that the latter approach could not be
feasible for mitochondria from all cell types or tissues, e.g., it was
noticed that mitoplasts from mice heart were very fragile and only
method in which mitoplasts were formed directly in the recording
chamber was used in this case [12].
Two types of substance administration techniques for testing
the pharmacological properties of channels can be employed. One
of them is perfusion with the use of an external perfusion pump
feeding the test substances from the second perfusion pipette, with
a larger diameter than that used as a recording pipette (the record-
ing pipette is inserted into the perfusion pipette tip from which the
substance is administered). In the second approach, substances are
directly added to the recording chamber.
Succesful patch-clamp on mitoplast depends on the correct
identification of the very object. For a trained eye, a mitoplast is
easily recognized by the round shape and characteristic outer mem-
brane cap under a phase-contrast microscope. More objective
methods to identify mitoplast could be used during training, like
fluorescent staining with potential sensitive dyes, which accumulate
in mitochondria [30]. However, this requires the presence of a
fluorescent microscope in the patch-clamp rig, and in addition,
hydrophobic dyes could impact the activity of a channel. Therefore,
at the end of our protocol, we described single-mitoplast PCR, a
technique that can be employed with a standard patch-clamp setup
to correlate the visual appearance of the object with the content of
mitochondrial DNA (mtDNA) [31]. Altogether, the protocol can
be helpful especially for the novice electrophysiologist in mastering
the technique of recording the activity of mitochondrial ion
channels.
2 Materials
All chemicals need to be of the highest purity available. Ultrapure
water is systematically used (resistance >18 MΩcm at 25 C).
2.1 Stock Solutions All solutions should be kept in the fridge or frozen when indicated.
1. 1.5 M KCl: Weigh 55.9 g of potassium chloride (KCl). Add
water to 500 mL and mix.
2. 0.1 M CaCl
2
: Weigh 5.5 g of calcium chloride hexahydrate
(CaCl
2
·H
2
O). Add water to 250 mL and mix.
3. 0.5 M HEPES, pH ¼7.2: Weigh 11.9 g of 4-(2-hydroxyethyl)-
1-piperazineethanesulfonic acid (HEPES). Add water to a vol-
ume of 90 mL and mix. Adjust pH to 7.2 with potassium
hydroxide (KOH). Finally, add water to a volume of 100 mL.
238 Piotr Bednarczyk et al.
4. 0.1 M EGTA: Weigh 3.8 g of ethylene glycol-bis(β-aminoethyl
ether)-N,N,N0,N0-tetraacetic acid (EGTA). Add water to a
volume of 200 mL. Adjust pH to 7.2 with potassium hydroxide
(KOH) while mixing, and finally add water to a volume of
250 mL.
2.2 Mitochondria
Isolation
1. Preparation solution (250 mM sucrose, 5 mM HEPES,
pH ¼7.2): Weigh 21.4 g sucrose, add 2.5 mL of 0.5 M
HEPES stock solution. Add water to a volume of 200 mL.
Mix and adjust pH to 7.2 with potassium hydroxide (KOH).
Add water to a volume of 250 mL, mix, and aliquot in 15 mL
Falcon tubes. Store at 20 C.
2. Storage solution (150 mM KCl, 10 mM HEPES, pH ¼7.2):
Mix 10 mL of 1.5 M KCl stock solution and 2 mL of 0.5 M
HEPES stock solution. Add water to a volume of 90 mL and
mix. Adjust pH to 7.2 with potassium hydroxide (KOH), and
add water to a volume of 100 mL.
3. Glass potter homogenizer No. 19, 1 mL capacity.
4. Refrigerated benchtop centrifuge with rotors for 15 mL Falcon
and Eppendorf tubes.
2.3 Mitoplast
Preparation
1. Hypotonic solution (5 mM HEPES, 100 μM CaCl
2
,
pH ¼7.2): Mix 2.5 mL of 0.5 M HEPES stock solution and
250 μL of 0.1 M CaCl
2
stock solution. Add water to a volume
of 200 mL. Mix, adjust pH to 7.2 with potassium hydroxide
(KOH), and add water to a volume of 250 mL.
2. Hypertonic solution (750 mM KCl, 30 mM HEPES, 100 μM
CaCl
2
,pH¼7.2): Mix 125 mL of 1.5 M KCl stock solution
with 7.5 mL of 0.5 M HEPES stock solution and 250 μLof
0.1 M CaCl
2
stock solution. Add water to a volume of 200 mL.
Mix, adjust pH to 7.2 with potassium hydroxide (KOH), and
add water to a volume of 250 mL.
3. Sucrose solution (1.5 M sucrose, 30 mM HEPES, 100 μM
CaCl
2
,pH¼7.2): Weigh 12.84 g of sucrose, add 750 μLof
0.5 M HEPES stock solution and 25 μL of 0.1 M CaCl
2
stock
solution. Add water to a volume of 25 mL and mix. Aliquot in
Eppendorf tubes and keep frozen at 20 C.
2.4 Patch-Clamp
Recordings
1. Isotonic solution (150 mM KCl, 10 mM HEPES, and 100 μM
CaCl
2
,pH¼7.2): Mix 25 mL of 1.5 M KCl stock solution
with 5 mL of 0.5 M HEPES stock solution and 250 μLof
0.1 M CaCl
2
stock solution. Add water to a volume of 200 mL.
Mix and adjust pH to 7.2 with potassium hydroxide (KOH),
and add water to a volume of 250 mL.
Patch-Clamp Technique for the Inner Mitochondrial Membrane 239
2. Low-calcium solution, free 1 μM CaCl
2
(150 mM KCl, 10 mM
HEPES, 1 mM EGTA and 0.752 mM CaCl
2
,pH ¼7.2): Mix
25 mL of 1.5 M KCl stock solution with 5 mL of 0.5 M
HEPES stock solution, 2.5 mL of 0.1 M EGTA stock solution,
and 1.88 mL of 0.1 M CaCl
2
stock solution. Add water to a
volume of 200 mL. Mix and adjust pH to 7.2 with potassium
hydroxide (KOH), and add water to a volume of 250 mL.
3. 0.22 μm syringe filters.
4. Patch-clamp borosilicate glass capillaries, 1.5 mm O.
D. 0.86 mm I.D., e.g., Harvard Apparatus GC150-10.
5. Micropipette puller, e.g., Narishige PC-10, Sutter Instruments
P-1000.
6. Patch-clamp setup (Fig. 2) equipped with a phase-contrast
microscope with 300–400magnification, patch-clamp ampli-
fier, perfusion system (consisting of a holder with a multibarrel
glass tube, a peristaltic pump, and Teflon tubing).
Fig. 2 Photographs of the patch-clamp setup. (a) (1) multibarrel pipette of four-channel perfusion system with
tubings driven by a peristaltic pump (also shown in b); (2) a patch-clamp glass pipette with the recording AgCl
electrode inside (also shown in b). The pipette is connected to a water-filled U-tube with a three-way valve for
the application of the positive and negative pressure (not shown). In the back, a reference bath electrode with
a 3 M KCl salt-agar bridge can be seen as well. (b) (3) An in-house made anti-vibration table; (4) joystick
controllers of micromanipulators for positioning the multibarrel pipette of the perfusion system (left) and
patch-clamp pipette (right); (5) inverted microscope; (6) macromanipulator for multiple-channel perfusion
system with tubings; (7) macromanipulator with amplifier headstage and patch-clamp pipette holder; (8) four-
channel peristaltic pump; (9) oscilloscope; (10) digital-to-analog signal converter (digitizer); (11) amplifier;
(12) a monitor of PC for recording and analysis of raw data. Objects 1–8 are enclosed in Faraday’s cage
240 Piotr Bednarczyk et al.
2.5 Single-
Mitoplast PCR
1. Materials needed for mitochondria isolation (see Subheading
2.2).
2. Deoxyribonuclease I (DNase I).
3. Materials needed for mitoplasts preparation (see Subheading
2.3).
4. Isotonic solution, syringe filters, patch-clamp glass capillaries,
micropipette puller, patch-clamp setup (see Subheading 2.4).
5. Primers to amplify unique human mtDNA sequence (product
size 152 bp):
forward 50-CGAAAGGACAAGAGAAATAAGG-30,
reverse 50-CTGTAAAGTTTTAAGTTTTATGCG-30.
6. Primers to amplify human nuclear DNA (GADPH gene: prod-
uct size 268 bp):
forward 50-GAAGGTGAAGGTCGGAGTC-30,
reverse 50-GAAGATGGTGATGGGATTC-30.
7. PCR Reaction Mix: 12.5 μL Jump Start Red Taq Master Mix,
1μL mtDNA forward primer (10 μM), 1 μL mtDNA reverse
primer (10 μM), 1 μL nucDNA forward primer (10 μM), 1 μL
nucDNA reverse primer (10 μM), 0.5 μL DMSO, ~5 μLof
isotonic buffer containing mitoplast (mtDNA from mitoplast
serves as a template for PCR), water to the final volume of
25 μL.
8. Thermoblock.
9. Ultrasonic water bath.
10. Thermocycler.
3 Methods
3.1 Mitochondria
Isolation
1. Grow cells in 2–5 culture flasks (see Note 1).
2. All the following steps should be performed on ice or with
ice-cold buffers. Precool homogenizer with the pestle in an ice
bath before starting the procedure. Turn on the centrifuge and
preset to 4 C. Wash cells with PBS (twice is recommended)
and harvest them by scraping in 3.5 mL PBS per flask followed
by collecting cells in the 15 or 50 mL Falcon tube (depend on
cell culture flask).
3. Spin down cells at 400–800 gfor 10 min (depend on
cells type).
4. Resuspend the cell pellet in 2 mL of isolation solution and
transfer to the precooled Potter homogenizer with a glass
pestle.
Patch-Clamp Technique for the Inner Mitochondrial Membrane 241
5. Homogenize cell suspension with homogenizer on ice with
8–10 gentle strokes (for primary cultures 6–8 gentle strokes is
recommended). Transfer homogenate to the Eppendorf tube.
6. Spin down cell homogenate at 9200 gfor 10 min.
7. Discard the supernatant and resuspend the pellet thoroughly
by gentle pipetting. Centrifuge at 750–780 gfor 10 min.
8. Transfer the mitochondria containing supernatant to a new
Eppendorf tube and spin down mitochondria at 9200 gfor
10 min.
9. Resuspend the pelleted mitochondria in storage solution and
spin down at 9200 gfor 10 min. Finally, resuspend the
mitochondria in 300 μL of storage solution. Alternatively,
pelleted mitochondria could be resuspended directly in isola-
tion buffer. All of the steps should be performed at 4 C.
10. For each experimental day, fresh mitochondria should be
isolated unless frozen mitoplasts are to be used (see below).
3.2 Mitoplast
Preparation
Protocol A. Mitoplast preparation directly in the patch-clamp
recording chamber.
1. Place 2 mL of hypotonic solution in 5 cm plastic petri dish
serving as a recording chamber.
2. Add 2 μL of mitochondria suspension to the middle of the dish.
Observe swelling of the mitochondria and formation of mito-
plasts under the phase-contrast microscope of the patch-clamp
setup. Long-distance objective is required.
3. After approximately 1–5 min, add gently 0.5 mL of hypertonic
solution at the periphery of the dish to restore the isotonicity of
the medium.
4. Proceed to the patch-clamp experiment.
Protocol B. Mitoplast preparation for subsequent use or
freezing.
1. Pipette 40 μL of hypotonic solution into an Eppendorf
tube. Add 1 μL of isolated mitochondria. After 1–5 min when
mitochondria are swollen, add 10 μL of sucrose solution (see
Note 2). Place mitoplast on ice. Add 0.2–1 μL of mitoplast into
the recording chamber and proceed with the patch-clamp
experiment.
2. Aliquot 2–5 μL mitoplast into Eppendorf tubes, snap freeze in
liquid nitrogen, and keep at 80 C. Mitoplasts can be kept
frozen for up to a few months. When needed, place tube on ice
and use mitoplasts directly in patch-clamp experiments.
242 Piotr Bednarczyk et al.
3.3 Patch-Clamp
Recordings
1. Set up a patch-clamp puller and pull patch-clamp pipettes with
the resistance of 10–15 MΩwhen filled with isotonic solution.
2. Switch on the patch-clamp rig (Fig. 2) and set up a recording
chamber grounded by the reference electrode. For the record-
ing from mitoplast prepared according to Protocol A, fill the
chamber with hypotonic solution. Alternatively, if mitoplasts
were preformed according to Protocol B, fill the recording
chamber with isotonic solution.
3. Mount freshly pulled pipette filled with isotonic solution in the
pipette holder. Apply slight positive pressure, i.e., blow
(10–20 mmHg) to the recording pipette during all manipula-
tions prior to capturing a mitoplast (see Note 3).
4. Recognize mitoplasts added to the recording chamber by the
round shape, transparency, and presence of a “cap” (Fig. 3),
features that distinguish these structures from the cellular
debris that is also present in the preparation.
5. Using a micromanipulator move the recording pipette
towards a mitoplast and when in close proximity switch the
pressure from positive (blowing) into negative (sucking) in the
range of 20 to 40 mmHg. The Gigaohm seal should form
within few seconds after attachment of the mitoplast to the tip
of the recording pipette. After that time, release the pressure.
At this stage, the membrane patch could spontaneously be
formed at the tip of the pipette and the mitoplast. Observe
the tip of the recording pipette under the microscope. If the
whole mitoplast is still attached to the pipette, you can excise
membrane patch by tapping pipette holder to induce
vibrations.
Fig. 3 Phase-contrast image of a single mitoplast attached to the patch-clamp
pipette. Fractions of the outer mitochondrial membrane are visible as dark “cap”
on top of the swollen mitoplast. The image was taken using an inverted
microscope (Olympus IX71) equipped with ColorView IIIu camera
Patch-Clamp Technique for the Inner Mitochondrial Membrane 243
6. Run voltage protocol in the range of 60 mV to +60 mV or
apply one voltage to check for the presence of channel activity
(Fig. 4)(see Note 4).
7. Transfer the tip of the recording pipette into the opening of the
multibarrel glass tube to apply channel activators or blockers.
3.4 Single-
Mitoplast PCR
1. Isolate mitochondria according to the protocol described
above (Subheading 3.1) with the addition of DNase I (see
Note 5) to the cell suspension before homogenization to
destroy any DNA, which is not protected from the enzyme
activity by a lipid bilayer.
2. Prepare mitoplasts according to the protocol described in Sub-
heading 3.2.
3. Pull micropipettes with the resistance around 2 MΩ, when the
pipette tip is filled with isotonic solution. The tip of the pipette
should be large enough to suck in the mitoplast.
4. Install a pipette in the micromanipulator with only very the tip
filled with isotonic solution and close the suction port keeping
slight positive pressure. Add mitoplasts to the recording cham-
ber filled with either hypotonic solution and form mitoplast
Fig. 4 Representative recordings of the single-channel activity of mitochondrial large-conductance calcium-
activated potassium channel (mitoBK
Ca
). (a) The activity of three channels recorded at a constant voltage of
40 mV (matrix negative). (b) The activity of one channel blocked after perfusion of the tip of the recording
pipette with BK
Ca
channel inhibitor paxilline (1 μM). Both recordings were carried out in the presence of
100 μMCa
2+
.“” indicates a closed state of the channel
244 Piotr Bednarczyk et al.
according to Subheading 3.2, Protocol A or isotonic solution
to use preformed mitoplast according to Subheading 3.2, Pro-
tocol B. Using a micromanipulator move the pipette towards a
selected mitoplast, switch off positive pressure in the pipette
and pick out the mitoplast by application of negative pressure
(suction) to the pipette. Intact mitoplast should be sucked into
the pipette interior (Fig. 5).
5. Take the pipette out of the recording chamber and put its tip
into the PCR tube filled with PCR Reaction Mix. Break the
pipette tip which contains mitoplast, so that it falls into reaction
mix. Discard the rest of the pipette.
6. Incubate closed PCR tube at 95 C for 10 min to inactivate
DNAses.
7. Next, briefly sonicate PCR tube in the ultrasonic water bath to
allow efficient disruption of mitoplast and release of mtDNA.
8. Place PCR tube in a thermocycler and run the appropriate PCR
program (see Note 6).
9. Separate PCR products using a standard 2% agarose gel elec-
trophoresis and analyze the data.
4 Notes
1. Several cell lines have been successfully used in this protocol.
The number of flasks and growth conditions depend on the
type of cells used and should be determined experimentally.
The mitochondria could be also obtained from fresh animal or
plant tissues.
2. The use of sucrose solution reduces the formation of ice crys-
tals during the freezing of mitoplasts. In addition, the higher
density of the final preparation in comparison to isotonic
Fig. 5 Outline of the single-mitoplast PCR procedure. Pick up using a pipette transparent round object that has
a dark cap. Use only one pipette for an object. Carry out PCR reaction and analyze products by standard
agarose gel electrophoresis. If all PCR reactions containing pipette samples have a product corresponding to
mtDNA, you correctly identified all mitoplasts. As a negative control, you can suck in some solution without
any visible object. As positive reaction, use purified mtDNA at low concentration
Patch-Clamp Technique for the Inner Mitochondrial Membrane 245
solution directs the flow of mitoplasts to the bottom of the
recording chamber and helps at micromanipulations during
patch-clamping.
3. Pressure could be applied to the interior of the pipette by
mouth or by various types of manual or automatic systems.
Blowing into the micropipette during manipulations prevents
clogging by debris, which prevails in the mitoplast
preparations.
4. Exact voltage protocol and buffer composition depend on the
type of channel to be recorded. In this protocol, as an example,
buffer composition and voltage protocol are provided to record
the activity of mitoBK
Ca
channels.
5. The concentration of DNase I should be determined experi-
mentally and no product should be present when supernatant
fraction lacking mitoplasts is used as a template for PCR
reaction.
6. Example program for PCR:
95°C – 3 min
95°C – 30 s
57°C – 30 s 25×
72°C – 40 s
72°C – 5 min
4°C – forever
For positive control reactions, use isolated nuclear DNA
(nucDNA) and mtDNA. Prepare negative controls without
mtDNA and nucDNA. Use the same primer sets as for single-
mitoplasts PCR.
Acknowledgments
This work was supported in part by grants from the National
Science Center: 2016/21/B/NZ1/02769 to P.B., 2020/36/T/
NZ1/00116 to R.P.K., 2018/31/N/NZ1/00928 to A.W., and
2015/19/B/NZ1/02794 to P.K.
References
1. Goodman MB, Lindsay TH, Lockery SR,
Richmond JE (2012) Electrophysiological
methods for Caenorhabditis elegans neurobiol-
ogy. Methods Cell Biol 107:409–436. https://
doi.org/10.1016/B978-0-12-394620-1.
00014-X
2. Hodgkin AL, Huxley AF, Katz B (1952) Mea-
surement of current-voltage relations in the
membrane of the giant axon of Loligo. J
Physiol 116(4):424–448. https://doi.org/10.
1113/jphysiol.1952.sp004716
246 Piotr Bednarczyk et al.
3. Neher E, Sakmann B (1976) Single-channel
currents recorded from membrane of dener-
vated frog muscle fibres. Nature 260
(5554):799–802. https://doi.org/10.1038/
260799a0
4. Conforti L (2012) Chapter 20 - patch-clamp
techniques. In: Sperelakis N (ed) Cell physiol-
ogy source book, 4th edn. Academic Press, San
Diego, pp 369–381
5. Hamill OP, Marty A, Neher E, Sakmann B,
Sigworth FJ (1981) Improved patch-clamp
techniques for high-resolution current record-
ing from cells and cell-free membrane patches.
Pflugers Arch 391(2):85–100. https://doi.
org/10.1007/bf00656997
6. Pallotta BS, Magleby KL, Barrett JN (1981)
Single channel recordings of Ca
2+
-activated
K
+
currents in rat muscle cell culture. Nature
293(5832):471–474. https://doi.org/10.
1038/293471a0
7. Siemen D, Loupatatzis C, Borecky J,
Gulbins E, Lang F (1999) Ca
2+
-activated K
channel of the BK-type in the inner mitochon-
drial membrane of a human glioma cell line.
Biochem Biophys Res Commun 257
(2):549–554. https://doi.org/10.1006/bbrc.
1999.0496
8. Bednarczyk P, Koziel A, Jarmuszkiewicz W,
Szewczyk A (2013) Large-conductance Ca(
2
+
)-activated potassium channel in mitochon-
dria of endothelial EA.hy926 cells. Am J
Physiol Heart Circ Physiol 304(11):
H1415–H1427. https://doi.org/10.1152/
ajpheart.00976.2012
9. Chen CC, Cang C, Fenske S, Butz E, Chao YK,
Biel M, Ren D, Wahl-Schott C, Grimm C
(2017) Patch-clamp technique to characterize
ion channels in enlarged individual endolyso-
somes. Nat Protoc 12(8):1639–1658. https://
doi.org/10.1038/nprot.2017.036
10. Cao Q, Zhong XZ, Zou Y, Zhang Z, Toro L,
Dong XP (2015) BK channels alleviate lyso-
somal storage diseases by providing positive
feedback regulation of lysosomal Ca
2+
release.
Dev Cell 33(4):427–441. https://doi.org/10.
1016/j.devcel.2015.04.010
11. Li B, Jie W, Huang L, Wei P, Li S, Luo Z,
Friedman AK, Meredith AL, Han MH, Zhu
XH, Gao TM (2014) Nuclear BK channels
regulate gene expression via the control of
nuclear calcium signaling. Nat Neurosci 17
(8):1055–1063. https://doi.org/10.1038/
nn.3744
12. Frankenreiter S, Bednarczyk P, Kniess A, Bork
NI, Straubinger J, Koprowski P, Wrzosek A,
Mohr E, Logan A, Murphy MP, Gawaz M,
Krieg T, Szewczyk A, Nikolaev VO, Ruth P,
Lukowski R (2017) cGMP-elevating
compounds and ischemic conditioning provide
Cardioprotection against ischemia and reperfu-
sion injury via cardiomyocyte-specific BK chan-
nels. Circulation 136(24):2337–2355.
https://doi.org/10.1161/
CIRCULATIONAHA.117.028723
13. Bednarczyk P, Wieckowski MR,
Broszkiewicz M, Skowronek K, Siemen D,
Szewczyk A (2013) Putative structural and
functional coupling of the mitochondrial
BK
Ca
Channel to the respiratory chain. PLoS
One 8(6):e68125. https://doi.org/10.1371/
journal.pone.0068125
14. Bednarczyk P, Kicinska A, Laskowski M,
Kulawiak B, Kampa R, Walewska A,
Krajewska M, Jarmuszkiewicz W, Szewczyk A
(2018) Evidence for a mitochondrial
ATP-regulated potassium channel in human
dermal fibroblasts. Biochim Biophys Acta Bioe-
nerg 1859(5):309–318. https://doi.org/10.
1016/j.bbabio.2018.02.005
15. Toczylowska-Maminska R, Olszewska A,
Laskowski M, Bednarczyk P, Skowronek K,
Szewczyk A (2014) Potassium channel in the
mitochondria of human keratinocytes. J Invest
Dermatol 134(3):764–772. https://doi.org/
10.1038/jid.2013.422
16. Gururaja Rao S, Bednarczyk P, Towheed A,
Shah K, Karekar P, Ponnalagu D, Jensen HN,
Addya S, Reyes BAS, Van Bockstaele EJ,
Szewczyk A, Wallace DC, Singh H (2019)
BK
Ca
(Slo) channel regulates mitochondrial
function and lifespan in Drosophila melanoga-
ster. Cell 8(9):945. https://doi.org/10.3390/
cells8090945
17. Balderas E, Zhang J, Stefani E, Toro L (2015)
Mitochondrial BK
Ca
channel. Front Physiol
6:104. https://doi.org/10.3389/fphys.2015.
00104
18. Laskowski M, Augustynek B, Bednarczyk P,
Zochowska M, Kalisz J, O’Rourke B,
Szewczyk A, Kulawiak B (2019) Single-channel
properties of the ROMK-pore-forming sub-
unit of the mitochondrial ATP-sensitive potas-
sium channel. Int J Mol Sci 20(21):5323.
https://doi.org/10.3390/ijms20215323
19. Szabo
´I, Zoratti M (1992) The mitochondrial
megachannel is the permeability transition
pore. J Bioenerg Biomembr 24(1):111–117.
https://doi.org/10.1007/bf00769537
20. Urbani A, Giorgio V, Carrer A et al (2019)
Purified F-ATP synthase forms a Ca
2+
-depen-
dent high-conductance channel matching the
mitochondrial permeability transition pore.
Nat Commun 10(1):4341. https://doi.org/
10.1038/s41467-019-12331-1
21. Toczyłowska-Mamin
´ska R, Olszewska A,
Laskowski M, Bednarczyk P, Skowronek K,
Patch-Clamp Technique for the Inner Mitochondrial Membrane 247
Szewczyk A (2014) Potassium channel in the
mitochondria of human keratinocytes. J Invest
Dermatol 134(3):764–772. https://doi.org/
10.1038/jid.2013.422
22. Szabo I, Bock J, Jekle A, Soddemann M,
Adams C, Lang F, Zoratti M, Gulbins E
(2005) A novel potassium channel in lympho-
cyte mitochondria. J Biol Chem 280
(13):12790–12798. https://doi.org/10.
1074/jbc.M413548200
23. Bednarczyk P, Kowalczyk JE, Beresewicz E,
Dolowy K, Szewczyk A, Zablocka B (2010)
Identification of a voltage-gated potassium
channel in gerbil hippocampal mitochondria.
Biochem Biophys Res Comm 397
(3):614–620. https://doi.org/10.1016/j.
bbrc.2010.06.011
24. De Marchi U, Sassi N, Fioretti B,
Catacuzzeno L, Cereghetti GM, Szabo I, Zor-
atti M (2009) Intermediate conductance Ca
2+
-
activated potassium channel (KCa
3.1
) in the
inner mitochondrial membrane of human
colon cancer cells. Cell Calcium 45
(5):509–516. https://doi.org/10.1016/j.
ceca.2009.03.014
25. Walewska A, Kulawiak B, Szewczyk A,
Koprowski P (2018) Mechanosensitivity of
mitochondrial large-conductance calcium-acti-
vated potassium channels. Biochim Biophys
Acta Bioenerg 1859(9):797–805. https://doi.
org/10.1016/j.bbabio.2018.05.006
26. Szabo I, Zoratti M (2014) Mitochondrial
channels: ion fluxes and more. Physiol Rev 94
(2):519–608. https://doi.org/10.1152/
physrev.00021.2013
27. Testai L, Rapposelli S, Martelli A, Breschi MC,
Calderone V (2015) Mitochondrial potassium
channels as pharmacological target for cardio-
protective drugs. Med Res Rev 35
(3):520–553. https://doi.org/10.1002/med.
21332
28. Kampa RP, Kicinska A, Jarmuszkiewicz W,
Pasikowska-Piwko M, Dolegowska B,
Debowska R, Szewczyk A, Bednarczyk P
(2019) Naringenin as an opener of mitochon-
drial potassium channels in dermal fibroblasts.
Exp Dermatol 28(5):543–550. https://doi.
org/10.1111/exd.13903
29. De Marchi U, Checchetto V, Zanetti M et al
(2010) ATP-sensitive cation-channel in wheat
(Triticum durum Desf.): identification and
characterization of a plant mitochondrial chan-
nel by patch-clamp. Cell Physiol Biochem 26
(6):975–982. https://doi.org/10.1159/
000324010
30. Dolga AM, Netter MF, Perocchi F et al (2013)
Mitochondrial small conductance SK2 chan-
nels prevent glutamate-induced oxytosis and
mitochondrial dysfunction. J Biol Chem 288
(15):10792–10804. https://doi.org/10.
1074/jbc.M113.453522
31. Kicinska A, Augustynek B, Kulawiak B,
Jarmuszkiewicz W, Szewczyk A, Bednarczyk P
(2016) A large-conductance calcium-regulated
K
+
channel in human dermal fibroblast mito-
chondria. Biochem J 473(23):4457–4471.
https://doi.org/10.1042/BCJ20160732
248 Piotr Bednarczyk et al.
Chapter 19
Assessment of Mitochondrial Protein Glutathionylation
as Signaling for CO Pathway
Ana S. Almeida, Cla
´udia Figueiredo-Pereira, and Helena L. A. Vieira
Abstract
Protein glutathionylation is a posttranslational process that regulates protein function in response to redox
cellular changes. Furthermore, carbon monoxide-induced cellular pathways involve reactive oxygen species
(ROS) signaling and mitochondrial protein glutathionylation. Herein, it is described as a technique to
assess mitochondrial glutathionylation due to low concentrations of CO exposure. Mitochondria are
isolated from cell culture or tissue, followed by an immunoprecipitation assay, which allows the capture
of any glutathionylated mitochondrial protein using a specific antibody coupled to a solid matrix that binds
to glutathione antigen. The precipitated protein is further identified and quantified by immunoblotting
analysis.
Key words Glutathionylation, Carbon monoxide, Mitochondria, Glutathione, Immunoprecipitation
1 Introduction
Protein glutathionylation is a posttranslation mechanism involved
in redox response, which consists on the regulated formation of
mixed disulfides between protein thiol and glutathione disulfide
(GSSG) due to glutathione redox changes [1,2]. The progressive
glutathionylation of key proteins can be a molecular switch by
which cells respond in an immediate and reversible fashion to
oxidative stress by protecting cysteine residues [1,2]. Still, it can
alter protein activity, presenting a physiological signaling function,
in the same way as the phosphorylation process. Mitochondria are
key organelles for reactive oxygen species (ROS) generation, thus
protein glutathionylation can be crucial for protecting mitochon-
dria from this source of oxidative damage. Furthermore, changes in
the redox state of mitochondrial proteins through thiol modifica-
tions can transduce redox signals and modulate mitochondrial
activity [3].
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_19,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
249
Several examples of mitochondrial protein glutathionylation or
de-glutathionylation are described in the literature: (1) glutathio-
nylation of complex II decreases after myocardial ischemia, limiting
its electron transfer activity [4]; (2) glutathionylation of specific
cysteine residues (C136 and C155) regulates activity of carnitine/
acylcarnitine carrier [5]; (3) glutathionylation of complex I protects
against oxidative stress and is mediated by thiyl radical [6]; (4) ANT
(ATP/ADP translocator) glutathionylation prevents cell death by
improving its activity and limiting mitochondrial membrane per-
meabilization [7]; and (5) degradation of mitochondrial thymidine
kinase-2 is modulated by glutathionylation [8].
Carbon monoxide (CO) is endogenously produced through
the cleavage of heme group by heme oxygenase activity (HO),
presenting several biological properties: anti-inflammatory, anti-
proliferative, and antiapoptotic, for review [9,10]. Cell redox
responses, such as ROS signaling, appear to be tightly involved in
CO-induced pathways [11], namely in: anti-inflammation in
macrophages [12]; cytoprotection in cardiomyocytes [13]; in neu-
rons [14] or in astrocytes [7]; cardioprotection [15]; and antipro-
liferation in airway smooth muscle cells [16]. CO also modulates
levels of oxidized glutathione, signaling through mitochondrial
protein glutathionylation [7].
Herein a method for assessing CO-induced mitochondrial pro-
tein glutathionylation is described, in particular glutathionylation
of the mitochondrial inner membrane protein ANT (ATP/ADP
translocator). ANT presents critical thiol groups in cysteine resi-
dues (cysteine 56, 159, and 256), which can be oxidized and/or
derivatized in order to modulate the pore-forming activity of ANT
and cell death control [17,18]. The described protocol can be used
with other mitochondrial proteins. For assessing CO-induced
mitochondrial protein glutathionylation, two different sources of
mitochondria are used from cell culture (cell lines or primary
cultures of astrocytes) and brain cortex.
2 Materials
Prepare all solutions using ultrapure water (prepared by purifying
deionized water to attain a sensitivity of 18 MΩcm at 25 C) and
analytical grade reagents. Prepare and store all the reagents at 4 C
(unless indicated otherwise).
2.1 Mitochondria
Isolation from Cell
Culture
1. Phosphate buffer saline (PBS): 1.54 M NaCl, 34 mM
Na
2
HPO
4
,20mMKH
2
PO
4
, pH 9.4. In 900 mL of water,
dissolve 90 g of NaCl, 4.83 g of Na
2
PO
4
, and 2.72 g of
KH
2
PO
4
. Mix and adjust pH. Make up to 1 L with water.
250 Ana S. Almeida et al.
2. Hypotonic buffer: 0.15 mM MgCl
2
, 10 mM KCl, 10 mM Tris-
HCl, pH 7.6. Weigh 1.43 mg of MgCl
2
, 74.56 mg of KCl, and
156.6 mg of Tris-HCl. Add water to a volume of 90 mL. Mix
and adjust pH. Make up to 100 mL with water. Store at 4 C.
3. Homogenate buffer (2): 0.6 M sucrose, 10 mM TES,
0.4 mM EGTA, pH 7.2. Weigh 41.07 g of sucrose, 458.5 mg
of TES, and 30.43 mg of EGTA. Add water to a volume of
190 mL. Mix and adjust pH. Make up to 200 mL with water.
Store at 4 C.
4. Homogenate buffer (1): Dilute 1:2 homogenate buffer (2)
in water.
2.2 Mitochondria
Isolation from Brain
Cortex
1. MIB buffer: 225 mM mannitol, 75 mM sucrose, 1 mM EGTA,
5 mM HEPES, pH 7.4. Weigh 10.25 g of mannitol, 6.42 g of
sucrose, 75.09 mg of EGTA, and 279.88 mg of HEPES. Add
water to a volume of 220 mL. Mix and increase pH until 8 in
order to dissolve EGTA. Adjust pH to 7.4, make up to 250 mL
and store at 4 C.
2. Brain mitochondrial buffer (complex I): 125 mM KCl, 2 mM
K
2
HPO
4
, 1 mM MgCl
2
,1μM EGTA, 20 mM Tris–HCl,
5 mM glutamate, 5 mM malate, pH 7.2. Add to 220 mL,
2.33 g of KCl, 73.6 mg of K
2
HPO
4
, 50.8 mg of MgCl
2
,
25 μL of EGTA, 10 mM stock solution, 788 mg of tris-HCl,
233.92 mg of glutamate, and 167.63 mg of malate. Mix and
adjust pH to 7.2 at 37 C. Make up to 250 mL with water and
store at 4 C(see Note 1).
3. Percoll gradient: Dilute stock solution of Percoll in MIB buffer
to obtain final Percoll concentrations of 15%, 24%, and 40%
(v/v). Mix 0.75 mL of Percoll stock solution with 4.25 mL of
MIB buffer, for 15% concentrated solution. In order to prepare
the 24% and the 40% concentrated solutions, pipette 1.2 mL of
Percoll and 3.8 mL of MIB and 2 mL of Percoll and 3 mL of
MIB, respectively.
2.3 CO Treatment 1. CORM-A1 solution: Prepare a 5 mM solution of CORM-A1
(Sigma-Aldrich) in water. Filtrate the solution with 0.22 μM
filter, aliquot, and store at 20 C. For each use, an aliquot
should be thawed and rapidly added into the culture.
2. CO gas solution: Saturate PBS by bubbling 100% of CO gas for
30 min to produce 10
3
M stock solution. Hundred percent
CO was purchased as compressed gas (Linde, Germany). Fresh
stock solutions of CO gas should be prepared each day and
sealed carefully (see Note 2).
2.4 Immuno-
precipitation
1. 10% Triton X-100: Dilute 10 μL of Triton X-100 in 90 μLof
water.
2. PBS (see Subheading 2.1,item 1).
Mitochondrial Protein Glutathionylation 251
3. Loading buffer: 10% (v/v) glycerol; 10 mM DTT; 0.005%
(w/v) Bromophenol Blue. To prepare 20 mL, weigh 30 mg
of DTT and 0.001 mg of Bromophenol Blue. Solubilize both
in 18 mL of water. Add 2 mL of glycerol, mix, and store at
4C.
2.5 Immunoblotting 1. T-TBS buffer: 0.25 M Tris-HCL; 0.75 M NaCl. Weigh 7.88 g
of Tris-HCl and 8.76 g of NaCl. Solubilize in 1 L of water.
2. Blocking buffer: T-TBS with 5% (w/v) milk. Weigh 5 g of milk
and dilute in 100 mL of T-TBS buffer.
3. Running buffer (10): 0.25 M Tris, 1.92 M glycine, 35 mM
SDS. Weigh 30 g of Tris, 144 g of glycine, and 10 g of SDS.
Solubilize all the components in 1 L of water.
4. Running buffer (1): Dilute 100 mL of running buffer (10)
in 900 mL of water.
5. Transfer buffer: Running buffer with 20% (v/v) of methanol.
Add 200 mL of methanol to 800 mL of running buffer (1).
3 Methods
All the steps should be carried out at 4 C, unless indicated
otherwise.
3.1 CO Treatment 1. Cell culture: Add CO gas solution to culture medium, to a final
concentration of 50–100 μM. If you are using CORM-A1
solution, add it to the culture medium to a final concentration
of 12.5–25 μM(see Note 3). At the required time point after
CO exposure, proceed to mitochondrial isolation from cell
culture (Subheading 3.2).
2. Tissue: After isolation from tissue, add CO gas solution
(or CORM-A1 solution) directly to isolated mitochondria.
Incubate mitochondria at 37 C and proceed to immunopre-
cipitation at the different time points after CO exposure (Sub-
heading 3.4).
3.2 Mitochondria
Isolation from Cell
Culture
Adapted from Vieira et al. [19].
1. Inoculate 175 cm
2
T-flasks with primary cell culture of astro-
cytes or a cell line culture.
2. Maintain cells in culture until achieve the confluence.
3. Wash cell culture (175 cm
2
T-flask) with 5 mL PBS at 4 C, in
order to eliminate any serum.
4. Trypsinize the cells by adding 5 mL trypsin, followed by 5 min
incubation at 37 C.
252 Ana S. Almeida et al.
5. Collect the cells in 10 mL of culture medium and centrifuge at
200 gfor 10 min at 4 C.
6. Discard supernatant, wash the cells with 10 mL of PBS, and
centrifuge at 200 gfor 10 min at 4 C.
7. Discard supernatant, add 3.5 mL of hypotonic buffer, and
incubate at 4 C during 5 min.
8. Add an equal volume (3.5 mL) of homogenization buffer twice
concentrated (2) to a final volume of 7 mL.
9. Homogenize samples with a Dounce glass homogenizer at
4C(see Note 4).
10. Remove the sample to a 50 mL tube, wash glass homogenizer
with homogenization buffer (1) and add it to the sample.
11. Centrifuge cell extracts at 900 gfor 10 min at 4 C (to r-
emove nuclei and unbroken cells).
12. Remove supernatant to a clean tube and centrifuge at
10,000 gfor 10 min at 4 C.
13. Resuspend mitochondrial pellet in 100 μL of homogenization
buffer (1) and quantify the total amount of protein.
3.3 Mitochondria
Isolation from Brain
Tissue
The non-synaptic mitochondria isolation protocol was adapted by
Queiroga et al. [7]from Kristia
´n and colleagues and Sims [20–22].
1. Sacrifice one male Wistar rat (see Notes 5 and 6) by cervical
dislocation.
2. Remove cerebellum and underlying structures (only cortex is
used). Isolate cortex 1 min after death.
3. Wash the cortex in MIB in a petri dish and cut it in small pieces.
4. Homogenize cortex manually ten times with tissue homoge-
nizer and centrifuge the tissue extract at 1300 gfor 3 min at
4C.
5. Keep the supernatant, resuspend the pellet, and recentrifuge at
1300 gfor 3 min at 4 C.
6. Pool together the two supernatant and centrifuge at
21,000 gfor 10 min at 4 C in ultracentrifugation tubes.
7. Resuspend pellet in 3.5 mL of 15% Percoll solution and add it
over the gradient (Fig. 1).
8. Centrifuge the gradient at 31,700 gfor 8 min at 4 C.
9. Remove mitochondrial fraction from layer 24% and 40% with a
syringe. Add MIB buffer to mitochondrial fraction to wash
Percoll out by centrifugation at 16,700 gfor 10 min at 4 C.
10. Resuspend pellet in 10 mL of MIB buffer supplemented with
BSA 5 mg/mL and centrifuge at 6800 gfor 10 min at 4 C.
Mitochondrial Protein Glutathionylation 253
11. Remove supernatant and resuspend mitochondria in 100 μLof
MIB without EGTA.
12. Quantify total amount of protein.
3.4 Immuno-
precipitation
of Proteins in Isolated
Mitochondria
1. Prepare microtubes with 50–100 μg of isolated mitochondria
into 100 μL of homogenization buffer (1).
2. Permeabilize mitochondria by adding 5 μL of Triton X-100 at
10% (final concentration 0.5%).
3. Incubate mitochondria with 20 μL of anti-GSH (ViroGen,
USA) during 1 h 30 min at 37 C.
4. Perform the immunoprecipitation by adding 15 μL of Protein
A/G PLUS-Agarose beads (Santa Cruz Biotechnology, UK)
and incubate them for 30 min at 37 C with extremely gentle
shaking.
5. Discard the supernatant after 10 min of centrifugation at
10,000 g. Wash the pellet with PBS, followed by centrifuga-
tion at 500 gfor 10 min, five times.
6. Resuspend the pellet (proteins attached to the beads) in 40 μL
of loading buffer and freeze at 20 C for further immunoblot
analysis.
3.5 Immunoblotting 1. Load the samples on a 12% SDS-PAGE gel in order to separate
the proteins under reducing electrophoresis conditions. Run
the electrophoresis at fixed voltage of 135–150 V during
30 min.
2. Electrically transfer the proteins to a nitrocellulose membrane
(fixed 500 mA for 1 h).
3. Incubate the membrane at RT for 1 h with blocking buffer.
4. Dilute 1:1000 the primary antibody (anti-ANT, Abcam, UK)
in blocking buffer and incubate the membrane for 2 h at RT.
5. Wash the membrane with T-TBS three times during 10 min.
Fig. 1 Percoll gradient schema. Pipet 1.7 mL of 40% Percoll solution followed by
3.7 mL 24% Percoll and, finally, 3.5 mL 15% Percoll. Mitochondrial content will
be found between 24% and 40% fractions of the gradient after centrifugation
254 Ana S. Almeida et al.
6. Incubate the blot with HRP-labeled anti-mouse IgG antibody
(Abcam, UK), 1:5000 diluted in blocking buffer, for 1 h at RT.
7. Wash the membrane with T-TBS three times during 10 min.
8. Develop the blot using ECL (enhanced chemiluminescence)
detection system (Fig. 2).
9. The area and intensity of bands (Fig. 2) can be quantified by
densitometry analysis and presented as a percentage relative to
control (100%) without any treatment. In this example, it can
be observed that CO treatment increased the amount of glu-
tathionylated ANT (30 kDa), compared to control.
4 Notes
1. EGTA’s concentration can go up to 15 μM, in order to obtain
more consistent results.
2. The concentration of CO in solution was determined spectro-
photometrically by measuring the conversion of deoxymyoglo-
bin to carbon monoxymyoglobin [23].
3. Homogenize samples with the Dounce glass homogenizer
25 times with the loose-fitted pestle and then more 25 times
with the tight-fitted pestle at 4 C.
4. Use CO saturated solution immediately after opening the vial,
about 2 or 3 min. CO releases very easily, changing its final
concentration.
5. Animals are allowed water and food ad libitum for the 24 h
before death.
6. From one male Wistar rat (300–350 g) one might obtain 5 mg
of non-synaptic mitochondria.
Fig. 2 Example of immunoblotting film image obtained after immunoprecipitation
of glutathionylated proteins (α-GSH) of a mitochondrial isolate of astrocytes. The
blot was incubated with primary antibody against ANT, which is the target
glutathionylated mitochondrial protein
Mitochondrial Protein Glutathionylation 255
Acknowledgments
This work was supported by the Portuguese Fundac¸a
˜o para a
Cie
ˆncia e Tecnologia (FCT) for the grants FCT-ANR/
NEUNMC/0022/2012 and UID/Multi/04462/2013, I&D
2015-2020 iNOVA4Health – Programme in Translacional Medi-
cine; and for ASA’s SFRH/BD/78440/2011 and CFP’s SFRH/
BD/106057/2015 fellowships.
References
1. Gallogly MM, Mieyal JJ (2007) Mechanisms of
reversible protein glutathionylation in redox
signaling and oxidative stress. Curr Opin Phar-
macol 7:381–391
2. Popov D (2014) Protein S-glutathionylation:
from current basics to targeted modifications.
Arch Physiol Biochem 120:123–130
3. Mailloux RJ, Treberg JR (2016) Protein
S-glutathionlyation links energy metabolism
to redox signaling in mitochondria. Redox
Biol 8:110–118. https://doi.org/10.1016/j.
redox.2015.12.010
4. Chen YR, Chen CL, Pfeiffer DR, Zweier JL
(2007) Mitochondrial complex II in the post-
ischemic heart. J Biol Chem 282:32640
5. Giangregorio N, Palmieri F, Indiveri C (2013)
Glutathione controls the redox state of the
mitochondrial carnitine/acylcarnitine carrier
Cys residues by glutathionylation. Biochim
Biophys Acta Gen Sub 1830:5299–5304
6. Kang PT, Zhang L, Chen CL, Chen J, Green
KB, Chen YR (2012) Protein thiyl radical med-
iates S-glutathionylation of complex I. Free
Radic Biol Med 53:962–973
7. Queiroga CSF, Almeida AS, Martel C,
Brenner C, Alves PM, Vieira HLA (2010) Glu-
tathionylation of adenine nucleotide translo-
case induced by carbon monoxide prevents
mitochondrial membrane permeabilization
and apoptosis. J Biol Chem
285:17077–17088. https://doi.org/10.
1074/jbc.M109.065052
8. Sun R, Eriksson S, Wang L (2012) Oxidative
stress induced S-glutathionylation and proteo-
lytic degradation of mitochondrial thymidine
kinase 2. J Biol Chem 287:24304–24312
9. Ryter SW, Choi AMK (2016) Targeting heme
oxygenase-1 and carbon monoxide for thera-
peutic modulation of inflammation. Transl Res
167:7–34
10. Figueiredo-Pereira C, Dias-Pedroso D, Soares
NL, Vieira HLA (2020) CO-mediated cyto-
protection is dependent on cell metabolism
modulation. Redox Biol 32:101470
11. Almeida AS, Figueiredo-Pereira C, Vieira HLA
(2015) Carbon monoxide and mitochondria-
modulation of cell metabolism, redox response
and cell death. Front Physiol 6:33
12. Zuckerbraun BS, Chin BY, Bilban M, de Costa
d’Avila J et al (2007) Carbon monoxide signals
via inhibition of cytochrome c oxidase and gen-
eration of mitochondrial reactive oxygen spe-
cies. FASEB J 21:1099–1106
13. Suliman HB, Carraway MS, Ali AS, Reynolds
CM, Welty-wolf KE, Piantadosi CA (2007)
The CO / HO system reverses inhibition of
mitochondrial biogenesis and prevents murine
doxorubicin cardiomyopathy. J Clin Invest
117:3730–3741
14. Almeida AS, Soares NL, Vieira M, Gramsber-
gen JB, Vieira HLA (2016) Carbon monoxide
releasing molecule-A1 (CORM-A1) improves
neurogenesis: increase of neuronal differentia-
tion yield by preventing cell death. PLoS One
11:e0154781
15. Scragg JL, Dallas ML, Wilkinson JA, Varadi G,
Peers C (2008) Carbon monoxide inhibits
L-type Ca2+ channels via redox modulation
of key cysteine residues by mitochondrial reac-
tive oxygen species. J Biol Chem
283:24412–24419
16. Taille
´C, El-Benna J, Lanone S, Boczkowski J,
Motterlini R (2005) Mitochondrial respiratory
chain and NAD(P)H oxidase are targets for the
antiproliferative effect of carbon monoxide in
human airway smooth muscle. J Biol Chem
280:25350–25360
17. Costantini P, Chernyak BV, Petronilli V, Ber-
nardi P (1996) Modulation of the mitochon-
drial permeability transition pore by pyridine
nucleotides and dithiol oxidation at two sepa-
rate sites. J Biol Chem 271:6746–6751
18. Costantini P, Belzacq A-SS, Vieira HLA, Lar-
ochette N et al (2000) Oxidation of a critical
thiol residue of the adenine nucleotide translo-
cator enforces Bcl-2-independent permeability
transition pore opening and apoptosis.
256 Ana S. Almeida et al.
Oncogene 19:307–314. https://doi.org/10.
1038/sj.onc.1203299
19. Vieira HLA, Boya P, Cohen I, Hamel CE et al
(2002) Cell permeable BH3-peptides over-
come the cytoprotective effect of Bcl-2 and
Bcl-X(L). Oncogene 21:1963–1977
20. Kristian T, Fiskum G (2004) A fluorescence-
based technique for screening compounds that
protect against damage to brain mitochondria.
Brain Res Protoc 13:176–182
21. Kristia
´n T, Gertsch J, Bates TE, Siesjo
¨BK
(2000) Characteristics of the calcium-triggered
mitochondrial permeability transition in
nonsynaptic brain mitochondria: effect of
cyclosporin A and ubiquinone O. J Neurochem
74:1999–2009
22. Sims NR (1990) Rapid isolation of metaboli-
cally active mitochondria from rat brain and
subregions using percoll density gradient cen-
trifugation. J Neurochem 55:698–707
23. Motterlini R, Clark JE, Foresti R,
Sarathchandra P, Mann BE, Green CJ (2002)
Carbon monoxide-releasing molecules: charac-
terization of biochemical and vascular activities.
Circ Res 90:e17–e24
Mitochondrial Protein Glutathionylation 257
Chapter 20
3D Optical Cryo-Imaging Method: A Novel Approach
to Quantify Renal Mitochondrial Bioenergetics Dysfunction
Shima Mehrvar, Amadou K. S. Camara, and Mahsa Ranji
Abstract
Mitochondrial dysfunction contributes to various injuries and diseases. A mechanistic understanding of
how dysfunctional mitochondria modulates metabolism is of paramount importance. Three-dimensional
(3D) optical cryo-imager is a custom-designed device that can quantify the volumetric bioenergetics of
organs in small animal models. The instrument captures the autofluorescence of bioenergetics indices
(NADH and FAD) from tissues at cryogenic temperature. The quantified redox ratio (NADH/FAD) is
used as an optical indicator of mitochondrial redox state.
Key words Mitochondria, Redox state, Optical imaging, Bioenergetics, Fluorescence imaging
1 Introduction
1.1 Mitochondrial
Dysfunction
Mitochondria play central roles in various key cellular processes
such as ATP production (bioenergetics), the regulation of calcium
homeostasis, cell death pathways, and also act as both source and
scavenger of reactive oxygen species (ROS). Mitochondrial dys-
function is manifested in the derangement of any of these physio-
logical processes. Thus, mitochondrial dysfunction inevitably leads
to cellular damage, which has been linked to various diseases, such
as Parkinson [1], obesity [2], Alzheimer [3], and cancer [4]. This
connection to mitochondria suggests that various therapeutic inter-
ventions targeting mitochondria could lead to protection against
cellular injury [3,5–8].
Multiple approaches have been developed to assess mitochon-
drial dysfunction. For instance, mitochondrial function and dys-
function can be determined with isolated mitochondrial assays such
as mitochondrial respiratory control [9]. In an intact cell assay, cell
respiratory control can provide the rate of ATP production, the rate
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_20,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
259
of proton leak, the coupling efficiency, the maximum respiratory
rate, the respiratory control ratio, and the spare respiratory capacity
[10].
31
Phosphorus nuclear magnetic resonance spectroscopy has
also given insights into the bioenergetics state in vivo by providing
an indicator of mitochondrial oxidative phosphorylation [11].
1.2 Optical Metabolic
Imaging
Living tissues can be studied using optical imaging. A wide range of
optical imaging techniques can be utilized to visualize tissue
morphologies or to assess metabolic processes [12]. Optical meta-
bolic imaging of tissue is classified into three main categories [13]:
oxygenation imaging, fluorescence imaging of exogenous markers,
and tissue autofluorescence imaging of the mitochondrial
metabolites.
Oxygenation imaging is performed by measuring blood oxyge-
nation, for example, muscle blood oxygen consumption can be
measured in vivo using diffuse optical imaging and spectroscopy
[14]. Tissue oxygen consumption correlates with cytochrome oxi-
dase levels in the tissues [15]. Therefore, whole-body respiration
can be correlated with the overall rate of mitochondrial electron
transport. However, the inference of mitochondrial dysfunction
from changes in oxygen consumption is difficult due to the com-
plexity of the whole organism and tissues [10].
Fluorescence imaging of exogenous markers allows tracking of
cellular metabolic processes by using various metabolic markers,
such as MitoSOX [16]. However, autofluorescence imaging of
redox ratios can provide mitochondrial redox state of the tissue
without the need for exogenous tagging. Two mitochondrial bio-
energetics markers, namely, reduced nicotinamide adenine dinucle-
otide (NADH) and oxidized flavin adenine dinucleotide (FAD), are
autofluorescents. NADH and FAD can be measured by fluores-
cence metabolic imaging techniques pioneered by Chance et al.
[17]. 3D optical cryo-imaging provides a 3D measurement of
NADH and FAD. The cryogenic temperature enables us to have
higher quantum yield and preserves the metabolic state of the
tissue. The ratio of these fluorophores (NADH/FAD), the redox
ratio, provides a quantitative marker of the mitochondrial redox
state of tissues. This work describes how 3D optical cryo-imaging
can be utilized to quantitatively assess mitochondrial redox state,
enabling us to determine altered mitochondrial metabolism in
tissues, particularly in disease state.
2 Materials
The 3D fluorescence cryo-imager was custom-designed in the
Biophotonics Laboratory at the University of Wisconsin-
Milwaukee. A computer with a graphical user interface (GUI)
runs image acquisition, and the cryo-images are processed with a
260 Shima Mehrvar et al.
written code (see Subheading 3). Figure 1shows a schematic view of
the system describing the optical and mechanical components of
the instrument.
2.1 Optical
Components
The optical components are located outside of the freezer and are
the parts of two main light paths, namely the excitation light path
and the emission light path.
1. The excitation light path (Fig. 2) includes:
(a) Mercury arc lamp (see Note 1) as the light source.
(b) Cold mirror.
(c) Motorized filter wheel referred to the excitation filter
wheel.
(d) Controller for the filter wheel.
(e) Band-pass optical filters for the selective excitation of
fluorophores in the tissue (see Note 2).
(f) Reflective mirror for air-guide of the light to the tissue.
2. The emission light path (Fig. 3) includes:
(a) Motorized filter wheel named as emission filter wheel.
(b) Controller for the filter wheel.
Fig. 1 A schematic view of 3D optical Cryo-imager. Major components of the system are labeled (reproduced
from ref. 19 with permission from Springer)
3D Optical Cryo-Imaging 261
Fig. 2 A schematic view of excitation path for the optical design of 3D optical Cryo-imager. Major components
of the system are (1) Mercury-arc lamp, (2) Cold mirror, (3) Excitation filter wheel, and (4) Reflecting mirror
Fig. 3 A schematic view of two designs for the emission path of 3D optical Cryo-imager. (a) the filter wheel is
in front of the zoom lens and camera. (b) the filter wheel is between the zoom lens and camera. Major
components of the emission path are (1) Emission filter wheel, (2) Zoom lens, and (3) CCD Camera
262 Shima Mehrvar et al.
(c) Band-pass optical filters for the selective collection of
emitted autofluorescence from the tissue (see Note 2).
(d) Zoom lens.
(e) Image recordings system, charge-coupled device (CCD)
camera (see Note 3).
3. The emission path of the optical design can be arranged in two
formats:
(a) Design A: This is the most common design (Fig. 3a) that
we use to ensure a large field of view, up to 40 mm (see
Note 4). The filter wheel comes in front of the lens, and
there is no space between the lens and camera. This
allowed us to use comparably large (50 mm) optical filters
and increase the efficiency of fluorescence light collection
resulting to higher image contrast.
(b) Design B: The emission filter wheel can come between the
lens and the camera (Fig. 3b). Then, the camera and lens
can get closer to the tissue with working distance as low as
100 mm, for capturing higher resolution images (see Note
5). We are working on optimizing the optical design that
can increase the resolution of the images and the efficacy
of light collection.
2.2 Mechanical
Components
The mechanical part of the instrument works as an automotive
cryo-microtome system to sequentially slice the tissue. The
mechanical parts include:
1. Freezer (40 C).
2. Microtome blade.
3. Metallic sample holder.
4. AC motor for driving the blade (see Note 6).
5. Stepper motor for the movement of sample carrier.
6. Sensors for detecting the blade location and safety (see Note 7).
7. Sample holder.
8. Motorized XY scanning stage (used in the case of raster scan-
ning see Note 8).
2.3 Sample
Preparation
Requirements
Sample preparation includes sample freezing, black mounting
medium (BMM) preparation, and sample embedding.
1. Supplies needed for sample freezing:
(a) Isopentane (2 methyl butane).
(b) Liquid nitrogen.
(c) 80 C freezer.
3D Optical Cryo-Imaging 263
2. Supplies needed for BMM preparation:
(a) Polyvinyl alcohol.
(b) Distilled water.
(c) India ink or carbon black powder.
3 Methods
3.1 Black Mounting
Medium (BMM)
Preparation
1. Use a reliable fluorescent-free mounting medium.
2. Dilute polyvinyl alcohol and distilled water in a beaker with a
ratio of 1:2 and placed on medium heat for 1 h.
3. Gradually add carbon black powder or India ink until there is
no gray appearance.
4. Cool the BMM in a refrigerator before use.
3.2 Kidney Sample
Preparation
1. On the day of the tissue harvest, deeply anesthetize the animals.
Before tissue harvesting, flush the blood by infusing cold iso-
tonic saline via a catheter placed in the aorta (see Note 9).
2. After complete flushing, remove the kidneys quickly and drop
them into liquid nitrogen chilled isopentane. After 2 min in the
cooled isopentane, move the kidney to liquid nitrogen and
then store at 80 C freezer.
3. One day before initiating the cryo-imaging, place the frozen
kidney on a bed of frozen BMM on top of an aluminum holder
to keep the tissue in place, and then cover with more BMM.
Embed the kidney samples in a way that for each sagittal slice,
there will corresponding sagittal-sectioned images. Allow the
whole sample block to rest in the 80 C freezer for a day.
3.3 Cryo-Imaging
Procedure
3.3.1 Mounting
1. Mount the metal plate on the sample stage inside the freezer,
ready to be sliced and imaged. In the GUI for image acquisi-
tion, the following parameters should be set:
2. Define the Z-resolution in the GUI which is typically 30 μm.
3. Save and label the folder where the images are saved.
4. Set the location of the excitation and emission filters in the filter
wheels.
5. Set the exposure times.
3.3.2 Image Acquisition 1. Slice the tissue with a defined slicing size.
2. Set the excitation and emission filters for NADH.
3. Capture the NADH image and save for the specific slice.
4. Set the excitation and emission filters for FAD.
5. Capture the FAD image and save for the specific slice.
6. Repeat steps 1–5, until the entire sample is sliced and imaged
(see Note 10).
264 Shima Mehrvar et al.
3.4 Image
Processing
The stack of NADH and FAD images from sagittal kidney slices
makes the three-dimensional (3D) cryo-images of the kidney
(Fig. 4a). The image processing is performed using the following
steps:
1. Calibrate the images using flat field images (see Note 11).
2. Subtract the background low-intensity voxels with threshold-
ing (see Note 12).
3. Calculate the redox ratio (NADH/FAD) voxel by voxel
(Fig. 4a).
4. Quantify the 3D rendered image using the histogram of the
redox ratio, which is a distribution of voxel intensities through
the whole volume (Fig. 4b).
5. Calculate the mean of the redox ratio histograms as the optical
marker for mitochondrial redox state (NADH/FAD) of the
kidney.
6. Repeat the procedure (1–5) for all the samples imaged.
7. Analyze the statistical significance between different groups of
the specific animal model.
8. Analyze the heterogeneity of redox ratio regionally for some
applications [18,19].
3.5 Data
Interpretation
Changes in mitochondrial redox state due to various injuries or
diseases have been studied using 3D optical cryo-imaging [18–
26]. Larger redox ratio (RR) suggests a more reduced and less
Fig. 4 3D optical Cryo-imaging reveals the effect of partial body irradiation (PBI) and lisinopril treatment on the
mitochondrial redox ratio. (a) Representative three-dimensional rendered images of kidneys from each
treatment: Nonirradiated (Control), partial body irradiated without (PBI), and with lisinopril treatment
(PBI + Lisino). The fluorescence patterns for NADH, FAD, and the tissue redox ratio (NADH/FAD) are shown.
(b) The corresponding intensity histogram distributions of the whole kidney redox ratio (reproduced from ref.
19 with permission from Springer)
3D Optical Cryo-Imaging 265
oxidized mitochondria, while smaller RR suggests less reduced and
more oxidized mitochondria. Comparing the mean redox ratio of
the injured group with the control group, two scenarios may
happen:
First, the RR of the injured tissue is significantly less than the
RR of the control. This happens during disease state, such as in
diabetic wounds [22], diabetic-related injuries in kidneys [23],
radiation-induced injuries to kidneys [19], and ischemia-
reperfusion injuries to hearts [18], livers [20], and lungs [24]. In
such a situation, the injured tissue shows relatively more oxidized
mitochondria when compared to their corresponding controls.
These observations suggest that the oxidized mitochondria cause
perturbations that interfere with the biochemical machinery of
oxidative phosphorylation (OXPHOS), resulting in impaired ATP
production [8,18,27]. Impaired ATP production can be a sign of
mitochondrial dysfunction due to cellular injuries and/or disease to
the organ.
Second, the RR of injured tissue is significantly greater than RR
of the control. For example, the redox state of tissues increases due
to ischemic insult to organs [18,20,24]. Limiting O
2
supplies
during ischemia prevents the oxidation of NADH by mitochondria
electron transport chain, and thus NADH builds up, causing the
increase in redox ratio. The transgenic rat models with cytosolic
p67
phox
[26] or Nox4 [25] mutation also show increased mito-
chondrial RR in kidneys that may contribute to the protective
effects observed in salt-sensitive rats.
As an example of redox ratio data interpretation in disease and
treatment, here is a glimpse on the role of mitochondrial redox
state in the development and treatment of renal radiation injury,
which has been reported in reference [19]. In that study, the impact
of irradiation on mitochondrial redox ratio was compared in three
groups of rat kidneys: (1) nonirradiated controls, (2) leg out partial
body irradiated (PBI), and (3) leg out PBI followed by lisinopril
treatment (PBI+Lisino). The details of the animal injury model and
treatment with lisinopril to mitigate the irradiation-induced injuries
can be found in the following reports [28–30] as described briefly
below.
Representative examples of 3D cryo-images of the NADH and
FAD fluorescence signals and their redox ratios (NADH/FAD) are
shown in Fig. 4a. Figure 4b illustrates the corresponding redox
ratio histograms of the same three representative kidneys from each
group of rats. Comparing PBI rats to controls, lower NADH and
higher FAD fluorescence signals were observed throughout the
kidney, resulting in a lower redox ratio. These results suggest that
irradiation oxidizes renal mitochondrial redox state and alters mito-
chondrial bioenergetics in the rat model of irradiation-induced
kidney damage. When the PBI rats were treated with lisinopril
after irradiation, the entire kidneys exhibited higher levels of
266 Shima Mehrvar et al.
redox ratios, i.e., reduced redox state when compared to kidneys of
rats exposed to PBI without lisinopril. This result suggests that
lisinopril mitigates irradiation damage by attenuating the oxidation
of mitochondria leading to increase redox ratio and preserve mito-
chondrial function in the kidney.
In conclusion, 3D optical cryo-imager introduces a novel
method to study the correlation between disease progression and
changes in the volumetric RR levels as a marker of mitochondrial
oxidative state [18–26]. Furthermore, 3D optical cryo-imaging
sets a stage for studying and evaluating treatment options and
their effects on mitochondrial redox state of injured tissue
[19]. The 3D volumetric redox ratio images can also provide
information on the heterogeneity in the distribution in mitochon-
drial redox state [18], which will provide insights into how different
parts of an organ responds differentially to stress/injury.
4 Notes
1. LED light sources can be another choice. However, the LED
should have high luminous intensity specifically for NADH and
FAD excitation wavelengths. The mercury arc lamps’ spectrum
has two peaks at the excitation wavelengths of NADH and
FAD, making it a good choice to perform autofluorescence
imaging.
2. The optical filters can be chosen to have a pair of excitations
and emissions of any fluorophores. NADH and FAD are
excited at 350 nm and 430 nm, respectively. The
corresponding emission spectra of NADH and FAD peaks at
460 nm and 530 nm, respectively.
3. The camera should have high quantum efficiency because the
autofluorescence signal is weak in biological tissues. Using a
low sensor-size, such as ~4 μm, helps us to achieve higher
resolution images.
4. The downside of this design is that the working distance cannot
be lower than 250 mm, due to the size of emission filter wheel
that should come between the lens and tissue.
5. The emission filter wheel makes an unavoidable space between
the camera and zoom lens that decreases the collection light
efficiency and makes the field of view smaller. Therefore, this
design can only be used when the samples are small, i.e., the
field of view <10 mm. One way to circumvent this problem is
to use a collimating emission-port adapter.
6. The AC motor should be strong enough and have high torque
that can handle hard tasks in freezer temperature. The motor’s
oil also needs to be changed regularly with a low-viscosity oil.
3D Optical Cryo-Imaging 267
7. The safety is considered regarding the movement of the blades,
and the motor does not move the blades while the freezer’s
door is open.
8. When magnifying the image, there will be a small field of view.
In the case of larger samples, the whole tissue may not be fitted
in the image. Raster scanning is implemented by subdividing
the image into images from multiple smaller fields of view.
These single images will then be stitched together to construct
the whole image [31]. This approach improves the resolution
of images. However, the imaging time can be increased to the
point that makes the approach unpractical.
9. Blood in the samples significantly diminishes the autofluores-
cence signals from NADH and FAD. This may introduce var-
iations in the results. This quenching effect of blood can be
prevented by flushing the blood out of the kidneys.
10. The NADH and FAD fluorophores are excited and detected
sequentially and separately. Their emission spectra do not over-
lap, which allows for selective detection of fluorescence
between the two fluorophores.
11. The calibration is needed to minimize the day-to-day variations
in light intensity and mirror angle. It also can compensate for
the distortions caused by the nonuniformity of the illumina-
tions pattern.
12. Very low-intensity tissue voxels or low-quality BMM can cause
the background low-intensity voxels to have a similar intensity
as the tissue voxels. This can cause problem in background
subtraction. This problem can be circumvented by image pro-
cessing steps: (A) calculating a thresholding mask, (B) keeping
the biggest connected component of the image associated with
the tissue and removing the small objects associated with the
background, and (C) filling the mask image to make sure that
all the voxels within the tissue are accounted for.
Acknowledgments
This work is supported by UWM RGI 101x379.
References
1. Abou-Sleiman PM, Muqit MM, Wood NW
(2006) Expanding insights of mitochondrial
dysfunction in Parkinson’s disease. Nat Rev
Neurosci 7(3):207
2. de Mello AH, Costa AB, Engel JDG, Rezin GT
(2018) Mitochondrial dysfunction in obesity.
Life Sci 192:26–32
3. Onyango IG, Dennis J, Khan SM (2016) Mito-
chondrial dysfunction in Alzheimer’s disease
and the rationale for bioenergetics based thera-
pies. Aging Dis 7(2):201
4. Hsu C-C, Tseng L-M, Lee H-C (2016) Role of
mitochondrial dysfunction in cancer progres-
sion. Exp Biol Med 241(12):1281–1295
268 Shima Mehrvar et al.
5. Camara AK, Lesnefsky EJ, Stowe DF (2010)
Potential therapeutic benefits of strategies
directed to mitochondria. Antioxid Redox Sig-
nal 13(3):279–347
6. Cano Sanchez M, Lancel S, Boulanger E,
Neviere R (2018) Targeting oxidative stress
and mitochondrial dysfunction in the treat-
ment of impaired wound healing: a systematic
review. Antioxid Redox Signal 7(8):98
7. Park J-S, Davis RL, Sue CM (2018) Mitochon-
drial dysfunction in Parkinson’s disease: new
mechanistic insights and therapeutic perspec-
tives. Curr Neurol Neurosci Rep 18(5):21
8. Camara AK, Bienengraeber M, Stowe DF
(2011) Mitochondrial approaches to protect
against cardiac ischemia and reperfusion injury.
Front Physiol 2:13
9. Funes HA, Blas-Garcia A, Esplugues JV, Apos-
tolova N (2015) Efavirenz alters mitochondrial
respiratory function in cultured neuron and
glial cell lines. J Antimicrob Chemother 70
(8):2249–2254. https://doi.org/10.1093/
jac/dkv098
10. Brand MD, Nicholls DG (2011) Assessing
mitochondrial dysfunction in cells. Biochem J
435(2):297–312
11. Conley KE, Ali AS, Flores B, Jubrias SA,
Shankland EG (2016) Mitochondrial NAD
(P)H in vivo: identifying natural indicators of
oxidative phosphorylation in the 31P magnetic
resonance spectrum. Front Physiol 7:45.
https://doi.org/10.3389/fphys.2016.00045
12. Ntziachristos V (2010) Going deeper than
microscopy: the optical imaging frontier in
biology. Nat Methods 7(8):603
13. Wall MA, Heaster TM, Tilbury K, Choi WJ,
Roblyer D, Wang R, Skala M, Liu JT (2018)
Metabolic imaging approaches: optical imag-
ing. In: Imaging and Metabolism. Springer,
New York, pp 99–126
14. Kadamati P, Sugar JJ, Quirk BJ, Mehrvar S,
Chelimsky GG, Whelan HT, Chelimsky TC,
Ranji M (2018) Near-infrared spectroscopy
muscle oximetry of patients with postural
orthostatic tachycardia syndrome. J Innov
Opt Health Sci 11(05):1850026
15. Yonetani T, Ray G (1965) Studies on cyto-
chrome oxidase. J Biol Chem 240
(8):3392–3399
16. Ghanian Z, Mehrvar S, Jamali N, Sheibani N,
Ranji M (2018) Time-lapse microscopy of oxi-
dative stress demonstrates metabolic sensitivity
of retinal pericytes under high glucose condi-
tion. J Biophotonics 11:e201700289
17. Chance B, Schoener B, Oshino R, Itshak F,
Nakase Y (1979) Oxidation-reduction ratio
studies of mitochondria in freeze-trapped
samples. NADH and flavoprotein fluorescence
signals. J Biol Chem 254(11):4764–4771
18. Ranji M, Motlagh MM, Salehpour F, Sepehr R,
Heisner JS, Dash RK, Camara AKS (2016)
Optical Cryoimaging reveals a heterogeneous
distribution of mitochondrial redox state in
ex vivo Guinea pig hearts and its alteration
during ischemia and reperfusion. IEEE J Transl
Eng Health Med 4:1–10
19. Mehrvar S, MFl C, Medhora M, Camara AKS,
Ranji M (2019) Optical metabolic imaging for
assessment of radiation-induced injury to rat
kidney and mitigation by lisinopril. Ann
Biomed Eng 47(7):1564–1574
20. la Cour MF, Mehrvar S, Kim J, Martin A, Zim-
merman MA, Hong JC, Ranji M (2017) Opti-
cal imaging for the assessment of hepatocyte
metabolic state in ischemia and reperfusion
injuries. Biomed Opt Express 8
(10):4419–4426
21. la Cour MF, Mehrvar S, Heisner JS, Motlagh
MM, Medhora M, Ranji M, Camara AKS
(2018) Optical metabolic imaging of irradiated
rat heart exposed to ischemia-reperfusion
injury. J Biomed Opt 23(1):1–9
22. Mehrvar S, Rymut KT, Foomani FH,
Mostaghimi S, Eells JT, Ranji M, Gopalakrish-
nan S (2019) Fluorescence imaging of mito-
chondrial redox state to assess diabetic wounds.
IEEE J Transl Eng Health Med 7:1–9
23. Maleki S, Sepehr R, Staniszewski K,
Sheibani N, Sorenson CM, Ranji M (2012)
Mitochondrial redox studies of oxidative stress
in kidneys from diabetic mice. Biomed Opt
Express 3(2):273–281
24. Sepehr R, Staniszewski K, Maleki S, Ranji M,
Jacobs ER, Audi S (2012) Optical imaging of
tissue mitochondrial redox state in intact rat
lungs in two models of pulmonary oxidative
stress. J Biomed Opt 17:046010
25. Cowley AW Jr, Yang C, Zheleznova NN,
Staruschenko A, Kurth T, Rein L, Kumar V,
Sadovnikov K, Dayton A, Hoffman M, Ryan
RP, Skelton MM, Salehpour F, Ranji M, Geurts
A (2016) Evidence of the importance of Nox4
in production of hypertension in dahl salt-
sensitive rats. Hypertension 67(2):440–450
26. Salehpour F, Ghanian Z, Yang C, Zheleznova
NN, Kurth T, Dash RK, Cowley AW Jr, Ranji
M (2015) Effects of p67 phox on the mito-
chondrial oxidative state in the kidney of dahl
salt-sensitive rats: optical fluorescence 3-D
cryoimaging. Am J Physiol Renal Physiol 309
(4):F377–F382
27. Stowe DF, Camara AK (2009) Mitochondrial
reactive oxygen species production in excitable
cells: modulators of mitochondrial and cell
3D Optical Cryo-Imaging 269
function. Antioxid Redox Signal 11
(6):1373–1414
28. Fish BL, Gao F, Narayanan J, Bergom C,
Jacobs ER, Cohen EP, Moulder JE, Orschell
CM, Medhora M (2016) Combined hydration
and antibiotics with Lisinopril to mitigate acute
and delayed high-dose radiation injuries to
multiple organs. Health Phys 111
(5):410–419. https://doi.org/10.1097/hp.
0000000000000554
29. Medhora M, Gao F, Gasperetti T, Narayanan J,
Khan AH, Jacobs ER, BLJHp F (2019)
Delayed effects of acute radiation exposure
(Deare) in juvenile and old rats: mitigation by
lisinopril. Health Phys 116(4):529–545
30. Jacobs ER, Narayanan J, Fish BL, Gao F, Har-
mann LM, Bergom C, Gasperetti T, Strande
JL, MJHp M (2019) Cardiac remodeling and
reversible pulmonary hypertension during
pneumonitis in rats after 13-Gy partial-body
irradiation with minimal bone marrow sparing:
effect of lisinopril. Health Phys 116
(4):558–565
31. MasoudiMotlagh, M. (2014). Cryoimaging-
Microscopy Implementation for 3D Optical
Imaging. Electrical Engineering, University of
Wisconsin-Milwaukee. Master of Science.
270 Shima Mehrvar et al.
Chapter 21
Simultaneous Quantification of Mitochondrial ATP and ROS
Production Using ATP Energy Clamp Methodology
Liping Yu, Brian D. Fink, and William I. Sivitz
Abstract
Several methods are available to measure ATP production by isolated mitochondria or permeabilized cells
but have several limitations, depending upon the particular assay employed. These limitations may include
poor sensitivity or specificity, complexity of the method, poor throughput, changes in mitochondrial inner
membrane potential as ATP is consumed, and/or inability to simultaneously assess other mitochondrial
functional parameters. Here we describe a novel nuclear magnetic resonance (NMR)-based assay that can
be carried out with high efficiency in a manner that alleviates the above problems.
Key words ATP, Mitochondria, Superoxide, Reactive oxygen species, H
2
O
2
, NMR, Bioenergetics
1 Introduction
Here we describe a novel, highly sensitive and specific nuclear
magnetic resonance (NMR)-based ATP assay that can be carried
out with reasonably high throughput using small amounts of mito-
chondrial isolates or permeabilized cells (see Note 1). There are
several advantages of this method. First, the assay allows for simul-
taneous fluorescent measurements. For example, it is possible to
quantify the production of reactive oxygen species (ROS) simulta-
neously with ATP production measurement, as we discuss below.
Another major advantage of the assay is that it avoids the problem
of changing mitochondrial membrane potential (ΔΨ) while ADP is
concerted to ATP, as occurs in conventional assays. In contrast to
conventional assays, ΔΨ in our assay is clamped at fixed levels
determined by the amount of ADP added.
1.1 2DOG ATP
Energy Clamp
To assess mitochondrial functional parameters at fixed ΔΨ, we use
excess 2-deoxyglucose (2DOG) (see Note 2) and hexokinase
(HK) to generate an “ATP energy clamp” (Fig. 1). The conversion
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_21,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
271
of 2DOG to 2DOG phosphate (2DOGP) occurs rapidly and irre-
versibly, thereby effectively clamping ADP concentrations and ΔΨ
dependent on the amount of exogenous ADP added (see Note 3).
This enables titration of membrane potential at different fixed
values with mitochondria in respiratory states ranging from state
4 (no ADP, maximal potential) to state 3 (high levels of ADP,
reduced potential) (Fig. 2).
1.2 Use of the 2DOG
ATP Energy Clamp to
Quantify ATP
Production in Isolated
Mitochondria and
Simultaneous
Assessment of H
2
O
2
Production
Mitochondria are added to individual wells of 96-well plates in a
total volume of typically 60 μL and incubated at 37 C in respira-
tory buffer in the presence of HK, 2DOG, and the desired concen-
tration of ADP. [6-
13
C]2DOG is used in two-dimensional
(2D)
13
C/
1
H-HSQC NMR-based assays, while unlabeled 2DOG
is used in one-dimensional (1D)
1
H NMR-based assays (see
below). After incubation for the desired time, the contents of the
microplate wells are removed to tubes on ice containing oligomycin
to inhibit ATP synthase. The tubes are then centrifuged, and the
supernatants are held at 20
C until NMR analysis. To prepare the
NMR sample for measurement of ATP production by mitochon-
dria, the assay supernatants are added to 5 mm (OD) standard
NMR tubes along with appropriate NMR buffer (see below).
ATP production rates are calculated based on the percent con-
version of 2DOG to 2DOGP as determined by NMR, the initial
matrix
ATP
synthase
2DOG 2DOG phosphate
ATP AD P
ADP
ATP
HK
IMM
intermembrane
space
Fig. 1 The 2-deoxyglucose (2DOG) energy clamp. Saturating amounts of 2-deoxyglucose (2DOG) and
hexokinase (HK) recycle ATP back to ADP by rapidly and irreversibly converting 2DOG into 2-deoxyglucose
phosphate (2DOGP). The resulting ADP availability is clamped at levels determined by the amount of ADP
added. IMM inner mitochondrial membrane
272 Liping Yu et al.
ADP (µM)
510 20 40 80
1 3 10 30 100 300
A
Heart
Liver
B
1007550250
Time (min)
Time (min)
ADP (µM)
295
300
290
285
280
)Vm( laitnetop edortcelE
272
274
270
268
266
)Vm( laitnetop edortcelE
100604020080
Fig. 2 Computer tracings of inner membrane potential vs. time obtained by incubating normal mouse liver
mitochondria, 0.5 mg/mL (panel a) or heart mitochondria, 0.25 mg/mL (panel b), fueled by the combined
Clamp Methodology 273
2DOG concentration, incubation volume, and incubation time.
The assay clearly has several advantages over existing methods (see
Note 4).
In order to simultaneously assess H
2
O
2
production, mitochon-
drial incubations can be carried out in the presence of DHPA.
Moreover, it is possible to use other fluorescent probes to assess
mitochondrial parameters such as membrane potential and calcium
uptake as long as the probes do not interfere with the NMR
detection.
1.3 Utilization in
Recent Studies
We have used this assay in several recent studies designed to assess
muscle, heart, liver, brain, and brown adipose tissue mitochondrial
function under normal and perturbed metabolic conditions. By
clamping ΔΨ, we found that ATP production by muscle mitochon-
dria isolated from insulin deficient rodents was limited not only by
reduced respiration but also by an inability to harness membrane
potential for ATP production [1]. We also showed that ROS pro-
duction per unit of ATP generated was greater in these diabetic
muscle mitochondria [1]. In further work, we observed that muscle
mitochondria isolated from high-fat fed obesity-prone mice was
associated with a greater oxidative cost of ATP production [2]. In
a further study of a rodent model of type 2 diabetes, we observed
that ATP production by succinate-energized muscle mitochondria
was impaired due to decreased utilization of membrane potential
and that more ADP was required for peak respiration [3].
In more recent work, we found that membrane potential pri-
marily determines the relationship of complex II energized-
respiration to ADP concentrations by isolated muscle mitochondria
[4]. We further observed that oxygen flux in succinate-energized
mitochondria peaked at lower levels of added ADP (higher poten-
tial) but actually decreased at higher ADP (lower potential). Fur-
ther work revealed that this biphasic pattern of oxygen flux versus
ADP concentrations was due to membrane potential dependent
oxaloacetate inhibition of succinate dehydrogenase [5,6]. Of
note, the abovementioned biphasic respiratory pattern was depen-
dent on tissue type from which mitochondria were isolated, being
barely evident in liver mitochondria but prominent for brain, mus-
cle, and heart mitochondria [4].
ä
Fig. 2 (continued) substrates of 5 mM succinate + 5 mM glutamate + 1 mM malate. ADP was added in
incremental amounts to generate the final total recycling nucleotide phosphate concentrations (indicated by
arrows). Potential was determined using a tetraphenylphosphonium (TPP) electrode [2]. After each addition, a
plateau potential was reached, consistent with recycling at a steady ADP concentration and generation of a
stepwise transition from state 4 to state 3 respiration. Note that the potential shown on the y-axis represents
electrode potential (not mitochondrial potential). The actual ΔΨ follows a similar pattern after calculation
using the Nernst equation based on the distribution of tetraphenylphosphonium (TPP) external and internal to
mitochondria
274 Liping Yu et al.
In another study, we examined the effect of calcium on enhanc-
ing respiration by mitochondria energized by complex I substrates
in the presence of varying concentrations of ADP. We found that
the stimulatory effect of calcium on mitochondrial function was
substrate dependent and most prominent over intermediate respi-
ratory status [7]. On the other hand, ROS production was posi-
tively and continuously associated with increasing calcium
concentrations [7]. In that work, we further observed that the
inhibition of respiration by calcium at higher concentrations can
occur independent of opening of the mitochondrial permeability
transition pore.
Finally, we have used this 2DOG ATP energy clamp method-
ology to study liver and heart mitochondrial function in mice fed
with high dietary fat [2] and the effect of uncoupling protein I on
respiration in brown adipose tissue [5].
2 Materials
2.1 Isolation of
Mitochondria
1. Isolation medium: 0.25 M sucrose, 5 mM HEPES (pH 7.2),
0.1 mM EDTA, 0.1% BSA (fatty acid-free).
2. Purification medium: 30% v/v Percoll
®
. Dilute three parts
Percoll
®
with seven parts of isolation medium. 2.4 mL of
Percoll
®
+ 5.6 mL of isolation medium ¼8 mL, sufficient for
two centrifuge tubes. Keep on ice.
3. Beckman XL-80 ultracentrifuge or similar instrument, pre-
cooled to 4 C, SW60 swinging-bucket rotor with caps and
greased O-ring seals, polyallomer centrifuge tubes ~4.2 mL
max capacity per tube.
2.2 Assay Incubation 1. Respiration medium: 0.3% fatty acid-free BSA, 120 mM KCl,
1 mM EGTA, 5 mM KH
2
PO
4
, 2 mM MgCl
2
, 10 mM HEPES,
pH 7.2.
2. Microplate with 96-wells (e.g., a Costar #3792 black round-
bottom plate).
3. 10 U/mL hexokinase (HK), 5 mM 2-deoxyglucose (2DOG)
or [6-
13
C]2DOG, 10-acetyl-3,7-dihydroxyphenoxazine
(DHPA or Amplex Red, Invitrogen), horseradish peroxidase
(HRP).
2.3 Sample
Preparation for NMR
Spectroscopy
1. Sample dilution buffer: 120 mM KCl, 5 mM KH
2
PO
4
,2mM
MgCl
2
, pH 7.2.
2. Deuterium oxide (D
2
O).
3. Standard 7-in. (length) 5 mm (outer diameter) NMR tubes.
Clamp Methodology 275
2.4 NMR
Spectroscopy
1. NMR spectrometer equipped with a dual or triple resonance
probe and capable of acquiring
1
H and
1
H/
13
C HSQC NMR
spectra.
2. NMR spectrometer equipped with an auto sample changer,
thus capable of continuous data acquisition of multiple sam-
ples, for example, 60 samples.
3 Methods
3.1 Isolation of
Mitochondria from
Tissue
1. Carry out all procedures on ice or at 4 C.
2. Harvest the tissue and rinse in isolation medium.
3. Homogenize up to 1 g of tissue in 10–15 mL of isolation
medium using a Potter-Elvehjem-type tissue grinder in an ice
bucket. Fibrous tissues should be minced with scissors to aid
tissue disruption. The Teflon pestle is mounted on a drill set to
approximately 300 rpm. Four to six passes are typically
required. Optionally, a subsequent pass of the homogenate
through a ground-glass-style homogenizer can increase the
mitochondrial yield from fibrous tissues.
4. Centrifuge homogenate at 500 gfor 10 min (low speed spin)
(see Note 5).
5. Transfer the supernatant to Sorvall-type tube (Oak Ridge screw
cap). Discard the pellet and centrifuge at 10,000 gfor 10 min
(high-speed spin). Discard supernatant.
6. Wash the mitochondrial pellet with isolation medium
without BSA.
7. Resuspend the final pellet at ~50% v/v in isolation medium
without BSA.
3.2 Further
Purification of Isolated
Mitochondria (See
Note 6)
1. Add 3.8 mL of 30% Percoll
®
solution to each SW60 polyallo-
mer tube on ice.
2. Resuspend mitochondrial crude prep pellets in 0.1 mL of
isolation medium containing BSA.
3. Lay the mitochondria on top of the Percoll
®
solution and insert
the tubes into the buckets.
4. Use a balance to precisely equalize the mass of the buckets +
tubes + lids, adding isolation medium containing BSA to adjust
the mass. Ensure that the contents of the tubes are within
3 mm of the top of the tubes.
5. Securely attach the bucket caps. A liquid sample within a sealed
bucket will be protected from the vacuum applied to the cen-
trifuge chamber.
276 Liping Yu et al.
6. Hang the buckets on the precooled SW61 rotor at 4 C and
spin for 30 min at 30,000 rpm (~90,000 g).
7. The pure mitochondria band appears near the bottom of the
tube, just above a clear and dense mass of Percoll®. Remove all
contaminating fractions above the mitochondria band with a
pipette.
8. Transfer the mitochondria band to a 1.5 mL centrifuge tube.
9. Add 1 mL of isolation medium without BSA. Spin in a micro-
fuge at 8000 gfor 5 min at 4 C. Remove the supernatant. If
subjecting the mitochondria to endogenous Ca
2+
depletion,
then stop here and proceed with a calcium depletion protocol
(see Note 7). Otherwise, resuspend the pellet in 1 mL of
isolation medium without BSA and spin in a microfuge at
8000 gfor 5 min at 4 C.
10. Resuspend the final washed pellet in BSA-free isolation
medium and keep on ice.
3.3 Assay Incubation 1. Warm a 96-well microplate to 37 C for 10 min. If carrying out
simultaneous fluorescent detection (for example, for reactive
oxygen species), warm the plate in the plate reader and set the
plate reader gain as desired.
2. Preload all reagents (before adding mitochondria) to wells with
1.2respiration medium (upon subsequent addition of mito-
chondria the medium will be 1) in a total volume of 50 μL.
For rat hind limb muscle mitochondria [1], we used the fol-
lowing final or 1assay concentrations: 5 mM succinate +
5 mM glutamate + 1 mM malate (or other mitochondrial fuel
selection and concentrations as desired), 10 U/mL hexokinase
(HK), 5 mM 2-deoxyglucose (2DOG) (using [6-
13
C]2DOG
for 2D NMR detection or unlabeled 2DOG for 1D NMR
detection, see below), and ADP at desired concentration
(up to 100 μM). For simultaneous assay of reactive oxygen
species, add 20 μM 10-acetyl-3,7-dihydroxyphenoxazine
(DHPA or Amplex Red, Invitrogen) plus 5 U/mL of horse-
radish peroxidase (HRP).
3. Wells containing no added substrate should be present on the
plate to serve as background control. Reserve some wells for
inclusion of an H
2
O
2
standard curve, typically containing a
range of H
2
O
2
concentrations up to a maximum of 5–10 μM.
These standard curve wells do not have added mitochondria.
4. Two to three wells should be included in the plate to serve as
positive controls and/or standards for NMR quantification of
the amount of 2DOGP. These wells have the same final total
volume of 60 μL as the others, but only contain 5 mM 2DOG
or [6-
13
C]2DOG (depending on whether 1D or 2D NMR
methods are used for measurement, respectively, see below),
Clamp Methodology 277
10 U/mL hexokinase, and 10 mM ATP in 1respiration
medium. The use of twofold molar excess of ATP with respect
to 2DOG is to ensure full conversion of 2DOG to 2DOGP.
5. To start the assay, add 10 μLof6concentrated mitochondria
suspended in 1respiration medium. The well has a total
volume of 60 μL now. The plate reader program is started
and fluorescence determined at a frequency of once per minute
or greater. Final mitochondrial concentrations are typically
0.1–0.5 mg/mL.
6. After incubation, typically for 5–30 min, harvest the wells by
transferring the well contents to 500 μL tubes containing 1 μL
of 120 μM oligomycin. Then immediately centrifuge the tubes
at 10,000 gfor 4 min at 4 C.
7. Transfer the supernatants to new labeled tubes and hold them
at 20 C until NMR sample assembly.
3.4 Fluorescent
Assessment of H
2
O
2
Production
1. As indicated above, the assay wells contain DHPA (20 μM) and
HRP (5 U/mL).
2. Monitor DHPA fluorescence over the duration of the assay at
544 nm excitation and 590 nM emission.
3. Determine the average fluorescence over the duration of the
assay for all points in the standard curve. Subtract the average
background fluorescence from all values for the standard curve.
The background value is defined as fluorescence in the absence
of H
2
O
2
(the zero H
2
O
2
point in the standard curve).
4. For all wells containing mitochondria, determine the slope of
fluorescence as a function of time (see Note 8). It is not neces-
sary to subtract a background fluorescence value in order to
determine slope for these wells since the background is
constant.
5. Use curve fitting software to assess background-subtracted
standard curve fluorescence as a function of the molar amount
of H
2
O
2
present in the well. Then use the fitted curve equation
to convert slope values for the wells containing mitochondria
to molar values (e.g., pmol per min).
3.5 Processing the
Well Contents for
NMR-Based ATP Assay
1. Add 0.39 mL of sample dilution buffer, 50 μL of deuterium
oxide (D
2
O), and 40 μL of assay well supernatant to a standard
7-in. (length) 5 mm (outer diameter) NMR tube.
2. Deliver the prepared NMR samples (kept at 4 C) to the NMR
facility.
3.6 NMR
Spectroscopy for
Quantifying ATP
Production
1. In our studies, we use a Bruker Avance II 500 MHz NMR
spectrometer equipped with a 5 mm TXI triple resonance
non-cryoprobe operating at 37C. The spectrometer is also
equipped with an automatic sample changer capable of holding
278 Liping Yu et al.
a maximum of 60 samples. The amount of ATP produced by
the mitochondria is quantified by measuring the amount of
2DOGP produced from 2DOG in the presence of hexokinase
as described above.
2. For precise measurement of the amount of 2DOGP formed,
duplicate or triplet control samples are prepared during the
assay (see Subheading 3.3). The control samples have the
same total volume of 60 μL as the other samples, but only
contain 5 mM 2DOG or [6-
13
C]2DOG (depending on
whether 1D or 2D NMR methods are used for measurement,
respectively. See Note 9), 10 U/mL hexokinase, and 10 mM
ATP in 1respiration medium. These control samples are then
subjected to the same protocol for NMR sample preparation
(see Subheading 3.5). Due to the presence of excess ATP, these
control samples have the 2DOG being fully converted into
2DOGP. Therefore, these control samples serve as standards
for quantification of 2DOGP formation within the mitochon-
drial samples and are also used to check for assay reproducibility
since duplicate or triplet control samples are used.
3. Load the control samples and mitochondrial samples onto the
sample changer. We can run 60 samples continuously without
interruption at this spectrometer.
4. Use a control sample or mitochondrial sample to lock, tune,
and shim. Save the optimized shimming parameters which
serve as the starting shimming setting for the subsequent auto-
matic robot run. Also, use this sample to calibrate
1
H and
13
C
channel pulse widths (
13
C pulse calibration is needed only
when [6-
13
C]2DOG is used).
5. For Bruker Topspin software, start the automation program
ICONNMR. Set up the automatic robot run by choosing 1D
1
H NMR experiment if unlabeled 2DOG is used or choosing
2D
13
C/
1
H HSQC NMR experiment if [6-
13
C]2DOG is used
in the samples. Enter appropriate pulse program parameters
such as pulse power level, pulse width, relaxation delay, number
of scans, water presaturation parameters, number of t1 incre-
ments. Also, set to tune
1
H channel and shim on every sample.
It takes about 25 min to collect either a 1D
1
H spectrum or a
2D
13
C/
1
H HSQC spectrum for each sample. Start the
robot run.
6. Transfer the collected NMR data to another Linux computer
where the acquired data are processed by using NMRPipe
package [8] and analyzed using NMRView [9].
7. Using the assigned
1
H and
13
C NMR resonances of 2DOG and
2DOGP [1], measure the peak intensities in NMRView of
2DOGP in the processed control and mitochondrial samples.
Clamp Methodology 279
8. Quantify the amount of 2DOGP present in the mitochondrial
samples by comparing the peak intensity of 2DOGP of the
mitochondrial sample with that of the control/standard sam-
ples. Since a fixed amount of 2DOG (5 mM) is used in the assay
incubation (see Subheading 3.3), the amount of 2DOGP
formed corresponds to the amount of reduction of 2DOG
and thus can be expressed as percentage of conversion from
2DOG to 2DOGP. The amount of 2DOGP determined for
the mitochondrial samples corresponds to the amount of ATP
produced by the mitochondria. Representative data derived
from the analysis by 1D and 2D NMR methods are shown in
Fig. 3.
3.7 Calculation of
ATP Production Rates
1. Determine the percent conversion of 2DOG to 2DOGP. From
the known initial 2DOG concentration and percent conver-
sion, determine the molar amount of 2DOGP formed which
2DOG
2DOGP
ADP
No subs
1000 uM
100 uM
30 uM
10 uM
1.0 uM
0.1 uM
0.0 uM
2DOGP
0%
98%
76%
68%
44%
10%
0%
0%
1.41.61.8
1
H (ppm)
1
H (ppm)
β
13
C (ppm)
α
β
α
Con trol: bla ck
Diabetic: red 2DOG
2DOGP
AB
Fig. 3 Quantification of ATP generated by mitochondria via measurement of 2DOGP converted from 2DOG by
1D
1
H NMR method (a) and 2D
13
C/
1
H HSQC NMR method (b). In panel a, representative data are derived from
the incubation of isolated gastrocnemius muscle mitochondria from a control rat. Mitochondria (0.1 mg/mL)
are incubated for 20 min in respiration medium containing 5 mM 2DOG, excess hexokinase, variable
concentrations of ADP, and fueled with 5 mM succinate, 5 mM glutamate, and 1 mM malate. It clearly
shows that the percent 2DOGP formed by the mitochondrial incubation increases with increasing amount of
ADP used in the incubation. No subs mean that substrates of succinate, glutamate, and malate are not added.
In panel b, overlay of representative 2D
1
H/
13
C HSQC spectra is shown for the H6/C6 region of 2DOG and
2DOGP of the samples that contain mitochondria from a control (black) and diabetic (red) rat. Mitochondria
(0.1 mg/mL) are incubated for 20 min in respiration medium containing 5 mM [6-
13
C]2DOG, excess
hexokinase, and 1 mM ADP, and fueled with 5 mM succinate, 5 mM glutamate, and 1 mM malate. The
cross peaks are labeled, and 1D slices through the cross peaks of 2DOGP are included and clearly show that
the control mitochondria produce more 2DOGP or ATP than the diabetic mitochondria
280 Liping Yu et al.
equals to the molar amount of ATP generated. Calculate ATP
production rates in the microplate assay wells based on the
volume in microplate, the amount used in NMR, and
incubation time.
4 Notes
1. The cytoplasm of permeabilized cells is replaced by the respira-
tory media enabling assessment of mitochondrial function
independent of cytoplasmic events. The methods described
herein are for isolated mitochondria. Permeabilized cells can
also be used by adapting the methodology beginning in Sub-
heading 3.3.
2. Cytoplasmic hexokinase is competitively inhibited by 2-deox-
yglucose, but this is not an issue in isolated mitochondria or
permeabilized cells wherein the cytoplasm is replaced by the
incubation medium.
3. The 2DOG clamp has been used in the past to assess
mitochondrial-bound HK activity at constant ADP [10], but
not to assess mitochondrial physiology or ATP production as
described herein. Mitochondrial HK is not an issue in our assay
since the amount added HK is in excess.
4. There are several advantages to our ATP assay. First, ΔΨ is
clamped, allowing assessment of ATP as a function of its direct
driving force (i.e., ΔΨ). Second, the assay is sensitive enough to
measure ATP production in small numbers of mitochondria.
We found that the 1D
1
H NMR method is 34-fold more
sensitive and the 2D
1
H/
13
C HSQC NMR method is
41-fold more sensitive when compared to 1D
31
P NMR for
ATP detection [1]. The higher sensitivity of the 2D NMR
method is due to the fact that the chemical shifts of the two
H6 protons of the β-anomeric form of 2DOGP are degenerate,
resulting in detection of one single C6/H6 HSQC cross peak
with high intensity. Third, both the 1D and 2D NMR spectra
are highly specific. Fourth, throughput is quite good since we
add mitochondria to multiple wells of a 96-well plate, incubate,
spin off the mitochondria, and then save the samples for NMR
analysis. Fifth, a powerful aspect is that we can assess mitochon-
drial ROS simultaneously with ATP quantification by NMR
since the ROS probe does not interfere with mitochondrial
ATP production or with NMR detection of 2DOGP for ATP
quantification [1]. Therefore, the ATP assay described herein
has clear advantages over conventional methods. Fluorescent
and bioluminescent measurements are sensitive, but lack speci-
ficity and may be confounded by background interference or
variations in light emission [11]. Quantifying ATP by phos-
phorous NMR is not sensitive enough and requires long
Clamp Methodology 281
acquisition time unless large amounts of mitochondria are
used. High-pressure liquid chromatography is precise, but
cumbersome. The ATP:O ratio is often considered representa-
tive of ATP production. However, ATP itself is not measured
and the ratio can be altered by any condition that affects
uncoupling, ATP synthase, or respiration.
5. To increase the mitochondrial yield with fibrous tissues, save
the first low-speed pellet for resuspension and regrinding using
the ground-glass-type homogenizer. Spin the homogenate at
low speed. Combine the supernatants from both low-speed
spins prior to the high-speed spin.
6. Crude mitochondria pellets obtained by standard methods can
be further purified using a self-generating Percoll
®
gradient.
We use a published method [12] described for liver mitochon-
dria (which we adapted for heart and skeletal muscle mitochon-
dria) by using a centrifugal force of 95,000 gto establish the
gradient. Others recommend only 30,000 g. In our initial
attempts by using 30,000 g, we did not get acceptable results
as evidenced by loose, fluffy, and diffuse bands or mitochondria
that stayed only at the top of the tube. When we increased the
centrifugal force to 95,000 g, we obtained excellent separa-
tion of mitochondria from contaminants. In this way, the mito-
chondrial band (near the bottom of tube) is clearly separated
from less-dense contaminants and broken mitochondria (upper
and middle bands, respectively).
7. Calcium depletion can be carried out as follows: Percoll-
purified mitochondria (0.8–1.2 mg) are incubated for 6 min
at 37 C in 1 mL of ionic respiratory buffer (105 mM KCl,
10 mM NaCl, 5 mM Na
2
HPO
4
, 2 mM MgCl
2
,10mM
HEPES pH 7.2, 1 mM EGTA, 0.1 mM malate, 3.2 mM
ATP, 0.2% defatted BSA). The mitochondria are then pelleted
at 10,000 gand the pellet washed twice with ice-cold isola-
tion medium (0.25 M sucrose, 0.1 mM EDTA, 3 mM HEPES,
pH 7.25).
8. The observed increase in Amplex Red fluorescence over time is
usually quite linear. The slope for each well can be converted to
molar H
2
O
2
per unit time per mg of mitochondrial protein
with the aid of the standard curve.
9. The 2D NMR method is preferred since it is more sensitive and
has almost no background interference.
Acknowledgments
This work was supported by Veterans Affairs Medical Research
Funds, by the National Institute of Health [5R01HL073166],
and by the Iowa Affiliate Fraternal Order of the Eagles.
282 Liping Yu et al.
References
1. Yu L, Fink BD, Herlein JA, Sivitz WI (2013)
Mitochondrial function in diabetes: novel
methodology and new insight. Diabetes 62
(6):1833–1842. https://doi.org/10.2337/
db12-1152
2. Yu L, Fink BD, Herlein JA, Oltman CL, Lamp-
ing KG, Sivitz WI (2014) Dietary fat, fatty acid
saturation and mitochondrial bioenergetics. J
Bioenerg Biomembr 46:33–44. https://doi.
org/10.1007/s10863-013-9530-z
3. Fink BD, Bai F, Yu L, Sivitz WI (2016)
Impaired utilization of membrane potential by
complex II-energized mitochondria of obese,
diabetic mice assessed using ADP recycling
methodology. Am J Physiol Regul Integr
Comp Physiol 311(4):R756–Rr63. https://
doi.org/10.1152/ajpregu.00232.2016
4. Bai F, Fink BD, Yu L, Sivitz WI (2016)
Voltage-dependent regulation of complex II
energized mitochondrial oxygen flux. PLoS
One 11(5):e0154982. https://doi.org/10.
1371/journal.pone.0154982
5. Fink BD, Yu L, Sivitz WI (2019) Modulation
of complex II-energized respiration in muscle,
heart, and brown adipose mitochondria by
oxaloacetate and complex I electron flow.
FASEB J 33(11):11696–11705. https://doi.
org/10.1096/fj.201900690R
6. Fink BD, Bai F, Yu L, Sheldon RD, Sharma A,
Taylor EB et al (2018) Oxaloacetic acid med-
iates ADP-dependent inhibition of mitochon-
drial complex II-driven respiration. J Biol
Chem 293(51):19932–19941. https://doi.
org/10.1074/jbc.RA118.005144
7. Fink BD, Bai F, Yu L, Sivitz WI (2017) Regu-
lation of ATP production: dependence on cal-
cium concentration and respiratory state. Am J
Physiol Cell Physiol 313(2):C146–Cc53.
https://doi.org/10.1152/ajpcell.00086.2017
8. Delaglio F, Grzesiek S, Vuister GW, Zhu G,
Pfeifer J, Bax A (1995) NMRPipe: a multidi-
mensional spectral processing system based on
UNIX pipes. J Biomol NMR 6(3):277–293
9. Johnson BA, Blevins RA (1994) NMR view: a
computer program for the visualization and
analysis of NMR data. J Biomol NMR 4
(5):603–614. https://doi.org/10.1007/
bf00404272
10. da Silva WS, Gomez-Puyou A, de Gomez-
Puyou MT, Moreno-Sanchez R, De Felice
FG, de Meis L et al (2004) Mitochondrial
bound hexokinase activity as a preventive anti-
oxidant defense: steady-state ADP formation as
a regulatory mechanism of membrane potential
and reactive oxygen species generation in mito-
chondria. J Biol Chem 279
(38):39846–39855. https://doi.org/10.
1074/jbc.M403835200
11. Manfredi G, Spinazzola A, Checcarelli N, Naini
A (2001) Assay of mitochondrial ATP synthesis
in animal cells. Methods Cell Biol 65:133–145
12. Hovius R, Lambrechts H, Nicolay K, de Kruijff
B (1990) Improved methods to isolate and
subfractionate rat liver mitochondria. Lipid
composition of the inner and outer membrane.
Biochim Biophys Acta 1021(2):217–226
Clamp Methodology 283
Chapter 22
High-Throughput Image Analysis of Lipid-Droplet-Bound
Mitochondria
Nathanael Miller, Dane Wolf, Nour Alsabeeh, Kiana Mahdaviani,
Mayuko Segawa, Marc Liesa, and Orian S. Shirihai
Abstract
Changes to mitochondrial architecture are associated with various adaptive and pathogenic processes.
However, quantification of changes to mitochondrial structures is limited by the yet unmet challenge of
defining the borders of each individual mitochondrion within an image. Here, we describe a novel method
for segmenting primary brown adipocyte (BA) mitochondria images. We describe a granular approach to
quantifying subcellular structures, particularly mitochondria in close proximity to lipid droplets: peridroplet
mitochondria. In addition, we lay out a novel machine-learning-based mitochondrial segmentation method
that eliminates the bias of manual mitochondrial segmentation and improves object recognition compared
to conventional thresholding analyses. By applying these methods, we discovered a significant difference
between cytosolic and peridroplet BA mitochondrial H
2
O
2
production and validated the machine-learning
algorithm in BA via norepinephrine-induced mitochondrial fragmentation and comparing manual analyses
to the automated analysis. This approach provides a high-throughput analysis protocol to quantify ratio-
metric probes in subpopulations of mitochondria in adipocytes.
Key words Image analysis, Machine learning, Mitochondria, Brown adipocyte morphology
1 Introduction
1.1 Background The protocols here provide tools to segment and quantify lipid
droplet number, morphology, and labeling intensity in brown adi-
pocytes. We describe a novel approach to segment mitochondria
within these same cells to investigate mitochondria immediately
adjacent to lipid droplets versus mitochondria in the rest of the
cytosol. Imaging data can provide not only morphological mea-
surements [1] but can also be used to quantify membrane potential
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_22,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
Nathanael Miller and Dane Wolf contributed equally to this work.
Marc Liesa and Orian Shirihai contributed equally as co-corresponding authors.
285
(ΔΨ), mitochondrial mass, ROS production, and calcium concen-
trations, among others. Automated image analysis provides an
unbiased, repeatable, and high-throughput method to quantify
high-resolution micrographs. The use of macros, a set of instruc-
tions for a computer to perform repeatedly, is the most amenable
way to increase image analysis throughput [2]. A macro reduces the
amount of time to analyze a single image while ensuring that
subsequent images are analyzed in an identical manner.
Brown adipocytes (BA) provide a unique model for the study of
mitochondrial interaction with lipid droplets due to its high density
of frequently overlapping mitochondria in the cytosol packed
tightly around lipid droplets. BA are specialized to utilize energy
to produce heat upon activation by norepinephrine [3,4]. This
specialization depends on changes to mitochondrial dynamics.
Recent work has shown that peridroplet mitochondria are distinct
from cytosolic mitochondria [5]. Previous work in other cell types
has focused on quantifying the entire network or manual classifica-
tion of network morphology [6–10].
Due to considerations of cost and transparency, the open-
source programs ImageJ (https://imagej.nih.gov/ij/) and its
plugin-rich counterpart FIJI [11] (http://fiji.sc/) are utilized for
these analysis protocols and data. FIJI provides several plugins not
available in stock ImageJ that greatly expedite analysis. Chief
among these tools is the WEKA trainable segmentation plugin
[12], which enables rapid training and deployment of machine-
learning segmentation protocols. Tools of this nature are incredibly
resource intensive and, as such, must be used on computers with
sufficient processing and graphical power. Both ImageJ and FIJI are
open-source and widely documented, providing ideal tools for
scientific use.
Similar tools have been utilized to quantify mitochondria,
albeit in a more limited manner. Koopman et al. introduced the
concept of using a machine-learning classifier to morphologically
characterize and bin mitochondria [13]. This technique has been
further refined by other groups [7,8]. Both of these approaches did
not fully quantify mitochondrial morphology, but rather quantified
the population of each type of mitochondrion, e.g.,
tubular vs. punctate. Previous work has described what parameters
best define mitochondrial morphology [1]. While the conventional
approach of manual thresholding for segmentation is valid, it is
subject to the analysts’ bias [9]. Combining these above
approaches—using machine-learning to segment mitochondria
which are then individually quantified using established para-
meters—provides the most robust and accurate quantification to
date. While paid software platforms exist to perform a similar
segmentation routine [14], the methods provided here are cross-
platform, adaptable and usable within ImageJ or FIJI. Additionally,
the data outputs from these two methods are not identical and may
in fact complement each other.
286 Nathanael Miller et al.
1.2 Training Images
Used for WEKA
Training Set
Images used for training of the classifier were high resolution and
super resolution micrographs. Training was performed on manually
cropped individual cells from different fields of view
obtained from more than four independent experiments. Mito-
chondria in these cells were labelled with: MitoTracker Green and
tetramethylrhodamine, ethyl ester, perchlorate (TMRE). This
broad training set increased accuracy and flexibility of segmenta-
tion. Further details regarding the images can be found in methods.
1.3 Mitochondria
Morphological
Parameters That
Describe
Mitochondrial
Networking
Imaging data reported in multiple studies [1,7,13,15–18] have
demonstrated that mitochondrial morphology can be described by
various shape descriptors (Table 1). Chief among these shape
descriptors is the circularity measurement (Circ), which compares
the perimeter of the region of interest to the perimeter of a perfect
circle of the same area. This measurement is primarily used to
describe branching of the network because it equals 1 when mea-
suring a perfect circle, and <1 for a starfish shape. Form factor (FF)
is another frequently used shape descriptor and is simply the inverse
of circularity, meaning a starfish shape has an FF much greater than
Table 1
Common mitochondrial morphology parameters
Parameter Information Units Notes
Fluorescence
intensity
(FI)
Brightness of
fluorophore
A.U. Can be the sum or average of the fluorescence intensity
contained in a region of interest (ROI). Ratios of
multiple fluorophores are possible
Area (A) Cross-sectional
area of region
Pixels
(px),
μm
2
Perimeter (P) Distance around
region
Pixels
(px),
μm
Circularity
(Circ)
Branching
4πðAreaÞ
Perimeter2
A.U. 0 <Circ 1;
Circ is 1 for a perfect circle; less than 1 for branched
structures
Circ ¼FF
1
Form factor
(FF)
Branching
Perimeter2
4πðAreaÞ
A.U. 1 FF <1;
FF is 1 for a perfect circle; more than 1 for branched
structures
FF ¼Circ
1
Aspect ratio
(AR)
Elongation
Long Axis
Short Axis
A.U. AR measures elongation, not branching
Solidity Concavity
Object Area
Convex Hull Area
A.U. Creates a fully convex shape (hull) surrounding the object
and measures the ratio of the object area to the convex
hull area
High-Throughput Mitochondrial Image Analysis 287
1, while a perfect circle still has an FF of 1. Another frequent shape
descriptor used for analysis of mitochondrial morphology is the
aspect ratio (AR). AR measures the ratio of the long-axis length
of the region of interest (ROI) to the short-axis length and is a
measure of elongation of the object. Solidity measures the concav-
ity of the ROI, which can also be interpreted as branching or
connectivity. Utilizing these parameters, mitochondria may be
described in morphological as well as fluorescent detail.
1.4 Segmentation
of Cytosolic
Mitochondria Versus
Peridroplet
Mitochondria Reveals
Significant Differences
in H
2
O
2
Levels
Brown adipocytes (BA) provide an ideal test case for these tools.
Not only do the mitochondria in the cytosol differ from those
surrounding lipid droplets [5] but also BA mitochondria fragment
upon norepinephrine-induced thermogenesis [3]. Consequently,
the accurate quantification of these phenomena can demonstrate
the validity of our image analysis protocols. By selectively labeling
lipid, we can visualize lipid droplets in order to segment the mito-
chondria around them. Here, we will describe our new segmenta-
tion method for BAT mitochondria using a lipid label as the base
for the analysis.
The original image contains four channels: nile red (lipid),
roGFP2-Orp1 (H
2
O
2
sensor, two channels), and bright field
(BF). The BF channel is immediately discarded upon beginning
analysis; it is a quality control channel included for manual review
(see Note 1). The roGFP2 probe is a redox-sensitive GFP2 fused to
Orp1, which is itself a H
2
O
2
-specific antioxidant enzyme from yeast
that allows roGFP2 to report on H
2
O
2
levels. We used roGFP2-
Orp1 targetted to the mitochondrial matrix, to measure H
2
O
2
levels inside the mitochondria [19]. Nile red is a lipophilic dye
that labels lipids and is used to locate and segment lipid droplets
within BAT cells (Fig. 1).
When BA images were segmented to separate subcellular popu-
lations of mitochondria, we found that cytosolic mitochondria
produce significantly more H
2
O
2
than peridroplet mitochondria
(Fig. 2). The addition of menadione, a mitochondrial toxicant that
induces ROS formation via redox cycling [20], was used as a
positive control and elicited a significant increase in H
2
O
2
levels
in treated cells.
1.4.1 Trainable WEKA
Segmentation Is
Significantly More Effective
at Mitochondrial
Segmentation Compared
to Conventional
Thresholding Methods
Machine learning is a complex field with multiple algorithmic
approaches, each with tradeoffs between speed, accuracy, processor
utilization, and other system requirements. The FastRandomForest
[21] algorithm attempts to mitigate some of the system require-
ments and speed up computation, which is part of why it is com-
monly used in this context. WEKA provides a graphical user
interface (GUI) useful for training a classifier. Within this GUI,
the user defines the number and names of each class to be trained
(in this example, the two classes are “mitochondria” and “back-
ground”). After defining and naming classes, the user trains the
288 Nathanael Miller et al.
classifier, which entails using any of FIJI’s selection tools (line,
rectangle, circle, etc.) to define regions of an image belonging to
a certain class. In Fig. 3, this iterative process is demonstrated, while
training the classifier to distinguish mitochondrial signal from
background noise. Figure 3f represents the final classification of
the image, showing mitochondria in red and background in green.
The final set of instructions for segmentation is stored in a file called
a classifier with extension “.model.” This published classifier is the
product of over 100 training images containing tens to hundreds of
discrete data points and used for subsequent analysis of new image
sets (Fig. 4).
Fig. 1 Schematic representation of quantification of peridroplet and cytosolic mitochondria in brown adipo-
cytes (BA). The original multichannel image is split into its component channels, which are then filtered for
analysis. The lipid droplet is labeled with nile red dye, and this image region is used for recognition and
segmentation throughout the analysis. The nile red image is binarized and used multiple times in the analysis.
First, it is used to subtract fluorescence bleed-through in the roGFP2 channel from the nile red. Then, the lipid
droplet region is dilated ntimes, with nbeing empirically determined by the user to encompass peridroplet
mitochondria. This dilated lipid region is then used to quantify the mitochondrial signal contained within. The
cytosolic analysis builds on the previous steps, utilizing the dilated lipid region to remove any mitochondrial
fluorescence in that region from the cytosolic image. The cytosolic image is then binarized, segmented, and
measured similarly to the peridroplet image. Scale bar shown in original image is 10 μm and is intentionally
left out during analysis. Manual steps are outlined in red
High-Throughput Mitochondrial Image Analysis 289
1.4.2 Validation
of Machine-Learning
Classifier
The WEKA segmentation classifier was compared to human recog-
nition and segmentation of mitochondria in BA (Fig. 5).
While human manual analysis exhibited the largest dynamic range
(1.3 AR, 0.31 Circ for Manual vs 0.008 AR, 0.08 Circ for thresh-
olding) and provided the most robust results (Fig. 5e), threshold-
ing provided the least robust results—thresholding is indeed
hampered by the airy haze of background signal surrounding
objects, inherently generated by fluorescence microscopy. WEKA
segmentation limited the magnitude of the decreases in mitochon-
drial connectivity induced by norepinephrine (NE) and measured
by Circularity and Solidity, as well as limiting the decrease in mito-
chondrial aspect ratio (AR) (Fig. 5b–e): AR only showed a
non-statistically significant 15% decrease in NE treated BA
(Fig. 5b). This limitation detecting the decrease in AR is caused by
an artifact inherent to the WEKA classifier, which can segment an
exceedingly long mitochondrion, mostly observed in non-
stimulated BA, as 2–3 smaller ones. However, this bias toward
splitting large mitochondria is necessary to maximize mitochondrial
recognition, as evidenced by WEKA segmentation’s trend to detect
more mitochondrial objects. While it may be possible to reduce the
impact of this over-segmentation artifact by restricting the training
set to BA images, rather than a training set with images from other
cell types, this restriction would impede the use of the classifier in
other cell types. Alternatively, due to the higher throughput of
analysis enabled by the WEKA classifier and macro allowing
machine learning, larger sample sizes could also offset this artifact.
Fig. 2 BA mitochondria produce different amounts of H
2
O
2
depending on their
localization in relation to lipid droplets. BA cytosolic mitochondria (CM) produce
significantly more H
2
O
2
than their peridroplet (PDM) counterparts. Addition of
menadione increased ROS production significantly in both mitochondrial populations
290 Nathanael Miller et al.
1.5 Benefits of a
Software Macro
Because manual analysis of hundreds of images is time-consuming,
the WEKA classifier needed to be incorporated into a high-
throughput analysis method. The analysis workflow into which it
was incorporated is shown in Fig. 4. WEKA segmentation provided
segmentation of mitochondria with closer ROIs to the real objects
(with less effect of background signal), as well as better separation
of almost-touching objects compared to conventional filtering and
thresholding methods. The output of the classifier is a probability
map: an image where the pixel intensity value represents the com-
puter’s certainty from 0 to 1 that a specific pixel belongs to a certain
class (in Fig. 4, the “mitochondria” class is shown). A threshold for
Fig. 3 Representation of iterative training of the machine learning classifier using the WEKA trainable
segmentation plugin for FIJI. Once an image is opened in FIJI, WEKA trainable segmentation can be called
from the plugins menu. WEKA is capable of using several machine learning algorithms, with the most fitting
being FastRandomForest considering time and processor utilization. Options set within the WEKA GUI can
control the parameters used for machine learning, though several of these require additional plugins. Once
open, the user defines the number and name of each class of object. Once the classes are defined, the user
can use any selection tool available in FIJI (line, rectangle, circle, etc.) to define what image objects belong to
which class. Upon clicking “train classifier,” the computer then calculates the best segmentation algorithm
and returns a map of the classes overlaid over the training image which can be used to refine the class
segmentation with subsequent rounds of assignment and training. Once WEKA returns a satisfactory
segmentation of the image, the user can finalize the classifier to be used on new data sets. Scale bar
shown in original image is 10 μm and is intentionally left out during analysis. Manual steps are outlined in red
High-Throughput Mitochondrial Image Analysis 291
Fig.4Workflowfor a WEKA-based mitochondrial segmentationandanalysis.The original image is initially
filteredtoremoveunevenbackground using the rolling ballalgorithm with a diameter of 50pixels (px). This
facilitatesbetterrecognitionofmitochondria by the WEKA classifier. Shown at this branch point for compari-
sonistraditional threshold-based segmentation of mitochondriainthesame image.Oncethe image is
classifiedwiththeWEKAclassifier,it returns a probability map. A probabilitymapisthealgorithm’sbestguess
atwhichclasseachpixel in an image belongsto,inthiscasemitochondria or background,withaprobability
from0to1.The user manually thresholdsthecertaintyoftheclassificationfor bestmitochondrialsegmenta-
tion(thoughthisstepmay be automated withvalidation).Afterthethresholdingstep, the imageisbinarized,
themitochondrialobjects segmented for measurement, and thenmeasured.Scalebar shown in original
imageis10μmand isintentionallyleftoutduringanalysis. Manualstepsareoutlinedin red
Fig. 5 WEKA segmentation detects mitochondrial fragmentation in NE-activated BA. A clear trend was
exhibited by traditional thresholding to detect fewer mitochondrial objects per cell due to background
fluorescence and airy haze surrounding mitochondrial objects in the images (a). WEKA segmentation detected
significantly more mitochondrial objects than traditional thresholding but was not significantly different from
manual ROI creation. The WEKA classifier more aggressively splits long filaments than a human analyst,
leading to a trend of increased mitochondrial objects in the basal (nonstimulated) condition compared to
manual analysis (a). This largely unavoidable bias is also evidenced by the significant decrease of AR with
WEKA segmentation when compared to manual analysis (b). This artifact inherent to the algorithm
High-Throughput Mitochondrial Image Analysis 293
this probability map can be manually or automatically assigned.
However, a manual step here ensures data quality, as a human
checkpoint is evaluating the classifier output (see Note 1).
When mitochondria in INS-1 cells were classified as in BA
using WEKA segmentation, the number of detected mitochondria
increased twofold (Fig. 6a). This increase is attributable to the
improvement in the classification as noise of the background sig-
nal located between proximal mitochondria, when compared to
conventional thresholding. This improved classification results in
fewer groups of single mitochondria in close apposition being clas-
sified as a single large mitochondrion. Because of this increased
separation of mitochondria and classification of background signal
as noise, mitochondrial perimeter is significantly decreased in
WEKA-segmented measurements (Fig. 6b). Consequently, mito-
chondrial aspect ratio is increased (Fig. 6c), while circularity and
solidity are decreased (Fig. 6d,e). See Note 2 for further clarification
of these parameters.
The advantage of using macros is a large increase in through-
put. The macros contained herein are intended to be useful tools
that can be freely edited, modified, or incorporated into additional
workflows.
2 Materials
2.1 Primary Brown
Adipocytes Culture
Primary brown adipocytes (BA) were generated by differentiating
pre-adipocytes isolated from BAT as previously described [3,5,22,
23] and will not be described in detail, but only culture conditions.
2.1.1 Primary
Preadipocytes Isolated
from BAT Growth Media
1. DMEM supplemented with 20% newborn calf serum (NCS).
2. 4 mM glutamine.
3. 10 mM HEPES.
4. 0.1 mg/mL sodium ascorbate.
5. 50 U/mL penicillin.
6. 50 mg/mL streptomycin.
Fig. 5 (continued) necessitates a larger sample size for observed trends to be significant. However, Circ (c)
and Solidity (d) exhibit significant increases with NE-activation of brown fat, suggesting reduced mitochondrial
connectivity upon activation and validating the WEKA-based segmentation approach. WEKA segmentation
increases dynamic range of measurements in both circularity and AR, error not shown because dynamic range
was calculated from the averages shown in panels (a)–(d). (e) Each bar is the mean of >10 cells per condition
from three separate experiments with error bars representing SEM; p0.05 by one-way ANOVA: (*) significant
difference between basal and NE conditions of same analysis method; (@) significant difference vs. threshold
analysis of same condition; (#) significant difference vs. WEKA analysis of same condition; ($) significant
difference vs. manual analysis of same condition
294 Nathanael Miller et al.
Fig. 6 Comparison of mitochondrial morphological measurements segmented with WEKA segmentation versus
traditional thresholding. WEKA segmentation is significantly better at recognizing, separating, and quantifying
individual INS-1 cell mitochondria (a). Because of this more accurate segmentation, the mitochondrion is
measured much more closely to its actual boundaries instead of the airy haze surrounding it. As a result, the
perimeter readout is significantly reduced (b). This more accurate segmentation provides more accurate
measurements of Aspect Ratio (c), Circularity (d), and Solidity (e), and cross-sectional area (f). Better object
recognition leads to improved dynamic range of measurements. Each point represents an individual image,
error bars SEM. *p0.05 by t-test
High-Throughput Mitochondrial Image Analysis 295
2.1.2 Primary
Preadipocytes
Differentiation Media to BA
1. Preadipocyte growth media, plus:
(a) 1 μM rosiglitazone maleate.
(b) 4 nM human recombinant insulin.
2.1.3 Adenoviral
Transduction to Deliver
H
2
O
2
Reporters
1. Adenovirus encoding roGFP2-Orp1 targeted to the mitochon-
drial matrix [19] were generated in house using Gateway Clon-
ing into pAdeno-CMV-V5, amplified and titrated at Welgen
Inc as viral particles/mL. Two thousand viral particles per BA
were used (MOI). See Note 3 for discussion about effective
concentration/MOI for adenoviral vectors.
2. Preadipocytes isolated from 12- to 16-week-old C57BL/6J
mice differentiated to BA.
3. BA differentiation media.
2.2 Microscopy *For any BA activation experiment, norepinephrine (NE) is added
to the media at the final concentration of 1 μM.
2.2.1 Basic
Requirements of a Confocal
Microscopy System
1. Zeiss LSM 710 or LSM880 microscope equipped with a plan-
apochromat 100(NA ¼1.4) oil immersion objective. Alter-
natively, this level of detail is generally acquired by an objective
lens of 63magnification or higher, with an NA of 1.4. See
Note 4 for more discussion about minimum requirements.
2. Digital zoom of 1 or above.
3. Image resolution of at least 1024 1024 px.
4. Live-cell incubation equipment (37 C, 8% CO
2
) are required
due to the extended duration of these experiments.
2.2.2 Basic
Requirements of Images
1. Maximum pixel size utilized for analysis was 0.07 μm/px.
Figures 3and 4illustrate lower quality and lower resolution
(0.165 μm/px) which can still be used for this analysis. See
Note 5 for more discussion.
2.2.3 Fluorophores
and Imaging Settings
1. BA with roGFP2-Orp1 expression: four channels, with Green
(reduced roGFP2), Red (Nile red; lipid), T-PMT (DIC), Violet
(Oxidized roGFP2), see Note 6.
(a) Channel 1: Green.
lLaser: 488 nm, power varies by sample.
lDetector wavelengths: 500–550 nm.
(b) Channel 2: Red.
lLaser: 543 nm, power varies by sample.
lDetector wavelengths: 620–670 nm.
296 Nathanael Miller et al.
(c) Channel 3: T-PMT.
lLaser: 488 nm, power varies by sample.
lDetector wavelengths: N/A, captures transmitted DIC
image.
(d) Channel 4: Magenta.
lLaser: 405 nm, power varies by sample.
lDetector wavelengths: 500–550 nm.
2.3 Analysis
2.3.1 Training a WEKA
Classifier
1. Trainable WEKA segmentation uses two file types: ARFF (raw
data to build the classifier) and MODEL (the classifier). First
open FIJI, then an image, then trainable WEKA segmentation,
then the ARFF file with [load data] to edit or add to your
training data set, see Note 7.
2. Use any selection tool available in FIJI to select pixels in an
image and add to a class on the right side of the WEKA
interface. Save as ARFF file with [save data], then open a new
image and continue to train.
3. Once the training data set is complete, click [train classifier]
and the resulting MODEL file can be called from WEKA to
classify the open image.
4. Save the classifier with [save classifier].
5. The classifier can now be used on any open image after opening
WEKA interface and calling it with [load classifier]. It can also
be inserted into a macro and called as part of a repeated opera-
tion. This is how it is utilized in this chapter.
2.3.2 Using a Macro 1. With FIJI open, navigate to the macro menu [plu-
gins !macros !install], then select the appropriate macro
for analysis.
2. Go to [plugins !macros], then the newly installed macro
should appear at the bottom of the list. Click it to run and
follow any prompts from the macro itself. These will vary from
procedure to procedure.
3. The macros contained in this chapter are available online at
bioRxiv [24]. See Note 8.
3 Methods
3.1 Primary
Pre-adipocytes
Transfer to an
Imaging Plate
1. Wash 6-well plate containing cells with 1 mL PBS three times.
2. Add 400 μL of accutase to each well just to cover the bottom.
3. Place in incubator for 5 min.
4. Add 4 mL of pre-adipocyte media to each well to quench the
accutase and collect cells.
High-Throughput Mitochondrial Image Analysis 297
5. Count cells.
6. Calculate the dilution for a plating density of 60,000 cells per
dish in a glass-bottom 35 mm confocal imaging dish.
7. Incubate for 2 days, then add differentiation media and trans-
duce cells.
8. After plating for imaging, differentiate cells for 7 days, chang-
ing the medium every other day. See Note 9. Volumes are
typically 2 mL per dish, but should be adjusted to account for
evaporation.
3.2 Transduction
of BA with Adenovirus
Encoding H
2
O
2
Reporters
1. Transduce BA on differentiation days 0–3.
2. Replace BA media with the same volume of media with 2000
viral particles per cell. See Note 3.
3. Place cells in 37 C8%CO
2
incubator for at least 8 h, see
Note 3.
4. After 8–12 h, aspirate viral-particle containing media, and
replace with BA differentiation media.
3.3 Microscopy 1. Microscope settings are described above in Subheading 2.2.3.
2. Record multiple cells’ (a minimum of 20 per experiment)
locations using the positions’ interface of the microscope.
This way, the same cells can be imaged before and after stimu-
lation with norepinephrine. See Note 10.
3. After locating the cells of interest and recording their position,
image each cell at its basal state.
4. After acquiring basal images of all cells, add 1 μM norepineph-
rine (NE) from ampoule to dish very gently using a 2 μL
pipette. Do not touch the dish with your pipette; this is critical
to preserve the matching between the recorded positions and
the cells imaged before treatment. See Note 11.
5. Mix the media in the plate with NE with a 1 mL pipette. Pipette
gently up and down, without touching the dish, ten times to
mix the media and NE.
6. Immediately begin imaging, and image each position at least
twice after stimulation.
3.4 Image Analysis
(See Note 12)
3.4.1 Brown Adipocyte
(BA) Analysis
1. Crop all individual BA out of the image. The macro assumes a
single cell per image. While there are methods to automatically
crop out individual cells, none have worked as of this writing
consistently on BA.
2. The macro used for this analysis is included in the online
supplement at bioRxiv [24] and is named “mito roGFP peri-
droplet and cytosolic mitos manual region check FINAL.ijm.”
Its steps are visualized in Fig. 1.
298 Nathanael Miller et al.
3. Set up FIJI as in Subheading 2.3.2 and load the appropriate
macro.
4. Once run, navigate the macro to the root folder for analysis.
5. The macro will perform the first steps for lipid droplet and
peridroplet mitochondria recognition, then have a step to
click [OK] to continue if the region recognition was correct.
6. After approval of segmentation, the macro will perform the
second half of the analysis for cytosolic mitochondria.
7. The output after it completes running for all files in the folder is
a .csv file, openable in Excel.
3.4.2 Using WEKA
Classifiers
1. Basic manipulation of WEKA files is discussed in Subheading
2.3.2. This is for single-image utilization of the classifier.
2. The macro discussed herein is available on bioRxiv [24]. It is
titled “WEKA SEGMENTATION BASIC 1 channel manual
threshold region check copypaste to exce.ijm.”
3. This macro assumes cropped images of individual cells. This
must be done manually before the next step.
4. Install the macro as in Subheading 2.3.2.
5. Run the macro. Its basic workflow is shown in Fig. 4.
6. Once started, the macro asks for the folder containing the
images to be analyzed, and for the correct classifier/MODEL
file, in that order. Navigate to and select these parameters.
7. After every cell, the macro will display a results table, which can
then be copy and pasted into excel to work with as any data
table.
8. Editing this macro is an excellent way to increase flexibility and
learn basic automation skills.
4 Notes
1. While a manual step is not required and certainly slows the
analysis, it is a good quality control step. Other approaches
could be saving an intermediate image for review at a later
date, or other methods of validating output.
2. As an analogy, if outlining an object left a 1 mm gap between
the object and its outline, the object would be measured as
being more round (circularity approaching 1, aspect ratio
approaching 1, and solidity approaching 1). However, if the
outline is more accurate and leaves only a 0.1 mm gap between
the object and the outline, the outline’s shape will more closely
match the shape of the object, and its measurements would
diverge from 1 because the object would be measured as less
High-Throughput Mitochondrial Image Analysis 299
round. This increased accuracy of segmentation is illustrated in
Fig. 4by comparing the traditionally segmented image to the
WEKA-segmented image.
3. The multiplicity of infection, or MOI, is an approximation of
adenovirus particles needed to transduce one cell. This value
will change with the expression of the adenoviral receptor in
the cell surface and the method used to titrate the virus.
Indeed, some methods quantify both active and inactive viral
particles. It is imperative that you test your MOI for every new
batch of viruses and cells. Other practical considerations here
are the variable nature of primary cells, differentiation effi-
ciency, and their short lifetime in culture. As a result, we usually
used this rule of thumb: BAT need 10more viral particles for
transduction efficiency than INS-1 or pancreatic cells. We
would then titrate the MOI around that number until we
found the ideal MOI for our desired transduction efficiency.
For a quick guide: we used 1000–2000 MOI for INS1 cells,
5000–10,000 MOI for primary BA, and 1000–2000 MOI for
HepG2 cells. These high MOIs are explained by the viral
titration method, quantifying total viral particles, which
included nonfunctional viruses as well. The MOI is determined
for a specific transduction time—we frequently used over-
night/8 h/12 h. Basic criteria to select a specific MOI included
least cell death and highest number of expressing cells. Some
experiments only required 30% transduction (i.e., imaging
where you can select for the expressing cells). Respirometry
and other experiments require >70%. These steps must be
taken before the actual experiments are performed.
4. While possible to analyze BA images from widefield micro-
scopes, the higher signal-to-noise ratio achieved with a confo-
cal system is preferred, in addition to more resolution.
Confocal microscopes, especially the Zeiss Airyscan, provide
higher resolution and signal-to-noise ratio. Widefield and
spinning disk microscopes are less suited in that regard. How-
ever, the authors have performed roGFP2 experiments on
spinning disk microscopes with some degree of success; the
dynamic range was less than on a confocal and resolution only
sufficient for whole-cell quantification. High resolu-
tion becomes more important when comparing subcellular
regions like a peridroplet vs. cytosolic mitochondrial compart-
ment. In simpler terms, more resolution is almost always better
when measuring parameters dependent on fluorescence
intensity.
5. Some enhancements can improve poor quality images for anal-
ysis, such as a previously published filtering method utilizing a
median filter [25] or utilizing built-in background subtraction
in ImageJ, such as the rolling ball algorithm. Poor signal-to-
300 Nathanael Miller et al.
noise ratio, low fluorescence in samples, and many other
caveats of fluorescent microscopy influence the efficacy of
image analysis. A recent publication details the ideal input
image quality and common problems and artifacts encountered
with analyses of this type [14]. For instance, out-of-focus
images and images with resolutions less than 0.165 μm/px
may not yield accurate results—the algorithms have not been
validated at lower resolutions.
6. For best spectral separation, use separate tracks for each
channel.
7. When creating a new classifier, always save both the ARFF (with
[save data]) and MODEL (with [save classifier]) files; some
updates to WEKA or FIJI occasionally break how classifiers
behave, but you can always rebuild your classifier from the
underlying ARFF in the new version to fix compatibility. It is
not necessary to save the classifier (MODEL) until the training
is complete, but always maintain a backup copy of the ARFF.
When a new version of WEKA disrupts compatibility, just open
the ARFF file in the WEKA interface, and click [train classifier].
The new MODEL file should be compatible.
8. Macros can be viewed and edited by anyone. This is the easiest
way to learn to automate analyses: read and understand existing
analysis workflows. Go to [plugins !macros !edit] and open
the appropriate macro. The macros available on the bioRxiv
link are supplemental material and are intended for use to
repeat these analyses on new images. Because the macro and
protocol published here are designated for use with images
collected by our lab, some tweaking may be required to use
with images composed of different channel orders. Changing
channel order in a macro is straightforward, but you must
decipher the file naming conventions of your microscope
acquisition software. We will not cover them all here, but for
instance “T1-C1-Z1.tiff” usually suggests an image at the first
time point, with the first acquisition channel, in the first Z-
plane of a stack. Adjusting file names and pointers is straight-
forward in FIJI as long as you clearly understand basic multidi-
mensional file names. This issue will certainly be evident when
running the BA analysis macro; the channels of our acquisition
could be different from the one in other labs. This macro
specifically references channels by the convention “CN-Origi-
nalImageName” where “N” is the channel number.
9. Differentiation can be assessed by lipid droplet count and size
(nile red or oil red O staining), expression of UCP1 by Western
blot, or oligomycin insensitive respiration after NE
injection [23].
High-Throughput Mitochondrial Image Analysis 301
10. BA differentiate to different degrees. Typically, a cell is consid-
ered differentiated if a majority of its cytosol contains lipid
droplets as defined by the nile red channel. When imaging,
we analyzed differentiated cells instead of undifferentiated pre-
adipocytes. Also see Note 9.
11. The skill of mixing medium on the microscope without touch-
ing the dish takes much practice, especially when working
through an incubation system. Some tips specific to the Zeiss
LSM 880 confocal microscope outfitted with a clear Pecon
environmental control box system follow. When ready to add
and mix, open the top of the incubation chamber, push the
transmitted light arm back, and brace your dominant arm on
the top of the incubation chamber. Slowly lower your pipette
into the sample until surface tension meets the tip and you can
begin to aspirate. Move the tip down with the surface of the
medium until it approaches the bottom of the dish, but do not
touch the bottom. There will be a minute volume remaining.
Re-add medium on the side of the dish, again utilizing surface
tension to reduce disruption. Adding small volumes follows a
similar procedure but does not require moving with the surface
of the medium. Mixing is the same procedure—but combine
the above steps for aspirating and adding medium.
12. All macros discussed and available from this chapter assume
single-cell images cropped from larger fields of view
(if necessary). This step could be automated, but we have yet
to find a suitably accurate tool to crop individual cells without a
significant number of errors. The intent of providing these
macros is to facilitate training and learning; they represent
years of learning and optimization and can be building blocks
of future macros or protocols. FIJI provides a macro recorder
to start learning the macro language. It also supports Python if
there is experience. The individual will almost always need to
edit/tweak these macros for channel order or data output
columns if input or output requirements differ from our own.
References
1. Joshi MS, Crouser ED, Julian MW et al (2000)
Digital imaging analysis for the study of
endotoxin-induced mitochondrial ultrastruc-
ture injury. Anal Cell Pathol 21:41–48
2. Mutterer J, Rasband W (2012) ImageJ macro
language programmer’s reference guide
v1.46d. RSB Homepage 1–45
3. Wikstrom JD, Mahdaviani K, Liesa M et al
(2014) Hormone-induced mitochondrial fis-
sion is utilized by brown adipocytes as an
amplification pathway for energy expenditure.
EMBO J 33:418–436. https://doi.org/10.
1002/embj.201385014
4. Zingaretti MC, Crosta F, Vitali A et al (2009)
The presence of UCP1 demonstrates that met-
abolically active adipose tissue in the neck of
adult humans truly represents brown adipose
tissue. FASEB J 23:3113–3120. https://doi.
org/10.1096/fj.09-133546
5. Benador IY, Veliova M, Mahdaviani K et al
(2018) Mitochondria bound to lipid droplets
have unique composition, bioenergetics, and
302 Nathanael Miller et al.
dynamics that support lipid droplet expansion.
Cell Metab 27:869
6. Mahdaviani K, Benador IY, Su S et al (2017)
Mfn2 deletion in brown adipose tissue protects
from insulin resistance and impairs thermogen-
esis. EMBO Rep 18:1123. https://doi.org/
10.15252/embr.201643827
7. Leonard AP, Cameron RB, Speiser JL et al
(2015) Quantitative analysis of mitochondrial
morphology and membrane potential in living
cells using high-content imaging, machine
learning, and morphological binning. Biochim
Biophys Acta 1853:348–360. https://doi.
org/10.1016/j.bbamcr.2014.11.002
8. Valente AJ, Maddalena LA, Robb EL et al
(2017) A simple ImageJ macro tool for analyz-
ing mitochondrial network morphology in
mammalian cell culture. Acta Histochem
119:315–326. https://doi.org/10.1016/j.
acthis.2017.03.001
9. Cribbs JT, Strack S (2009) Functional charac-
terization of phosphorylation sites in dynamin-
related protein 1. Methods Enzymol
457:231–253. https://doi.org/10.1016/
S0076-6879(09)05013-7
10. Chaudhry A, Shi R, Luciani DS (2019) A pipe-
line for multidimensional confocal analysis of
mitochondrial morphology, function, and
dynamics in pancreatic β-cells. Am J Physiol
Metab 318:E87–E101. https://doi.org/10.
1152/ajpendo.00457.2019
11. Schindelin J, Arganda-Carreras I, Frise E et al
(2012) Fiji: an open-source platform for
biological-image analysis. Nat Methods
9:676–682. https://doi.org/10.1038/
nmeth.2019
12. Arganda-Carreras I, Kaynig V, Rueden C et al
(2017) Trainable Weka Segmentation: a
machine learning tool for microscopy pixel
classification. Bioinformatics 33:2424–2426.
https://doi.org/10.1093/bioinformatics/
btx180
13. Koopman WJH, Visch H-J, Smeitink JAM,
Willems PHGM (2006) Simultaneous quanti-
tative measurement and automated analysis of
mitochondrial morphology, mass, potential,
and motility in living human skin fibroblasts.
Cytom Part A 69A:1–12. https://doi.org/10.
1002/cyto.a.20198
14. Harwig MC, Viana MP, Egner JM et al (2018)
Methods for imaging mammalian mitochon-
drial morphology: a prospective on Mito-
Graph. Anal Biochem 552:81. https://doi.
org/10.1016/j.ab.2018.02.022
15. Nguyen A, Beyersdorf J, Riethoven J-J, Pan-
nier AK (2016) High-throughput screening of
clinically approved drugs that prime
polyethylenimine transfection reveals modula-
tion of mitochondria dysfunction response
improves gene transfer efficiencies. Bioeng
Transl Med 1:123–135. https://doi.org/10.
1002/btm2.10017
16. Molina AA, Wikstrom JD, Stiles L et al (2009)
Mitochondrial networking protects beta-cells
from nutrient-induced apoptosis. Diabetes
58:2303–2315. https://doi.org/10.2337/
db07-1781
17. Wikstrom JD, Katzman SM, Mohamed H et al
(2007) Beta-cell mitochondria exhibit mem-
brane potential heterogeneity that can be
altered by stimulatory or toxic fuel levels. Dia-
betes 56:2569–2578. https://doi.org/10.
2337/db06-0757
18. Twig G, Graf SA, Wikstrom JD et al (2006)
Tagging and tracking individual networks
within a complex mitochondrial web with
photoactivatable GFP. Am J Physiol Cell
Physiol 291:C176–C184. https://doi.org/
10.1152/ajpcell.00348.2005
19. Morgan B, Sobotta MC, Dick TP (2011) Mea-
suring E GSH and H 2O 2 with roGFP2-based
redox probes. Free Radic Biol Med
51:1943–1951. https://doi.org/10.1016/j.
freeradbiomed.2011.08.035
20. Criddle DN, Gillies S, Baumgartner-Wilson
HK et al (2006) Menadione-induced reactive
oxygen species generation via redox cycling
promotes apoptosis of murine pancreatic acinar
cells. J Biol Chem 281:40485–40492. https:/
/
doi.org/10.1074/jbc.M607704200
21. Breiman L (2001) Random forests. Mach
Learn 45:5–32. https://doi.org/10.1023/
A:1010933404324
22. Cannon B, Nedergaard J (2001) Cultures of
adipose precursor cells from brown adipose
tissue and of clonal brown-adipocyte-like cell
lines. Methods Mol Biol 155:213–224.
https://doi.org/10.1385/1-59259-231-
7:213
23. Assali EA, Jones AE, Veliova M et al (2018)
NCLX prevents cell death during adrenergic
activation of the brown adipose tissue. bioR-
xiv:464339. https://doi.org/10.1101/
464339
24. Miller N, Wolf D, Alsabeeh N et al (2020)
High-throughput image analysis of lipid-
droplet-bound mitochondria. bioR-
xiv:985929. https://doi.org/10.1101/2020.
03.10.985929
25. Smith DD, Kovats S, Lee TD, Cano L (2006)
Median filter algorithm for estimating the
threshold of detection on custom protein
arrays. Biotechniques 41:74–78. https://doi.
org/10.2144/000112204
High-Throughput Mitochondrial Image Analysis 303
Chapter 23
Cell Energy Budget Platform for Multiparametric
Assessment of Cell and Tissue Metabolism
Dmitri B. Papkovsky and Alexander V. Zhdanov
Abstract
Specific bioenergetic signature reports on the current metabolic state of the cell, which may be affected by
metabolic rearrangement, dysfunction or dysregulation of relevant signaling pathways, altered physiological
condition or energy stress. A combined analysis of respiration, glycolytic flux, Krebs cycle activity, ATP
levels, and total biomass allows informative initial assessment. Such simple, high-throughput, multipara-
metric methodology, called cell energy budget (CEB) platform, is presented here and demonstrated with
particular cell and tissue models. The CEB uses a commercial fluorescent lanthanide probe pH-Xtra™to
measure extracellular acidification (ECA) associated with lactate (L-ECA) and combined lactate/CO
2
(T-ECA), a phosphorescent probe MitoXpress
®
-Xtra to measure oxygen consumption rate (OCR), a
bioluminescent ATP kit, and an absorbance-based total protein assay. All the assays are performed on a
standard multi-label reader. Using the same readouts, the CEB approach can be extended to more detailed
mechanistic studies, by targeting specific pathways in cell bioenergetics and measuring other cellular
parameters, such as NAD(P)H, Ca
2+
, mitochondrial pH, membrane potential, redox state, with conven-
tional fluorescent or luminescent probes.
Key words Cell metabolism, Bioenergetics, Oxidative phosphorylation, Glycolysis, Extracellular
acidification, Respiration, Oxygen consumption rate, Cell energy budget, O
2
and pH sensitive probes,
Time-resolved fluorescence, ATP assay
1 Introduction
ATP levels in healthy mammalian cells are usually maintained at
constant levels (1–10 mM depending on the cell type [1]), by the
tight regulation of ATP production and consumption. The main
energy-producing pathways, including oxidative phosphorylation
(OXPHOS), glycolysis, the Krebs cycle, pentose phosphate path-
way (PPP), glutaminolysis and β-oxidation of fatty acids, and sub-
strate level phosphorylation (SLP), all work in a coordinated
manner to maintain optimal energy status of the cell (Fig. 1).
Each pathway normally has a significant spare capacity outside the
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_23,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
305
basal range (resting cell), and multiple regulatory mechanisms
which allow the different pathways to compensate each other
under stress or increased energy demand, so that the cell retains
optimal ATP levels under changing conditions. Conversely, signifi-
cant drops in ATP indicate serious bioenergetic problems leading to
altered function and ultimately to cell death.
Under a disease state associated with perturbed signaling or
metabolism, spare capacity of one or several pathways may be
reduced, thus decreasing cell adaptability to stress conditions
[2]. Traditional “floating bar” indicators such as cellular ATP,
ADP/ATP ratio, NAD(P)H, mitochondrial and cytosolic Ca
2+
,
ΔpH
m
, membrane potentials (ΔΨp and ΔΨm), reactive oxygen
species (ROS), and redox state are useful, but they provide narrow
information on cell bioenergetics. Simultaneous measurement of
extracellular acidification (ECA), total cellular ATP, oxygen con-
sumption rate (OCR), and biomass under several different condi-
tions allows for more systemic assessment of cell bioenergetics
(OXPHOS, glycolysis, and Krebs cycle) and susceptibility of cells
to stress [3–5].
According to the cell energy budget (CEB), two ECA assays are
carried out in standard 96/384-well plates, one with unsealed
samples (lactate-mediated, L-ECA, while CO
2
is allowed to escape)
and another with samples sealed under oil (total, T-ECA, lactate
OxPhos
Cell
metabolism
PPP
Glycolysis Krebs
cycle
Glutaminolysis
Respiration
Krebs
cycle
Biosynthesis
Ion
balance
ATP
balance
Redox state
O
2
Lactate
CO
2
OCR
T-ECA
L-ECA
ATP
ATP
measurement
Protein content
/
/
/
/
Pyruvate
carboxylation
Reductive
carboxylation
NAD(P)
+
/NAD(P)H
turnover
Fig. 1 The concept of cell energy budget analysis. Cell bioenergetics, ion balance, redox state, and
biosynthesis are maintained by the main metabolic pathways, which include glycolysis, glutaminolysis,
pentose phosphate pathway, the Krebs cycle, and OXPHOS. Through these pathways, cells consume
metabolic substrates, O
2
, and ADP and produce ATP, lactate, and CO
2
. Depending on substrate availability,
activity of key enzymes, energy demand, or pharmacological treatments, these pathways contribute differ-
ently to cellular function. This can be assessed using CEB platform which is based on a parallel measurement
of L-ECA, T-ECA, OCR, and ATP values and sample biomass (total protein content)
306 Dmitri B. Papkovsky and Alexander V. Zhdanov
and CO
2
combined), thus allowing to devise the glycolytic and
non-glycolytic (predominantly Krebs cycle derived) ECA compo-
nents. In addition, ATP, OCR, and total protein levels are measured
in parallel or multiplexed manner. The measurements are typically
performed in 30–50 sample wells which include different treat-
ments, necessary controls, and replicates. Following quick prelimi-
nary analysis of cell bioenergetics (Fig. 2a), the whole set of CEB
assays can be fitted on one plate (Fig. 2b). However, for a larger
panel of samples, the two-plate format is better, with ECA and total
protein measured sequentially on one plate, and OCR and ATP—
on the other. For the initial examination of the OXPHOS and
glycolytic fluxes, we also recommend kinetic analysis of ATP con-
tent in cells deprived of glucose or treated with OXPHOS
inhibitors.
The ECA and OCR assays use the long-decay photolumines-
cent probes pH-Xtra™and MitoXpress
®
-Xtra (Agilent Technolo-
gies, Santa Clara, CA), for which fluorescence lifetime (LT) is
A
A
A
AA
A
F/O
F/O
F/O
F/O
F/O
F/O
Wells with cells Protein analysis
Wells without cells
A
MMock treatment
Antimycin A
A
F/O
A
F/O
MMM M
F/O
1 2 3 4 5 6 7 8 9 10 11 12
A
B
C
D
E
F
G
H
Unsealed Sealed
M
Glc
Glc/Gln/Pyr
MM
MMM
A
F/O
A
F/O
A
A
A
AA
A
F/O
F/O
F/O
F/O
F/O
F/O
A
F/O
A
F/O
MMM M
M
MM
MMM
A
F/O
A
F/O
ATP
L-ECA T-ECA OCR
FCCP/Oligomycin
M
A
F/O
M
A
F/O
T-ECA
Plate 1
Plate 2
Assays groupped on plate 1
Assays groupped on plate 2
A
A
AAA
AA A
MMM M
1 2 3 4 5 6 7 8 9 10 11 12
M
MM
MMM
AA
A
A
AAA
AA A
MMM M
M
MM
MMM
AA
M
A
M
A
Glc
Gal
M
A
M
A
0h 0.5 h 1h 2h 3h
ATP
AB
Fig. 2 Recommended plate layouts for CEB experiments. (a) A quick assessment of ATP dynamics is
performed at the beginning of CEB analysis in order to check for glycolysis and OXPHOS deficiency or limited
spare capacity. Cells are incubated in RM or RM-gal for up to 5 h with or without AntA. (b) A complete four-
parametric analysis with two conditions (e.g., media with Glc alone and Glc/Gln/Pyr), 2 treatments (e.g., AntA
(AA), FCCP/OM (F/O) plus mock (M)) and duplicated measurements can be performed on one plate. Protein
assay performed in the wells highlighted in orange allows normalization for total cell biomass. Wells without
cells are used for correction of temperature-dependent changes in probe fluorescence. The “two plate” format
can be used to analyze more conditions, cell types, or replicates (adjustable). In this case, the L-ECA, T-ECA,
and protein assays are performed on the first plate (CO
2
-free conditions), while the OCR and ATP assays are
performed on the second plate
Multi-Parametric Cell Energy Budget Platform 307
measured by rapid lifetime determination (RLD) method (two
time-resolved fluorescence (TR-F) intensity signals are measured
at two delay times [6]). State-of-the-art multi-label plate readers,
including PHERAstar, CLARIOstar or Omega series (BMG Lab-
tech), Victor (PerkinElmer) or Synergy (BioTek) families, normally
support the TR-F and RLD modes and contain necessary filters for
pH-Xtra™and MitoXpress
®
-Xtra probes. Measured LT signals and
derived pH/[H
+
] and ECA, O
2
, and OCR values are calculated
automatically by vendor software (BMG Labtech) or by manual or
semiautomated post-processing of raw fluorescent data (PE, Bio-
Tek), using calibration functions provided here or in the original
publications.
Total ATP and total protein/biomass content are measured
with standard chemiluminescence- and absorbance-based kits/
reagents, such as Promega and Pierce, respectively. The CEB plat-
form can be extended with some other cell-based assays, e.g., with
probes for mitochondrial membrane potential, ROS, Ca
2+
. This
suite of assays provides detailed information on cell metabolic state,
with high-sample throughput, adequate sensitivity, and flexibility.
So far, analytical performance and practical potential of the CEB
platform have been demonstrated in a number of metabolic and
signaling studies performed by different labs [7–10].
Here we describe a four-parametric CEB platform based on the
parallel measurement of ATP, ECA, OCR, and total ATP levels,
normalized for cell biomass/total protein content. We provide
standard step-by-step protocols for preparation and execution of
individual CEB assays, processing of raw experimental data, deter-
mination of the key metabolic parameters, and comparative analysis
of cell metabolism under different conditions. Analytical potential
and flexibility of CEB planform are exemplified with several cell and
tissue models and conditions.
2 Materials
2.1 Critical
Equipment
1. Standard multi-label reader capable of TR-F/RLD, lumines-
cence, and absorbance measurements in 96/384-well plates,
equipped with temperature control, red-sensitive photodetec-
tor (up to 700 nm), software for kinetic assays, and a set of
optical filters:
lExcitation for pH-Xtra™and MitoXpress
®
-Xtra:
340–390 nm (optimum at 380 nm).
lEmission for pH-Xtra™: 615 5 nm.
lEmission for MitoXpress
®
: 640–670 nm (optimum at
650 nm).
308 Dmitri B. Papkovsky and Alexander V. Zhdanov
lLuminescence for ATP kit: No filter (empty slot, excitation
lamp—OFF).
lAbsorbance for protein level analysis using BCA (bicincho-
ninic acid) assay: 560–580 nm.
Recommended instruments are FLUOstar Omega (BMG
Labtech, Germany), Synergy H1 (BioTek, USA), and Victor
4 (PerkinElmer, USA) families (see Note 1).
2. Standard microbiological safety cabinet, tabletop class II, with
HEPA filter.
3. CO
2
and CO
2
-free incubators.
4. pH-meter.
5. Analytical balances.
6. Automatic pipettes P2, P20, P100, and P1000.
7. Standard centrifuges for 15/50 mL and 1.5 mL tubes (Eppen-
dorf, Hettich).
2.2 Cells
and Reagents
1. PC12 rat pheochromocytoma cell line from ATCC (LGC stan-
dards) (see Note 2).
2. pH-Xtra™probe (PH-100) (Agilent).
3. MitoXpress
®
-Xtra HS OCR kit (MX-200) (Agilent).
4. CellTiter-Glo
®
ATP measurement kit (Promega,
Madison, WI).
5. BCA™Protein Assay kit (Pierce, Rockford, III).
6. Several 96-well plates, clear polystyrene, tissue culture grade.
7. One white 96-well plate.
8. Tissue culture flasks, 75 cm
2
, sterile.
9. 1.5 mL Eppendorf tubes.
10. 50 mL tubes.
11. 20 mL plastic syringe with 22 G needle.
12. Dulbecco’s Modified Eagle’s Medium (DMEM) without phe-
nol red and glucose.
13. Roswell Park Memorial Institute (RPMI) 1640 medium.
14. Horse serum (HS).
15. Fetal bovine serum (FBS).
16. HEPES solution, 1 M, pH 7.2.
17. 10,000 U/mL penicillin, 10,000 μg/mL streptomycin
solution.
18. Sodium pyruvate, 100 mM solution.
19. L-Glutamine, 200 mM solution.
20. D(+)-Glucose.
Multi-Parametric Cell Energy Budget Platform 309
21. D(+)-Galactose.
22. Phosphate-buffered saline tablets.
23. 0.25% trypsin/1 mM EDTA solution.
24. Collagen type IV from human placenta.
25. Nerve growth factor (NGF), 7S from mouse submaxillary
glands.
26. Carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone
(FCCP).
27. Antimycin A (AntA).
28. Oligomycin (OM).
29. Dimethyl sulfoxide (DMSO).
2.3 Solutions Prepare all solutions using ultrapure water (electrical resistivity of
18 MΩcm at 25 C) and analytical/molecular biology grade
reagents and store them as required (see Notes 3 and 4).
1. High serum medium (HSM): RPMI 1640 supplemented with
10% HS, 5% FBS, 100 U/mL penicillin/100 μg/mL strepto-
mycin (P/S), and 10 mM HEPES, 500 mL.
2. Low serum medium (LSM): RPMI 1640 supplemented with
1% HS, P/S, HEPES and 100 nM NGF, 500 mL.
3. “pH medium” (PM): Powder DMEM reconstituted in deio-
nized water, filter-sterilized and supplemented with 10 mM
glucose, 2 mM L-glutamine, 1 mM sodium pyruvate, and
NGF, 500 mL.
4. “Respiration medium” (RM) is PM buffered with 20 mM
HEPES, 500 mL.
5. “Respiration medium with galactose” (RM-gal) is RM, in
which glucose is replaced with galactose, 500 mL.
6. PBS is prepared by reconstituting tablets in deionized water,
and sterilized by autoclaving, 0.2–1 L.
3 Methods
3.1 T-ECA and L-ECA
Assays
The L-ECA assay measures the glycolytic component of medium
acidification (i.e., lactate), using pH–Xtra™probe [11] and
unsealed samples with cells in a 96/384-well plate. The T-ECA
assay measures acidification due to both lactate and CO
2
(produced
predominantly by the Krebs cycle and pentose-phosphate path-
way), using samples sealed with mineral oil (see Note 5) and
pH-Xtra™probe. Phosphorescence LT of pH-Xtra™probe
increases with [H
+
] elevation (or pH reduction).
310 Dmitri B. Papkovsky and Alexander V. Zhdanov
1. Grow PC12 cells in suspension in a 75cm
2
flask in HSM at low
passage numbers.
2. Harvest the cells by centrifugation in a 50 mL Falcon tube at
150–200 gfor 5 min (see Note 6), aspirate the supernatant.
3. Resuspend the cells in 20 mL of PBS, centrifuge again and
aspirate the supernatant.
4. Resuspend the cells in 1 mL of trypsin, incubate at 37 C for
90 s, then resuspend again using a P1000 pipette. Neutralize
trypsin with 10 mL of LSM.
5. Using a 20 mL syringe, pass cell suspension 8–10 times
through a 22½ G needle to break cell aggregates (see Note 7).
6. Count cells and dilute to a concentration 2.5 10
5
cells/mL
with LSM.
7. Seed the cells on a collagen IV coated 96-well plate at 5 10
4
cells/well (see Note 8). Adjust the final volume of medium to
200 μL. Fill the remaining (outer) wells with PBS (200 μL) to
maintain uniform humidity and temperature across the plate.
8. Differentiate the cells in a humidified incubator at 5% CO
2
,
37 C for 4–5 days, changing LSM every 48 h (see Note 9).
9. Replace spent medium with 200 μL of RM and place the plate
in CO
2
-free incubator for 2.5 h (37 C).
10. Switch on the reader and warm it up to 37 C. Select measure-
ment settings for ECA and program the instrument to measure
wells designated for ECA assays.
11. Reconstitute the content of 1 vial of pH-Xtra™probe (10 μg)
in 10 mL of PM.
12. Prepare stock solutions of all drugs for cell treatment using PM
(see Note 10).
13. Warm up mineral oil at 37 C.
14. Aspirate RM and add 150 μL of PM to each well with cells.
Replace PBS with 150 μL of PM in the wells used as “no cell”
control. Leave the plate in the same conditions for 30 min (see
Notes 9 and 11).
15. Replace spent PM with 100 μL of PM containing the probe.
Add drug stock solutions (2–10 μL) according to the plate
layout shown in Fig. 2b.
16. Gently add to the wells dedicated for T-ECA analysis 150 μLof
prewarmed mineral oil.
17. Quickly insert the plate in the TR-F reader, start kinetic ECA
measurements for 60–90 min (see Table 1) to generate raw
signal profiles of pH-Xtra™probe (Fig. 3a).
18. When the measurements are completed, take out the plate.
Multi-Parametric Cell Energy Budget Platform 311
3.2 OCR Assay
3.2.1 Rates of Cellular
Oxygen Consumption
Relative rates of cellular O
2
consumption (OCRs) are measured
with MitoXpress
®
-Xtra probe [5,12,13] in cells plated on 96/
384-well plate and sealed with mineral oil (see Note 5). Probe
phosphorescence signals are reversibly quenched by O
2
and deple-
tion of sample O
2
due to cell respiration increases probe signal.
1. Seed and differentiate PC12 cells as in Sect. 3.1,steps 1–8.
Table 1
Typical instrument settings used in CEB assays
Assay method PerkinElmer (Victor4)
BMG Labtech FLUOstar
Omega BioTek Synergy H1
T-ECA/L-ECA Mode: TR-F
Excitation: 340 40 nm
(Eu filter)
Emission: 615 nm
(Eu filter)
Delay time 1: 100 μs
Delay time 1: 300 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats
every 2–3 min
Temperature: 37 C
Mode: TR-F
Excitation: 380 20 nm
Emission: 615 8.5 nm
Delay time 1: 100 μs
Delay time 1: 300 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats
every 2–3 min
Temperature: 37 C
Mode: TR-F
Excitation: 380 20 nm or
360 40 nm
Emission: 620 10 nm
Delay time 1: 100 μs
Delay time 1: 300 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats every
2–3 min
Temperature: 37 C
OCR Mode: TR-F
Excitation: 340 40 nm
(Eu filter)
Emission: 642 10 nm
(Eu filter)
Delay time 1: 30 μs
Delay time 1: 70 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats
every 2–3 min
Temperature: 37 C
Mode: TR-F
Excitation: 380 20 nm
Emission: 655 50 nm
Delay time 1: 30 μs
Delay time 1: 70 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats
every 2–3 min
Temperature: 37 C
Mode: TR-F
Excitation: 380 20 nm
Emission: 645 15 nm
Delay time 1: 30 μs
Delay time 1: 70 μs
Gate time 1: 30 μs
Gate time 2: 30 μs
Integration time: 1 s/well
Repeats: 30–60 repeats every
2–3 min
Temperature: 37 C
Total ATP Mode:
Chemiluminescence
Emission aperture: Large
Filter: none
Integration time: 0.5 s
Repeat: No (end-point)
Mode:
Chemiluminescence
Filter: none
Integration time: 1 s
Repeat: No
Mode: Chemiluminescence
Filter: none
Integration time: 1 s
Repeat: No
Total protein
(biomass)
Mode: Absorbance
Filter: 560–580 nm
Repeat: No
Mode: Absorbance
Filter: 560–580 nm
Repeat: No
Mode: Absorbance
Filter: 560–580 nm
Repeat: No
312 Dmitri B. Papkovsky and Alexander V. Zhdanov
2. In the wells designated for OCR assay (including the wells used
as “no cell” control), replace LSM with 200 μL of RM and
leave the plate for 30 min in a CO
2
incubator (see Note 11).
3. Switch on the reader, select OCR assay method and measure-
ments of designated wells on the plate (Table 1).
4. Take a vial of MitoXpress
®
-Xtra probe (10 nM) and reconsti-
tute its contents in 100 μL of RM (100 μM stock). Dilute
probe stock 1:500 in PM to produce 1solution (200 nM).
5. Prepare drug stocks in RM. (see Note 10).
AB
5
10
15
20
25
0 10 20304050 607080
100
150
200
250
300
350
400
450
0 1020304050607080
0
50
100
150
200
250
300
010 20 30 40 50 60 70 80
No cell (corrected)
LT drift
100
150
200
250
300
350
400
450
0 1020304050607080
No cell Glc
Glc/Gln/Pyr
Phosphorescence [cps*10
4
]
LT [μs]
LT [μs]
Corrected LT [μs]
Time [min] Time [min]
Time [min]
Time [min]
C
DF
6.4
6.6
6.8
7.0
7.2
7.4
0
5E-08
1E-07
1.5-07
2E-07
2.5E-07
3E-07
0 20406080
Extracellular pH
Time [min]
pH
H
+
6.7
6.8
6.9
7.0
7.1
7.2
7.3
20 25 30 35 40 45 50
4E-8
6E-8
8E-8
1E-7
1.2E-7
1.4E-7
1.6E-7
1.8E-7
2E-7
pH
H
+
[M]
Extracellular pH
Time [min]
H
+
E
ABCDE / F
H
+
[M]
pH = 7.44
Glc
0.004
0.008
0.012
L-ECA [pH/min]
Glc/Gln/Pyr
0
0.002
0.006
0.01
G
G
t1 t2
Fig. 3 Example of data processing for kinetic ECA assays. (a) Profiles of raw TR-F intensity signal (F
1
) produced
by dPC12 cells in Glc/Gln/Pyr (top curve), Glc (middle) media, and blank without cells (bottom). (b) Lifetime
(LT) profiles derived from (a). (c) Correction curve (nonspecific drift of probe signal due to temperature and gas
equilibration) is determined from the LT profile of blank samples (bottom). (d) Corrected LT profiles for
different samples. Blank gives a straight line at ~200 μs which corresponds to pH ¼7.44. E. Profiles of pH and
[H
+
] derived from (d). (f) Linear sections corresponding to t
1
t
2
window in (e) used for calculation of ECA
rates (slopes ΔpH/Δtor Δ[H
+
]/Δt). (g) Calculated mean L-ECA values (four replicates for each condition—
error bars) show a drastic increase upon Gln/Pyr deprivation due to compensatory utilization of glycolysis.
Measured on Victor
2
reader (PE)
Multi-Parametric Cell Energy Budget Platform 313
6. Warm up mineral oil at 37 C.
7. Replace RM in OCR assay wells with 100 μL of RM containing
1probe to all assay wells.
8. Add 2–10 μL of drug stock solutions to designated wells
according to plate map (Fig. 2).
9. Gently add to all OCR assay wells 150 μL of mineral oil pre-
warmed at 37 C.
10. Quickly insert the plate in the TR-F reader, start kinetic OCR
measurements for 60–90 min to generate profiles of cell respi-
ration (resemble Fig. 3a).
11. When the measurements are completed, take out the plate.
3.2.2 Conduction
of Measurements
Using Animal Tissue
Measurements were conducted using settings described in Sect.
3.2.1. All protocols involving animals were approved by the Uni-
versity College Cork Animal Experimentation Ethics Committee;
experiments were conducted under license from the Irish Govern-
ment in accordance with national and EU legislation (European
Directive 2010/63/EU). Age- and weight-matched adult male
C576Bl/J mice (Charles River Laboratories, UK) were exposed
to chronic intermittent hypoxia (CIH, 14 days), comprising alter-
nating periods of hypoxia (90 s; 5–6.5% O
2
) and normoxia (210 s,
21% O
2
) for 12 cycles/h, 8 h/day (N¼6), as described previously
[14]. In parallel, sham animals (N¼6) were exposed to normoxia.
Animals were anesthetized by 5% isoflurane inhalation in O
2
and
euthanized by cervical dislocation followed by immediate dissec-
tion of tissue samples. Three fragments of diaphragm muscle were
excised from each animal.
1. Quickly wash tissues with Krebs buffer containing 10 mM
glucose and 2 mM glutamine.
2. Carefully remove an excess of the washing medium.
3. Weigh fragments of the tissue and place into wells of 96-well
plates (Fig. 5a).
4. Add 100 μL of OCR medium containing 200 nM MitoX-
press
®
-Xtra dye in each well.
5. Quickly seal all wells, including “no tissue” controls, with
150 μL of prewarmed mineral oil.
6. Repeatedly measure fluorescence in TR-F mode at 37 C for
~60 min (see Note 12), take out the plate.
3.3 ATP Analysis Normally, cells tend to maintain steady ATP levels, if permitted by
spare capacity of their energy fluxes. CellTiter-Glo
®
Assay provides
end-point measurement of total cellular ATP levels, which reflect
cell energy status and viability. The assay does not show actual rates
of ATP production and consumption, but demonstrates steady-
314 Dmitri B. Papkovsky and Alexander V. Zhdanov
state ATP levels at the time of cell lysis. Easy to perform, CellTiter-
Glo
®
Assay is the first choice for pilot examination of potential
deficiencies in the Krebs cycle, OXPHOS, and glycolytic fluxes
[10](see Note 13). For this, cellular ATP dynamics should be
analyzed in RM and RM-gal media, with or without OXPHOS
inhibition (e.g., with AntA), as exemplified in Fig. 2a (see Note 14).
1. Seed and differentiate PC12 cells as described in Sect. 3.1,
steps 1–8.
2. Incubate the cells with metabolic effectors according to the
plan of the experiment in 100 μL of the medium. See plate map
in Fig. 2b as an example.
3. Prepare CellTiter-Glo
®
reagent as per manufacturer’s protocol.
4. Add 100 μL of this reagent to each ATP assay well (200 μL
total volume) (see Note 15).
5. Shake the plate intensively for 2 min.
6. Transfer samples into wells of a white 96-well plates (Greiner
Bio One).
7. Switch on the reader, select ATP assay method (Table 1) and
program it to measure designated wells on the plate.
8. Insert the plate and measure chemiluminescence in designated
wells (end-point, one scan).
9. Take out the plate when finished.
3.4 Total Protein
(Biomass) Analysis
1. Prepare a set of BSA protein standards (from BCA kit) in
Eppendorf tubes as per manufacturer’s instructions using cell
lysis buffer (see Note 16).
2. Add cell lysis buffer to designated wells (20–25 μL/well, 2–6
wells, Fig. 2) and incubate the plate on ice for 15 min. Collect
cell lysates in 1.5 mL Eppendorf tubes, pooling repeats for the
same condition.
3. Centrifuge the lysates at 14,000 gfor 10 min.
4. Dispense 25 μL aliquots of each lysate into wells of a clear
96-well plate in duplicates.
5. Dispense protein standards from BCA kit (0–2.0 mg/mL
range) to designated wells on the plate.
6. Mix reagents A and B from BCA kit, add 200 μL aliquots to all
samples. Mix well and incubate the plate at 37 C for 30 min.
7. Switch on the reader, select protein assay method (Table 1) and
program it to measure BCA assay wells.
8. Insert the BCA plate and measure absorbance in each assay well
(one scan).
9. Remove the plate when finished and discard it.
Multi-Parametric Cell Energy Budget Platform 315
According to this protocol, a complete CEB analysis with one
cell line and several different conditions/treatments can be per-
formed in approximately 5–6 h, including plate preparation, pre-
incubation in CO
2
-free conditions, drug treatment, measurement
of L-ECA, T-ECA, OCR, ATP, and biomass.
3.5 Data Processing
and CEB Analysis
Effects of different conditions and drug treatments on cell bioen-
ergetics can be assessed qualitatively by examining raw signal pro-
files of the pH-Xtra™and MitoXpress-Xtra
®
probes, which reflect
relative changes in ECA and OCR levels. However, correct and
quantitative comparison of different cell types or complex condi-
tions requires calculation of ECA, OCR, and ATP values. Normali-
zation of these values for total protein content accounts for the
differences in cell morphology and numbers, especially upon pro-
longed manipulation or treatment of cells [5]. Once the L-ECA,
T-ECA, OCR, and ATP are calculated, the differences or changes in
glycolytic activity, respiration, and CO
2
turnover can be
determined.
1. Take the ECA assay data file and convert its TR-F intensity
readings (see profiles in Fig. 3a) into LT values by applying the
following transformation [6](see Note 17):
LT ¼t2t1
ðÞ=ln F1=F2
ðÞ,
where F
1
and F
2
are pairs of TR-F intensity signals at delay
times t
1
and t
2
.
2. Plot LT profiles for all sample wells (Fig. 3b).
3. Produce average LT profile for each cell type/condition.
4. Take “average LT drift” profile (“no cells” wells—Fig. 3c) and
subtract it from all sample wells. This will produce corrected LF
profiles (Fig. 3d)(see Note 18).
5. Transform corrected LT profiles into pH profiles (Fig. 3e)
using the following equation:
pH ¼1893:4‐LTðÞ=227:54:
Alternatively, conversion into [H
+
] scale can be made
([H
+
]¼10
pH
) (Fig. 3e).
6. Select linear initial parts on the resulting pH/[H
+
] profiles
(here—20–50 min, Fig. 3f) and calculate their slopes (ΔpH/
Δtor Δ[H
+
]/Δtsee Fig. 3g). This gives T-ECA and L-ECA
values.
7. Take the OCR assay data file and process it according to steps
1–4above.
8. Convert the resulting LT profiles for all sample wells into O
2
concentration profiles, using the following transformation:
O2
½¼4455:46 eLT=7:48284:
316 Dmitri B. Papkovsky and Alexander V. Zhdanov
9. Select linear parts on these respiration profiles (resemble
Fig. 3e) and calculate their slopes (ΔO
2
/Δt), which give aver-
age OCR values.
10. Take the ATP assay data file and calculate mean ATP signals for
each condition or sample type (Fig. 2).
11. Take the total protein assay data file and generate a calibration
curve from absorbance values for protein standards.
12. Calculate protein content for each sample type or condition
used (Fig. 2).
13. Normalize the L-ECA, T-ECA, OCR, and ATP values for
corresponding total protein content at each condition (Fig. 4).
14. Calculate the contribution of CO
2
to the acidification (i.e.,
Krebs cycle activity), e.g., as (T-ECA L-ECA) or
(T-ECA L-ECA)/T-ECA 100% (Fig. 4b).
15. Assess spare capacity of glycolysis, Krebs cycle, and OXPHOS
[5] by comparing L-ECA, T-ECA, and OCR values for the
resting and FCCP/OM- or AntA-treated cells (Fig. 4b)(see
Note 19).
16. Finally, calculate OCR/L-ECA ratios (Fig. 4b), which reflect
relative contribution of glycolysis and OXPHOS to ATP
production [5].
3.6 Examples
and Interpretation
of CEB Data
1. The initial demonstration of CEB analysis is with dPC12 cells
grown on glucose and on glucose/glutamine/pyruvate, which
gave the following L-ECA values: 0.009 and 0.003 pH/min
(Fig. 3g), or after normalization for total protein (20 μg in all
sample wells)—0.45 and 0.15 pH/min/mg protein (Fig. 4a).
Their OCRs were 1.7 and 4.2 nmol/min/mg protein, and
ATP—0.85 and 1.0 a.u. (Fig. 4b), respectively. Cells supplied
with glucose alone showed increased glycolysis, which com-
pensates for low OXPHOS flux in the absence of glutamine.
Cells grown on complete medium produced significantly more
CO
2
than on glucose (Fig. 4b). The OCR/L-ECA ratio
reports on shifts in the glycolytic and mitochondrial ATP pro-
duction for different cells or treatments.
The effects of mitochondrial uncoupling and inhibition on
T-ECA, L-ECA, and OCR are shown in Fig. 4c, d, exemplify-
ing spare glycolytic and respiratory capacity. Treatments with
FCCP/OM and AntA both elevate ECA (Fig. 4c), but the
inhibition of complex III increases L-ECA to a much higher
degree than the uncoupling by two major reasons. First, by
inhibiting the electron transport chain, AntA massively
decrease amounts of pyruvate oxidized by mitochondria;
instead, pyruvate is converted to lactate, which, upon extrusion
from cells, contributes to ECA. Second, in the presence of
AntA, the F1Fo ATP synthase becomes one of the main
Multi-Parametric Cell Energy Budget Platform 317
consumers of cellular ATP [15], thus further accelerating gly-
colytic flux. In contrast, FCCP/OM double-treatment acti-
vates pyruvate consumption and prevents the reversal of F1Fo
ATP synthase; as a result, the rate of pyruvate-to-lactate con-
version cannot reach maximal levels. On the other hand,
T-ECA rates for FCCP/OM- and AntA-treated cells are quite
0
0.5E-08
1.5E-08
2.5E-08
3.5E-08
4.5E-08
Mock OM AntA
ECA in COX-deficient HCT116 cells
(H
+
×min
-1
×mg
-1
protein)
E
p = 0.01
∆
CO2
Glc / Gln / Pyr
L-ECA
T-ECA
Glc
Pyr
Mock
0
0.1
0.2
0.3
0.4
0.5
0.6
L-ECA T-ECA
Resting
AntA
pHΔ
L
FCCP
Mt
Lactate
Pyr
FCCP AntA
Glc
ΔpH
CO2
∆pH
CO2
ECA (pH×min
-1
×mg
-1
protein)
CO
2
OCR (nmole×min
-1
×mg
-1
protein)
0
2
4
6
8
10
pH
CD
Resting
AntA
FCCP
Glc
0.2
0.4
0.6
ECA (pH×min
-1
×mg
-1
protein)
Glc/Gln/Pyr
0
0.1
0.3
0.5
p = 0.0008
AB
OCR (nmole×min
-1
×mg protein)
0
1
2
3
4
5
Normalised ATP (a.u.)
0
0.2
0.4
0.6
0.8
1
1.2
0.05
0.1
0.15
∆ECA, T-ECA-L-ECA
(pH×min
-1
×mg
-1
protein)
0
0.2
Normalised
OCR / L-ECA ratio (a.u.)
0
0.2
0.4
0.6
0.8
1
1.2
0.13
Glc
Glc/Gln/Pyr
L-ECA
T-ECA
p = 0.007
p = 0.04
p = 0.02
p = 0.014 p = 0.026
Fig. 4 Advanced CEB analysis in PC12 (a–d) and COX-deficient HCT116 cells (e). (a) The difference between
T-ECA and L-ECA rates reflects the contribution of CO
2
release to acidification; increased L-ECA values in
Gln/Pyr free conditions show the activation of glycolytic flux. (b) Upon Gln/Pyr deprivation OCR decreases,
although ATP levels remain practically unchanged thanks to activated glycolysis. As (a) suggests, the
difference between T-ECA and L-ECA (ΔECA) in the medium with Glc/Gln/Pyr is significantly higher than in
the medium supplemented with glucose only. The decrease in OCR-to-L-ECA ratio in the medium containing
no Gln or Pyr also indicates a shift towards glycolytic ATP production. (c) Upon mitochondrial inhibition (5 μM
AntA), cells convert more Pyr to lactate thus increasing the ECA. The difference in lactate production between
cells with inhibited and activated respiration is shown as ΔpH
L
.ΔpH
CO2
demonstrates the contribution of CO
2
to T-ECA in cells with uncoupled mitochondria (1 μM FCCP/10 μM OM). Upon AntA treatment, ΔpH
CO2
is
minor. (d) OCR analysis reveals activation and inhibition of respiration in cells treated with FCCP/OM and AntA,
respectively. (e) Effects of the metabolites and mitochondrial modulators on ECA in COX-deficient HCT116
cells. ECA rate analysis. Media containing different combinations of Glc (10 mM), Pyr (1 mM), and Gln (2 mM)
were added to the cells 1 h prior to the analysis. Oligomycin (OM, 10 μM) and AntA (5 μM) were added
immediately before the analysis. ΔCO
2
demonstrates the contribution of CO
2
in T-ECA (shown for Mock
control only). N¼4 for all experimen ts ( p-values are presented, t-test)
318 Dmitri B. Papkovsky and Alexander V. Zhdanov
similar because in uncoupled cells a decrease in lactate extru-
sion is compensated by an increase in CO
2
release by NADH-
producing enzymes of the Krebs cycle. For the same cells and
conditions, OCR analysis showed drastic decrease in respira-
tion upon inhibition and increase upon uncoupling of mito-
chondria (Fig. 4d).
2. For some tissue models, e.g., cells with mitochondrial malfunc-
tions, such as fumarate hydratase or cytochrome c oxidase
(COX) deficiency, OCR assay may not be very informative
[10,16,17], while ATP and ECA assays remain essential
tools for studying such cells.
We found that in SCO2
/
(COX-deficient) colonic carci-
noma HCT116 cells, F1Fo ATP synthase maintains mitochon-
drial polarization (ΔΨm and ΔpH) by hydrolyzing ATP
generated by glycolysis and substrate level phosphorylation
[15]. Interestingly, total amounts of ATP in WT and
COX-deficient cells were identical, showing that very efficient
mechanisms control ATP levels even in the absence of
OXPHOS. In COX-deficient cells, ATP pool was rapidly
depleted when no glucose was available, while glutamine with-
drawal, as well as mitochondrial uncoupling or inhibition, had
negligible effect on ATP levels.
More informative was ECA assay, which reports on both
glycolytic rates and the balance of (de)carboxylation reactions.
As expected, treatment of COX-deficient cells with OM
decreased L-ECA rate. Indeed, the glycolytic flux was supposed
to decrease because it did not have to supply the inhibited F1Fo
ATPase with ATP. More intriguing were the substantial
increase in L-ECA and the decrease in total CO
2
production
in cells treated with AntA (Fig. 4e). We proposed that complex
III partially retains its activity, and residual electron and H
+
fluxes help maintain the mitochondrial polarization, as con-
firmed by the changes in ROS and ΔΨm[15].
Similar to cells with normal glycolysis and OXPHOS,
COX-deficient cells slightly increased their L-ECA rate in glu-
tamine () conditions, suggesting that their glycolysis com-
pensates for the loss of an additional source of ATP (Fig. 4e).
With OXPHOS inhibited, the observed effect was attributed to
the inhibition of the SLP flux, which is largely driven by gluta-
mine-to-α-KG conversion [18].
3. CEB platform was originally designed for cell models; however,
it is also applicable for the analysis of animal tissues.
Recently, we applied OCR analysis to confirm our finding
that colonic epitheliocytes decrease O
2
consumption in the
rodent model of colitis [19], but large data variability did not
allow OCR to reveal significant changes in respiration in colitis
samples in standard statistical tests. We realized that unknown
Multi-Parametric Cell Energy Budget Platform 319
“dead” volumes inside the colon contribute to the total weight
of tissue samples and impede accurate calculations. We
hypothesized that for other tissues, for which the actual weight
can be measured more accurately, data variability should be
lower. Indeed, the OCR analysis of the effects of CIH on
mouse diaphragm tissue was seen to work well. Individual
fragments of the diaphragm ranged 5–8 mg showed almost
linear relationship between their OCR and weight. Fragments
outside this range usually produced outliers, and therefore
were excluded from statistical analysis. We comprehended
that for statistically accurate comparison of the two groups,
changes in corrected LT values (see Fig. 3c) were more appro-
priate than OCR values. Indeed, due to the hyperbolic analyti-
cal relationship between probe LT and O
2
concentration [20],
LT values had a lower “noise” during the first 25 min of the
experiment, from which the slopes were calculated (Fig. 5b).
We found that the increase in LT values (μs/min/mg of tissue)
was more pronounced for the sham group (Fig. 5c). A negative
trend in CIH group suggested a moderate inhibition of the
OCR in hypoxic diaphragm ( p¼0.11, t-test). This result
aligns well with the finding that CIH induces diaphragm mus-
cle fatigue in rats [21]. We believe that the observed decrease in
tissue respiration is due to oxidative stress-induced inhibition
of mitochondrial function and mitophagy, which both have
been shown for hypoxic diaphragm in a similar mouse
model [22].
0
0.02
0.04
0.06
0.08
0.1
Sham CIH
0
5
10
15
20
25
30
35
40
45
0 5 10 15 20 25 30 35
10.8 mg
7.2 mg
3.2 mg
ABC
LT (μs)
Time (min)
Normalised changes in LT
(μs × min
-1
× mg
-1
)
p = 0.11
abc
1
2
3
4
5
Fig. 5 Analysis of OCR by mouse diaphragm tissue samples. (a) A photograph of tissue samples in 96-well
plate: 1–5 are different animals, a–c are replicates for each animal. (b) Corrected lifetime (LT) profiles for
diaphragm fragments of different size (10.8, 7.2, and 3.2 mg), excised from one sham animal. Dashed line at
25 min shows LT values used for OCR calculation. (c) Quantitative analysis of the effect of chronic intermittent
hypoxia (CIH) on respiration in mouse diaphragm (N¼6 for both sham and CIH groups). Negative trend in LT
values in CIH group is shown ( p¼0.11, t-test). Each data point represents an average of 2–3 technical
replicates (different fragments of the same diaphragm)
320 Dmitri B. Papkovsky and Alexander V. Zhdanov
3.7 Conclusions Overall, the multiparametric CEB platform allows high-
throughput quantitative assessment of basal activity and spare
capacity of the main metabolic pathways involved in cell bioener-
getics, providing detailed information about their changes and
underlying regulatory mechanisms.
4 Notes
1. When choosing an instrument for CEB, ensure that it is spec-
trally compatible with the pH and O
2
probes, provides suffi-
cient sensitivity (signal-to-blank ratio, time resolution),
scanning speed, and uniform temperature control of the
plate, and supports the RLD mode with two TR-F intensity
readings at different delay times [7]. Some readers allow direct
calculation of probe LT by the software (FLUOstar Omega,
Synergy H1) or in Excel from the intensity profiles (Victor2,
4, 5). Victor2 4, 5 reader Europium filter set (340 nm/
615 nm) and Samarium filter set (340 nm/642 nm) are ideal
for pH-Xtra™and MitoXpress
®
-Xtra probes, respectively.
Luminescence analysis is required for ATP measurement;
absorbance at 560, 595, and 750 nm is used for total protein
measurement using BCA, Bradford, and Lowry methods,
respectively.
2. Many common cell lines and primary cells are suitable for CEB
analysis. In PC12 cells both glycolysis and OCR are very active.
However, for certain cells types (e.g., highly glycolytic), OCR
measurements can be problematic, as illustrated by the Exam-
ple 2 (Sect. 3.6) using COX-deficient HCT116 cells.
3. Perform all operations with cells under the laminar flow hood.
For cell culture, all reagent solutions, unless supplied sterile,
should be autoclaved (with high pressure saturated steam at
121 C for 20 min) or filtered through sterile filter (0.22 μm).
4. NGF is used for PC12 cell differentiation, it should be stored at
20 C and added to RM or PM immediately prior to applica-
tion to the cells. For many cell lines, NGF is not required;
however, other media, combination of substrates, growth and
differentiation factors can be used.
5. High-viscosity mineral oil acts as a barrier for diffusion of gases
(O
2
,CO
2
) to and from the sample. Partial penetration still
occurs through plastic body of the plate and oil layer.
6. Optimal cell numbers depend on the cell type, size, metabolic
activity, and experimental conditions. Preliminary experiments
can be conducted to optimize seeding density for ECA and
OCR assays.
Multi-Parametric Cell Energy Budget Platform 321
7. This may not be necessary for other cell types.
8. Collagen coating improves attachment of PC12 cells to plastic
surface (optional for adherent cells). Normally, sterile stock
solution of collagen IV (0.1%) in 0.25% acetic acid is prepared
and used at 0.01% in 0.1% acetic acid (12–24 h incubation
at RT).
9. While replacing the media or washing the cells, make sure that
cells are not washed away from the wells. Automatic pipette
(P200 or P1000) is preferred over vacuum pump or multi-well
dispenser.
10. Drug solutions can be prepared as concentrated stocks
(10–50). After drug addition, the total volume of medium
should be equal in all samples.
11. At this stage, cells can be pretreated (for 30 min to 3 h) with
different drugs, media (e.g., glucose, glutamine, or pyruvate
deprivation).
12. The best reproducibility and accuracy are achieved when the
total number of tissue fragments analyzed simultaneously does
not exceed nine (three animals).
13. Using kinetic ATP analysis, OXPHOS flux can be probed using
RM or other medium, in which glucose is substituted with an
equivalent amount of galactose (RM-gal). Glycolytic flux can
be examined using RM supplemented with OXPHOS inhibi-
tors. Tis assay can be performed with or without FBS.
14. An end-point (e.g., 3 h) or kinetic ATP analysis in RM-gal
medium can also be used to assess mitochondrial toxicity of
new pharmaceutical entities [23].
15. For quantitative analysis, ATP standards should be included to
generate a calibration curve.
16. Cell lysis buffer contains 150 mM NaCl, 50 mM HEPES
(pH ¼7.5), 1 mM EDTA, and 1% IGEPAL
®
CA-630. Alter-
native buffers (e.g., RIPA buffer compatible with BCA kit) or
protein assays (e.g., Bradford or Lowry method with absor-
bance measurements at 595 and 750 nm, respectively) can also
be used.
17. FLUOstar Omega reader data analysis software (MARS) pro-
vides automatic processing of raw fluorescent data, back-
ground correction, calculation of ECA/OCR values, and
graphical representation. For PerkinElmer readers data proces-
sing is performed manually, using standard MS Excel templates
for ECA and OCR assays (e.g., Agilent website).
18. Proper correction should produce flat LT profiles for “no cell”
samples with pH close to 7.4 (see Fig. 3c, d).
322 Dmitri B. Papkovsky and Alexander V. Zhdanov
19. AntA is a potent inhibitor of mitochondrial complex III (cyto-
chrome c reductase); FCCP is a mitochondrial uncoupler,
capable of efficient transferring H
+
across the inner mitochon-
drial membrane; OM is an inhibitor of both direct and reversed
activities of mitochondrial complex V (F1Fo ATP synthase).
Acknowledgments
Support of this work by the Science Foundation Ireland, grant
12/RC/2276_P2, is gratefully acknowledged.
References
1. Beis I, Newsholme EA (1975) The contents of
adenine nucleotides, phosphagens and some
glycolytic intermediates in resting muscles
from vertebrates and invertebrates. Biochem J
152:23–32
2. Lin MT, Beal MF (2006) Mitochondrial dys-
function and oxidative stress in neurodegener-
ative diseases. Nature 443:787–795
3. Hynes J, Natoli E, Will Y (2009) Fluorescent
pH and oxygen probes of the assessment of
mitochondrial toxicity in isolated mitochondria
and whole cells. Curr Protoc Toxicol 2
(16):11–22
4. Ferrick DA, Neilson A, Beeson C (2008)
Advances in measuring cellular bioenergetics
using extracellular flux. Drug Discov Today
13:268–274
5. Zhdanov AV, Favre C, O’Flaherty L, Adam J,
O’Connor R, Pollard PJ, Papkovsky DB
(2011) Comparative bioenergetic assessment
of transformed cells using a cell energy budget
platform. Integr Biol (Camb) 3:1135–1142
6. Ballew RM, Demas J (1989) An error analysis
of the rapid lifetime determination method for
the evaluation of single exponential decays.
Anal Chem 61:30–33
7. Favre C, Zhdanov A, Leahy M, Papkovsky D,
O’Connor R (2010) Mitochondrial pyrimidine
nucleotide carrier (PNC1) regulates mitochon-
drial biogenesis and the invasive phenotype of
cancer cells. Oncogene 29:3964–3976
8. Zhdanov AV, Dmitriev RI, Papkovsky DB
(2012) Bafilomycin A1 activates
HIF-dependent signalling in human colon can-
cer cells via mitochondrial uncoupling. Biosci
Rep 32:587–595
9. Zhdanov AV, Dmitriev RI, Papkovsky DB
(2011) Bafilomycin A1 activates respiration of
neuronal cells via uncoupling associated with
flickering depolarization of mitochondria. Cell
Mol Life Sci 68:903–917
10. O’Flaherty L, Adam J, Heather LC, Zhdanov
AV, Chung Y-L, Miranda MX, Croft J, Olpin S,
Clarke K, Pugh CW, Griffiths J, Papkovsky D,
Ashrafian H, Ratcliffe PJ, Pollard PJ (2010)
Dysregulation of hypoxia pathways in fumarate
hydratase-deficient cells is independent of
defective mitochondrial metabolism. Hum
Mol Genet 19:3844–3851
11. Hynes J, O’Riordan TC, Zhdanov AV, Uray G,
Will Y, Papkovsky DB (2009) In vitro analysis
of cell metabolism using a long-decay pH-sen-
sitive lanthanide probe and extracellular acidifi-
cation assay. Anal Biochem 390:21–28
12. Hynes J, Floyd S, Soini AE, O’Connor R, Pap-
kovsky DB (2003) Fluorescence-based cell via-
bility screening assays using water-soluble
oxygen probes. J Biomol Screen 8:264–272
13. Zhdanov AV, Ogurtsov VI, Taylor CT, Pap-
kovsky DB (2010) Monitoring of cell oxygena-
tion and responses to metabolic stimulation by
intracellular oxygen sensing technique. Integr
Biol (Camb) 2:443–451
14. Lucking EF, O’Halloran KD, Jones JF (2014)
Increased cardiac output contributes to the
development of chronic intermittent hypoxia-
induced hypertension. Exp Physiol
99:1312–1324
15. Zhdanov AV, Andreev DE, Baranov PV, Pap-
kovsky DB (2017) Low energy costs of F1Fo
ATP synthase reversal in colon carcinoma cells
deficient in mitochondrial complex IV. Free
Radic Biol Med 106:184–195
16. Sung HJ, Ma W, Wang P-y, Hynes J, O’Rior-
dan TC, Combs CA, McCoy JP, Bunz F, Kang
J-G, Hwang PM (2010) Mitochondrial respi-
ration protects against oxygen-associated DNA
damage. Nat Commun 1:5
Multi-Parametric Cell Energy Budget Platform 323
17. Matoba S, Kang J-G, Patino WD, Wragg A,
Boehm M, Gavrilova O, Hurley PJ, Bunz F,
Hwang PM (2006) p53 regulates mitochon-
drial respiration. Science 312:1650–1653
18. Kiss G, Konrad C, Doczi J, Starkov AA,
Kawamata H, Manfredi G, Zhang SF, Gibson
GE, Beal MF, Adam-Vizi V (2013) The nega-
tive impact of α-ketoglutarate dehydrogenase
complex deficiency on matrix substrate-level
phosphorylation. FASEB J 27:2392–2406
19. Zhdanov AV, Okkelman IA, Golubeva AV,
Doerr B, Hyland NP, Melgar S, Shanahan F,
Cryan JF, Papkovsky DB (2017) Quantitative
analysis of mucosal oxygenation using ex vivo
imaging of healthy and inflamed mammalian
colon tissue. Cell Mol Life Sci 74:141–151
20. Dmitriev RI, Papkovsky DB (2012) Optical
probes and techniques for O2 measurement
in live cells and tissue. Cell Mol Life Sci
69:2025–2039
21. Shortt CM, Fredsted A, Chow HB, Williams R,
Skelly JR, Edge D, Bradford A, O’Halloran KD
(2014) Reactive oxygen species mediated dia-
phragm fatigue in a rat model of chronic inter-
mittent hypoxia. Exp Physiol 99:688–700
22. Giordano C, Lemaire C, Li T, Kimoff RJ, Pet-
rof BJ (2015) Autophagy-associated atrophy
and metabolic remodeling of the mouse dia-
phragm after short-term intermittent hypoxia.
PLoS One 10:e0131068
23. Rana P, Aleo MD, Gosink M, Will Y (2019)
Evaluation of in vitro mitochondrial toxicity
assays and physicochemical properties for pre-
diction of organ toxicity using 228 pharmaceu-
tical drugs. Chem Res Toxicol 32:156–167
324 Dmitri B. Papkovsky and Alexander V. Zhdanov
Chapter 24
Fluorescence-Based Assay For Measuring OMA1 Activity
Julia Tobacyk and Lee Ann MacMillan-Crow
Abstract
Mitochondrial fusion depends on proteolytic processing of the dynamin-related GTPase protein, OPA1,
which is regulated by the mitochondrial zinc metalloproteinase, OMA1. Last year we published a report
describing a novel approach to directly measure the enzymatic activity of OMA1 in whole cell lysates. This
fluorescence-based reporter assay utilizes an eight amino acid peptide sequence referred to as the S1
cleavage site where OMA1 cleaves within OPA1 and is flanked by a fluorophore and quencher. In this
chapter, we provide additional insight into the OMA1 activity assay.
Key words OMA1, Mitochondria, Protease, Fusion, Fluorescence-based reporter assay
1 Introduction
Mitochondria are dynamic organelles that continually undergo
fission and fusion to maintain normal mitochondrial health [1–
3]. Fusion is regulated by a dynamin-related GTPase called optic
atrophy protein (OPA1), which is involved in fusion of the inner
mitochondrial membrane. OPA1 exists in different splice isoforms
that can be proteolytically processed at distinct sites (S1 and S2) [4]
and its proteolytic cleavage results in the long (L-OPA1) and short
(S-OPA1) forms of OPA1. The balance between the L-OPA1 and
S-OPA1 forms of OPA1 is regulated by two metalloproteases called
OMA1 and YME1L [5], which cleave OPA1 at S1 and S2 sites,
respectively. YME1L constitutively cleaves OPA1 leading to a bal-
anced accumulation of L- and S-OPA1 protein forms, which sup-
ports normal mitochondrial fusion [6]. OMA1 is a stress-induced
protease, and different stress stimuli such as ROS production, heat
shock, and ATP depletion have been shown to activate OMA1
[5,7–9].
Understanding the precise mechanistic and functional roles of
OMA1 has been difficult, since there is no crystal structure or
specific OMA1 inhibitors available. Recently, we have published a
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_24,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
325
report describing a sensitive method to measure OMA1 enzymatic
activity for the first time [10]. Many researchers measure the L- and
S-OPA1 via immunoblotting to indirectly measure OMA1 activity.
A drawback using this technique is that it does not take into
account additional substrates that are involved in the proteolytic
processing of OPA1 including YME1L [5,11,12]. Researchers
have also used a nonspecific metal chelator, o-phenanthroline, to
validate the role for OMA1 in various cell models [11]; however,
both YME1L and OMA1 are effectively inhibited by
o-phenanthroline [11]. To address the specific role of OMA1,
researchers have also relied on OMA1 knockout models [13–
15]. Several studies show that knocking out OMA1 in vivo has
protective effects in renal injury [15], heart failure [14], and neu-
rodegeneration [16] suggesting that OMA1 might be a therapeutic
target for many diseases. Use of OMA1 knockout models in vitro
and in vivo has verified the importance of OMA1 during disease
and its regulatory role in mitochondrial quality control. Although
these genetic techniques and models offer new opportunities to
study the mechanism of OMA1, there are some potential limita-
tions. For example, unknown compensatory or redundancy
mechanisms might be activated during genetic manipulations and
can complicate the interpretation in characterizing the precise role
of OMA1. The newly described direct, real-time OMA1 activity
assay will be an important tool to further characterize this mito-
chondrial protease. Additionally, we anticipate that this high-
throughput assay could help to identify specific pharmacological
agents that will inhibit and/or activate OMA1.
During the development of our OMA1 activity assay, we used
cellular knockout models as tools to validate OMA1 specificity.
Using both mouse embryonic fibroblast OMA1 knockout cell
line and transient OMA1 siRNA knockdown models, we detected
a ~50% reduction in OMA1 activity using our assay [10]. Clearly,
further development is needed since these results suggest that the
remaining 50% of the activity is most likely due to other proteases
cleaving the fluorogenic reporter peptide. It is important to note
that YME1L knockdown did not alter OMA1 activity revealing that
the nonspecific component is not due to YME1L [10]. We are
currently addressing these issues by optimizing the fluorogenic
reporter peptide and evaluating whether lysing procedures or addi-
tion of specific protease inhibitors might reduce the nonspecific
component. In this chapter, we provide a detailed description of
the OMA1 activity assay. The goal of this chapter is to provide
readers with technical considerations for efficient assay design and
its usage across multiple types of samples and under different
conditions.
326 Julia Tobacyk and Lee Ann MacMillan-Crow
2 Materials
Prepare all solutions using ultrapure water and analytical grade
reagents. Prepare and store all reagents at 4 C (unless indicated
otherwise).
2.1 Fluorogenic
Reporter Peptide
(AFRATDHG): Serves
as the Substrate
for the OMA1 Assay
The OMA1 assay utilizes a fluorogenic eight amino acid peptide
that incorporates the OPA1 S1 cleavage site where OMA1 cleaves
[17] (Fig. 1a). This peptide includes an MCA (7-Methoxycou-
marin-4-ylacetyl) fluorophore on the amino end and the fluores-
cence quencher DNP (2,4-Dinitrophenyl) on the carboxy
terminus. When the peptide is cleaved by OMA1, the DNP
quencher is released resulting in MCA fluorescence that is
measured spectrofluorometrically (excitation/emission of
325/392) using a plate reader (Fig. 1b). We custom order this
Fig. 1 The methodology behind the OMA1 activity assay. (a). OMA1, a stress-
induced mitochondrial protease, cleaves OPA1 at the S1 site between arginine
(194) and alanine (195) resulting in a “long” L-OPA1 and a “short” S-OPA1. (b). A
custom-made peptide (AFRATDHG) was synthesized based on the rat OMA1 S1
cleavage site within OPA1. OMA1 cleavage removes the quenching of DNP,
which results in increased MCA fluorescence. OMM outer mitochondrial mem-
brane, IMS inner mitochondrial space, IMM inner mitochondrial membrane, S1
site 1, S-OPA1 short OPA1), L-OPA1 long OPA1, MCA 7-Methoxycoumarin-4-
ylacetyl, DNP 2,4-Dinitrophenyl, Lys lysine
OMA1 Activity Assay 327
peptide from LifeTein LLC. Upon arrival, the lyophilized peptide is
resuspended in DMSO to a final concentration of 1 mM and stored
in 50 μL aliquots at 80 C. We have shown that a working
concentration of 5 μM produces a linear response [10]. Protect
from light when handling the fluorogenic reporter peptide.
2.2 Protein Sample/
Unknown
The OMA1 assay can utilize protein from various sources, and it is
important to optimize sample processing and quantity used in the
assay within each laboratory (see Note 1). For example, in our labo-
ratory, we have shown sufficient OMA1 activity using 5 μgofprotein
isolated using a RIPA lysis method and renal cells (see Note 1).
2.3 TPEN (Used as an
OMA1 Inhibitor)
50 mM master stock solution of N,N,N0,N0-Tetrakis
(2-pyridylmethyl)ethylenediamine (TPEN): Add 10.67 mg of
TPEN and dissolve in 0.5 mL of 100% ethanol (see Note 2).
2.4 OMA1 Buffer OMA1 assay buffer: 50 mM Tris–HCl, 40 mM KCl. Add 15 mL of
100 mM Tris–HCl and 1.2 mL 1 mM of KCl. Make up to 30 mL
with water. Store at 4 C. This buffer is good for long-term use.
2.5 96-Well
Microplate
Use a black 96-well, solid bottom, non-treated, polystyrene,
opaque microplate (costar) with no lid for the OMA1 activity
assay (product number: 3915).
2.6 Plate Reader Use a fluorescent plate reader available at your institution. Since
MCA has been chosen as the fluorophore, be sure the plate reader
can read fluorescence at excitation/emission of 325/392 nm or if
the plate reader is filter based, you can select 320/405 nm settings.
3 Methods
Carry out all procedures on ice unless otherwise specified.
3.1 Procedure 1. Before starting the OMA1 activity assay, measure the protein
concentration of your samples (see Note 1). Preheat the plate
reader to 37 C prior to reading.
2. Set up a template by arranging samples in a 96-well plate
(Fig. 2). If you are adding additional treatments that involve
adding a new compound into the assay (e.g., screening for
drugs that modulate OMA1), it is important to determine
whether the compound or its solvent reacts with the fluoro-
genic reporter substrate or possesses intrinsic fluorescence (see
Note 3).
Add reagents to each well of microplate in the following
order: (1) OMA1 buffer, (2) protein samples, (3) TPEN, and
(4) to start reaction 5 μM of fluorogenic substrate peptide.
328 Julia Tobacyk and Lee Ann MacMillan-Crow
Each sample should be in triplicate. Make sure to include:
(1) Blank (OMA1 buffer only), (2) substrate alone
(AFRATDHG + OMA1 buffer), (3) samples without TPEN
(OMA1 buffer + AFRATDHG + protein) and samples with
protein (OMA1 buffer + AFRATDHG + protein + TPEN) as
well as any compound controls (Fig. 2).
3. Measure relative fluorescence at excitation/emission of
325/392 nm or if the plate reader is filter based, select
320/405 nm settings, read fluorescence every 1 min for
30 min at 37 C.
3.2 Analysis 1. At each time point, measure the average fluorescence reported
as relative fluorescence unit (RFU). Since the fluorogenic
reporter substrate has modest intrinsic RFU, it is important
to subtract this value (AFRATDHG alone) at each time point
from the sample average (Sample X) to give you the corrected
RFU for each sample.
Corrected sample RFU½¼Sample X½AFRATDHG alone½:
2. Next, to obtain the OMA1 activity for each condition, you will
utilize the TPEN values. Specifically, OMA1 activity is the
difference between samples with and without TPEN:
OMA1 activity RFU½¼corrected sample X½
corrected sample XþTPEN½:
Fig. 2 Suggested layout of 96-well black microplate for the OMA1 activity assay. In this hypothetical example,
the intent was to screen compounds for OMA1 inhibitors. For this experiment, the following conditions were
used: (1) untreated sample, (2) untreated sample + TPEN, and (3) untreated sample + Drug X (e.g., an OMA1
inhibitor). It is important to include appropriate controls that will test for possible interferences in the assay
OMA1 Activity Assay 329
3. Report final activity by calculating the rate (slope) over time
(30 min) using linear regression:
Rate RFU=min½¼
OMA1 activity½
Time :
4. For the statistical analysis, compare rates between each condi-
tion using appropriate statistical tests.
4 Notes
1. During sample preparation, protein can be harvested from
numerous sources (e.g., cells, tissues, microorganisms) using
a variety of protein isolation methods (e.g., sonication,
mechanical homogenization, use of detergent-based buffers).
In addition, each lab typically uses a combination of different
protease inhibitors during the protein isolation procedure. In
our studies, we have predominantly used radioimmunoprecipi-
tation assay (RIPA) lysis buffer supplemented with protease
inhibitors to extract protein from cell lines and rat tissues
[10]. Our preliminary studies have shown variability in
OMA1 activity across different methods of cell lysing techni-
ques. We recommend users to compare different methods of
isolating proteins and selection of protease inhibitors within
their laboratory practices. In our first report, we showed that
5μg of protein resulted in a linear increase in fluorescence over
a 30 min time frame [10]. Since OMA1 activity will likely vary
between different cell lines and tissue types, we suggest
performing a dose–response curve to determine the optimal
protein concentration needed to achieve a linear increase in
RFU over 30 min (Fig. 3. Line A). If too much protein was
included, the reaction will saturate (Fig. 3. Line B), and if too
little protein was used, the reaction will have a lag phase (Fig. 3.
Line C).
2. Since o-phenanthroline has intrinsic fluorescence that inter-
feres with the MCA fluorophore, TPEN was selected as an
OMA1 inhibitor for the OMA1 activity assay. There may be
variability in TPEN inhibition between different cell lines and
tissue types. In our first report, we used a variety of cell types
and showed that 200 μM TPEN inhibits ~90% of fluorescence
[10]. We recommend each user perform a dose–response curve
to assess the optimal working concentration of TPEN that
results in a minimum of 80% inhibition.
3. As with any fluorescence-based enzymatic assay, it is important
to note that the OMA1 assay may be subjected to interferences
resulting in false-positive and/or false-negative readings. The
sources of interference include: (1) fluorogenic reporter
330 Julia Tobacyk and Lee Ann MacMillan-Crow
substrate interacts nonspecifically with a solute (e.g., a drug
used in the assay may cleave the fluorogenic reporter peptide),
(2) a solute possesses intrinsic fluorescence and it is absorbed at
the 325/392 emission/excitation wavelength, and (3) solvent
or solute interacts with the protein sample. For the last source
of interference, we determined that the addition of DMSO
above 1% results in decreased RFU. These types of interfer-
ences can be detected by measuring the fluorescence signal
without the presence of cell lysates (Fig. 2). It is important to
include internal controls of the fluorogenic reporter peptide
with a solvent/solute/drug to determine whether there is an
interaction and include the solvent/solute/drug alone in
OMA1 assay buffer to determine if it possesses intrinsic
fluorescence.
Acknowledgments
Dr. Lee Ann MacMillan-Crow and UAMS have a financial interest
in the technology discussed in this chapter. This financial interest
has been reviewed and approved in accordance with the UAMS
conflict of interest policies.
References
1. Wai T, Langer T (2016) Mitochondrial dynam-
ics and metabolic regulation. Trends Endocri-
nol Metab 27(2):105–117
2. Anand R, Langer T, Baker MJ (2013) Proteo-
lytic control of mitochondrial function and
morphogenesis. Biochim Biophys Acta 1833
(1):195–204
3. Chen H, Chan DC (2009) Mitochondrial
dynamics--fusion, fission, movement, and
Fig. 3 Example of protein dose–response curve in OMA1 activity assay. It is
important to determine the optimal amount of protein that results in a linear
increase in RFU over the 30 min time course (Line A). Line B depicts what would
occur with too low protein was included in the assay. Line C depicts what would
occur if too much protein was added to the substrate, resulting in a saturated
reaction
OMA1 Activity Assay 331
mitophagy—in neurodegenerative diseases.
Hum Mol Genet 18(R2):R169–R176
4. Olichon A et al (2006) Mitochondrial dynam-
ics and disease, OPA1. Biochim Biophys Acta
1763(5–6):500–509
5. Anand R et al (2014) The i-AAA protease
YME1L and OMA1 cleave OPA1 to balance
mitochondrial fusion and fission. J Cell Biol
204(6):919–929
6. Ruan Y et al (2013) Loss of Yme1L perturbates
mitochondrial dynamics. Cell Death Dis 4:
e896
7. Baker MJ et al (2014) Stress-induced OMA1
activation and autocatalytic turnover regulate
OPA1-dependent mitochondrial dynamics.
EMBO J 33(6):578–593
8. Ehses S et al (2009) Regulation of OPA1 pro-
cessing and mitochondrial fusion by m-AAA
protease isoenzymes and OMA1. J Cell Biol
187(7):1023–1036
9. Head B et al (2009) Inducible proteolytic inac-
tivation of OPA1 mediated by the OMA1 pro-
tease in mammalian cells. J Cell Biol 187
(7):959–966
10. Tobacyk J et al (2019) The first direct activity
assay for the mitochondrial protease OMA1.
Mitochondrion 46:1–5
11. Rainbolt TK et al (2016) Reciprocal degrada-
tion of YME1L and OMA1 adapts
mitochondrial proteolytic activity during stress.
Cell Rep 14(9):2041–2049
12. Rainbolt TK, Saunders JM, Wiseman RL
(2015) YME1L degradation reduces mito-
chondrial proteolytic capacity during oxidative
stress. EMBO Rep 16(1):97–106
13. Quiros PM et al (2012) Loss of mitochondrial
protease OMA1 alters processing of the
GTPase OPA1 and causes obesity and defective
thermogenesis in mice. EMBO J 31
(9):2117–2133
14. Acin-Perez R et al (2018) Ablation of the stress
protease OMA1 protects against heart failure in
mice. Sci Transl Med 10(434):eaan4935
15. Xiao X et al (2014) OMA1 mediates OPA1
proteolysis and mitochondrial fragmentation
in experimental models of ischemic kidney
injury. Am J Physiol Renal Physiol 306(11):
F1318–F1326
16. Korwitz A et al (2016) Loss of OMA1 delays
neurodegeneration by preventing stress-
induced OPA1 processing in mitochondria. J
Cell Biol 212(2):157–166
17. Ishihara N et al (2006) Regulation of mito-
chondrial morphology through proteolytic
cleavage of OPA1. EMBO J 25
(13):2966–2977
332 Julia Tobacyk and Lee Ann MacMillan-Crow
Chapter 25
Studying Mitochondrial Network Formation by In Vivo and In
Vitro Reconstitution Assay
Wanqing Du, Xiangjun Di, and Qian Peter Su
Abstract
Mitochondria change their morphologies from small isolated vesicles to large continuous networks across
the cell cycles. The mitochondrial network formation (MNF) plays an important role in maintaining
mitochondrial DNA integrity and interchanging mitochondrial materials. The disruption of the mitochon-
drial network affects mitochondrial functions, such as ATP production, integration of metabolism, calcium
homeostasis, and regulation of apoptosis, leading to the abnormal development and several human diseases
including neurodegenerative disease. In this unit, we describe the method of studying MNF, which is driven
by microtubule-dependent motor protein, by in vivo imaging and single-molecule in vitro reconstitution
assays.
Key words Mitochondrial network formation (MNF), KIF5B, Single-molecule, In vitro reconstitu-
tion system
1 Introduction
Mitochondrial fusion, fission, and movement are the basic mechan-
isms for maintaining mitochondrial dynamics and morphology
across the cell cycle [1]. Our previous research found that KIF5B-
mediated mitochondrial tubulation is an additional mechanism for
regulating mitochondrial morphological dynamics [2]. We pro-
posed a modified model for mitochondrial network formation:
the dynamic tubulation of mitochondria, driven by Kinesin-1
motor along microtubules, gives rise to highly dynamic nano-
tubules, and mitofusin-mediated fusion of these tubules forms
lattices which eventually interconnect to generate the mitochon-
drial networks in cells or by an in vitro reconstitution system
[2]. Here, we summarized the protocols of studying mitochondrial
network formation (MNF) with in vivo imaging and more focused
on the single-molecule in vitro reconstitution system. The system
contains purified mitochondria as the membrane source and in vitro
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_25,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
333
polymerized microtubules together with the purified motor pro-
tein KIF5B to provide the driving force [3,4]. In addition, this
in vitro assay can also be widely applied to reconstitute the auto-
lysosome tubulation during autophagy [5], as well as other
biological processes driven by kinesin motors [6,7] along micro-
tubules and myosin motors [8] along actin filaments.
2 Materials
All the solutions and buffers used in this study should be prepared
using analytical grade reagents and dissolved in ddH
2
O, followed
by filtration with the 0.2 μm syringe filters to remove all the
impurities, which may contain autofluorescence during the total
internal reflection fluorescence (TIRF) microscopy imaging. We do
not add sodium azide (NaN
3
) to the solutions. Prepare and store all
reagents, solutions, and buffers at room temperature (unless indi-
cated otherwise). Diligently follow all waste disposal regulations
when disposing of waste materials.
2.1 Observation
of MNF in Cells
1. Culture medium: DMEM (high glucose) supplemented with
10% fetal bovine serum (FBS), penicillin, streptomycin, and
GlutaMAX-I™.
2. Phosphate-buffered saline (PBS): pH 7.4, 135 mM NaCl,
4.7 mM KCl, 10 mM Na
2
HPO
4
, 2 mM NaH
2
PO
4
.
3. Amaxa Cell Line Nucleofector II Kit.
4. Normal Rat Kidney Epithelial (NRK) cells or Mouse Embry-
onic Fibroblast (MEF) cells.
5. Plasmid: TOM20-GFP.
6. Plasmid: Mito-YFP.
7. Microscope: Spinning disk confocal microscope equipped with
488, 561, and/or 647 nm laser.
2.2 In Vitro
Reconstitution System
2.2.1 Protein Purification
1. ddH
2
O: MilliQ, Millipore, 18.2 MΩcm at 25 C.
2. Wash buffer: 20 mM Tris–HCl, pH 7.5, 500 mM NaCl,
30 mM imidazole.
3. Elution buffer: 50 mM Tris–HCl, pH 8.0, 150 mM NaCl,
250 mM imidazole.
4. Storage buffer: 50 mM HEPES-KOH, pH 7.4, 300 mM NaCl,
1 mM MgCl
2
, 10% (w/v) sucrose, 50 mM ATP (see Note 1).
Store at 4 C.
5. Sf9 cells.
6. Culture medium for insect cells.
7. Ni Sepharose 6 Fast Flow.
334 Wanqing Du et al.
2.2.2 Mitochondria
Isolation
1. Phosphate-buffered saline (PBS): pH 7.4, 135 mM NaCl,
4.7 mM KCl, 10 mM Na
2
HPO
4
, 2 mM NaH
2
PO
4
.
2. H-S buffer: 10 mM HEPES-KOH, pH 7.4, 320 mM sucrose,
5 mM MgSO
4
, 1 mM EGTA, 1 mM DTT, and protease
inhibitors (see Note 2). Store at 4 C.
3. Dounce homogenizer.
4. Optima MAX-XP ultracentrifuge.
5. 5 mL ultracentrifuge tubes.
2.2.3 In Vitro
Reconstitution Assay
1. Coverslips (No.1, thickness 170 μm, size 24 50 mm) and
slides (size 24 60 mm).
2. Double-sided tapes (see Note 3).
3. Ultrasonic cleaner.
4. Acetone.
5. 1 M KOH solution in ddH
2
O.
6. HTS tubulin powder (see Note 3). Store at 20 C before
dissolving in buffer.
7. HiLyte 647-labeled tubulin powder (see Note 3). Store at
20 C before dissolving in buffer.
8. General tubulin buffer (see Note 3). Store at 4 C.
9. Tubulin glycerol buffer (see Note 3). Store at 4 C.
10. Motility assay buffer (MAB)/BRB80: 80 mM PIPES-KOH,
pH 6.8, 1 mM MgCl
2
, 1 mM EGTA. Store at 4 C for up to
1 month.
11. 3 mg/mL casein solution in MAB. Aliquot and store at 80 C
for long-term storage; once taken out, store at 4 C for up to
1 month (see Note 4).
12. GLOX solution: 60 mg/mL glucose oxidase, 6 mg/mL cata-
lase in PBS with 40% (v/v) glycerol. Divide into aliquots of
200 μL and store at 20 C for up to 2 months; once thawed,
store at 4 C for up to 1 week (see Note 5).
13. ATP working solution (with an ATP regeneration system and
an oxygen scavenger system): 10 mM phosphocreatine (PC),
300 μg/mL creatine kinase (CK), 20 μM taxol, 10 mM DTT,
0.1 mg/mL casein, 2.5% glucose, 100GLOX, and 20 μM
ATP in MAB (see Note 6).
14. Syringe and syringe filter (0.2 μm).
15. Inverted microscope with 561 nm and 640 nm lasers, high NA
(>1.40) TIRF objective, TIRF illumination mode, EMCCD
or sCMOS as detector.
16. CM-DiI and/or MitoTracker for mitochondrial membrane
staining.
17. Dimethylformamide (DMF) or dimethyl sulfoxide (DMSO) as
solvent for dye.
Mitochondrial Network Formation 335
3 Methods
3.1 Cell Transfection
and Observation
1. Culture cells in culture medium and check the condition by
stereoscopic microscope.
2. Transfect cells with 5μg DNA (TOM20-GFP or Mito-YFP)
via an Amaxa Nucleofector II using solution T (for NRK cells)
or a MEF2 kit (for all other cells) and programs for each cell
line individually.
3. Culture the transfected cells in growth medium for further
analysis and visualize them by spinning disk microscopy.
4. Visualize the mitochondrial network formation (MNF) using a
live-cell imaging system on a confocal microscope (FV1000,
Olympus). Adjust the pinhole of the confocal microscope to
80–120 μm (Fig. 1).
3.2 In Vitro
Reconstitution System
3.2.1 Motor Protein
Purification
1. Use pFastBac DUAL as the vector for baculovirus expression.
This vector does not contain a tag for purification, therefore
introduce a histidine (6) tag before the first codon of full-
length KIF5B. Clone the histidine-tagged KIF5B coding
sequence between the BamH I/XbaI sites of the pFastBac
DUAL plasmid.
2. Express the full-length KIF5B motor protein using the Bac-to-
Bac expression system. Grow the Sf9 cells in Lonza media to a
density of ~2 10
6
cells/mL and then incubate the cells with
virus containing the full-length KIF5B construct. After 60 h,
collect and lyse the cells by freeze–thaw cycles, and centrifuge
the cells at 4 C.
Fig. 1 NRK cells expressing the mitochondrial marker Mito-YFP were visualized
by spinning disk microscopy with 515 nm laser. Scale bar, 10 μm
336 Wanqing Du et al.
3. Bond the soluble fraction in batches to Ni-NTA agarose and
wash the resin with wash buffer. Elute the His6-tagged KIF5B
with elution buffer. Concentrate the eluate and store it in
storage buffer at 80 C(see Note 7).
3.2.2 Mitochondrial
Isolation
1. Wash fresh rat liver (~2 g) or 20 dishes (15 cm in diameter) of
cultured MEF cells with PBS for several times, homogenize it
with a Dounce homogenizer in 4 mL of H-S buffer.
2. Centrifuge at 1000 gfor 10 min at 4 C.
3. Discard the pellet and the centrifuge the supernatant at
11,400 gfor 20 min at 4 C.
4. Resuspend the pellet in 500 μL H-S buffer and apply it on the
top of a HEPES-buffered sucrose step density gradient
(0.3 mL of 2.3 M, 1.7 mL of 1.7 M, and 1.5 mL of 1 M).
5. Centrifuge the gradient at 100,000 gfor 30 min at 4 C.
6. Recover 500 μL of 1 M/1.7 M interface, dilute threefold with
H-S buffer, and centrifuge for 10 min at 11,400 g. Repeat
this wash step three times.
7. Characterize the purification by Western blot towards Tim23
and Kif5b, as well as transmission electron microscopy (TEM)
imaging for mitochondria morphology (Fig. 2).
3.2.3 Flow Chamber
Assembly
1. Clean the coverslips in a staining jar by sonicating in acetone for
30 min, then in 1 M KOH for 30 min. Then wash the coverslips
with ddH
2
O for three times and store in ddH
2
O to keep the
surface hydrophilic.
2. Assemble the flow chamber used for the single-molecule
in vitro reconstitution assay (shown in Fig. 3, containing four
individual channels) with cleaned coverslips and slides just
before the experiments. Use strips of double-sided tapes
(length ~ 40 mm, width ~2 mm, thickness ~200 μm, total
volume ~15 μL) to stick the coverslips and slides together and
to isolate the flow channels. Use wrapped filter papers to absorb
and change the solution along the chamber/channel.
ab
Fraction
KIF5B
TIM23
Fig. 2 Purification of mitochondria. (a) MEF cells were homogenized and centrifuged in OptiPrep density
gradient medium. The distribution of KIF5B and the mitochondrial marker TIM23 in the fractions was
monitored by Western blotting. (b) TEM analysis of purified mitochondria from rat liver. Scale bar, 0.5 μm.
(Images from ref. [2])
Mitochondrial Network Formation 337
3.2.4 Preparation
of Polymerized Microtubule
Filaments
1. Dissolve the dark and/or HiLyte 647 tubulin powder in gen-
eral tubulin buffer and tubulin glycerol buffer with GTP as
described in the manufacturer’s manual to give a final concen-
tration of 4 mg/mL protein. Aliquot the tubulin solution and
store in the 80 C freezer after quick freezing in liquid
nitrogen, protecting from light.
2. Mix the dark tubulin and dye-labeled tubulin at a molar ratio of
25:1. Add 1 mM GTP and 20 μM taxol to the tubulin solution.
Incubate and polymerize the tubulin mix in a 37 C water bath
for 30 min followed by 30 min centrifugation at 20,000 gto
remove the tubulin dimers or short oligomers. Resuspend the
pellet of microtubule filaments in MAB/BRB80 containing
20 μM taxol and keep the tube in a 37 C water bath for at
least 1 day before using (see Note 8).
3. Check the length of the microtubule filaments by direct visual-
ization with 640 nm TIRF illumination on an inverted
microscope.
3.2.5 Gliding Assays
to Confirm the Activity
of Purified Motor Proteins
1. Prepare the gliding assay chambers by using hydrophilic cover-
slips. Incubate 15 μL of ~1 mg/mL full-length KIF5B in flow
chamber channels for 5 min.
2. Block the kinesin-coated coverslips with 50 μL of 3 mg/mL
casein for 5 min, and then incubate the chamber channel with
50 μL of 40 nM HyLite 647-labeled microtubules (see Note 9).
Check the density of microtubule filaments by direct visualiza-
tion with 640 nm TIRF illumination on an inverted
microscope.
3. Add an ATP solution with an ATP regeneration system and an
oxygen scavenger system subsequently into the chamber.
4. Record image sequences every 500 ms with TIRF illumination
mode under 640 nm laser excitation (Fig. 4).
Pipette tips
Channel #1
Channel #2
Channel #3
No.1 Coverslip
Double-side tape
Channel #4
The flow chamber for in vitro reconstitution assay
Slides
Flow
Wrapped filter
paper
Fig. 3 The flow chamber used in the single-molecule in vitro reconstitution assay. For each channel, length
~40 mm, width ~2 mm, thickness ~200 μm, total volume ~15 μL
338 Wanqing Du et al.
3.2.6 In Vitro
Reconstitution of MNF
1. Incubate and coat the channels with 15 μLof10μg/mL anti-
tubulin antibody for 5 min, wash and block the channel with
50 μL of 3 mg/mL casein for 5 min, allowing the fluorescent-
labeled microtubule filaments to be immobilized on the cover-
slips (see Note 9). Check the length and density of the filaments
by direct visualization with 640 nm TIRF illumination on an
inverted microscope.
2. Label the mitochondria with CM-Dil or MitoTracker as
described by the manufacturer’s instruction. For CM-Dil, add
0.5 μL mitochondria and 0.3 μL CM-Dil into 50 μL PBS. Mix
and incubate at 37 C for 3 min; centrifuge at 12,000 gfor
3 min, 4 C. Resuspend pellet with 50 μL MAB buffer.
3. Incubate full-length KIF5B (80 nM) with ~0.4 mg/mL mito-
chondria for 10 min on ice.
4. Incubate the motor-coated mitochondria into the
microtubule-coated flow chambers. Check the density of the
mitochondria by direct visualization with 561 nm TIRF
illumination.
5. Add 60 μL of ATP solution containing 20 μM ATP, 20 μM
taxol, 1 mg/mL casein, an ATP regeneration system, and an
oxygen scavenger system into the chamber channel.
6. Visualize the formation of mitochondrial nanotubes and the
networks using a Nikon TIRF microscope under 561 nm TIRF
illumination with 500 ms interval time (Fig. 5), as well as
scanning electron microscope (SEM).
Single frame Maximum intensity
DyLight647
Microtubule
Fig. 4 Gliding assay to confirm the activity of purified KIF5B motor proteins. The single frame (left panel) and
maximum intensity projection (right panel) of the microtubule gliding track trajectory. The images were
obtained every 500 ms for 60 s on a Nikon TIRF microscope under 640 nm laser illumination and processed
with ImageJ. Scale bar, 10 μm
Mitochondrial Network Formation 339
4 Notes
1. When preparing the storage buffer for KIF5B protein, add the
ATP freshly just before using.
2. When preparing the H-S buffer for mitochondria purification,
add the protease inhibitors freshly just before using.
3. We recommend using the tapes of 200 μm thickness to assem-
ble the flow chamber for the in vitro reconstitution assay. We
recommend using the double-sided tape from 3 M (Cat.
No. 200MP). We also recommend using the HTS tubulin,
HiLyte 647 tubulin, general tubulin buffer, and tubulin glyc-
erol buffer all from Cytoskeleton Corporation.
4. When preparing the casein solution in MAB, the pH value will
be reduced after the casein powder is dissolved. Adjust the pH
back to 6.8 with KOH powder.
5. When preparing the 100 GLOX solution, weigh and dissolve
60 mg glucose oxidase and 6 mg catalase powder in 500 μL
PBS on a vortex shaker and avoid bubbles for few minutes.
Centrifuge the yellow solution at 20,000 gfor 1 min and
Isolated mitochondria from
MEF cells Rat liver
-ATP
+ ATP
20μM
Fluorescent image contrast was enhanced to better visualize the nanotube.
CM-DiI
a
b
SEM image after in vitro
tubulation/MNF
Fig. 5 KIF5B drives mitochondrial network reformation in vitro. (a) Purified mitochondria were labeled with
CM-DiI. Highly concentrated mitochondria were incubated with KIF5B, transferred into flow chamber channels
coated with polymerized microtubules, and visualized in the presence of ATP. Images were collected with a
TIRF microscope. Scale bar, 10 μm. (b) Channels from (a) was disassembled and mitochondria were analyzed
by scanning electron microscope (SEM). Scale bar, 2 μm
340 Wanqing Du et al.
discard the dark-colored pellet. Finally, add 500 μL 80% glyc-
erol, mix and aliquot the solution into 100–200 μL. Store the
GLOX solution at 20 C.
6. Prepare the ATP working solution freshly before each experi-
ment and keep it on ice before adding it into the chamber.
Immediately before adding the solution to the chamber, add
the ATP and GLOX. Remember to warm the ATP solution
before adding it into the chamber.
7. Add ATP to the storage buffer freshly before usage. To avoid
multiple freeze–thaw cycles, we recommend making protein ali-
quots of no more than 5 μL per tube (every experiment needs
~0.4 μL kinesin solution). Store the protein aliquots in the
80 C freezer after quickly freezing them in liquid nitrogen.
8. Prewarm the buffers and centrifuge before dealing with micro-
tubules because low temperatures cause rapid depolymerization
of microtubule filaments. Cut the pipette tips when resuspend-
ing the microtubule pellet to avoid shearing forces. Sometimes
the microtubule filaments are not long enough (<5μm) for the
in vitro assay. If so, just leave the resuspended solution in a 37 C
water bath and the filaments will anneal again at 37 C.
9. The microtubule filaments used in gliding assay should be cut
by quickly pipette shearing, while the filaments used in MNF
assay should be transferred gently by cut pipette tips.
Acknowledgments
This work was supported by the grants from the Australia National
Health and Medical Council (NHMRC, APP1177374 to Q.P.S.),
the Australia National Heart Foundation (NHF, 102592 to Q.P.S),
the University of Technology Sydney’s Grant for IBMD (Q.P.S.)
and the China Scholarship Council (No. 201706170028 to X.D.).
References
1. Chan DC (2012) Fusion and fission: interlinked
processes critical for mitochondrial health. Annu
Rev Genet 46:265–287
2. Wang C, Du W, Su QP et al (2015) Dynamic
tubulation of mitochondria drives mitochondrial
network formation. Cell Res 25(10):1108–1120
3. Su QP, Du W, Ji Q et al (2016) Vesicle size
regulates nanotube formation in the cell. Sci
Rep 6:24002
4. Du W, Su QP (2019) Single-molecule in vitro
reconstitution assay for kinesin-1-driven mem-
brane dynamics. Biophys Rev 11(3):319–325
5. Du W, Su QP, Chen Y et al (2016) Kinesin
1 drives autolysosome tubulation. Dev Cell 37
(4):326–336
6. Guan R, Zhang L, Su QP et al (2017) Crystal
structure of Zen4 in the apo state reveals a miss-
ing conformation of kinesin. Nat Commun
8:14951
7. Su QP, Ju LA (2018) Biophysical nanotools for
single-molecule dynamics. Biophys Rev 10
(5):1349–1357
8. Shen M, Zhang N, Zheng S et al (2016) Cal-
modulin in complex with the first IQ motif of
myosin-5a functions as an intact calcium sensor.
Proc Natl Acad Sci U S A 113(40):
E5812–E5820
Mitochondrial Network Formation 341
Chapter 26
Extraction of Functional Mitochondria Based on Membrane
Stiffness
Md Habibur Rahman, Qinru Xiao, Shirui Zhao, An-Chi Wei, and Yi-Ping Ho
Abstract
The abnormal functionality of mitochondria has been linked to many life-threatening diseases such as
cancers, failure of cardiovascular functions, and neurodegenerative disorders. Therefore, in vitro analysis of
mitochondria has garnered great interest for understanding the mechanism of mitochondrial dysfunction-
related disease development and therapeutics. However, due to the intrinsic heterogeneity of cell mem-
brane stiffness, it remains challenging to standardize the protocols for the extraction of mitochondria and
adequate disruption of the cellular membrane while retaining the functionality of mitochondria. We have
previously developed a microfluidics-based cell shredder capable of serving the purpose. In this protocol, we
describe the step-by-step procedures to empirically identify the threshold shear stress using this
microfluidics-based cell shredder for mitochondrial extraction. The optimal shear stress to disrupt human
embryonic kidney cell (HEK 293) and mice muscle cell (C2C12) has been characterized at around 16.4 Pa,
whereas cell lines with stiffer membrane stiffness, for example, neuroblastoma cells (SH-SY5Y), require
27.4 Pa to effectively lyse the cells. This protocol also provides detailed procedures to determine the quality
of extracted mitochondria based on the membrane potential and the integrity of extracted mitochondria. A
comparison with the widely employed Dounce homogenizer has shown that the proposed microscale cell
shredder can yield at least 40% more functional mitochondria and retain higher integrity regarding
extracted mitochondria than the counterparts extracted from Dounce homogenizer, especially for low
cell concentrations (5–20 10
4
cells/mL) and small sample volume (<200 μL).
Key words Cell membrane stiffness, Membrane disruption, Mitochondrial membrane potential,
Mitochondrial integrity, Shear stress
1 Introduction
Mitochondrial dysfunction has been linked to several human diseases
including cardiovascular diseases [1], cancers [2], neurodegenerative
disorders [3], and premature aging [4]. Therefore, analysis of mito-
chondrial functions becomes imperative to understand the role of
mitochondrial dysfunction throughout the disease development.
Existing investigations rely heavily on in vitro analysis where the
mitochondria are extracted upon cellular membrane disruption either
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_26,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
343
physically or chemically. Chemical lysis of cells may involve membrane
digesting enzymes, which are also prone to impair mitochondrial
membrane [5], given that the cellular membrane and mitochondrial
membrane are composed of similar constructs. Physical disruption of
the cellular membrane is typically conducted by mechanically shearing
the cell membrane by a Teflon-glass apparatus such as Dounce
homogenizer, where the degree of cell disruption is collectively deter-
mined by the grinding speed and number of strokes. As the magni-
tude of applied pressure by the homogenization is relatively
uncontrollable and largely depends on operator variation, the
extracted mitochondria may not be consistent in quality. As reported
by Zeng et al., pestle-based homogenization may damage the mito-
chondrial integrity up to 40% even at the optimal number of strokes
with consistent grinding speed [6].
We have previously developed a microfluidics-based cell shred-
der (Fig. 1a), based on different membrane stiffness of
Fig. 1 Microscale cell shredder for cell disruption to extract functional mitochondria. (a) Schematic of the cell
disruption while cells are hydrodynamically stressed in the cross-junction microchannel. (b) Numerical
estimation of extensional stress surrounding the stagnation point in the cross-junction channel. (c) Overview
of the microfluidics chip. (d) High-speed image of cell deformation obtained deforming HEK 293 cells using a
high-speed camera (Dimax CS3, PCO AG, Kelheim, Germany) (Partially reproduced from ref. [10] with
permission from Springer Nature)
344 Md Habibur Rahman et al.
mitochondrial and cellular membrane [7–9], to selectively disrupt
cellular membrane while keeping mitochondrial membrane intact
[10]. Herein, this protocol describes step-by-step procedures to
empirically observe the threshold shear stress to disrupt the model
cell lines such as human embryonic kidney (HEK 293), mouse
muscle cells (C2C12), and neuroblastoma (SH-SY5Y). As a com-
parison, the optimization process for traditionally used Dounce
homogenizer is also described. Based on the characterizations of
membrane potentials and integrity, the microfluidics-based cell
shredder is able to yield higher quantity of mitochondria and retain
better mitochondria integrity especially when low concentration of
sample is considered.
2 Materials
Prepare all the solutions in ultrapure water at room temperature
unless otherwise specified.
1. Hypotonic reticulocyte standard buffer (RSB): 10 mM NaCl,
1.5 mM MgCl
2
, 10 mM Tris–HCl, pH 7.5. For a 300 mL
buffer, mix 100 mL ultrapure water and 300 μL of 1 M Tris–
HCl (pH 7.5) in a glass beaker. Add 175.3 mg NaCl and
42.84 mg MgCl
2
powder. Mix well by a magnetic stirrer
(about 30 min) and adjust the pH to 7.5 with HCl. Bring the
volume up to 300 mL with ultrapure water. Store at 4 C.
2. Mitochondria storage (MS) buffer: 10 mM Tris–HCl, pH 6.7,
10 mM KCl, 0.15 mM MgCl
2
, 1 mM PMSF, and 1 mM DTT.
For a 300 mL buffer, mix 100 mL ultrapure water and 300 μL
of 1 M Tris–HCl (pH 6.7) in a glass beaker. Add 4.28 mg
MgCl
2
, 223.65 mg KCl, 46.28 mg DTT, and 52.26 mg PMSF.
Mix well by a magnetic stirrer (about 30 min) and adjust the
pH to 7.5 with HCl. Bring the volume up to 300 mL with
ultrapure water. Store at 4 C.
3. MitoTracker Red (Thermo Fisher): Add 100 μL DMSO to
50 μg of MitoTracker Red dye powder to make 1 mM storage
solution. Prepare a stock solution of 100 μM by diluting the
storage solution in PBS. Preserve the stock and storage solu-
tion in dark at 20 C.
4. Citrate synthase (CS) assay (Sigma Aldrich): Bring the CS assay
buffer to room temperature before use. Reconstruct CS devel-
oper (provided by the manufacturer) by adding 1 mL of CS
assay buffer. Aliquot and store at 20 C and use within
2 months of reconstruction. Reconstruct the substrate mixture
by adding 220 μL of ultrapure water. Mix well by pipetting up
and down, aliquot and store 20 C and use within 2 months
of reconstruction.
Mitochondrial Extraction 345
3 Methods
3.1 Design
and Fabrication
of the Microfluidics-
Based Cell Shredder
1. Mask production: Design a microchannel capable of producing
proper extensional stress (see Note 1 and exampled simulation
to estimate the shear stress surrounding the stagnation point in
Fig. 1b) for shearing the membrane of mammalian cells. Print
the design on a transparency mask.
2. Fabrication of SU8 master mold: Conduct the photolithogra-
phy process according to the manufacturer’s suggestion. Spin
coat negative photoresist SU8–2075 on a 400 silicon wafer at
4000 rpm (~394 g) for 30 s to obtain a thickness of 60 μm.
Place the SU8-coated wafer onto a UV aligner, and make sure
that the distance between the transparency mask and the wafer
is proper. Expose the SU8 with proper energy dose based on
the power density of employed UV aligner. After exposure,
develop the structures by SU8 developer. Bake the developed
structures on a hot plate accordingly.
3. PDMS molding and chip fabrication:Mix polydimethylsilox-
ane (PDMS) at a ratio of 10:1 (PDMS prepolymer: curing
agent) thoroughly. Degas the well-mixed PDMS mixture in a
vacuum chamber. Pour the PDMS onto the SU8 master mold
and cure the PDMS at 60 C for 1 h. Peel off the cured PDMS
slab from the master mold, punch inlet and outlet holes with a
1 mm OD hole puncher. Seal the channel with a glass coverslip
using oxygen plasma, rendering the assembled chip as shown in
Fig. 1c.
3.2 Cell Culture 1. Grow and maintain human embryonic kidney cells (HEK293,
ATCC) and mouse skeletal muscle cells (C2C12, ATCC) in
Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen).
Supplement the growth medium with fetal bovine serum
(10%) (Invitrogen), penicillin (100 units/mL), and streptomy-
cin (100 μg/mL).
2. Maintain neuroblastoma cells (SH-SY5Y) in DMEM/F-12,
supplemented with fetal bovine serum (10%), penicillin
(100 units/mL), and streptomycin (100 μg/mL).
3. Maintain the cell culture at 37 C in a humidified 5% CO
2
atmosphere.
3.3 Cell Disruptions
and Mitochondria
Extraction
1. All the cell disruption procedures and centrifugation should be
operated at 4 C.
2. The cells are harvested from a confluent flask by trypsinization
before implementing the procedures of cell disruption and
mitochondria extraction. In this chapter, previously validated
protocols by Dounce homogenizer [11] and the procedures to
operate a microscale cell shredder presented by our group [10]
are described.
346 Md Habibur Rahman et al.
3. Prepare a stock solution of cells at 5 10
6
cells/mL in PBS by
counting the cells using a standard hemocytometer.
4. Prepare a working solution of cells at a concentration of
110
6
cells/mL in 3 mL of RSB hypotonic buffer.
5. Resuspend the cells in 3 mL of RSB hypotonic buffer at a
concentration of 1 10
6
cells/mL.
6. Incubate the cells-containing RSB solution at 37 C for about
5–10 min. Validate the swelling progress by phase-contrast
microscopy (20magnification).
7. The swelled cells are used for subsequent procedures for cell
homogenization.
3.3.1 Optimization
of Mitochondria Extraction
by the Dounce
Homogenizer
1. Make the homogenizer ready prior to the extractions (see
Note 2).
2. Transfer 200 μL of swelled cells (as obtained from step 6 of
Subheading 3.3) to the cylinder of the homogenizers. Apply
five strokes to disrupt and obtain cell lysates (see Note 3).
3. Transfer all the cell lysates to a 1 mL tube. Record the number
of strokes applied to the homogenization (e.g., 5strokes).
4. Repeat steps 1–3with titrated strokes (recommended interval:
five strokes) up to 30 strokes and label the tubes accordingly.
5. Observe the degree of homogenization by phase-contrast
microscopy (see Note 4).
6. Centrifuge at 1,000 gfor 5 min twice to remove unbroken
cells, cell debris, and nuclei (see Note 5).
7. Carefully transfer the supernatant to a clean 1 mL tube without
disturbing the pelleted debris.
8. Centrifuge the supernatant at 15,000 gfor 15 min. Discard
the supernatant, resuspend the mitochondrial palette in 1MS
buffer, and proceed to the further characterization of the
extracted and isolated mitochondria.
9. Identify the optimum number of strokes by measuring the total
protein yield (detailed in Subheading 3.5) and the percentage
of functional mitochondria as described in Subheading 3.6 (see
Note 6).
3.3.2 Optimization
of Mitochondria Extraction
by the Microscale Cell
Shredder
1. Collect 1 mL of swelled cells (as obtained from step 6 in
Subheading 3.3) into a 1 mL syringe connected with a tubing
assembly composed of a 22AWG needle and 24 AWG PTFE
tube (see Note 7).
2. Use another syringe-tubing assembly to withdraw 1 mL of RSB
hypotonic buffer.
3. Connect the two syringe-tubing assemblies to the two inlets of
the microfluidic cell shredder as illustrated in Fig. 1c.
Mitochondrial Extraction 347
4. Place the two syringes onto a syringe pump.
5. Connect another tubing assembly from the outlet to a
collection tube.
6. Set the pump at volumetric flow rates designated for different
shear stress (for the design shown in this protocol, the
employed volumetric flow rates are 20, 40, 60, 80, and
100 μL/min).
7. Collect the cell homogenates produced at different flow rates
into different tubes.
8. Follow the steps 5–8 in Subheading 3.3.1 and resuspend the
isolated mitochondria in 1MS buffer.
9. Identify the optimum flow rate by measuring the total protein
yield (detailed in Subheading 3.5) and the percentage of func-
tional mitochondria as described in Subheading 3.6 (see
Note 8).
3.4 Cell Disruption
Efficiency Determined
by Flow Cytometer
1. Load untreated cells (intact cells) at the concentrations of
110
6
cells/mL into a flow cytometer.
2. Profile a forward versus side scatter (FSC-SSC) plot by at least
10,000 events to obtain reliable statistics. Record the total
number of cells detected.
3. Analyze the cell lysates obtained from different conditions of
disruptions using the same settings of the flow cytometer.
4. Compare the FSC-SSC plots obtained from cell lysates with the
profile obtained from the intact cells to calculate the disruption
efficiency (ratio of the number of disrupted cells to the total
recorded cells from untreated cells).
3.5 Determination
of Total Protein Yield
(Bradford Assay)
1. Bring all the reagents to room temperature before the assay.
2. For a 96-well reaction, load 150 μL of the Bradford reagent
into the well, add 10 μL of extracted mitochondria sample, and
incubate at 37 C for 15 min.
3. Measure the absorbance at 595 nm using a microplate reader.
4. Convert the absorbance measured from the samples to protein
concentration using a standard curve constructed by bovine
serum albumin (BSA) in the concentration range of 10–50 μg/
mL.
3.6 Characterization
of Mitochondrial
Membrane Potential
1. For a 96-well reaction, load 100 μL of unstained mitochondria
to a well as a control and background signal.
2. For each condition, load 97.5 μL of isolated mitochondria
solution to a well, add 2.5 μL of MitoTracker dye (stock
solution as described in item 2 of Subheading 2).
3. Incubate the stained sample at 37 C for 10 min in dark.
348 Md Habibur Rahman et al.
4. Analyze the sample using a flow cytometer by detecting at least
10,000 events with proper fluorescence channels as suggested
by the manufactures.
5. Determine the fluorescently positive events by subtracting the
background fluorescent signals and proper gating (see Note 9).
3.7 Characterization
of Mitochondrial
Integrity (Citrate
Synthase Assay)
1. Follow the manufacturer’s protocol to construct a standard
curve of CS activity against absorbance using glutathione
(GSH) (supplied by CS Assay Kit).
2. For a 96-well reaction, load 10 μL of the extracted mitochon-
dria sample and subsequently add 43 μL of CS assay buffer,
5μL of reconstructed developer, and 2 μL of substrate mixture.
For blank/background samples, add 45 μL of CS assay buffer,
5μL of the reconstructed developer (no substrate mixture),
and 10 μL of ultrapure water.
3. Incubate at 37 C in dark.
4. Measure the absorbance at 412 nm continuously until the
reading reaches a plateau. Compare the measured absorbance
to the GSH standard curve (as constructed in step 1) and
obtain the CS activity based on the standard curve. Determine
“initial CS activity” immediately after adding the reaction mix-
ture and “total CS activity” when the reading reached a plateau
(typically within 20–30 min at a 5 min interval in each
readings).
5. Mark the difference between “total CS activity” and “initial CS
activity” as “latent activity.”
6. Estimate the mitochondrial integrity by the ratio between the
“latent activity” and “total enzymatic activity” for each sample
tested (see Note 10).
4 Notes
1. Calculate the mean shear stress at the cross section of the
channel following Newton’s law of viscosity [12], and estimate
the extensional stress numerically by COMSOL Multiphysics
®
as detailed in our previously reported work [10]. Based on
recent findings, soft mammalian cells can be permanently
deformed by applied mean shear stress in the range of
2–10 Pa [13]. Therefore, a microfluidic chip capable of
providing a mean shear stress in the range 2–25 Pa and an
extensional stress in the cross-junction geometry in the range
of 25–130 Pa is suggested (Fig. 1b). While the parameters in
this protocol are optimized for HEK 293, C2C12, and
SH-SY5Y, Table 1provides a list of Young’s moduli of other
Mitochondrial Extraction 349
cell lines as a reference. For the demonstration purposes, an
image of deformed HEK 293 cells when passing through the
cross-junction microchannel is shown in Fig. 1d.
2. Prior to the experiment, sterilize the Dounce homogenizer and
dry at 80 C overnight. Perform cell disruptions while keeping
the Dounce homogenizer on ice. Wash the cylinder and pestle
of the Dounce homogenizer with 75% ethanol and ultrapure
water between different conditions of cell disruptions such as
changing the number of strokes or changing the cell
concentrations.
3. Use a tightly fitted pestle and mortar to ensure maximum
disruptions. For the application of a stroke, press the pestle
straight down to the glass tube, maintain firm and constant
grinding speed for the consideration of consistency.
4. We suggest observing membrane disruption under 20or
40magnification using a phase-contrast microscopy. Under
the field, dark spherical objects (relatively larger in size) are the
nuclei. Occasionally, nonspecific binding of nucleotides would
create a long and sticky-looking strand. Small organelles
including mitochondria are normally observed dark and
granular-like with a much smaller size than nuclei. Staining
with trypan blue may help to increase the contrast of the images
and for the identification of cellular membrane disruption.
5. After the centrifugation, cell debris shall be precipitated at the
bottom. The supernatant containing mitochondria and other
subcellular organelles shall look hazy white.
6. For cell lines with relatively softer membrane stiffness such as
HEK293 and C2C12 (Table 1), ten strokes have been
observed as optimal to extract functional mitochondria.
Figure 2a shows the optimization for the number of strokes
based on total protein yield (detailed in Subheading 3.5) and
functional mitochondria (detailed in Subheading 3.6)
extracted from HEK 293 cell lines. As observed from the
Table 1
Young’s moduli of different cell lines
Cells Young modulus (characterized by AFM measurement)
Human embryonic kidney cells (HEK293) 2.1 0.1 kPa [16]
Mice muscle cells (C2C12) 2.97 0.1 kPa [16]
Human mesenchymal stem cells (hMSC) 2.5 1.8 kPa [17]
Human neuroblastoma cells (SH-SY5Y) 5.2 0.1 kPa [14]
Porcine cardiomyocytes 1–10 KPa [18]
Porcine heart mitochondria 17–50 KPa [8]
350 Md Habibur Rahman et al.
results, total protein yield reaches a saturation after ten strokes,
while MitoTracker positive events slightly drop upon increas-
ing the number of strokes. Therefore, ten strokes are selected as
the optimal condition for HEK 293 cells. However, a cell line
with relatively stiffer cell membrane, for example, neuroblas-
toma SH-SY5Y, which stiffness is 5.2 0.1 kPa [14], about
twice higher then HEK293, requires 20 strokes for the efficient
disruption (Fig. 2b).
7. For a 24AWG PTFE tube, the perfectly fitted needle size is
23AWG. Take extra care when inserting the needle into the
inner region of the tube. We suggest connecting the syringe to
the tube with a blunt needle (Luer-lock) for the safety concern
and to avoid protruding the tube.
8. Figure 3a shows the optimization for the microfluidics-based
cell shredder to disrupt HEK 293 cells. With the increased
shear stress (i.e., applied shear stress), the total protein yield
increases as well due to the further damage on cell membrane.
Similarly, extended shear stress larger than ~15 Pa would ham-
per the mitochondrial functionality (Fig. 3a). A similar magni-
tude of shear stress (60 μL/min flow rate for this case) is also
considered optimal to disrupt the C2C12 cell lines as the
membrane stiffness of C2C12 cell line is close to that of
HEK293. For cells with stiffer membrane, for example, neuro-
blastoma (SH-SY5Y), the volumetric flow rate shall be
increased to 100 μL/min, which corresponded to 27.47 Pa
mean shear stress (Fig. 3b).
Fig. 2 Optimization of the number of strokes applied by the pastel of the Dounce homogenizer for cell lines of
different membrane stiffness (a) HEK 293 and (b) neuroblastoma (SH-SY5Y) cell lines. Results plotted at
mean SD (n¼3 independent experiments) (Reproduced from ref. [10] with permission from Springer
Nature)
Mitochondrial Extraction 351
9. Figure 4shows the gating to determine the MitoTracker Red
positive events of isolated samples. Background fluorescent
signal was subtracted (from the unstained sample) to obtain
Fig. 3 Optimization of the applied shear stress by the microscale cell shredder for functional mitochondria
extraction from (a) HEK 293 cells and (b) neuroblastoma (SH-SY5Y) cell. Results plotted at mean SD (n¼3
independent experiments) (Reproduced from ref. [10] with permission from Springer Nature)
Fig. 4 Determination of MitoTracker Red fluorescent positive events from the isolated mitochondria samples.
The upper-right quadrant notes the percentage of fluorescent positive events and the number of positive
events. The data presented were obtained by disrupting HEK 293 cells at the concentration of 20 10
4
cells/
ml under the optimal condition of (a) Dounce homogenizer (ten strokes) (b) microscale cell shredder at a flow
rate of 60 μL/min
352 Md Habibur Rahman et al.
the percentage of MitoTracker Red positive events. The per-
centage of MitoTracker events and concentrations of the
detected sample has been recorded (as shown on the upper-
right quadrant of the plot). Concentrations of functional mito-
chondria (per unit volume) have been obtained by multiplying
the percentage of MitoTracker Red positive events with the
number of positive events detected per unit volume. Figure 5
Fig. 5 Quantity of extracted mitochondria was determined by MitoTracker Red staining of the extracted
sample. Cells were disrupted by the two approaches for cell lines with soft cell membrane (a) HEK 293, (b)
C2C12 and cell membrane with relatively stiffer membrane, and (c) SH-SY5Y cells. Results were plotted as
mean SD (n¼3 independent experiments, *P <0.05, **P <0.01, ***P <0.001) (Reproduced from ref.
[10] with permission from Springer Nature)
Mitochondrial Extraction 353
shows the concentrations of functional mitochondria at differ-
ent cell concentrations using both approaches for HEK 293,
C2C12, and SH-SY5Y cell lines.
10. Citrate synthase (CS) release serves as a reverse correlation of
mitochondrial membrane integrity, as the CS is located in the
inner membrane of the double pair membrane of mitochon-
dria. Therefore, the activity of released CS enzymes is consid-
ered an exclusive marker to quantify the integrity of the
extracted mitochondria [6,15].
Acknowledgments
This work is supported in part by the Shun Hing Institute of
Advanced Engineering (Project #BME-p2-17) and the Research
Committee Funding (MD18764) provided by the Chinese Univer-
sity of Hong Kong. A.C.W. would also like to acknowledge the
support provided by the Ministry of Science and Technology in
Taiwan (MOST-107-2636-B-002-001).
References
1. Chistiakov DA, Shkurat TP, Melnichenko AA
et al (2018) The role of mitochondrial dysfunc-
tion in cardiovascular disease: a brief review.
Ann Med 50:121–127
2. Vyas S, Zaganjor E, Haigis MC (2016) Mito-
chondria and cancer. Cell 166:555–566
3. Johri A, Beal MF (2012) Mitochondrial dys-
function in neurodegenerative diseases. J Phar-
macol Exp Ther 342:619–630
4. Scheibye-Knudsen M, Croteau DL, Bohr VA
(2013) Mitochondrial deficiency in Cockayne
syndrome. Mech Ageing Dev 134:275–283
5. Franko A, Baris OR, Bergschneider E et al
(2013) Efficient isolation of pure and func-
tional mitochondria from mouse tissues using
automated tissue disruption and enrichment
with anti-TOM22 magnetic beads. PLoS One
8(12):e82392
6. Bahnemann J, Kayo S, Wahrheit J et al (2014)
In search of an effective cell disruption
method to isolate intact mitochondria from
Chinese hamster ovary cells. Eng Life Sci
14:161–169
7. Hochmuth RM (2000) Micropipette aspira-
tion of living cells. J Biomech 33:15–22
8. Li Y, Honda S, Iwami K et al (2015) Analysis of
mitochondrial mechanical dynamics using a
confocal fluorescence microscope with a bent
optical fibre. J Microsc 260:140–151
9. Mela MJ (1967) Elastic-mathematical theory
of cells and mitochondria in swelling process:
the membranous stresses and modulus of elas-
ticity of the egg cell of sea urchin, Strongylocen-
trotus purpuratus. Biophys J 7:95–110
10. Rahman H, Xiao Q, Zhao S et al (2018)
Demarcating the membrane damage for the
extraction of functional mitochondria. Micro-
syst Nanoeng 4:39
11. Clayton DA, Shadel GS (2014) Isolation of
mitochondria from tissue culture cells. Cold
Spring Harb Protoc 2014:1109–1111
12. Byron Bird R, Warren E, Stewart ENL (2007)
Transport phenomena, 2nd edn. John Wiley
and Sons, New York, NY
13. Tanzeglock T, Soos M, Stephanopoulos G,
Morbidelli M (2009) Induction of mammalian
cell death by simple shear and extensional
flows. Biotechnol Bioeng 104:360–370
14. Fang Y, Iu CYY, Lui CNP et al (2014) Inves-
tigating dynamic structural and mechanical
changes of neuroblastoma cells associated
with glutamate-mediated neurodegeneration.
Sci Rep 4:1–10
15. Boutagy NE, Pyne E, Rogers GW et al (2015)
Isolation of mitochondria from minimal quan-
tities of mouse skeletal muscle for high
throughput microplate respiratory measure-
ments. J Vis Exp 105:53217
354 Md Habibur Rahman et al.
16. Li M, Liu L, Xiao X et al (2016) Effects of
methotrexate on the viscoelastic properties of
single cells probed by atomic force microscopy.
J Biol Phys 42:551–569
17. Darling EM, Topel M, Zauscher S et al (2008)
Viscoelastic properties of human
mesenchymally-derived stem cells and primary
osteoblasts, chondrocytes, and adipocytes. J
Biomech 41:454–464
18. Jacot JG, Kita-Matsuo H, Wei KA et al (2010)
Cardiac myocyte force development during dif-
ferentiation and maturation. Ann N Y Acad Sci
1188:121–127
Mitochondrial Extraction 355
Chapter 27
A Protocol for Untargeted Metabolomic Analysis: From
Sample Preparation to Data Processing
Amanda L. Souza and Gary J. Patti
Abstract
Untargeted metabolomics has rapidly become a profiling method of choice in many areas of research,
including mitochondrial biology. Most commonly, untargeted metabolomics is performed with liquid
chromatography/mass spectrometry because it enables measurement of a relatively wide range of physio-
chemically diverse molecules. Specifically, to assess energy pathways that are associated with mitochondrial
metabolism, hydrophilic interaction liquid chromatography (HILIC) is often applied before analysis with a
high-resolution accurate mass instrument. The workflow produces large, complex data files that are
impractical to analyze manually. Here, we present a protocol to perform untargeted metabolomics on
biofluids such as plasma, urine, and cerebral spinal fluid with a HILIC separation and an Orbitrap mass
spectrometer. Our protocol describes each step of the analysis in detail, from preparation of solvents for
chromatography to selecting parameters during data processing.
Key words Metabolomics, Metabolites, Profiling, Liquid chromatography, Mass spectrometry, High-
resolution, Accurate mass, HILIC, Data-dependent acquisition, Quality assurance, Quality control
1 Introduction
Metabolomics, or metabolite profiling, is the quantitative study of
endogenous and exogenous small molecules from a biological sys-
tem [1,2]. Unlike genomics and proteomics, which do not assess
enzyme activity directly, metabolomics surveys biochemical pheno-
types and therefore provides unique insight into health and disease.
As such, metabolomics has advanced our understanding of numer-
ous physiological processes and disease pathologies. Successful
applications of metabolomics have ranged from biomarker discov-
ery and precision medicine to nutrition and mitochondrial bioen-
ergetics [3,4].
Generally, there are two analytical strategies for profiling meta-
bolites. The first, referred to as targeted metabolomics, applies
methods that have been optimized to quantify a defined set of
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_27,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
357
molecules. The approach is well suited for research questions that
require measuring only a small number of analytes (e.g., does this
drug affect a specific pathway of interest? do these samples contain a
given pesticide? is this compound a marker for a particular pheno-
type?). Targeted metabolomics can be applied to quantitate less
than a few dozen compounds, but some workflows have been
developed to monitor several hundred metabolites [5].
At the other end of the spectrum is untargeted metabolomics,
which aims to measure all the small molecules in a sample.
Although untargeted metabolomics strives to be global and unbi-
ased, the physiochemical diversity of the metabolome limits the
number of compounds that can be measured in a single experiment
[6]. Selection of solvents, separation strategies, and instrumenta-
tion platforms strongly influence which metabolites can be detected
[7]. Given that the goal of untargeted metabolomics is to survey as
many metabolites as possible, methods that are optimal for only
specific classes of compounds are typically not applied as they often
necessitate a decrease in overall metabolome coverage. In this sense,
there is a tradeoff in metabolomics between molecular coverage and
method optimization for specific compounds. Hence, before apply-
ing the protocol presented here, we suggest considering whether
untargeted metabolomics is necessary for a given study. Detailed
descriptions of suitable applications of untargeted metabolomics
(such as biomarker discovery and hypothesis generation) have been
provided elsewhere [8].
After determining that a study will benefit from the application
of untargeted metabolomics, the next step is to select an experi-
mental workflow to conduct the analysis. There are many well-
established protocols involving different solvents, columns, mass
spectrometers, NMR systems, and processing software [9–16]. In
general, each has its own advantages and it is therefore typical for
investigators to apply multiple methods to a single biological sam-
ple. Liquid chromatography/mass spectrometry (LC/MS) is the
most widely used analytical platform for untargeted metabolomics
because of its high sensitivity and its ability to detect many com-
pounds without chemical derivatization. To improve coverage, it
has become increasingly common to integrate analysis of lipophilic
compounds by reversed-phase liquid chromatography (RP-LC)/
MS with analysis of water-soluble compounds by hydrophilic inter-
action liquid chromatography (HILIC)/MS. Notably, however,
there are multiple variations of such integrated strategies [17–20].
Herein, we present a protocol to perform untargeted metabo-
lomics on biofluids such as plasma, urine, and cerebral spinal fluid
(CSF) by using HILIC/MS. The method, which is an adaptation of
a previous protocol for targeted metabolomics, uses a Waters Atlan-
tis™HILIC Silica column coupled to an Orbitrap mass spectrom-
eter [21]. Each piece of the analytical pipeline is described in detail,
from sample extraction to data processing (Fig. 1). We would like
358 Amanda L. Souza and Gary J. Patti
to emphasize that this is one workflow for performing untargeted
metabolomics, but there are many variations that are also widely
employed in the field [22–28]. Considering this, it is worth noting
that the protocol presented is highly modular. The HILIC column
described, for example, can be exchanged for a different HILIC
column with minimal modifications. Alternatively, instead of using
an Orbitrap instrument, analysis can be performed with a quadru-
pole time-of-flight mass spectrometer or a triple quadrupole mass
spectrometer. Similarly, rather than using Thermo Scientific™
Compound Discoverer™software, data processing can be per-
formed with any number of commercial or freely available software
platforms. Some opportunities where the protocol is easily adapted
are noted within the steps that follow.
The protocol presented here is highly amenable to the investi-
gation of mitochondrial biology. Mitochondrial dysfunction can
lead to numerous metabolic alterations, both at the cellular level
and at the organismal level [29–33]. These changes result not only
from decreased ATP production by oxidative phosphorylation but
also from disruptions in other mitochondrial pathways related to
fatty acid, amino acid, nucleotide, and heme metabolism. Addition-
ally, given their role in cellular signaling, defective mitochondria
may affect any number of biochemical processes within the cell. In
the context of an organism, altered mitochondrial function can lead
to the accumulation of metabolites in biofluids such as blood
[34]. When the mitochondrial enzyme medium-chain acyl-CoA
dehydrogenase (MCAD) is deficient, for example, it leads to the
accumulation of medium-chain length acylcarnitines in blood.
Indeed, most newborns are screened for MCAD deficiency by
using dried blood spots analyzed with mass spectrometry
[35]. We have applied the protocol below to assay many metabo-
lites relevant to mitochondrial function, such as medium-chain
length acylcarnitines. Although steps are described for analysis of
biofluids, the extraction procedure can be easily adapted to study
cells or tissues.
Untargeted Workflow
Experimental
Design
Sample
Collection
Sample
Preparation
& Extraction
Data Processing
LC/MS
Unbiased
Peak
Detection
Unknown
Annotation
Statistics
Pathway
Association
Fig. 1 Steps implemented in the untargeted metabolomic workflow, from experimental design to data
processing
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 359
2 Materials
1. LC aqueous mobile phase A (0.1% formic acid, 10 mM ammo-
nium formate): To yield a total volume of 1 L, prepare the
solution with the following steps. Using an appropriate analyt-
ical balance, weigh 0.631 g of ammonium formate. Transfer
ammonium formate to a clean 1 L HPLC glass bottle. Using a
1 L graduated cylinder, transfer 999 mL of LC/MS-grade
water to the HPLC bottle. Using a pipette, transfer 1 mL of
LC/MS-grade formic acid, 99.0+%, to the HPLC bottle. Man-
ually vortex or sonicate to mix. Connect the HPLC bottle to
the LC system via the inlet tubing. Prime and purge the solvent
channel according to the manufacturer’s recommendations.
Add a label to the bottle listing its content, filling date, and
expiration date. This solution will expire approximately
1 month after preparation.
2. LC organic mobile phase B (0.1% formic acid in acetonitrile):
To yield a total volume of 1 L, prepare the solution with the
following steps. Using a 1 L graduated cylinder, transfer
999 mL of LC/MS-grade acetonitrile to a clean 1 L HPLC
glass bottle. Using a pipette, transfer 1 mL of LC/MS-grade
formic acid, 99.0+%, to the HPLC bottle. Manually vortex or
sonicate to mix. Connect the HPLC bottle to the LC system via
the inlet tubing. Prime and purge the solvent channel accord-
ing to the manufacturer’s recommendations. Add a label to the
bottle listing its content, filling date, and expiration date. This
solution will expire approximately 1 month after preparation.
3. Extraction solvent (acetonitrile:methanol:formic acid (74.9,
24.9, 0.2, v/v/v)): This organic solvent solution is formulated
to extract hydrophilic polar metabolites from the sample
matrix. Two selected stable-labeled metabolites will be
incorporated at a later step (see Internal Standard Extraction
Solution) for quality control (QC) purposes (see Note 1). To
yield a total volume of 500 mL, prepare the solution with the
following steps. Using a 250 mL graduated cylinder, transfer
125 mL of LC/MS-grade methanol into a 500 mL glass media
bottle. Using a 500 mL graduated cylinder, add 375 mL of
LC/MS-grade acetonitrile to the 500 mL glass bottle. Using a
pipette, transfer 1 mL of LC/MS-grade formic acid, 99.0+%,
to the 500 mL glass bottle. Manually vortex or sonicate to mix.
This is a stop point where the solution can be stored at 20 C
for up to 1 month after preparation. The total volume can be
adjusted based on the required volume of extraction solution
needed for analysis.
4. Internal standard stock solutions: Two isotope labeled amino
acids, L-Phenylalanine-d8 and L-Valine-d8 (Cambridge Isotope
Laboratories), should be used as internal standards (IS) to
360 Amanda L. Souza and Gary J. Patti
monitor each sample for QC purposes (see Note 2). These two
IS can be substituted for different stable-labeled metabolites as
needed. Each IS is initially prepared as a concentrated stock
solution at a nominal concentration of 1000 μg/mL. These
concentrated stocks are used to make the internal standard
extraction solution (next step). To yield a total volume of
5 mL, prepare the solution with the following steps. Using an
appropriate analytical balance, individually weigh out 5 mg of
L-Phenylalanine-d8 and L-Valine-d8 into a 7 mL glass scintilla-
tion vial. To each vial, add 4 mL of LC/MS-grade water and
1 mL of LC/MS methanol totaling 5 mL to achieve a solution
with a nominal concentration of 1000 μg/mL. Vortex each vial
to completely dissolve and mix. Solution stability differs by
selected metabolites. This is a stop point where the solution
can be stored at 20 C for up to several weeks after prepara-
tion. When required, allow IS stock solution to reach room
temperature before use.
5. Internal standard extraction solution: The IS extraction solu-
tion combines isotope-labeled internal standards with the
extraction solvent from the previous two steps, extraction sol-
vent and IS stock solutions (see Note 3). The ionization effi-
ciencies of L-Phenylalanine-d8 and L-Valine-d8 differ, thus
slightly different concentrations are used to minimize interfer-
ence with endogenous metabolite levels. To yield a total vol-
ume of 250 mL extraction solution with a nominal
concentration of 0.1 μg/mL of L-Phenylalanine-d8 and
0.2 μg/mL of L-Valine-D8, prepare the solution with the
following steps. Using a 250 mL graduated cylinder, transfer
250 mL of extraction solvent to a 250 mL glass media bottle.
Pipette 25 μLofL-Phenylalanine-d8 IS stock solution into the
glass bottle. Pipette 50 μLofL-Valine-d8 IS stock solution into
the glass bottle. Vortex to mix. Solution stability differs by
selected metabolites. This is a stop point where the IS extrac-
tion solution can be stored at 20 C for up to several weeks
after preparation.
6. Mixture of reference metabolites: A neat mixture of reference
metabolites is used to qualitatively assess LC and MS para-
meters before, during, and at the end of the LC/MS run by
evaluating metrics like retention time, mass accuracy, MS
response by peak area, precision, chromatographic peak
shape, multistage fragmentation, and other relevant parameters
(see Note 2). Compounds selected for the reference mixture
should include metabolites of known detection for the specific
protocol. Methods applying different extraction solutions or
chromatographic stationary phases for separation should incor-
porate metabolites that are expected to be detected when using
the modified protocol. When applied to plasma, urine, and
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 361
CSF, this HILIC/MS protocol typically detects the following
metabolites, which are a good starting point for a reference
mixture: glycine, trimethylamine N-oxide, alanine, β-alanine,
sarcosine, dimethylglycine, γ-aminobutyric acid (GABA), urei-
dopropionic acid, choline, creatine, creatinine, betaine, valine,
leucine, isoleucine, asparagine, ornithine, methionine, histi-
dine, asymmetric dimethylarginine (ADMA), symmetric
dimethylarginine (SDMA), 1-methylnicotinamide, kynurenic
acid, citrulline, glutamine, glutamic acid, thiamine, phospho-
choline, thyroxine, and carnitine. Using the same approach for
preparing the concentrated IS stock solution, concentrated
reference metabolite stock solutions should be prepared. To
yield a total volume of 5 mL of reference metabolite stock
solution, prepare the solution with the following steps. Using
an appropriate analytical balance, individually weigh out 5 mg
of each metabolite (purified powdered stock) (Sigma-Aldrich)
into a 7 mL glass scintillation vial. To each vial, add 2.5 mL of
LC/MS-grade water and then 2.5 mL of LC/MS-grade meth-
anol for a total volume of 5 mL to achieve a solution with a
nominal concentration of 1000 μg/mL (see Note 4). Vortex
each vial to completely dissolve and mix. Solution stability
differs by selected metabolites. This is a stop point where the
concentrated metabolite stock solutions can be stored at
20 C for up to several weeks after preparation. To yield a
total volume of 50 mL of reference metabolite mix containing
30 metabolites, each at a nominal concentration of 0.2 μg/mL,
prepare the solution with the following steps. Using a 50 mL
graduated cylinder, transfer 49.7 mL of extraction solvent to a
50 mL glass media bottle. Pipette 10 μL of each reference
metabolite solution (total volume equals 300 μL) into the
glass bottle. Vortex to mix. Solution stability differs by selected
metabolites. This is a stop point where the neat mixture of
reference metabolites can be stored at 20 C for up to several
weeks after preparation.
7. Pooled QC stock sample: A pooled QC sample consisting of
matrix material has multiple purposes, which include facilitat-
ing the annotation and identification of unknown compounds,
evaluation of reproducibility using select known metabolites
(from the reference metabolite mixture), and normalization of
detected compounds to adjust for temporal drift in signal
response (see Notes 2 and 5). A pooled QC sample consists
of a specified volume taken from each experimental sample that
is combined into one stock pooled sample (see Note 2). The
required volume of pooled QC sample depends on the injec-
tion frequency (e.g., how intermittently the pooled QC sample
is analyzed and the total number of samples in the worklist). It
is recommended that the required volume of the pooled QC
362 Amanda L. Souza and Gary J. Patti
sample be calculated before moving to the next step. Remove
experimental samples from storage in the 80 C freezer (see
Note 6) and allow them to thaw on ice (covered with foil) or in
a4C refrigerator (see Note 7). Depending on the total vol-
ume of pooled sample to collect, use a 1.5 or 2 mL microcen-
trifuge tube or a 15 mL centrifuge tube. From each
experimental sample, using a pipette, transfer the same volume
into the stock pooled sample. For instance, from each of
50 experimental samples, transfer 10 μL of sample into the
collection tube for a total of 500 μL. Vortex for 1 min at 4 C
to mix. To minimize freeze–thaw cycles, aliquot an appropriate
volume of the pooled QC sample into labeled 1.5 mL micro-
centrifuge tubes prior to the experimental analysis (see Note 8).
Always handle samples on ice. Store all samples at 80 C(see
Note 6).
3 Method
3.1 LC/MS Protocol 1. LC/MS protocol for positively charged polar metabolites: This
protocol is designed to measure hydrophilic metabolites from
biofluids such as plasma, urine, and cerebral spinal fluid (CSF)
with HILIC/MS (see Notes 5, and 9–11). The procedure has
been adapted from a prior publication to support untargeted
metabolite profiling by high-resolution accurate mass (HRAM)
mass spectrometry [9]. Minor modifications were also made for
improved outcomes, based on experience. The LC/MS system
should be cleaned, calibrated, and checked for system suitabil-
ity according to the manufacturer’s operating manual before
preparing and extracting experimental samples (see Note 12)
[36]. To evaluate LC/MS performance over extended time
periods using a designated matrix material, a long-term refer-
ence matrix can assess system suitability, also termed quality
assurance (see Note 1). The long-term reference matrix is
separated from experimental samples but provides historical
data points for instrument assessment. Solvents, additives,
and reference material should be LC/MS-grade and of the
highest purity.
3.2 LC Method Setup 1. The following parameters outline the analytical conditions for
the Thermo Scientific™Vanquish™UHPLC system (Thermo
Fisher Scientific) configured with a Vanquish Horizon Binary
Pump H, a Vanquish Column Compartment H, a Vanquish
Split sampler HT, and installed with Thermo Scientific™Stan-
dard Instrument Integration (SII) for Xcalibur™interface
driver for LC software control.
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 363
2. Use mobile phase A (0.1% formic acid, 10 mM ammonium
formate) and mobile phase B (0.1% formic acid in acetonitrile)
as prepared in the Materials section.
3. Metabolites are separated with gradient elution using the
Waters Atlantis HILIC Silica column, 2.1 150 mm, 3 μm
(Waters Corporation).
4. Prior to the use of a new column, equilibration is essential to
ensure optimal chromatographic separation. As recommended
by the manufacturer of the Atlantis HILIC Silica column, pass
50 column volumes of 50% acetonitrile/50% water (100 min at
0.25 mL/min) followed by 20 column volumes of initial
mobile phase conditions (40 min at 0.25 mL/min) through
the column.
5. Method setup parameters for the Vanquish™UHPLC system
are listed in Table 1. Perform gradient elution by using the
linear steps outlined in Table 2.
3.3 MS
Method Setup
1. The following parameters outline the analytical conditions for
the Thermo Scientific™Q Exactive™HF Hybrid-Quadrupole
Orbitrap™mass spectrometer (Thermo Fisher Scientific) uti-
lizing the QE HF Tune Instrument Control Software and
high-purity nitrogen (99.999%) connection. The Orbitrap
mass analyzer provides HRAM detection of charged molecules.
Table 1
Vanquish UHPLC system setup parameters
Column temperature 30 C
Column compartment mode Still air
Column compartment equilibration time 0.5 min
Autosampler temperature 4 C
Autosampler puncture offset 1 μm
Autosampler wash speed 10 μL/s
Autosampler wash mode Both
Autosampler wash time 2 s
Autosampler dispense speed 1.0 μL/s
Autosampler draw speed 0.5 μL/s
Injection volume 10 μL
Total run time 32 min
Pump flow rate 0.25–0.4 mL/min
Pump curve 5
364 Amanda L. Souza and Gary J. Patti
Fragmentation of isolated precursor ions is achieved with
high-energy collision-induced dissociation (HCD) to generate
fragmentation spectra containing fragment ions.
2. For improved method robustness, a diverter valve can be
incorporated into the hardware configuration post-column to
divert LC flow between MS during acquisition and waste dur-
ing re-equilibration and the initial isocratic hold. Diverter valve
setup and timing steps are outlined in Table 3.
3. Data collection for untargeted metabolomic analysis involves
two acquisition modes: full-scan (FS) acquisition mode and
data-dependent acquisition (DDA) mode. In this protocol,
FS parameter settings include ion detection with high-
resolution for all experimental samples. These MS1 data are
used for quantitative profiling analysis. DDA is then applied to
the pooled QC sample (or another representative sample, like a
group pooled sample or a related reference matrix sample) to
generate fragmentation spectra to aid in unknown identifica-
tion via spectral matching against open-source or in-house
spectral libraries.
Table 2
UHPLC gradient method
Step Total time (min) Flow rate (μL/min) Mobile phase B (% vol)
0 0 250 95
1 0.5 250 95
2 10.5 250 40
3 15.0 250 40
4 17.0 250 95
5 18.0 400 95
6 30.5 400 95
7 31.5 250 95
8 32.0 250 95
Table 3
MS diverter valve configuration
Time (min) Direction of flow
0.0 To waste
0.3 To MS
16.0 To waste
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 365
4. To minimize and eliminate the potential selection of experi-
mentally unrelated background ions in the DDA injection such
as chemical noise, contaminants, or analytes resulting from
sample handling, an exclusion list containing specified masses
can be incorporated in the acquisition method. These masses
will not be triggered for subsequent fragmentation even if the
ion is present in the MS1 spectrum. A list of unrelated back-
ground masses can be generated via an FS solvent blank injec-
tion (e.g., extraction solvent).
5. Table 4provides method parameters for the global MS source
settings applied in both acquisition methods: FS and DDA. FS
method parameters used to generate quantitative profiling data
for all experimental samples are found in Table 5. This method
Table 4
Q Exactive HF method global settings
Source type
Heated electrospray
ionization (H-ESI)
Lock mass Off
MS method duration 16.5 min
Spray voltage (V) +3500
Sheath gas (Arb) 45
Auxiliary gas (Arb) 8
Sweep gas (Arb) 1
Ion transfer tube (capillary) temperature (C) 350
Vaporizer temperature (C) 300
S-lens RF level (%) 35
Table 5
Q Exactive HF FS method conditions
Acquisition mode Full MS
Polarity Positive
Microscans 1
Resolution setting 120K
AGC target 1e6
Maximum injection time (ms) 100
Scan range (m/z) 67–800
Spectrum data type Profile
366 Amanda L. Souza and Gary J. Patti
is denoted as FS. DDA method parameters used to generate
fragmentation spectra for annotation and identification of
unknown compounds are found in Table 6. This method is
denoted as DDA.
3.4 Acquisition
Sequence Order
1. The LC/MS injection sequence for data acquisition is ordered
to allow for system conditioning, qualitative assessment, QC
monitoring, and analyte normalization to the reference matrix.
The sequence table is a list that consists of acquisition informa-
tion to acquire the data set such as sample information, acqui-
sition method, and sample location for the LC autosampler.
Table 7provides a guided list for the LC/MS sequence setup
for sample injection order.
2. Solvent blanks flush and equilibrate the chromatographic col-
umn after instrument standby with no flow. Solvent blanks also
reduce the potential for analyte carryover.
3. The neat reference metabolite mix is injected at the beginning
and end of the run for qualitative evaluation (e.g., retention
time, peak shape, chromatographic resolution).
Table 6
Q exactive HF DDA method conditions
FS Acquisition mode Full MS/dd-MS2 (TopN)
Polarity Positive
Exclusion On
Microscans 1
Resolution setting 120K
AGC target 1e6
Maximum injection time (ms) 100
Scan range (m/z) 67–800
Spectrum data type Profile
MS2 Resolution setting 15 K
AGC target 1e5
Maximum injection time (ms) 50
Loop count 1
Top N 5
Isolation window (m/z) 1.5
Fixed first mass (m/z)50
Collision energy mode Stepped
Collision energies (%) 20, 50, 100
Spectrum data type Profile
Underfill ratio (%) 10
Intensity threshold 1e5
Exclude isotopes On
Dynamic exclusion (sec) 8
If idle... Pick others
Exclusion list parameters Mass (m/z) ppm
Polarity Positive
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 367
Table 7
LC/MS sequence setup for sample injection order
Injection number Sample MS acquisition method Acquisition scheme
1 Solvent Blank 1 FS System Conditioning
2 Solvent Blank 2 FS System Conditioning
3 Reference Metabolite Mix 1 FS Qualitative Assessment
4 Solvent Blank 3 FS System Conditioning
5 Solvent Blank 4 FS System Conditioning
6 Pooled Sample Conditioning 1 FS System Conditioning
7 Pooled Sample Conditioning 2 FS System Conditioning
8 Pooled Sample Conditioning 3 FS System Conditioning
9 Pooled Sample 1 FS MS1 Profiling
10 Experimental Sample 1 FS MS1 Profiling
11 Experimental Sample 2 FS MS1 Profiling
12 Experimental Sample 3 FS MS1 Profiling
13 Experimental Sample 4 FS MS1 Profiling
14 Experimental Sample 5 FS MS1 Profiling
15 Experimental Sample 6 FS MS1 Profiling
16 Experimental Sample 7 FS MS1 Profiling
17 Experimental Sample 8 FS MS1 Profiling
18 Experimental Sample 9 FS MS1 Profiling
19 Experimental Sample 10 FS MS1 Profiling
20 Pooled Sample 2 FS MS1 Profiling
21 Experimental Sample 11 FS MS1 Profiling
... ... ... ...
63 Experimental Sample 50 FS MS1 Profiling
64 Pooled Sample 6 FS MS1 Profiling
65 Solvent Blank 5 FS System Conditioning
66 Solvent Blank ID FS Background Ion ID
67 Pooled Plasma ID 00 FS System Conditioning
68 Pooled Sample ID 01 DDA Unknown Annotation
69 Reference Metabolite Mix 2 FS Qualitative Assessment
70 Solvent Blank 6 FS System Conditioning
368 Amanda L. Souza and Gary J. Patti
4. The pooled QC sample is multipurposed. Initially, this sample
is injected to condition the stationary phase with matrix sam-
ple. Three injections are sufficient to accomplish conditioning.
We recommend injecting the pooled QC sample at the begin-
ning and end of experiments, as well as every ten experimental
samples for compound normalization. Finally, DDA is applied
to a pooled sample to collect fragmentation spectra of sample
relevant ions for unknown compound identification.
5. The run order of experimental samples must be randomized to
ensure that the comparison of metabolite concentrations in the
subsequent analysis is not confounded by temporal drift in
instrument performance.
3.5 Biofluid Sample
Preparation
and Metabolite
Extraction
for Positively Charged
Polar Metabolites
1. Allow frozen samples to thaw on ice (see Note 7). Cover with
foil to shield from light. Large volumes of starting material
require longer wait times.
2. Transfer an appropriate volume of sample into a labeled 1.5 mL
microcentrifuge tube as outlined in Table 8.
3. Transfer the required volume of IS extraction solution to the
sample for metabolite extraction by protein precipitation.
4. Vortex samples for 1 min at 4 C.
5. Centrifuge samples at 10000 rcf for 10 min at 4 C. A protein
pellet is formed at the bottom of the microcentrifuge tube and
the hydrophilic metabolites are partitioned into the
supernatant.
6. Transfer 60 μL of organic supernatant into a labeled LC auto-
sampler vial containing a deactivated (silanized) low volume
glass insert (Thermo Fisher Scientific) (see Note 13). Avoid
disturbing the protein pellet when transferring the supernatant.
Tightly secure a screw cap with a PTFE/silicone septum
(Thermo Fisher Scientific) to seal the autosampler vial.
7. The LC/MS protocol for positively charged polar metabolites
should be employed for these extracted samples. Post-data
processing is then performed on the raw LC/MS data.
Table 8
Metabolite extraction dilution factors by sample type
Biofluid sample Dilution factor Sample volume (μL) IS extraction solution (μL)
Plasma 10 10 90
Urine 5 20 80
CSF 5 20 80
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 369
3.6 Data Analysis 1. Several open-source and commercial software packages are
available for the analysis of untargeted metabolomic LC/MS
data (see Note 14)[12,14,16,37–41]. While each program
follows a series of general processing steps (peak detection,
statistical analysis, and unknown annotation and identifica-
tion), implementation and execution of each step may vary
resulting in slightly different outcomes. Here, Compound Dis-
coverer software v3.1 is employed (Thermo Fisher Scientific)
(see Notes 15 and 16). The following is a step-by-step guide
for generating a Study, an Analysis, and the Results Table in the
data processing program (see Note 17).
2. Launch the New Study and Analysis wizard from initial Start
Page in the application window to begin.
3. Define a Study Name and Study Folder to allocate where the
processed data should reside. In the processing section of this
window, select the preexisting workflow template titled Untar-
geted Metabolomics with Statistics Detect Unknowns with
ID using Online Databases and mzLogic. Click Next to
continue in the wizard.
4. Xcalibur RAW files are added in the Input File Selection
section. Via the Add Files dialog box, browse to and select
the RAW files to be added for processing. Select Open and
then Next to continue.
5. In the Input File Characterization section, a sample type is
assigned to each RAW file. Options include Sample,Blank,
and Identification Only. For metabolomic experiments with
defined characteristics like specified groups (e.g., genetic var-
iants), disease state, time, etc., a Study Factor may be gener-
ated to designate RAW files to a respective group. This step is
necessary for statistical analysis. Multiple study factors can be
applied based on experimental design. The example in Fig. 2
displays the user-defined Study Factor titled Phenotype for an
experiment comparing two groups of Zucker diabetic fatty
(ZDF) rats. Within the Phenotype, two study factors are
assigned as Lean and Fatty. For this study, each RAW file
assigned as a Sample is then further assigned as Lean or Fatty.
Click Next to continue in the wizard.
6. The Sample Group and Ratios section enables differential
analysis by defining sample groups to compare and group ratios
to include in the results file. Sample RAW files are assigned to a
comparison group automatically using the Study Variables sec-
tion or manually using the Manual Ratio Generator. Defined
ratios are listed along with user-defined groups and associated
RAW files per group. Figure 3is an example of the Sample
Group and Ratios section from the ZDF experiment consisting
of three sample groups, Lean, Fatty and n/a, and one ratio of
370 Amanda L. Souza and Gary J. Patti
Fatty compared to Lean. The n/a sample group includes RAW
files assigned as Blank and Identification Only. Files in the n/a
sample group are excluded from the differential analysis. Click
Finish to save the study and close the wizard.
Fig. 2 Input file table with assigned study factors. This table demonstrates how to implement a study factor
within the Input File Characterization page. This example includes the user-defined study factor Phenotype
where experimental samples are allocated as Lean or Fatty
Fig. 3 Defining sample groups and ratios for differential analysis. Using the ZDF rat experiment as an example,
three sample groups are defined automatically by selecting Phenotype in the Study Variables check box. The
comparison ratio is defined by selecting the control group to compare to
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 371
7. Completing the wizard generates a Study page and populates
the Analysis pane. A Study consists of the following four tabs:
Study Definition, Input Files, Samples, and Analysis Results.
The Analysis pane lists the selected processing workflow, the
Results File name, and selected RAW files for processing in
addition to the Run button. When an Analysis is actively
open, two additional tabs are displayed in the Analysis pane:
Grouping & Ratios and Workflows. Figure 4displays the Study
pages, Analysis pages, and Analysis pane in the application.
8. Before submitting the analysis to the Job Queue for processing,
review and modify the processing Workflow Tree as needed
(see Note 18). The Workflow Tree is an assembly of nodes and
edges where each node is a processing function (Fig. 5). Edges
are directional and establish the flow of processing steps. The
program only allows connections for logical associations. The
untargeted metabolomic processing workflow template follows
a series of general steps: peak detection, statistical analysis, and
annotation for unknown compounds. Clicking a node popu-
lates associated parameter settings on the left panel, allowing
users to manually adjust values. Note, selecting Advanced
Parameters enables access to additional settings within a
node. Default parameter settings for the metabolomic template
are a good starting point for processing. Depending on the
data, however, modifications to these settings may be
necessary.
9. In Analysis pane, rename the Results File to the desired file
name. Then submit the analysis to the Job Queue by selecting
the Run button.
Fig. 4 Study page and analysis pane. After completing the wizard, a tab is generated consisting of the study
pages (Study Definition, Input Files, Samples, and Analysis Results), the analysis pages (Grouping & Ratios,
Workflows) and the Analysis pane. Within the Analysis pane, users can assign the Result File name
372 Amanda L. Souza and Gary J. Patti
10. Within the Analysis Results tab, select the completed analysis
from the list and click Open Results. The default results file
layout includes the Chromatogram View,Mass Spectrum
View, and the Main Table of the Compounds tab (Fig. 6).
The main table provides information about each compound
across the dataset (see Notes 19 and 20). Clicking on a specific
compound returns the associated chromatogram overlay and
mass spectrum. Annotations are determined based on the user
input in the Assign Compound Annotations node (see Note
21). Supporting data for the compound annotation is viewed
by opening Related Tables (accessible from the lower left of
the Main Table by selecting Show Related Tables).
Fig. 5 A processing workflow tree. Nodes connected by edges indicate the selected functions for data
processing. The workflow represented here is for the workflow template titled “Untargeted Metabolomics with
Statistics Detect Unknowns with ID using Online Databases and mzLogic.” The workflow includes retention
time alignment, unknown peak detection and ion association, gap filling, detection of background components
unrelated to experimental samples, prediction of elemental composition, ChemSpider database searching,
mzCloud spectral library matching, pathway mapping, and statistical analysis
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 373
11. Key icons displayed at the top of the program provide plotting
capabilities and visualization tools useful for analyzing untar-
geted metabolomic data. Relevant plots include Chromato-
grams, Mass Spectra, Trend Charts, Descriptive Statistics,
Differential Analysis via volcano plot, Principal Component
Analysis (PCA), Partial Least Squares–Discriminant Analysis
(PLS-DA), and Hierarchical Cluster Analysis to associate com-
pounds with similarity via a heat map array, Retention Time
Corrections, and biochemical pathway mapping. The Trend
Chart view enables box-and-whisker charts or a trend line
chart to aid in viewing metabolite differences among samples.
Figure 7displays icons for data review in the application
toolbar.
Fig. 6 Results page default layout. The Main Table of the Compounds tab includes information for each of the
detected compounds including annotation, elemental composition, molecular weight, peak areas, results from
database searching and library matching, and univariate statistical results. Selecting a specific compound
returns the associated chromatogram overlay and mass spectra
Fig. 7 Icons in the application toolbar supporting data review. When reviewing a Results File, data review icons
are active. Selection of an icon brings the view to the front
374 Amanda L. Souza and Gary J. Patti
12. Multiple annotation tools are available in the Compound Dis-
coverer software including prediction of elemental composi-
tion using HRAM spectra, database searching via elemental
composition or m/z via the ChemSpider™chemical structure
database, and spectral library matching to a custom in-house
spectral library (Thermo Scientific™mzVault™offline appli-
cation) or the online Thermo Scientific™mzCloud spectral
library. The mzLogic algorithm can also be applied to prioritize
the list of candidate chemical structures from the ChemSpider
database search based on the experimental fragmentation spec-
tra. Annotation tools are processed independently; however,
priority is given to annotations with matching fragmentation
spectra compared to HRAM MS1 alone. A consensus approach
is applied by considering multiple annotation sources via the
Annotation Sources column. Collectively, these annotation
tools increase confidence in the annotation assignment. For
compound identification, confirmation by purified reference
standard is necessary.
4 Notes
1. Refer to the Metabolomics Quality Assurance & Quality Con-
trol Consortium website for guidelines and recommendations
in quality assurance and quality control measures for untar-
geted metabolomics, mQACC guidelines: https://epi.grants.
cancer.gov/Consortia/mQACC/.
2. Definitions for terms used in this protocol can be found in
Box 1.
3. With any LC/MS analysis, analytical controls should be
incorporated to monitor injection-to-injection variability. Ana-
lytical controls allow for the detection of outliers and other
oddities resulting from the LC/MS analysis, which may be
considered for removal from the data set to subsequently
increase confidence in the qualitative measurements. When
using IS as detailed above, peak areas of stable-labeled meta-
bolites can be evaluated between sample injections. To mini-
mize variation in the IS during sample preparation, the IS
should be added to the extraction solution instead of directly
to the experimental sample. This approach eliminates an addi-
tional step in the preparation procedure.
4. When dissolving purified powder stock of individual reference
metabolites, a combination of water and methanol is sug-
gested. Organic solvent is mixed into the composition to pre-
vent the solution from freezing when stored at 20 C. Storage
in liquid form makes this solution easier to work with practice.
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 375
Depending on the metabolite (and its associated pKa value),
different ratios of water to organic may be necessary as well as
incorporating additives to acidify or basify the solution to
ensure complete dissolution of the compound.
5. Before starting any metabolomic analysis, it is advised to con-
duct a repeatability and reproducibility experiment to establish
analytical precision and method robustness. The pooled QC
sample or a reference matrix sample is ideal for this method
validation step. A repeatability experiment is where one sample
is prepared and extracted once, and then repeatedly injected
onto the system to test variability arising from the LC/MS
procedure alone. Typically, a larger volume of starting material
is needed to support multiple injections. While the number of
repetitions should reflect the actual experiment, this may not
be feasible for metabolomic experiments with >50 samples.
Nevertheless, a good starting point for a repeatability experi-
ment is 20–30 repetitions. A reproducibility experiment is
where the same sample is prepared and extracted multiple
times to evaluate the sample preparation procedure in addition
to analytical precision, which effectively tests the analyst’s abil-
ity to prepare samples. Here, the number of replicates prepared
should reflect the total number of samples prepared at once,
often considered a sample batch. A good starting point to test
reproducibility is 50 replicates.
6. Metabolites are susceptible to degradation by enzymatic and
nonenzymatic reactions. All samples should be stored in a
80 C freezer to reduce the possibility of degradation.
7. Always handle thawed samples on ice. Never leave at room
temperature. Use microcentrifuge tube racks directly on ice
to organize and prepare samples.
8. Minimize the number of freeze–thaw cycles to biofluid sam-
ples. When possible, pre-aliquot necessary volumes in micro-
centrifuge tubes when samples are thawed. This may be
applicable when creating pooled QC samples, long-term refer-
ence matrix samples, and when the experimental sample needs
to be extracted for multiple LC/MS methods (e.g., HILIC and
reversed phase).
9. The sample preparation procedure was designed with the min-
imal necessary number of steps involved. Each additional step
may introduce method variability, which can impact the exper-
imental outcome.
10. There are several common laboratory tools that can minimize
variability. A repeater pipette (Thermo Fisher Scientific) is best
to aliquot the IS extraction solution for a consistent volume. A
multitube vortexer is ideal to mix multiple samples
376 Amanda L. Souza and Gary J. Patti
simultaneously for a defined time period. A positive displace-
ment single-channel pipette (Gilson) is best when transferring
highly organic solutions. For dry-down and reconstitution
steps, an automated solvent evaporation system (Biotage) is
best for rapid removal of solvent of multiple samples
simultaneously.
11. Consistent sample collection, handling, and preparation
improves the quality of the results. Any procedures applied to
one sample should be applied to all samples. This eliminates
potential confounding factors that may impact metabolite con-
centration levels. Such known factors are blood collection
tubes, the analyst who prepared samples, samples being
prepared in different batches.
12. Before starting any LC/MS analysis, evaluate instrument
response using a system suitability sample (neat standard mix-
ture of reference metabolites). Although problems can arise
with both the LC and the MS, the LC tends to require more
attention. A common LC issue is low back pressure caused by a
tubing leak. The chromatographic column may need replace-
ment, perhaps from a blockage or end of column lifetime.
13. Be sure to remove any air bubbles introduced into supernatant
within autosampler vials. This will avoid aspirating air from the
autosampler vial and subsequently introducing air into the LC
system.
14. For untargeted metabolomic data processing, we generally
recommend only comparing the same or similar sample types.
For instance, human plasma samples should be analyzed and
compared to human plasma. Conversely, human plasma sam-
ples should not be compared to urine. While outside the scope
of this protocol, the same idea applies to other sample types like
tissue or cells.
15. Data processing software should be installed on an appropriate
processing computer. It is not recommended to install the
processing software on the LC/MS acquisition computer.
The following configuration is recommended for Compound
Discoverer software v3.1: dual 8-core processor (for example,
2Intel™Xeon™Gold 6134 CPU at 3.20 GHz), 64 GB
RAM, 1 TB SSD hard drive for operating system, second 3 TB
hard drive for data storage, DVD-ROM and USB drive, and
two 27 in. UHD monitors (display monitor resolution of
3840 2160).
16. After software installation, within the taskbar Help option, it is
recommended to execute Communication Tests to verify
access to external mass spectral databases: BioCyc, Kyoto Ency-
clopedia of Genes and Genomes™(KEGG) Pathway,
mzCloud™spectral library, and the ChemSpider database.
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 377
17. Use the Compound Discoverer software Metabolomics Tuto-
rial for guidance when getting started with the program. The
tutorial provides step-by-step guidelines for processing an
example data set titled ZDF consisting of Xcalibur RAW files
generated from ZDF rat plasma.
18. The processing Workflow Tree in the Compound Discoverer
software is the basis for how raw data are processed. Depending
on the objective of the experiment, the user can define the
processing steps. There are key parameters that can be impact-
ful to the results outcome. The Minimum Peak Intensity
within the “Detect Compounds” node establishes a threshold
for chromatographic peak detection. Increasing or decreasing
this value will affect the total number of compounds returned
in the results. Users should take caution when decreasing this
number due to the increased potential for false-positive results
arising from erroneous peaks close to baseline detection.
Figure 8provides a table of recommended values to set as a
starting point depending on which mass spectrometer was used
for data collection.
19. FS mass spectral data of extracted biological samples is highly
dense and complex in nature. A three-dimensional array of
components is generated based on mass-to-charge ratio (m/
z), chromatographic retention time, and signal intensity to
create a complex matrix of information. Unbiased peak detec-
tion reports an exhaustive list of features, where a feature is an
ion measured at a specific retention time with a defined mas-
s-to-charge ratio (m/z). While features are available in the
Compound Discoverer software, compounds are the preferred
outcome because associated features are clustered by deisotop-
ing and de-adducting ions representing the same molecule into
a compound [42,43]. Ion associations are displayed within the
Results View in the Mass Spectrum by ions highlighted in
green.
Fig. 8 Recommended minimum peak intensity range for the Detect Compounds node of the Compound
Discoverer software. Depending on the mass spectrometer used for data acquisition, corresponding peak
intensity values should be selected for unknown peak detection
378 Amanda L. Souza and Gary J. Patti
20. Within the Compound Discoverer software, ions determined
to be background ions based on a solvent blank are filtered
from the Main Table. Turning off the filter returns background
ions to the Main Table.
21. Untargeted metabolomic analysis brings the added challenge
of annotation and identification of unknown compounds.
Guidelines for reporting identification of unknown metabolites
based on analytical measurements are available [44,45]. Multi-
ple annotation resources like database searching and spectral
library matching are available in the Compound Discoverer
software to increase confidence in the annotation assignment
of unknown compounds [46].
Box 1 Definitions of Common Terms
Experimental sample: Complex biological fluid for LC/MS anal-
ysis including human or animal plasma, urine, or CSF.
Pooled sample: A sample representing multiple experimental sam-
ples. The pooled sample is generated by combing the same
volume of biofluid from each experimental sample into one
stock vial.
Reference metabolite mixture: A mixture of metabolites in a
solvent compatible with the initial LC conditions at low
concentration.
Internal standard: A chemical substance, typically a stable-
labeled metabolite, added at a constant amount to each experi-
mental sample used for analytical quality control. Peak area is
plotted to visualize variations and potential outliers.
Long-term reference matrix: A biological sample available in large
quantity, which is separate from and not related to the experi-
mental sample used as an analytical reference for LC/MS assess-
ment. Applied to document datapoints over long periods of time
for historical record keeping. This reference matrix is often
plasma.
Acknowledgments
The authors would like to thank Dr. Clary Clish (Broad Institute of
Harvard and MIT) for discussion and insights that helped develop
this protocol.
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 379
References
1. Oliver SG, Winson MK, Kell DB, Baganz F
(1998) Systematic functional analysis of the
yeast genome. Trends Biotechnol 16:373–378
2. Cho K, Mahieu NG, Johnson SL, Patti GJ
(2014) After the feature presentation: technol-
ogies bridging untargeted metabolomics and
biology. Curr Opin Biotechnol 28:143–148
3. Wishart DS (2019) Metabolomics for investi-
gating physiological and pathophysiological
processes. Physiol Rev 99:1819–1875
4. Tebani A, Bekri S (2019) Paving the way to
precision nutrition through metabolomics.
Front Nutr 6:41
5. Nikolskiy I, Siuzdak G, Patti GJ (2015) Dis-
criminating precursors of common fragments
for large-scale metabolite profiling by triple
quadrupole mass spectrometry. Bioinformatics
31:2017–2023
6. Patti GJ (2011) Separation strategies for untar-
geted metabolomics. J Sep Sci 34:3460–3469
7. Ivanisevic J, Want EJ (2019) From samples to
insights into metabolism: uncovering biologi-
cally relevant information in LC-HRMS meta-
bolomics data. Meta 9(12):308
8. Patti GJ, Yanes O, Siuzdak G (2012) Innova-
tion: metabolomics: the apogee of the omics
trilogy. Nat Rev Mol Cell Biol 13:263–269
9. Dunn WB, Bailey NJC, Johnson HE (2005)
Measuring the metabolome: current analytical
technologies. Analyst 130:606–625
10. Kind T, Tsugawa H, Cajka T, Ma Y, Lai Z,
Mehta SS, Wohlgemuth G, Barupal DK, Sho-
walter MR, Arita M, Fiehn O (2017) Identifi-
cation of small molecules using accurate mass
MS/MS search. Mass Spectrom Rev 37
(4):513–532
11. Dunn WB, Ellis DI (2005) Metabolomics: cur-
rent analytical platforms and methodologies.
Trends Anal Chem 24(4):285–294
12. Mahieu NG, Genenbacher JL, Patti GJ (2016)
A roadmap for the XCMS family of software
solutions in metabolomics. Curr Opin Chem
Biol 30:87–93
13. Olivon F, Grelier G, Roussi F, Litaudon M,
Touboul D (2017) MZmine 2 data-
preprocessing to enhance molecular network-
ing reliability. Anal Chem 89:7836–7840
14. Lommen A, Kools HJ (2012) MetAlign 3.0:
performance enhancement by efficient use of
advances in computer hardware. Metabolomics
8:719–726
15. Dunn WB, Broadhurst DI, Atherton HJ,
Goodacre R, Griffin JL (2011) Systems level
studies of mammalian metabolomes: the role of
mass spectrometry and nuclear magnetic reso-
nance spectroscopy. Chem Soc Rev
40:387–426
16. Uppal K, Soltow QA, Strobel FH, Pittard WS,
Gernert KM, Yu T, Jones DP (2013) xMSana-
lyzer: automated pipeline for improved feature
detection and downstream analysis of large-
scale, non-targeted metabolomics data. BMC
Bioinformatics 14:15
17. Schwaiger M, Schoeny H, El Abiead Y,
Hermann G, Rampler E, Koellensperger K
(2019) Merging metabolomics and lipidomics
into one analytical run. Analyst 144:220–229
18. Naser FJ, Mahieu NG, Wang L, Spalding JL,
Johnson SL, Patti GJ (2018) Two complemen-
tary reversed-phase separations for comprehen-
sive coverage of the semipolar and nonpolar
metabolome. Anal Bioanal Chem 410
(4):1287–1297
19. Contrepois K, Jiang L, Snyder M (2015) Opti-
mized analytical procedures for the untargeted
metabolomic profiling of human urine and
plasma by combining hydrophilic interaction
(HILIC) and reverse-phase liquid chromatog-
raphy (RPLC)-mass spectrometry. Mol Cell
Proteomics 14(6):1684–1695
20. Ivanisevic J, Zhu ZJ, Plate L, Tautenhahn R,
Chen S, O’Brien PJ, Johnson CH, Marletta
MA, Patti GJ, Siuzdak G (2013) Toward
‘omic scale metabolite profiling: a dual
separation-mass spectrometry approach for
coverage of lipid and central carbon metabo-
lism. Anal Chem 85(14):6876–6884
21. Roberts LD, Souza AL, Gerszten RE, Clish CB
(2012) Targeted metabolomics. Curr Protoc
Mol Biol 98(1):1–34
22. Snyder NW, Khezam M, Mesaros CA,
Worth A, Blair IA (2013) Untargeted metabo-
lomics from biological sources using ultraper-
formance liquid chromatography-high
resolution mass spectrometry (UPLC-
HRMS). J Vis Exp 75:1–8
23. Want EJ (2018) LC-MS untargeted analysis.
In: Theodoridis AG, Gika HG, Wilson ID
(eds) Metabolic profiling, methods and proto-
cols. Humana, New York, NY, pp 99–116
24. University of California San Diego. Metabolo-
mics workbench, general protocols. https://
www.metabolomicsworkbench.org/
protocols/general.php. Accessed 12 Jan 2019
25. European Molecular Biology Laboratory
(2019) Metabolomics core facility, protocols
used for LC-MS analysis. https://www.embl.
de/mcf/metabolomics-core-facility/
protocols/. Accessed 12 Jan 2019
380 Amanda L. Souza and Gary J. Patti
26. Saigusa D, Okamura Y, Motoike IN, Katoh Y,
Kurosawa Y, Saijyo R, Koshiba S, Yasuda J,
Motohashi H, Sugawara J, Tanabe O,
Kinoshita K, Yamamoto M (2016) Establish-
ment of protocols for global metabolomics by
LC-MS for biomarker discovery. PLoS One 11
(8):e0160555
27. Gika HG, Zisi C, Theodoridis G, Wilson ID
(2016) Protocol for quality control in meta-
bolic profiling of biological fluids by U
(H)PLC-MS. J Chromatogr B Anal Technol
Biomed Life Sci 1008:15–25
28. Knee JM, Rzezniczak TZ, Barsch A, Guo KZ,
Merritt TJS (2013) A novel ion pairing
LC/MS metabolomics protocol for study of a
variety of biologically relevant polar metabo-
lites. J Chromatogr B Anal Technol Biomed
Life Sci 936:63–73
29. Esterhuizen K, van der Westhuizen FH, Louw
R (2017) Metabolomics of mitochondrial dis-
ease. Mitochondrion 35:97–110
30. Barshop BA (2004) Metabolomic approaches
to mitochondrial disease: correlation of urine
organic acids. Mitochondrion 4:521–527
31. Shaham O, Slate NG, Goldberger O, Xu Q,
Ramanathan A, Souza AL, Clish CB, Sims
KB, Mootha VK (2010) A plasma signature of
human mitochondrial disease revealed through
metabolic profiling of spent media from
cultured muscle cells. Proc Natl Acad Sci U S
A 107(4):1571–1575
32. Leoni V, Strittmatter L, Zorzi G, Zibordi F,
Dusi S, Garavaglia B, Venco P, Caccia C,
Souza AL, Deik A, Clish CB, Rimoldi M,
Ciusani E, Bertini E, Nardocci N, Mootha
VK, Tiranti V (2012) Metabolic consequences
of mitochondrial coenzyme A deficiency in
patients with PANK2 mutations. Mol Genet
Metab 105(3):463–471
33. Yao CH, Wang R, Wang Y, Kung CP, Weber
JD, Patti GJ (2019) Mitochondrial fusion sup-
ports increased oxidative phosphorylation dur-
ing cell proliferation. Elife 8:pii:e41351
34. Mandal R, Chamot D, Wishart DS (2018) The
role of the Human Metabolome Database in
inborn errors of metabolism. J Inherit Metab
Dis 41:329–336
35. Feuchtbaum L, Carter J, Dowray S, Currier RJ,
Lorey F (2012) Birth prevalence of disorders
detectable through newborn screening by
race/ethnicity. Genet Med 14:937–945
36. Thermo Fisher Scientific, Exactive Series
Operating Manual, BRE0012255, Revision A,
April 2017. https://assets.thermofisher.com/
TFS-Assets/CMD/manuals/man-
bre0012255-exactive-series-manbre0012255-
en.pdf. Accessed 16 Feb 2019
37. Tsugawa H, Cajka T, Kind T, Ma Y, Higgins B,
Ikeda K, Kanazawa M, VanderGheynst J,
Fiehn O, Arita M (2015) MS-DIAL: data-inde-
pendent MS/MS deconvolution for compre-
hensive metabolome analysis. Nat Methods
12:523–526
38. Chong J, Soufan O, Li C, Caraus I, Li S,
Bourque G, Wishart DS, Xia J (2018) Meta-
boAnalyst 4.0: towards more transparent and
integrative metabolomics analysis. Nucleic
Acids Res 46:W486–W494
39. Mahieu NG, Spalding J, Patti GJ (2015)
Warpgroup: increased precision of metabolo-
mic data processing by consensus integration
bound analysis. Bioinformatics 32:268–275
40. Blazenovic I, Kind T, Ji J, Fiehn O (2018)
Software tools and approaches for compound
identification of LC-MS/MS data in metabo-
lomics. Meta 8(2):31
41. Stanstrup J, Broeckling CD, Helmus R,
Hoffmann N, Mathe E, Naake T, Nicolotti L,
Peters K, Rainer J, Salek RM, Schulze T, Schy-
manski EL, Stravs MA, Thevenot EA,
Treutler H, Weber RJM, Willighagen E,
Witting M, Neumann S (2019) The metaRbo-
lomics toolbox in bioconductor and beyond.
Meta 9(10):200
42. Souza A, Tautenhan R (2018) Features or
compounds? A data reduction strategy for
untargeted metabolomics to generate mean-
ingful data (Report no. TN65204-EN
0418S). Thermo Fisher Scientific. https://
assets.thermofisher.com/TFS-Assets/CMD/
Technical-Notes/tn-65204-lc-ms-untargeted-
metabolomics-tn65204-en.pdf. Accessed
16 Feb 2020
43. Mahieu NG, Patti GJ (2017) Systems-level
annotation of a metabolomics data set reduces
25000 features to fewer than 1000 unique
metabolites. Anal Chem 89:10397–10406
44. Fiehn O, Robertson D, Griffin J, van der
Werf M, Nikolau B, Morrison N, Sumner LW,
Goodacre R, Hardy NW, Taylor C, Fostel J,
Kristal B, Kaddurah-Daouk R, Mendes P, van
Ommen B, Lindon JC, Sansone S (2007) The
metabolomics standards initiative (MSI).
Metabolomics 3:175–178
45. Sumner LW, Amberg A, Barrett D, Beale MH,
Beger RD, Daykin CA, Fan T, Fiehn O,
Goodacre R, Griffin JL, Hankemeier T, Hardy
NW, Harnly J, Higashi RM, Kopka J, Lane AN,
Lindon JC, Marriott P, Nicholls AW, Reily
MD, Thaden J, Viant M (2007) Proposed
A Protocol for Untargeted Metabolomic Analysis: From Sample Preparation to... 381
minimum reporting standards for chemical
analysis. Metabolomics 3:231–241
46. Souza A, Ntai I, Tautenhan R (2018) Acceler-
ated unknown compound annotation with
confidence: from spectra to structure in untar-
geted metabolomics experiments (Report
no. AN65362-EN 1218M). Thermo Fisher
Scientific. https://assets.thermofisher.com/
TFS-Assets/CMD/Application-Notes/an-
65362-ms-compound-annotation-an65362-
en.pdf. Accessed 16 Feb 2020
382 Amanda L. Souza and Gary J. Patti
Chapter 28
A Method for Analysis of Nitrotyrosine-Containing Proteins
by Immunoblotting Coupled with Mass Spectrometry
Matej Kohutiar and Adam Eckhardt
Abstract
Nitrotyrosine formation is caused by presence of reactive oxygen and nitrogen species. Nitration is a very
selective process leading to specific modification of only a few tyrosines in protein molecule. 2D electro-
phoresis and western blotting techniques coupled with mass spectrometry are common methods used in
analysis of proteome. Here we describe protocol for analysis of peroxynitrite-induced protein nitration in
isolated mitochondria. Mitochondrial proteins are separated by 2D electrophoresis and transferred to
nitrocellulose membrane. Membranes are then incubated with antibodies against nitrotyrosine. Positive
spots are compared with corresponding Coomassie-stained gels, and protein nitration is confirmed with
mass spectrometry techniques.
Key words Mitochondria, Nitrotyrosine, 2D electrophoresis, Immunoblotting, Mass spectrometry
1 Introduction
Nitrotyrosine formation is posttranslational modification occurring
in the presence of nitrating agents. Nitration is not random and
leads to selective modification of some tyrosine residues in the
polypeptide chain [1,2]. Under physiological conditions, only a
few tyrosines are nitrated [3]. Mitochondria are major source of
reactive oxygen, and nitrogen species in the cell and their overpro-
duction leads to oxidative damage [4–6]. Inner mitochondrial
membrane contains high amounts of hydrophobic proteins in com-
parison with other membranes in the cell [7,8]. This can be a
limiting factor for their separation during electrophoresis and
membrane transfer in more polar solvents and buffers.
2D electrophoresis is a proteomic technique commonly cou-
pled with mass spectrometry. Separated proteins can be blotted to
membrane and analyzed with various probes. In our experiment,
we exposed mitochondria isolated from bovine heart to peroxyni-
trite [9–11]. The goal was to identify proteins the most endangered
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_28,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
383
by nitration process. For this purpose, mitochondrial proteins were
resolved by 2D electrophoresis using technique described in [12]
with some modifications [10,13]. In the first dimension, sample
was focused on 13 cm strip (pH 3–10). In the second dimension,
sodium dodecyl sulfate polyacrylamide gel electrophoresis
(SDS-PAGE) was used. Resolved proteins were then transferred
to nitrocellulose membrane and incubated with antibodies against
nitrotyrosine. Nitrated proteins were detected by chemilumines-
cent reaction. Immunodetection of nitrotyrosine is more sensitive
than MS analysis. This explains why there are more proteins found
to be nitrated in immunoblots. On the other hand, cross reactivity
cannot be excluded, and protein nitration should be confirmed by
MS methods. In our experiment, nanoscale liquid chromatography
coupled to tandem mass spectrometry (nLC-MS/MS) was used.
Isolation and identification of proteins in spots was performed as
described in [14] with some modifications [10].
Figure 1shows 2D electrophoresis and immunoblots of pro-
tein sample incubated with peroxynitrite for 1 h. Immunoblots are
shown as negatives to enable better comparison with 2D gel. Data
obtained from MS analysis are summarized in Table 1. In described
protocol, we present a lot of recommendations and practical tips to
obtain reproducible results from analysis of sample containing
nitrated mitochondrial proteins.
2 Materials
Use analytical grade reagents and ultrapure water (purified deio-
nized water). Avoid vigorous stirring during preparation of solu-
tions with detergents (SDS, TWEEN) to minimize bubble
formation.
2.1 2D
Electrophoresis
1. Resolving gel buffer: 1.5 M TRIS–HCl, 0.4% SDS, pH 8.8.
Dissolve 18.15 g of TRIS base and 0.4 g of SDS in ~80 mL of
water. Adjust pH with 30% HCl (~2 mL 35% HCl). Diluted
HCl (15%) can be used closer to the final pH. Fill up to 100 mL
with water. Store at 4 C.
2. Acrylamide/bis solution: 29.2% acrylamide, 0.8% N,N0-meth-
ylenebisacrylamide. Dissolve 29.2 g of acrylamide in ~50 mL
water. Then add 0.8 g of N,N0-methylenebisacrylamide and
after dissolving, fill up the volume to 100 mL. Use acrylamide
solution within 1 month. Prepared amount is enough for six
gels (160 140 1 mm). Caution: acrylamide is neurotoxin!
Use gloves and work in hood.
3. Ammonium persulfate: 10% solution in water (see Note 1).
4. N,N,N,N0-tetramethylethylenediamine (TEMED).
384 Matej Kohutiar and Adam Eckhardt
5. Running buffer: TRIS-taurine with SDS. Dissolve 18 g of
TRIS base and 150 g of taurine in ~700 mL water. Add 6 g
of SDS and wait until dissolved. Fill up the volume to 1 L with
water and store at 4 C. Before use, dilute buffer with 5 L of
cold water (see Note 2).
6. Trichloroacetic acid (TCA): 10% (w/v) trichloroacetic acid,
20 mM dithiothreitol (DTT) in acetone. Dissolve 1 g of tri-
chloroacetic acid in 8 mL of cold acetone. Then add 30.8 mg of
DTT and fill up to 10 mL with acetone.
Fig. 1 2D electrophoresis of mitochondrial proteins isolated from bovine heart. Mitochondria were exposed to
0.5 mM peroxynitrite for 1 h. Spots indicated with circles have been subjected to nLC-MS/MS. Spots indicated
by rectangles were chosen as a control for nLC-MS/MS. (a) control (silver staining); (b) Immunoblot of control;
(c) 0.5 mM peroxynitrite for 1 h (Coomassie staining); (d) Immunoblot of nitrated proteins. (Reproduced from
[10] with permission from Physiol Res)
Immunoblotting Coupled with Mass Spectrometry 385
Table 1
List of proteins identified in bovine heart mitochondria resolved by 2D electrophoresis [10]
Spot
Accession
number Protein
NT
position Nitration
Protein
score
Sequence
coverage [%]
1 P02769 Serum albumin 355 Yes 1582.6 35.9
2 P00829 ATP synthase subunit beta,
mitochondrial
395 Yes 2292.9 67.2
3 P31800 Cytochrome b-c1 complex subunit
1, mitochondrial
Possible 839.2 25.2
4 P31800 Cytochrome b-c1 complex subunit
1, mitochondrial
Possible 649.5 19.2
5 P17694 NADH dehydrogenase [ubiquinone]
iron-sulfur 2, mitochondrial
Possible 1141.8 47.3
6 P49410 Elongation factor Tu, mitochondrial Possible 1002.5 34.1
7 P60712 Actin, cytoplasmic 1 Possible 1073.4 8.8
8 P04394 NADH dehydrogenase [ubiquinone]
flavo 2, mitochondrial
Possible 996.8 47.8
9 Q3T149 Heat shock beta-1 Possible 884.6 47.8
10 P31800 Cytochrome b-c1 complex subunit
1, mitochondrial
Possible 515.8 19.4
11 P42028 NADH dehydrogenase [ubiquinone]
iron-sulfur 8, mitochondrial
Possible 527.8 31.6
12 P11966 Pyruvate dehydrogenase E1
component subunit beta,
mitochondrial
63; 67 Yes 1039.6 35.7
13 Q29RK1 Citrate synthase, mitochondrial 381 Yes 646.9 24.7
14 Q29RZ0 Acetyl-CoA acetyltransferase,
mitochondrial
326 Yes 174.8 15.6
A P15690 NADH-ubiquinone oxidoreductase
75 kDa subunit, mitochondrial
No 1627.3 41.0
B P31081 60 kDa heat shock, mitochondrial No 2460.4 58.3
C P85100 Myosin light chain 3 No 753.0 57.8
D P23709 NADH dehydrogenase [ubiquinone]
iron-sulfur 3, mitochondrial
No 1144.2 60.3
E P23709 NADH dehydrogenase [ubiquinone]
iron-sulfur 3, mitochondrial
No 277.0 21.5
F Q3SZE5 Myosin regulatory light chain
2, ventricular/cardiac muscle
isoform
No 573.8 52.4
(continued)
386 Matej Kohutiar and Adam Eckhardt
7. 20 mM DTT in acetone.
8. Bromophenol blue: 0.05% solution in water.
9. Rehydration buffer/lysis buffer: 6 M urea, 2 M thiourea, 4%
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfo-
nate hydrate (CHAPS), 3% TRITON X-100, 0.4% DTT, 0.5%
ampholyte solution. Dissolve 18 g of urea and 7.6 g of thiourea
in 20 mL water. Add 2 g of CHAPS, 0.2 g of DTT, 1.5 mL of
TRITON-X, 0.25 mL of Pharmalytes (pH 3–10), and 2 mL of
bromophenol blue solution. Fill up to 50 mL with water. Lysis
buffer can be stored frozen in aliquots at 20 C for months.
10. Equilibration buffer I: 6 M urea, 0.375 M TRIS–HCl, 2% SDS,
20% glycerol, 2% DTT, pH 8.8. Dissolve 18 g of urea in 20 mL
water. Add 2.27 g of TRIS base and 1 g of SDS. Add 10 mL
glycerol and adjust pH with diluted HCl to 8.8. Fill up to
50 mL. Buffer can be stored in aliquots at 20 C for months.
Just before use, dissolve 0.2 g of DTT in 10 mL of fresh/
thawed aliquot. Use 10 mL per one strip.
11. Equilibration buffer II: 6 M urea, 2 M thiourea, 0.375 M
TRIS–HCl, 2% SDS, 20% glycerol, 2% DTT, 2.5% iodoaceta-
mide, pH 7.5. Dissolve 18 g of urea and 7.6 g of thiourea in
20 mL water. Then add 2.27 g of TRIS–HCl and 1 g of SDS.
Add 10 mL of glycerol and adjust pH with diluted HCl to 7.5.
Fill up to 50 mL. Buffer can be stored in 10 mL aliquots at
20 C for months. Just before use, dissolve 0.25 g of iodoa-
cetamide in 10 mL of fresh/thawed aliquot. Use 10 mL per
one strip.
12. Agarose: 0.7% solution in running buffer. Weigh 0.35 g of
agarose and pour over around 40 mL of running buffer. Add
0.3 mL of 0.05% bromophenol blue. Heat beaker until solu-
tion is cleared. Transfer to graduated cylinder and fill up to
50 mL with running buffer.
Table 1
(continued)
Spot
Accession
number Protein
NT
position Nitration
Protein
score
Sequence
coverage [%]
G P00426 Cytochrome c oxidase subunit 5A,
mitochondrial
No 359.7 26.3
H P00428 Cytochrome c oxidase subunit 5B,
mitochondrial
No 328.6 35.7
I P13620 ATP synthase subunit d,
mitochondrial
No 979.5 68.9
Accession number—specific protein number in database http://www.uniprot.org [15]; NT position—position of nitra-
tion in protein sequence
Immunoblotting Coupled with Mass Spectrometry 387
13. Protein standards were prepared according to the manufac-
turer’s instructions (low-range standards from Bio-Rad).
14. Paraffin oil.
15. Immobiline DryStrip GE Healthcare 13 cm pH 3–10.
16. Multiphor II (Amersham Biosciences).
17. Hoefer SE 600 Ruby electrophoretic unit (Amersham
Biosciences).
2.2 Immunoblotting 1. Transfer buffer: 0.015 M Na
2
B
4
O
7
, 0.1 M H
3
BO
3
, 0.1% SDS,
pH 8.3–8.4. Dissolve 5.72 g of Na
2
B
4
O
7
· 10H
2
O, 6.48 g of
H
3
BO
3
, and 1 g of SDS in ~900 mL water. The pH should be
between 8.3 and 8.4. Adjust pH with solid H
3
BO
3
if necessary.
Fill up to 1000 mL with water.
2. Concentrated phosphate-buffered saline (PBS): Dissolve 160 g
of NaCl, 4 g of KCl, 72.6 g of Na
2
HPO
4
· 12H
2
O and 4.8 g of
KH
2
PO
4
in water. Fill up to 1000 mL. Prepared PBS buffer is
20concentrated.
3. PBST: PBS with 0.1% TWEEN. Pipette 2 mL of TWEEN into
2 L volumetric flask with 200 mL of concentrated PBS. Dilute
with water and gently mix on the stirrer. Fill up to the mark
with water (see Note 3).
4. Blocking solution: 5% nonfat dry milk in PBST. Dissolve 7.5 g
of nonfat dry milk in 150 mL of PBST.
5. Diluent solution: 1% nonfat dry milk in PBST. Dissolve 0.6 g of
nonfat dry milk in 60 mL of PBST (see Note 4).
6. Primary antibody: 1:5000 in diluent solution. Primary anti-
body was prepared in our laboratory [16–19], but commercial
probes can be also used. Before immunoblotting, optimal dilu-
tions should be found (see Note 5).
7. Secondary antibody: Polyclonal rabbit anti-mouse immuno-
globulins/HRP (Dako, Glostrup, Denmark) 1:10,000 in dilu-
ent solution.
8. Hybond
®
ECL™nitrocellulose membranes.
9. Semidry transfer unit TE 70 (Amersham Biosciences).
2.3 Staining 1. Coomassie Brilliant Blue: Mix 1000 mL of methanol, 200 mL
of glacial acetic acid, and 800 mL of water. Prepared solution
can be used for fixation, staining, and also for destaining. Fix
gel for 1 h in ~150 mL of mixture. In the meantime, prepare
0.3% Coomassie by dissolving 0.45 g of dye in 150 mL of
mixture. Immerse gel in dye for 1 h. Gradually destain the gel
as required (see Note 6).
2. Silver staining: For silver staining, technique described in [20]
was used.
388 Matej Kohutiar and Adam Eckhardt
3. Ponceau staining: Dissolve 0.2 g of Ponceau S in 8 mL of
glacial acetic acid and fill up to 200 mL with water. Stain the
membrane with transferred proteins for 30 min. Use water as
destaining solution. Ponceau S staining is reversible.
2.4 Chemilumine-
scence Detection
1. Working solution: 2.5 mM luminol, 0.1 mM p-coumaric acid,
0.009% H
2
O
2
in 0.1 M TRIS–HCl. Dissolve 0.45 g of luminol
in 10 mL of DMSO. Pipette 1 mL aliquots and store them at
20 C until needed. Dissolve 0.074 g of p-coumaric acid in
5 mL of DMSO. Pipette 0.5 mL aliquots and store them at
20 C. Dissolve 1.2 g of TRIS base in 80 mL water, adjust pH
to 8.8, and fill up the volume to 100 mL with water. Just before
use, mix luminol and p-coumaric acid aliquot with 50 mL of
buffer. Add 300 μLof3%H
2
O
2
to remaining 50 mL of
prepared buffer. Mix solutions together and use immediately.
2. Charge-coupled device camera (Kodak Image Station 4000R).
2.5 In-Gel Digestion
and nLC-MS/MS
Analysis
1. Destaining buffer: 0.1 M NH
4
HCO
3
/acetonitrile (1:1; v/v).
2. Reduction buffer: 10 mM DTT in 0.1 M NH
4
HCO
3
.
3. Alkylation buffer: 55 mM iodoacetamide in 0.1 M NH
4
HCO
3
.
4. Lysis buffer: Trypsin treated with N-tosyl-L-phenylalanine
chloromethyl ketone (TPCK) 13 μg/mL in 0.1 M NH
4
HCO
3
.
5. Extraction buffer: 5% formic acid/acetonitrile (1:2; v/v).
6. Sample buffer: 2% formic acid.
7. Mobile phase: A (water) and B (acetonitrile), both containing
0.1% (v/v) formic acid.
8. Precolumns: NS-MP-10 Biosphere C18 and NS-AC-11-C18
Biosphere C18 (NanoSeparations, Nieuwkoop, Netherlands).
3 Methods
3.1 Sample
Preparation
and Isoelectric
Focusing
1. Precipitate ~0.5 mL of the sample with 2 mL of TCA for
45 min at 20 C. Centrifuge at 10,500 gfor 10 min at
4C. Wash the pellet three times with DTT in acetone and
centrifuge after each wash step (see Note 7).
2. Dissolve pellet in lysis buffer at room temperature (see Note 8).
Centrifuge again if necessary and for further step use
supernatant.
3. Rehydrate strip in rehydration buffer according to the manu-
facturer’s instructions (see Note 9).
4. Load sample in alkaline region using cup loading method (see
Note 10). Cover strips with liquid paraffin.
Immunoblotting Coupled with Mass Spectrometry 389
5. Carry out isoelectric focusing: 1 h at 100 V, 2 h at 300 V, 1 h at
1000 V, and 16 h at 3500 V (see Note 11).
6. After focusing, remove strips and wash them from excess par-
affin. Strips can be frozen and stored at 20 C until needed.
7. Incubate strip for 20 min in equilibration buffer I. Rinse
quickly with water and equilibrate in equilibration buffer II
(see Note 12). Rinse with water.
3.2 SDS-PAGE 1. For 12% resolving gel, mix 18.2 mL of resolving buffer, 28 mL
of acrylamide/bis solution, and 23.8 mL water. Degas with
helium for 10–15 min. Add 35 μL of TEMED and 350 μLof
ammonium persulfate. Mix gently. Immediately cast the gel
within gel cassettes. Overlay with butanol and let polymerize
at 8 C overnight (see Note 13).
2. Apply strip on prepared gel. Put a small square of filter paper
soaked with protein standards in the corner. Cover strip with
~3 mL of agarose (see Note 14).
3. Carry out electrophoresis in TRIS-taurine system for 6–8 h at
8C (180 V, 80 mA), until the blue dye front reaches the end
of the gel.
4. After electrophoresis is complete, transfer the gel from plate to
a container with transfer buffer. Cover the container and
remove excess SDS by gentle shaking. Change buffer after
20 min and shake again for 10 min.
5. Stain the gel not intended to transfer with Coomassie or silver.
3.3 Immunoblotting
and Chemilumine-
scence Detection
1. Cut nitrocellulose membrane and filter paper exactly to the
gel size.
2. Prepare semidry transfer unit. Rinse cathode and anode with
water. Put the plastic mask on the cathode and form
gel-membrane sandwich (see Note 15).
3. Put exactly on the top of the mask ten pieces of soaked filter
cuts. Remove excess transfer buffer by creating a gentle pres-
sure using clean glass test tube. This procedure also removes air
bubbles from the sandwich. Put another ten cuts and repeat
procedure. Align nitrocellulose membrane exactly on the top.
Do not touch the membrane; work with gloves to avoid con-
tamination of surface. Lay down gel and put on the top a few
filter cuts. Roll the surface to remove excess buffer and air
bubbles. Form sandwich using remaining cuts.
4. Carry out protein transfer at current density 0.8 mA cm
2
for
90 min at 8 C(see Note 16).
5. After blotting is completed, transfer efficiency can be checked
by reversible staining with Ponceau S.
390 Matej Kohutiar and Adam Eckhardt
6. Put nitrocellulose membrane into a container with blocking
buffer. Incubate overnight at 8 C. Afterwards, wash mem-
brane twice with PBS buffer and incubate again in fresh block-
ing buffer for 1 h at room temperature, shaking occasionally
(see Note 17).
7. Wash membrane with PBST for 15 min and then twice for
5 min.
8. Incubate membrane with anti-nitrotyrosine antibody for 60 min
shaking it very gently at room temperature (see Note 18).
9. Wash membrane with PBST for 15 min and then twice for
5 min.
10. Incubate membrane with secondary antibody for 60 min shak-
ing it very gently at room temperature.
11. Wash membrane with PBST for 15 min and then twice for
5 min.
12. Pour chemiluminescent reagent over the membrane and incu-
bate for 2 min.
13. Record fluorescence signal using charge-coupled device camera
(see Note 19).
3.4 In-Gel Digestion 1. Excise protein spots (~1–3 mm in diameter) from Coomassie-
stained gels (see Note 20).
2. Bleach the spots by occasional shaking in 100 μL of destaining
buffer for 1 h.
3. Dehydrate gel pieces in 300 μL of acetonitrile for 5 min.
Remove acetonitrile and dry pieces in a vacuum centrifuge for
5 min.
4. Add 50 μL of reduction buffer and incubate for 1 h at 56 C
with gently shaking.
5. Dehydrate gel pieces in 300 μL of acetonitrile for 5 min.
Remove acetonitrile and dry gel pieces in a vacuum centrifuge
for 5 min.
6. Add 50 μL of alkylation buffer and incubate for 45 min in
the dark.
7. Dehydrate gel pieces in 300 μL of acetonitrile for 5 min.
Remove acetonitrile and dry gel pieces in a vacuum centrifuge
for 5 min.
8. Cool samples at 4 C and fully submerge dehydrated gel in
50–100 μL of digestion buffer. After 1 h of hydration, place
samples to air circulation thermostat and incubate overnight at
37 C(see Note 21).
Immunoblotting Coupled with Mass Spectrometry 391
9. Extract the resulting tryptic peptides by 10 min sonication and
afterwards, aspirate liquid to new labeled tube.
10. Fully submerge hydrated gel in 50–100 μL of extraction buffer.
Sonicate for 10 min and after then, aspirate liquid to the same
tube used in the step 9.
11. Repeat the step 10 once more.
12. Concentrate the solution to dryness in a vacuum centrifuge or
lyophilizator. Dried extracts can be stored at 80 C until
needed (see Note 22).
3.5 Analysis
of Tryptic Digests
from Spots
with nLC-MS/MS
1. Dissolve dried protein digests in 25 μL sample buffer.
2. Vortex for 5 s and sonicate for 10 min.
3. Centrifuge at 12,000 gfor 10 min.
4. Carefully aspirate 20 μL of supernatant to inserts for vials for
MS. Be careful not to pipette any parts of pellet.
5. Inject 5 μL of peptide mixture on a precolumn. In our experi-
ment, NS-MP-10 Biosphere C18 and NS-AC-11-C18 Bio-
sphere C18 column were used.
6. Separate peptides via linear gradient between mobile phase A
and B. Start by running the system with 5% mobile phase B,
followed by gradient elution to 30% B at 70 min. After that,
follow by gradient elution to 50% B in 10 min and then by
gradient elution to 100% B in 8 min. Finally, elute the column
with 100% B for 2 min. Before next run, equilibrate the col-
umns with 5% B for 10 min.
7. Use nano-electrospray ionization (ESI) in positive mode. ESI
voltage: +4.5 kV, scan time: 1.3 Hz, nebulizer pressure: 40 kPa,
drying gas N
2
: 1 L/min, drying gas temperature: 160 C.
8. Scan a mass-to-charge ratio (m/z) from 100 to 2200 and use a
scan time 1 s during the whole process. As the reference (inter-
nal mass lock), use the monocharged ion m/z1221.9906
C
24
H
19
F
36
N
3
O
6
P
3
.
9. Process data using ProteinScape software as described in
[21]. Identify proteins by correlation of tandem mass spectra
to the NCBInr proteins and/or SwissProt using the MASCOT
searching engine (www.matrixscience.com)[22]. Restrict the
proper taxonomy (e.g., Bos Taurus) to remove the redundancy
of protein identifications. Choose the enzyme parameter as
“trypsin.” Allow one missed cleavage and use an initial peptide
mass tolerance of 10.0 ppm for MS and 0.05 Da. Assume
tyrosine to be nitrated, cysteine to be carbamidomethylated,
proline and lysine to be hydroxylated, methionine to be oxi-
dated (all these possible modifications set variable). Set charge
of monoisotopic peptide to 1+, 2+, and 3+. Select the Peptide
392 Matej Kohutiar and Adam Eckhardt
Decoy option during the data search process to remove false-
positive results (see Note 23).
10. Accept only significant hits (mascot score for peptide 20 and
for protein 80) as defined by MASCOT probability analysis
(p<0.05) and identify at least two unique peptides to match
for each protein.
4 Notes
1. Ammonium persulfate should be always prepared fresh.
2. Running buffer can be used repeatedly.
3. Use wide tip because TWEEN is very viscous liquid.
4. During incubation, membrane should be completely immersed
in antibody solution. In our laboratory, we use trays exactly
corresponding to the membrane dimensions.
5. In our experiment, we incubated nitrated BSA with primary
antibody (1:1000, 1:2500, 1:5000, 1:10,000, 1:50,000) fol-
lowed by incubation with secondary antibody (1:5000,
1:10,000, 1:25,000, 1:50,000, 1:100,000). The best results
were obtained for 1:5000 dilution of primary and 1:10,000
dilution of secondary antibody.
6. Coomassie can be removed from destaining solution by active
carbon. Use again after filtration.
7. Protein concentration in our sample was 15–20 mg/mL.
Washing the sample with cold acetone will remove lipids.
8. Hydrophobic mitochondrial proteins require more time. Dis-
solution can take 0.5–3 h.
9. Rehydrate strip overnight at room temperature.
10. For application sample in or cup loading method can be used.
Sample in method is better for smaller amounts of protein
(hydrophobic proteins in 2-DE are better resolved). Cup load-
ing method is better for higher amounts of protein, but the
sample should be applied in alkaline region of strip. The
amount of loaded protein is limited by its solubility in lysis
buffer and also by capacity of the cup. Avoid buffer crystalliza-
tion in the cup by adding a few drops of liquid paraffin on its
surface.
11. Keep the temperature during focusing at 20 C using circular
thermostat to avoid protein carbamylation.
12. Just before use, dissolve DTT and iodoacetamide in equilibra-
tion buffers. Put strip into glass test tube and pour buffer.
Equilibrate on orbital shaker in lying position.
Immunoblotting Coupled with Mass Spectrometry 393
13. In case of need, degassing can be also carried out by using
water aspirator or vacuum pump with continual stirring of
acrylamide/bis solution at low speed. Degas at least for
30–45 min. We recommend preparing acrylamide gels a day
before planned SDS-PAGE.
14. Avoid contact with the gel surface. To optimize strip position
on separating gel, use tweezers and steel spatula. Before pour-
ing of agarose, check temperature by the hand. Agarose solu-
tion should not be hot. Use pipette and work quickly because
agarose easily solidifies.
15. Prepare 40 filter cuts, 20 for each side of blotting sandwich.
Soak cuts and membrane in transfer buffer. Soak cuts for 1 h.
Membrane is ready to use after 10 min of soaking in transfer
buffer.
16. Borate buffer is more suitable for transfer in comparison with
commonly used buffer containing methanol. In our experi-
ments, the use of methanol buffer led to significant increase
of voltage, which negatively affected transfer efficiency. Using
of borate buffer did not lead to adverse increase of voltage
above 10 V.
17. Membrane should be perfectly blocked before incubation with
antibodies. Usual recommended time for membrane blocking
is 1 h at room temperature. Longer incubation did not influ-
ence subsequent steps.
18. Increase of room temperature (more than 25 C) can adversely
affect antigen–antibody interaction.
19. If no or just a weak fluorescence is detected, prolong recording
time to 10–15 min. If there is still just a weak signal, block
membrane again in blocking solution and repeat steps 5–9.It
is always better to repeat incubation with antibodies than
repeat whole experiment.
20. The best way how to excise protein spot is to use shortened
pipette tips (usual 10–200 μL tip is shortened ~1 cm from the
narrow end). Excised spots should be bigger than 1 1mm
because smaller pieces could clog pipette tips in subsequent
procedures.
21. The cooling procedure is used for better absorption of trypsin
into gel. It is possible to incubate protein spot for shorter time
(3 h), however complete digestion of protein is not assured.
22. For improved purification of samples before MS analysis “stage
tips” can be used [23].
23. Set possible modifications in a way, where they are expected
(e.g., collagen type I is highly hydroxylated). This will increase
the amount of detected peptides. It is preferable to set variable
394 Matej Kohutiar and Adam Eckhardt
modifications because in the sample there could be modified
and also nonmodified peptides (e.g., due to the worse steric
conditions).
Acknowledgments
This work was supported by grant of GACR No. P303/11/0298
and GACR No. 15-01948S.
References
1. Abello N, Kerstjens HAM, Postma DS et al
(2009) Protein tyrosine nitration: selectivity,
physicochemical and biological consequences,
denitration, and proteomics methods for the
identification of tyrosine-nitrated proteins. J
Proteome Res 8(7):3222–3238. https://doi.
org/10.1021/pr900039c
2. Ischiropoulos J (2003) Biological selectivity
and functional aspects of protein tyrosine nitra-
tion. Biochem Biophys Res Commun
305:776–783. https://doi.org/10.1016/
S0006-291X(03)00814-3
3. Radi R (2013) Protein tyrosine nitration: bio-
chemical mechanisms and structural basis of
functional effects. Acc Chem Res 46
(2):550–559. https://doi.org/10.1021/
ar300234c
4. Pfeiffer S, Mayer B, Hemmens B (1999) Nitric
oxide:chemical puzzles posed by a biological
messenger. Angew Chem Int 38:1714–1731.
https://doi.org/10.1002/(SICI)1521-3773(
19990614)38:12<1714::AID-ANIE1714>3.
0.CO;2-3
5. Szabo
´C, Ischiropoulos H, Radi R (2007) Per-
oxynitrite:biochemistry, pathophysiology and
development of therapeutics. Nat Rev Drug
Discov 6(8):662–680. https://doi.org/10.
1038/nrd2222
6. Weidinger A, Kozlov AV (2015) Biological
activities of reactive oxygen and nitrogen spe-
cies: oxidative stress versus signal transduction.
Biomolecules 5(2):472–484. https://doi.org/
10.3390/biom5020472
7. Brugiere S, Kowalski S, Ferro M et al (2004)
The hydrophobic proteome of mitochondrial
membranes from Arabidopsis cell suspensions.
Phytochemistry 65(12):1693–1707. https://
doi.org/10.1016/j.phytochem.2004.03.028
8. Bartesaghi S, Ferrer-Sueta G, Peluffo G et al
(2007) Protein tyrosine nitration in hydro-
philic and hydrophobic environments. Amino
Acids 32(4):501–515. https://doi.org/10.
1007/s00726-006-0425-8
9. Kohutiar M, Ivica J, Vyta
´s
ˇek R et al (2016)
Comparison of the effects of tert-butyl hydro-
peroxide and peroxynitrite on the oxidative
damage to isolated beef heart mitochondria.
Physiol Res 65(4):617–626
10. Kohutiar M, Eckhardt A, Miks
ˇı
´k A et al (2018)
Proteomic analysis of peroxynitrite-induced
protein nitration in isolated beef heart mito-
chondria. Physiol Res 67:239–250. https://
doi.org/10.33549/physiolres.933608
11. Haas DW, Elliott WB (1963) Oxidative phos-
phorylation and respiratory control in digito-
nin fragments of beef heart mitochondria. J
Biol Chem 238:1132–1136
12. Rabilloud T (2008) Mitochondrial proteomics:
analysis of a whole mitochondrial extract with
two-dimensional electrophoresis. Methods
Mol Biol 432:83–100. https://doi.org/10.
1007/978-1-59745-028-7_6
13. Westermeier R, Naven T (2002) Proteomics in
practice: a laboratory manual of proteome anal-
ysis. Wiley-VCH, Weinheim
14. Shevchenko A, Tomas H, Havlis
ˇJ et al (2006)
In-gel digestion for mass spectrometric charac-
terization of proteins and proteomes. Nat Pro-
toc 1:2856–2860. https://doi.org/10.1038/
nprot.2006.468
15. The UniProt Consortium (2016) UniProt: the
universal protein knowledgebase. Nucleic
Acids Res 45:D158–D169. https://doi.org/
10.1093/nar/gkw1099
16. Fis
ˇa
´rkova
´B, Vyta
´s
ˇek R, Mı
´kova
´D et al (2004)
Hyperoxia attenuated nitrotyrosine concentra-
tion in the lung tissue of rats with experimental
pneumonia. Physiol Res 53(5):487–492
17. Herget J, Wilhelm J, Novotna
´J et al (2000) A
possible role of the oxidant tissue injury in the
development of hypoxic pulmonary hyperten-
sion. Physiol Res 49(5):493–501
18. Wilhelm J, Vyta
´s
ˇek R, Uhlı
´k J et al (2016)
Oxidative stress in the developing rat brain
due to production of reactive oxygen and
Immunoblotting Coupled with Mass Spectrometry 395
nitrogen species. Oxid Med Cell Longev
2016:5057610. https://doi.org/10.1155/
2016/5057610
19. Za
´mec
ˇnı
´k J, Vyta
´s
ˇek R, Vencovsky
´Jetal
(2011) Immunolocalization of protein-bound
3-nitrotyrosine in inflammatory myopathies.
Cesk Patol 47(2):62–65
20. Blum H, Beier H, Gross HJ (1987) Improved
silver staining of plant proteins, RNA and
DNA in polyacrylamide gels. Electrophoresis
8(2):93–99. https://doi.org/10.1002/elps.
1150080203
21. Eckhardt A, Ja
´gr M, Pataridis S et al (2014)
Proteomic analysis of human tooth pulp: pro-
teomics of human tooth. J Endod 40
(12):1961–1966. https://doi.org/10.1016/j.
joen.2014.07.001
22. Mascot search engine: protein identification
software for mass spec data. http://www.
matrixscience.com/. Accessed 16 Aug 2016.
23. Rappsilber J, Mann M, Ishihama Y (2007) Pro-
tocol for micro-purification, enrichment,
pre-fractionation and storage of peptides for
proteomics using StageTips. Nat Protoc 2
(8):1896–1906
396 Matej Kohutiar and Adam Eckhardt
Chapter 29
In Vivo Visualization and Quantification of Mitochondrial
Morphology in C. elegans
R. de Boer, R. L. Smith, W. H. De Vos, E. M. M. Manders,
and H. van der Spek
Abstract
Caenorhabditis elegans is a highly versatile model system, intensively used for functional, genetic, cyto-
metric, and integrative studies. Due to its simplicity and large muscle cell number, C. elegans has frequently
been used to study mitochondrial deficiencies caused by disease or drug toxicity. Here we describe a robust
and efficient method to visualize and quantify mitochondrial morphology in vivo. This method has many
practical and technical advantages above traditional (manual) methods and provides a comprehensive
analysis of mitochondrial morphology.
Key words Mitochondria, Morphology, Caenorhabditis elegans, Confocal microscopy, GFP, Image
analysis, Cytometry
1 Introduction
Mitochondria are highly dynamic tubular organelles that continu-
ously remodel by fusion and fission in a regulated manner. The
balance between these opposing events determines the morphology
of the mitochondrial network, with excessive fission resulting in
mitochondrial fragmentation and increased fusion leading to
extended and highly interconnected mitochondria. Mitochondrial
dynamics play an essential role in cell division, cell survival, cellular
redox status, mtDNA rescue, and mitochondrial protein quality
control [1,2], and specific morphological changes in mitochondrial
structure and organization are considered to reflect cellular
responses to stress and pathological conditions in worms, mice,
and humans [3,4]. For example, changes in mitochondrial struc-
ture and function are known to occur in age-associated disorders
such as Parkinson’s disease, sarcopenia, and in metabolic diseases,
including heart disease and diabetes mellitus [5,6] and also as a
direct result of treatment with therapeutic drugs [7].
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_29,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
397
The nematode Caenorhabditis elegans has proven to be one of
the most versatile model organisms for the elucidation of molecular
pathways implicated in many human diseases and ageing [8–
10]. Wild-type C. elegans has a relatively short life span of
2 weeks, enabling researchers to rapidly assess the effects of differ-
ent mutations or treatments on mitochondrial function or longev-
ity [11,12]. Mitochondrial research in C. elegans has yielded
insights in the genetic regulation of ageing and mitochondrial
function, and it has provided a vast array of mutants to study
these effects [13–15]. C. elegans has many technological advan-
tages. It is small, transparent, and has a fast reproductive rate, fully
described development, anatomy, and a completely sequenced
genome. Moreover, many tools exist to manipulate C. elegans,
including CRISPR/Cas9 tools for endogenous labeling of cellular
proteins with fluorescent markers [16]. Although limited to less
than 1000 cells, C. elegans has highly differentiated tissues such as
skin, nerves, muscles, reproductive organs, and gut tissue, and
mitochondrial function is well-conserved. The preferred tissue to
study mitochondria is muscle as it has well-organized and extensive
mitochondrial networks to provide ATP for their high energy
demands.
Specific cell-permeable dyes, such as MitoTracker, or TMRE
are available for visualizing mitochondria in C. elegans. However,
due to the cuticle, the uptake is slow, and specific staining of
MitoTracker is lost upon if the mitochondrial membrane potential
is disrupted. To avoid this, one can make use of fluorescent proteins
that can be targeted to the mitochondria. Constitutive expression
guarantees mitochondrial staining under all conditions and during
all nematode developmental stages. Additionally, fluorescent pro-
teins are highly stable, remain localized, and allow in vivo imaging.
Using a strain expressing mito::GFP in muscle cells, we established
a protocol for reliable and robust quantification of the mitochon-
drial network. This assay is easily amenable to upscaling and imple-
mentation in high-throughput assays.
2 Materials
Ensure that all solutions are sterile before use. Typically, the nema-
tode growth medium does not contain any antibiotics and is there-
fore prone to contamination. Work in a sterile environment when
handling the worms. Transgenic strains and chemical waste must be
disposed of correctly. Special care is to be taken when handling
sodium azide (NaN
3
) as it is toxic. C. elegans strains can be
obtained from the Caenorhabditis Genetics Centre (https://www.
cbs.umn.edu/cgc/strains).
398 R. de Boer et al.
1. Transgenic C. elegans strain JJ1271 injected with expression
construct mito::GFP expressed from the myo-3 promoter,
which is mitochondrial matrix specific, and transformation
marker rol-6, which distinguishes transgenic animals by a roll-
ing phenotype [17](see Note 1).
2. Temperature-controlled stove (20 C).
3. Nematode growth medium (NGM): NaCl (3 g), Bacto-
peptone (2.5 g), Agar (20 g), dd H
2
O (1 L), 1 M CaCl
2
(1 mL), 1 M MgSO
4
(1 mL), potassium phosphate pH 6.0
(KPO
4
) (25 mL). Autoclave. Add 5 mg/mL cholesterol in 95%
EtOH (1 mL).
4. OP50 Escherichia coli (OD ~ 2.0) in LB: Bacto Trypton (10 g),
Bacto Yeast (5 g), NaCl (5 g), dd H
2
O (1 L). Autoclave.
5. M9: NaCl (5 g), Na
2
HPO
4
(6 g), KH
2
PO
4
(3 g), dd H
2
O
(1 L). Autoclave and add 1 M MgSO
4
(1 mL).
6. Worm picker and platinum wire tip (Genesee Scientific, San
Diego, California, USA).
7. MatTek glass-bottomed petri dishes: P35G-1.0-14C (Corning
35 mm dish with 14 mm glass diameter and coverslip
No. 1 (0.13–0.16 mm thick)) Depth of well ¼0.70–0.75 mm
(see Note 2).
8. 10 mM sodium azide (NaN
3
) in M9. Use ice cold.
9. Microscopy coverslips: Menzel-Gl€
aser, 18 18 mm, No. 1 (see
Note 3).
10. Confocal microscope with camera (see Subheading 3.3). Argon
laser; 488 nm excitation and 525–550 nm emission.
11. GFP reference (green chroma fluorescent slide: 2273-G. Ted
Pella, Inc.).
12. FIJI (https://fiji.sc/), a packaged version of ImageJ freeware
(W.S. Rasband, U.S.A., National Institutes of Health,
Bethesda, Maryland, USA, http://rsb.info.nih.gov/ij).
3 Methods
3.1 Culture
of C. elegans
Worms are cultured on nematode growth medium (NGM) plates
seeded with OP50 Escherichia coli. Culture at 20 C is preferred as
other temperatures induce stress and adaptation, which can alter
mitochondrial morphology. For detailed descriptions of nematode
culture and physiology, see http://www.wormbook.org or www.
wormbase.org/species/c_elegans.
1. Pour NGM plates (~20 mL per 9 cm Ø petri dish) and let
harden at room temperature. Once solid, seed the plates with
500 μL fresh OP50 culture by pipetting and spreading the
Mitochondrial Morphology Quantification 399
bacteria evenly over the plate surface area with a sterile Dri-
galski spatula (see Note 4).
2. Synchronous populations of worms are obtained by alkaline
hypochlorite treatment of gravid adults [18]. Synchronize
~100 gravid adults.
3. Place the eggs overnight in 10 mL M9 in an Erlenmeyer while
gently shaking at 20 C, so as to let all the eggs hatch to L1
larvae (see Note 5).
4. Collect the larvae the following morning in a 15 mL conical
tube and centrifuge at 1.4 gfor 2 min. Aspirate the M9
without disturbing the pellet.
5. Wash the worm pellet in 10 mL M9 to rid the culture of debris,
and centrifuge at 1.4 gfor 2 min before aspirating most of
the M9.
6. Plate the worms sparsely on fresh OP50 seeded plates with a
glass pipette; approximately 200 worms per 9 cm Ø petri dish.
Incubate for 48 h at 20 C to the L4 stage.
7. When having reached the L4 stage, transfer the worms to a
15 mL conical tube by pouring 10 mL M9 onto the plates and
gently swirling to dislodge the worms.
8. Centrifuge the worms at 1.4 gfor 2 min. Aspirate most of the
M9 and then using a glass pipette, transfer the worms to fresh
OP50 seeded plates (in this case containing the compound of
interest). For RNAi seed, the NGM plates with the E. coli
transformant strain carrying the desired gene target construct.
9. Incubate further at 20 C for the desired exposure time (In this
case 24 h) (see Note 6).
3.2 Microscopy Slide
Preparation
1. Keeping the MatTek slides clean, pipette 120 μL of ice-cold
10 mM NaN
3
into the glass-bottomed well.
2. Pick approximately 50 worms and place them inside the NaN
3
droplet.
3. Cover the well gently with a glass coverslip (Fig. 1). Wait for
approximately 10 min for the worms to become paralyzed.
4. Your samples are ready for immediate imaging (see Note 7).
3.3 Imaging
of Mitochondria
It is important to have optimized your confocal microscope settings
beforehand, and once set, to keep them constant at all times.
Settings mentioned here were used for a Nikon A1 confocal micro-
scope and are indicative. Because the sample is suspended in a
water-based solution, the preferred objective is a water immersion
objective. Additionally, a water immersion objective provides supe-
rior focus in depth because it minimalizes aberration under these
circumstances. We use a Plan Apo 60WI objective with a numeri-
cal aperture of 1.27. Keep the acquisition settings such as laser
400 R. de Boer et al.
power (3/5.8 kW/cm
2
), pinhole size (2 A.U./55 μm) and HV
value (85), pixel dwell time (1.9 μs) and pixel size (0.2 μm/pixel)
identical between images.
1. Place the glass-bottom dish on the viewing platform with the
glass-bottom facing the microscope objective.
2. Using widefield microscopy, locate a single worm and align the
focal center just behind the clearly distinguishable area of the
posterior bulb (Fig. 2)(see Note 8).
Fig. 1 Preparation of the MatTek glass-bottomed petri dishes. (1) Pipette 120 μL of the M9 worm suspension
into the glass-bottomed well. (2) Cover the well with a coverslip
Fig. 2 Region of interest (red). Align the focal center just behind the clearly
distinguishable area of the posterior bulb (yellow)
Mitochondrial Morphology Quantification 401
3. Using the scanning function (to avoid bleaching) on the con-
focal microscope, locate the bottom of the well and set this
point as the z-stack’s lower limit. Now locate the topside of the
muscle quadrant using the fast scanning function and set this
point as the z-stack’s upper limit (see Note 9).
4. Set the z-stack slices at 1 μm intervals (see Note 10) and acquire
the images at the optimal acquisition settings.
5. Collect a minimal of 10 z-stacks, i.e., ten worms, per condition,
to ensure statistically relevant sampling.
6. After every imaging session, acquire an image stack from a
reference fluorescence slide, which can be used for correcting
spatial and temporal fluctuations in illumination intensity (see
Subheading 3.4).
3.4 Automated
Image Analysis (See
Note 11)
Images of mitochondria are segmented by means of a dedicated
image-processing pipeline. Before proceeding, inspect images visu-
ally to only retain those of consistent quality. The following image
processing steps are performed before feature extraction:
1. Uninformative slices or slices with reflections are removed from
the image stacks. This can be done automatically using an
image quality criterion that only retains slices with an intensity
covariance (stdev/mean) >1.
2. A flat-field correction is performed to buffer for spatial inten-
sity variations (illumination heterogeneity). To this end, the
image is divided by the corresponding image from the refer-
ence slide.
3. The flat-field corrected stack is then projected according to the
maximum pixel intensity. This allows capturing the majority of
the mitochondria in one image. A potential disadvantage of this
procedure is superposition of mitochondria from different
levels in the worm, although we found this effect to be limited
due to the coarse axial sampling.
4. A duplicate image of the maximum projection image is
pre-processed, by background subtraction (rolling ball
radius ¼15) and local contrast enhancement (block size ¼15,
slope ¼3), so as to buffer intensity variations between the
different objects of interest (mitochondria).
5. Mitochondria are then specifically enhanced by means of a
multi-scale Laplacian operator [19](see Note 12).
6. The enhanced image is binarized according to an autothre-
sholding procedure (Yen or Isodata), yielding a mask that can
be used for analyzing the mitochondria in the original image.
Before doing so, the mask should be filtered to only retain
objects of a predefined size (>7 pixels), this is to avoid noise
or debris from skewing the results.
402 R. de Boer et al.
Using the mask, shape, and intensity metrics, individual mito-
chondria are extracted from the original image as well as the total
number of mitochondria. In addition, general texture metrics can
be calculated from a gray-level co-occurrence matrix analysis. We
specifically calculate the average texture parameters over a horizon-
tal and a vertical GLCM matrix with a 1 pixel offset (a reference
pixel and its immediate neighbor) (see Note 13).
3.5 Mitochondrial
Quantification
After feature extraction, results are summarized per worm or per
condition, by averaging individual mitochondrial metrics. Dedi-
cated statistical analyses can be performed in MATLAB 2017
®
or
in R freeware. In a first approach, individual parameters can be
statistically compared between conditions, by means of pairwise
students t-tests or, in case of non-normal distributions, Wilcoxon
rank sum tests. Subsequently, a more holistic cluster analysis can be
performed, integrating all relevant features. To this end, the data
set is first standardized (values are converted to Z-scores), so as to
avoid differences in magnitude or range from pulling the weight
too much towards one particular variable. Using the standardized
data set, the different conditions (e.g., chemical treatments or
RNAi) are then clustered using Euclidean distance as distance
metric and the average value as linkage value for establishing the
dendrogram. The same is done for the features and the final output
is displayed in a two-dimensional clustergram, color-coded by the
z-value (Fig. 3). This representation allows for quickly resolving
conditions with similar effects on mitochondrial morphology.
Fig. 3 Mitochondrial morphological changes caused by 10 mM NaN
3
. Mitochondrial networks slowly deterio-
rate over time and become more fragmented. a¼10 min, b¼30 min, c¼60 min, d¼120 min, e¼150 min,
f¼180 min, g¼240 min, h¼Typical mitochondrial morphology after vulvar gonad protrusion
Mitochondrial Morphology Quantification 403
4 Notes
1. Other strains such as SJ4103 [20] also service and are easier to
maintain, yet the strain mentioned is preferred due to the fact
that it lacks autofluorescent and birefringent gut granules, and
the mitochondrial network in control animals is neatly
organized and clearly distinguishable, making image analysis
straightforward.
2. Glass-bottomed petri dishes have the advantage over tradi-
tional agar-pad microscopy slides as they do not need to be
made beforehand, are sterile, do not dry out the worms, and
importantly minimalize light refraction and background during
image acquisition.
3. The type is irrelevant for imaging; they are merely needed to
seal off the well.
4. Add the chemical of interest before the medium is poured and
mix well with a magnetic bead stirrer, making sure that the
chemical can withstand temperatures above ~50 C before-
hand. Pour the NGM plates in a sterile environment when
the medium has cooled to ~50 C, or you can comfortably
hold the medium flask. The agarose in the NGM plates solidi-
fies during cooling, which takes approximately 30 min. NGM
plates without chemical additions can be stored at 4 C for
approximately 2 months in a sealed container or petri dish
sleeve. Plates including drugs or other chemicals should be
used as quickly as possible, depending on the compound’s
stability. Seeded plates can be kept for several days in a cool,
dark, and dry environment. Ensure that the OP50 can grow
normally when exposed to your chemical of interest. If this is
not the case, OP50 stocks can be 5concentrated, spread on
the NGM plates, and “inactivated” by UV exposure, before
drying at room temperature and placing the worms on the
plates.
5. Because there is no food present, the larvae halt growth at the
L1 stage. L1 worms can be kept for approximately 48 h in these
conditions before use.
6. During this period, worms become adults and lay eggs. There-
fore, the plates are prone to become overcrowded, rapidly
reducing the E. coli food source. In this case, worms can be
sterilized at the L4 stage by adding 5-fluoro-20-deoxyuridine
(FUdR) to the agar [21], or if FUdR may interfere with the
experiments, meshed using a Sefar Nitex μM filter (Sefar AG
Filtration Solutions, Heiden, Switzerland) every 24 h to rid the
culture of progeny.
404 R. de Boer et al.
7. Immediate analysis is essential as NaN
3
inhibits the respiratory
enzyme cytochrome oxidase and therefore affects mitochon-
dria in the long run. After approximately 1 h of exposure to
10 mM NaN
3
worm, mitochondrial networks become disen-
gaged and they fragment (Fig. 4). This is particularly evident in
muscles adjacent to the vulva, and prolonged NaN
3
exposure
can cause gonad protrusion through the vulva (Fig. 3). NaN
3
is, however, still preferred above other immobilizing agents as
its effects are rapid. Frequently used anesthetizing agents for
in vivo analysis include levamisole (Tetramisole hydrochloride)
or aldicarb, but these compounds show full paralyzing effects
only after hours [22]. Fixating agents, such as formaldehyde,
can also be used [23]. However, the effects of these chemicals
and their relatively long-term incubation time to obtain full
immobilization may affect mitochondrial morphology.
8. This area contains somatic muscle cells nr. 7, 8, and 9 from the
D-lineage, which are present at hatching and have the highest
exposure to the compound through ingestion.
9. C. elegans has distinct muscle segments which are divided into
two dorsal and two ventral quadrants, with two rows of muscle
cells per quadrant (http://www.wormatlas.org/hermaphro
dite/muscleintro/MusIntroframeset.html). Each quadrant in
Fig. 4 Heatmap obtained after analysis of a set of images from C. elegans worms treated with different
chemical compounds or RNAi. The columns represent different features of mitochondrial shape and intensity
as well as image texture, and the rows represent the different treatments. On the right, representative images
are shown for the most dominant phenotypical patterns of mitochondrial networks (normal, fragmented, and
complex)
Mitochondrial Morphology Quantification 405
an adult hermaphrodite is approximately 8 μm thick at the area
just behind the posterior bulb. Typically, ~25 slices at 1 μm
intervals will sufficiently cover the quadrant closest to the
bottom coverslip. Because of the orientation of the muscle
quadrants (see Note 10), it can be quicker to scan for the top
of the worm and set this as the z-stack’s upper limit (~50 slices
at 1 μm intervals). This limit no longer needs to be adjusted for
each worm, thus speeding up image acquisition, although it
will provide you with many useless images.
10. This should provide you with approximately 15–25 images per
z-stack, of which, on average, six images provide sufficient
muscle cross-sectional surface area so they can be used for
image analysis. Because the nematode strain described here
has the roller phenotype, the nematode’s rows of muscle are
likely slightly corkscrewed when paralyzed. Depending on the
position of the nematode in the well, the z-stack segmentation
may not be optimal to obtain large surface area images of the
muscle quadrants. In this case, select a different worm.
11. Traditional mitochondrial morphology analysis methods rely
on blind-scoring by the researcher and, although adequate,
have limited descriptive power due to the amount of variables
that can be taken into account [23]. In addition, these scoring
methods are time-consuming, labor-intensive, and due to their
lack of sensitivity, a considerable amount of worms or muscle
cells needs to be imaged before sufficient statistical power can
be attained.
12. This requires the FeatureJ plugin by Erik Meijering (http://
www.imagescience.org/meijering/software/featurej/), which
is part of the FIJI package but should be downloaded and
installed when using ImageJ.
13. This requires the GLCM_Texture plugin by Julio Cabrera
(http://rsbweb.nih.gov/ij/plugins/texture.html).
References
1. Bess AS, Crocker TL, Ryde IT et al (2012)
Mitochondrial dynamics and autophagy aid in
removal of persistent mitochondrial DNA
damage in Caenorhabditis elegans. Nucleic
Acids Res 40:7916–7931
2. Shutt TE, McBride HM (2013) Staying cool in
difficult times: mitochondrial dynamics, quality
control and the stress response. Biochim Bio-
phys Acta 1833:417–424
3. Yasuda K, Ishii T, Suda H et al (2006)
Age-related changes of mitochondrial structure
and function in Caenorhabditis elegans. Mech
Ageing Dev 127:763–770
4. Jendrach M, Pohl S, Vo
¨th M et al (2005)
Morpho-dynamic changes of mitochondria
during ageing of human endothelial cells.
Mech Ageing Dev 126:813–821
5. Desler C, Hansen TL, Frederiksen JB et al
(2012) Is there a link between mitochondrial
reserve respiratory capacity and aging? J Aging
Res 2012:192503
6. Galloway CA, Yoon Y (2013) Mitochondrial
morphology in metabolic diseases. Antioxid
Redox Signal 19:415–430
7. de Boer R, Smith RL, De Vos WH et al (2015)
Caenorhabditis elegans as a model system for
406 R. de Boer et al.
studying drug induced mitochondrial toxicity.
PLoS One 10:e0126220
8. Markaki M, Tavernarakis N (2010) Modeling
human diseases in Caenorhabditis elegans. Bio-
technol J 5:1261–1276
9. Culetto E, Sattelle DB (2000) A role for Cae-
norhabditis elegans in understanding the func-
tion and interactions of human disease genes.
Hum Mol Genet 9:869–877
10. Lezzerini M, Budovskaya Y (2013) A dual role
of the Wnt signaling pathway during aging in
Caenorhabditis elegans. Aging Cell 13:8
11. Luz AL, Rooney JP, Kubik LL et al (2015)
Mitochondrial morphology and fundamental
parameters of the mitochondrial respiratory
chain are altered in Caenorhabditis elegans
strains deficient in mitochondrial dynamics
and homeostasis processes. PLoS One 10:
e0130940
12. Koopman M, Michels H, Dancy BM et al
(2016) A screening-based platform for the
assessment of cellular respiration in Caenor-
habditis elegans. Nat Protoc 11:1798–1816
13. Tsang WY, Lemire BD (2003) The role of
mitochondria in the life of the nematode, Cae-
norhabditis elegans. Biochim Biophys Acta
1638:91–105
14. Addo MG, Cossard R, Pichard D et al (2010)
Caenorhabditis elegans, a pluricellular model
organism to screen new genes involved in mito-
chondrial genome maintenance. Biochim Bio-
phys Acta 1802:765–773
15. Bratic I, Hench J, Trifunovic A (2010) Caenor-
habditis elegans as a model system for mtDNA
replication defects. Methods 51:437–443
16. Dickinson DJ, Pani AM, Heppert JK et al
(2015) Streamlined genome engineering with
a self-excising drug selection cassette. Genetics
200:1035–1049
17. Labrousse AM, Zappaterra MD, Rube DA et al
(1999) C. elegans dynamin-related protein
DRP-1 controls severing of the mitochondrial
outer membrane. Mol Cell 4:815–826
18. Sulston J, Hodgkin J (1988) Methods. In:
Wood W (ed) The nematode Caenorhabditis
elegans. Cold Spring Harbor Laboratory
Press, New York, NY, pp 287–606
19. De Vos WH, Van Neste L, Dieriks B et al
(2010) High content image cytometry in the
context of subnuclear organization. Cytom
Part A J Int Soc Anal Cytol 77:64–75
20. Benedetti C, Haynes CM, Yang Y et al (2006)
Ubiquitin-like protein 5 positively regulates
chaperone gene expression in the mitochon-
drial unfolded protein response. Genetics
174:229–239
21. Mitchell DH, Stiles JW, Santelli J et al (1979)
Synchronous growth and aging of Caenorhab-
ditis elegans in the presence of fluorodeoxyur-
idine. J Gerontol 34:28–36
22. Ramot D, Johnson BE, Berry TL et al (2008)
The Parallel Worm Tracker: a platform for mea-
suring average speed and drug-induced paraly-
sis in nematodes. PLoS One 3:e2208
23. Giacomotto J, Brouilly N, Walter L et al (2013)
Chemical genetics unveils a key role of mito-
chondrial dynamics, cytochrome c release and
IP3R activity in muscular dystrophy. Hum Mol
Genet 22:4562–4578
Mitochondrial Morphology Quantification 407
Chapter 30
Assessing Impact of Platinum Complexes on Mitochondrial
Functions
Suxing Jin and Xiaoyong Wang
Abstract
Platinum-based antitumor drugs play important roles in the clinical treatment of various tumors. Never-
theless, some deficiencies such as poor targeting ability, low bioavailability, in vivo deactivation, drug
resistance, and side effects undermine the efficacy of these drugs. Mitochondria are important organelles
which regulate the energy metabolism, physiological function, life span, and survival of the cells. Regulating
or interfering with mitochondrial metabolism is of great significance in the prevention or treatment of
cancers. Thus, a series of mitochondrion-targeted platinum complexes were prepared by modifying triphe-
nylphosphine (TPP
+
) through chemical modifications, which endow traditional platinum drugs with new
properties and mechanisms through interfering with mitochondrial DNA (mtDNA), mitochondrial mem-
brane potential (MMP), mitochondrial morphology, mitochondrial bioenergetics, or production of reactive
oxygen species (ROS), thereby opening a new path for the clinical application of platinum drugs. Here we
introduce the synthesis of some TPP
+
-modified platinum (II, IV) complexes in details and the detection
method of the activity parameters related to the mitochondrial functions.
Key words Platinum complex, Mitochondrion, Triphenylphosphine, Synthesis, Detection
1 Introduction
Platinum-based metallodrugs are the most widely used antitumor
drugs in the clinic. Cisplatin, for example, was approved by the
Food and Drug Administration (FDA) of the United States as an
antitumor drug for clinical use in the treatment of cancers in 1979
and is widely used in the treatment of cervical cancer, bladder
cancer, head and neck cancer, non-small cell lung cancer, and so
on [1,2]. Although platinum drugs have achieved huge success in
cancer treatment, they also encountered problems such as systemic
toxicity and drug resistance [3,4]. To solve these problems and
provide better chemotherapy to cancers, we need to explore and
develop novel metallodrugs and to study new mechanisms of plati-
num drugs.
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_30,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
409
Mitochondria are tiny organelles in cells, which provide almost
all the energy needed for cell operation by producing adenosine
triphosphate (ATP) [5]. Mitochondria are widely involved in signal
transduction, energy metabolism, autophagy, apoptosis, and other
cell processes, which is essential to maintain the normal physiologi-
cal functions of the cells; furthermore, there are direct or indirect
connections between the functions [6]. Calcium balance, for exam-
ple, is related to the production of ATP, which is in turn related to
the activation of calcium-sensitive citrate cycle enzyme, the produc-
tion of ROS, the opening of mitochondrial transition pore, and the
reduction of membrane potential that causes apoptosis [7]. Thus, it
is important to regulate or intervene mitochondrial metabolic pro-
cess in the prevention or treatment of cancers. Moreover, the
excessive production of reactive oxygen species (ROS) in mito-
chondria could cause damage to mitochondrial DNA (mtDNA)
and mitochondrial dysfunction, which would lead to cell apoptosis.
The status of mitochondria (number, size, etc.) will change accord-
ingly when cells are damaged. It was observed that the morphology
of mitochondria in tumor cells is different from that in normal cells,
and the function of mitochondria can be regulated by affecting the
morphology or membrane integrity of mitochondria. The imbal-
ance of fusion and division could also lead to mitochondrial dys-
function, fragmentation, increase of ROS and ATP production,
followed by decreased membrane potential (MMP), which was
regarded as a starting point for inducing apoptosis [8]. In addition,
the membrane permeability may increase due to the high expres-
sion of mitochondrial membrane proteins Bak and Bax, leading to
the release of proapoptotic proteins such as cytochrome c
[9,10]. Considering the multiple functions of metal complexes,
intervention of mitochondrial morphology, MMP, ROS, and pro-
teins may be accomplished through tactful design of platinum
complexes, which have showed great potential in antitumor
applications.
Mitochondria have double-layer membranes, and the mem-
brane potential is negative inside and positive outside, which
could be exploited for targeting mitochondria. Selective accumula-
tion of platinum drugs in mitochondria can intervene the status or
function of mitochondria [11]. The MMP of cancer cells is higher
than that of normal ones on the account of metabolic changes,
which gives rise to the accumulation of positively charged com-
pounds in mitochondria of cancer cells up to 10 times higher than
that in normal cells [12]. Therefore, cationic molecules are prefer-
entially taken up by tumor cells, leaving normal cells free of dam-
age. Consequently, mitochondrion-targeting groups have two
major characteristics: (1) delocalized positive charge(s) and
(2) high lipophilicity. Triphenylphosphine cation (TPP
+
) is the
most widely used such group for mitochondrial targeting
410 Suxing Jin and Xiaoyong Wang
[13]. Advantages of TPP
+
have been well recognized, which make
it easy to be introduced into platinum complexes by chemical
synthesis [14].
Herein we describe the synthesis of different types of
TPP
+
-modified TPP
+
-Pt
II
and TPP
+
-Pt
IV
complexes, and the
experimental methods for testing their effects on mtDNA, MMP,
bioenergetics, mitochondrial morphology, ROS, and apoptosis-
related proteins, etc. In-depth investigations on the functions of
mitochondria interfered by these platinum complexes were carried
out according to these methods.
2 Materials
Prepare all solutions using ultrapure water and analytical grade
reagents. Prepare and store all reagents at room temperature unless
indicated otherwise.
2.1 Equipment 1. Three-necked round bottom flasks (150 and 250 mL).
2. Microtubes (1.5 mL, nonpyrogenic and Rnase-/Dnase-free).
3. Centrifuge tubes (10 mL).
4. Glass homogenizer.
2.2 Primer The primer sequences of the selected genes in RT-qPCR [15]:
1. AS1-F: 50-CCCTAACACCAGCCTAACCA-30
2. AS1-R: 50-AAAGTGCATACCGCCAAAAG-30
3. BS1-F: 50-CATGCCCATCGTCCTAGAAT-30
4. BS1-R: 50-ACGGGCCCTATTTCAAAGAT-30
5. CS1-F: 50-TCCAACTCATGAGACCCACA-30
6. CS1-R: 50-TGAGGCTTGGATTAGCGTTT-30
7. DS1-F: 50-ACTACAACCCTTCGCTGACG-30
8. DS1-R: 50-GCGGTGATGTAGAGGGTGAT-30
9. AL1-F: 50-CTGTTCTTTCATGGGGAAGC-30
10. AL1-R: 50-AAAGTGCATACCGCCAAAAG-30
11. BL1-F: 50-CATGCCCATCGTCCTAGAAT-30
12. BL1-R: 50-TGTTGTCGTGCAGGTAGAGG-30
13. CL1-F: 50-CACACGAGAAAACACCCTCA-30
14. CL1-R: 50-CTATGGCTGAGGGGAGTCAG-30
15. DL1-F: 50-CCCTTCGCCCTATTCTTCAT-30
16. DL1-R: 50-GCGTAGCTGGGTTTGGTTTA-30
17. mt-ND1-F: 50-ATATGAAGTCACCCTAGCCAT-30
Impact of Platinium Complexes on Mitochondria 411
18. mt-ND1-R: 50-CTGAGACTAGTTCGGACTCCC-30
19. mt-ND2-F: 50-CGGACAATGAACCATAACCAA-30
20. mt-ND2-R: 50-GTTTAATCCACCTCAACTGCC-30
21. mt-ND3-F: 50-GCCCTACAAACAACTAACCTG-30
22. mt-ND3-R: 50-ATTCGGTTCAGTCTAATCCTT-30
23. mt-ND4-F: 50-TCTGTGCTAGTAACCACGTTC-30
24. mt-ND4-R: 50-AAAACCCGGTAATGATGTCG-30
25. mt-ND4L-F: 50-ACTAGTATATCGCTCACACC-30
26. mt-ND4L-R: 50-CTAGTATGGCAATAGGCACA-30
27. mt-ND5-F: 50-CTTACCACCCTCGTTAACCC-30
28. mt-ND5-R: 50-ATAACTTCTTGGTCTAGGCACA-30
29. mt-ND6-F: 50-ATATACTACAGCGATGGCTA-30
30. mt-ND6-R: 50-AATCCTACCTCCATCGCTA-30
31. mt-CYB-F: 50-TTATTGACTCCTAGCCGCAGA-30
32. mt-CYB-R: 50-TAGTACGGATGCTACTTGTCCA-30
33. mt-CO1-F: 50-AATAGGAGCTGTATTTGCCAT-30
34. mt-CO1-R: 50-AGAAAGTTAGATTTACGCCGAT-30
35. mt-CO2-F: 50-CTTTACATAACAGACGAGGTCA-30
36. mt-CO2-R: 50-TTGAAGATTAGTCCGCCGTA-30
37. mt-CO3-F: 50-CCACTCCTAAACACATCCGTA-30
38. mt-CO3-R: 50-GCCAATAATGACGTGAAGTCC-30
39. mt-ATP6-F: 50-CAACACTAAAGGACGAACCTG-30
40. mt-ATP6-R: 50-TTAATCTTAGAGCGAAAGCCTA-30
41. mt-ATP8-F: 50-TGCCCCAACTAAATACTACCG-30
42. mt-ATP8-R: 50-ATGAATGAAGCGAACAGAT-30
43. REF-F: 50-GGAGCGAGATCCCTCCAAAAT-30
44. REF-R: 50-GGCTGTTGTCATACTTCTCATGG-30
2.3 Solutions 1. XF assay medium (Seahorse Bioscience): The constituents are
based on Dulbecco’s Modified Eagle’s Medium (DMEM). No
sodium bicarbonate (tissue culture buffering agent), glucose,
glutamine/GlutaMAX, or sodium pyruvate is present.
2. Test solution for oxygen consumption rate (OCR) assay: Add
glucose (25 mM) and pyruvate (1 mM, Sigma-Aldrich) into
the XF assay medium and adjust the pH to 7.4.
3. Test solution for extracellular acidification rate (ECAR) assay:
Add glutamine (2 mM, Sigma-Aldrich) into the XF assay
medium and adjust the pH to 7.4.
412 Suxing Jin and Xiaoyong Wang
4. XF glycolysis stress test kit:
(a) Glucose solution: Add XF assay medium (1 mL, see
item 1) to glucose (0.45 g) to reach a final concentration
of 2.5 mM.
(b) Oligomycin solution: Add dimethyl sulfoxide (DMSO,
180 μL) to oligomycin to reach a final concentration of
5 mM.
(c) 2-Deoxy-D-glucose (2-DG) solution: Add XF assay
medium (16 mL, see item 1) to 2-DG, mix well until
clear, and keep at 37 C until the solution becomes yellow.
Add the medium to make the constant volume be 18 mL.
Adjust the pH to 7.4 with 1 M NaOH.
5. XF cell mito stress test kit:
(a) Oligomycin solution: see item 4b.
(b) Carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone
(FCCP) solution: Add DMSO (180 μL) to FCCP to
reach a final concentration of 2.5 mM.
(c) Rotenone solution: Add DMSO (180 μL) to rotenone to
reach a final concentration of 2.5 mM.
(d) Antimycin solution: Add DMSO (180 μL) to antimycin to
reach a final concentration of 2.5 mM.
6. Radio immunoprecipitation assay (RIPA) lysis buffer: The
main components are Tris–HCl (50 mM, pH 7.4), NaCl
(150 mM), 1% Triton X-100, 1% sodium deoxycholate, 0.1%
sodium dodecyl sulfate (SDS), inhibitors such as sodium
orthovanadate, sodium fluoride, leupeptin, ethylenediaminete-
traacetic acid (EDTA), and so on.
7. Cell lysis buffer: Add DTT (1 mM, 1 μL), PMSF (1 mM,
10 μL), NaF (0.1 mM, 10 μL) and 100protease inhibitor
cocktail (10 μL) to RIPA lysis buffer.
8. Polymerize acrylamide containing 0.1% SDS (Acryl/Bis solu-
tion 29:1, 40%) solution to form SDS-PAGE gel under the
induction of ammonium persulfate and four methyl ethylene-
diamine. Perform protein electrophoresis using a noncontinu-
ous system with the stacking gel of 5% (80 v) and the separation
gel of 5% (60–212 kDa), 7.5% (30–120 kDa), 10%
(18–75 kDa), 12.5% (15–60 kDa), 15% (15–43.5 kDa)
(120 v), respectively.
9. Transfer buffer: Dissolve Tris–HCl (30.29 g) and glycine
(144 g) in ultrapure water (1 L) to prepare 10transfer buffer.
Mix 10transfer buffer (100 mL), methanol (200 mL), and
ultrapure water (700 mL) to obtain 1transfer buffer.
10. Phosphate-buffered saline (PBS): Dissolve NaCl (80.0 g), KCl
(2.0 g), Na
2
HPO
4
· 12H
2
O (36.3 g), and KH
2
PO
4
(2.7 g) in
Impact of Platinium Complexes on Mitochondria 413
ultrapure water (1 L) to prepare 10PBS buffer. Mix 10PBS
buffer (100 mL) and ultrapure water (900 mL) to obtain 1
PBS buffer.
11. PBST buffer: Add Tween-20 (1 mL) in 1PBS buffer (1 L).
12. Blocking buffer: Add skim milk (2 g) in PBST buffer (40 mL).
2.4 Fluorescent Dyes 1. MitoSOX™working solution: Dissolve a vial of MitoSOX™
mitochondrial superoxide indicator in DMSO (13 μL) to make
a MitoSOX™reagent stock solution (5 mM). Dilute the Mito-
SOX™reagent stock solution in PBS or incomplete cell culture
to make a MitoSOX™reagent working solution (5 μM).
2. JC-1 working solution: Dilute an appropriate amount of JC-1
(200) at a ratio of 50 μL JC-1 (200) to 8 mL ultrapure
water and dissolve it by a vigorous vortex. Add JC-1 buffer
solution (5, 2 mL) to get the JC-1 working solution.
3. JC-1 staining buffer: Dilute an appropriate amount of JC-1
(5) at a ratio of 1 mL JC-1 (5) to 4 mL ultrapure water to
get the JC-1 working solution and place it on ice.
2.5 Cell Culture
Components
1. HeLa and A 549 cells (ATCC, USA).
2. Cell culture medium: DMEM culture medium with 10% fetal
bovine serum; store it at 4 C.
3. Trypsin: 0.25% trypsin in PBS buffer; store it at 4 C.
4. Washing buffer: PBS of pH 7.2 without calcium and
magnesium.
5. Cell culture dishes (60 and 100 mm) and 6-well plates
(Corning).
6. Standard 35 mm cell culture dish with glass bottom.
3 Methods
3.1 Synthesis
of TPP
+
-Pt
II
(a)
1. Stir p-methylbenzaldehyde (9.00 mL) and 2-acetylpyridine
(9.00 mL) in NaOH (2%) aqueous solution (150 mL) at
room temperature for 8 h to obtain a pale yellow solid.
2. Add 2-acetylpyridine (9.00 mL) to the above solid-liquid sys-
tem, then adjust the concentration of NaOH aqueous solution
to 20%. Raise the temperature to 50 C and stir the solution
continuously for 8 h to obtain a brown-red oil.
3. Solidify the oil and remove the aqueous phase. Add ethanol
(200 mL) to dissolve the oil, obtaining a dark red clear liquid.
Add ammonium acetate (50.00 g) in batches under reflux at
60 C for 5 h. Evaporate part of the ethanol as the raw materials
react almost completely and crystallize to obtain needle-like
crystals. Recrystallize in ethanol to obtain the modified TPP
ligand (TPP-TPy).
414 Suxing Jin and Xiaoyong Wang
4. Add aqueous solution of K
2
PtCl
4
(5 mL) dropwise to the
DMSO solution of TPP-TPy (5 mL) and stir for 12 h at
110 C. Add the mixture to concentrated HCl to precipitate
the product. Redissolve the product in DMSO and add it
dropwise to acetone (50 mL). Wash the product with diethyl
ether and acetone twice to obtain the TPP
+
-modified platinum
complex TPP
+
-Pt
II
(a) [16].
3.2 Synthesis
of TPP
+
-Pt
II
(b)
1. Add K
2
CO
3
(3.37 g) into 2-(chloromethyl)pyridine hydro-
chloride (8.00 g) solution to adjust the pH to 7.0. Extract
the reaction mixture with diethyl ether, and dry with anhydrous
Na
2
SO
4
and concentrate. Dissolve the residue (5.60 g) in
1,4-dioxane (25 mL) and add triphenylphosphine (11.54 g).
Heat the mixture at 110 C, reflux for 12 h, and filter to obtain
the solid product. Wash the solid with diethyl ether and dry
under vacuum to get the modified ligand o-PPh
3
CH
2
PyCl.
2. Dissolve o-PPh
3
CH
2
PyCl (150 mg) in anhydrous N,N-
dimethylformamide (DMF, 5 mL) and allow to react with
AgNO
3
(66 mg) under stirring for 5 h at 25 C. Centrifuge
the reaction solution to obtain the yellow supernatant contain-
ing o-PPh
3
CH
2
PyNO
3
.
3. Stir cisplatin (150 mg) and AgNO
3
(80 mg) in anhydrous
DMF (3 mL) overnight in the dark at 45 C and obtain a pale
yellow solution of cis-[Pt(NH
3
)
2
Cl(DMF)](NO
3
) after
centrifugation.
4. Drop the above o-PPh
3
CH
2
PyNO
3
solution into [cis-Pt
(NH
3
)
2
Cl(DMF)](NO
3
) solution and stir in the dark at 55 C
for 48 h. Filter the resulting golden solution and evaporate.
Rinse the oily substance by dichloromethane (DCM), extract
with hot methanol (100 mL), and concentrate the extract to
5 mL. Add extra diethyl ether to get a light yellow precipitate.
Wash the precipitate with DCM and diethyl ether, and dry it
in vacuum to obtain the final product TPP
+
-Pt
II
(b) [17].
3.3 Synthesis
of TPP
+
-Pt
IV
(a)
1. Add hydrogen peroxide (30 wt%, 20 mL) to the suspension of
cisplatin (400 mg) in H
2
O (12 mL) at 60 C. After 4 h, cool
the bright yellow solution at room temperature overnight to
afford yellow crystals. Filter the crystals and wash them with
cold water to obtain oxoplatin.
2. Add EDC · HCl (211 mg) and NHS (127 mg) to the solution
of (4-carboxybutyl)triphenylphosphonium bromide (433 mg)
in acetonitrile (15 mL) and stir at room temperature for 12 h to
get a colorless solution. Remove acetonitrile by rotary evapora-
tion to yield a colorless raw product. Dissolve the product in
DCM and wash it with water three times. Collect the organic
layer and add anhydrous Na
2
SO
4
to remove water. Finally,
remove the solvent to gain the modified ligand TPP-NHS.
Impact of Platinium Complexes on Mitochondria 415
3. Add a solution of TPP-NHS ester (194 mg) in anhydrous
DMSO (5 mL) to the suspension of oxoplatin (100 mg) in
anhydrous DMSO (10 mL) with vigorous stirring. Stir the
mixture at 30 C for 72 h. Remove DMSO by addition of
excessive diethyl ether. Extract the product with methanol
and wash twice with methanol and ether. Dry the light yellow
solid TPP
+
-Pt
IV
(a) in vacuum [18].
3.4 Synthesis
of TPP
+
-Pt
IV
(b)
1. Stir oxoplatin (50 mg) in DMF with (4-carboxybutyl)triphe-
nylphosphonium bromide (200 mg), triethylamine (46 mg),
and O-(benzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
tetrafluoroborate (TBTU, 144 mg) for 48 h. After filtration,
collect the filtrate and remove DMF under high vacuum.
2. Add ethanol and water to the residue to gain the precipitation
of desired product. Dissolve the crude product in methanol to
obtain TPP
+
-Pt
IV
(b) by multiple precipitation in diethyl
ether [18].
3.5 Mitochondrial
Uptake
1. Seed the cancer cells into a 10 cm petri dish and incubate
overnight to 60–70% confluence.
2. Replace the cell culture medium with fresh growth medium
and treat the cells with above mitochondrial-targeted platinum
complex (see Subheadings 3.1–3.4,the same below) at the
desired concentrations and 37 C for 24 h in an atmosphere
of 5% CO
2
and 95% air.
3. Collect the cells, suspend the cellular precipitate in mitochon-
drial isolation solution (1 mL), and cool on ice for 10 min.
4. Homogenize the cell suspension for 30 strokes using a tight
pestle on ice and then centrifuge at 600 gand 4 C for
10 min (see Note 1).
5. Place the supernatant in a fresh tube and centrifuge again at
12,000 rpm (15,300 g) and 4 C for 15 min to obtain the
mitochondrial deposition.
6. Digest the cell lysis solution by concentrated nitric acid
(100 μL) at 95 C for 2 h, hydrogen peroxide (30%, 50 μL)
at 95 C for 1.5 h, and concentrated hydrochloric acid (50 μL)
at 37 C until the total volume is less than 50 μL. Add ultrapure
water to each sample to reach 1 mL.
7. Test the Pt content in each sample by ICP-MS.
3.6 Determination
of Mitochondrial
Superoxide (mtSOX)
1. Seed cancer cells in a 6-well plate (Corning) at a density of
10
5
cells/mL and incubate overnight to 60–70% confluence.
2. Replace the cell culture medium with fresh growth medium,
and incubate the cells with the complex at the desired concen-
trations at 37 C for 24 h in an atmosphere of 5% CO
2
and
95% air.
416 Suxing Jin and Xiaoyong Wang
3. Wash the cells three times with PBS buffer and incubate with
incomplete growth medium under 5 μM of mtSOX probe
MitoSOX™at 37 C for 10 min in situ (see Notes 2 and 3).
4. Trypsinize the cells, wash with PBS three times and resuspend
in PBS (500 μL).
5. Analyze the cells on the BD FACSCalibur flow cytometer
within 1 h.
3.7 Mitochondrial
Membrane Potential
(JC-1 Assay)
1. Seed cancer cells in a glass bottom cell culture dish (φ2 mm,
NEST) containing 1 mL of growth medium at 40% confluence.
2. Replace the cell culture medium with fresh growth medium
and incubate the cells with the complex with desired concen-
trations at 37 C for 24 h.
3. Add the JC-1 working solution and incubate at 37 C for
20 min. Wash the cells thrice with JC-1 staining buffer before
imaging (see Note 4).
4. Carry out the imaging by a confocal laser scanning fluorescence
microscopy (Zeiss LSM710) with a 63objective lens.
Record the fluorescence of green channel (λ
ex
¼488 nm,
λ
em
¼510–545 nm) and red channel (λ
ex
¼543 nm,
λ
em
¼575–630 nm), respectively (see Fig. 1). Quantify the
fluorescence intensities and correct total cell fluorescence
(CTCF) values through ImageJ.
3.8 Mitochondrial
Morphology
1. Plate cancer cells in 10 cm dishes and cultured at 37 C for
18 h.
2. Incubate the cells with indicated concentrations of complexes
for 48 h, trypsinize, and collect the cells.
3. Fix the obtained pellets with 2.5% glutaraldehyde at 4 C over-
night, wash the cells several times with PBS and further fix with
1% OsO
4
.
4. Dehydrate the samples using solutions of acetone (50%, 75%,
90%, and 100%) prior to impregnation in increasing concen-
trations (25%, 50%, 75%, and 100%) of resin in acetone over a
period of 24 h.
5. Cut the cells into small segments with ultramicrotome, stain
with 2% aqueous uranyl acetate and lead citrate.
6. Wash the sections with ultrapure water.
7. Observe the samples under transmission electron microscopy
(JEOL JEM-1011) after drying (see Fig. 2).
3.9 mtDNA Damage 1. Seed cancer cells into 6-well plates, culture in cell culture
medium with 10% (v/v) FBS at 37 C for 24 h.
Impact of Platinium Complexes on Mitochondria 417
2. Replace the cell culture medium with fresh growth medium
and incubate the cells with complexes on the desired concen-
trations at 37 C for 24 h.
3. Wash the attached cells twice with PBS (4 C), harvest by
trypsinization (0.5 mL), and wash with PBS (1 mL).
4. Lyse the cell pallets in DNAzol reagent (1 mL), and extract the
genomic DNA from the lysate with pure ethanol (0.5 mL) by
incubating the sample at room temperature for 1–3 min.
5. Determine the amount of DNA with Nanodrop 1000 at
260 nm. Quantify the level of mtDNA damage in each tested
Fig. 2 TEM images of mitochondria in A549 cells after treatment with TPP
+
-Pt
IV
(b) for 24 h. The mitochondria
in control group are identified with typical features, including the well-defined integral double membranes and
regular cristae. The morphology of mitochondria in TPP
+
-Pt
IV
(b)-group changed significantly, including the
damage of mitochondria with distortion of cristae and partial or total cristolysis, even vacuoles
Fig. 1 Fluorescence changes of HeLa cells after treatment with the complex for 24 h and staining by JC-1 dye.
JC-1 remains as monomers that emit green fluorescence when MMP is dissipated in apoptotic or abnormal
cells, while assembles to form aggregates with red fluorescence when MMP is high in normal cells
418 Suxing Jin and Xiaoyong Wang
region of the mitochondrial genome by comparing the relative
amplification of two mtDNA fragments (long fragments and
small ones) locating in the same mitochondrial genomic
region, with the shorter fragments as internal normalization
controls.
6. Perform the PCR using total DNA (5 ng, 1 μL), forward
primer (1 μL), reverse primer (1 μL), SsoFast EvaGreen Super-
mix (5 μL), and nuclease-free water (2 μL, see Note 5). Con-
ditions for the long fragments are 95 C for 10 min, followed
by 39 cycles at 95 C for 10 s, 60 C for 10 s, and 72 C for
50 s. Perform the cycling for the short reaction at 95 C for
10 min, followed by 39 cycles at 95 C for 10 s, 62 C for 10 s,
and 72 C for 10 s.
7. Calculate the mtDNA lesion using the reported methods:
Lesion rate [lesion per 10 kb DNA] ¼(1 2
(ΔCq,longΔCq,short)
)10,000 [bp]/size of
long fragment [bp].
ΔC
q,long
¼C
q,sample
C
q,reference
;ΔC
q,short
¼C
q,control
C
q,reference
.
3.10 Transcription
of Mitochondrial Genes
1. Seed cancer cells into 6-well plates, culture in cell culture
medium with 10% (v/v) FBS at 37 C for 24 h.
2. Replace the cell culture medium with fresh growth medium
and incubate the cells with the complex on the desired con-
centrations at 37 C for 24 h.
3. Wash the attached cells twice with PBS (4 C), add trizol
reagent (1 mL), and keep in dark for 10 min.
4. Pipette the solution repeatedly and collect the suspension in
1.5 mL centrifuge tube. Add chloroform (200 μL) and shake
for 30 s, and then leave the tube at room temperature for
5 min.
5. Centrifuge the solution at 12,000 rpm (15,300 g) for 15 min
at 4 C, collect the upper water phase into a clean centrifuge
tube, add isopropanol (500 μL) again and mix with the solu-
tion. Allow the solution to stand for 10 min, and then centri-
fugate at 12,000 rpm (15,300 g) and 4 C for 15 min to
remove the supernatant.
6. Add ethanol (75%, 1 mL) and mix. Centrifuge the solution at
7500 rpm (5970 g) and 4 C for 5 min and remove the
supernatant. Repeat the operation twice and dry the
precipitation.
7. Add ultrapure water (50 μL) to dissolve the RNA precipitation.
Determine the amount of RNA on Nanodrop 1000.
Impact of Platinium Complexes on Mitochondria 419
8. Add 5iScript reaction mix (4 μL), iScript reverse transcriptase
(1 μL), RNA (1 μg), and a certain amount of ultrapure water to
a centrifuge tube. Reverse-transcribed approximately 1 μgof
total RNA to cDNA.
9. Analyze the expression of 13 mitochondrial code genome pro-
teins by RT-qPCR. Perform the PCR by using cDNA (1 μL),
forward primer (1 μL), reverse primer (1 μL), SsoFast Eva-
Green Supermix (5 μL), and nuclease-free water (2 μL, see
Note 5). The amplification program is 30 s at 95 C, followed
by 49 cycles of 5 s at 95 C, and 5 s at 60 C. At the end of the
amplification, assess the specificity of the gene by a melting
curve between 65 and 95 C. Calculate the results according
to the 2
ΔΔCt
method [see Subheading 3.9,step 7].
3.11 Mitochondrial
Bioenergetics
3.11.1 Oxygen
Consumption Rate (OCR)
1. Seed cancer cells in XFe24-well cell culture plates at a density of
10
4
cells per well and then incubate for 24 h.
2. Replace the cell culture medium with fresh growth medium
and treat the cells with the complex for 18 h.
3. Remove the cell culture medium and wash the cells with XF
assay medium (see Subheading 2.3,item 2,same as below)
thrice. Add XF assay medium (525 μL) to each well again (see
Note 6).
4. Incubate the plate at 37 CinaCO
2
-free incubator for 1 h
before the measurement.
5. Add oligomycin (1.0 μM, 75 μL), an inhibitor of ATP synthase
complex, trifluorocarbonylcyanide phenylhydrazone (FCCP,
1.0 μM, 75 μL), an uncoupler of ATP synthesis, and a mixture
(75 μL) of antimycin-A (1.0 μM), an inhibitor of complex III,
and rotenone (1.0 μM), an inhibitor that prevents the transfer
of electrons from the Fe–S center in complex I to ubiquinone,
to A, B, and C holes, sequentially.
6. Determine OCR on a Seahorse XFe24 Cell Bioanalyzer (Sea-
horse Biosciences). Record the data during the measurement
and analyze them using the average of four baseline rates and
up to five test rates [19](see Fig. 3).
3.11.2 Extracellular
Acidification Rate (ECAR)
1. Seed cancer cells in XFe24-well cell culture plates at a density of
10
4
cells per well and incubate for 24 h.
2. Replace the cell culture medium with fresh growth medium
and treat the cells with the complex for 18 h.
3. Remove the cell culture medium and wash the cells with XF
assay medium (see Subheading 2.3,item 3,same as below)
thrice. Finally, add XF assay medium (525 μL) in each well
again (see Note 6).
420 Suxing Jin and Xiaoyong Wang
4. Incubate the plate at 37 CinaCO
2
-free incubator for 1 h
before the measurement.
5. Add glucose (10 mM, 75 μL), oligomycin (1 μM, 75 μL), and
2-DG (130 mM, 75 μL) to A, B, and C holes, sequentially.
6. Determine ECAR on a Seahorse XFe24 Cell Bioanalyzer (Sea-
horse Biosciences) and record the data during the measure-
ment. Analyze the data using the average of four baseline rates
and up to five test rates [19](see Fig. 3).
3.12 Proteins
Relevant
to Mitochondrion-
Mediated Apoptosis
1. Seed the cancer cells into 10 cm petri dish and incubate over-
night to 60–70% confluence.
2. Treat the cells with the desired concentration of drugs, collect
and wash twice with ice-cold PBS.
3. Lyse the cell pellets in cell lysis buffer (100 μL) on ice for
30 min. Centrifuge the lysate at 12,000 rpm (15,300 g)
and 4 C for 15 min to remove the cell debris (see Note 7).
4. Electrophorese the lysate (40–60 μg) on a 10–12% SDS-PAGE
gel by Bio-Rad Mini-PROTEAN Tetra System (80 V, 20 min;
100 V, 1 h).
5. Transfer proteins to PVDF (Millipore, 0.22 μm) membranes
in transfer buffer containing 0.033% SDS (90 V, 1.5 h) (see
Notes 8 and 9).
6. Block the PVDF membranes at ambient temperature in block-
ing buffer for 1 h (see Note 10).
7. Incubate the membrane with monoclonal antibody (anti-
IMMT, Cyto c, PDK2, Bad, Bax, Bcl-2, caspase-3, caspase-9,
p53, α-tubulin, etc.) at 4 C overnight (see Note 11).
Fig. 3 (a) Overall mitochondrial OCR profiles of HeLa cells in response to the complex at 18 h. OCR for basal
respiration ¼OCR
initial
OCR
antimycin A/rotenone
, OCR for ATP production ¼OCR
basal
OCR
oligomycin
, OCR for
proton leak ¼OCR
oligomycin
OCR
antimycin A/rotenone
.(b) ECAR of HeLa cells in response to the complex at 18 h.
ECAR for glycolysis ¼ECAR
oligomycin
ECAR
glycose
, ECAR for glycolytic capacity ¼ECAR
2-DG
ECAR
glycose
,
ECAR for glycolytic reserve ¼ECAR
2-DG
ECAR
oligomycin
Impact of Platinium Complexes on Mitochondria 421
8. Wash the membranes with PBST for three times and incubate
with the peroxidase-conjugated anti-rabbit IgG or anti-mouse
IgG as secondary antibody in washing buffer at room tempera-
ture for 1 h (see Note 11).
9. Visualize the position of proteins by chemiluminescence HRP
substrate (Millipore).
4 Notes
1. All the steps relating to mitochondrial separation should be
carried out on ice or at 4 C, and the solutions should be
precooled at 4 C.
2. Probe vials should be warmed to room temperature before
opening.
3. The concentration of MitoSOX™reagent working solution
should not exceed 5 μM.
4. JC-1 (200) must be thoroughly dissolved and mixed with
ultrapure water before adding JC-1 staining buffer (5). Pre-
paring JC-1 staining buffer (1) and then adding JC-1 (200)
will make JC-1 difficult to be fully dissolved and thus affect the
subsequent detection.
5. The preparation of solutions in PCR experiments should be
performed on ice.
6. In the Seahorse experiment, XF assay medium should be
heated to 37 C in advance, and then corresponding drugs be
added and pH be adjusted to 7.4.
7. All the lysed cell samples should be stored at 20 C.
8. The stacking position of sponge, filter paper (special filter paper
for transfer film), transfer film, and gel should be checked
carefully to ensure the correct transfer direction.
9. The relative sizes of sponge, filter paper (special filter paper for
transfer film), transfer film, and gel should be examined thor-
oughly to avoid short circuit.
10. The membrane should be in contact with the blocking solution
completely in order to avoid leaving the blocking solution for a
long time.
11. All the antibodies were diluted with PBST.
12. Synthetic methods and the properties of the mitochondrion-
targeted platinum complexes may vary more or less; the above
representative experimental procedures may need some mod-
ifications accordingly.
422 Suxing Jin and Xiaoyong Wang
Acknowledgments
We acknowledge the National Natural Science Foundation of
China (Grants 31570809, 21877059).
References
1. Johnstone TC, Suntharalingam K, Lippard SJ
(2016) The next generation of platinum drugs:
targeted Pt(II) agents, nanoparticle delivery,
and Pt(IV) prodrugs. Chem Rev
116:3436–3486
2. Deo KM, Ang DL, McGhie B et al (2018)
Platinum coordination compounds with
potent anticancer activity. Coord Chem Rev
375:148–163
3. Wang XY, Guo ZJ (2013) Targeting and deliv-
ery of platinum-based anticancer drugs. Chem
Soc Rev 42:202–224
4. Wang XH, Wang XY, Guo ZJ (2019) Stimuli-
responsive therapeutic metallodrugs. Chem
Rev 119:1138–1192
5. Zhang W, Zhang SL, Hu XH et al (2015)
Targeting tumor metabolism for cancer treat-
ment: is pyruvate dehydrogenase kinases
(PDKs) a viable anticancer target? Int J Biol
Sci 11:1390–1400
6. Chipuk JE, Bouchier-Hayes L, Green DR
(2016) Mitochondrial outer membrane per-
meabilization during apoptosis: the innocent
bystander scenario. Cell Death Differ
13:1396–1402
7. Santo-Domingo J, Demaurex N (2010) Cal-
cium uptake mechanisms of mitochondria.
Biochim Biophys Acta (BBA) Bioenerg
1797:907–912
8. Youle RJ, Bliek AM (2012) Mitochondrial fis-
sion, fusion, and stress. Science
337:1062–1065
9. Martinou JC, Youle RJ (2011) Mitochondria
in apoptosis: Bcl-2 family members and mito-
chondrial dynamics. Dev Cell 21:92–101
10. Souers AJ, Leverson JD, Boghaert ER et al
(2013) ABT-199, a potent and selective Bcl-2
inhibitor, achieves antitumor activity while
sparing platelets. Nat Med 19:202–208
11. Fulda S, Galluzzi L, Kroemer G (2010) Target-
ing mitochondria for cancer therapy. Nature
9:447–464
12. Chen LB (1988) Mitochondrial membrane
potential in living cells. Annu Rev Cell Biol
4:155–181
13. Weinberg SE, Chandel NS (2015) Targeting
mitochondria metabolism for cancer therapy.
Nat Chem Biol 11:9–15
14. Zielonka J, Joseph J, Sikora A et al (2017)
Mitochondria-targeted triphenylphospho-
nium-based compounds: syntheses, mechan-
isms of action, and therapeutic and diagnostic
applications. Chem Rev 117:10043–10120
15. Guo Y, Guo ZJ, Wang XY et al (2019) Enhanc-
ing Cytotoxicity of a monofunctional platinum
complex via a dual-DNA-damage approach.
Inorg Chem 58:13150–13160
16. Wang K, Wang XY, Guo ZJ et al (2019)
Restraining cancer cells by dual metabolic inhi-
bition with a mitochondrion-targeted platinum
(II) complex. Angew Chem Int Ed
58:4638–4643
17. Zhu ZZ, Guo ZJ, Wang XY et al (2019)
Mitochondrion-targeted platinum complexes
suppressing lung cancer through multiple
pathways involving energy metabolism. Chem
Sci 10:3089–3095
18. Jin SX, Guo ZJ, Wang XY et al (2018) Impact
of mitochondrion-targeting group on the reac-
tivity and cytostatic pathway of platinum
(IV) complexes. Inorg Chem
57:11135–11145
19. Zhang J, Nuebel E, Koehler CM et al (2012)
Measuring energy metabolism in cultured cells,
including human pluripotent stem cells and
differentiated cells. Nat Protoc 7:1068–1085
Impact of Platinium Complexes on Mitochondria 423
Chapter 31
In Silico Modeling of the Mitochondrial Pumping Complexes
with Markov State Models
Roger Springett
Abstract
The mechanism of proton pumping by the mitochondrial electron transport chain complexes is still
enigmatic after decades of research. Recently, there has been interest in in silico Markov state models to
model the mitochondrial pumping complexes at the microscopic level, and this chapter describes the
methods of constructing and simulating such models.
Key words Markov state models, Mitochondria, Proton pumping, Rate constants, Gillespie algorithm
1 Introduction
The mitochondrial electron transport chain operates by Mitchell’s
hypothesis of oxidative phosphorylation [1] in which redox free
energy is first converted into a proton motive force (ΔP) by the
three proton pumping complexes. The proton motive force is then
used by the rotary ATP synthase and the ATP transport system to
maintain the cytosolic phosphorylation potential far from equilib-
rium. The pumping complexes have been very amenable to study
because the heme redox centers can be followed with optical spec-
troscopy [2], the iron sulfur (FeS) centers with electron paramag-
netic resonance spectroscopy, and charge movement with time-
resolved electrometry [3]. However, it has been 40 years since
Mitchell’s hypothesis was accepted by the scientific community,
25 years since the first atomic resolution crystal structures of cyto-
chrome oxidase [4] and the bc
1
complex [5] were published, and
10 years for complex I [6], but the mechanism of proton pumping
is still enigmatic. Currently missing is an in silico model that can
provide a framework to understand the underlying principles used
by the complexes to pump protons. Molecular dynamics is a popu-
lar method to model proton pumping function, particularly water
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_31,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
425
and proton movement, but it cannot model electron transfer events
or operate on the time scale of pumping. Markov state models are
an alternative modeling paradigm, and several models have been
published attempting to simulate complex I [7,8], the bc
1
complex
[9–11], and cytochrome oxidase [12,13] at the microscopic level.
The goal of this chapter is to describe the methods of construction
and simulation of Markov state models with an emphasis on the
mitochondrial pumping complexes. The Ransac et al. model of the
bc
1
complex [9,10] is used to illustrate the process.
2 Ransac Model of the bc
1
Complex
The bc
1
complex operates via the modified Q cycle (Fig. 1) in which
ubiquinol (UQH
2
) binds at the Qo center close to the P-side of the
membrane. The two electrons are bifurcated, so that the first elec-
tron is passed to the iron sulfur center (FeS) in the iron sulfur
protein (ISP), and the second is passed to low potential b-heme
(b
L
). The ISP is mobile and is able to move between the Qo center
(b
L
conformation) and c
1
(c
1
conformation) where the first electron
can be transferred to heme c
1
and then to cytochrome c(Cytc) that
binds on the P-side of the membrane. The second electron is passed
across the membrane to the high potential b-heme (b
H
) and then
Fig. 1 A cartoon of the bc
1
complex showing the reaction sequence of the
modified Q cycle. Electron transfers are shown in black, ISP movement in green,
and substrate movement shown in gray
426 Roger Springett
used to reduce ubiquinone (UQ) bound at the Qi center to semi-
ubiquinone (UQS) on the N-side of the membrane. This process is
then repeated to provide a second electron to reduce UQS to
UQH
2
which is released into the membrane.
3 Markov State Models
A state model is one in which the model exists in discrete states and
transitions between states stochastically according to a set of rate
constants that describe a set of reactions. A Markov state model is a
state model that has no memory of previous states, and the prob-
abilities that it will transition from one state to the next are only
dependent on the current state and not the previous states it has
visited.
The Michaelis–Menten model of invertase [14] is an example
of a very simple Markov state model. The states and elementary
reactions of the model are shown in reaction 1where the substrate
(S) binds to an enzyme (E) to form the enzyme-substrate complex
(ES). The bound substrate is converted to the product (P) to form
the enzyme-product complex (EP), and then the product is
released.
ð1Þ
In this example, the model can exist in three states (E, ES, and
EP) with three elementary reactions connecting the states. The
probability at any instant in time that it will transition to another
state in an infinitesimal time interval of δtis given by kδtwhere kis
the rate constant. For instance, when the model is in state EP, the
probability of releasing the product and transitioning to E is k
+3
δt
and for the product to revert to substrate to form ES is k
2
δt. The
model is a Markov model because the transition probabilities are
independent of the previous history: when the model is in state EP,
the probability of it transitioning to state E or ES is independent of
whether it was previously in state ES and the substrate converted to
product or was in state E and the product bound.
From a theoretical perspective, a Markov state model is a good
model of an enzyme because an enzyme exists in different states,
e.g., unbound, substrate bound, or product bound, with the
enzyme transitions between states stochastically governed by ther-
mal energy and the activation energy barriers, and the transition
probabilities are independent of the previous history of the enzyme.
The precise correspondence between the enzyme kinetics predicted
by the model and those experimentally determined for invertase is
evidenced that a Markov state models can be used to model enzyme
function.
Modeling the Mitochondrial Pumps 427
Examining the chemistry of the invertase reaction, in which
sucrose is hydrolyzed to glucose and fructose, immediately shows
that reaction 1is a simplification and that the EP state in reaction 1
actually has two products bound (glucose and fructose) which most
likely can be released or bound independently (reaction 2).
ð2Þ
This is an example of granularity: how close the model must
replicate the enzyme mechanism. This is particularly difficult to
address for the conformational changes of an enzyme because
molecular dynamics emphasizes that the protein conformation is a
continuous variable, whereas a Markov state model has discrete
states. This will be addressed in Subheading 9.
Another difference between reactions 1and 2is that the former
is a linear reaction mechanism, whereas the latter is branched. The
complexity of the proton pumps means that the reaction mecha-
nism will be highly branched, and this is contrary to the typical
narrative describing the reaction mechanism (see the example in
Subheading 2).
4 States and Reactions
The first task in developing a Markov state model is to define the
states of the model. A rational approach is to define centers and the
possible states of the centers based on the structure and function of
the complex. The hemes and FeS center of the bc
1
complex are
redox centers which can be either oxidized or reduced. The Qo and
Qi centers can either be empty or have UQH
2
, UQS, or UQ
bound. Likewise, the Cytcbinding site is a center which can either
be empty or have reduced or oxidized Cytcbound. In the enzyme,
the ISP moves between the c
1
and the b
L
conformation, and this can
be captured in the model as conformational center which has either
the c
1
or b
L
conformation.
The state of the model is described by a state vector which
describes the state of each center. For example, the centers in the
state vector of the Ransac model are b
L
,b
H
, FeS, c
1
, Qo, Qi, Cytc,
and ISP. The model has 2 2224432¼1536
states, and a typical state vector could be described as <R, R, O, O,
UQH
2
, Empty, R, b
L
>for the state with b
L
and b
H
reduced, FeS
and c
1
oxidized, the Qo site bound with UQH
2
, the Qi site empty,
reduced Cytcbound at the Cytccenter, and the ISP is in the b
L
conformation.
The Ransac model combined the electron and proton transfer
in the oxidation of UQH
2
and so only needed to define UQH
2
,
428 Roger Springett
UQS, and UQ bound at the Qo and Qi centers. A more granular
model could separate the electron and proton transfers and define
all the oxidation intermediates.
The next step is to define the rate constants connecting the
states in the model. The large number of states of a typical model
means that there are a very large number of rate constants which
cannot be hand coded and must be generated automatically from a
simpler and more functional definition of an elementary reaction.
For instance, the electron transfer reaction from b
L
to b
H
in the bc
1
complex connects states with b
L
reduced and b
H
oxidized to one
with b
L
oxidized and b
H
reduced with the other centers not chang-
ing. In terms of state vectors, the initial and final states of the
reaction could be described as <R,O,*,*,*,*,*,*>and <O,R,*,*,
*,*,*,*>, respectively, where the star means any value for the center
that does not change in the final state. This reaction connects
224432¼384 pairs of states in the Ransac
model. The 16 reactions of the Ransac model and their initial and
final state vectors are given in Table 1. The large number of reac-
tions in the model allows any state in the model to transition to
multiple other states, so that the reaction sequence is highly
branched.
5 Estimation of Rate Constants
Each reaction has a forward and reverse rate constants (k
f
and k
r
,
respectively) which are related to the ΔG
0
of the reaction by:
ΔG0¼kBTln kf
kr
ð1Þ
where k
B
is Boltzmann’s constant and Tis the absolute temperature
in Kelvin. For one-electron transfers, the ΔG
0
is given by:
ΔG0¼ Ea
mEd
m
ð2Þ
where E
ma
and E
md
are the midpoint potentials of the electron
acceptor and donor, respectively. For intra-protein proton transfers,
the ΔG
0
is given by:
ΔG0¼kBTln 10ðÞpKaapK ad
ð3Þ
where pK
aa
and pK
ad
are the pK
a
of the acceptor and donor site
respectively. Where the proton transfer is a proton binding from the
bulk media, the same expression can be used with the pK
a
of the
bulk set to zero. It is often more convenient to use the chemical
potential of protons (μH¼k
B
Tln(10)pH) and the chemical
potential of association (μH
a
¼k
B
Tln(10)pK
a
) instead of the
pH and pK
a
, respectively. In this case, the ΔG
0
for proton transfer
is:
Modeling the Mitochondrial Pumps 429
ΔG0¼þ μHa
aμHd
a
ð4Þ
At the first sight, the large number of rate constants, which are
typically difficult to measure experimentally, mean that the model
has numerous adjustable parameters which reduce its predictive
power. However, many rate constants can be estimated. For
instance, the rate constants for electron transfer through a protein
can be estimated from the Moser–Dutton ruler [15] in which the
rate constant in the exergonic direction is given by:
log kex
ðÞ¼15 0:6D3:1ΔG0λ
2
λð5Þ
where Dis the edge to edge distance in Åof the delocalized
electron ring around the redox center and λis the reorganization
energy in eV, generally assumed to be 0.7 eV. This only requires the
ΔG
0
, which is usually well known, and the distance, which can be
obtained from the atomic structure. The rate constant in the end-
ergonic direction can be calculated from the ΔG
0
and the rate
Table 1
The 16 reactions of the Ransac model and their initial and final state vectors
Reaction Initial state Final state
MB UQH
2
at Qo <*,*,*,*,E,*,*,*><*,*,*,*,UQH
2
,*,*,*>
MB UQ at Qo <*,*,*,*,,E,*,*,*><*,*,*,*,UQ,*,*,*>
ET UQH
2
to FeS <*,*,O,*,UQH
2
,*,*,B><*,*,R,*,UQS,*,*,B>
ET UQS to FeS <*,*,O,*,UQS,*,*,B><*,*,R,*,UQ,*,*,B>
ET UQH
2
to b
L
<*,O,*,*,UQH
2
,*,*,*><*,R,*,*,UQS,*,*,*>
ET UQS to b
L
<*,O,*,*,UQS,*,*,*><*,R,*,*,UQ,*,*,*>
ET b
L
to b
H
<O,R,*,*,*,*,*,*><R,O,*,*,*,*,*,*>
ET b
H
to UQ at Qi <R,*,*,*,*,UQ,*,*><O,*,*,*,*,*UQS,*,*>
ET b
H
to UQSi <R,*,*,*,*,UQS,*,*><O,*,*,*,*,UQH
2
,*,*>
MB UQH
2
at Qi <*,*,*,*,*,E,*,*><*,*,*,*,*,UQH
2
,*,*>
MB UQ at Qi <*,*,*,*,*,E,*,*><*,*,*,*,*,UQ,*,*>
PM Rieske b
L
to c
1
<*,*,*,*,*,*,*,B><*,*,*,*,*,*,*,C>
ET FeS to c
1
<*,*,R,O,*,*,*,C><*,*,O,R,*,*,*,C>
PB reduced Cytc <*,*,*,*,*,*,E,*><*,*,*,*,*,*,R,*>
PB oxidized Cytc <*,*,*,*,*,*,E,*><*,*,*,*,*,*,O,*>
ET c
1
to Cytc <*,*,*,R,*,*,O,*><*,*,*,O,*,*,R,*>
MB molecular binding, ET electron transfer, PM protein movement, and PB protein binding. The centers of the state
vectors are (b
H
,b
L
, FeS, c
1
, Qo, Qi, Cytc, ISP) with Ooxidized, Rreduced, Eempty, Bb
L
conformation, Cc
1
conformation
430 Roger Springett
constant in the exergonic direction using Eq. (1). The sensitivity of
the rate with separation is used by the bc
1
complex to prevent short
circuits: the FeS center is 8.9 Åfrom the Qo center when the ISP is
in the b
L
position allowing very rapid electron transfer from UQH
2
to FeS, but FeS is more than 28 Åfrom c
1
in this conformation
preventing further transfer to c
1
at significant rates. When the ISP
moves to the c
1
position, the situation is reversed and the FeS center
is now 11.1 Åfrom c
1
but too far from the ubiquinol intermediates
at Qo for significant electron transfer.
To the first approximation, substrate binding can be assumed
to be diffusion limited [16], so that the on rate constant (k
on
)is10
9
to 10
10
M
1
s
1
for small molecules. The off rate constant (k
off
) can
then be calculated from K
d
¼k
off
/k
on
. The concentrations of
NADH in the matrix and UQ/UQH
2
in the membrane are in
the millimolar range [17], so the forward rates will be >10
6
. This
is much larger than the turnover number, and the substrate binding
reactions are likely to operate close to equilibrium and unlikely to
be kinetically limiting. However, the concentration of free protons
in the matrix is only 10
8
M when the matrix has a pH of 8.0 giving
diffusion-limited on-rates of 10
1
to 10
2
per second. This is slower
than typical turnover rates of the bc
1
complex (150e
/s) and
CytOx (40e
/s) [17]. However, the protons are strongly buffered
by inorganic phosphate, citrate, and bicarbonate and other matrix
small molecules, so that concentration of exchangeable protons is
in the millimolar range compared to the nanomolar range for free
protons. Furthermore, proton binding can occur on disassociation
of a water molecule at the binding site and ejection of the hydroxyl.
These effects possibly explain how proton binding can be much
faster than diffusion rates estimated using the free H
+
concentration.
Using a very simple model, Minneart [18] estimated the k
on
and k
off
of Cytcbinding to CytOx to be 40 10
6
M
1
s
1
and
1200 s
1
, respectively, giving a K
d
of 30 μM. The applicability of
this model to CytOx has recently been questioned, and a review of
the literature suggests that the k
on
is an order of magnitude greater
and the K
d
closer to 1 μM[13].
Proton transfers across a hydrogen bond are very fast and have
been estimated to occur at a rate of 10
11
to 10
13
s
1
[19]. There is
no equivalent to the Moser–Dutton rule for intra-protein proton
transfers, and the rates must be estimated to fit experimental data.
6 The Membrane Potential and Charge Transfer Reactions
The membrane potential across the inner membrane (ΔΨ) is the
result of charge separation across the dielectric membrane. For a
simple homogeneous dielectric, the charge separation generates an
electric field (E) within the membrane given by E¼ΔΨ/D, where
Modeling the Mitochondrial Pumps 431
Dis the dielectric thickness. The free energy change for a charge
qmoving a distance dparallel to this electric field is given by
ΔG¼qEd, so that ΔG¼αqΔΨ, where αis the fractional
distance perpendicular to the membrane in the direction from the
P-side to the N-side. The situation is more complicated for a
protein which is likely to have a heterogeneous relative permittivity,
and so αbecomes the fractional dielectric distance, that is, the
distance weighted by the relative permittivity along the path. A
reasonable approximation where no experimental data is available
is to assume that the fractional dielectric distance is equal to the
fractional physical distance measured from the protein structure.
This term must be added to the ΔG
0
of an electron or proton
transfer reaction, so that the ΔG
0
for electron and proton transfers
becomes:
ΔG0¼ Ea
mEd
m
αqΔΨð6Þ
ΔG0¼þ μHa
aμHd
a
αqΔΨð7Þ
where qis 1 for an electron and +1 for a proton. This additional
contribution to the ΔG
0
will change the activation barrier and the
rate constants of the reaction. In the absence of knowing precisely
how the activation barrier will be affected, this term can be equally
split between the forward and reverse rate constants, so they
become:
k0
f¼kfeþ1=2αqΔΨ=kBT
k0
r¼kre1=2αqΔΨ=kBTð8Þ
where k
f
and k
r
are the rate constants in the absence of a membrane
potential.
7 Thermodynamics and Interactions
A major difference between the model and the enzyme is that the
model uses forward and reverse rate constants to define the ΔG
0
between states, whereas each state has a defined energy in
the enzyme and the rate constants are determined by the ΔG
0
and
the activation energy (ΔG
{
) between states. The latter ensures that
the energy is a state function and that the ΔG
0
between any two
states is independent of the path taken between the states. It is
possible to define the rate constants for reactions between states in
the model such that the ΔG
0
between two states at the ends of two
separate paths is different. If this occurs, the model breaks the
second law of thermodynamics, and a thermodynamically inconsis-
tent model cannot accurately simulate the enzyme.
If there are no interactions between centers, then a model will
generally be thermodynamically consistent but, when there are
interactions, the rate constants must be implemented very carefully.
432 Roger Springett
For instance, there is anti-cooperativity between the b
L
and b
H
hemes in the bc
1
complex [20], so that the midpoint potential of
the b
L
heme depends on whether the b
H
heme is oxidized or
reduced, and vice-versa. The origin of this anti-cooperativity is
likely to be the electrostatic repulsion between the two hemes,
which raises the energy of the state when both the hemes are
reduced. This means that the rate constants for transfer of an
electron between UQH
2
at the Qo center and b
L
will depend on
the oxidation state of b
H
and the rate constants for the electron
transfer between b
H
and UQ or UQS at the Qi center will depend
on the oxidation state of b
L
. The model will become thermodynam-
ically inconsistent if the rate constants for these reactions are not
specified very carefully.
Another example is that the structure of the bc
1
complex sug-
gests that a hydrogen bond will form between UQH
2
bound at the
Qo center and the ISP [21]. This will affect the K
d
,ΔG
0
, and rate
constants of UQH
2
binding depending on whether the ISP is in the
c
1
or b
L
conformation as this hydrogen bond can only form in the
latter case. Furthermore, it will also affect the rate constants of ISP
movement because the ΔG
0
of this reaction will change depending
on the intermediate bound at the Qo center.
Interactions will be an important feature of proton pumping
models because it is likely that these interactions facilitate the
coupling of electron and proton transfers. In particular, the mecha-
nism of coupling in CytOx is thought to be electrostatic in origin
[22] such that the charge from the pump proton affects the mid-
point potentials of the hemes changing the electron transfer equi-
librium, and the charge from the electrons affect the pK
a
’s of the
proton binding sites changing the proton-transfer
equilibrium [23].
8 Simulations
The numerical model can be simulated deterministically by using
either coupled ordinary differential equations (ODE) or stochasti-
cally by using the Gillespie algorithm [24].
The ODE approach assembles the chemical master equation
(CME) to calculate the time-dependent probability of finding an
individual enzyme of the reaction mixture in a particular state (P
σ
)
where σis the state. This probability is related to the concentration
of enzymes in state σ(C
σ
)byP
σ
¼C
σ
/C
T
, where C
T
is the
concentration of the enzyme in the reaction mixture. The chemical
master equation is a series of linear equations relating the rate of
change in the probability of being in each state to the probability of
all the states via a set of rate constants. It has the form:
Modeling the Mitochondrial Pumps 433
d
dt
P1
P2
P3
⋮
PN
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5
¼
Pkn,1 k1,2 k1,3 k1,N
k2,1 Pkn,2 k2,3 k2,N
k3,1 k4,1 Pkn,3 k3,N
⋮⋮⋮⋱⋮
kN,1 kN,2 kN,3 Pkn,N
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5
P1
P2
P3
⋮
PN
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5ð9Þ
where Nis the number of states, the sums are over all states n, and
k
r,c
is the rate constant of the reaction which links states rand c. The
rate constant is the k
f
of the reaction if state ris the reactant and
state cis the product of the reaction, and it is the k
r
if vice-versa.
Note that k
n,n
is zero because no reaction has the same state as
reactant and product. Many of the off-diagonal elements are also
zero for a large model because there are many pairs of states not
linked by reactions. The chemical master equation can be written in
matrix form as:
db
P
dt¼b
Kb
Pð10Þ
where b
Pis a column matrix of probabilities and b
Kis the rate
constant matrix. The rows of b
Ksum to zero, so that the determi-
nant of b
Kis zero. The physical meaning of this can be understood
by summing the linear equations of Eq. (9) where the left-hand side
would be the rate of change of the sum of all the probabilities. As
the probabilities must sum to 1, the left-hand side is zero and hence
the right-hand side must also be zero for any set of state
probabilities.
For very simple models, the CME can be solved analytically,
but the solution can be very complex (see [13] for the analytic
solution of a six state model), but is usually solved numerically.
The time evolution of the CME can be solved by the finite differ-
ence method where the probabilities of the states after some time
434 Roger Springett
step δt,b
PtþδtðÞ, is related to the probabilities before the step,
b
PtðÞ,by.
b
PtþδtðÞ¼
b
PtðÞþb
Kb
PtðÞδtð11Þ
The step size must be very small compared to the fastest reac-
tion to ensure the finite difference is a good approximation.
The CME can also be solved in the steady state in which the
left-hand side of Eq. (9) is zero. The CME then provides only
N1 equations (the final equation is equal to the sum of the
other equations) and so cannot be used directly to solve for the
Nprobabilities, and an additional equation is needed. This last
equation can be the constraint that the sum of the probabilities of
all the states must be 1. This allows an NNmatrix to be
constructed of the form:
0
0
0
⋮
1
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5
¼
Pkn,1 k1,2 k1,3 k1,N
k2,1 Pkn,2 k2,3 k2,N
k3,1 k4,1 Pkn,3 k3,N
⋮⋮⋮⋱⋮
111 1
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5
P1
P2
P3
⋮
PN
2
6
6
6
6
6
6
4
3
7
7
7
7
7
7
5
ð12Þ
and the probabilities solved for by matrix inversion.
Typically, an experimental technique does not measure the state
probabilities themselves but rather measures the oxidation state of
the redox centers or the rate of consumption or production of
metabolites. The probability that a redox center is reduced can be
calculated by summing the probabilities of all the states which have
that center reduced. Similarly, the probability that a ubiquinone
intermediate is present in the Qo or Qi center can be calculated
from the sum of the probabilities of states with the intermediate at
the center. The net metabolite flux, J
n
, through a particular reaction
can be calculated from:
Jn¼XkfPikrPfð13Þ
where the sum is over all the pairs of states the reaction connects,
and P
i
and P
f
are the probabilities of the pairs of initial and final
states, respectively.
Solving the CME numerically is an excellent approach for small
models but impractical for large models due to the size of the
matrix. For instance, the Ransac model has 1536 states and requires
Modeling the Mitochondrial Pumps 435
a matrix of 18 megabytes, assuming each matrix element is a double
precision floating point number. This is quite feasible for a modern
desktop PC but modeling the dimer would require a model of
1536 1536 states and a matrix of 40.5 terabytes.
The Gillespie algorithm [24] is an alternative simulation method
that does not require large quantities of memory. It was originally
designed to follow chemical reactions where there are a finite number
of reactant molecules, but it is also suited to follow a single enzyme as
it passes through its possible states performing catalysis. The simula-
tion starts with the enzyme in an initial state and calculates the dwell
time and subsequent state based on the rate constants and two
random numbers. As such it is a stochastic method, and no two
simulations will produce precisely the same time evolution. This is
exactly the way an individual enzyme turns over and the Gillespie
algorithm provides the framework in which to explore how catalytic
function emerges from the enzyme structure.
The algorithm is very simple to implement. First, the reactions
of the model are split into a list of all the possible half-reactions (the
forward half-reaction and the reverse half-reaction) each with a
single rate constant. This list is then parsed to determine which
reactions are possible from the current state and the rate constants
of all the possible half-reactions are summed to give the total
relaxation rate (Σk) from the current state. A random number
between 0 and 1, R
1
, is created and the dwell time set to ln
(R
1
)/Σk. A second random number between 0 and 1, R
2
,is
generated, and the list of half-reactions is parsed again to find
which reaction spans R
2
Σkin the summation (see Fig. 2for a
graphical representation of the reaction choice). This reaction is
chosen and executed to give the next state. This process is then
repeated to create the time evolution of a single enzyme. If this
process were repeated for an infinite number of enzymes, then the
averaged data would precisely reproduce the time evolution
simulated using the chemical master equation. For finite number
of enzymes, the data contains stochastic noise which can be reduced
by increasing the number of simulations; typically 10
4
to 10
6
simulations are required to calculate a time evolution with good
signal to noise ratio.
Fig. 2 Graphical example of how a reaction is chosen with the Gillespie
algorithm. Five reactions are available to the state with rate constants k
1
,...,
k
5
as indicated by the length of the bar. Σkis the sum of these constants, and, in
this case, R
2
Σkchooses reaction 2
436 Roger Springett
Enzyme catalysis is an ergodic process in that the probability of
a single enzyme in an infinite pool of catalyzing enzymes being in a
particular state at an instant in time is equal to the probability that a
single enzyme will be in the same state when sampled at random
over an infinite period of time. In essence, each enzyme passes
through all its possible states over time but spends longer in some
states than others depending on the rate constants. This property
allows the steady state of enzymes in a reaction mixture to be
calculated from the time evolution of a single enzyme over a long
period of time. Typically, the model must be followed for 10
7
reactions to give an estimate of the steady state and 10
9
reactions
for publication quality data.
The oxidation state of redox centers must be calculated during
the simulation by summing the time that the model dwells with the
redox center reduced and then normalizing to total time of the
simulation. The flux through the reaction pathway can be calcu-
lated by counting the number of times a particular reaction was
executed and normalizing to the simulation time.
9 Protein Conformation
Conformational changes of the pumping complexes are likely to
prove critical to the proton pumping mechanism. The pumping by
the membrane domain subunits of complex I is thought to undergo
conformational changes coupled to the reduction of UQ [25–27],
the ISP of the bc
1
complex moves by tethered diffusion from the b
L
conformation to the c
1
conformation as part of the electron transfer
process [28,29], and helix-X of CytOx subunit I has been found to
unwind under certain conditions with a concomitant rotation of
the heme afarnesyl side chain [30,31].
Molecular dynamics highlight that the protein is in constant
motion, so the conformational state of the protein is a continuous
variable of very high dimensionality rather than the discrete steps
used by a Markov state model. The movement of a peripheral
residue can be expected to have little effect on catalysis, whereas
certain movements will be critical. These important movements can
be projected onto one or more intuitive variables, such as the
diffusion path of the ISP, and the movement discretized
[32]. The mitochondrial ADP/ATP carrier cycles between confor-
mations in which the central substrate binding site is made accessi-
ble to the cytosolic side, to one in which access to the binding site is
blocked to both sides, and to one where the binding site is accessi-
ble to the matrix made. These continuous conformational changes
were modeled as 21 steps, and the simulations were able to repro-
duce the kinetic parameters of a series of mutants with good quan-
titative accuracy [33]. While the rate constants and free energy
profile of this conformational change were estimated in that
Modeling the Mitochondrial Pumps 437
study, in the future, it would be possible to calculate them from
molecular dynamic simulations [32]. Such profiles have been esti-
mated for the movement of the ISP [29] and binding of ATP to the
ADP/ATP carrier [34].
References
1. Mitchell P (1979) Keilin’s respiratory chain
concept and its chemiosmotic consequences.
Science 206(4423):1148–1159
2. Chance B, Williams GR (1956) The respiratory
chain and oxidative phosphorylation. Adv
Enzymol Relat Areas Mol Biol 17:65–134
3. Verkhovsky MI et al (2001) Charge transloca-
tion coupled to electron injection into oxidized
cytochrome c oxidase from Paracoccus denitri-
ficans. Biochemistry 40(24):7077–7083
4. Iwata S et al (1995) Structure at 2.8 A resolu-
tion of cytochrome c oxidase from Paracoccus
denitrificans. Nature 376(6542):660–669
5. Xia D et al (1997) Crystal structure of the
cytochrome bc1 complex from bovine heart
mitochondria. Science 277(5322):60–66
6. Efremov RG, Baradaran R, Sazanov LA (2010)
The architecture of respiratory complex
I. Nature 465(7297):441–445
7. Ransac S, Arnarez C, Mazat JP (2010) The
flitting of electrons in complex I: a stochastic
approach. Biochim Biophys Acta 1797
(6-7):641–648
8. Ransac S, Heiske M, Mazat JP (2012) From in
silico to in spectro kinetics of respiratory com-
plex I. Biochim Biophys Acta 1817
(10):1958–1969
9. Ransac S, Parisey N, Mazat JP (2008) The
loneliness of the electrons in the bc1 complex.
Biochim Biophys Acta 1777(7-8):1053–1059
10. Ransac S, Mazat JP (2010) How does antimy-
cin inhibit the bc(1) complex? A part-time
twin. Biochim Biophys Acta 1797
(12):1849–1857
11. Kim N, Ripple MO, Springett R (2012) Mea-
surement of the mitochondrial membrane
potential and pH gradient from the redox
poise of the Hemes of the bc1 complex. Bio-
phys J 102(7):1194–1203
12. Kim N, Ripple MO, Springett R (2011) Spec-
tral components of the alpha-band of cyto-
chrome oxidase. Biochim Biophys Acta 1807
(7):779–787
13. Rocha M, Springett R (2018) Electron transfer
between cytochrome c and the binuclear center
of cytochrome oxidase. J Theor Biol
460:134–141
14. Michaelis L, Menten ML (1913) Die Kinetik
der Invertinwirkung. Biochem Z 49:333–369
15. Moser CC et al (2006) Electron tunneling
chains of mitochondria. Biochim Biophys Acta
1757(9–10):1096–1109
16. Alberty RA, Hammes GG (1958) Application
of the theory of diffusion-controlled reactions
to enzyme kinetics. J Phys Chem 62
(2):154–159
17. Rocha M, Springett R (2019) Measuring the
functionality of the mitochondrial pumping
complexes with multi-wavelength spectros-
copy. Biochim Biophys Acta Bioenerg 1860
(1):89–101
18. Minnaert K (1961) The kinetics of cytochrome
c oxidase. I. the system: cytochrome
c-cytochrome oxidase-oxygen. Biochim Bio-
phys Acta 50:23–34
19. Victoria D, Burton R, Crofts AR (2013) Role
of the -PEWY-glutamate in catalysis at the Q
(o)-site of the Cyt bc(1) complex. Biochim
Biophys Acta 1827(3):365–386
20. Shinkarev VP, Crofts AR, Wraight CA (2001)
The electric field generated by photosynthetic
reaction center induces rapid reversed electron
transfer in the bc1 complex. Biochemistry 40
(42):12584–12590
21. Link TA (1997) The role of the ’Rieske’ iron
sulfur protein in the hydroquinone oxidation
(Q(P)) site of the cytochrome bc1 complex.
The ’proton-gated affinity change’ mechanism.
FEBS Lett 412(2):257–264
22. Rich PR et al (1986) Coupling of charge and
proton movement in cytochrome c oxidase.
Biochim Biophys Acta 1275:91–95
23. Belevich I et al (2007) Exploring the proton
pump mechanism of cytochrome c oxidase in
real time. Proc Natl Acad Sci U S A 104
(8):2685–2690
24. Gillespie DT (1977) Exact stochastic simula-
tion of coupled chemical-reactions. J Phys
Chem 81(25):2340–2361
25. Efremov RG, Sazanov LA (2012) The coupling
mechanism of respiratory complex I—A struc-
tural and evolutionary perspective. Biochim
Biophys Acta 1817(10):1785–1795
438 Roger Springett
26. Efremov RG, Sazanov LA (2011) Structure of
the membrane domain of respiratory complex
I. Nature 476(7361):414–420
27. Brandt U (2011) A two-state stabilization-
change mechanism for proton-pumping com-
plex I. Biochim Biophys Acta 1807
(10):1364–1369
28. Zhang Z et al (1998) Electron transfer by
domain movement in cytochrome bc1. Nature
392(6677):677–684
29. Izrailev S et al (1999) Steered molecular
dynamics simulation of the Rieske subunit
motion in the cytochrome bc(1) complex. Bio-
phys J 77(4):1753–1768
30. Ishigami I et al (2017) Crystal structure of
CO-bound cytochrome c oxidase determined
by serial femtosecond X-ray crystallography at
room temperature. Proc Natl Acad Sci U S A
114(30):8011–8016
31. Shimada A et al (2017) A nanosecond time-
resolved XFEL analysis of structural changes
associated with CO release from cytochrome c
oxidase. Sci Adv 3(7):e1603042
32. Prinz JH et al (2011) Markov models of molec-
ular kinetics: generation and validation. J
Chem Phys 134(17):174105
33. Springett R et al (2017) Modelling the free
energy profile of the mitochondrial ADP/ATP
carrier. Biochim Biophys Acta 1858
(11):906–914
34. Dehez F, Pebay-Peyroula E, Chipot C (2008)
Binding of ADP in the mitochondrial
ADP/ATP carrier is driven by an electrostatic
funnel. J Am Chem Soc 130
(38):12725–12733
Modeling the Mitochondrial Pumps 439
Chapter 32
Monitoring the Mitochondrial Presequence Import Pathway
In Living Mammalian Cells with a New Molecular Biosensor
Maxime Jacoupy, Emeline Hamon-Keromen, and Olga Corti
Abstract
Most mitochondrial proteins are encoded by the nuclear genome, synthesized in the cytosol, and imported
into the organelle. Mitochondrial protein import is therefore vital for the maintenance of mitochondrial
function and cell survival. Alterations in this process are suspected to contribute to various diseases,
including neurodegenerative disorders, such as Alzheimer’s disease and Parkinson’s disease. Our under-
standing of the cytosolic signaling mechanisms and posttranslational modifications regulating the mito-
chondrial import process is still in its infancy and hampered by the lack of tools for its dynamic assessment in
cells. We recently engineered an inducible molecular biosensor for monitoring one of the main mitochon-
drial import routes, the so-called presequence pathway, using a quantitative luminescence-based readout.
Here, we provide basic guidelines for using this probe in common cell types of general use in the scientific
community: HEK293T cells, human fibroblasts, and mouse primary neurons. These guidelines can serve as
a starting point for the development of more elaborated protocols for the dynamic investigation of the
presequence import pathway and its regulation in relevant physiological and pathological conditions.
Key words Biosensor, Mitochondrial protein import, Presequence pathway, TOM machinery, Biolu-
minescence assay
1 Introduction
Mitochondria exert crucial functions in metabolism, the produc-
tion of ATP, cellular signaling, the response to stress, and the
control of apoptosis. They originated from the incorporation by
an ancestor eukaryotic host cell of a prokaryote related to
α-proteobacteria, over 1.5 billion years ago [1]. Virtually all their
genetic information was transferred to the nuclear genome during
evolution. In humans, only 13 of the over 1500 proteins are
encoded by the mitochondrial genome. Mitochondrial proteins
encoded by nuclear genes and synthesized by cytosolic ribosomes
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_32,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
Maxime Jacoupy and Emeline Hamon-Keromen contributed equally to this work.
441
harbor specific signals that target them to receptors of the outer
mitochondrial membrane (OMM). From there, they reach the
appropriate mitochondrial subcompartment through five major
import pathways [2–5]. Nearly two thirds of all mitochondrial
proteins are translocated into the organelle by an N-terminal mito-
chondrial targeting signal (MTS), via the so-called presequence
pathway [6]. The MTS is recognized by the TOM20 and
TOM22 receptors [7,8] of the translocase of outer mitochondrial
membrane (TOM) and directs the transfer of the protein across the
translocation channel, TOM40 [3,9]. The protein is then pulled
into the matrix through the TIM23 channel of the presequence
translocase of the IMM (TIM23 complex). This process depends
on the mitochondrial transmembrane potential (Δψmit) [10–14]
and the presequence translocase-associated motor (PAM), includ-
ing the ATP-driven mitochondrial heat shock protein
70 (mtHsp70). The protein is then generally processed by the
matrix mitochondrial processing peptidase MPP [3,15,16],
which removes the MTS, and by other peptidases, before being
folded into its mature form with the help of mtHsp70 and other
chaperones.
Mitochondrial protein import has emerged in recent years as a
process with remarkable plasticity in response to changes in meta-
bolic requirements, stress, and disease [17]. Despite these
advances, relatively little is known about the signaling cascades
and cytosolic mechanisms regulating mitochondrial protein import
in mammalian cells. The study of such mechanisms has been ham-
pered by the lack of tools for the assessment of mitochondrial
protein import in living cells. We recently developed a novel che-
mogenetic tool for assessing the presequence import pathway [18],
exploiting the properties of the Renilla reniformis green fluores-
cent (RGFP) and luciferase (Rluc) proteins. In this anthozoan
organism, bioluminescence is generated by a high-efficiency reso-
nance energy transfer process involving interaction between RGFP
and Rluc and conversion of the blue light emitted by the excited
Rluc substrate coelenterazine (CLZ) into green light. This process
can be reproduced in mammalian cells expressing a fusion construct
between RGFP and Rluc; it is most efficient when the RGFP moiety
is placed at the N-terminus of the fusion protein whereas it is
inhibited by short N-terminal extensions [19]. In addition to the
shift in emission wavelength, the interaction between RGFP and
Rluc is associated with a strong cooperative effect, leading to a
40-fold increase in the intensity of the bioluminescent signal in
the presence of low-quantum yield Rluc substrates, such as
CLZ400A.
Based on these principles, we designed two inducible modular
biosensors (Probes 1 and 2) targeted to mitochondria by a classical
MTS, predicted to yield a strong bioluminescent signal in the
presence of CLZ400A only after import and the subsequent
442 Maxime Jacoupy et al.
proteolytic removal of the MTS in the mitochondrial matrix
[18]. Each probe consists of (1) the N-terminal cleavable MTS
from the human dihydrolipoamide dehydrogenase (DLD), fol-
lowed by (2) an RGFP-Rluc fusion separated by a short linker
region containing a cMyc-tag, and (3) a C-terminal conditional
FKBP-based destabilizing domain (DD), targeting the probe for
proteasomal degradation under basal conditions, and stabilizing it
in the presence of the small molecule Shield1. Probe 2 also contains
a PEST degradation motif (rich in proline (P), glutamic acid (E),
serine (S), and threonine (T)), as a means of optimizing the signal-
to-noise ratio and limiting potential toxicity for specific applica-
tions. We also generated a cytosolic control probe based on the
backbone of Probe 1 but lacking the MTS (Probe 3).
The validation of these tools has been described in [18]. Here,
we provide basic guidelines for using these probes in three cell types
(HEK293T cells, human fibroblasts, and mouse primary neurons),
as a starting point for developing more elaborated protocols
adapted to the users’ models of interest and specific biological
questions.
2 Materials
2.1 Recombinant
Vectors
See Fig. 1for a graphical representation of the vectors.
2.1.1 pCI-Neo
Mammalian Expression
Vectors Encoding Probes
1, 2, and 3
1. pCI-Neo-Probe 1 (MTS—RGFP—cMyc-tag—Rluc—DD).
2. pCI-Neo-Probe 2 (MTS—RGFP—cMyc-tag—Rluc—DD—
PEST).
3. pCI-Neo-Probe 3 (RGFP—cMyc-tag—Rluc—DD).
2.1.2 Recombinant
Lentiviral Vector Encoding
Probe 1
1. rLV-EF1-Probe 1 (MTS—RGFP—cMyc-tag—Rluc—DD)
(see Note 1).
2.2 Materials
and Reagents
2.2.1 Equipment and Cell
Culture Material
1. Neon Transfection System, Invitrogen, or an equivalent device.
2. SpectraMax i3x microplate reader, Molecular Devices, or a
similar device recording bioluminescent and fluorescent signals
(see Note 2).
3. Black 96-well cell culture plates, coated with poly-D-lysine,
with clear bottom (Greiner Bio-One 655,946, or equivalent);
384-well plates can also be used (see Note 3).
4. Multichannel pipettes (8 or 12 channels) and corresponding
reagent reservoirs.
Monitoring the Mitochondrial Presequence Import Pathway In Living... 443
2.2.2 Cell Culture Media,
Buffers, Reagents
1. B-27 supplement (50, Gibco).
2. Dulbecco’s Phosphate-Buffered Saline without calcium and
magnesium (DPBS, Gibco).
3. Dulbecco’s Modified Eagle Medium with high glucose (4.5 g/L),
GlutaMAX supplement, pyruvate (DMEM-GlutaMAX, Gibco).
4. Fetal bovine serum (FBS).
5. L-Glutamine (200 mM, 100, Gibco).
6. Hank’s Balanced Salt Solution without calcium and magne-
sium (HBSS).
7. HEPES 1 M.
8. Neurobasal A medium (Gibco).
Fig. 1 Mammalian expression plasmids and recombinant lentiviral vector for monitoring mitochondrial import
through the presequence pathway in mammalian cells. (a–c) Maps of the mammalian pCI-Neo vectors
encoding Probes 1–3 showing the domain composition of each probe. (d) Map of recombinant lentivirus
encoding Probe 1 under control of the house-keeping EF1-αpromoter. MTS: Mitochondrial Targeting
Sequence, RGFP: Green Fluorescent Protein from Renilla reniformis, Rluc: luciferase from Renilla reniformis,
DD: FKBP-based destabilizing domain, PEST: destabilizing motif rich in proline (P), glutamic acid (E), serine (S),
and threonine (T). All the maps have been designed using the SnapGene software (Insightful Science; https://
snapgene.com)
444 Maxime Jacoupy et al.
9. Opti-MEM (Gibco).
10. Penicillin/Streptomycin (Gibco).
11. Trypsin-EDTA (0.05%, 1, Gibco).
2.2.3 Transfection
Reagents
1. Lipofectamine 2000 (Invitrogen).
2. Lipofectamine RNAiMax (Invitrogen).
3. Neon™Transfection System 100 μL Kit (Invitrogen).
2.2.4 Chemical
Compounds
1. Carbonyl cyanide 3-chlorophenylhydrazone (CCCP, Sigma-
Aldrich) (see Note 4).
2. Coelenterazine 400A (CLZ400A, Santa Cruz) (see Note 5).
3. Shield1 (0.5 mM, Takara) (see Note 6).
2.2.5 Immuno-
cytochemistry
1. Normal Goat Serum (NGS, Gibco).
2. Paraformaldehyde 37%.
3. Triton X-100.
4. Anti-GFP antibody (Abcam, ab13970).
3 Methods
When designing your experiment, we recommend to include three
to six wells per condition of interest, including conditions treated
with CCCP and not treated with Shield1 as controls for back-
ground luminescence. Do not use perimeter wells and fill them
with medium or sterilized water to avoid evaporation throughout
the plate.
3.1 Cell Culture:
Seeding, Transfection
or Lentiviral
Transduction,
Induction
of the Probes
These protocols have been optimized for experiments in HEK293T
lines, human primary fibroblasts, and mouse primary cortical neu-
rons (see Note 7 and Fig. 2a).
3.1.1 HEK293T Cells 1. Plate 20,000 cells per well in black, PDL-coated 96-well plates.
Use 200 μL per well of DMEM-GlutaMAX supplemented with
10% FBS and Penicillin/Streptomycin (see Note 8) as culture
medium.
2. After 24 h, transfect cells with pCI-Neo-Probe 2 and, as a
control, with pCI-Neo-Probe 3 (see Note 9).
(a) Use Lipofectamine 2000 for transfection with plasmids
only and follow the manufacturer’s instructions. As an
indication, start with a total of 175 ng plasmid DNA,
use a pCI-Neo-Probe to empty vector ratio of 1:6
Monitoring the Mitochondrial Presequence Import Pathway In Living... 445
Fig. 2 Graphical illustration of the protocol for the presequence mitochondrial import assay, adapted to
HEK293T cells, primary human fibroblasts, and primary cortical neurons. (a) HEK293T cells (left panel) are
seeded at day 1 (D1), transfected using liposomes ( ) at day 2 (D2), and treated with Shield1/CCCP on day
446 Maxime Jacoupy et al.
(see Note 10), and 0.6 μL of Lipofectamine 2000 in a total
of 40 μL Opti-MEM per well. In case of co-transfection
with another protein-encoding plasmid, you may start
your optimization with 25 ng of pCI-Neo-Probe, 75 ng
of the plasmid of interest, and 75 ng of empty vector (see
Note 11).
(b) For co-transfection with pCI-Neo-Probe and siRNAs, use
Lipofectamine RNAiMax and follow the manufacturer’s
instructions. As an indication, start with a total of 35 ng of
plasmid DNA, use a pCI-Neo-Probe to empty vector/
plasmid of interest of 1:6 (see Note 10), 35 pM siRNA,
and 0.6 μL of lipofectamine RNAiMax in a total of 40 μL
Opti-MEM per well.
3. Add transfection mixture to each well, without changing the
medium.
4. Perform reporter activity assay 48–72 h after transfection.
3.1.2 Human Primary
Fibroblasts
1. Electroporate human primary fibroblasts with pCI-Neo-Probe
1(see Note 12), following the manufacturer’s instructions with
the following settings: one pulse of 1400 V during 20 ms. As an
indication, start with a total of 175 ng of plasmid DNA per well
and use a pCI-Neo-Probe to empty vector/plasmid of interest
ratio of 2:1 (see Note 10).
2. Seed cells on black, PDL-coated 96-well plates at a density of
20,000 per well. Use 200 μL per well of DMEM-GlutaMAX
supplemented with 10% FBS without antibiotics as culture
medium.
3. Perform reporter activity assay 48–72 h after transfection.
3.1.3 Mouse Primary
Cortical Neurons
1. Prepare mouse primary cortical neurons according to standard
procedures (see Note 13).
ä
Fig. 2 (continued) 4 (D4). Primary human fibroblasts (middle panel) are transfected by electroporation ( ),
seeded at D1, and treated with Shield1/CCCP on D3. Primary mouse neurons (right panel) are seeded at D1,
transduced with recombinant lentivirus ( ) on D7, and treated with Shield1/CCCP at D14 or later time points
(see Note 15). (b) RGFP signals and background bioluminescence are measured for HEK293T cells and human
fibroblasts before CLZ400A treatment. For primary neurons, bioluminescence is acquired first, and cells are
immediately fixed and immuno-stained for RGFP fluorescence recording. After analysis (c), the data can be
represented as line (d) or bar plots (e). (This figure was created using Servier Medical Art templates, which are
licensed under a Creative Commons Attribution 3.0 Unported License; https://smart.servier.com. The image on
panel (d) is based on data presented in [18])
Monitoring the Mitochondrial Presequence Import Pathway In Living... 447
2. Plate the neurons at a density of 50,000 per well in black,
PDL-coated 96-well plates. Cultivate the cells in 200 μL per
well of Neurobasal A medium supplemented with 2% FBS, 1%
L-Glutamine, Penicillin/Streptomycin, and 1B27.
3. Change half of the medium every 2 days.
4. After 7 days in vitro, transduce the cells with the lentiviral
vector expressing Probe 1 (see Note 12). As an indication,
start with MOIs between 1 and 10 (see Note 14).
5. Perform reporter activity assay between day 7 and day 10 post-
transduction (see Note 15).
3.2 Mitochondrial
Import Reporter Assay
3.2.1 Induction
of the Mitochondrial
Import Probe
1. Treat the cells with 0.5 μM Shield1 6 h before bioluminescence
recording (see Note 16).
2. Treat selected wells with CCCP (1–10 μM) as a control (see
Note 17).
3.2.2 RGFP
Fluorescence Assay
For each well, the bioluminescent signal emitted by the probe will
be normalized to its fluorescent signal, to correct for potential
inter-well differences in terms of transfection/transduction effi-
ciency. Different procedures will be used to record fluorescent
signals, depending on their intensity in the cell types of interest,
as will become clear in the following paragraphs. For the following
steps, use a multichannel pipette for media changes (see Note 18
and Fig. 2b).
For HEK293T cells and primary fibroblasts:
The fluorescent signal will be recorded immediately before the
bioluminescence reporter assay, as follows:
1. Wash cells three times with 100 μL of HBSS—20 mM HEPES,
pre-warmed at 37 C, and supplemented with 20 mM HEPES,
leaving 100 μL in each well at the end of the procedure.
2. Record the total RGFP fluorescence from each well with a
microplate spectrophotometer (see Note 2), at an excitation
wavelength of 470 nm and an emission wavelength of 510 nm.
For primary cortical neurons:
In this cell type, the fluorescence signal may be too low to be
detected as described above. In this case, after assaying for Rluc-
dependent bioluminescence, proceed as follows:
1. Wash cells three times with DPBS.
2. Fix cells with a 4% paraformaldehyde solution diluted in DPBS
for 20 min at room temperature (RT).
3. Permeabilize cells with a 0.2% Triton X-100 solution diluted in
DPBS for 10 min at RT, under gentle agitation.
448 Maxime Jacoupy et al.
4. Wash cells three times with DPBS.
5. Block nonspecific binding of antibodies with a solution of 10%
NGS diluted in DPBS.
6. Incubate cells with anti-GFP primary antibody diluted in 2%
NGS overnight at 4 C.
7. Wash cells with DPBS and incubate with secondary antibody in
2% NGS diluted in DPBS for 45 min at RT. Protect from light
(see Note 19).
8. Measure total RGFP fluorescence with a spectrophotometer.
3.2.3 Luciferase
Bioluminescence Reporter
Assay (Rluc)
For HEK293T cells and primary fibroblasts:
1. Add CLZ400A diluted in 100 μL of HBSS—20 mM HEPES,
pre-warmed at 37 C, at a final concentration of 10 μM for cell
lines, and 25 μM for primary fibroblasts.
2. Measure total bioluminescence before and immediately after
addition of CLZ400A (see Note 20). As an indication, we
recommend to collect bioluminescence signals every second
for 5 min.
For primary cortical neurons:
1. Wash cells three times with 100 μL of HBSS—20 mM HEPES,
pre-warmed at 37 C, and supplemented with 20 mM HEPES,
leaving 100 μL in each well at the end of the procedure.
2. Follow the procedure described in step 1, using 25 μM
CLZ400A.
3. Follow the procedure in Subheading 3.2.2,step 2. to immu-
nostain the cells with anti-GFP antibodies.
3.2.4 Analysis
of the Results
1. Subtract any background bioluminescence from the signals
recorded after the addition of CLZ400A to the culture
medium (see Note 20).
2. Normalize bioluminescence signals to the total RGFP fluores-
cence in each well.
3. Subtract the normalized mean signal obtained for cells treated
with CCCP (see Note 21 and Fig. 2c).
4. Option 1:
(a) Represent the data normalized as indicated above
obtained over the 5 min recording period as line plots
(Fig. 2d).
5. Option 2:
(a) Represent bioluminescence peaks as bar plots. Consider
obtaining peak values as mean of the normalized biolumi-
nescence signals recorded during the first 100 s (see
Fig. 2e).
Monitoring the Mitochondrial Presequence Import Pathway In Living... 449
4 Notes
1. Lentiviral vector was produced as previously described [18,20]
at titers of approximately 1 10
6
TU/μL, as estimated by
quantitative PCR (qPCR) [21]. Store aliquots at 80 C and
do not freeze and thaw more than once.
2. In our paper, we used the automated Functional Drug
Screening System (FDSS) from Hamamatsu in most of our
experiments [18]. For fluorescence recordings, automatic
microscope settings can also be used, such as CellInsight™
CX5 High Content Screening (HCS).
3. You can also use 384-well plates, but in our hands, inter-well
reproducibility is lower.
4. Prepare a stock solution at 20 mM CCCP in DMSO. Aliquot in
dark tubes and store at 20 C. Discard aliquot after first use.
5. Prepare a stock solution at 50 mM CLZ400A in DMSO.
Aliquot in dark tubes and store at 20 C. Do not freeze and
thaw more than three times.
6. Store Shield1 at 20 C. Purchase small amounts and test after
3–6 months of storage, as Shield1 is unstable.
7. You can adapt the protocol to other cell types.
8. For transfection with RNAiMAX, do not add antibiotics.
9. In our hands, transient transfection of HEK293T cells with
pCI-Neo-Probe 1 leads to mitochondrial toxicity, as judged
by the loss of intensity of the mitochondrial fluorescent dye
tetramethyl rhodamine methyl ester (TMRM) in cells with
high expression levels of the probe [18]. Depending on the
cell type of interest, the most adequate probe may be chosen.
10. In our hands, this ratio limits the toxicity of the probe and
yields optimal results in terms of signal/noise ratio. You may
have to adapt it to your experimental conditions.
11. If necessary, empty vector can be replaced by other protein-
encoding plasmids of interest.
12. In our hands, expression of Probe 2 is extremely low in primary
fibroblasts and primary neurons. We therefore recommend
using pCl-Neo-Probe 1 in these cells.
13. This protocol has been adapted for primary mouse cortical
neurons dissected at E14.5. Our dissection protocol is adapted
from Scirretta et al. [22].
14. According to your needs, you may transduce the cells at other
time points, including before seeding.
450 Maxime Jacoupy et al.
15. The time of transduction may be adapted according to your
needs, but should not be shorter than 7 days.
16. Depending on your needs, you may adapt the time of induc-
tion (Shield1) for optimal signal/noise ratio. Induction times
should be kept as short as possible for the signal of the probe to
reliably reflect the import process.
17. You may consider optimizing the CCCP concentration and
renewing the toxin, depending on the time of induction
(Shield1).
18. If possible, use an electric multichannel pipette or be quick
when changing the medium to avoid drying of the cells. Be
careful not to generate bubbles, as these will disturb signal
recording.
19. Wrap with aluminum foil to avoid bleaching of RGFP
fluorescence.
20. Depending on the cell type and equipment used, you may want
to check for background bioluminescence before addition of
CLZ400A.
21. In our hands, a bioluminescence ratio of about 5 is observed
between conditions without and with CCCP for induction
times (Shield1) of 6 h and CCCP concentrations of 10 μM,
without renewal.
Acknowledgments
We thank the ICM iVector facility for generating and producing
lentiviral particles. This work has received support from Institut
national de la sante´ et de la recherche me´dicale,Association France
Parkinson,Fondation de France (Engt 2012 00034508), Fondation
ICM,“Investissements d’avenir” ANR-10-IAIHU-06, Institut de
Recherches Servier/Les Laboratoires Servier, the Innovative Med-
icines Initiative 2 Joint Undertaking under grant agreement No
821522 (PD-MitoQUANT; this Joint Undertaking receives sup-
port from the European Union’s Horizon 2020 research and inno-
vation program and EFPIA and Parkinson’s UK). The material
presented and views expressed here reflect the author’s view and
neither IMI nor the European Union, EFPIA, or any Associated
Partners are responsible for any use that may be made of the
information contained herein. M.J. and E.H.K. were supported
by fellowships from the French Ministry of Higher Education and
Research and by Association France Parkinson.
Monitoring the Mitochondrial Presequence Import Pathway In Living... 451
References
1. Archibald JM (2015) Endosymbiosis and
eukaryotic cell evolution. Curr Biol 25(19):
R911–R921
2. Chacinska A, Koehler CM, Milenkovic D,
Lithgow T, Pfanner N (2009) Importing mito-
chondrial proteins: machineries and mechan-
isms. Cell 138:628–644
3. Neupert W, Hermann JM (2007) Transloca-
tion of proteins into mitochondria. Annu Rev
Biochem 76:723–749
4. Pfanner N, Warscheid B, Wiedemann N (2019)
Mitochondrial proteins: from biogenesis to
functional networks. Nat Rev Mol Cell Biol
20(5):267–284
5. Wiedemann N, Pfanner N (2017) Mitochon-
drial machineries for protein import and assem-
bly. Annu Rev Biochem 86:685–714
6. Vo
¨gtle FN, Wortelkamp S, Zahedi RP,
Becker D, Leidhold C, Gevaert K et al (2009)
Global analysis of the mitochondrial
n-proteome identifies a processing peptidase
critical for protein stability. Cell 139:428–439
7. Abe Y, Shodai T, Muto T, Mihara K, Torii H,
Nishikawa S et al (2000) Structural basis of
presequence recognition by the mitochondrial
protein import receptor Tom20. Cell 100
(5):551–560
8. van Wilpe S, Ryan MT, Hill K, Maarse AC,
Meisinger C, Brix J et al (1999) Tom22 is a
multifunctional organizer of the mitochondrial
preprotein translocase. Nature 401
(6752):485–489
9. Melin J, Schulz C, Wrober L, Bernhard O,
Chacinska A, Jahn O et al (2014) Presequence
recognition by the tom40 channel contributes
to precursor translocation into the mitochon-
drial matrix. Mol Cell Biol 34(18):3473–3485
10. Banerjee R, Gladkova C, Mapa K, Witte G,
Mokranjac D (2015) Protein translocation
channel of mitochondrial inner membrane
and matrix-exposed import motor communi-
cate via two-domain coupling protein. Elife 4:
e11897
11. Demishtein-Zohary K, Gu¨nsel U, Marom M,
Benerjee R, Neupert W, Azem A et al (2017)
Role of Tim17 in coupling the import motor
to the translocation channel of the mitochon-
drial presequence translocase. Elife 6:e22696
12. Kang PJ, Ostermann J, Shilling J, Neupert W,
Craig EA, Pfanner N (1990) Requirement for
hsp70 in the mitochondrial matrix for translo-
cation and folding of precursor proteins.
Nature 348(6297):137–143
13. Schulz C, Rehling P (2014) Remodelling of
the active presequence translocase drives
motor-dependent mitochondrial protein trans-
location. Nat Commun 5:4349
14. Ting S-Y, Yan NL, Schilke BA, Craig EA
(2017) Dual interaction of scaffold protein
Tim44 of mitochondrial import motor with
channel-forming translocase subunit Tim23.
Elife 6:e23609
15. Fukasawa Y, Tsuji J, Fu S-C, Tomii K,
Horton P, Imai K (2015) MitoFates: improved
prediction of mitochondrial targeting
sequences and their cleavage sites. Mol Cell
Proteomics 14(4):1113–1126
16. Gakh O, Cavadini P, Isaya G (2002) Mitochon-
drial processing peptidases. Biochim Biophys
Acta 1592(1):63–77
17. Harbauer AB, Zahedi R, Sickmann A,
Pfanner N, Meisinger C (2014) The protein
import machinery of mitochondria-a regu-
latory hub in metabolism, stress, and disease.
Cell Metab 19(3):357–372
18. Jacoupy M, Hamon-Keromen E, Ordureau A,
Erpapazoglou Z, Coge F, Corvol JC et al
(2019) The PINK1 kinase-driven ubiquitin
ligase Parkin promotes mitochondrial protein
import through the presequence pathway in
living cells. Sci Rep 9(1):11829
19. Molinari P, Casella I, Costa T (2008) Func-
tional complementation of high-efficiency res-
onance energy transfer: a new tool for the study
of protein binding interactions in living cells.
Biochem J 409:251–261
20. Zennou V, Petit C, Guetard D, Nerhbass U,
Montagnier L, Charneau P (2001) The HIV-1
DNA flap stimulates HIV vector-mediated cell
transduction in the brain. Nat Biotechnol
19:446–450
21. Barczak W, Suchorska W, Rubis B, Kulcenty K
(2015) Universal real-time PCR-based assay
for lentiviral titration. Mol Biotechnol
57:195–200
22. Sciarretta C, Minichiello L (2010) The prepa-
ration of primary cortical neuron cultures and a
practical application using immunofluorescent
cytochemistry. Methods Mol Biol
633:221–231
452 Maxime Jacoupy et al.
Correction to: Measuring Mitochondrial Hydrogen Peroxide
Levels and Glutathione Redox Equilibrium in Drosophila
Neuron Subtypes Using Redox-Sensitive Fluorophores
and 3D Imaging
Lori M. Buhlman, Petros P. Keoseyan, Katherine L. Houlihan,
and Amber N. Juba
Correction to:
Chapter 8 in: Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine,
Methods in Molecular Biology Series, vol. 2276,
https://doi.org/10.1007/978-1-0716-1266-8_8
The first name of the author Katherine L. Houlihan was unfortunately published with an
error. This has now been corrected.
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8_33,
©Springer Science+Business Media, LLC, part of Springer Nature 2022
The updated version of this chapter can be found at
https://doi.org/10.1007/978-1-0716-1266-8_8
C1
INDEX
A
AcetylCoA ........................................................2, 131, 165
Acidic phosphatase................................37, 43, 46, 47, 99
ACO1........................................................... 88, 90, 91, 95
Acrylamide................................... 93, 105–107, 109, 228,
231, 384, 390, 394, 413
Actin/Hsp70/sugar kinase superfamily ........................ 88
Actin filaments............................................................... 334
Adenosine diphosphate (ADP) .......................... 4, 13–15,
18, 44, 57, 69, 73, 75, 77, 79, 81, 155, 157, 161,
179, 201, 271, 272, 274, 275, 277, 280, 281,
306, 437, 438
Adenoviral transduction ............................................... 296
Adipocytes ............................................................ 285, 288
ADP, see Adenosine diphosphate (ADP)
ADP-ribose-metabolizing enzyme............................... 16 6
ADP-ribosylation ................................................. 165–171
Aedes aegypti................................... 69–73, 75, 78, 79, 83
Affinity-capture.............................................................. 145
Alkaline phosphatases ......................................37, 99, 112
Alzheimer .................................................... 1, 6, 100, 259
Anaerobic glycolysis ........................................................1 9
Animal models............................................................... 265
Antimycin A................................................ 14–17, 20, 70,
74, 180, 186, 190, 310, 420
Anti-ADP-ribose antibody ........................................... 167
Anti-BiP/GRP78 ............................................................ 36
Anti-catalase .................................................................... 36
Anti-Lamp-1.................................................................... 36
Anti-TOM20................................................................... 36
Anti-TOM22.............................................................34, 41
Apaf-1 ............................................................................2 16
Apoptosis .................................................... 2, 19, 90, 143,
188, 215–218, 220, 410, 421, 422, 441
Apoptotic cells...................................................... 215–222
Aspect ratio.................................................. 288, 294, 299
Astrocytoma ..................................................................23 6
Ataxias............................................................................ 19 4
ATO1 ............................................................................... 88
ATO2 ............................................................................... 88
ATP synthase ................................................... 3, 4, 13–15,
17, 18, 68, 75, 190, 217, 221, 271, 282,
317–319, 323, 386, 387, 420, 425
ATP/ADP translocator.................................................2 50
ATP synthesis ............................................. 3, 4, 7, 13, 14,
16–21, 73, 75, 420
Autofluorescence.................................194–197, 201, 266
B
Bak ........................................................................ 216, 410
Basic phosphatase........................................ 43, 46, 47, 49
BAT, see Brown adipose tissue (BAT)
Bax ............................................................... 216, 410, 421
Bc
1
complex ............................................... 425, 426, 428,
429, 431, 433, 437
BCA™protein assay kit................................................3 09
Bcl-2 family ...................................................................2 16
β-dodecylmaltoside
β-oxidation ..............................................2 , 3, 16, 20, 129
Bicinchoninic assay.......................................................... 63
Bid .................................................................................. 216
Bioenergetics ....................................................... 3, 11–13,
20, 22, 23, 68, 103, 129, 130, 132, 153–163,
165, 194, 259–268, 306, 316, 318, 357, 411,
420, 421
BIOLOG ......................................................133–136, 140
Bioluminescence..................................442, 447–449, 451
Biosensors..................................... 11, 184, 187, 441–451
Biotinylated oligonucleotides....................................... 14 5
Bisacrylamide.......................................105, 106, 109, 231
BK
Ca
...................................................................... 236, 244
Blood oxygenation........................................................25 9
Blue Native polyacrylamide gel electrophoresis
(BN-PAGE) ...................................................6, 7, 9
Blue-Native Gel Electrophoresis ..............................5, 231
BN-PAGE, see Blue Native polyacrylamide gel
electrophoresis (BN-PAGE)
Bradford assay ................................................47, 156, 162
Brain slices ............................................................ 193–201
Brown adipose tissue (BAT)............................... 274, 275,
286, 288, 294, 298–300, 302
C
Caenorhabditis elegans (C. elegans) ................... 397–406
Calcium Green .................................................... 174, 176,
177, 180, 188
Calcium homeostasis..................................................... 259
Calmodulin.................................................................... 174
Volkmar Weissig and Marvin Edeas (eds.), Mitochondrial Medicine: Volume 2: Assessing Mitochondria,
Methods in Molecular Biology, vol. 2276, https://doi.org/10.1007/978-1-0716-1266-8,
©Springer Science+Business Media, LLC, part of Springer Nature 2021
453
Cancers ...................................................2, 20, 31, 37, 41,
90, 100, 130, 140, 211, 224, 259, 343, 371, 409,
410, 416–418, 420, 421
Carbon monoxide (CO)...................................... 249–256
Carbonyl cyanide p-trifluoromethoxyphenylhydrazone
(FCCP) ........................................... 13, 15, 17, 70,
73, 75, 77, 82, 176, 180, 186, 190, 195–201,
307, 310, 317, 318, 323, 413, 420
Carboxyatractyloside.................................................70, 79
Cardiolipin...................................................................6, 58
Cardiomyocytes................................................... 154, 162,
236, 250, 350
Carnitine/acylcarnitine carrier .....................................25 0
Caspase-8.......................................................................21 6
Caspase-9.............................................................. 216, 421
Caspase-dependent apoptosis........................................... 2
Catalase ..............................................................37, 43, 46,
47, 50, 54, 335, 340
Cell death ...............................................87, 88, 130, 205,
215–224, 250, 259, 300, 306
CEB, see Cell Energy Budget (CEB)
Cell Energy Budget (CEB) ................................ 306–308,
312, 316–320
CellRox™............................................205, 207, 208, 210
Cell shredder ........................................................ 344, 347
CellTiter-Glo®............................................ 309, 313, 315
Cellular energy metabolism..........................................1 33
Cerebral spinal fluid (CSF)................................... 22, 358,
362, 363, 369
Cervical dislocation .............................155, 198, 253, 313
Charge transfer reactions.............................................. 431
Citrate synthase (CS) .......................................... 5, 35, 36 ,
77, 88, 131, 345, 349, 354, 386
CLARIOstar ..................................................................30 8
Clark electrode ................................................................ 37
Clear-native (CN) ...............................106, 107, 110, 227
CLZ400A ................................... 442, 445, 447, 449–451
CO, see Carbon monoxide (CO)
Coagulation ....................................................................... 2
Coelenterazine ..................................................... 442, 445
Coenzyme A.................................................................... 16
Complex I........................................................ 2, 3, 73–75,
131, 133, 194, 195, 222, 224, 250, 251, 275,
420, 425, 426, 437
Complex II ............................................... 3, 14, 131, 133,
194, 222, 224, 250, 274
Complex III.................................................. 3, 14, 74, 76,
190, 317, 319, 323, 420
Complex IV............................................... 3, 14, 195, 201
Compound Discoverer™software .............................. 359
Coomassie G250........................ 103–105, 107, 109, 111
Coronavirus ....................................................................... 2
COVID-19 ........................................................................ 2
COX, see Cytochrome c oxidase (COX)
C2C12, see Mice muscle cell (C2C12)
CRISPR/Cas9............................................................... 398
Cristae..................................................3, 32, 36, 131, 133
Cryo-imaging ....................................................... 259–268
CIT2, see Peroxisomal citrate synthase (CIT2)
Cit2p-GFP detection ...................................................... 93
Cysteine proteases......................................................... 216
Cytochrome C............................. 3, 5, 13, 69, 75, 81–83,
131, 215–217, 219, 220, 222, 323, 410, 426
Cytochrome c oxidase (COX).............................. 3, 5, 70,
74, 76, 77, 82, 179, 319, 387
Cytokines .....................................................................2, 20
Cytosolic Ca2
+
..................................................... 189, 306
D
Damage-associated molecular patterns (DAMPs) ........ 58
DAMPs, see Damage-associated molecular patterns
(DAMPs)
DAPI staining................................................................ 170
Deoxynucleoside 5’-triphosphates (dNTPs) ..............143,
144, 146, 148, 150
DHE, see Dihydroethidium (DHE)
Diabetes ................................................... 1, 130, 274, 397
Differential centrifugations............................... 12, 32, 34,
35, 45, 179, 182, 231
Digitonin ........................................... 6, 13, 68, 105, 107,
110, 140, 175, 176, 180, 216–218, 221, 224,
229–233
Digitonin-permeabilized plasma membrane ..... 218, 219,
221, 222
Dihydroethidium (DHE) ............................................. 205
Dihydrolipoamide dehydrogenase (DLD) .................. 4 43
dNTPs, see Deoxynucleoside 5’-triphosphates (dNTPs)
Drosophila .......................................................67, 114–126
Drosophila melanogaster (D. melanogaster) ................. 67,
69–73, 75, 78, 80, 82, 116, 236
Drp1, see Dynamin-related protein 1 (Drp1)
Dynamin related GTPase.............................................. 325
Dynamin-related protein 1 (Drp1) .............................. 204
E
Electron transport chain (ETC) ................................2, 23,
103–112, 129, 194, 197, 227, 266, 317, 425
Endothelial cells ............................................................ 236
Energy clamp........................................................ 271–282
Epigenetic alterations.................................................... 130
Epilepsy.......................................................................... 194
ETC, see Electron transfer chain (ETC)
Extracellular acidification............................ 306, 314, 420
F
FACS.............................................................203–211, 218
FastRandomForest algorithm....................................... 288
454 MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index
FCCP, see Carbonyl cyanide
p-trifluoromethoxyphenylhydrazone (FCCP)
F0F1-ATP synthase.............................................. 3, 14, 15
FIJI.................................... 154, 155, 157–159, 169, 286,
289, 297, 299, 301, 302, 399, 406
Fiji-ImageJ............................................................ 155, 286
Flow cytometry ....................................................... 21, 62,
63, 204, 205, 221
Fluorescent assessment of H
2
O
2
production .............. 278
Fluorescence imaging .......................................... 155, 259
Fluorescence quenching ................................................. 11
Form Factor (FF)................................................. 286–288
FreeMitos......................................................................... 57
FRET-based cameleons.................................................1 74
Fumarate hydratase .............................................. 131, 319
Fura-2 .................................................................. 174, 178,
181, 182, 189
G
GAL4 ........................................................... 115, 116, 120
GCamPs......................................................................... 17 4
GECOs ..........................................................................1 74
Gillespie algorithm............................................... 433, 436
Glucose-6-phosphatase............................... 37, 46, 47, 51
Glucose-6-phosphate assay............................................. 43
Glutaminolysis......................................... 1 7, 20, 305, 306
Glutathione disulfide (GSSG) ............................. 114, 249
Glutathione redox equilibrium ........................... 114–126
Glutathionylation ................................................. 249, 250
Glycolytic fluxes .................................................. 307, 315,
318, 319, 322
Glycolytic rates .................................................11, 22, 319
Grx 1 ...............................................................................11 4
GSSG, see Glutathione disulfide (GSSG)
H
H
2
O
2
reporters .................................................... 296, 298
Head dissection............................................................... 71
HEK 293, see Human embryonic kidney cell
(HEK 293)
Heme-oxygenase ...........................................................2 50
HIF1α................................................................................ 2
High-resolution accurate mass spectrometry
(HRAM) .......................................... 363, 364, 375
High resolution clear native electrophoresis ............... 103
High-throughput mitochondrial image
analysis ...................................................... 285–302
HILIC, see Hydrophilic interaction liquid
chromatography (HILIC)
HRAM, see High-resolution accurate mass spectrometry
(HRAM)
Human embryonic kidney cell (HEK 293) ................344,
345, 349–354
Huntington ....................................................................... 2
Hybrid electrophoresis................................ 104, 105, 111
Hydrophilic interaction liquid chromatography
(HILIC) .......................... 358, 359, 362–364, 376
I
IDH1 ........................................................... 88, 90, 91, 95
ImageJ......................................................... 155, 286, 300,
339, 406, 417
Imaging of mitochondria .................................... 156, 400
IMM, see Inner Mitochondrial Membrane (IMM)
Immunoblotting .................................. 90, 103–112, 252,
254, 255, 326, 383–394
Immunoprecipitation..........................251, 252, 254, 255
Inflammation..................................................................... 2
In-gel complex V assay.................................................. 231
In-gel digestion .................................................... 389, 391
Inner mitochondrial membrane (IMM).....................3, 4,
21, 68, 75, 79, 81, 107, 131, 216, 235–246, 272,
323, 325, 327, 383, 442
Insect tissues..............................................................67–83
Ion channels ......................................................... 235–246
Iron sulfur protein (ESP) ............................................. 426
Isolation of mitochondria................................. 12, 34, 35,
58, 68, 179, 204, 231, 275, 276
Isopycnic density gradient centrifugation...................... 47
J
JC-1 ............................................................ 37, 42, 44, 46,
52, 53, 204, 413, 417, 422
K
Keratinocytes .................................................................2 36
KIF5B-mediated mitochondrial tubulation ................ 33 3
Kinesin-1........................................................................ 333
Krebs cycle.................................................. 129, 131, 194,
305–307, 310, 315, 317, 319
L
L-carnitine ....................................................................... 16
Leak respiration.........................................................17, 18
Lentiviral transduction.................................................. 445
Lipid-droplet-bound mitochondria .................... 285–302
Lipid droplet number ...................................................2 85
Lipid droplets ......................................................285, 286,
288, 299, 301, 302
Lipofectamine 2000............................................. 445, 447
Lipofectamine RNAiMax..................................... 445, 447
Liquid chromatography/mass spectrometry
(LC/MS) ................................................. 358, 360,
361, 363, 365, 368, 371, 376, 377
Lisinopril........................................................................ 265
MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index 455
M
Magnetic beads ................................................37, 41, 404
Malate ................................................... 14–16, 22, 69, 73,
131, 137, 139, 155, 157, 161, 219, 222, 224,
251, 274, 277, 280, 282
Marker enzyme assays ...............................................47–54
Markov state models ............................................ 425–438
MARylated proteins ...................................................... 167
Mass spectrometry (MS)....................................... 23, 104,
133, 205, 322, 345, 347, 348, 358, 359,
361–368, 377, 383–395
Mechanical permeabilization................ 71–73, 75, 80, 82
Membrane potentials ...................................... 20–22, 154,
158, 193, 204, 205, 207, 217, 221, 223,
272–274, 285, 306, 345, 410, 431, 432
Membrane stiffness .............................................. 343–354
Menten, M.L.................................................................24 7
Metabolic diseases.....................................................1, 397
Metabolism................................................. 2, 3, 7, 10, 12,
16, 17, 19, 20, 22, 23, 31, 34, 67, 68, 75, 114,
129–140, 143, 144, 165, 173, 175, 194, 204,
259, 305–322, 359, 410, 441
Metabolomics.........................................22, 23, 357, 358,
368, 371, 372, 376, 378
MetaMorph ................................................................... 155
Mice muscle cell (C2C12).................................. 345, 346,
349–351, 353, 354
Michaelis, L. .................................................................. 247
Microfluidics-based cell shredder........................ 345, 351
Microscale cell shredder.............................. 344, 346, 352
Microtubules .......................................333, 334, 338–341
MitoBK
Ca
............................................236, 237, 244, 246
Mitochondria enrichment.........................................47, 54
Mitochondria imaging.................................................. 15 6
Mitochondria isolation from brain cortex ................... 251
Mitochondria isolation from cell culture............ 250, 252
Mitochondrial DNA (mtDNA).................................6, 23,
24, 32, 77, 88, 144, 203, 238, 240, 245, 246,
397, 410, 411, 419
Mitochondrial DNA depletion syndromes.................. 1 44
Mitochondrial dNTP pools ................................. 143–151
Mitochondrial dysfunctions.......................................1–24,
34, 88–90, 130, 215–224, 259, 260, 266, 343,
359, 410
Mitochondrial extraction................................................ 61
Mitochondrial functions................................... 13 , 21–23,
33, 67–83, 88, 133, 134, 137, 154, 165, 194,
227, 228, 232, 259, 267, 274, 275, 281, 320,
343, 359, 398, 409–423
Mitochondrial fusion .....................................23, 325, 333
Mitochondrial heat-shock protein
70 (mtHsp700) ................................................. 442
Mitochondrial hydrogen peroxide ...................... 114–126
Mitochondrial impor t reporter assay ........................... 448
Mitochondrial integrity ....................................13, 81, 83,
188, 344, 349
Mitochondrial membrane potentials................20, 21, 37,
161, 175, 193–201, 204, 216–218, 220, 221,
271, 308, 398, 417
Mitochondrial morphology.................................... 33, 68,
286–288, 397–406, 410, 411, 417
Mitochondrial network formation (MNF)................. 333,
334, 336, 339, 341
Mitochondrial permeability transition pore
(mPTP) ..............................................................2 75
Mitochondrial presequence import pathway...... 441–451
Mitochondrial protein glutathionylation............ 249–255
Mitochondrial protein import......................................4 42
Mitochondrial pumping complexes .................... 425–438
Mitochondrial retrograde signaling .......................87–100
Mitochondrial supercomplexes ........................... 107, 110
Mitochondrial superoxide .................................. 205, 219,
413, 416, 417
Mitochondrial targeting signal (MTS) ............... 442–444
Mitochondrial ultrastructure.......................................... 32
Mitochondrial uptake .......................................... 188, 416
Mitochondria mass...................................... 204, 205, 207
Mitochondria purification ..................................... 61, 340
Mitochondria-specific dyes.................................. 204, 211
Mitofusin-mediated fusion ........................................... 33 3
Mitoplast isolation ........................................................ 179
Mitoplast patch-clamping............................................. 18 3
Mitoplast preparations ................................ 239, 242, 246
Mitoplasts ................................................... 179, 182, 189,
237, 238, 240, 242–246
Mito-roGFP2-Grx1 ............................114–116, 120, 125
Mito-roGFP2-Orp1............................115, 116, 120, 125
MitoSOX™................................................ 205, 207, 208,
219, 223, 224, 413, 417, 422
Mito-TASK3.................................................................. 236
MitoTracker™Green .......................................... 204, 207
Mitotracker™Orange ..................................................2 04
Mitotracker™Red ........................................................ 20 7
MitoXpress®-Xtra ............................................... 307–309,
312, 313, 318
Mono-ADP-ribosylhydrolase .......................................1 66
Mouse primary cortical neurons ......................... 445, 447
mPTP, see Mitochondrial permeability transition pore
(mPTP)
mtDNA, see Mitochondrial DNA (mtDNA)
MtDNA damage .................................................. 417–419
mtHsp70, see Mitochondrial heat-shock protein
70 (mtHsp700)
MTS, see Mitochondrial targeting signal (MTS)
Multi-color flow cytometric analysis ................... 203–211
Multi-parametric cell energy budget
platform .................................................... 305–323
456 MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index
Muscle mitochondria ..........................274, 277, 280, 282
Myocardial ischemia...................................................... 25 0
Myosine light chain kinase .................................. 174, 386
N
NADH autofluorescence ..................................... 197, 201
NADH/NAD
+
..............................................................2 , 3
NADH Ubiquinone Reductase........................................ 3
NADPH.................................................................. 21, 194
Nanoscale liquid chromatography ............................... 384
Nanotubules ..................................................................33 3
Native gel electrophoresis.................................... 103–112
Natural compounds ...................................................... 130
Nematodes................................................... 398, 399, 406
Neon™Transfection System........................................ 445
Neurodegenerative diseases.............................1, 6, 37, 41
N-formylated peptides .................................................... 58
Nile red........................................................ 288, 296, 301
Nitrotyrosine ........................................................ 383, 384
Nitrotyrosine-containing proteins ...................... 383–394
NMR Spectroscopy.............................275, 276, 278–280
NMR-based ATP assay ................................................. 278
Nobel Prizes ...................................................67, 235, 236
nucDNA, see Nuclear DNA (nucDNA)
Nuclear DNA (nucDNA) ................................... 144, 203,
240, 246
O
O2K
®
Oxygraph................................................................. 9
Obesity......................................................... 1, 32, 41, 259
Oligomycin....................................................5, 13–15, 17,
18, 70, 75, 79, 155, 157, 176, 186, 190, 233,
271, 278, 310, 318, 413, 420, 421
OMA1................................................................... 325–331
OMA1 knockout models.............................................. 326
1D
1
H N MR ................................................272, 279–281
1D
31
P NMR................................................................. 28 1
Optic Atrophy Protein (OPA1) .................... 23, 325–327
OPA1, see Optic Atrophy Protein (OPA1)
Optical metabolic imaging ........................................... 25 9
Orbitrap mass spectrometer ......................................... 35 8
ORCA-Flash4.0 ............................................................1 55
Oroboros ................................................9, 10, 15, 17, 21,
32, 72–74, 76, 78, 81
Orp1 ..................................................................... 114, 288
Oxidoreductases ................................................... 3–5, 386
OXPHOS................................. 2, 4–6, 14–16, 19–21, 24,
75, 77, 79, 80, 129, 266, 307, 315, 317, 319, 322
Oxygen consumption .............................. 4, 9, 15, 17, 37,
72–78, 80–83, 130, 216, 218, 221, 222, 259
Oxygen consumption rate (OCR) ......................... 68, 81,
204, 306–309, 312–314, 316–320, 322, 412, 420
Oxygenation imaging ................................................... 259
Oxygraphs..................................... 9, 10, 15, 17 , 218, 219
P
Parkinson’s disease........................................................ 39 7
PARylation.....................................................................16 6
Patch-clamp ........................................175, 179, 235–238,
240, 242, 243
Patch-clamp recording......................................... 184, 242
Pentose phosphate pathway (PPP) ............ 197, 305, 306
Pericams......................................................................... 174
Peri-droplet mitochondria............................................ 28 6
Peri-droplet mitochondrial H
2
O
2
production ............ 288
Permeabilization................................................10, 12, 13,
15, 16, 19, 21–23, 67–83, 134, 135, 167, 168,
175, 216, 221, 224, 250
Peroxisomal citrate synthase (CIT2)................. 88–92, 95
PHERAstar ....................................................................308
Phosphorylating respiration ........................14, 15, 17–19
PH-Xtra™...........................................307–311, 316, 318
Plasma ...................................................13, 37, 46, 49, 61,
68, 69, 156, 162, 216, 217, 221, 224, 236, 346,
358, 361, 363, 368, 369, 377, 378
Platelet-derived microvesicles...................................57, 58
Platelet-derived mitochondria..................................57, 64
Platelet mitochondria................................................57–65
Platinum complexes ............................................. 409–422
Porins............................................................................... 32
Pre-adipocytes ...................................................... 294, 302
Pre-sequence translocase-associated motor
(PAM) ................................................................44 2
Primary brown adipocytes............................................ 29 4
Primary fibroblasts ............................4, 17, 445, 447–450
Proapoptotic proteins ................................................... 215
Pro-caspase-9................................................................. 216
Proline................................................................69, 72, 79,
392, 443, 444
Protein measurements ........................................... 47, 318
Protein misfolding .......................................................... 88
Proton gradient...........................................................3, 52
Proton leaks........................................................... 4, 7, 14,
16, 18, 21, 75, 260
Protonmotive force.............................................. 3, 68 , 77
Purified motor proteins ....................................... 334, 338
Pyruvates........................................ 14–17, 22, 23, 70, 72,
73, 79, 82, 129, 131, 180, 185, 206, 219, 222,
224, 309, 310, 317, 318, 322, 386, 412, 444
R
Radio Immunoprecipitation Assay (RIPA).................322,
328, 330, 413
Ransac model ...............................................428–430, 435
RCR, see Respiratory control ratio (RCR)
Reactive Oxygen Species (ROS) ...............................2, 20,
21, 23, 77, 130, 154, 205, 207, 208, 210, 211,
216, 227, 249, 250, 259, 271, 277, 281, 288,
306, 308, 319, 410, 411
MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index 457
Real-time PCR ............................................ 90, 91, 94, 95
Recombinant vectors .................................................... 443
Redox balance ............................................................... 130
Redox cycling ................................................................288
Redox-sensitive green fluorescent protein
(roGFP2) ........................114, 120, 125, 288, 300
Renilla reniformis green fluorescent protein (RGFP)
442–444, 447–449, 451
Residual oxygen consumption ....................................... 77
Respirasomes .......................................................... 11, 103
Respirations ..........................................3, 7, 9–23, 33, 59,
63, 64, 69, 72, 73, 75, 77, 81, 83, 88, 129, 130,
137, 155, 186, 189, 195–197, 201, 204, 218,
219, 221–223, 229, 232, 274, 275, 277–280,
282, 310, 312, 314, 316–320
Respiration rates................................ 7, 11, 13–19, 22, 23
Respiratory chain electron flow........................... 129–140
Respiratory control ratio (RCR) ............ 14, 16, 232, 260
Respiratory Super-Complexes .......................................... 6
RGFP, see Renilla reniformis green fluorescent protein
(RGFP)
Rhod-2.................................................................. 174, 181
Rhodamine 123...................................195, 197–199, 204
roGFP2, see Redox-sensitive green fluorescent protein
(roGFP2)
ROS production................................................ 2, 20, 173,
206, 216, 271–282, 286, 325
ROS, see Reactive Oxygen Species (ROS)
Rotenone ....................................... 14, 15, 17, 20, 22, 70,
74, 155, 157, 167, 176, 219, 222, 229, 413, 420
Routine respiration ............................................ 17, 18, 20
RTG genes....................................................................... 88
RTg2p........................................................................88, 89
Rtg3 phosphorylation detection ..............................92, 93
S
Saccharomyces cerevisiae.............................................88, 91
SARS-CoV-2 ..................................................................... 2
SDS-PAGE, see Sodium dodecylsulfate polyacrylamide gel
electrophoresis (SDS-PAGE)
Seahorse®XF ................................................................... 11
Shear stress .......................................................... 345, 346,
348, 349, 351, 352
Signaling pathways............................................. 2, 67, 115
Single par ticle-tracking ........................................ 153–162
Single-mitoplast ..................................237, 238, 243, 245
SIRT1 ............................................................................1 30
SIRTs, see Sirtuins
Sirtuins (SIRTs)............................................................. 166
Skin fibroblasts ..........................................................5, 236
Sodium dodecylsulfate polyacrylamide gel electrophoresis
(SDS-PAGE)........................................95, 97, 254,
384, 390, 394, 413, 421
State 3u......................................................................13, 14
Streptavidin....................................................................1 45
Stress responses ....................................... 2, 19, 24, 87, 89
Substrate level phosphorylation (SLP) ............... 305, 319
Succinate dehydrogenase assay....................................... 42
Succinate-energized muscle mitochondria .................. 274
Succinate ubiquinone reductase....................................... 3
Superoxide ............................................................ 216, 224
Surface marker staining....................................... 204, 205,
207–209, 211
Synthasomes ..................................................................1 03
T
Tandem mass spectrometry .......................................... 384
TCA cycle .............................................2, 3, 7, 13, 16, 18,
20, 22, 23, 88, 129, 130, 132, 134, 194, 197
Tetramethylrhodamine ............................... 156, 204, 218
Tetrazolium redox dye................................ 131, 133, 134
Thermodynamics..............................................4 , 432–433
Thermostable DNA polymerase................................... 146
Thorax dissection ............................................................ 71
3d imaging............................................................ 113–127
3d optical cryo-imaging.............................. 259, 265, 267
Thrombosis ....................................................................... 2
Thymidine kinase-2.............................................. 144, 250
Thymidine phosphor ylase deficiency ........................... 144
TIM23 .................................................................. 337, 442
TIRF, see Total internal reflection fluorescence
microscopy imaging (TIRF)
Tissue auto-fluorescence imaging ................................ 259
Tissue macrophages ...................................................... 205
TMRM.................................................. 21, 154, 156–159,
161, 162, 450
TOM .............................................................................. 442
TOM20 .........................................................................4 42
TOM22 .........................................................................4 42
Total internal reflection fluorescence microscopy imaging
(TIRF)...................................... 334, 335, 338–340
TPP+ ........................................................ 21, 22, 410, 411
TPP, see Triphenylphosphine (TPP+)
TrackMate............................................154, 157, 159, 162
Transcription of mitochondrial genes................. 418, 420
Transfection reagents.................................. 178, 181, 187
Triphenylphosphine (TPP+)................................ 410, 415
Triphenylphosphonium .................................10, 415, 416
2D
13
C/
1
H HSQC NMR.......................... 272, 279, 280
2d electrophoresis ........................................104, 383–386
458 MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index
2-deoxyglucose (2DOG)........................... 271, 272, 274,
275, 277–281
2DOG, see 2-deoxyglucose (2DOG)
2DOG ATP energy clamp ..................271, 272, 274, 275
tRNAs .............................................................................. 32
Tyrosine hydroxylase (TH) ................................ 115, 116,
121, 123, 124
U
Ubiquinol ............................................................. 426, 431
Ubiquinol cytochrome c reductase .............................. 3, 5
Ubiquinone ........................................................... 3–5, 13,
386, 420, 427, 435
Ultima Gold
™
liquid scintillation cocktail.................... 146
Untargeted metabolomics ................................... 357–379
Untargeted metabolomics analysis...................... 357–397
Urine.................................... 22, 358, 361, 363, 369, 377
V
Vanquish™UHPLC system ...............................363, 364
W
WEKA......................................................... 286, 288–291,
297, 299, 301
WEKA segmentation .......................................... 290, 291,
294, 297, 299
Western blotting......................................... 7, 93, 98, 111,
217, 233, 337
Whole-body respiration ................................................2 59
X
XF Extracellular Flux Analyzer...................................9, 11
Y
Yeast .......................................................... 23, 70, 87–100,
114–117, 119, 176, 288, 399
Yeast protein extraction .................................................. 96
YME1L ................................................................. 325, 326
MITOCHONDRIAL MEDICINE:VOLUME 2: ASSESSING MITOCHONDRIA
Index 459
