Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Colson, L.; Kwon, Y.; Nam,
S.; Bhandari, A.; Maya, N.M.; Lu, Y.;
Cho, Y. Trends in Single-Molecule
Total Internal Reflection Fluorescence
Imaging and Their Biological
Applications with Lab-on-a-Chip
Technology. Sensors 2023,23, 7691.
https://doi.org/10.3390/
s23187691
Academic Editor: Sellamuthu Anbu
Received: 21 July 2023
Revised: 1 September 2023
Accepted: 3 September 2023
Published: 6 September 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
sensors
Review
Trends in Single-Molecule Total Internal Reflection
Fluorescence Imaging and Their Biological Applications with
Lab-on-a-Chip Technology
Louis Colson 1, Youngeun Kwon 2, Soobin Nam 2, Avinashi Bhandari 1, Nolberto Martinez Maya 1,
Ying Lu 1and Yongmin Cho 2 ,*
1Department of Systems Biology, Harvard Medical School, Boston, MA 02115, USA;
louiscolson@g.harvard.edu (L.C.); avinashi_bhandari@hms.harvard.edu (A.B.);
nolberto_martinezmaya@hms.harvard.edu (N.M.M.); ying_lu@hms.harvard.edu (Y.L.)
2Department of Chemical Engineering, Myongji University, Yongin 17058, Republic of Korea;
kye3526@mju.ac.kr (Y.K.); soobin019@mju.ac.kr (S.N.)
*Correspondence: yongmincho@mju.ac.kr; Tel.: +82-31-330-6385
Abstract:
Single-molecule imaging technologies, especially those based on fluorescence, have been de-
veloped to probe both the equilibrium and dynamic properties of biomolecules at the single-molecular
and quantitative levels. In this review, we provide an overview of the state-of-the-art advancements
in single-molecule fluorescence imaging techniques. We systematically explore the advanced imple-
mentations of
in vitro
single-molecule imaging techniques using total internal reflection fluorescence
(TIRF) microscopy, which is widely accessible. This includes discussions on sample preparation,
passivation techniques, data collection and analysis, and biological applications. Furthermore, we
delve into the compatibility of microfluidic technology for single-molecule fluorescence imaging,
highlighting its potential benefits and challenges. Finally, we summarize the current challenges and
prospects of fluorescence-based single-molecule imaging techniques, paving the way for further
advancements in this rapidly evolving field.
Keywords: single-molecule imaging; TIRF; fluorescence; data analysis; microfluidics
1. Introduction
Imaging techniques provide powerful tools to visualize and quantify molecular interac-
tions, cellular dynamics, and tissue architecture and are therefore instrumental in advancing
our understanding of biological systems [
1
–
12
]. Certain imaging techniques can directly
observe individual biomolecules such as oligonucleotides, proteins, and protein complexes.
These single-molecule imaging techniques can provide information on the heterogeneity of the
system which can often be difficult to determine using other methods. In recent years, single-
molecule imaging with total internal reflection fluorescence (TIRF) has gained significant
popularity due to its accessibility and high sensitivity in probing the properties of biomolecules.
By enabling the visualization and tracking of individual molecules in exceptional spatial and
temporal resolutions, TIRF-based single-molecule imaging has opened up new avenues for
studying complex biological processes, including protein folding, protein–protein interactions,
DNA replication, and cellular signaling [13–22].
In this review, we specifically explore
in vitro
single-molecule imaging with
TIRF [
11
,
23
–
25
]. In addition to discussing the technical aspects of single-molecule imaging,
this review surveys and highlights several exemplary applications of TIRF-based single-
molecule imaging, especially microfluidic-based approaches. By showcasing the diversity of
biological questions addressed using this technique, we aim to demonstrate its broad impact
across various fields, including molecularbiology, biophysics, and nanotechnology. Finally, we
address the potential prospects and challenges of fluorescence-based single-molecule imaging
Sensors 2023,23, 7691. https://doi.org/10.3390/s23187691 https://www.mdpi.com/journal/sensors
Sensors 2023,23, 7691 2 of 20
techniques. We also discuss the limitations and potential sources of artifacts in single-molecule
imaging experiments, as well as strategies to mitigate these issues.
2. Optical Systems for Single-Molecule Fluorescence Imaging
Fluorescence single-molecule imaging techniques rely on the utilization of optical radiation
to probe individual molecules within a liquid or solid sample. To achieve successful single-
molecule imaging, two key requirements must be met: (1) ensuring that resonant molecules are
spatially resolved by the detector, and (2) providing a sufficient signal-to-noise ratio (SNR) for
the single-molecule signal within a reasonable averaging time [
22
]. Consequently, a fundamental
prerequisite for conducting single-molecule observations is to dilute the concentration of the
target molecule of interest to exceedingly low levels (typically < 100 nM). The detection of single
molecules via fluorescence-based methods demands careful optimization of the signal-to-noise
ratio. Maximizing the signal requires the selection of an impurity molecule with the highest
possible fluorescence quantum efficiency.
This approach harnesses recent advancements in fluorescence imaging techniques,
including TIRF microscopy [
23
–
25
], super-resolution microscopy, and single-molecule lo-
calization microscopy [
26
–
32
]. TIRF microscopy, one of the most commonly employed
tools in single-molecule fluorescence microscopy, capitalizes on the principle of total inter-
nal reflection. This phenomenon occurs when a laser beam strikes the interface between
a medium with a higher refractive index (typically glass) and a medium with a lower
refractive index (such as a sample solution) at an angle greater than the critical angle
(Figure 1A). As a result, an evanescent wave is generated, which excites fluorophores in
the immediate vicinity of the interface, facilitating the visualization of single molecules
near the sample surface. TIRF microscopy is practically implemented by using either a
quartz prism or the microscope objective to generate the evanescent field and illuminate
surface-immobilized molecules (Figure 1B). TIRF microscopy offers exceptional optical
sectioning and background suppression, leading to a high signal-to-noise ratio.
Sensors 2023, 23, x FOR PEER REVIEW 2 of 19
impact across various fields, including molecular biology, biophysics, and nanotechnol-
ogy. Finally, we address the potential prospects and challenges of fluorescence-based sin-
gle-molecule imaging techniques. We also discuss the limitations and potential sources of
artifacts in single-molecule imaging experiments, as well as strategies to mitigate these
issues.
2. Optical Systems for Single-Molecule Fluorescence Imaging
Fluorescence single-molecule imaging techniques rely on the utilization of optical ra-
diation to probe individual molecules within a liquid or solid sample. To achieve success-
ful single-molecule imaging, two key requirements must be met: (1) ensuring that reso-
nant molecules are spatially resolved by the detector, and (2) providing a sufficient signal-
to-noise ratio (SNR) for the single-molecule signal within a reasonable averaging time [22].
Consequently, a fundamental prerequisite for conducting single-molecule observations is
to dilute the concentration of the target molecule of interest to exceedingly low levels (typ-
ically < 100 nM). The detection of single molecules via fluorescence-based methods de-
mands careful optimization of the signal-to-noise ratio. Maximizing the signal requires
the selection of an impurity molecule with the highest possible fluorescence quantum ef-
ficiency.
This approach harnesses recent advancements in fluorescence imaging techniques,
including TIRF microscopy [23–25], super-resolution microscopy, and single-molecule lo-
calization microscopy [26–32]. TIRF microscopy, one of the most commonly employed
tools in single-molecule fluorescence microscopy, capitalizes on the principle of total in-
ternal reflection. This phenomenon occurs when a laser beam strikes the interface between
a medium with a higher refractive index (typically glass) and a medium with a lower re-
fractive index (such as a sample solution) at an angle greater than the critical angle (Figure
1A). As a result, an evanescent wave is generated, which excites fluorophores in the im-
mediate vicinity of the interface, facilitating the visualization of single molecules near the
sample surface. TIRF microscopy is practically implemented by using either a quar
prism or the microscope objective to generate the evanescent field and illuminate surface-
immobilized molecules (Figure 1B). TIRF microscopy offers exceptional optical sectioning
and background suppression, leading to a high signal-to-noise ratio.
Figure 1. TIRF microscopy for single-molecule fluorescence imaging. (A) Principle of TIRF micros-
copy. (B) Types of TIRF microscopy: prism-type (P-TIRF) or objective-type (O-TIRF).
The evanescent field intensity, I(z), at a perpendicular distance z from the interface is
described by Equation (1).
𝐼𝑧= 𝐼
0exp
), (1)
𝑑= 𝜆
4𝜋𝑛
𝑠𝑖𝑛𝜃𝑛
/, (2)
where 𝐼0 represents the evanescent field intensity at the interface (Figure 1A). The char-
acteristic penetration depth (Equation (2)), d, is determined by the wavelength of incident
Figure 1.
TIRF microscopy for single-molecule fluorescence imaging. (
A
) Principle of TIRF mi-
croscopy. (B) Types of TIRF microscopy: prism-type (P-TIRF) or objective-type (O-TIRF).
The evanescent field intensity, I(z), at a perpendicular distance z from the interface is
described by Equation (1).
I(z)=I(0)exp−z
d, (1)
d=λ0
4πn2
1sin2θ−n2
2)−1/2, (2)
where
I(0)
represents the evanescent field intensity at the interface (Figure 1A). The charac-
teristic penetration depth (Equation (2)), d, is determined by the wavelength of incident
light (
λ0)
, refractive index of the medium through which the light initially passes (
n1
) and
in contact with the sample (
n2
), and incident angle
(θ)
. Typically ranging between 30 and
200 nm, the penetration depth defines the region within which fluorophores are effectively
excited by the evanescent wave.
Sensors 2023,23, 7691 3 of 20
In addition, the detection of individual fluorophores is a critical aspect of single-
molecule fluorescence imaging. Here, the numerical aperture (NA) is one of the key
parameters. High NA objectives are commonly used to maximize light collection and
detection efficiency. The specific NA value depends on the imaging setup and the desired
resolution and sensitivity. For conventional single-molecule fluorescence imaging, objec-
tives with NA values ranging from 1.2 to 1.49 are frequently employed. These objectives
offer a balance between high light collection efficiency and reasonable working distances.
They are suitable for imaging samples in various configurations, including liquid solutions,
solid surfaces, and biological specimens.
The choice of camera is another important factor in the detection of individual flu-
orophores. Ultimately, digital cameras capture the photons from individual fluorescent
molecules and convert the light into electrical signals. The cameras used for single-molecule
TIRF imaging tend to have quantum efficiencies above 80%, spectral range between
300 and 1100 nm, low readout noise, and millisecond readout speeds [
33
]. Electron multi-
plying charge coupled devices (EMCCDs) and the scientific complementary metal–oxide–
semiconductor (sCMOS) devices are the most common types of cameras used in single-
molecule imaging. Note that most of the biological applications discussed in Section 5use
an EMCCD camera. Some recent laboratory advances in imaging technology may further
improve the performance of scientific cameras across the broadband spectrum [34–36].
3. Sample Preparation
Sample preparation is a critical step in single-molecule imaging of biological molecules
for repeatable and reliable results. In this section, we define “sample preparation” as the
preparation of the imaging device (Figure 2) and any labeling of the biological molecules.
Generally, biomolecules non-specifically adhere to the surfaces of the imaging device. Thus,
the preparation of the imaging device includes a passivation step to reduce non-specific
binding and false-positive signals. Fluorophores provide a readout for interactions between
molecules or the functions of the reaction system. Preparing biological molecules for the
experiments includes a labeling step to conjugate fluorophores to molecules of interest and
a strategy to constrain the location of the molecules of interest in the imaging region [
37
,
38
].
There are methods that detect freely diffusing single molecules [
39
,
40
], but in this review,
we limit our scope to single-molecule strategies with immobilized molecules.
Sensors 2023, 23, x FOR PEER REVIEW 3 of 19
light (𝜆, refractive index of the medium through which the light initially passes (𝑛) and
in contact with the sample (𝑛), and incident angle 𝜃. Typically ranging between 30 and
200 nm, the penetration depth defines the region within which fluorophores are effectively
excited by the evanescent wave.
In addition, the detection of individual fluorophores is a critical aspect of single-mol-
ecule fluorescence imaging. Here, the numerical aperture (NA) is one of the key parame-
ters. High NA objectives are commonly used to maximize light collection and detection
efficiency. The specific NA value depends on the imaging setup and the desired resolution
and sensitivity. For conventional single-molecule fluorescence imaging, objectives with
NA values ranging from 1.2 to 1.49 are frequently employed. These objectives offer a bal-
ance between high light collection efficiency and reasonable working distances. They are
suitable for imaging samples in various configurations, including liquid solutions, solid
surfaces, and biological specimens.
The choice of camera is another important factor in the detection of individual fluor-
ophores. Ultimately, digital cameras capture the photons from individual fluorescent mol-
ecules and convert the light into electrical signals. The cameras used for single-molecule
TIRF imaging tend to have quantum efficiencies above 80%, spectral range between 300
and 1100 nm, low readout noise, and millisecond readout speeds [33]. Electron multiply-
ing charge coupled devices (EMCCDs) and the scientific complementary metal–oxide–
semiconductor (sCMOS) devices are the most common types of cameras used in single-
molecule imaging. Note that most of the biological applications discussed in Section 5 use
an EMCCD camera. Some recent laboratory advances in imaging technology may further
improve the performance of scientific cameras across the broadband spectrum [34–36].
3. Sample Preparation
Sample preparation is a critical step in single-molecule imaging of biological mole-
cules for repeatable and reliable results. In this section, we define “sample preparation”
as the preparation of the imaging device (Figure 2) and any labeling of the biological mol-
ecules. Generally, biomolecules non-specifically adhere to the surfaces of the imaging de-
vice. Thus, the preparation of the imaging device includes a passivation step to reduce
non-specific binding and false-positive signals. Fluorophores provide a readout for inter-
actions between molecules or the functions of the reaction system. Preparing biological
molecules for the experiments includes a labeling step to conjugate fluorophores to mole-
cules of interest and a strategy to constrain the location of the molecules of interest in the
imaging region [37,38]. There are methods that detect freely diffusing single molecules
[39,40], but in this review, we limit our scope to single-molecule strategies with immobi-
lized molecules.
Figure 2. Schematic representation of an imaging device. The construction of an imaging device
entails the integration of a microscope slide and a coverslip, employing double-sided tape for pre-
cision juxtaposition, followed by hermetic sealing with epoxy resin. The holes on the slide are used
as the inlet and outlet for solution exchange.
3.1. Surface Passivation
Established passivation techniques to prepare the imaging device rely on coating
chemically treated glass surfaces with biocompatible reagents such as polyethylene glycol
Figure 2.
Schematic representation of an imaging device. The construction of an imaging device
entails the integration of a microscope slide and a coverslip, employing double-sided tape for
precision juxtaposition, followed by hermetic sealing with epoxy resin. The holes on the slide are
used as the inlet and outlet for solution exchange.
3.1. Surface Passivation
Established passivation techniques to prepare the imaging device rely on coating
chemically treated glass surfaces with biocompatible reagents such as polyethylene glycol
(PEG), phospholipids, or Tween-20. In this section, we briefly describe the PEG, lipids, and
Tween-20 passivation methods. These methods produce similar results and are described in
depth elsewhere [
41
–
43
]. The PEG passivation method relies on the amino-silanization of
the glass surface. Usually, researchers use KOH to form alcohol groups on the glass surface.
Amino silane can react with these alcohol groups on the surface. This reaction (shown in
Figure 3A) takes the following form:
ROH +(OCH3)3SiOC H3→CH3OH.
To complete
Sensors 2023,23, 7691 4 of 20
the passivation, commercially available PEG ester molecules react with the silane groups
(Figure 3B) to PEGylate the surface of the coverslip [
41
]. This covalent passivation method
can withstand harsh protein denaturing conditions such as 8 M guanidinium chloride
(GdmCl) [
44
] and 4M urea [
45
]. For the Tween-20 passivation, a dichlorodimethylsilane
(DDS)-treated glass surface forms a hydrophobic coating that can be passivated with the
addition of the biocompatible surfactant Tween-20 [
42
]. For the passivation with phospho-
lipids, glass devices can be incubated with liposomes to form a fluid lipid bilayer. The
liposomes used for passivation in single-molecule TIRF experiments have been made from
lipids such as 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and egg phosphatidyl-
choline [46,47].
Sensors 2023, 23, x FOR PEER REVIEW 4 of 19
(PEG), phospholipids, or Tween-20. In this section, we briefly describe the PEG, lipids,
and Tween-20 passivation methods. These methods produce similar results and are de-
scribed in depth elsewhere [41–43]. The PEG passivation method relies on the amino-si-
lanization of the glass surface. Usually, researchers use KOH to form alcohol groups on
the glass surface. Amino silane can react with these alcohol groups on the surface. This
reaction (shown in Figure 3A) takes the following form: 𝑅𝑂𝐻 + 𝑂𝐶𝐻𝑆𝑖𝑂𝐶𝐻→
𝐶𝐻𝑂𝐻. To complete the passivation, commercially available PEG ester molecules react
with the silane groups (Figure 3B) to PEGylate the surface of the coverslip [41]. This cova-
lent passivation method can withstand harsh protein denaturing conditions such as 8 M
guanidinium chloride (GdmCl) [44] and 4M urea [45]. For the Tween-20 passivation, a
dichlorodimethylsilane (DDS)-treated glass surface forms a hydrophobic coating that can
be passivated with the addition of the biocompatible surfactant Tween-20 [42]. For the
passivation with phospholipids, glass devices can be incubated with liposomes to form a
fluid lipid bilayer. The liposomes used for passivation in single-molecule TIRF experi-
ments have been made from lipids such as 1,2-dioleoyl-sn-glycero-3-phosphocholine
(DOPC) and egg phosphatidylcholine [46,47].
Figure 3. Chemical diagrams of surface preparation steps. (A) Diagram of amino silanization of
glass surface. (B) Diagram of PEG-ylation step with heterobifunctional biotin–PEG–succinimidyl
carbonate (SC) ester. (C) Diagram of the interactions between avidin and biotin.
3.2. Surface Functionalization
Typically, surface-immobilized avidin, streptavidin, or neutravidin (from here on av-
idin will be used interchangeably with any of these forms of avidin) tether biotinylated
biomolecules to the imaging region. To immobilize avidin on the surface, biotins are in-
troduced onto the surface before or during the passivation step. For the PEG passivation
method, a percentage of the PEG molecules that passivate the surface contain a biotin
moiety on the opposite end from the ester group [43]. The ester group in the biotinylated
PEG molecule reacts with the amine group on the surface as shown in Figure 3B. For the
Tween-20 passivation method, researchers introduce biotinylated BSA to the imaging de-
vice prior to passivation. The biotinylated BSA adheres to the hydrophobic surface before
Tween-20 passivates the surface [42]. For the lipid passivation method, either a fraction of
the lipids will be biotinylated [46] or avidin will be directly applied to the imaging surface
before passivation [47]. Avidin binds to biotin molecules with high affinity (K
D
~10
−15
M)
[48]. This affinity comes from a number of hydrogen bonds formed between the amino
acids of avidin’s biotin-binding site and the biotin molecule as diagramed in Figure 3C.
Since avidin forms a tetramer, avidin molecules bound to the biotins on the imaging sur-
face still contain available biotin binding sites to immobilize biotinylated molecules.
3.3. Protein Biotinylation
One strategy to directly biotinylate proteins of interest involves introducing an
AviTag [49–51]. The BirA ligase recognizes the 15 amino acid AviTag and conjugates biotin
to the tag’s only lysine residue [52,53]. BirA biotinylation occurs through a two-step
Figure 3.
Chemical diagrams of surface preparation steps. (
A
) Diagram of amino silanization of
glass surface. (
B
) Diagram of PEG-ylation step with heterobifunctional biotin–PEG–succinimidyl
carbonate (SC) ester. (C) Diagram of the interactions between avidin and biotin.
3.2. Surface Functionalization
Typically, surface-immobilized avidin, streptavidin, or neutravidin (from here on avidin
will be used interchangeably with any of these forms of avidin) tether biotinylated biomolecules
to the imaging region. To immobilize avidin on the surface, biotins are introduced onto the
surface before or during the passivation step. For the PEG passivation method, a percentage
of the PEG molecules that passivate the surface contain a biotin moiety on the opposite end
from the ester group [
43
]. The ester group in the biotinylated PEG molecule reacts with the
amine group on the surface as shown in Figure 3B. For the Tween-20 passivation method,
researchers introduce biotinylated BSA to the imaging device prior to passivation. The bi-
otinylated BSA adheres to the hydrophobic surface before Tween-20 passivates the surface [
42
].
For the lipid passivation method, either a fraction of the lipids will be biotinylated [
46
] or
avidin will be directly applied to the imaging surface before passivation [
47
]. Avidin binds to
biotin molecules with high affinity (K
D
~10
−15
M) [
48
]. This affinity comes from a number of
hydrogen bonds formed between the amino acids of avidin’s biotin-binding site and the biotin
molecule as diagramed in Figure 3C. Since avidin forms a tetramer, avidin molecules bound
to the biotins on the imaging surface still contain available biotin binding sites to immobilize
biotinylated molecules.
3.3. Protein Biotinylation
One strategy to directly biotinylate proteins of interest involves introducing an
AviTag [
49
–
51
]. The BirA ligase recognizes the 15 amino acid AviTag and conjugates
biotin to the tag’s only lysine residue [
52
,
53
]. BirA biotinylation occurs through a two-step
reaction where biotin first reacts with ATP before the amine group of AviTag’s lysine
residue attacks the ester in biotin (Figure 4C). Overexpressing BirA ligase can biotinylate
the protein
in vivo
[
53
] or purified BirA can biotinylate the protein in an
in vitro
reaction
system [53,54]. Other direct coupling methods include using biotinylated peptides [55] or
ligating a biotinylated peptide to the C-terminus of a protein of interest [
14
]. Biotinylated
peptides or molecules with a high affinity to fusion tags can also be used to introduce
Sensors 2023,23, 7691 5 of 20
biotin to the proteins of interest [
56
]. Biotinylated antibodies [
57
,
58
] or biotinylated nu-
cleotide oligos [
59
–
61
] can indirectly couple proteins to the surface. Tethering the sample
to the imaging surface does not require high efficiency. The vast majority of unbiotiny-
lated molecules will not stick to the passivated surface and at saturating conditions, most
biotin-binding sites will bind to biotinylated biomolecules.
Sensors 2023, 23, x FOR PEER REVIEW 5 of 19
reaction where biotin first reacts with ATP before the amine group of AviTag’s lysine res-
idue aacks the ester in biotin (Figure 4C). Overexpressing BirA ligase can biotinylate the
protein in vivo [53] or purified BirA can biotinylate the protein in an in vitro reaction sys-
tem [53,54]. Other direct coupling methods include using biotinylated peptides [55] or
ligating a biotinylated peptide to the C-terminus of a protein of interest [14]. Biotinylated
peptides or molecules with a high affinity to fusion tags can also be used to introduce
biotin to the proteins of interest [56]. Biotinylated antibodies [57,58] or biotinylated nucle-
otide oligos [59–61] can indirectly couple proteins to the surface. Tethering the sample to
the imaging surface does not require high efficiency. The vast majority of unbiotinylated
molecules will not stick to the passivated surface and at saturating conditions, most biotin-
binding sites will bind to biotinylated biomolecules.
Figure 4. Chemical diagrams of common protein labeling strategies. (A) Thiol–maleimide reaction
represents a common way to label cysteine residues in proteins. (B) NHS ester reaction with lysine
provides a way to label proteins. (C) The process of AviTag biotinylation as a way to tether proteins
to immobilized avidin.
3.4. Protein Fluorescence Labeling
The requirements for fluorescent labeling are more stringent. One of the most com-
mon strategies conjugates fluorescent molecules with maleimide groups onto the thiol
groups on the cysteines of proteins (Figure 4A). This strategy works best for small proteins
without essential cysteine residues. Similarly, lysine can react with NHS ester groups on
fluorophores (Figure 4B). Recent studies also use unnatural amino acids to perform click
chemistry with fluorophores [54,62]. This method works well for large proteins or protein
complexes and for proteins that harbor essential cysteine residues. Additionally, the de-
velopment of high-affinity protein fusion tags allows N- or C-terminal fusion tags to pro-
vide a specific and simple way to bind a fluorophore to the target of interest [63]. Fluores-
cently labeled antibodies offer a potential alternative to these aforementioned methods.
For all of these, the choice of fluorophores and the labeling strategy will depend on the
Figure 4.
Chemical diagrams of common protein labeling strategies. (
A
) Thiol–maleimide reaction
represents a common way to label cysteine residues in proteins. (
B
) NHS ester reaction with lysine
provides a way to label proteins. (
C
) The process of AviTag biotinylation as a way to tether proteins
to immobilized avidin.
3.4. Protein Fluorescence Labeling
The requirements for fluorescent labeling are more stringent. One of the most com-
mon strategies conjugates fluorescent molecules with maleimide groups onto the thiol
groups on the cysteines of proteins (Figure 4A). This strategy works best for small proteins
without essential cysteine residues. Similarly, lysine can react with NHS ester groups on
fluorophores (Figure 4B). Recent studies also use unnatural amino acids to perform click
chemistry with fluorophores [
54
,
62
]. This method works well for large proteins or protein
complexes and for proteins that harbor essential cysteine residues. Additionally, the devel-
opment of high-affinity protein fusion tags allows N- or C-terminal fusion tags to provide a
specific and simple way to bind a fluorophore to the target of interest [
63
]. Fluorescently
labeled antibodies offer a potential alternative to these aforementioned methods. For all of
these, the choice of fluorophores and the labeling strategy will depend on the biological
application. Roy et al. [
33
] provide a practical overview of the TIRF-based single-molecule
experiments with Förster Resonance Energy Transfer (FRET).
Sensors 2023,23, 7691 6 of 20
4. Analysis Methods
Single-molecule data often exhibit inherent noise stemming from both the system
under study and the measurement instrument. This noise can manifest in various forms,
including sample stage drift [
64
,
65
], Gaussian fluctuations [
66
,
67
], non-Gaussian varia-
tions [
68
–
70
], diffusive behavior [
71
,
72
], and even undefined sources [
73
,
74
]. Complications
arise particularly when the nature of the underlying fluctuation is unknown, as it can po-
tentially follow either a Gaussian or non-Gaussian distribution. Consequently, extracting
meaningful information from single-molecule data poses significant challenges. In this
section, we will introduce key steps in the analysis with key examples (Figure 5).
Sensors 2023, 23, x FOR PEER REVIEW 6 of 19
biological application. Roy et al. [33] provide a practical overview of the TIRF-based sin-
gle-molecule experiments with Förster Resonance Energy Transfer (FRET).
4. Analysis Methods
Single-molecule data often exhibit inherent noise stemming from both the system
under study and the measurement instrument. This noise can manifest in various forms,
including sample stage drift [64,65], Gaussian fluctuations [66,67], non-Gaussian varia-
tions [68–70], diffusive behavior [71,72], and even undefined sources [73,74]. Complica-
tions arise particularly when the nature of the underlying fluctuation is unknown, as it
can potentially follow either a Gaussian or non-Gaussian distribution. Consequently, ex-
tracting meaningful information from single-molecule data poses significant challenges.
In this section, we will introduce key steps in the analysis with key examples (Figure 5).
Figure 5. Workflow of single-molecule imaging data analysis.
4.1. Point Spread Function Fiing
To achieve high spatial resolution, it is essential to precisely localize the position of
each detected fluorophore. This localization is typically performed using a technique
called point spread function (PSF) fiing, where the observed intensity distribution of a
fluorophore is fit to a mathematical model of the PSF. By accurately determining the center
of the PSF, the position of the fluorophore can be determined with sub-pixel precision,
enabling precise localization of single molecules. Sage et al. [29] comprehensively evalu-
ated software packages for single-molecule localization microscopy (SMLM). Many mod-
ules from these software packages would be usable for TIRF-based single-molecule fluo-
rescence imaging datasets.
4.2. Extracting Information from Signal
Among the most prevalent types of single-molecule imaging data are time series sig-
nals characterized by values ranging from zero to an upper limit. To extract the desired
information, several techniques have been developed to fit the noisy time series data to an
idealized model involving discrete steps and dwell times [73]. One widely employed
method is hidden Markov modeling (HMM) [75–79]. HMM enables the identification of
hidden (unobservable) states within a Markovian process, where the present and future
states depend solely on the current state, independent of the system’s prior states. The
idealized model from HMM is a reliable way to extract the FRET states, dwell times, and
rate constants from single-molecule time series data.
4.3. Deep Learning
Deep learning has emerged as a powerful tool for analyzing single-molecule fluores-
cence imaging data, particularly for handling large volumes of complex and noisy data
[80,81]. Specifically, deep learning algorithms, such as convolutional neural networks
(CNNs) and recurrent neural networks (RNNs), excel in recognizing paerns and features
in images that may be challenging for manual identification. CNNs are well-suited for
detecting and localizing individual molecules, while RNNs can analyze the temporal dy-
namics of fluorescence signals. Deep learning can also extract more intricate information
from single-molecule data, including classifying molecular states based on fluorescence
properties or predicting molecular interactions. Nonetheless, challenges exist, such as the
Figure 5. Workflow of single-molecule imaging data analysis.
4.1. Point Spread Function Fitting
To achieve high spatial resolution, it is essential to precisely localize the position
of each detected fluorophore. This localization is typically performed using a technique
called point spread function (PSF) fitting, where the observed intensity distribution of a
fluorophore is fit to a mathematical model of the PSF. By accurately determining the center
of the PSF, the position of the fluorophore can be determined with sub-pixel precision,
enabling precise localization of single molecules. Sage et al. [
29
] comprehensively evaluated
software packages for single-molecule localization microscopy (SMLM). Many modules
from these software packages would be usable for TIRF-based single-molecule fluorescence
imaging datasets.
4.2. Extracting Information from Signal
Among the most prevalent types of single-molecule imaging data are time series
signals characterized by values ranging from zero to an upper limit. To extract the desired
information, several techniques have been developed to fit the noisy time series data to
an idealized model involving discrete steps and dwell times [
73
]. One widely employed
method is hidden Markov modeling (HMM) [
75
–
79
]. HMM enables the identification of
hidden (unobservable) states within a Markovian process, where the present and future
states depend solely on the current state, independent of the system’s prior states. The
idealized model from HMM is a reliable way to extract the FRET states, dwell times, and
rate constants from single-molecule time series data.
4.3. Deep Learning
Deep learning has emerged as a powerful tool for analyzing single-molecule fluorescence
imaging data, particularly for handling large volumes of complex and noisy data [
80
,
81
].
Specifically, deep learning algorithms, such as convolutional neural networks (CNNs) and re-
current neural networks (RNNs), excel in recognizing patterns and features in images that may
be challenging for manual identification. CNNs are well-suited for detecting and localizing
individual molecules, while RNNs can analyze the temporal dynamics of fluorescence signals.
Deep learning can also extract more intricate information from single-molecule data, including
classifying molecular states based on fluorescence properties or predicting molecular interac-
tions. Nonetheless, challenges exist, such as the need for large, annotated datasets, which can
be time-consuming and costly to generate, as well as the risk of overfitting or underfitting
models, potentially leading to inaccurate or unreliable results. Liu et al. summarized the deep
learning application in single-molecule analysis [80].
Sensors 2023,23, 7691 7 of 20
5. Biological Application
Single-molecule imaging of
in vitro
systems proves to be a powerful tool for research-
ing conformational dynamics, protein folding, protein modifications, and protein interac-
tions. Results from single-molecule imaging investigations can lead to important insights
into biological processes such as transcription [
60
,
61
], protein synthesis [
82
,
83
], and protein
degradation [
13
,
54
,
63
]. Single-molecule methods offer unique insights into the heterogene-
ity of the sample. Observing the fluorescence of biomolecules
in vitro
further offers the
benefit of time scales from milliseconds to minutes, location of the biomolecules within
the system, tight control over the components in the system, and potential readouts for
conformational states. In this section, we describe some of the recent applications of
single-molecule TIRF imaging.
5.1. Conformation Dynamics
Elucidating protein conformational dynamics often reveals important mechanistic
insights into how proteins function. To observe conformational dynamics in real-time,
researchers can combine single-molecule TIRF with fluorescent resonance energy transfer
(FRET) [
33
]. FRET experiments estimate the efficiency of energy transfer, E, as described by
Equation (3).
E= 1+r
R06!−1
, (3)
In Equation (3), r represents the distance between one donor fluorophore and one
acceptor, and
R0
represents the Förster radius for a specific donor–acceptor pair at which
E=
0.5 [
33
]. To illustrate typical values for
R0
, the Förster radii for Cy3-Cy7, Cy3-Cy5, and
Cy5-Cy7 are 3.8 nm, 5.4 nm, and 6.2 nm, respectively [
84
]. The apparent FRET efficiency
and corrected FRET efficiency are calculated from the intensity of the fluorophores using
Equations (4) and (5).
Eapp =IA
IA+ID
, (4)
Ecorrected =1+γID
IA−1
, (5)
In Equations (4) and (5),
IA
represents the intensity of the acceptor fluorophore and
ID
represents the intensity of the donor fluorophore. In Equation (5),
γ
represents the correc-
tion factor [
33
]. The observed apparent FRET efficiency can be affected by dye orientation,
dye conjugation, and instrument factors, and thus only provides an approximation for the
distance between donor and acceptor dyes [
33
]. Nevertheless, by conjugating one acceptor
fluorophore and one donor fluorophore at carefully picked locations on a protein of interest,
FRET efficiencies can correspond to distinct conformational states (Figure 6A). In the last
three years, smFRET has been applied to uncover some of the conformational dynamics
of proteins such as CRISPR (clustered regularly interspaced short palindromic repeats)-
Cas9 (CRISPR-associated protein 9) from Streptococcus pyogenes [
51
], mouse metabotropic
glutamate receptor 2 (mGluR2) [62], and the yeast 26S proteasome [54].
The CRISPR-Cas9 RNA-guided endonuclease enzyme performs multiple steps and
conformational changes to cleave DNA [
51
]. In particular, the HNH and RuvC nuclease
domains on CRISPR-Cas9 cleave the target strand (TS) and nontarget strand (NTS) of
DNA, respectively [
51
]. To characterize the post-catalytic conformational changes in the
HNH domain with respect to the TS, Wang et al. developed a Cas9 construct with a
single cysteine residue on the HNH domain (Cas9
LD750
) [
51
]. Because the TS is relatively
stationary, labeling the TS with Cy3 and the HNH domain with LD750 provided a highly
sensitive reporter for the conformational changes in the HNH domain [
51
]. Choosing LD750
rather than Cy5 as the acceptor increases the sensitivity of the FRET efficiency at shorter
distances. Using this smFRET reporter with catalytically dead RuvC and HNH variants,
they found that only variants that could cleave the TS showed fluctuations between FRET
Sensors 2023,23, 7691 8 of 20
efficiencies (E) ~0.38 and ~0.61 [
51
]. Ultimately, this single-molecule fluorescence reporter
provided direct support for the high flexibility of the HNH domain post-DNA cleavage.
Sensors 2023, 23, x FOR PEER REVIEW 9 of 19
agreement with the FRET efficiency from bulk measurements and free-diffusion single-
molecule experiments [46]. They found the unfolding/folding of adenylate kinase involves
at least six states with an average trajectory length of 4.6 s and higher concentrations of
denaturant increase the probability of sequential transitions [46].
Figure 6. Application concepts. (A) A graph of ideal smFRET efficiencies (E) versus distance be-
tween fluorophores (r) for different fluorophore pairs. The Förster radii for Cy3-Cy7, Cy3-Cy5, Cy5-
Cy7, and Alexa488-Alexa555 are 3.8 nm, 5.4 nm, 6.2 nm, and 7 nm, respectively [84]. The green and
pink colors distinguish between domains labeled with donor and acceptor fluorophores. The small,
semi-transparent circles represent fluorophores that are not radiating light. The small circles with
white highlights represent fluorophores that are emiing light. (B) A diagram displaying the paths
(the lines connecting the circles) between folded, intermediate, and unfolded states (the circles). The
different colors illustrate the distinct smFRET efficiency peaks for each state. (C) A representation
of freely diffusing fluorescent biomolecules interacting with immobilized fluorescent biomolecules.
In the non-FRET case, green and pink colors separate biomolecules with spectrally distinct fluores-
cent signals. In the FRET case, pink biomolecules contain an acceptor fluor and green biomolecules
contain a donor fluor. (D) A diagram of an immobilized protein (pink) and an immobilized protein
with fluorescent post-translational modifications (PTMs) (green).
5.3. Protein Interactions
One of the more intriguing applications of single-molecule fluorescence experiments
is the detection of protein interactions (Figure 6C). This provides information on the dy-
namics, the interaction’s dependence on the system, and the interaction’s effects on the
function of proteins. Bibeau et al. demonstrated that yeast cofilin binds to actin filaments
independent of curvature, but their results suggest actin curvature may facilitate cofilin
dissociation [85]. They also show that cofilin clusters grow asymmetrically with the
growth towards the pointed end of the actin being twice as fast as the growth rate towards
the barbed end [85]. These results offered novel insights into cofilin interactions with actin
and cofilin clusters under different conditions.
Asher et al. [57] observed the dynamics between a model GPCR protein, human V
2
vasopressin receptor, and β-arrestin 1. They show that the β-arrestin 1 C-terminal tail
binds to its own N-terminal positively charged groove to block interaction with the phos-
phorylated C-terminal of the human vasopressin receptor [57]. Immobilized β-arrestin 1
was labeled with LD555p and LD655 to directly observe the distance of β-arrestin 1′s C-
terminal tail to the N-terminal groove. Alone, β-arrestin 1 demonstrates a stable high
Figure 6.
Application concepts. (
A
) A graph of ideal smFRET efficiencies (E) versus distance between
fluorophores (r) for different fluorophore pairs. The Förster radii for Cy3-Cy7, Cy3-Cy5, Cy5-Cy7,
and Alexa488-Alexa555 are 3.8 nm, 5.4 nm, 6.2 nm, and 7 nm, respectively [
84
]. The green and
pink colors distinguish between domains labeled with donor and acceptor fluorophores. The small,
semi-transparent circles represent fluorophores that are not radiating light. The small circles with
white highlights represent fluorophores that are emitting light. (
B
) A diagram displaying the paths
(the lines connecting the circles) between folded, intermediate, and unfolded states (the circles). The
different colors illustrate the distinct smFRET efficiency peaks for each state. (
C
) A representation of
freely diffusing fluorescent biomolecules interacting with immobilized fluorescent biomolecules. In
the non-FRET case, green and pink colors separate biomolecules with spectrally distinct fluorescent
signals. In the FRET case, pink biomolecules contain an acceptor fluor and green biomolecules contain
a donor fluor. (
D
) A diagram of an immobilized protein (pink) and an immobilized protein with
fluorescent post-translational modifications (PTMs) (green).
In addition to elucidating conformational dynamics, smFRET is also used to charac-
terize the propagation of conformational changes. Activation of metabotropic glutamate
receptors (mGluRs) through binding the excitatory neurotransmitter L-glutamate results in
local and global conformational changes that propagate through the ligand-binding Venus
flytrap (VFT) domain, cysteine-rich domain (CRD), and 7-transmembrane (7TM) domain
to reach the intracellular G protein-binding interface [
62
]. To construct a smFRET reporter
on the CRD conformational changes, Liauw et al. incorporated an unnatural amino acid at
amino acid 548 in the CRD of the mouse mGluR2 [
62
]. A copper-catalyzed azide–alkyne
click reaction labeled the proteins with either Cy3 or Cy5 and labeled mGluR2 with a
C-terminal FLAG-tag were immobilized in the imaging region with biotinylated FLAG
antibodies. These smFRET studies uncovered that the CRD domain dynamically transitions
between two intermediate FRET states (E~0.51 and E~0.71) and two FRET states corre-
sponding to the inactive (E~0.31 and predominant population in the absence of an agonist)
and active states (E~0.89) of mGluR2 [
62
]. Further, they labeled a glutamate-binding
defective monomer with an N-terminal SNAP-tag and BG-ATTO488 (SNAP-tag substrate)
fluorophore to determine that heterodimers predominantly reside in the E~0.51 [
62
]. Their
findings provide evidence that shows mGluR activation proceeds through multiple states
including one state where one of the ligand binding domains is inactive.
Sensors 2023,23, 7691 9 of 20
Similarly, Jonsson et al. established, using an unnatural amino acid incorporation
strategy, the first single-molecule observations of the conformational changes between the
engagement-competent s1 state and the processing-competent non-s1 states of the large,
multi-subunit ~2.5 MDa yeast 26S proteasome [
54
]. The distance between the subunit Rpn9
and the N-terminal of subunit Rpt5 on the 26S proteasome decreases by ~3 nm when the
26S proteasome goes from the s1 state to the non-s1 state. By labeling Rpn9 with LD555 and
Rpt5 with LD655, the s1 state corresponded to a distinct smFRET Eof ~0.3 and the non-S1
state corresponded to a distinct E of ~0.75 [
54
]. Using this system, they observed that the
26S predominantly resided in the low FRET s1 state [
54
]. The addition of substrate biased
the 26S proteasome to the non-S1 state [
54
]. Substrates with higher thermodynamic stability
increased the frequency of high FRET non-S1 states to return briefly to the low FRET s1
state [
54
]. Furthermore, the presence of tetra-ubiquitin chains allosterically stabilized the s1
state and reduced the rate of the s1-to-non-s1 transition by ~3-fold, suggesting ubiquitin
chain binding to the 26S proteasome might promote substrate engagement and degradation
initiation [
54
]. In these examples, researchers leveraged the advantages of single-molecule
fluorescence imaging to characterize conformational dynamics, new conformational states,
and allosteric regulators.
5.2. Protein Folding/Unfolding
Single-molecule fluorescence experiments can identify distinct intermediate states and
characterize the transition paths between each of the states during folding or unfolding
(Figure 6B). Free-diffusion single-molecule experiments ameliorate concerns of artifacts
from immobilizing a protein but limit the time of observation to milliseconds. To increase
the observation time scale and mimic free-diffusion conditions, Pirchi et al. encapsulated
adenylate kinase in a biotinylated lipid vesicle and tethered the lipid vesicle to the surface
through biotin–streptavidin–biotin–PEG surface interactions [
46
]. The average FRET effi-
ciency from the lipid vesicle-constrained single-molecule experiments showed agreement
with the FRET efficiency from bulk measurements and free-diffusion single-molecule ex-
periments [
46
]. They found the unfolding/folding of adenylate kinase involves at least six
states with an average trajectory length of 4.6 s and higher concentrations of denaturant
increase the probability of sequential transitions [46].
5.3. Protein Interactions
One of the more intriguing applications of single-molecule fluorescence experiments is
the detection of protein interactions (Figure 6C). This provides information on the dynamics,
the interaction’s dependence on the system, and the interaction’s effects on the function of
proteins. Bibeau et al. demonstrated that yeast cofilin binds to actin filaments independent
of curvature, but their results suggest actin curvature may facilitate cofilin dissociation [
85
].
They also show that cofilin clusters grow asymmetrically with the growth towards the
pointed end of the actin being twice as fast as the growth rate towards the barbed end [
85
].
These results offered novel insights into cofilin interactions with actin and cofilin clusters
under different conditions.
Asher et al. [
57
] observed the dynamics between a model GPCR protein, human
V
2
vasopressin receptor, and
β
-arrestin 1. They show that the
β
-arrestin 1 C-terminal
tail binds to its own N-terminal positively charged groove to block interaction with the
phosphorylated C-terminal of the human vasopressin receptor [
57
]. Immobilized
β
-arrestin
1 was labeled with LD555p and LD655 to directly observe the distance of
β
-arrestin 1
0
s C-
terminal tail to the N-terminal groove. Alone,
β
-arrestin 1 demonstrates a stable high FRET
state indicating interactions between the C-terminal and N-terminal groove. The addition
of phosphomimetic C-terminal peptides from human vasopressin receptors transitioned
the high FRET states to a lower FRET state indicating displacement of the C-terminal and
N-terminal groove. A full-length chimera receptor protein when bound with the agonist
epinephrine also demonstrated transitions to a lower FRET state albeit with shorter dwell
times in the lower state.
Sensors 2023,23, 7691 10 of 20
Poyton et al. investigated the interactions between nucleosomes and chromatin remod-
eler, SWR1 to understand the timing of histone and DNA dynamics when SWR1 mediates
histone H2A exchange for H2A.Z [
59
]. They find that most SWR1 binding events do not
lead to H2A exchange [
59
]. However, when exchange occurs, H2A remains in complex with
SWR1–nucleosome complex for tens of seconds after being displaced and DNA rewrapping
takes about 1.4 s.
To understand the mechanism of the drug ataluren on the eukaryotic ribosome, Huang
et al. labeled the peptide and tRNA. They found that ataluren binds to the ribosome and
competes with the release factor complex (RFC) [
86
]. In the absence of ataluren, 20 nM of
RFC resulted in a 50% maximum effect on peptide and tRNA dissociation with an effective
concentration (EC50) of 20 nM [
86
]. With the ataluren concentration at 1000 uM, the EC50
for RFC increases to 100 nM. This indicates that ataluren plays a role in regulating RFC
activity and alters the dissociation of peptides and tRNA from the ribosome.
Roca et al. used single-molecule fluorescence imaging to investigate the binding of
small RNA to the RNA-binding protein Hfq [
49
]. The content of the small RNA and the
binding interface on Hfq determined the effectiveness of the small RNA binding to Hfq [
49
].
5.4. Protein Post-Translational Modifications
Post-translational modifications of proteins regulate the activity and destruction of
proteins in the cell. Single-molecule
in vitro
fluorescence imaging can be used to directly
observe the addition of post-translational modifications as in Figure 6D or elucidate the
effects of post-translational modifications on the system. Recently, this technique has been
applied to studying ubiquitination [13,50] and phosphorylation [57].
Branigan et al. directly observed that ubiquitin transfer proceeds from a high FRET
signal corresponding to the closed ring conformational state of the E2 ubiquitin ligase [
50
].
Lu et al. elucidated the dynamics of ubiquitination by the E3 ligase APC, where APC dis-
plays a biphasic transfer to substrates [
13
,
14
]. Initially, APC adds three to five ubiquitins to
the substrate within the first five seconds. After, APC slowly elongates the ubiquitin chains.
The results from single-molecule fluorescence imaging of
in vitro
systems provided critical
information on protein structural states, protein interactions, and protein modifications
that are difficult to obtain any other way. The investigations highlighted in this section are
summarized in Table 1.
Sensors 2023,23, 7691 11 of 20
Table 1. Overview of biological applications.
Target
Biomolecule(s)
Fluorophore
Labeling Fluorophores Biotin
Conjugation Surface Analysis Software Results Camera Ref
Conformational Dynamics
Streptococcus
pyogenes CRISPR
Cas9 Cysteine–maleimide Cy3
LD750 Biotinylated DNA PEG Custom Cas90s HNH domain exhibits
dynamics coupled with
non-target strand cleavage EMCCD Wang [51]
Saccharomyces
cerevisiae
26S Proteasome
Cysteine–maleimide
Unnatural amino acid
“click” chemistry
Cy3
LD555
LD655
AviTag fusion reacted
with BirA in vitro PEG SPARTAN Ubiquitin chain binding to the
26S proteasome reduces the rate
of conformational transitions EMCCD Jonsson [54]
Mus musculus
metabotropic
glutamate receptor 2
and 3
Unnatural amino acid
“click” chemistry Cy3
Cy5 Commercial biotinylated
antiFLAG antibody PEG smCamera software Metabotropic glutamate receptor
2 displays four sequential
conformational states EMCCD Liauw [63]
Protein Folding
Escherichia coli
adenylate kinase Cysteine–maleimide Alexa 488
Atto 590 Biotinylated
phosphoethanolamine Lipids Custom MATLAB
folding of adenylate kinase
involves at least 6 states with
sequential and
non-sequential transitions
SPAD Pirchi [46]
Protein Interactions
Saccharomyces
cerevisiae
cofilin on actin Cysteine–maleimide Alexa 488
Alexa 647 Biotinylated actin Tween 20 TrackMate
MATLAB
ImageJ
Cofilin clusters grow 2 times
faster towards actin’s pointed
end versus barbed end EMCCD Bibeau [87]
Bovine β-arrestin1 Cysteine–maleimide LD555p
LD655 Strep-tag fusion PEG SPARTAN
β-arrestin1 tail displacement by
phosphorylated C-terminal
receptor requires GPCR agonist sCMOS Asher [58]
S. cerevisiae
Histone and SWR1 Cysteine–maleimide Cy3
Cy5
Cy7 Biotinylated DNA PEG Custom MATLAB
H2A remains in complex with
SWR1–nucleosome complex for
tens of seconds after
H2A.Z displacement
EMCCD Poyton [60]
Release factor
complex (RFC)
Lysine
hydroxysuccinimide
(NHS) ester
Cy3
Atto 647 mRNA biotinylated at
30end PEG ImageJ
Python
Ataluren, a translation
readthrough-inducing drug, acts
as a competitive inhibitor EMCCD Huang [88]
sRNA ChiX
sRNA DsrA
Escherichia coli Hfq
sRNA chaperone
50sRNA-free primary
amine-NHS ester Cy3
Cy5
AviTag fusion
biotinylated by
endogenous BirA Tween 20 Imscroll in MATLAB
Sometimes two sRNAs can stably
bind to Hfq. Most replacement
occurs when a strongly
competitive sRNA, ChiX,
replaces a moderately
competitive sRNA, DsrA.
EMCCD Roca [49]
Sensors 2023,23, 7691 12 of 20
Table 1. Cont.
Target
Biomolecule(s)
Fluorophore
Labeling Fluorophores Biotin
Conjugation Surface Analysis Software Results Camera Ref
Protein post-translational
Human Ubc13 E2
ubiquitin ligase Cysteine–maleimide Cy3B
Alexa 647 AviTag fusion reacted
with BirA in vitro PEG Interactive data
language (IDL)
Ubiquitin transfer proceeds from
high FRET signal corresponding
to the closed conformation
of ubc13
EMCCD Branigan [50]
Human
anaphase-promoting
complex E3
ubiquitin ligase
Cysteine–maleimide Alexa 488
DyLight 550
Alexa 647
Intein-mediated protein
ligation (IPL) of
biotin-containing peptide
to the C terminus
PEG Custom MATLAB Anaphase-promoting complex
displays biphasic activity EMCCD Lu [56]
Sensors 2023,23, 7691 13 of 20
6. Application of Lab-on-a-Chip Techniques for Single-Molecule Fluorescence Imaging
Lab-on-a-chip techniques have emerged as valuable tools in single-molecule fluores-
cence imaging, offering numerous advantages and enabling new possibilities for experi-
mental design and analysis [
87
–
93
]. Lab-on-a-chip devices provide unparalleled control
and manipulation capabilities, enabling precise management of fluid flow and sample
handling in single-molecule imaging experiments through the utilization of a laminar
flow regime. The controlled flow within microfluidic channels not only facilitates efficient
sample processing but also aids in reducing background noise by removing unbound or
non-specifically bound molecules, thereby enhancing the SNR. Moreover, lab-on-a-chip
platforms leverage miniaturized channels and chambers to create controlled microenvi-
ronments for the delivery, mixing, and incubation of samples, as well as the manipulation
of individual molecules. In this section, we will provide a concise introduction to several
key lab-on-a-chip-based platforms and highlight their recent applications in biological re-
search [
94
]. These platforms have successfully overcome the limitations of single-molecule
imaging while also improving experimental modalities and data quality.
6.1. Enhance Signal-to-Noise Ratio
One advantage of single-molecule imaging over ensemble studies is its superior time
domain resolution for investigating molecular dynamics. However, this advantage can
be compromised by the limited photostability of singlet exciton emission, which is prone
to bleaching and blinking due to factors like O
2
and intersystem crossing. The stochastic
fluctuations resulting from blinking are unrelated to the underlying biological behavior.
To overcome these challenges, microfluidics has been employed by incorporating oxygen
scavengers and triplet quenchers into the imaging buffer [
95
,
96
]. By carefully designing
the setup, this approach has recently facilitated the shortest observation durations [
97
].
It effectively addresses the limitation of time domain resolution posed by fluidic speed,
particularly during fluidic mixing. A different method was shown [
98
], wherein the imag-
ing channels were integrated with those consistently supplied with nitrogen ventilation
(Figure 7A). Furthermore, sophisticated microfluidic architectures can reduce flow velocities
immediately after mixing, enabling longer optical interrogations [99].
6.2. Increase Sample Concentration
As mentioned, single-molecule fluorescence imaging techniques are limited to using
pico- to nanomolar concentrations to ensure that only single molecules are resonant within
the laser-probed volume and provide a sufficient SNR. However, many biologically rele-
vant processes occur at micromolar level concentrations, necessitating a reduction in the
conventional observation volume by three orders of magnitude. Here, arrays of zero-mode
waveguides (ZMWs) consisting of subwavelength holes in a metal film provide a means to
increase sample concentrations to the micromolar range while confining the observation
volume to zeptoliter dimensions (Figure 7B) [
100
–
102
]. This breakthrough enables studies
in the physiological concentration range and has been successfully applied in real-time,
protein–protein interactions [102]. ZMWs have also been utilized to investigate ribosome-
mediated translation processes, allowing the observation of tRNA transit in real-time at
physiological concentrations [
103
]. Additionally, ZMWs have demonstrated versatility in
studying biomolecular interactions, protein receptor diffusion, and oligomerization on
living cell membranes [94].
6.3. On-Chip Single-Molecule Manipulation
On-chip devices have been developed with the capability to spatially modulate indi-
vidual molecules with nanometer or even sub-nanometer sensitivities. A notable example
is the microfluidic-based “DNA curtain”, which has recently emerged as an elegant on-chip
tool for investigating DNA–protein interactions [
16
–
18
,
47
,
104
]. Illustrated in Figure 7C,
this technique involves driving DNA molecules that are tethered to a fluidic lipid bilayer
on the surface. These molecules drift downstream under the influence of flow until they
Sensors 2023,23, 7691 14 of 20
encounter a thin layer of metal, which serves as a diffusion barrier. Consequently, the DNA
molecules align with each other, forming what is referred to as a DNA curtain [
47
]. This
fluidic-chip setup has proven highly successful in unraveling the searching modes of a
DNA repair complex at DNA damage and elucidating the disruption of a transcription
complex by a DNA translocase at the single-molecule level [
18
]. In addition, Alwan et al.
utilized a microfluidics-based single-molecule live cell fluorescence imaging to study the
spatiotemporal dynamics of selectin ligands on the membrane tethers and slings during
cell rolling (Figure 7D) [105].
Sensors 2023, 23, x FOR PEER REVIEW 13 of 19
Figure 7. Examples of lab-on-a-chip applications for single-molecule fluorescence imaging. (A) Pho-
tobleaching is reduced by deoxygenation via gas diffusion through porous channel walls in a mi-
crofluidic device. Reprinted with permission from Ref. [98]. Copyright 2009 American Chemical
Society. (B) Real-time imaging of single-molecule fluorescence with a ZMW for the study of protein–
protein interaction. (C) Microfluidic-based DNA curtain platform allows parallel data acquisition of
individual protein−DNA interactions in real time. Adapted with permission from Ref. [104]. Copy-
right 2008 American Chemical Society. (D) A microfluidics-based single-molecule live cell fluores-
cence imaging platform for the study of spatiotemporal dynamics of selectin–ligand interactions
during cell rolling.
6.4. Microenvironment Control
Lab-on-a-chip technology also affords precise control over the microenvironment
surrounding single molecules. Variables such as temperature, pH, and chemical gradients
can be precisely manipulated within microfluidic devices, providing valuable insights
into the impact of different conditions on the behavior and functionality of biomolecules.
This level of control allows for the investigation of dynamic processes under various phys-
iological or pathological conditions, mimicking complex biological environments. For ex-
ample, Zhang et al. studied the in situ conformational response of single biomolecules
such as DNA to a change in environmental solution conditions [106]. This level of control
allows researchers to probe biomolecular interactions, enzymatic activities, and other dy-
namic processes with exceptional temporal resolution.
Moreover, lab-on-a-chip devices possess the remarkable capability of automation
when integrated with other techniques. This integration not only minimizes experimental
bias but also facilitates high-throughput screening, data acquisition, and analysis, which
are indispensable for conducting large-scale single-molecule studies. By automating mi-
crofluidic processes, researchers can streamline their experiments, achieve consistent and
reliable results, and analyze vast amounts of data efficiently.
7. Conclusions and Future Directions
Despite the remarkable advancements in fluorescence-based single-molecule imag-
ing techniques, several limitations still exist, which present opportunities for further
Figure 7.
Examples of lab-on-a-chip applications for single-molecule fluorescence imaging.
(
A
) Photobleaching is reduced by deoxygenation via gas diffusion through porous channel walls
in a microfluidic device. Reprinted with permission from Ref. [
98
]. Copyright 2009 American
Chemical Society. (
B
) Real-time imaging of single-molecule fluorescence with a ZMW for the study
of protein–protein interaction. (
C
) Microfluidic-based DNA curtain platform allows parallel data
acquisition of individual protein
−
DNA interactions in real time. Adapted with permission from
Ref. [
104
]. Copyright 2008 American Chemical Society. (
D
) A microfluidics-based single-molecule
live cell fluorescence imaging platform for the study of spatiotemporal dynamics of selectin–ligand
interactions during cell rolling.
6.4. Microenvironment Control
Lab-on-a-chip technology also affords precise control over the microenvironment
surrounding single molecules. Variables such as temperature, pH, and chemical gradients
can be precisely manipulated within microfluidic devices, providing valuable insights
into the impact of different conditions on the behavior and functionality of biomolecules.
This level of control allows for the investigation of dynamic processes under various
physiological or pathological conditions, mimicking complex biological environments. For
example, Zhang et al. studied the in situ conformational response of single biomolecules
such as DNA to a change in environmental solution conditions [
106
]. This level of control
allows researchers to probe biomolecular interactions, enzymatic activities, and other
dynamic processes with exceptional temporal resolution.
Sensors 2023,23, 7691 15 of 20
Moreover, lab-on-a-chip devices possess the remarkable capability of automation
when integrated with other techniques. This integration not only minimizes experimental
bias but also facilitates high-throughput screening, data acquisition, and analysis, which
are indispensable for conducting large-scale single-molecule studies. By automating mi-
crofluidic processes, researchers can streamline their experiments, achieve consistent and
reliable results, and analyze vast amounts of data efficiently.
7. Conclusions and Future Directions
Despite the remarkable advancements in fluorescence-based single-molecule imaging
techniques, several limitations still exist, which present opportunities for further devel-
opment and improvement. One of the major challenges in single-molecule imaging is
photobleaching, which refers to the irreversible loss of fluorescence caused by repeated
excitation. This phenomenon poses limitations on the observation time and hampers the
investigation of long-lived biological processes. Another critical aspect of precise single-
molecule imaging is the efficient and specific labeling of biomolecules with fluorophores.
However, existing labeling methods may introduce artifacts, alter the natural behavior
of molecules, or impact their functionality. Future investigations should aim to enhance
labeling techniques, striving for high efficiency, specificity, and minimal disruption to the
biological system at hand. This pursuit encompasses developing novel labeling strategies,
including genetically encoded tags and chemical modification approaches, which afford
improved targeting capabilities. Simultaneous imaging of multiple molecular species or
different structural components within complex systems holds immense value. However,
spectral overlap among fluorophores presents challenges in reliable multi-color imaging.
Future endeavors involve designing and synthesizing fluorophores with narrower emis-
sion spectra and refining spectral separation techniques. Additionally, the development of
advanced imaging setups, detection algorithms, and novel fluorophore combinations will
enable more precise and efficient multi-color imaging experiments.
Furthermore, while fluorescence-based single-molecule imaging offers impressive
spatial and temporal resolution, advancements are sought to observe dynamic molecular
processes at an even finer scale. Innovations in super-resolution techniques like SMLM
or stimulated emission depletion (STED) microscopy can push the boundaries of spatial
resolution [
27
–
29
,
31
,
107
–
111
]. Similarly, the progress in ultrafast imaging methods and the
design of detectors with heightened sensitivity and speed will facilitate the study of rapid
molecular dynamics with heightened temporal resolution.
As single-molecule imaging experiments generate increasingly complex and volumi-
nous datasets, there arises a need for sophisticated data analysis techniques and integration
with other omics data. Future investigations should concentrate on developing advanced
analysis algorithms, machine learning approaches, and statistical modeling methods to
extract comprehensive insights from acquired data. Integrating single-molecule imaging
data with other techniques such as genomics, proteomics, or structural biology will provide
a holistic understanding of biological processes, facilitating the correlation of molecular
behavior with higher order cellular functions.
Expanding fluorescence-based single-molecule imaging to
in vivo
settings and dy-
namic cellular environments poses significant challenges. Factors like autofluorescence,
scattering, motion artifacts, and physiological conditions present formidable hurdles. Fu-
ture directions should explore strategies to address these obstacles, including the design of
biocompatible fluorophores, advanced imaging approaches to mitigate tissue scattering,
and imaging techniques capable of capturing real-time dynamics in living systems.
One of the first biological applications of single-molecule TIRF was published in
1995 [
112
]. The fluorophores (Cy3 and Cy5), the labeling strategy (cysteine–maleimide),
and the type of camera (CCD) used in the first application are still often used today [
112
].
However, these fluorophores and labeling strategies do not work for every biological sys-
tem. Now, researchers have more choices (i.e., passivation, conjugation, immobilization,
and analysis strategies) to apply single-molecule TIRF to almost any biological system.
Sensors 2023,23, 7691 16 of 20
Certain steps still pose challenges, notably the site-specific labeling of large proteins and
protein complexes. Nevertheless, the expanded options, at the very least, provide ways
to address these complexities. For example, the unnatural amino acid conjugation strat-
egy enables the application of single-molecule TIRF to large biomolecules such as the
2.5 MDa 26S proteasome [
54
]. Additionally, the liposome immobilization strategy indirectly
constrained the location of the protein to improve the folding/unfolding measurements of
proteins [
46
]. From a survey of the recent literature on biological applications, one of the
last remaining areas to improve is the ability to visualize diverse protein post-translational
modifications. Future studies should aim to develop fluorescent reporters for glycosylation,
phosphorylation, and methylation.
Lab-on-a-chip devices offer researchers a powerful tool to scale up the application
of single-molecule fluorescence imaging. These microfluidic platforms provide unprece-
dented capabilities, enabling precise fluid flow and microenvironmental control, leading to
enhanced signal-to-noise ratios and improved data quality. Moreover, the automation and
design of lab-on-a-chip devices have the potential to substantially enhance data collection
throughput, potentially accommodating hundreds of conditions and samples within one
device. To enable these single-molecule devices, researchers should simplify the pipeline of
in vitro single-molecule fluorescence imaging and tailor the design of the devices.
Author Contributions:
Conceptualization, L.C., Y.L. and Y.C.; writing—original draft preparation,
L.C. and Y.C.; writing—review and editing, Y.K., S.N., A.B., N.M.M. and Y.L.; supervision, project
administration, Y.C.; funding acquisition, Y.C. All authors have read and agreed to the published
version of the manuscript.
Funding: This work was supported by the 2023 Research Fund of Myongji University.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: No new data were created in this work.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Lee, K.; Kim, K.; Jung, J.; Heo, J.; Cho, S.; Lee, S.; Chang, G.; Jo, Y.; Park, H.; Park, Y.; et al. Quantitative phase imaging techniques
for the study of cell pathophysiology: From principles to applications. Sensors 2013,13, 4170–4191. [CrossRef] [PubMed]
2.
Royer, L.A.; Lemon, W.C.; Chhetri, R.K.; Wan, Y.; Coleman, M.; Myers, E.W.; Keller, P.J. Adaptive light-sheet microscopy for
long-term, high-resolution imaging in living organisms. Nat. Biotechnol. 2016,34, 1267–1278. [CrossRef] [PubMed]
3.
Wassie, A.T.; Zhao, Y.; Boyden, E.S. Expansion microscopy: Principles and uses in biological research. Nat. Methods
2019
,16,
33–41. [CrossRef] [PubMed]
4. Becker, W. Fluorescence lifetime imaging–techniques and applications. J. Microsc. 2012,247, 119–136. [CrossRef] [PubMed]
5.
Fu, D.; Zhou, J.; Zhu, W.S.; Manley, P.W.; Wang, Y.K.; Hood, T.; Wylie, A.; Xie, X.S. Imaging the intracellular distribution of
tyrosine kinase inhibitors in living cells with quantitative hyperspectral stimulated Raman scattering. Nat. Chem.
2014
,6, 614–622.
[CrossRef]
6.
Lu, F.K.; Basu, S.; Igras, V.; Hoang, M.P.; Ji, M.; Fu, D.; Holtom, G.R.; Neel, V.A.; Freudiger, C.W.; Fisher, D.E.; et al. Label-free DNA
imaging in vivo with stimulated Raman scattering microscopy. Proc. Natl. Acad. Sci. USA 2015,112, 11624–11629. [CrossRef]
7.
Roeffaers, M.B.; Zhang, X.; Freudiger, C.W.; Saar, B.G.; van Ruijven, M.; van Dalen, G.; Xiao, C.; Xie, X.S. Label-free imaging of
biomolecules in food products using stimulated Raman microscopy. J. Biomed. Opt. 2010,15, 066016. [CrossRef]
8.
Shi, L.; Zheng, C.; Shen, Y.; Chen, Z.; Silveira, E.S.; Zhang, L.; Wei, M.; Liu, C.; de Sena-Tomas, C.; Targoff, K.; et al. Optical
imaging of metabolic dynamics in animals. Nat. Commun. 2018,9, 2995. [CrossRef]
9.
Jalili, N.; Laxminarayana, K. A review of atomic force microscopy imaging systems: Application to molecular metrology and
biological sciences. Mechatronics 2004,14, 907–945. [CrossRef]
10.
Kherlopian, A.R.; Song, T.; Duan, Q.; Neimark, M.A.; Po, M.J.; Gohagan, J.K.; Laine, A.F. A review of imaging techniques for
systems biology. BMC Syst. Biol. 2008,2, 74. [CrossRef]
11.
Shashkova, S.; Leake, M.C. Single-molecule fluorescence microscopy review: Shedding new light on old problems. Biosci. Rep.
2017,37, BSR20170031. [CrossRef]
12.
Cho, Y.; Zhao, C.L.; Lu, H. Trends in high-throughput and functional neuroimaging in Caenorhabditis elegans. Wiley Interdiscip.
Rev. Syst. Biol. Med. 2017,9, e1376. [CrossRef] [PubMed]
Sensors 2023,23, 7691 17 of 20
13.
Lu, Y.; Lee, B.-h.; King, R.W.; Finley, D.; Kirschner, M.W. Substrate degradation by the proteasome: A single-molecule kinetic
analysis. Science 2015,348, 1250834. [CrossRef] [PubMed]
14.
Lu, Y.; Wang, W.; Kirschner, M.W. Specificity of the anaphase-promoting complex: A single-molecule study. Science
2015
,348, 1248737.
[CrossRef]
15.
Rief, M.; Žoldák, G. Single-molecule mechanical studies of chaperones and their clients. Biophys. Rev.
2022
,3, 041301. [CrossRef]
16.
Georgescu, R.E.; Yao, N.Y.; O’Donnell, M. Single-molecule analysis of the Escherichia coli replisome and use of clamps to bypass
replication barriers. FEBS Lett. 2010,584, 2596–2605. [CrossRef] [PubMed]
17.
Visnapuu, M.-L.; Greene, E.C. Single-molecule imaging of DNA curtains reveals intrinsic energy landscapes for nucleosome
deposition. Nat. Struct. Mol. Biol. 2009,16, 1056–1062. [CrossRef] [PubMed]
18.
Gorman, J.; Wang, F.; Redding, S.; Plys, A.J.; Fazio, T.; Wind, S.; Alani, E.E.; Greene, E.C. Single-molecule imaging reveals
target-search mechanisms during DNA mismatch repair. Proc. Natl. Acad. Sci. USA 2012,109, E3074–E3083. [CrossRef]
19.
Hon, J.; Lu, Y. Single-molecule methods for measuring ubiquitination and protein stability. Methods Enzymol.
2019
,619, 225–247.
20.
van Ginkel, J.; Filius, M.; Szczepaniak, M.; Tulinski, P.; Meyer, A.S.; Joo, C. Single-molecule peptide fingerprinting. Proc. Natl.
Acad. Sci. USA 2018,115, 3338–3343. [CrossRef]
21.
Gust, A.; Zander,A.; Gietl, A.; Holzmeister, P.; Schulz, S.; Lalkens, B.; Tinnefeld, P.; Grohmann, D. A starting point for fluorescence-
based single-molecule measurements in biomolecular research. Molecules 2014,19, 15824–15865. [CrossRef] [PubMed]
22.
Moerner, W.; Fromm, D.P. Methods of single-molecule fluorescence spectroscopy and microscopy. Rev. Sci. Instrum.
2003
,74,
3597–3619. [CrossRef]
23. Axelrod, D. Total internal reflection fluorescence microscopy. Methods Cell Biol. 1989,30, 245–270.
24. Axelrod, D. Total internal reflection fluorescence microscopy in cell biology. Traffic 2001,2, 764–774. [CrossRef] [PubMed]
25.
Schneckenburger, H. Total internal reflection fluorescence microscopy: Technical innovations and novel applications. Curr. Opin.
Biotechnol. 2005,16, 13–18. [CrossRef]
26.
Endesfelder, U.; Heilemann, M. Art and artifacts in single-molecule localization microscopy: Beyond attractive images.
Nat. Methods 2014,11, 235–238. [CrossRef]
27.
Mortensen, K.I.; Churchman, L.S.; Spudich, J.A.; Flyvbjerg, H. Optimized localization analysis for single-molecule tracking and
super-resolution microscopy. Nat. Methods 2010,7, 377–381. [CrossRef]
28.
Wu, Y.-L.; Tschanz, A.; Krupnik, L.; Ries, J. Quantitative data analysis in single-molecule localization microscopy. Trends Cell Biol.
2020,30, 837–851. [CrossRef]
29.
Sage, D.; Kirshner, H.; Pengo, T.; Stuurman, N.; Min, J.; Manley, S.; Unser, M. Quantitative evaluation of software packages for
single-molecule localization microscopy. Nat. Methods 2015,12, 717–724. [CrossRef]
30.
Khater, I.M.; Nabi, I.R.; Hamarneh, G. A review of super-resolution single-molecule localization microscopy cluster analysis and
quantification methods. Patterns 2020,1, 100038. [CrossRef]
31.
Lelek, M.; Gyparaki, M.T.; Beliu, G.; Schueder, F.; Griffié, J.; Manley, S.; Jungmann, R.; Sauer, M.; Lakadamyali, M.; Zimmer, C.
Single-molecule localization microscopy. Nat. Rev. Methods Primers 2021,1, 39. [CrossRef]
32.
Nicovich, P.R.; Owen, D.M.; Gaus, K. Turning single-molecule localization microscopy into a quantitative bioanalytical tool.
Nat. Protoc. 2017,12, 453–460. [CrossRef] [PubMed]
33. Roy, R.; Hohng, S.; Ha, T. A practical guide to single-molecule FRET. Nat. Methods 2008,5, 507–516. [CrossRef]
34.
Imran, A.; Zhu, Q.; Sulaman, M.; Bukhtiar, A.; Xu, M. Electric-Dipole Gated Two Terminal Phototransistor for Charge-Coupled
Device. Adv. Opt. Mater. 2023, 2300910. [CrossRef]
35.
Chen, H.; Liu, H.; Zhang, Z.; Hu, K.; Fang, X. Nanostructured Photodetectors: From Ultraviolet to Terahertz. Adv. Mater.
2016
,28,
403–433. [CrossRef] [PubMed]
36.
Lanza, M.; Smets, Q.; Huyghebaert, C.; Li, L.-J. Yield, variability, reliability, and stability of two-dimensional materials based
solid-state electronic devices. Nat. Commun. 2020,11, 5689. [CrossRef]
37.
Ha, T.; Tinnefeld, P. Photophysics of fluorescent probes for single-molecule biophysics and super-resolution imaging. Annu. Rev.
Phys. Chem. 2012,63, 595–617. [CrossRef]
38. Ha, T. Probing Nature’s Nanomachines One Molecule at a Time. Biophys. J. 2016,110, 1004–1007. [CrossRef]
39.
Wilson, H.; Wang, Q. ABEL-FRET: Tether-free single-molecule FRET with hydrodynamic profiling. Nat. Methods
2021
,18, 816–820.
[CrossRef]
40.
Ye, Y.; Blaser, G.; Horrocks, M.H.; Ruedas-Rama, M.J.; Ibrahim, S.; Zhukov, A.A.; Orte, A.; Klenerman, D.; Jackson, S.E.; Komander,
D. Ubiquitin chain conformation regulates recognition and activity of interacting proteins. Nature
2012
,492, 266–270. [CrossRef]
41.
Chandradoss, S.D.; Haagsma, A.C.; Lee, Y.K.; Hwang, J.-H.; Nam, J.-M.; Joo, C. Surface Passivation for Single-molecule Protein
Studies. J. Vis. Exp. 2014,86, 50549.
42.
Hua, B.; Han, K.Y.; Zhou, R.; Kim, H.; Shi, X.; Abeysirigunawardena, S.C.; Jain, A.; Singh, D.; Aggarwal, V.; Woodson, S.A.; et al.
An improved surface passivation method for single-molecule studies. Nat. Methods 2014,11, 1233–1236. [CrossRef]
43.
Paul, T.; Ha, T.; Myong, S. Regeneration of PEG slide for multiple rounds of single-molecule measurements. Biophys. J.
2021
,120,
1788–1799. [CrossRef] [PubMed]
44.
Yoo, J.; Louis, J.M.; Chung, H.S. Diverse Folding Pathways of HIV-1 Protease Monomer on a Rugged Energy Landscape. Biophys.
J. 2019,117, 1456–1466. [CrossRef] [PubMed]
Sensors 2023,23, 7691 18 of 20
45.
Chung, H.S.; McHale, K.; Louis, J.M.; Eaton, W.A. Single-Molecule Fluorescence Experiments Determine Protein Folding
Transition Path Times. Science 2012,335, 981–984. [CrossRef] [PubMed]
46.
Pirchi, M.; Ziv, G.; Riven, I.; Cohen, S.S.; Zohar, N.; Barak, Y.; Haran, G. Single-molecule fluorescence spectroscopy maps the
folding landscape of a large protein. Nat. Commun. 2011,2, 493. [CrossRef]
47. Granéli, A.; Yeykal, C.C.; Prasad, T.K.; Greene, E.C. Organized arrays of individual DNA molecules tethered to supported lipid
bilayers. Langmuir 2006,22, 292–299. [CrossRef]
48. Green, N.M. Avidin. 1. The use of (14-C)biotin for kinetic studies and for assay. Biochem. J. 1963,89, 585–591. [CrossRef]
49.
Roca, J.; Santiago-Frangos, A.; Woodson, S.A. Diversity of bacterial small RNAs drives competitive strategies for a mutual
chaperone. Nat. Commun. 2022,13, 2449. [CrossRef]
50.
Branigan, E.; Carlos Penedo, J.; Hay, R.T. Ubiquitin transfer by a RING E3 ligase occurs from a closed E2~ubiquitin conformation.
Nat. Commun. 2020,11, 2846. [CrossRef]
51.
Wang, Y.; Mallon, J.; Wang, H.; Singh, D.; Hyun Jo, M.; Hua, B.; Bailey, S.; Ha, T. Real-time observation of Cas9 postcatalytic
domain motions. Proc. Natl. Acad. Sci. USA 2021,118, 1. [CrossRef] [PubMed]
52.
Beckett, D.; Kovaleva, E.; Schatz, P.J. A minimal peptide substrate in biotin holoenzyme synthetase-catalyzed biotinylation.
Protein Sci. 1999,8, 921–929. [CrossRef] [PubMed]
53.
Cull, M.G.; Schatz, P.J. Biotinylation of proteins
in vivo
and
in vitro
using small peptide tags. Methods Enzymol.
2000
,326, 430–440.
[PubMed]
54.
Jonsson, E.; Htet, Z.M.; Bard, J.A.M.; Dong, K.C.; Martin, A. Ubiquitin modulates 26S proteasome conformational dynamics and
promotes substrate degradation. Sci. Adv. 2022,8, eadd9520. [CrossRef]
55.
Chatterjee, S.; Ade, C.; Nurik, C.E.; Carrejo, N.C.; Dutta, C.; Jayaraman, V.; Landes, C.F. Phosphorylation Induces Conformational
Rigidity at the C-Terminal Domain of AMPA Receptors. J. Phys. Chem. B 2019,123, 130–137. [CrossRef]
56.
Ciftci, D.; Huysmans, G.H.M.; Wang, X.; He, C.; Terry, D.; Zhou, Z.; Fitzgerald, G.; Blanchard, S.C.; Boudker, O. Single-molecule
transport kinetics of a glutamate transporter homolog shows static disorder. Sci. Adv. 2020,6, eaaz1949. [CrossRef]
57.
Asher, W.B.; Terry, D.S.; Gregorio, G.G.A.; Kahsai, A.W.; Borgia, A.; Xie, B.; Modak, A.; Zhu, Y.; Jang, W.; Govindaraju, A.; et al.
GPCR-mediated β-arrestin activation deconvoluted with single-molecule precision. Cell 2022,185, 1661–1675.e16. [CrossRef]
58. Tran, T.; Mitra, J.; Ha, T.; Kavran, J.M. Increasing kinase domain proximity promotes MST2 autophosphorylation during Hippo
signaling. J. Biol. Chem. 2020,295, 16166–16179. [CrossRef]
59.
Poyton, M.F.; Feng, X.A.; Ranjan, A.; Lei, Q.; Wang, F.; Zarb, J.S.; Louder, R.K.; Park, G.; Jo, M.H.; Ye, J.; et al. Coordinated DNA
and histone dynamics drive accurate histone H2A.Z exchange. Sci. Adv. 2022,8, eabj5509. [CrossRef]
60.
Park, S.R.; Hauver, J.; Zhang, Y.; Revyakin, A.; Coleman, R.A.; Tjian, R.; Chu, S.; Pertsinidis, A. A Single-Molecule Surface-Based
Platform to Detect the Assembly and Function of the Human RNA Polymerase II Transcription Machinery. Structrue
2020
,28,
1337–1343.e4. [CrossRef]
61.
Song, E.; Uhm, H.; Munasingha, P.R.; Hwang, S.; Seo, Y.-S.; Kang, J.Y.; Kang, C.; Hohng, S. Rho-dependent transcription
termination proceeds via three routes. Nat. Commun. 2022,13, 1663. [CrossRef] [PubMed]
62.
Liauw, B.W.-H.; Afsari, H.S.; Vafabakhsh, R. Conformational rearrangement during activation of a metabotropic glutamate
receptor. Nat. Chem. Biol. 2021,17, 291–297. [CrossRef] [PubMed]
63.
Fang, R.; Hon, J.; Zhou, M.; Lu, Y. An empirical energy landscape reveals mechanism of proteasome in polypeptide translocation.
eLife 2022,11, e71911. [CrossRef] [PubMed]
64.
Zhou, Y.; Fard, A.P.; Davies, A. Characterization of instrument drift using a spherical artifact. Precis. Eng.
2014
,38, 443–447.
[CrossRef]
65.
Wang, Y.; Schnitzbauer, J.; Hu, Z.; Li, X.; Cheng, Y.; Huang, Z.-L.; Huang, B. Localization events-based sample drift correction for
localization microscopy with redundant cross-correlation algorithm. Opt. Express 2014,22, 15982–15991. [CrossRef]
66. Davenport, W.B.; Root, W.L. Random Signals and Noise; McGraw-Hill: New York, NY, USA, 1958.
67.
Kou, S.C.; Xie, X.S. Generalized Langevin equation with fractional Gaussian noise: Subdiffusion within a single protein molecule.
Phys. Rev. Lett. 2004,93, 180603. [CrossRef]
68. Jakeman, E. On the statistics of K-distributed noise. J. Phys. A Math. General. 1980,13, 31. [CrossRef]
69.
Chan, R.; Yang, H.; Zeng, T. A two-stage image segmentation method for blurry images with poisson or multiplicative gamma
noise. SIAM J. Imaging Sci. 2014,7, 98–127. [CrossRef]
70. Kassam, S.A. Signal Detection in Non—Gaussian Noise; Springer Science & Business Media: Berlin, Germany, 2012.
71.
Shnerb, N.; Bettelheim, E.; Louzoun, Y.; Agam, O.; Solomon, S. Adaptation of autocatalytic fluctuations to diffusive noise.
Phys. Rev. E 2001,63, 021103. [CrossRef]
72.
Flegg, M.B.; Rüdiger, S.; Erban, R. Diffusive spatio-temporal noise in a first-passage time model for intracellular calcium release.
J. Chem. Phys. 2013,138, 154103. [CrossRef]
73.
Colomb, W.; Sarkar, S.K. Extracting physics of life at the molecular level: A review of single-molecule data analyses. Phys. Life
Rev. 2015,13, 107–137. [CrossRef] [PubMed]
74.
Rieckh, G.; Tkaˇcik, G. Noise and information transmission in promoters with multiple internal states. Biophys. J.
2014
,106,
1194–1204. [CrossRef] [PubMed]
75.
Yuan, J.; He, K.; Cheng, M.; Yu, J.; Fang, X. Analysis of the Steps in Single-Molecule Photobleaching Traces by Using the Hidden
Markov Model and Maximum-Likelihood Clustering. Chem.—An Asian J. 2014,9, 2303–2308. [CrossRef] [PubMed]
Sensors 2023,23, 7691 19 of 20
76.
Blanco, M.; Johnson-Buck, A.; Walter, N. Hidden Markov modeling in single-molecule biophysics. In Encyclopedia of Biophysics;
Springer: Berlin/Heidelberg, Germany, 2013; pp. 971–975.
77.
Tao, L.; Ozarkar, S.; Beck, J.M.; Bhandawat, V. Statistical structure of locomotion and its modulation by odors. Elife
2019
,8, e41235.
[CrossRef]
78.
Okamoto, K.; Sako, Y. Variational Bayes analysis of a photon-based hidden Markov model for single-molecule FRET trajectories.
Biophys. J. 2012,103, 1315–1324. [CrossRef]
79.
Sungkaworn, T.; Jobin, M.-L.; Burnecki, K.; Weron, A.; Lohse, M.J.; Calebiro, D. Single-molecule imaging reveals receptor–G
protein interactions at cell surface hot spots. Nature 2017,550, 543–547. [CrossRef]
80.
Liu, X.; Jiang, Y.; Cui, Y.; Yuan, J.; Fang, X. Deep learning in single-molecule imaging and analysis recent advances and prospects.
Chem. Sci. 2022,13, 11964–11980. [CrossRef]
81.
Wang, Q.; He, H.; Zhang, Q.; Feng, Z.; Li, J.; Chen, X.; Liu, L.; Wang, X.; Ge, B.; Yu, D. Deep-learning-assisted single-molecule
tracking on a live cell membrane. Anal. Chem. 2021,93, 8810–8816. [CrossRef]
82.
Prokhorova, I.; Altman, R.B.; Djumagulov, M.; Shrestha, J.P.; Urzhumtsev, A.; Ferguson, A.; Chang, C.-W.T.; Yusupov, M.;
Blanchard, S.C.; Yusupova, G.; et al. Aminoglycoside interactions and impacts on the eukaryotic ribosome. Proc. Natl. Acad. Sci.
USA 2017,114, E10899–E10908. [CrossRef]
83.
Chen, C.; Stevens, B.; Kaur, J.; Cabral, D.; Liu, H.; Wang, Y.; Zhang, H.; Rosenblum, G.; Smilansky, Z.; Goldman, Y.E.; et al.
Single-Molecule Fluorescence Measurements of Ribosomal Translocation Dynamics. Mol. Cell 2011,42, 367–377. [CrossRef]
84.
Lee, S.; Lee, J.; Hohng, S. Single-Molecule Three-Color FRET with Both Negligible Spectral Overlap and Long Observation Time.
PLoS ONE 2010,5, e12270. [CrossRef] [PubMed]
85.
Bibeau, J.P.; Gray, S.; De La Cruz, E.M. Clusters of a Few Bound Cofilins Sever Actin Filaments. J. Mol. Biol.
2021
,433, 166833.
[CrossRef] [PubMed]
86.
Huang, S.; Bhattacharya, A.; Ghelfi, M.D.; Li, H.; Fritsch, C.; Chenoweth, D.M.; Goldman, Y.E.; Cooperman, B.S. Ataluren binds to
multiple protein synthesis apparatus sites and competitively inhibits release factor-dependent termination. Nat. Commun.
2022
,
13, 2413. [CrossRef] [PubMed]
87.
Bhagat, A.A.; Bow, H.; Hou, H.W.; Tan, S.J.; Han, J.; Lim, C.T. Microfluidics for cell separation. Med. Biol. Eng. Comput.
2010
,48,
999–1014. [CrossRef]
88.
Kuntaegowdanahalli, S.S.; Bhagat, A.A.; Kumar, G.; Papautsky, I. Inertial microfluidics for continuous particle separation in spiral
microchannels. Lab. Chip 2009,9, 2973–2980. [CrossRef]
89.
Lu, Y.; Yang, L.; Wei, W.; Shi, Q. Microchip-based single-cell functional proteomics for biomedical applications. Lab. Chip
2017
,17,
1250–1263. [CrossRef]
90.
Shields, C.W.; Reyes, C.D.; Lopez, G.P. Microfluidic cell sorting: A review of the advances in the separation of cells from debulking
to rare cell isolation. Lab. Chip 2015,15, 1230–1249. [CrossRef]
91.
Xi, H.D.; Zheng, H.; Guo, W.; Ganan-Calvo, A.M.; Ai, Y.; Tsao, C.W.; Zhou, J.; Li, W.; Huang, Y.; Nguyen, N.T.; et al. Active
droplet sorting in microfluidics: A review. Lab. Chip 2017,17, 751–771. [CrossRef]
92. Yin, H.; Marshall, D. Microfluidics for single cell analysis. Curr. Opin. Biotechnol. 2012,23, 110–119. [CrossRef]
93.
Young, E.W.; Beebe, D.J. Fundamentals of microfluidic cell culture in controlled microenvironments. Chem. Soc. Rev.
2010
,39,
1036–1048. [CrossRef]
94.
Zhao, Y.; Chen, D.; Yue, H.; French, J.B.; Rufo, J.; Benkovic, S.J.; Huang, T.J. Lab-on-a-chip technologies for single-molecule studies.
Lab. Chip 2013,13, 2183–2198. [CrossRef] [PubMed]
95.
Rasnik, I.; McKinney, S.A.; Ha, T. Nonblinking and long-lasting single-molecule fluorescence imaging. Nat. Methods
2006
,3,
891–893. [CrossRef] [PubMed]
96.
Vasdekis, A.E.; Laporte, G.P. Enhancing single molecule imaging in optofluidics and microfluidics. Int. J. Mol. Sci.
2011
,12,
5135–5156. [CrossRef]
97.
Campos, L.A.; Liu, J.; Wang, X.; Ramanathan, R.; English, D.S.; Munoz, V. A photoprotection strategy for microsecond-resolution
single-molecule fluorescence spectroscopy. Nat. Methods 2011,8, 143–146. [CrossRef] [PubMed]
98.
Lemke, E.A.; Gambin, Y.; Vandelinder, V.; Brustad, E.M.; Liu, H.-W.; Schultz, P.G.; Groisman, A.; Deniz, A.A. Microfluidic device
for single-molecule experiments with enhanced photostability. J. Am. Chem. Soc. 2009,131, 13610–13612. [CrossRef] [PubMed]
99.
Gambin, Y.; VanDelinder, V.; Ferreon, A.C.; Lemke, E.A.; Groisman, A.; Deniz, A.A. Visualizing a one-way protein encounter
complex by ultrafast single-molecule mixing. Nat. Methods 2011,8, 239–241. [CrossRef] [PubMed]
100.
Levene, M.J.; Korlach, J.; Turner, S.W.; Foquet, M.; Craighead, H.G.; Webb, W.W. Zero-mode waveguides for single-molecule
analysis at high concentrations. Science 2003,299, 682–686. [CrossRef]
101.
Zhu, P.; Craighead, H.G. Zero-mode waveguides for single-molecule analysis. Annu. Rev. Biophys.
2012
,41, 269–293. [CrossRef]
102.
Miyake, T.; Tanii, T.; Sonobe, H.; Akahori, R.; Shimamoto, N.; Ueno, T.; Funatsu, T.; Ohdomari, I. Real-time imaging of single-
molecule fluorescence with a zero-mode waveguide for the analysis of protein
−
protein interaction. Anal. Chem.
2008
,80,
6018–6022. [CrossRef]
103.
Uemura, S.; Aitken, C.E.; Korlach, J.; Flusberg, B.A.; Turner, S.W.; Puglisi, J.D. Real-time tRNA transit on single translating
ribosomes at codon resolution. Nature 2010,464, 1012–1017. [CrossRef]
104.
Fazio, T.; Visnapuu, M.-L.; Wind, S.; Greene, E.C. DNA curtains and nanoscale curtain rods: High-throughput tools for single
molecule imaging. Langmuir 2008,24, 10524–10531. [CrossRef] [PubMed]
Sensors 2023,23, 7691 20 of 20
105.
Al Alwan, B.; AbuZineh, K.; Nozue, S.; Rakhmatulina, A.; Aldehaiman, M.; Al-Amoodi, A.S.; Serag, M.F.; Aleisa, F.A.;
Merzaban, J.S.; Habuchi, S. Single-molecule imaging and microfluidic platform reveal molecular mechanisms of leukemic
cell rolling. Commun. Biol. 2021,4, 868. [CrossRef] [PubMed]
106.
Zhang, C.; Jiang, K.; Liu, F.; Doyle, P.S.; van Kan, J.A.; van der Maarel, J.R. A nanofluidic device for single molecule studies with
in situ control of environmental solution conditions. Lab. A Chip 2013,13, 2821–2826. [CrossRef] [PubMed]
107.
Ouyang, W.; Aristov, A.; Lelek, M.; Hao, X.; Zimmer, C. Deep learning massively accelerates super-resolution localization
microscopy. Nat. Biotechnol. 2018,36, 460–468. [CrossRef] [PubMed]
108.
Small, A.; Stahlheber, S. Fluorophore localization algorithms for super-resolution microscopy. Nat. Methods
2014
,11, 267–279.
[CrossRef]
109.
Möckl, L.; Moerner, W. Super-resolution microscopy with single molecules in biology and beyond–essentials, current trends, and
future challenges. J. Am. Chem. Soc. 2020,142, 17828–17844. [CrossRef]
110.
Sydor, A.M.; Czymmek, K.J.; Puchner, E.M.; Mennella, V. Super-resolution microscopy: From single molecules to supramolecular
assemblies. Trends Cell Biol. 2015,25, 730–748. [CrossRef]
111.
Shivanandan, A.; Deschout, H.; Scarselli, M.; Radenovic, A. Challenges in quantitative single molecule localization microscopy.
FEBS Lett. 2014,588, 3595–3602. [CrossRef]
112.
Funatsu, T.; Harada, Y.; Tokunaga, M.; Saito, K.; Yanagida, T. Imaging of single fluorescent molecules and individual ATP
turnovers by single myosin molecules in aqueous solution. Nature 1995,374, 555–559. [CrossRef]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
