Identification of a Ryanodine Receptor in Rat Heart Mitochondria

Abstract

Recent studies have shown that, in a wide variety of cells, mitochondria respond dynamically to physiological changes in cytosolic Ca(2+) concentrations ([Ca(2+)](c)). Mitochondrial Ca(2+) uptake occurs via a ruthenium red-sensitive calcium uniporter and a rapid mode of Ca(2+) uptake. Surprisingly, the molecular identity of these Ca(2+) transport proteins is still unknown. Using electron microscopy and Western blotting, we identified a ryanodine receptor in the inner mitochondrial membrane with a molecular mass of approximately 600 kDa in mitochondria isolated from the rat heart. [(3)H]Ryanodine binds to this mitochondrial ryanodine receptor with high affinity. This binding is modulated by Ca(2+) but not caffeine and is inhibited by Mg(2+) and ruthenium red in the assay medium. In the presence of ryanodine, Ca(2+) uptake into isolated heart mitochondria is suppressed. In addition, ryanodine inhibited mitochondrial swelling induced by Ca(2+) overload. This swelling effect was not observed when Ca(2+) was applied to the cytosolic fraction containing sarcoplasmic reticulum. These results are the first to identify a mitochondrial Ca(2+) transport protein that has characteristics similar to the ryanodine receptor. This mitochondrial ryanodine receptor is likely to play an essential role in the dynamic uptake of Ca(2+) into mitochondria during Ca(2+) oscillations.

Full text

Identification of a Ryanodine Receptor in Rat Heart Mitochondria*
Received for publication, February 16, 2001, and in revised form, April 9, 2001
Published, JBC Papers in Press, April 10, 2001, DOI 10.1074/jbc.M101486200
Gisela Beutner‡, Virendra K. Sharma‡, David R. Giovannucci, David I. Yule, and Shey-Shing Sheu§
From the Department of Pharmacology and Physiology, University of Rochester, School of Medicine and Dentistry,
Rochester, New York 14642
Recent studies have shown that, in a wide variety of
cells, mitochondria respond dynamically to physiologi-
cal changes in cytosolic Ca
2ⴙ
concentrations ([Ca
2ⴙ
]
c
).
Mitochondrial Ca
2ⴙ
uptake occurs via a ruthenium red-
sensitive calcium uniporter and a rapid mode of Ca
2ⴙ
uptake. Surprisingly, the molecular identity of these
Ca
2ⴙ
transport proteins is still unknown. Using electron
microscopy and Western blotting, we identified a ryan-
odine receptor in the inner mitochondrial membrane
with a molecular mass of approximately 600 kDa in mi-
tochondria isolated from the rat heart. [
3
H]Ryanodine
binds to this mitochondrial ryanodine receptor with
high affinity. This binding is modulated by Ca
2ⴙ
but not
caffeine and is inhibited by Mg
2ⴙ
and ruthenium red in
the assay medium. In the presence of ryanodine, Ca
2ⴙ
uptake into isolated heart mitochondria is suppressed.
In addition, ryanodine inhibited mitochondrial swelling
induced by Ca
2ⴙ
overload. This swelling effect was not
observed when Ca
2ⴙ
was applied to the cytosolic frac-
tion containing sarcoplasmic reticulum. These results
are the first to identify a mitochondrial Ca
2ⴙ
transport
protein that has characteristics similar to the ryanodine
receptor. This mitochondrial ryanodine receptor is
likely to play an essential role in the dynamic uptake of
Ca
2ⴙ
into mitochondria during Ca
2ⴙ
oscillations.
Mitochondria play a central role in numerous fundamental
cellular processes ranging from the generation of ATP to the
regulation of cytosolic Ca
2⫹
homeostasis and apoptosis (1–3).
Impairment of intracellular Ca
2⫹
homeostasis and mitochon-
drial function has been implicated in the development of neu-
rodegenerative diseases, diabetes, and cardiomyopathy (4–6).
Measuring changes in mitochondrial Ca
2⫹
concentrations
([Ca
2⫹
]
m
)
1
in living cells with highly sensitive fluorescence
dyes and targeted luminescence probes has drawn new atten-
tion on the regulation of mitochondrial Ca
2⫹
homeostasis and
its biological implications (7–9). Several studies show that, in
many cell types, mitochondria respond dynamically to physio-
logical oscillations of free [Ca
2⫹
]
c
(10–14).
In cardiac muscle cells, mitochondria also respond to phys-
iological changes in [Ca
2⫹
]
c
(15). However, controversy re-
mains whether the mitochondrial Ca
2⫹
uptake mechanisms
can sequester Ca
2⫹
rapidly on a beat-to-beat basis (15). Es-
tablished Ca
2⫹
uptake mechanisms are the CaUP and rapid
mode of Ca
2⫹
uptake (1, 16). The mitochondrial CaUP is
activated by Ca
2⫹
concentrations greater than 10
␮
M, which
are usually only achieved within cytosolic microdomains (9).
In isolated liver mitochondria, rapid mode of Ca
2⫹
uptake
responds with mitochondrial Ca
2⫹
uptake in response to
Ca
2⫹
pulses of less than 300 nM(16). Although these Ca
2⫹
uptake mechanisms are kinetically and pharmacologically
well characterized, their molecular identity has yet to be
determined.
An intriguing observation is the considerable similarity in
biochemical and pharmacological properties between the mito-
chondrial CaUP and the SR-RyR (1, 17). Both, the CaUP and
the SR-RyR are activated by changes in [Ca
2⫹
]
c
and inhibited
by adenine nucleotides, Mg
2⫹
, and RR. The strikingly similar
properties of these proteins led to our hypothesis that mito-
chondria contain a RyR. Here we show with immunological,
biochemical, pharmacological, and physiological techniques
that heart mitochondria contain a functional RyR within the
IMM. This mRyR underlies fast mitochondrial Ca
2⫹
uptake
and, therefore, is uniquely positioned to regulate dynamically
Ca
2⫹
-mediated cellular processes such as ATP production dur-
ing heartbeat (18).
EXPERIMENTAL PROCEDURES
Materials—Ryanodine was purchased from Calbiochem and [
3
H]ry-
anodine from Amersham Pharmacia Biotech. Antibodies against RyR
(clone 34C), developed by J. Airey and J. Sutko, were obtained from the
Developmental Studies Hybridoma Bank maintained by the Depart-
ment of Biological Sciences of the University of Iowa (Iowa City, IA) and
from RBI/Sigma. The antibodies against the voltage-dependent anion
channel (VDAC) and the sarco- and endoplasmic reticulum Ca
2⫹
-
ATPase (SERCA) were obtained from Calbiochem and Santa Cruz,
respectively. All other chemicals were purchased from Sigma unless
noted.
Isolation of Rat Heart Mitochondria—Heart mitochondria were iso-
lated in isotonic ice-cold mannitol/sucrose buffer (M/S buffer; in mM: 225
mannitol, 10 sucrose, 0.5 EGTA, 1 glutathione; 10 HEPES, pH 7.4) by
differential centrifugation and subsequent purification on a Percoll
gradient (19). The final two washes were done in EGTA-free buffer. The
isolated mitochondria were stored on ice and used for experiments for
up to 4 h after finishing the isolation procedure. Experiments were done
with mitochondria having intact RR- and carbonyl cyanide 3-chlorophe-
nylhydrazone-sensitive Ca
2⫹
uptake mechanisms, as measured with a
Ca
2⫹
-sensitive microelectrode.
Preparation of Mitochondrial Subfractions—Mitochondrial subfrac-
tions were prepared as described previously (20). Briefly, isolated mi-
* This work was supported by National Institutes of Health Grant
HL-33333, National Institutes of Health Grant DK-54568, American
Heart Association Grant 9920244T, and American Heart Association
Grant 0050839T. The costs of publication of this article were defrayed
in part by the payment of page charges. This article must therefore be
hereby marked “advertisement” in accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
‡ These authors contributed equally to this work.
§ To whom correspondence should be addressed: Dept. of Pharmacology
and Physiology, Box 711, University of Rochester, School of Medicine and
Dentistry, 601 Elmwood Ave., Rochester, NY 14642. Tel.: 716-275-3381;
Fax: 716-273-2652; E-mail: sheyshing_sheu@urmc.rochester.edu.
1
The abbreviations used are: [Ca
2⫹
]
m
, mitochondrial Ca
2⫹
concen-
tration; [Ca
2⫹
]
c
, cytosolic Ca
2⫹
concentration; CaUP, mitochondrial cal-
cium uniporter; RR, ruthenium red; SR, sarcoplasmic reticulum; RyR,
ryanodine receptor, SR-RyR, sarcoplasmic ryanodine receptor; mRyR,
mitochondrial ryanodine receptor; SERCA, sarco- and endoplasmic re-
ticulum Ca
2⫹
-ATPase; CPA, cyclopiazonic acid; IMM, inner mitochon-
drial membrane; OMM, outer mitochondrial membrane; VDAC, volt-
age-gated anion channel; MOPS, 4-morpholinepropanesulfonic acid;
SDH, succinate dehydrogenase; M/S, mannitol/sucrose; OGB-5N, Ore-
gon Green Bapta 5N.
THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 24, Issue of June 15, pp. 21482–21488, 2001
© 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
This paper is available on line at http://www.jbc.org21482
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

tochondria from 2 rat hearts were osmotically shocked first in 10 mM
Na
2
HPO
4
/NaH
2
PO
4
for 20 min on ice followed by the addition of sucrose
(20% final concentration). Mitochondrial membranes were disrupted by
sonication (two times for 30 s), and eventually intact mitochondria were
removed by centrifugation at 7,000 ⫻g. The supernatant was trans-
ferred as a final layer onto a continuous sucrose gradient (from 60% to
30% in 10 mMHEPES, pH 7.4, plus protease inhibitor mixture (Roche
Molecular Biochemicals) and centrifuged for at least8hat70,000 ⫻g
to separate mitochondrial membrane vesicles. The mitochondrial sub-
fractions were tested for specific marker proteins (succinate dehydro-
genase (SDH) for IMM, creatine kinase for contact sites between IMM
and outer mitochondrial membrane (OMM), and the VDAC for OMM).
The characterized fractions were diluted 2-fold with M/S buffer, sup-
plemented with protease inhibitor mixture (Roche), and centrifuged for
90 min at 300,000 ⫻gto sediment the mRyR. The pellet was resus-
pended in a small volume of M/S buffer and stored in liquid nitrogen
until needed.
Immunogold Labeling Methods—For immunogold labeling of RyR,
isolated heart mitochondria were embedded in Lowicryl mixture at
⫺20 °C. Gold labeling of ultrathin sections of embedded mitochondria
against RyR was performed as described previously (21, 22).
Denaturing SDS-Gel Electrophoresis—For the immunological detec-
tion of the RyR and SERCA, 100
␮
g of protein was loaded on a 5%
SDS-polyacrylamide gel. The separated proteins were transferred onto
a nitrocellulose membrane for 90 min at 100 V. For VDAC, 50
␮
gof
protein was loaded onto a 12% SDS-polyacrylamide gel and transferred
for 45 min at 100 V onto a nitrocellulose membrane. Western blots were
performed using the Aurora chemiluminescence assay (ICN) with an
alkaline phosphatase-linked secondary antibody.
[
3
H]Ryanodine Binding—For binding assays, 100
␮
g of mitochon-
drial protein were incubated with different concentrations of [
3
H]ryano-
dine in 0.5 ml of binding buffer (in mM: 170 KCl, 0.02 CaCl
2
, 10 MOPS,
pH 7.0) for 16 h at 25 °C. For nonspecific binding, Ca
2⫹
was replaced by
6mMEGTA. At the end of the incubation time, the reaction mixture was
filtered under reduced pressure through glass fiber filters (Whatman)
and washed with ice-cold buffer (170 mMKCl, 10 mMMOPS).
FIG.1.Immunogold labeling and Western blot analysis of a RyR in the IMM. A, immunogold labeling of RyR (arrows) in the IMM of
isolated adult rat heart mitochondria (original magnification, ⫻50,000). Mitochondria were isolated as described under “Experimental Procedures.”
The final sediment containing the purified mitochondria was fixed in paraform aldehyde and embedded in low temperature resin prior to
immunogold labeling. B, as a negative control the gold-labeled secondary antibody was applied in absence of the RyR antibody, yielding no
significance detection of RyR protein. C, Western blot analysis against RyR of mitochondrial subfractions from isolated and osmotically shocked
heart mitochondria. Characterized subfractions were centrifuged for 90 min at 300,000 ⫻gto sediment the RyR. C, SR containing cytosol; M,
purified intact mitochondria; CS, contact sites; anti-RyR, labeling against RyR; anti-SERCA and anti-VDAC, Western blots against SERCA and
VDAC as negative controls for contamination of the IMM with cytosolic membrane fragments and OMM. Arrows indicate the position and the
molecular weight of the marker molecules. D, cytosol and mitochondria of four independent performed preparations were probed against SERCA.
Lanes 1C– 4C represent the cytosolic fractions obtained after the first centrifugation of the homogenate at 12,000 ⫻g.Lanes 1M–4M represent the
corresponding mitochondria after Percoll purification. C, cytosol; M, mitochondria.
FIG.2. [
3
H]Ryanodine binding to
isolated mitochondria in the pres-
ence or absence of modulators of the
RyR. All assays were done in triplicate,
and all figures represent a mean of at
least three independent experiments. A,
ryanodine binding to isolated heart mito-
chondria. Data were fit for a single class
of binding sites by Scatchard plot (inset).
B
max
was 398 ⫾12 fmol/mg of protein,
and the K
d
was 9.52 nM(r⫽0.97 for
linear regression). B, bound [
3
H]ryano-
dine; F, free [
3
H]ryanodine. B,Ca
2⫹
de-
pendence of [
3
H]ryanodine binding. 100
␮
g of mitochondrial protein was incu-
bated with various concentration of Ca
2⫹
in presence of 9 nM[
3
H]ryanodine. C,in
-
hibition of [
3
H]ryanodine binding with
different concentrations of MgCl
2
.D,in
-
hibition of [
3
H]ryanodine binding by RR.
Mitochondrial Ryanodine Receptor 21483
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

Measurements of Mitochondrial Net Calcium Uptake in Isolated Rat
Heart Mitochondria with a Ca
2⫹
-selective Electrode—Mitochondria (1
mg) were diluted in 1 ml of M/S buffer containing 10
␮
MEGTA, and the
free Ca
2⫹
concentration was calculated to be 60
␮
M. Net Ca
2⫹
uptake
was measured by monitoring changes in external Ca
2⫹
concentration in
the reaction medium using a Ca
2⫹
-selective electrode (Microelectrode
Inc., Bedford, NH).
Rhod-2 Loading of Isolated Mitochondria—Mitochondria were sus-
pended in M/S buffer containing 10
␮
MEGTA (to chelate Ca
2⫹
impuri-
ties in distilled water) to a final concentration of ⬃10 mg of protein/ml.
After incubation in 5
␮
Mrhod-2 AM (Molecular Probes) for 10 min at
room temperature, mitochondria were washed three times to remove
external rhod-2. Mitochondria were used for experiments 30 min after
finishing the loading protocol to ensure cleavage of the rhod-2 AM into
the membrane-impermeable rhod-2 free acid.
Flash Photolysis of Caged Ca
2⫹
and Measurement of [Ca
2⫹
]
m
Dynam-
ics—For [Ca
2⫹
]
m
measurement, isolated, rhod-2-loaded mitochondria in
a 0.5-
␮
l droplet were adhered to a glass coverslip and mounted on the
stage of an Eclipse TE200 microscope equipped with a SuperFluor 40X
(1.3 numeric aperture oil immersion objective, Nikon). This solution
was mixed with 5
␮
l of “caged-Ca
2⫹
” solution containing (in mM: 130
KCl, 10 HEPES, 10 succinate, 10 o-nitrophenyl EGTA, 5 CaCl
2
,2
Mg-ATP, 1.2 MgCl
2
, pH 7.2). For the pulsatile photolytic release of Ca
2⫹
in the external media, a 1-ms flash of ⬃250 J of UV light (340–360 nm)
was produced at 0.2 Hz. Determination of relative changes in [Ca
2⫹
]
m
was performed using digital imaging microscopy with a monochrom-
eter-based system and high speed CCD camera (T.I.L.L.-Photonics).
Rhod-2 labeled mitochondria were excited at 530 ⫾15 nm and fluores-
cence emission collected using a 565 nm long pass filter (Chroma).
Images were acquired at a rate of 2–4 Hz without binning and dis-
played as ⌬F/F
0
, where ⌬F/F
0
⫽100[(F⫺F
0
)/F
0
], and Fis the recorded
fluorescence and F
0
was obtained from the average of 15 sequential
frames prior to uncaging. Estimation of the evoked changes in external
Ca
2⫹
([Ca
2⫹
]
ext
) was performed using the low affinity Ca
2⫹
indicator,
Oregon Green Bapta 5N (OGB-5N; Molecular Probes). Using a range of
standard solutions buffered at set Ca
2⫹
concentrations (Molecular
Probes) and the equation [Ca
2⫹
]
ext
⫽K
d
[(F⫺F
min
)/(F
max
⫺F)], the
dissociation constant (K
d
), F
min
, and F
max
were determined to be 21
␮
M,
57, and 1050, respectively. For measurement of [Ca
2⫹
]
ext
,a5-
␮
l droplet
containing 50
␮
MOGB-5N was excited at 488 ⫾15 nm and fluorescence
emission collected through a 525 ⫾25 nm band pass filter (Chroma).
Mitochondrial Swelling—Mitochondrial swelling was induced as de-
scribed previously (23). Briefly, isolated heart mitochondria (1 mg of
protein) were diluted in 1 ml of modified M/S buffer (in mM: 120 KCl, 65
mannitol, 30 sucrose, 10 succinate, 5 Na
2
HPO
4
/NaH
2
PO
4
, 10 HEPES,
pH 7.2). The absorbance was recorded with a spectrophotometer (Spec-
tronic) at 540 nm for 2–3 min to obtain a stable base line, followed by
the addition of Ca
2⫹
to induce mitochondrial swelling.
RESULTS
Immunological Detection of RyR in Isolated Heart Mitochon-
dria—Electron microscopic analysis revealed that ⬃70% of iso-
lated adult rat heart mitochondria, treated with gold-labeled
antibodies against SR-RyR, were labeled with 1–4 gold parti-
cles (Fig. 1A). The majority of the gold particles were found in
the cristae membrane of IMM (73.3% of total), although some
labeling was detected in the peripheral IMM. In contrast, no
significant labeling was found in the OMM and extramitochon-
drial membranes (3.5% of total counted gold particles). In ab-
sence of the RyR antibody, only 5% of the mitochondria showed
labeling with a maximum of 1 gold particle (Fig. 1B).
Using Western blot analysis, we confirmed the specific de-
tection of RyR-like protein in the IMM from osmotically
shocked rat heart mitochondria (Fig. 1C). Western blots per-
formed on the cytosolic and mitochondrial subfractions demon-
strated immunoreactivity against RyR in all fractions except
the OMM (Fig. 1C,anti-RyR). A positive signal against RyR
protein in the IMM was obtained in all preparations (n⫽6). In
all tested fractions, the RyR antibody labeled a protein of ⬃600
kDa. The purity of the IMM fraction was verified by the pres-
ence of SDH activity, an enzyme localized exclusively in the
IMM. Furthermore, this fraction was free of SERCA and VDAC
(Fig. 1C,anti-SERCA,anti-VDAC) in all preparations, indicat-
ing that the IMM fraction is devoid of any significant contam-
ination from SR and OMM fragments. In addition, SERCA
pump proteins were not detected in four independently isolated
mitochondrial preparations (Fig. 1D), indicating that even the
intact mitochondria had minimal SR contamination.
An extensive effort to exclude the possibility that the mRyR
FIG.3.Ryanodine inhibits mitochondrial Ca
2ⴙ
uptake. Mitochondria were incubated with ryanodine, CPA, or dantrolene and added to M/S
buffer containing 60
␮
Mfree Ca
2⫹
, and the external Ca
2⫹
concentration was measured with a Ca
2⫹
-sensitive microelectrode. A, this panel shows
a representative experiment of ryanodine-inhibited mitochondrial Ca
2⫹
uptake. Mitochondria were preincubated with the indicated concentration
of ryanodine for 15 min at room temperature, and the changes of the Ca
2⫹
concentration in the M/S buffer were measured. B, after incubation with
10
␮
Mdantrolene for 10 min at room temperature, mitochondrial Ca
2⫹
uptake was inhibited by 55%. C, pretreatment of mitochondria with 30
␮
M
CPA for 10 min at room temperature did not affect mitochondrial Ca
2⫹
uptake. D, inhibition of mitochondrial Ca
2⫹
uptake in the presence or
absence of modulators of the RyR. Maximal Ca
2⫹
uptake in untreated heart mitochondria was set to be 100%. Dig, removal of OMM with 25
␮
g
of digitonin/mg of mitochondrial protein prior measuring Ca
2⫹
uptake; CPA,30
␮
MCPA; Rya, 100
␮
Mryanodine; Dig⫹Rya, digitonin treatment
plus incubation with 100
␮
Mryanodine; Dan,10
␮
Mdantrolene. Each bar represents the averaged value of at least three independent experiments,
and values are given ⫾S.E.
Mitochondrial Ryanodine Receptor21484
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

was not caused by contamination with SR-RyR was deemed
essential, because mitochondria are located in close proximity
to the SR in cardiac muscle cells (14, 24). The chosen marker
proteins, such as SERCA, VDAC, and SDH, are specific for
their intracellular location. Labeling of the RyR in the purified
IMM with a specific antibody is consistent with the result
obtained by electron microscopy, showing that the mRyR is
localized within the IMM (Fig. 1A). The molecular weight of the
detected RyR proteins in the IMM and the cross-reactivity with
the used SR-RyR specific antibody suggests that the mRyR is
structurally homologous to the SR-RyR.
[
3
H]Ryanodine Binding to Isolated Heart Mitochondria—
Various physiological and pharmacological effectors including
Ca
2⫹
, caffeine, Mg
2⫹
, and RR modulate ryanodine binding to
the SR-RyR (25, 26). To characterize the pharmacological prop-
erties of mRyR, we studied [
3
H]ryanodine binding to heart
mitochondria in presence of these modulators. [
3
H]Ryanodine
bound to isolated heart mitochondria in presence of 20
␮
MCa
2⫹
with an apparent affinity (K
d
)of9.8⫾2.1 nM(n⫽3, in
triplicate for all binding experiments; Fig. 2A). Depending on
the experimental conditions employed, it has been reported
that SR-RyR exhibited an apparent K
d
between 2 and 200 nM
for ryanodine (27, 28). The maximal density of mRyR binding
sites (B
max
) was 398.4 ⫾12 fmol/mg of protein (n⫽3; Scat-
chard plot in Fig. 2A), which is ⬃10 times less than that
described for [
3
H]ryanodine binding to purified SR membranes
under similar experimental conditions (28, 29). To confirm that
[
3
H]ryanodine binding was not due to contamination by SR,
binding was performed with equal amounts of proteins from
intact isolated mitochondria and mitochondrial subfractions.
The results showed that ⬃90 ⫾5% (n⫽3) of the total [
3
H]ry-
anodine bound to mitochondria was due to binding to the IMM
and not to other fractions.
The amount of [
3
H]ryanodine binding to mitochondria as a
function of free Ca
2⫹
concentration in the assay media was
biphasic. Binding increased at pCa between 5 and 7 and de-
creased at pCa between 3 and 4 with maximal binding at pCa
5.3 (Fig. 2B). Surprisingly, unlike the cardiac SR-RyR, caffeine
did not enhance mitochondrial ryanodine binding (n⫽3). Caf-
feine-insensitive RyR, however, have been described in canine
salivary glands (30) and in Jurkat cells (31).
Mg
2⫹
is known to decrease the B
max
for [
3
H]ryanodine bind-
ing and to inhibit SR Ca
2⫹
efflux in skeletal and cardiac muscle
(26, 32, 33). This is consistent with single-channel studies,
FIG.4.Ca
2ⴙ
dynamics of individual
or small clusters of isolated, rhod-2-
loaded heart mitochondria induced
by photolytic release of caged Ca
2ⴙ
.
A, raw fluorescence image and the corre-
sponding pseudocolored digital images of
isolated mitochondria evoked by repeti-
tive flash photolysis of 10 o-nitrophenyl
EGTA, 50% bound with Ca
2⫹
, in the ex-
ternal solution. Each image displayed is
the first frame (320 ms) following the
flash lamp artifact and ⌬Freflect rises in
[Ca
2⫹
]
m
.B, flash application (dots) pro-
duced micromolar Ca
2⫹
transients in the
extramitochondrial solution as estimated
with OGB-5N (K
d
⫽21
␮
M). C, represent-
ative fluorescence recordings showing the
flash-induced changes in [Ca
2⫹
]
m
as
shown in Afor both control (black trace)
and 100
␮
Mryanodine-treated (red trace)
mitochondria. Breaks in records indicate
removal of flash lamp artifacts. D, aver-
aged data comparing the cumulative ⌬F
produced by each flash for matched con-
trol and ryanodine-treated mitochondria.
⌬Fis the difference of measured fluores-
cence directly prior and after flash appli-
cation. Data from control and ryanodine-
treated mitochondria were normalized to
the maximal cumulative ⌬Ffor the con-
trol response of each experiment. Data
are expressed as mean ⫾S.E., and pval-
ues ⬍0.05 by paired ttest were considered
statistically significant (asterisks). Data
are normalized to the maximal ⌬Ffor the
control response of each experiment.
Mitochondrial Ryanodine Receptor 21485
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

where millimolar concentrations of Mg
2⫹
reduced the open
probability of the RyR and maintained the channel in a closed
state (25, 26, 32). Accordingly, we studied the effects of Mg
2⫹
on
[
3
H]ryanodine binding in isolated heart mitochondria and ob-
served a 50% inhibition in presence of 0.33 mMMg
2⫹
(n⫽3,
Fig. 2C). Under comparable experimental conditions, it has
been shown that up to 1 mMMg
2⫹
did not have any inhibitory
effects on ryanodine binding in cardiac SR-RyR (34–36). This
suggests that, in cardiac muscle cells, the mRyR is more sen-
sitive to Mg
2⫹
inhibition of [
3
H]ryanodine binding than the
SR-RyR.
RR inhibits the release of Ca
2⫹
from the SR in skeletal and
cardiac muscles by decreasing the open probability of the RyR
(26, 37). After RR treatment, [
3
H]ryanodine binding to isolated
rat heart mitochondria was strongly inhibited with IC
50
⫽105
nM(n⫽3, Fig. 2D). This suppression of mitochondrial [
3
H]ry-
anodine binding by RR is much more potent than that observed
in cardiac SR-RyR, which had IC
50
values between 290 and
1,000 nM(34, 35). Consistent with the binding data, we have
shown that RR (1–5
␮
M) blocks mitochondrial Ca
2⫹
uptake
without much effect on SR Ca
2⫹
release in chemically skinned
cardiac myocytes (14). These results indicate that significant
differences exist in the potency of RR in inhibiting mRyR and
SR-RyR.
The binding data provide pharmacological evidence of the
existence of mRyR in the IMM. They also show that there are
distinct differences between mRyR and SR-RyR with respect to
their abundance and their sensitivities to caffeine, Mg
2⫹
, and
RR. Therefore, the mRyR and SR-RyR would operate at differ-
ent capacities under similar conditions and could be regulated
and modulated differentially.
Ryanodine Inhibition of Mitochondrial Ca
2⫹
Uptake—We
next investigated functional aspects of mRyR in the sequestra-
tion of Ca
2⫹
using two different methods. Adding mitochondria
to a buffer containing 60
␮
Mfree Ca
2⫹
caused a significant
mitochondrial Ca
2⫹
uptake, as measured by a decrease in the
extramitochondrial Ca
2⫹
concentration with a Ca
2⫹
-selective
microelectrode (Fig. 3A,control). In the presence of 100 or 10
␮
Mryanodine, mitochondrial Ca
2⫹
uptake was suppressed by
60 ⫾2.7% or 41.2 ⫾1.9%, respectively (Fig. 3A,n⫽15).
Removal of the OMM with digitonin (25
␮
g/mg of mitochondrial
protein for 30 s) to minimize possible SR-contamination had no
significant effect on mitochondrial Ca
2⫹
uptake in the presence
or absence of ryanodine (Fig. 3D).
The same inhibitory effect in mitochondrial Ca
2⫹
uptake was
observed with dantrolene, a compound that has been shown to
inhibit the skeletal muscle SR-RyR and therefore Ca
2⫹
release
(17). The effect of dantrolene on the cardiac SR-RyR is still
controversial (38–40). However, incubation of isolated mito-
chondria with 10
␮
Mdantrolene decreased mitochondrial Ca
2⫹
uptake by 55.9 ⫾7.8% (Fig. 3, Band D;n⫽5). Finally, the
presence of 30
␮
Mcyclopiazonic acid (CPA), an inhibitor of
SERCA, did not alter mitochondrial Ca
2⫹
uptake compared
with untreated mitochondria (96 ⫾2.4%, n⫽3), indicating
little contamination of the SR-RyR (Fig. 3, Cand D).
The relatively slow response time of Ca
2⫹
microelectrodes
made them unable to follow the rapid mitochondrial Ca
2⫹
uptake during Ca
2⫹
pulses. To investigate whether the mRyR
contributes to rapid Ca
2⫹
uptake, we measured the Ca
2⫹
re-
sponses of single or small clusters of isolated, rhod-2-loaded
mitochondria using high speed digital imaging in combination
with pulsatile flash photolysis of caged Ca
2⫹
. As shown in Fig.
4(Aand C), Ca
2⫹
uptake was stimulated in rhod-2-loaded
isolated heart mitochondria after flash photolysis of caged
Ca
2⫹
in the external solution. The slow decay in the fluores-
cence may result from absence of Na
⫹
in the solution that
inhibited Na
⫹
-dependent Ca
2⫹
efflux. In Fig. 4B, a low affinity
Ca
2⫹
indicator Oregon Green Bapta-5N was used in separate
experiments to confirm that pulsatile flash photolysis resulted
FIG.5.Ryanodine inhibits Ca
2ⴙ
-induced mitochondrial swelling. In all experiments, 1 mg of mitochondrial protein were dissolved in
modified M/S buffer and swelling was induced by the addition of 100
␮
MCa
2⫹
.A, this figure shows a representative concentration-response curve
of ryanodine-inhibited mitochondrial swelling. Heart mitochondria were preincubated for 15 min at room temperature with the indicated
concentration of ryanodine. Ca
2⫹
was added at time 0 (arrow), and the change in absorbance at 540 nm was followed. Data are expressed as the
ratio of the absorbance at any given time (A) divided by the base-line absorbance at time 0 (Ao). B, averaged changes of the inhibitory effect of
ryanodine (Rya) on mitochondrial swelling. Each bar represents at least six independently performed experiments. Data are expressed as mean
values ⫾S.E., and pvalues ⬍0.01 by paired ttest were considered statistically significant (asterisks).
FIG.6. Mitochondrial swelling in the presence or absence of
modulators of the RyR or SERCA. Each bar represents the averaged
decrease in absorbance of at least three independently performed ex-
periments. Experiments were performed as in Fig. 5. Ca
2⫹
, 100
␮
M
Ca
2⫹
;CPA, incubation with 50
␮
MCPA for 10 min at room temperature
prior the addition of Ca
2⫹
;Dan, preincubation with 10
␮
Mdantrolene
for 10 min; Rya,20
␮
Mryanodine. C, SR containing cytosolic fraction.
Data are expressed as mean values ⫾S.E.
Mitochondrial Ryanodine Receptor21486
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

in elevations in extramitochondrial Ca
2⫹
that reached to their
peak value within sampling time of one frame (250–500 ms).
After pre-incubating mitochondria with 100
␮
Mryanodine, a
significant component of the evoked change in mitochondrial
fluorescence was suppressed (Fig. 4C). The cumulative change
in fluorescence (⌬F) evoked by each flash was 0.14 ⫾0.02,
0.34 ⫾0.05, 0.58 ⫾0.08, 0.85 ⫾0.04, and 1 for control; and
0.05 ⫾0.02, 0.12 ⫾0.04, 0.25 ⫾0.07, 0.40 ⫾0.09, and 0.60 ⫾
0.13 for ryanodine-treated mitochondria (n⫽6, Fig. 4D). The
magnitude of [Ca
2⫹
]
m
at higher number of flashes in the con-
trol solution could be an underestimation, due to the saturation
of rhod-2 (K
d
⫽0.6–0.8
␮
M). This could account for the de-
crease in the slope of control curve between flash number 4 and
5 in Fig. 4D. Finally, this mitochondrial Ca
2⫹
uptake was not
due to release of Ca
2⫹
from SR contamination, because repet-
itive UV flashes applied to rhod-2-loaded mitochondria in a
droplet containing 100
␮
MD-myo-inositol 1,4,5-trisphosphate,
P
4(5)
-1-(2-nitrophenyl)ethyl ester, a concentration of caged ino-
sitol 1,4,5-trisphosphate shown to evoke robust Ca
2⫹
release
from reticular stores (41), induced no measurable ⌬F.
The ability of mitochondria to sequester significant amount
of Ca
2⫹
through a ryanodine inhibitory pathway suggests that
the mRyR could play an important role in buffering high con-
centrations of [Ca
2⫹
]
c
. Consistent with our experiments, a de-
crease in [Ca
2⫹
]
m
in the presence of ryanodine has been ob-
served in A10 cells when perfused with more then 1
␮
MCa
2⫹
(42). Moreover, the ability of mitochondria to respond instan-
taneously to fast Ca
2⫹
pulses through a ryanodine inhibitable
pathway suggests that the mRyR may be responsible for Ca
2⫹
sequestration during heartbeats.
Inhibition of Mitochondrial Swelling by Ryanodine—Exces-
sive accumulation of Ca
2⫹
in the mitochondrial matrix depo-
larizes the mitochondrial membrane potential and disrupts
fundamental mitochondrial functions like oxidative phospho-
rylation and ATP production, which results in opening of the
mitochondrial permeability transition pore in isolated heart
and liver mitochondria (23, 43). This is accompanied by mito-
chondrial swelling and can be measured by the decrease of
absorbance at 540 nm. Inducing mitochondrial swelling in iso-
lated intact heart mitochondria with 100
␮
MCa
2⫹
led to a
decrease in absorbance of 19.5 ⫾3.4% (n⫽8). Upon preincu-
bation with 2–20
␮
Mryanodine, we observed a concentration-
dependent inhibition of Ca
2⫹
-induced mitochondrial swelling
(Fig. 5A). Interestingly, less then 2
␮
Mryanodine significantly
enhanced mitochondrial swelling in all experiments (Fig. 5, A
and B;n⫽6). CPA had no effect on the inhibition of mitochon-
drial swelling after treatment with 20
␮
Mryanodine (Fig. 6). In
addition, the concentration dependence in ryanodine-mediated
inhibition of mitochondrial swelling by ryanodine was also not
altered by CPA treatment (data not shown). An accelerated
mitochondrial swelling in the presence of less than 2
␮
Mryan-
odine or the inhibition of swelling by higher ryanodine concen-
trations is in agreement with the literature, where low ryano-
dine concentrations switch the SR-RyR into an open state and
higher concentrations keep the Ca
2⫹
channel in the RyR closed
(17, 44).
Dantrolene (10
␮
M) was as effective as 20
␮
Mryanodine in
blocking mitochondrial swelling (Fig. 6). Finally, the SR-con-
taining cytosolic fraction itself revealed no significant changes
in absorbance after the addition of Ca
2⫹
(Fig. 6). In these
experiments the same amount of cytosolic protein and experi-
mental protocol was used as that described for mitochondrial
swelling. Therefore, the possibility that the effect of mitochon-
drial swelling could be mimicked by cytosolic components like
the SR was excluded.
Mitochondrial swelling due to the opening of the mitochon-
drial permeability transition pore has been implicated in trig-
gering apoptosis and necrosis of several cell types (45). The
ability of ryanodine to prevent such mitochondrial swelling by
blocking influx of Ca
2⫹
may provide some insights into the
development of novel therapeutic agents for mitochondria-me-
diated cell injury and death.
DISCUSSION
The present study demonstrates that mitochondria contain a
RyR within the IMM, which shares several, similar biochemi-
cal, pharmacological, and physiological properties with both
the SR-RyR and the CaUP. Based on the results presented
here, a contamination of the purified mitochondria by SR-RyR
can be excluded for several reasons. 1) Preparations of isolated
mitochondria were free of detectable amounts of SERCA pro-
tein. 2) The separated OMM fraction tested positive for VDAC
but negative for SDH activity. 3) The separated IMM tested
positive for SDH but negative for the OMM protein VDAC and
the SR protein SERCA. 4) CPA had no effect on mitochondrial
Ca
2⫹
uptake and mitochondrial swelling. 5) [
3
H]Ryanodine
binding was more sensitive to Mg
2⫹
and RR inhibition. 6)
Caffeine had no effect on ryanodine binding.
The localization of mRyR in the IMM may raise the question
of whether the RyR, like other intracellular membrane pro-
teins, could have multiple intracellular locations. It has been
shown that the inositol 1,4,5-triphosphate receptor is localized
in the nucleus (46, 47) and the plasma membrane (48) in
addition to the ER and SR. Likewise, proteins of the cell death-
regulating Bcl-2 family have been shown to be localized in the
OMM, the nucleus membrane, and the ER membranes (49). A
FIG.7.A model of Ca
2ⴙ
dynamics in
cardiac myocytes. Mitochondria sense
microdomains with high Ca
2⫹
following
activation of voltage-gated Ca
2⫹
channels
(VGCC) and Ca
2⫹
release from intracellu-
lar stores. These raises could trigger fast
mitochondrial Ca
2⫹
uptake via the mRyR
and/or CaUP, which may ultimately af-
fect physiological and pathological
processes.
Mitochondrial Ryanodine Receptor 21487
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

Na
⫹
/Ca
2⫹
exchanger is localized in the IMM but also in the
plasma membrane (1).
An increase in [Ca
2⫹
]
c
activates the RyR in the SR or ER to
release Ca
2⫹
from intracellular stores due to an outwardly
directed Ca
2⫹
electrochemical gradient. Conversely, the in-
wardly directed Ca
2⫹
electrochemical gradient in mitochondria
could result in Ca
2⫹
-induced Ca
2⫹
uptake following the activa-
tion of mRyR. Finally, because of the pharmacological similar-
ities to the CaUP, it is tempting to speculate that the mRyR is
the CaUP.
The proximity between the mitochondria and other Ca
2⫹
transport proteins, such as SR- or ER-RyR and L-type Ca
2⫹
channels, would allow mitochondria to sense microdomains
with Ca
2⫹
concentrations sufficient to open the mRyR (Fig. 7).
Ca
2⫹
released by SR-RyR has been shown to activate mitochon-
drial Ca
2⫹
uptake and subsequently causes an increase in
electron transport chain activity and NAD(P)H fluorescence
(50). The rapid influx of Ca
2⫹
into the mitochondria would also
contribute to local cytosolic Ca
2⫹
buffering, thus regulating
biological processes such as ATP production and intracellular
Ca
2⫹
signaling. Conversely, dysregulation in a mRyR may play
an important role in the development of disease states such as
heart failure and neurodegeneration.
Acknowledgments—The antibody against RyR developed by J. Airey
and J. Sutko was obtained from the Developmental Studies Hybridoma
Bank maintained by the Department of Biological Sciences of the Uni-
versity of Iowa (Iowa City, IA). Electron microscopy was done by Karen
de Mesy Jensen (Department of Pathology, University of Rochester,
Rochester, NY). We thank Drs. R. T. Dirksen, P. M. Hinkle, C. Franzini-
Armstrong, current and past members of the Sheu laboratory for help-
ful comments, critical reading, and discussions on the manuscript.
REFERENCES
1. Gunter, T. E., Gunter, K. K., Sheu, S-S., and Gavin, C. E. (1994) Am. J.
Physiol. 267, C313–C339
2. Duchen, M. R. (2000) J. Physiol. (London)529, 57–68
3. Bernardi, P., Petronilli, V., Di Lisa, F., and Forte, M. (2001) Trends Biochem.
Sci. 26, 112–117
4. Nishikawa, T., Edelstein, D., Du, X. L., Yamagishi, S., Matsumura, T.,
Kaneda, Y., Yorek, M. A., Beebe, D., Oates, P. J., Hammes, H. P., Giardino,
I., and Brownlee, M. (2000) Nature 404, 787–790
5. Beal, M. F. (1996) Curr. Opin. Neurobiol. 6, 661–666
6. Wallace, D. C. (1999) Science 283, 1482–1488
7. Babcock, D. F., Herrington, J., Goodwin, P. C., Park, Y. B., and Hille, B. (1997)
J. Cell Biol. 136, 833–844
8. Sheu, S.-S., and Jou, M. J. (1994) J. Bioenerg. Biomembr. 26, 487–493
9. Rizzuto, R., Brini, M., Murgia, M., and Pozzan, T. (1993) Science 262, 744–747
10. Simpson, P. B., and Russell, J. T. (1996) J. Biol. Chem. 271, 33493–33501
11. Hajnoczky, G., Robb-Gaspers, L. D., Seitz, M. B., and Thomas, A. P. (1995) Cell
82, 415–424
12. Chacon, E., Ohata, H., Harper, I. S., Trollinger, D. R., Herman, B., and
Lemasters, J. J. (1996) FEBS Lett. 382, 31–36
13. Duchen, M. R., Leyssens, A., and Crompton, M. (1998) J. Cell Biol. 142,
975–988
14. Sharma, V., Ramesh, V., Franzini-Armstrong, C., and Sheu, S.-S. (2000)
J. Bioenerg. Biomembr. 32, 97–104
15. Hueser, J., Blatter, L. A., and Sheu, S.-S. (2000) J. Bioenerg. Biomembr. 32,
27–33
16. Sparagna, G. C., Gunter, K. K., Sheu, S.-S., and Gunter, T. E. (1995) J. Biol.
Chem. 270, 27510–27515
17. Franzini-Armstrong, C., and Protasi, F. (1997) Physiol. Rev. 77, 699–729
18. McCormack, J. G., and Denton, R. M. (1993) Biochem. Soc. Trans. 21, 793–799
19. Rehncrona, S., Mela, L., and Siesjo, B. K. (1979) Stroke 10, 437–446
20. Ohlendieck, K., Riesinger, I., Adams, V., Krause, J., and Brdiczka, D. (1986)
Biochim. Biophys. Acta 860, 672–689
21. Herrera, G. A. (1989) Ultrastruct. Pathol. 13, 485–499
22. de Mesy Jensen, K. L., and di Sant’Agnese, P. A. (1992) Ultrastruct. Pathol. 16,
51–59
23. Petronilli, V., Cola, C., Massari, S., Colonna, R., and Bernardi, P. (1993)
J. Biol. Chem. 268, 21939–21945
24. Ogata, T., and Yamasaki, Y. (1997) Anat. Rec. 248, 214–223
25. Pessah, I. N., Stambuk, R. A., and Casida, J. E. (1987) Mol. Pharmacol. 31,
232–238
26. Meissner, G., Darling, E., and Eveleth, J. (1986) Biochemistry 25, 236–244
27. Inui, M., Saito, A., and Fleischer, S. (1987) J. Biol. Chem. 262, 15637–15642
28. Lindsay, A. R., and Williams, A. J. (1991) Biochim. Biophys. Acta 1064,
89–102
29. Kijima, Y., Saito, A., Jetton, T. L., Magnuson, M. A., and Fleischer, S. (1993)
J. Biol. Chem. 268, 3499–3506
30. Yamaki, H., Morita, K., Kitayama, S., Imai, Y., Itadani, K., Akagawa, Y., and
Dohi, T. (1998) J. Dent. Res. 77, 1807–1816
31. Hakamata, Y., Nishimura, S., Nakai, J., Nakashima, Y., Kita, T., and Imoto,
K. (1994) FEBS Lett. 352, 206–210
32. Meissner, G., and Henderson, J. S. (1987) J. Biol. Chem. 262, 3065–3073
33. Liu, W., Pasek, D. A., and Meissner, G. (1998) Am. J. Physiol. 274, C120–V128
34. Pessah, I. N., Waterhouse, A. L., and Casida, J. E. (1985) Biochem. Biophys.
Res. Commun. 128, 449–456
35. Holmberg, S. R., and Williams, A. J. (1990) Biochim. Biophys. Acta 1022,
187–193
36. Zimanyi, I., and Pessah, I. N. (1991) J. Pharmacol. Exp. Ther. 256, 938–946
37. Xu, L., Tripathy, A., Pasek, D. A., and Meissner, G. (1999) J. Biol. Chem. 274,
32680–32691
38. Fruen, B. R., Mickelson, J. R., and Louis, C. F. (1997) J. Biol. Chem. 272,
26965–26971
39. Meissner, A., Min, J. Y., Haake, N., Hirt, S., and Simon, R. (1999) Eur. J. Heart
Fail. 1, 177–186
40. Fratea, S., Langeron, O., Lecarpentier, Y., Coriat, P., and Riou, B. (1997)
Anesthesiology 86, 205–215
41. Giovannucci, D. R., Groblewski, G. E., Sneyd, J., and Yule, D. I. (2000) J. Biol.
Chem. 275, 33704–33711
42. Nassar, A., and Simpson, A. W. (2000) J. Biol. Chem. 275, 23661–23665
43. Halestrap, A. P., and Davidson, A. M. (1990) Biochem. J. 268, 153–160
44. McGrew, S. G., Wolleben, C., Siegl, P., Inui, M., and Fleischer, S. (1989)
Biochemistry 28, 1686–1691
45. Bernardi, P., Scorrano, L., Colonna, R., Petronilli, V., and Di Lisa, F. (1999)
Eur. J. Biochem. 264, 687–701
46. Mak, D. O., and Foskett, J. K. (1994) J. Biol. Chem. 269, 29375–29378
47. Putney, J. W. (1997) Cell Calcium 21, 257–261
48. Munger, S. D., Gleeson, R. A., Aldrich, H. C., Rust, N. C., Ache, B. W., and
Greenberg, R. M. (2000) J. Biol. Chem. 275, 20450–20457
49. Kroemer, G. (1997) Nat. Med. 3, 614–620
50. Szalai, G., Csordas, G., Hantash, B. M., Thomas, A. P., and Hajnoczky, G.
(2000) J. Biol. Chem. 275, 15305–15313
Mitochondrial Ryanodine Receptor21488
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from

Shey-Shing Sheu
Gisela Beutner, Virendra K. Sharma, David R. Giovannucci, David I. Yule and
Identification of a Ryanodine Receptor in Rat Heart Mitochondria
doi: 10.1074/jbc.M101486200 originally published online April 10, 2001
2001, 276:21482-21488.J. Biol. Chem.
10.1074/jbc.M101486200Access the most updated version of this article at doi:
Alerts:
When a correction for this article is posted• When this article is cited•
to choose from all of JBC's e-mail alertsClick here
http://www.jbc.org/content/276/24/21482.full.html#ref-list-1
This article cites 50 references, 24 of which can be accessed free at
at University of Rochester on February 19, 2016http://www.jbc.org/Downloaded from