Am J Physiol Cell Physiol
279: C724–C733, 2000.
Differential Ca2⫹ sensitivity of skeletal and cardiac
muscle ryanodine receptors in the presence of calmodulin
BRADLEY R. FRUEN,1 JENNIFER M. BARDY,1 TODD M. BYREM,2
GALE M. STRASBURG,2 AND CHARLES F. LOUIS1
1
Department of Biochemistry, Molecular Biology, and Biophysics, University of Minnesota,
Minneapolis, Minnesota 55455; and 2Department of Food Science and Human Nutrition,
Michigan State University, East Lansing, Michigan 48824
Received 20 September 1999; accepted in final form 4 April 2000
sarcoplasmic reticulum; calcium release channel; excitationcontraction coupling
muscle, the increase in
myoplasmic Ca2⫹ that triggers contraction reflects the
activation of ryanodine receptor (RyR) channels in the
sarcoplasmic reticulum (SR). The RyR isoforms expressed in mammalian skeletal muscle (RyR1) and
cardiac muscle (RyR2) share 66% homology at the
IN BOTH SKELETAL AND CARDIAC
Address for reprint requests and other correspondence: B. R.
Fruen, 6–155 Jackson Hall, 321 Church St. SE, Minneapolis, MN
55455 (E-mail: [email protected]).
C724
amino acid level and exhibit similar responses to a
number of physiological and pharmacological channel
effectors (33). Ca2⫹ itself is a principal endogenous
effector of RyR channels, and activating as well as
inhibitory Ca2⫹ binding sites within the primary sequence of these channel proteins are suggested by
studies describing the Ca2⫹ dependence of channel
activity in isolated preparations (20). In situ, however,
important actions of Ca2⫹ may be dependent on Ca2⫹
binding not only to RyR channels but also to RyRassociated proteins (19). In this regard, calmodulin
(CaM), the ubiquitous intracellular Ca2⫹ sensor, is now
recognized as an integral component of the intact RyR1
channel complex (35, 36).
CaM binds to RyR1 channels both in the absence and
in the presence of micromolar Ca2⫹ (24, 34, 37). In the
presence of micromolar Ca2⫹, CaM binding is associated with RyR1 inhibition, whereas at nanomolar Ca2⫹
concentrations, CaM binding is associated with RyR1
activation (11, 12, 34). Factors that may regulate
CaM’s interactions with RyR1 channels remain largely
undefined, however, and consequently, the magnitude
of CaM’s effects on RyR1 activity has varied markedly
in studies that utilize different experimental conditions or preparations (8, 11, 26, 38). In addition, it
remains unclear how the CaM-dependent regulation of
RyR1 channels may relate to channel regulation by
Ca2⫹ (i.e., Ca2⫹-induced Ca2⫹ release, or CICR) or by
allosteric interactions with transverse tubule voltage
sensors (i.e., mechanical coupling). Importantly, detailed studies of CaM’s functional interactions with
RyR channels have to date focused on the RyR1 isoform, and in particular, no study has yet determined
whether, in the presence of nanomolar Ca2⫹, CaM may
also bind to and activate cardiac RyR2 channels. Thus
the role of CaM in the physiological mechanisms that
initiate SR Ca2⫹ release in skeletal and cardiac muscle
remains unclear.
In the present study, we examined the relationship
between CaM- and Ca2⫹-dependent mechanisms of
RyR channel activation with the use of SR vesicles
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
0363-6143/00 $5.00 Copyright © 2000 the American Physiological Society
http://www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Fruen, Bradley R., Jennifer M. Bardy, Todd M. Byrem, Gale M. Strasburg, and Charles F. Louis. Differential Ca2⫹ sensitivity of skeletal and cardiac muscle ryanodine
receptors in the presence of calmodulin. Am J Physiol Cell
Physiol 279: C724–C733, 2000.—Calmodulin (CaM) activates the skeletal muscle ryanodine receptor Ca2⫹ release
channel (RyR1) in the presence of nanomolar Ca2⫹ concentrations. However, the role of CaM activation in the mechanisms that control Ca2⫹ release from the sarcoplasmic reticulum (SR) in skeletal muscle and in the heart remains
unclear. In media that contained 100 nM Ca2⫹, the rate of
45
Ca2⫹ release from porcine skeletal muscle SR vesicles was
increased approximately threefold in the presence of CaM (1
␮M). In contrast, cardiac SR vesicle 45Ca2⫹ release was
unaffected by CaM, suggesting that CaM activated the skeletal RyR1 but not the cardiac RyR2 channel isoform. The
activation of RyR1 by CaM was associated with an approximately sixfold increase in the Ca2⫹ sensitivity of [3H]ryanodine binding to skeletal muscle SR, whereas the Ca2⫹ sensitivity of cardiac SR [3H]ryanodine binding was similar in the
absence and presence of CaM. Cross-linking experiments
identified both RyR1 and RyR2 as predominant CaM binding
proteins in skeletal and cardiac SR, respectively, and
[35S]CaM binding determinations further indicated comparable CaM binding to the two isoforms in the presence of
micromolar Ca2⫹. In nanomolar Ca2⫹, however, the affinity
and stoichiometry of RyR2 [35S]CaM binding was reduced
compared with that of RyR1. Together, our results indicate
that CaM activates RyR1 by increasing the Ca2⫹ sensitivity
of the channel, and further suggest differences in CaM’s
functional interactions with the RyR1 and RyR2 isoforms
that may potentially contribute to differences in the Ca2⫹
dependence of channel activation in skeletal and cardiac
muscle.
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
prepared from porcine skeletal and cardiac muscle.
Our results indicate that CaM activation of RyR1 channels may reflect an increased sensitivity of these channels to activation by Ca2⫹. In addition, our results
demonstrate differences in CaM’s functional interactions with the RyR1 and RyR2 channel isoforms that
may potentially contribute to differences in the Ca2⫹
dependence of channel activation that characterize
skeletal and cardiac muscle excitation-contraction
(E-C) coupling.
EXPERIMENTAL PROCEDURES
mined (S. P. J. Brookes, Carleton University, Ottawa, Canada). Equilibrium [3H]ryanodine binding was determined
after collection of SR vesicles on Whatman glass fiber filters.
Nonspecific binding was measured in the presence of 20 ␮M
nonradioactive ryanodine. Data are expressed as percentages
of the maximal [3H]ryanodine binding capacity of the SR
vesicle preparation, as determined in media containing 450
mM KCl, 10 mM Na2ATP, and 100 ␮M Ca2⫹ (12.9 ⫾ 1.3
pmol/mg SR protein for skeletal muscle, n ⫽ 5; 3.7 ⫾ 1.1
pmol/mg SR protein for cardiac muscle, n ⫽ 4). Determinations of half-maximally activating concentrations (Ka) of
CaM or Ca2⫹ were based on fits to the Hill equation (SigmaPlot software; Chicago, IL).
[125I]CaM cross-linking. Mammalian CaM Gln-143-Cys
was site-specifically derivatized at Cys-143 with the photoactivatable cross-linking agent benzephenone-4-maleimide
(32) followed by Bolton-Hunter iodination (38). Iodinated,
derivatized CaM (125I-Bz-CaM) was dialyzed and concentrated in 1 mM HEPES (pH 7.4) with Centricon-10 concentrators (Millipore, Bedford, MA). SR vesicles (1 mg/ml) and
125
I-Bz-CaM (50 nM) were preincubated in the dark for 30
min in ice-cold buffer containing 150 mM NaCl, 50 mM
HEPES (pH 7.4), and either 1 mM EGTA (free Ca2⫹ ⱕ 10
nM) or 100 ␮M CaCl2 and then illuminated for 20 min on ice
in an ultraviolet cross-linker (␭max ⫽ 365 nm; Hoefer, San
Francisco, CA). After electrophoresis of pelleted SR vesicles,
cross-linked proteins were identified on dried, Coomassiestained gels by storage phosphorimaging (Molecular Imager
FX; BioRad, Hercules, CA), and bands corresponding to the
RyRs were quantified by densitometric analysis.
[35S]CaM binding. Mammalian CaM was metabolically
radiolabeled according to the procedures of Moore et al. (24).
Briefly, [35S]methionine was incorporated into CaM by bacterial expression with the use of mCaM cDNA subcloned into
the pET-28a expression vector (Novagen, Madison, WI). Expressed [35S]CaM was first purified by nickel affinity chromatography, and then followed by phenyl sepharose chromatography. The concentration of CaM was determined by
spectroscopy with the use of an extinction coefficient of 0.20
ml 䡠 mg⫺1 䡠 cm⫺1 in the presence of EGTA. SR vesicle
[35S]CaM binding was determined as described (24) in media
that contained 300 mM NaCl, 50 mM MOPS (pH 7.4), 100
␮g/ml BSA, 0.1% 3-[(3-cholamidopropyl)dimethylammonio]1-propanesulfonate, 1 mM EGTA, and 1.2 mM CaCl2 (for
high-Ca2⫹ media, as indicated). After a 2-h incubation period, SR vesicles were collected on Whatman GF/B filters and
washed 5⫻ with 3 ml of ice-cold binding buffer. Nonspecific
binding was determined in the presence of 5 ␮M unlabeled
CaM.
RESULTS
CaM regulation of 45Ca2⫹ release from SR vesicles.
The effect of CaM on Ca2⫹ release from SR vesicles
passively loaded with 45Ca2⫹ was examined at 36°C in
media approximating ionic conditions present in the
myoplasm of relaxed muscle. These media contained
120 mM potassium propionate, 10 mM K-PIPES (pH
7.0), 3 mM Na2AMP-PCP, and 3 mM MgCl2 (free Mg2⫹
is ⬃0.4 mM). In these initial experiments, ionized Ca2⫹
in the release media was buffered to 100 nM with
EGTA.
Figure 1A shows that CaM significantly increased
45
Ca2⫹ release from porcine skeletal muscle SR vesicles in these media (t1/2 for 45Ca2⫹ release is ⬃30% of
control in the presence of 1 ␮M CaM, Table 1). In
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Materials. Pigs were obtained from the University of Minnesota Experimental Farm. 45Ca2⫹ and [3H]ryanodine were
purchased from NEN Life Science Products (Boston, MA).
Unlabeled ryanodine and suramin were from Calbiochem (La
Jolla, CA). ␤,[␥]-Methyleneadenosine 5⬘-triphosphate [AMPPCP (a nonhydrolyzable ATP analog)] and porcine brain CaM
were from Sigma (St. Louis, MO).
Isolation of SR vesicles. Skeletal muscle SR vesicles were
isolated from pig longissimus dorsi muscle as described (7).
Briefly, vesicles obtained by differential ultracentrifugation
of a muscle homogenate were extracted with 0.6 M KCl and
subsequently fractionated on discontinuous sucrose gradients. The terminal cisternae-derived (i.e., “heavy”) SR vesicles that band at the 36–40% interface were collected and
stored frozen at ⫺70°C. Cardiac muscle SR vesicles were
isolated from porcine ventricular tissue (6). After homogenization in 10 mM NaHCO3, membranes were extracted in
0.6 M KCl and 20 mM Tris (pH 6.8) and then resuspended in
10% sucrose and stored frozen at ⫺70°C. All isolation buffers
contained a mixture of protease inhibitors.
45
Ca2⫹ release. SR vesicle 45Ca2⫹ release was assayed
essentially as described (21). Vesicles passively loaded with 5
mM 45Ca2⫹ (⫾ 1 ␮M CaM) were placed on the side of a
polystyrene tube that contained 120 mM potassium propionate, 10 mM K-PIPES (pH 7.0), 8.6 mM EGTA, 3 mM
Na2AMP-PCP, 3 mM MgCl2 (free Mg2⫹ is ⬃0.4 mM), 2 mM
Ca2⫹ acetate (free Ca2⫹ is ⬃100 nM), and ⫾ 1 ␮M CaM. Ca2⫹
release was initiated by rapid mixing and stopped at the
indicated times by rapid dilution into a release-inhibiting
medium (120 mM potassium propionate, 10 mM K-PIPES
(pH 7.0), 10 mM EGTA, 5 mM MgCl2, and 20 ␮M ruthenium
red) and then immediately collected on 0.45-␮m Millipore
filters. The fraction of total loaded 45Ca2⫹ that was not
released after 10-s incubations in a release medium that
promotes maximal RyR activation (450 mM KCl, 10 mM
K-PIPES (pH 7.0), 10 mM Na2ATP, and 10 ␮M Ca2⫹) was
considered background and was subtracted from all determinations (⬍12% total counts per minute for both skeletal and
cardiac SR). Sample means were compared with Student’s
t-test and were considered significantly different at P ⬍ 0.05.
Estimates of the time required for vesicles to release one-half
of their 45Ca2⫹ contents (t1/2) were based on fits to the
equation R ⫽ Rmax⫻t/(t1/2 ⫹ t), where R is Ca2⫹ released,
Rmax is maximal Ca2⫹ release, and t is time.
[3H]ryanodine binding. SR vesicles were incubated for 90
min at 36°C in media that contained 120 mM potassium
propionate, 10 mM K-PIPES (pH 7.0), 3 mM Na2AMP-PCP,
100 nM [3H]ryanodine, and a Ca2⫹ acetate-EGTA buffer set
to give the desired Ca2⫹ concentration. In experiments described in RESULTS (see Fig. 2, C and D, and Fig. 4), media also
contained MgCl2 at the concentrations indicated in the figure
legends. Free Ca2⫹ and Mg2⫹ concentrations were calculated
with the use of the computer program Bound and Deter-
C725
C726
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
contrast to CaM’s effect on 45Ca2⫹ release from skeletal
muscle SR, 45Ca2⫹ release from cardiac SR vesicles
was similar in the presence and absence of CaM (Fig.
1B). Caffeine, however, significantly increased 45Ca2⫹
release from both cardiac and skeletal muscle SR (Fig.
1, A and B). These results thus indicate that CaM
selectively activated 45Ca2⫹ release from skeletal muscle, but not cardiac, SR vesicles in media that contained 100 nM Ca2⫹.
For comparison, subsequent experiments examined
SR 45Ca2⫹ release in media that contained 500 ␮M
Ca2⫹ (Fig. 1, C and D). In the presence of 500 ␮M Ca2⫹,
the rate of SR 45Ca2⫹ release was increased such that
in the control media, vesicles released more than onehalf of their 45Ca2⫹ stores within 1 s. Nonetheless,
when CaM was included in the media, an approximate
Table 1. Effect of CaM
from SR vesicles
45
Ca2⫹ release
Skeletal
Cardiac
45
Ca
2⫹
Release, t1/ 2, s
2⫹
100 nM Ca
Control
CaM
Caffeine
500 ␮M Ca2⫹
Control
CaM
15.0 ⫾ 7.2 (3)
4.2 ⫾ 3.0 (3)
0.6 ⫾ 0.3 (3)
11.1 ⫾ 17 (3)
10.4 ⫾ 10 (3)
2.4 ⫾ 1.8 (3)
0.4 ⫾ 0.2 (3)
0.8 ⫾ 0.2 (3)
0.5 ⫾ 0.1 (4)
1.1 ⫾ 0.2 (4)
Values are means ⫾ SE. 45Ca2⫹ release was determined as described in EXPERIMENTAL PROCEDURES. Estimates of half-times (t1/ 2)
for 45Ca2⫹ release in the absence (control) or presence of either CaM
(1 ␮M) or caffeine (10 mM) are based on fits to the data presented in
Fig. 1 (number of preparations indicated in parentheses). SR, sarcoplasmic reticulum; CaM, calmodulin.
twofold increase in t1/2 for 45Ca2⫹ release was apparent
for both skeletal and cardiac SR (Table 1). These results are thus consistent with earlier reports that demonstrate similar inhibitory effects of CaM on SR Ca2⫹
release from skeletal (21) and cardiac (23) SR vesicles
in media containing micromolar Ca2⫹.
CaM effects on Ca2⫹ dependence of SR vesicle [3H]ryanodine binding. SR vesicle [3H]ryanodine binding was
used to further investigate the selective activation by
CaM of the skeletal muscle RyR1 compared with the
cardiac RyR2 isoform, and to further examine the relationship between CaM- and Ca2⫹-dependent mechanisms of RyR activation. Because ryanodine binds with
high affinity to the open state of RyR channels, changes
in [3H]ryanodine binding that occur in the presence of
RyR effectors provide a useful index of changes in RyR
channel activity (3, 22). CaM’s effects on Ca2⫹ activation of [3H]ryanodine binding were initially examined
in media containing 3 mM Na2AMP-PCP and 3mM
MgCl2 (i.e., as in Fig. 1). In these Mg2⫹-containing
media, the threshold for Ca2⫹ activation of [3H]ryanodine binding to both skeletal muscle and cardiac SR
was ⬃0.1 ␮M Ca2⫹, although the maximal extent of
activation by micromolar Ca2⫹ was significantly less
for skeletal than for cardiac muscle SR (24% vs. 82%
activation, respectively; Fig. 2, A and B). The addition
of 1 ␮M CaM reduced the Ka for Ca2⫹ activation of
skeletal SR [3H]ryanodine binding to approximately
one-fourth of control and decreased the cooperativity of
Ca2⫹ activation (Table 2). In contrast, the Ca2⫹ dependence of cardiac SR [3H]ryanodine binding was unaffected by CaM (P ⫽ 0.5).
CaM’s effect on the Ca2⫹ dependence of [3H]ryanodine binding was also examined in media from which
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Fig. 1. Calmodulin (CaM) activation of
sarcoplasmic reticulum (SR) vesicle Ca2⫹
release in the presence of 100 nM Ca2⫹ (A
and B) and 500 ␮M Ca2⫹ (C and D). SR
vesicle 45Ca2⫹ release was determined as
described in EXPERIMENTAL PROCEDURES in
media that contained 120 mM potassium
propionate, 10 mM PIPES (pH 7.0), 3 mM
Na2AMP-PCP, and 3 mM MgCl2. CaM (1
␮M) or caffeine (10 mM) were added to the
media as indicated. Maximal ryanodine
receptor (RyR)-mediated 45Ca2⫹ release
[determined in media that contained 450
mM KCl, 10 mM K-PIPES (pH 7.0), 10
mM Na2ATP, and 10 ␮M Ca2⫹] was 33 ⫾
3.6 nmol/mg for skeletal SR and 20 ⫾ 4.8
nmol/mg for cardiac SR. Data are
means ⫾ SE from 3–4 independent experiments (performed in duplicate and using
different SR vesicle preparations). * Significantly different from release in control
media (P ⬍ 0.05).
C727
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
MgCl2 was omitted to optimize RyR activation by Ca2⫹
and obviate differential effects of Mg2⫹ on the RyR1
and RyR2 channel isoforms (17). The omission of Mg2⫹
increased the extent of Ca2⫹-activated [3H]ryanodine
binding to skeletal muscle SR but had little effect on
the threshold of activation (⬃100 nM Ca2⫹ in control
media; Fig. 2, C and D). Consequently, in the absence
of added CaM, the RyR1 and RyR2 channel isoforms
now displayed a similar Ca2⫹ dependence of activation
(Ka for Ca2⫹ is ⬃360 nM for both skeletal and cardiac
muscle SR; Table 2). This result is thus consistent with
the similar affinities of Ca2⫹ activation sites on RyR1
and RyR2 channels (e.g., Refs. 6, 10, 17). In the presence of CaM, however, the Ca2⫹ dependence of [3H]ryanodine binding to skeletal and cardiac muscle SR
differed markedly (Fig. 2, C and D). CaM reduced the
threshold for Ca2⫹ activation of skeletal muscle SR
[3H]ryanodine binding ⬃10-fold, decreasing the Ka for
Ca2⫹ from 360 nM to 60 nM. The Ca2⫹ sensitivity of
cardiac SR [3H]ryanodine binding, in contrast, was
again not significantly affected by CaM (Table 2; P ⫽
0.2). These data therefore demonstrate that CaM activation was associated with a pronounced shift in the
Ca2⫹ dependence of skeletal muscle SR [3H]ryanodine
binding to lower Ca2⫹ concentrations. Furthermore, in
the presence of CaM, the skeletal RyR1 and cardiac
RyR2 isoforms displayed a marked difference in their
Ca2⫹ sensitivities that was not apparent in the CaMfree media.
Modulation of CaM activation by CICR effectors. To
further investigate the relationship between CaM- and
Ca2⫹-dependent mechanisms of RyR1 activation, we
examined the modulation of CaM activation by effectors of CICR. In a previous report, we documented that
the extent of Ca2⫹ activation of RyR1 is strictly dependent on the presence of adenine nucleotide when media
are composed primarily of organic anions (6). Similarly, in a potassium propionate/PIPES medium con-
Table 2. Effect of CaM on the Ca2⫹ dependence of [3H]ryanodine binding
to skeletal and cardiac muscle SR vesicles
Skeletal
Ka Ca
⫹Mg
Control
CaM
⫺Mg2⫹
Control
CaM
2⫹
, ␮M
Cardiac
, ␮M
2⫹
nH
n
Ka Ca
nH
n
1.1 ⫾ 0.3
0.29 ⫾ 0.1*
2.1 ⫾ 0.7
1.0 ⫾ 0.3
3
3
0.67 ⫾ 0.1
0.57 ⫾ 0.1
1.9 ⫾ 0.1
1.8 ⫾ 0.1
4
4
0.36 ⫾ 0.02
0.06 ⫾ 0.01**
1.6 ⫾ 0.1
1.1 ⫾ 0.1
4
4
0.37 ⫾ 0.03
0.27 ⫾ 0.07
2.1 ⫾ 0.2
1.4 ⫾ 0.1*
4
4
2⫹
Values are means ⫾ SE. [3H]ryanodine binding was determined as described in EXPERIMENTAL PROCEDURES in the absence or presence of
1 ␮M CaM. Media contained 120 mM potassium propionate, 10 mM PIPES (pH 7.0), and 3 mM Na2AMP-PCP, either with or without 3 mM
MgCl2, as indicated. Estimates for half-maximally activating concentration (Ka ) and Hill coefficient (nH) are based on fits of the data
presented in Fig. 2 to the Hill equation. Significant differences from control are indicated at either the P ⬍ 0.05 (*) or P ⬍ 0.001 (**) level.
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Fig. 2. Effect of CaM on the Ca2⫹ dependence of
SR vesicle [3H]ryanodine binding. [3H]ryanodine
binding to normal skeletal muscle SR or cardiac
SR was determined in the absence (E) or presence (●) of 1 ␮M CaM. Media contained 120 mM
potassium propionate, 10 mM PIPES (pH 7.0),
and 3 mM Na2AMP-PCP, either with (A and B)
or without (C and D) 3 mM MgCl2. Data are
expressed as percentages of the maximal [3H]ryanodine binding capacity of the SR vesicle preparations (12.0 ⫾ 2.3 for skeletal SR; 3.9 ⫾ 0.8 for
cardiac SR). Solid lines are based on fits to the
Hill equation. Data are means ⫾ SE of 3–7
independent experiments (at least 3 different SR
vesicle preparations).
C728
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
taining 100 nM Ca2⫹, the extent of CaM activation of
RyR1 was also dependent on adenine nucleotide. Thus
in the absence of nucleotide, CaM activated skeletal
muscle SR [3H]ryanodine binding to ⬍5% of maximal
activation, whereas including 3 mM Na2AMP-PCP in
the medium increased the CaM-dependent activation
by ⬃20-fold (Fig. 3A). In the presence of Na2AMP-PCP,
the CaM-dependent activation of skeletal muscle SR
[3H]ryanodine binding was monophasic and suggested
that CaM may act at a discrete high-affinity site on the
RyR1. Caffeine (5 mM) also increased the extent of
CaM-activated [3H]ryanodine binding to skeletal muscle SR (Fig. 3A). Determinations of apparent affinities
and Hill coefficients for CaM activation of [3H]ryanodine binding suggested that AMP-PCP (Ka ⫽ 28 ⫾ 3.1
nM; nH ⫽ 1.5 ⫾ 0.2) and caffeine (Ka ⫽ 38 ⫾ 3.3 nM;
nH ⫽ 1.1 ⫾ 0.1) increased CaM activation of RyR1
independent of direct effects on CaM binding to activation sites on the channel protein.
Subsequent experiments examined the possibility
that AMP-PCP or caffeine might also promote CaM
activation of cardiac RyR2 channels in 100 nM Ca2⫹. In
media containing either AMP-PCP (3 mM) or caffeine
(10 mM), CaM significantly increased cardiac SR vesicle [3H]ryanodine binding (Fig. 3B; P ⬍ 0.05 in the
Fig. 4. Effect of CICR inhibitors on the CaM-dependent stimulation
of [3H]ryanodine binding to normal skeletal muscle SR vesicles.
[3H]ryanodine binding to SR vesicles was determined as described in
EXPERIMENTAL PROCEDURES in media that contained 120 mM potassium propionate, 10 mM PIPES (pH 7.0), 100 nM Ca2⫹, and 3 mM
Na2AMP-PCP (to optimize the CaM-dependent stimulation of
[3H]ryanodine binding). MgCl2 or 1,2-bis(2-aminophenoxy)ethaneN,N,N⬘,N⬘-tetraacetic acid (BAPTA) were included as indicated. Calculated ionized Mg2⫹ concentrations after addition of 2.5 mM and 10
mM MgCl2 were ⬃0.3 mM and ⬃3 mM, respectively. Calculated
ionized Ca2⫹ concentration in media containing 1 mM BAPTA was
⬃3 nM. Data are means ⫾ SE of 3 or 4 independent experiments (at
least 3 different SR vesicle preparations, duplicate determinations).
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Fig. 3. Effect of Ca2⫹-induced Ca2⫹ release (CICR) activators on
CaM-dependent activation of skeletal muscle (A) and cardiac (B) SR
[3H]ryanodine binding in the presence of 100 nM Ca2⫹. SR vesicle
[3H]ryanodine binding was determined as described in EXPERIMENTAL
PROCEDURES. Control media contained 120 mM potassium propionate, 10 mM PIPES (pH 7.0), and 100 nM Ca2⫹. Media were
supplemented with either Na2AMP-PCP (3 mM) or caffeine (5 mM in
A, 10 mM in B), as indicated. Data are means ⫾ SE of 3–4 independent experiments.
presence of 300 nM and 1 ␮M CaM). However, the
magnitude of the CaM-dependent activation of cardiac
SR [3H]ryanodine binding in these media was small
compared with the CaM-dependent activation of
[3H]ryanodine binding to skeletal muscle SR. Furthermore, compared with skeletal muscle SR (Fig. 3A), the
CaM-dependent activation of cardiac SR [3H]ryanodine
binding required higher concentrations of CaM and did
not plateau over the range of CaM concentrations examined. Thus, in contrast to RyR1, CaM activation of
RyR2 suggested a lower affinity or nonspecific interaction of CaM with the cardiac channel isoform.
The effect of CICR inhibitors on the CaM-dependent
activation of RyR1 channels was examined in media
containing 100 nM Ca2⫹ and 3 mM Na2AMP-PCP (Fig.
4). CaM activation of skeletal SR [3H]ryanodine binding displayed a marked sensitivity to inhibition by
physiological concentrations of Mg2⫹. The addition of
2.5 mM MgCl2 (free Mg2⫹ is ⬃0.3 mM) reduced the
CaM-dependent activation to ⬃25% of control, and 10
mM MgCl2 (free Mg2⫹ is ⬃3 mM) completely blocked
the activation of [3H]ryanodine binding by CaM. CaM
activation of [3H]ryanodine binding was similarly inhibited when Ca2⫹ was buffered to lower concentrations with 1,2-bis(2-aminophenoxy)ethane-N,N,N⬘,N⬘tetraacetic acid (BAPTA; Fig. 4, free Ca2⫹ is ⬃3 nM).
Both Mg2⫹ and BAPTA reduced the maximal extent of
CaM-activated [3H]ryanodine binding without affecting the Ka for CaM (24 ⫾ 5.8 nM in 2.5 mM MgCl2;
25 ⫾ 7.8 nM in 1 mM BAPTA), consistent with noncompetitive inhibition of CaM activation. These results
thus indicate that CaM activation of RyR1 was inhibited when activation of the channel by Ca2⫹ was
blocked and thereby further suggest that CaM activation may reflect an increase in the sensitivity of the
RyR1 to Ca2⫹, rather than a Ca2⫹-independent channel activation by CaM.
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
Fig. 5. Competitive inhibition by suramin of CaM-activated SR vesicle [3H]ryanodine binding. [3H]ryanodine binding to skeletal muscle
SR vesicles was determined as in the absence and presence of
suramin in media that contained 120 mM potassium propionate, 10
mM PIPES (pH 7.0), and 3 mM Na2AMP-PCP. Data are means ⫾ SE
of 2–4 independent experiments (triplicate determinations).
Fig. 6. Cross-linking of a photoactivatable derivative of [125I]CaM to
skeletal and cardiac muscle SR vesicles. SR vesicles from 2 different
skeletal SR preparations (left 4 lanes) and 2 different cardiac SR
preparations (right 4 lanes) were covalently labeled with 125I-BzCaM (50 nM) as described in EXPERIMENTAL PROCEDURES. Media
contained either ⬃10 nM Ca2⫹ or 100 ␮M Ca2⫹, as indicated. A:
autoradiogram of [125I]CaM-labeled SR proteins separated on 5–12%
linear gradient PAGE. Arbitrary optical density units of bands that
correspond to the RyRs were as follows: lane 1, 2,132; lane 2, 1,074;
lane 3, 3,396; lane 4, 1,587; lane 5, 2,600; lane 6, 5,460; lane 7, 3,745;
and lane 8, 7,561. B: the Coomassie blue-stained gel is shown to
document that the different lanes in A contained comparable
amounts of RyR protein. Arrows indicate locations of the two RyR
isoforms, free 125I-Bz-CaM, and the positions of molecular weight
markers. Experiment shown is representative of 5 independent experiments.
skeletal RyR1 and cardiac RyR2 isoforms, were present
in the different lanes. For both skeletal and cardiac SR
preparations, these ⬃565 kDa RyR proteins were the
predominant 125I-Bz-CaM cross-linked species, and for
both skeletal and cardiac SR, cross-linking was apparent whether media contained nanomolar or micromolar Ca2⫹. Nevertheless, these results suggested potential differences in the Ca2⫹ dependence of 125I-Bz-CaM
cross-linking to RyR1 and RyR2. Thus we found that
the cross-linking of 125I-Bz-CaM to RyR1 in nanomolar
Ca2⫹ was consistently increased relative to that in
micromolar Ca2⫹ (Fig. 6A). Conversely, the cross-linking of 125I-Bz-CaM to RyR2 in nanomolar Ca2⫹ was
reduced relative to that in micromolar Ca2⫹ (Fig. 6A).
It is possible that Ca2⫹-dependent conformational
changes in the 125I-Bz-CaM molecule influenced the
efficiency of cross-linking independent of actual effects
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Competitive inhibition of CaM activation by
suramin. In a recent report, Klinger and colleagues
(13) demonstrated that the purinergic antagonist
suramin reduced RyR1-CaM binding and blocked
RyR1 inhibition by Ca2⫹-CaM. Their results thus indicated that suramin might directly compete with Ca2⫹CaM binding to its inhibitory site on the RyR1 channel
protein. Their study did not, however, determine
whether suramin might also alter CaM’s interactions
with activation sites on the RyR1 in the presence of
submicromolar Ca2⫹. Accordingly, we examined the
effect of suramin on the CaM-dependent activation of
skeletal muscle SR [3H]ryanodine binding (Fig. 5). In
the presence of 10 ␮M suramin, CaM-dependent activation of skeletal muscle SR [3H]ryanodine binding
was shifted ⬃10-fold to higher CaM concentrations
(Ka ⬎ 300 nM CaM). Thus, in contrast to other RyR
effectors (Figs. 3 and 4), these data indicated that
suramin may modulate CaM activation via a direct
competition with CaM binding site(s) on the RyR1. The
increase of suramin to 30 ␮M, however, not only fully
blocked CaM-dependent activation of [3H]ryanodine
binding but also activated [3H]ryanodine binding twofold, suggesting that at higher concentrations, suramin
may have multiple effects on RyR1 function. Previously, activation of RyR1 and RyR2 channels by high
concentrations of suramin was associated with effects
on both single channel conductance and open probability of these channel proteins (29).
Radiolabeled CaM binding to skeletal and cardiac
muscle SR. To investigate potential differences in
CaM’s physical interactions with the skeletal RyR1
and cardiac RyR2 isoforms, initial experiments utilized
photoactivatable, iodinated CaM, site-specifically derivatized at Cys-143 with benzophenone-4-maleimide
(32). SR vesicles were preincubated in media that contained 50 nM 125I-Bz-CaM and either ⬃10 nM Ca2⫹ or
100 ␮M Ca2⫹ and then irradiated with UV light to
initiate cross-linking. The autoradiogram in Fig. 6A
identifies SR proteins cross-linked with 125I-Bz-CaM as
resolved by SDS-PAGE. The Coomassie stain of this gel
(Fig. 6B) confirms that approximately equivalent
amounts of ⬃565 kDa proteins, corresponding to the
C729
C730
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
on RyR CaM binding. In addition, recent evidence has
suggested that CaM binding determinations may be
affected by Bolton-Hunter iodination of CaM (24). Experiments that utilize expressed CaM metabolically
labeled with 35S ([35S]CaM), however, have indicated
high-affinity binding to RyR1 in both the presence and
absence of micromolar Ca2⫹, with a binding stoichiometry of ⬃1 CaM per channel subunit (24). We therefore
utilized [35S]CaM to further characterize CaM binding
to cardiac SR. Figure 7 shows cardiac SR [35S]CaM
binding in media in which Ca2⫹ was buffered to either
⬃10 nM or ⬃200 ␮M. Scatchard analysis (Fig. 7, inset)
indicates that, in the presence of micromolar Ca2⫹,
[35S]CaM binding to cardiac SR was consistent with a
single population of high-affinity sites (Kd ⫽ 15.8 ⫾ 1.8
nM, Bmax ⫽ 25.3 ⫾ 2.3 pmol/mg). In comparison, in
nanomolar Ca2⫹, [35S]CaM binding to cardiac SR was
reduced to one-fifth (Bmax ⫽ 4.9 ⫾ 0.9 pmol/mg), and
the apparent affinity of binding was also significantly
decreased relative to that in micromolar Ca2⫹ (Kd ⫽
83.6 ⫾ 11.4 nM; P ⬍ 0.5).
Table 3 directly compares the binding of [35S]CaM to
cardiac and skeletal SR. Parallel determinations of
[3H]ryanodine receptor density allowed for estimates of
the number of CaM binding sites per RyR tetramer. In
agreement with earlier findings (24), the extent of
[35S]CaM binding to skeletal muscle SR was similar in
DISCUSSION
CaM activates RyR1 channels in the presence of
nanomolar Ca2⫹ concentrations, and this might suggest that the CaM-dependent activation of RyR1 may
operate independently of Ca2⫹-dependent channel activation (i.e., CICR). On the contrary, our results indicate that CaM activation of RyR1 likely operates by
increasing the sensitivity of the Ca2⫹-dependent activation mechanism of this channel protein. Furthermore, the magnitude of CaM’s effect at Ca2⫹ concentrations present in resting muscle (⬃100 nM) is
consistent with a major role for CaM in controlling the
sensitivity of CICR in vivo. Finally, our results indicate
that CaM’s effect on the Ca2⫹ sensitivity of channel
activation may be far greater for the RyR1 than for the
RyR2 isoform. We therefore suggest that differential
interactions of CaM with RyR1 and RyR2 channels
may potentially contribute to differences in the Ca2⫹
dependence of SR Ca2⫹ release in skeletal and cardiac
muscle.
RyR1 activation by Ca2⫹ and CaM are regulated by
common effectors. In media containing 100 nM Ca2⫹
and adenine nucleotide, CaM activation of [3H]ryanodine binding to skeletal muscle SR vesicles was
monophasic (Ka is ⬃30 nM) and Hill coefficients for
activation indicated only weak cooperativity (nH is
⬃1.5). Thus despite early evidence that CaM may bind
to as many as four sites per subunit of the RyR1
tetramer (9, 34, 37), the observed activation is also
consistent with more recent data suggesting that there
is CaM action at a single high-affinity site within the
RyR1 primary sequence (Ref. 24, Table 3). Moreover,
Table 3. Comparison of [35S]CaM binding to skeletal and cardiac muscle SR vesicles
Skeletal SR
35
␮M Ca
nM Ca2⫹
2⫹
Cardiac SR
35
[ S]CaM bound, pmol/mg
CaM/RyR
n
[ S]CaM bound, pmol/mg
CaM/RyR
n
40 ⫾ 3.0
41 ⫾ 3.2
3.6 ⫾ 0.9
3.7 ⫾ 0.9
4
4
20 ⫾ 2.0
4.0 ⫾ 0.8*
5.0 ⫾ 0.9
0.9 ⫾ 1.0*
3
3
Values are means ⫾ SE. Binding of [35S]CaM was determined in media containing 300 mM NaCl, 50 mM MOPS (pH 7.4), 100 ␮g/ml BSA,
0.1% CHAPS, 1 mM EGTA, and 100 nM [35S]CaM. Media were supplemented with 1.2 mM CaCl2 for binding determinations in micromolar
Ca2⫹. Estimates of [35S]CaM binding per RyR tetramer (CaM/RyR) are based on the maximal SR vesicle [3H]ryanodine binding capacity,
determined as described in EXPERIMENTAL PROCEDURES. * Significantly less than binding in micromolar Ca2⫹, P ⬍ 0.04, Student’s t-test.
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
Fig. 7. [35S]CaM binding to cardiac SR vesicles. [35S]CaM binding to
SR vesicles was determined as described in EXPERIMENTAL PROCEDURES in media that contained 300 mM NaCl, 50 mM MOPS (pH 7.4),
100 ␮g/ml BSA, 0.1% 3-[(3-cholamidopropyl)dimethylammonio]-1propanesulfonate, and 1 mM EGTA (free Ca2⫹ is ⬃10 nM, E). As
indicated (●), media were also supplemented with 1.2 mM CaCl2
(free Ca2⫹ is ⬃200 ␮M). Binding parameters based on 3 independent
SR vesicle preparations were Bmax ⫽ 4.9 ⫾ 0.9 pmol/mg in nanomolar Ca2⫹ vs. 25.3 ⫾ 2.5 pmol/mg in micromolar Ca2⫹; Kd ⫽ 83.6 ⫾ 11
nM in nanomolar Ca2⫹ vs. 15.8 ⫾ 1.8 nM in micromolar Ca2⫹.
the presence of nanomolar and micromolar Ca2⫹, and
binding was consistent with approximately four
[35S]CaM binding sites per [3H]ryanodine binding site
in our skeletal SR preparations. Likewise, in media
that contained micromolar Ca2⫹, cardiac SR exhibited
approximately five [35S]CaM binding sites per [3H]ryanodine binding site (Table 3). In contrast to skeletal
SR, however, cardiac SR [35S]CaM binding was reduced to ⬃1 mol of [35S]CaM per [3H]ryanodine binding site in media containing nanomolar Ca2⫹. These
results thus suggest that the selective activation by
CaM of Ca2⫹ release from skeletal compared with
cardiac SR at nanomolar Ca2⫹ (Fig. 1, A and B) may be
associated with increased CaM binding to the RyR1
compared with the RyR2 isoform in nanomolar Ca2⫹.
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
cles (Fig. 1B, Table 1), suggesting that CaM effects on
cardiac SR [3H]ryanodine binding may reflect lower
affinity or possibly nonspecific interactions with RyR2
at nanomolar Ca2⫹. Consistent with this possibility,
the binding of [35S]CaM to cardiac SR vesicles was
reduced at nanomolar Ca2⫹ (Fig. 7, Table 3), indicating
that differential activation of RyR1 and RyR2 by CaM
may be associated with important differences in CaM
binding to the two channel isoforms at nanomolar
Ca2⫹. These results, therefore, suggest that the RyR2
isoform may lack a CaM activation site that is present
in RyR1. Alternatively, an RyR2 site may be occluded,
for example, by covalent modification of the channel
(24, 25) or by endogenous CaM that has remained
tightly bound. At micromolar Ca2⫹, however, skeletal
and cardiac SR displayed similar [35S]CaM binding
stoichiometries (Table 3), a result that is consistent
with earlier reports that document similar inhibitory
actions of Ca2⫹-CaM on RyR1 (21) and RyR2 (Ref. 30;
see also Fig. 1, C and D). Nevertheless, it remains to be
determined whether similar mechanisms may underlie
Ca2⫹-CaM inhibition of the RyR1 and RyR2 isoforms,
and whether these mechanisms may involve, for example, the modulation of Ca2⫹-dependent channel inhibition.
Potential role of CaM activation in skeletal muscle
E-C coupling. The primary, voltage-dependent mechanism responsible for activating SR Ca2⫹ release during
skeletal muscle E-C coupling almost certainly involves
a direct mechanical interaction between transverse
tubule voltage sensors and RyR1 channels that results
in RyR1 activation at resting Ca2⫹ concentrations (31,
33). Less certain is the degree to which this mechanism
may be dependent on endogenous effectors of CICR,
including Ca2⫹, Mg2⫹, ATP, and CaM (1). In muscle
fiber preparations, a major fraction of the SR Ca2⫹
released during an action potential may be attributed
to CICR (28, 31). Yet, paradoxically, CICR from isolated skeletal muscle SR may be largely suppressed
under ionic conditions that exist in vivo (e.g., Refs. 5
and 7). This paradox might be resolved if the sensitivity of CICR in skeletal muscle was in part controlled by
the transverse tubule voltage sensors. According to the
model proposed by Lamb and Stephenson (15, 16), SR
Ca2⫹ release is blocked by physiological Mg2⫹, and
during E-C coupling, voltage sensors activate Ca2⫹
release by promoting the dissociation of Mg2⫹ from
low-affinity sites on RyR1 channels. More recently,
Lacampagne and co-workers (14) demonstrated that
lowering myoplasmic Mg2⫹ in fiber preparations increased the frequency of spontaneous Ca2⫹ release
events (i.e., “sparks”) without altering the properties of
the individual release events. These effects were attributed to decreased Mg2⫹ block of high-affinity Ca2⫹
activation sites on RyR channels and a resultant shift
in the threshold for CICR nearer to resting Ca2⫹ (14).
In this regard, our results indicate that lowering the
Mg2⫹ concentration increased the maximal extent of
Ca2⫹-activated [3H]ryanodine binding to skeletal muscle SR but had comparatively little effect on the threshold for Ca2⫹ activation, except when CaM was also
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
that suramin competitively blocks not only CaM inhibition of RyR1 at micromolar Ca2⫹ (Ref. 13, data not
shown) but also CaM activation of RyR1 at nanomolar
Ca2⫹ (Fig. 5) further supports the possibility that CaM
activation and CaM inhibition reflect action at the
same site on the RyR1 channel protein (24, 34).
The extent of CaM activation of RyR1 was dependent
on the presence of adenine nucleotide. In the absence of
AMP-PCP, CaM activated normal skeletal muscle SR
vesicle [3H]ryanodine binding to ⬍5% of maximal,
whereas in the presence of 3 mM AMP-PCP, activation
of [3H]ryanodine binding by CaM approached 50% of
maximal activation (Fig. 3A). Conversely, CaM activation was inhibited by physiological concentrations of
Mg2⫹ (Fig. 4). Caffeine increased the extent of CaMdependent activation of [3H]ryanodine binding to skeletal muscle SR (Fig. 3A), whereas BAPTA (free Ca2⫹ is
⬃3 nM) noncompetitively inhibited CaM activation
(Fig. 4). Together these results demonstrate that CaM
activation of RyR1 is modulated by effectors of CICR
and further support the findings of Ikemoto and coworkers (11, 12) in indicating that CaM activation may
reflect an increased sensitivity of a CICR activation
mechanism. A similar role for CaM in sensitizing RyR
channels to activation by CICR was previously proposed to account for CaM activation of RyR channels in
sea urchin egg microsomes (18). Notably, CICR in
skeletal muscle has generally been considered to depend on an initial increase in myoplasmic Ca2⫹ above
resting concentrations (28, 31). However, these results
raise the possibility that CaM may provide a means by
which CICR may operate even at resting Ca2⫹.
Distinct functional interactions of CaM with RyR1
and RyR2. D-myo-inositol 1,4,5-trisphosphate receptors
[Ins(1,4,5)P3Rs] are intracellular Ca2⫹ release channels that exhibit important structural and functional
similarities with RyR channels (19), and differences
in CaM’s interactions with type 1 and type 3
Ins(1,4,5)P3Rs are postulated to contribute to differences in the Ca2⫹-dependent regulation of the two
Ins(1,4,5)P3R isoforms in situ (2, 27). Likewise, our
results demonstrate that CaM may differentially effect
the Ca2⫹-dependent activation of skeletal RyR1 and
cardiac RyR2 isoforms. In the presence of CaM, the
threshold for activation of skeletal muscle SR [3H]ryanodine binding was shifted to ⬃10-fold lower Ca2⫹
concentrations (Fig. 2C), and the apparent Ka for Ca2⫹
was decreased to near or below resting Ca2⫹ concentrations (Table 2). In comparison, the Ca2⫹ dependence
of cardiac SR [3H]ryanodine binding was only minimally affected by CaM, and in all media the Ka for Ca2⫹
activation of [3H]ryanodine binding to cardiac SR remained above resting Ca2⫹ concentrations (Fig. 2, Table 2). Although cardiac SR [3H]ryanodine binding was
significantly increased by CaM when Mg2⫹-free media
were supplemented with either caffeine or Na2AMPPCP (Fig. 3B), this activation was smaller in magnitude and required higher CaM concentrations than did
CaM activation of skeletal muscle SR [3H]ryanodine
binding under the same conditions. Moreover, CaM
failed to activate 45Ca2⫹ release from cardiac SR vesi-
C731
C732
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
We thank Drs. Z. Grabarek and S. L. Hamilton for providing the
mammalian CaM clone, and Dr. E. Balog for helpful discussions.
This work was supported by grants from the National Institutes of
Health (to C. F. Louis), the Muscular Dystrophy Association (to G. M.
Strasburg), and the American Heart Association (to B. R. Fruen).
REFERENCES
1. Anderson K and Meissner G. T-tubule depolarization-induced
SR Ca2⫹ release is controlled by dihydropyridine receptor- and
Ca2⫹-dependent mechanisms in cell homogenates from rabbit
skeletal muscle. J Gen Physiol 105: 363–383, 1995.
2. Cardy TJ and Taylor CW. A novel role for calmodulin: Ca2⫹independent inhibition of type-1 inositol trisphosphate receptors.
Biochem J 334: 447–455, 1998.
3. Chu A, Diaz-Munoz M, Hawkes MJ, Brush K, and Hamilton SL. Ryanodine as a probe for the functional state of the
skeletal muscle sarcoplasmic reticulum calcium release channel.
Mol Pharmacol 37: 735–741, 1990.
5. Endo M. Calcium release from the sarcoplasmic reticulum.
Physiol Rev 57: 71–108, 1977.
6. Fruen BR, Kane PK, Mickelson JR, and Louis CF. Chloridedependent sarcoplasmic reticulum Ca2⫹ release correlates with
increased Ca2⫹ activation of ryanodine receptors. Biophys J 71:
2522–2530, 1996.
7. Fruen BR, Mickelson JR, Shomer NH, Roghair TJ, and
Louis CF. Regulation of the sarcoplasmic reticulum ryanodine
receptor by inorganic phosphate. J Biol Chem 269: 192–198,
1994.
8. Fruen BR, Mickelson JR, and Louis CF. Dantrolene inhibition of sarcoplasmic reticulum Ca2⫹ release by direct and specific
action at skeletal muscle ryanodine receptors. J Biol Chem 272:
26965–26971, 1997.
9. Guerrini R, Menegazzi P, Anacardio R, Marastoni M,
Tomatis R, Zorzato F, and Treves S. Calmodulin binding
sites of the skeletal, cardiac, and brain ryanodine receptor Ca2⫹ channels: modulation by the catalytic subunit of
cAMP-dependent protein kinase? Biochemistry 34: 5120–
5129, 1995.
10. Gyorke S, Velez P, Suarez-Isla B, and Fill M. Activation of
single cardiac and skeletal ryanodine receptor channels by
flash photolysis of caged Ca2⫹. Biophys J 66: 1879–1886,
1994.
11. Ikemoto T, Iino M, and Endo M. Enhancing effect of calmodulin on Ca2⫹-induced Ca2⫹ release in the sarcoplasmic reticulum
of rabbit skeletal muscle fibres. J Physiol (Lond) 487: 573–582,
1995.
12. Ikemoto T, Takeshima H, Iino M, and Endo M. Effect of
calmodulin on Ca2⫹-induced Ca2⫹ release of skeletal muscle
from mutant mice expressing either ryanodine receptor type 1 or
type 3. Pflügers Arch 437: 43–48, 1998.
13. Klinger M, Freissmuth M, Nickel P, Stabler-Schwarzbart
M, Kassack M, Suko J, and Hohenegger M. Suramin and
suramin analogs activate skeletal muscle ryanodine receptor via
a calmodulin binding site. Mol Pharmacol 55: 462–472, 1999.
14. Lacampagne A, Klein MG, and Schneider MF. Modulation
of the frequency of spontaneous sarcoplasmic reticulum Ca2⫹
release events (Ca2⫹ sparks) by myoplasmic [Mg2⫹] in frog
skeletal muscle. J Gen Physiol 111: 207–224, 1998.
15. Lamb GD. Ca2⫹ inactivation, Mg2⫹ inhibition and malignant
hyperthermia. J Muscle Res Cell Motil 14: 554–556, 1993.
16. Lamb GD and Stephenson DG. Effect of Mg2⫹ on the control
of Ca2⫹ release in skeletal muscle fibres of the toad. J Physiol
(Lond) 434: 507–528, 1991.
17. Laver DR, Baynes TM, and Dulhunty AF. Magnesium
inhibition of ryanodine-receptor calcium channels: evidence
for two independent mechanisms. J Membr Biol 156: 213–229,
1997.
18. Lee HC, Aarhus R, and Graeff RM. Sensitization of calciuminduced calcium release by cyclic ADP-ribose and calmodulin.
J Biol Chem 270: 9060–9066, 1995.
19. Mackrill JJ. Protein-protein interactions in intracellular Ca2⫹release channel function. Biochem J 337: 345–361, 1999.
20. Meissner G. Ryanodine receptor/Ca2⫹ release channels and
their regulation by endogenous effectors. Annu Rev Physiol 56:
485–508, 1994.
21. Meissner G. Evidence of a role for calmodulin in the regulation
of calcium release from skeletal muscle sarcoplasmic reticulum.
Biochemistry 25: 244–251, 1986.
22. Meissner G and El-Hashem A. Ryanodine as a functional
probe of the skeletal muscle sarcoplasmic reticulum Ca2⫹ release
channel. Mol Cell Biochem 114: 119–123, 1992.
23. Meissner G and Henderson JS. Rapid calcium release from
cardiac sarcoplasmic reticulum vesicles is dependent on Ca2⫹
and is modulated by Mg2⫹, adenine nucleotide, and calmodulin.
J Biol Chem 262: 3065–3073, 1987.
24. Moore CP, Rodney G, Zhang JZ, Santacruz-Toloza L,
Strasburg GM, and Hamilton SL. Apocalmodulin and Ca2⫹
calmodulin bind to the same region on the skeletal muscle Ca2⫹
release channel. Biochemistry 38: 8532–8537, 1999.
25. Moore CP, Zhang JZ, and Hamilton SL. A role for cysteine
3635 of RYR1 in redox modulation and calmodulin binding.
J Biol Chem 274: 36831–36834, 1999.
26. O’Driscoll S, McCarthy TV, Eichinger HM, Erhardt W,
Lehmann-Horn F, and Herrmann-Frank A. Calmodulin sensitivity of the sarcoplasmic reticulum ryanodine receptor from
normal and malignant-hyperthermia-susceptible muscle. Biochem J 319: 421–426, 1996.
27. Patel S, Morris SA, Adkins CE, O’Beirne G, and Taylor
CW. Ca2⫹-independent inhibition of inositol trisphosphate receptors by calmodulin: redistribution of calmodulin as a possible
means of regulating Ca2⫹ mobilization. Proc Natl Acad Sci USA
94: 11627–11632, 1997.
28. Schneider MF. Control of calcium release in functioning skeletal muscle fibers. Annu Rev Physiol 56: 463–484, 1994.
29. Sitsapesan R and Williams AJ. Modification of the conductance and gating properties of ryanodine receptors by suramin. J
Membr Biol 153: 93–103, 1996.
30. Smith JS, Rousseau E, and Meissner G. Calmodulin modulation of single sarcoplasmic reticulum Ca2⫹-release channels
from cardiac and skeletal muscle. Circ Res 64: 352–359, 1989.
31. Stern MD, Pizarro G, and Rios E. Local control model of
excitation-contraction coupling in skeletal muscle. J Gen Physiol
110: 415–440, 1997.
32. Strasburg GM, Hogan M, Birmachu W, Thomas DD, and
Louis CF. Site-specific derivatives of wheat germ calmodulin.
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017
present (Fig. 2). Thus we suggest that if the mechanism that activates SR Ca2⫹ release in skeletal muscle
is dependent on a major shift in the Ca2⫹ threshold of
CICR, then this mechanism might also be dependent
on CaM-RyR1 interactions. In this view, our results
that indicate a greater effect of CaM on the Ca2⫹
sensitivity of the RyR1 compared with the RyR2 isoform would be consistent with the activation of Ca2⫹
release from the cardiac SR being more strictly dependent on an initial increase in myoplasmic Ca2⫹ above
resting levels (31, 33).
In conclusion, our results indicate that CaM activates RyR1 by increasing the Ca2⫹ sensitivity of the
channel. Although the significance of CICR in skeletal
muscle remains in question, these results provide additional evidence that CaM may play a critical role in
controlling the activation of RyR1 channels at resting
Ca2⫹ concentrations. Finally, apparent differences in
CaM’s interactions with the RyR2 isoform at submicromolar Ca2⫹ underscore the importance of characterizing the CaM binding domains of the different RyR
isoforms (24, 25, 36) and of further defining the potentially complex roles of CaM-RyR interactions in both
skeletal and cardiac muscle E-C coupling.
CALMODULIN ACTIVATION OF RYANODINE RECEPTORS
Interactions with troponin and sarcoplasmic reticulum. J Biol
Chem 263: 542–548, 1988.
33. Sutko JL and Airey JA. Ryanodine receptor Ca2⫹ release
channels: does diversity in form equal diversity in function?
Physiol Rev 76: 1027–1071, 1996.
34. Tripathy A, Xu L, Mann G, and Meissner G. Calmodulin
activation and inhibition of skeletal muscle Ca2⫹ release channel
(ryanodine receptor). Biophys J 69: 106–119, 1995.
35. Wagenknecht T and Radermacher M. Ryanodine receptors:
structure and macromolecular interaction. Curr Opin Struct Biol
7: 258–265, 1997.
C733
36. Wu Y and Hamilton SL. Functional interactions of cytoplasmic
domains of the skeletal muscle Ca2⫹ release channel. Trends
Cardiovasc Med 8: 312–319, 1998.
37. Yang HC, Reedy MM, Burke CL, and Strasburg GM.
Calmodulin interaction with the skeletal muscle sarcoplasmic
reticulum calcium channel protein. Biochemistry 33: 518–525,
1994.
38. Zhang J, Wu Y, Williams BY, Rodney G, Mandel F, Strasburg GM, and Hamilton SL. Oxidation of the skeletal muscle
Ca2⫹ release channel alters calmodulin binding. Am J Physiol
Cell Physiol 276: C46–C53, 1999.
Downloaded from http://ajpcell.physiology.org/ by 10.220.33.6 on June 17, 2017