Am J Physiol Cell Physiol 303: C567–C576, 2012.
First published July 11, 2012; doi:10.1152/ajpcell.00144.2012.
Changes in plasma membrane Ca-ATPase and stromal interacting molecule 1
expression levels for Ca2⫹ signaling in dystrophic mdx mouse muscle
Tanya R. Cully,1 Joshua N. Edwards,1 Oliver Friedrich,1 D. George Stephenson,2 Robyn M. Murphy,2
and Bradley S. Launikonis1
1
School of Biomedical Sciences, University of Queensland, Brisbane, Queensland, Australia; and 2Department of Zoology,
Latrobe University, Melbourne, Victoria, Australia
Submitted 26 April 2012; accepted in final form 28 June 2012
skeletal muscle; store-operated calcium entry; t-system; muscular
dystrophy; sarcoplasmic reticulum
are changes in the way Ca2⫹ is
handled by the fibers that lead to the loss of normal muscle
tissue and other deleterious effects (16, 32, 33). The plasma
membrane Ca-ATPase (PMCA) and also store-operated Ca2⫹
entry (SOCE) play a critical role in Ca2⫹ regulation at the
cellular level. Specializations of the skeletal muscle fiber for
excitation-contraction coupling (EC coupling) have allowed
this cell to evolve a rapidly activating SOCE mechanism (21)
and possibly place PMCA in the same microdomain. In skeletal muscle, the content of the internal Ca2⫹ store, the sarcoplasmic reticulum (SR), has been significantly increased (14,
26) for the purpose of conducting rapid and precise release of
IN MUSCULAR DYSTROPHY THERE
Address for reprint requests and other correspondence: B. S. Launikonis:
School of Biomedical Sciences, Univ. of Queensland, Brisbane, QLD 4072,
Australia (e-mail: [email protected]).
http://www.ajpcell.org
large amounts of Ca2⫹ for contraction. For tight voltage control of
the release of this Ca2⫹, ⬃80% of the plasma membrane of
skeletal muscle has become internalized as the tubular (t-) system,
forming standing junctions with the terminal cisternae of the SR.
These well-defined junctional membranes inside the fibers provide the physical contact required between the SR ryanodine
receptor (RyR) and the t-system voltage-sensor for EC coupling.
The confined cytoplasmic space between the junctional membranes also allows a tight coupling between the SOCE-conducting
molecules, stromal interacting molecule 1 (STIM1) and Orai1, the
SR Ca2⫹ content sensor, and t-system Ca2⫹ channel, respectively,
required for a fast-activated SOCE mechanism (13, 21). The
placement of PMCA at the junctional membranes would also
allow this protein to influence the evolution of Ca2⫹ signals in this
microdomain.
It has recently been reported that fast activating SOCE in
skeletal muscle is supported by a STIM1 isoform, STIM1L.
This isoform interacts with actin at the SR terminal cisternae
regardless of SR Ca2⫹ concentration ([Ca2⫹]SR) to form permanent clusters and colocalizes with Orai1 (6). This specialization and the presence of standing junctional membranes
through the skeletal muscle fiber dispense the need for protein
and membrane translocation following Ca2⫹ depletion, as must
occur in non-excitable cells (25), significantly speeding up
SOCE in muscle fibers compared with other cells. Of further
importance to Ca2⫹ signaling at junctional membranes, STIM1
and Orai1 in a complex with a newly identified protein called
partner of STIM1 (POST) have been shown to have an inhibitory effect on the activity of PMCA (18).
An upregulation of STIM1 and Orai1 in dystrophic mdx mouse
muscle has been observed (11). While somewhat controversial,
[Ca2⫹] levels at rest in dystrophic muscle are reportedly raised in
mdx muscle, which may lead to variations in Ca2⫹ regulatory
protein expression level changes (1). In response to different
[Ca2⫹] levels, multiple adaptive feedback mechanisms adjust
the levels of STIM1, PMCA, and sarco(endo)plasmic reticulum Ca2⫹-ATPase (SERCA: referred to as “SR Ca2⫹ pump” in
this study) in parallel to maintain Ca2⫹ signaling (1). This process
may be occurring in mdx muscle to maintain normal Ca2⫹
signaling under the stress of its pathophysiological state. In this
study, we probe the expression levels of PMCA and STIM1L
in healthy and dystrophic mdx mouse muscle and also determine the effect of the relative expression levels on Ca2⫹
signaling.
METHODS
Fiber isolation, experimental chamber, and loading of Ca2⫹-sensitive
fluorescent dyes. C57/BL6 (University of Queensland Biological Resources, Brisbane, Australia) and mdx (Animal Resource Centre,
0363-6143/12 Copyright © 2012 the American Physiological Society
C567
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Cully TR, Edwards JN, Friedrich O, Stephenson DG, Murphy RM,
Launikonis BS. Changes in plasma membrane Ca-ATPase and stromal
interacting molecule 1 expression levels for Ca2⫹ signaling in dystrophic
mdx mouse muscle. Am J Physiol Cell Physiol 303: C567–C576, 2012. First
published July 11, 2012; doi:10.1152/ajpcell.00144.2012.—The majority of
the skeletal muscle plasma membrane is internalized as part of the
tubular (t-) system, forming a standing junction with the sarcoplasmic
reticulum (SR) membrane throughout the muscle fiber. This arrangement facilitates not only a rapid and large release of Ca2⫹ from the SR
for contraction upon excitation of the fiber, but has also direct
implications for other interdependent cellular regulators of Ca2⫹. The
t-system plasma membrane Ca-ATPase (PMCA) and store-operated
Ca2⫹ entry (SOCE) can also be activated upon release of SR Ca2⫹. In
muscle, the SR Ca2⫹ sensor responsible for rapidly activated SOCE
appears to be the stromal interacting molecule 1L (STIM1L) isoform
of STIM1 protein, which directly interacts with the Orai1 Ca2⫹
channel in the t-system. The common isoform of STIM1 is STIM1S,
and it has been shown that STIM1 together with Orai1 in a complex
with the partner protein of STIM (POST) reduces the activity of the
PMCA. We have previously shown that Orai1 and STIM1 are upregulated in dystrophic mdx mouse muscle, and here we show that
STIM1L and PMCA are also upregulated in mdx muscle. Moreover,
we show that the ratios of STIM1L to STIM1S in wild-type (WT) and
mdx muscle are not different. We also show a greater store-dependent
Ca2⫹ influx in mdx compared with WT muscle for similar levels of SR
Ca2⫹ release while normal activation and deactivation properties were
maintained. Interestingly, the fiber-averaged ability of WT and mdx
muscle to extrude Ca2⫹ via PMCA was found to be the same despite
differences in PMCA densities. This suggests that there is a close
relationship among PMCA, STIM1L, STIM1S, Orai1, and also POST
expression in mdx muscle to maintain the same Ca2⫹ extrusion
properties as in the WT muscle.
C568
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
indicator of cytoplasmic [Ca2⫹] and left uncalibrated. During imaging, a
small volume of concentrated cyclopiazonic acid (CPA) was added to the
bathing solution to achieve a spatially averaged concentration of 100 ␮M
in the bath. This final concentration of CPA caused a net leak of Ca2⫹
from SR to the cytoplasm of intact fibers. Preliminary experiments with
lower concentrations of CPA (20 to 50 ␮M) with fibers loaded with
fluo-4 were imaged on a confocal microscope (as described below)
indicated that these concentrations of CPA were submaximal for blocking
the SR Ca2⫹ pump.
Ca2⫹ imaging in skinned fibers. The method of trapping fluorescent
dye in the sealed t-system for imaging Ca2⫹ movements across this
membrane has been described previously (20). Briefly, small bundles
of fibers from the EDL muscles were isolated and exposed to a “dye
solution” via a microcap pipette while still intact and under oil. The
dye-containing solution was a Na⫹-based physiological solution that
was applied before skinning of mouse fibers contained the following
(in mM): 145 NaCl, 3 KCl, 2.5 CaCl2, 2 MgCl2, 1 fluo-5N salt, and
10 HEPES (pH adjusted to 7.4 with NaOH). Dye was allowed 10 min
or more to diffuse throughout the t-system. Segments of individual fibers
were then isolated and mechanically skinned to completely remove the
surface membrane and trap dye in the sealed t-system (20). This procedure exposed the internal environment of the fiber to the very small
amount of the dye solution for a brief period of time, if any. Skinned
fibers were transferred to a custom-built experimental chamber with a
coverslip bottom, where they were bathed in an “internal solution.”
The standard internal solution contained the following (in mM):
48.5 HDTA2⫺ (Fluka, Buchs, Switzerland), 8 total ATP, 10 creatine
phosphate, 36 Na⫹, 126 K⫹, 8.5 total Mg2⫹, 1 total EGTA, 90
HEPES, 0.0001 Ca2⫹, and 0.02 rhod-2 (pH 7.1). This solution had an
osmolality of 295 ⫾ 5 mosmol/kgH2O and calculated free Mg2⫹
concetration ([Mg2⫹]) of 1 mM. Internal solutions with [Ca2⫹] raised
to 800 and 200 nM were used to load SR and t-system with Ca2⫹.
“Low Mg2⫹ solution” was similar to the standard internal solution,
but free [Mg2⫹] was reduced to 10 ␮M and was nominally free of
Ca2⫹. n- BTS (50 ␮M) was added to all solutions to suppress
contraction.
Confocal imaging. The experimental chamber containing intact
fibers was placed above the water immersion objective (⫻60, NA 1.0)
of the confocal laser scanning system (FV1000; Olympus, Tokyo,
Japan). Simultaneous acquisition of two images (F1 and F2) was
Add
CPA
A
Add
CPA
B
0.9
0.9
0.8
0.8
mdx
R
R
WT
0.7
0.7
0.6
0.6
0.5
0.5
time, min
time, min
2⫹
Fig. 1. Extrusion of cytoplasmic Ca via plasma membrane Ca-ATPase (PMCA) occurs at similar rates in wild-type (WT) and mdx fibers. Changes in
cytoplasmic fura-2 fluorescence ratio (R) from enzymatically isolated WT (A) and mdx (B) fibers during the addition of 100 ␮M cyclopiazonic acid (CPA) to
the 0 Ca2⫹, K⫹-based physiological solution are shown. Note the addition of CPA induces a rise in cytoplasmic Ca2⫹ via a net leak through the sarcoplasmic
reticulum (SR) Ca2⫹ release channel [ryanodine receptor (RyR)] during the blockade of the SR Ca2⫹ pump. The declining phase of the transient is due to the
extrusion of Ca2⫹ from the fibers by PMCA (see text).
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Perth, WA, Australia) mice were killed by cervical dislocation, and
the interossei muscles or the extensor digitorum longus (EDL) muscles were rapidly excised. All protocols were approved by the Animal
Ethics Committee of the University of Queensland. Interossei muscles
were placed in DMEM containing 1.5 mg/ml collagenase (type 1A,
Sigma) for 30 min at 30°C. Fibers were then washed with DMEM and
separated with a fire-polished glass pipette and placed into DMEM
supplemented with 5% fetal calf serum (GIBCO) and 1% penicillin
and streptomycin (GIBCO) and incubated at 30°C and 5% CO2
overnight to ensure temperature-induced reactive oxygen species
production was kept minimal (12). The isolated fibers were collected
from the bottom of the Petri dish with a Pasteur pipette and transferred
to the experimental chamber. To load fura-2 into the cytoplasm of
isolated fibers, they were exposed to a physiological solution containing the following (in mM): 4 K⫹, 140 Na⫹, 2 Ca2⫹, 1 Mg2⫹, 146 Cl⫺,
10 HEPES, and 0.01 fura-2 AM (pH adjusted to 7.4 with NaOH) for
20 min. The experimental chamber consisted of three wells, where the
central well had a thin coverslip as a base to allow for imaging of
isolated fibers. Two smaller wells on either side were connected by a
groove to the central well to allow solution exchange. To perform a
manual solution exchange, the experimental chamber was placed
under a compound microscope and solution exchanges were performed
by carefully adding a new solution to one small well and removing
solution from the other so that the new solution would be drawn in to the
central well while the original central well solution was drawn out.
Fibers loaded with fura-2 were imaged in a K⫹-based “depolarizing
solution” containing the following (in mM): 147 K⫹, 4.5 Mg2⫹, 156
Cl⫺, 10 HEPES, 1 EGTA, 0.05 n-benzyl-p-toluene sulphonamide
(n-BTS), and pH adjusted to 7.4 with KOH on an inverted Nikon
wide-field fluorescence microscope. Note that n-BTS suppresses contraction in fast-twitch muscle fibers without interfering with the Ca2⫹
movements (5). We note that the use of Cl⫺ as the anion in the 0 Ca2⫹
physiological solution may cause a small fiber volume change (consistent across all experiments), which could be avoided in future
studies by the use of an impermeant anion such as sulfate (3). Fura-2
was excited at 340 and 380 nm by sequential, 5-ms long flashes from
a DG4 monochromator directed through a ⫻20 air immersion lens
onto the preparation. Emitted light was collected by a camera attached
to a port of the microscope via a pathway back through the objective.
The ratio of the fluorescence signals (340/380 nm; R) was used as an
C569
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
PMCA activity in WT and mdx muscle. To isolate PMCA
activity in WT and mdx fibers from other cytoplasmic Ca2⫹
removal mechanisms, we tracked cytoplasmic fura-2 fluorescence during a CPA-induced Ca2⫹ transient with the fiber
bathed in a 0 Ca2⫹, K⫹-based physiological solution. Before
addition of CPA, individual fibers were equilibrated in the 0
Ca2⫹, K⫹-based physiological solution for several minutes
following transfer from a standard Ca2⫹-containing, Na⫹based physiological solution. Ca2⫹ was expected to leak from
the fiber in the absence of external Ca2⫹ (29). For this reason,
we expected initial [Ca2⫹]SR not to be the same in all fibers
exposed to CPA in the experiment described below. Fura-2
fluorescence, expressed as ratio R of fluorescence emitted from
fura-2 excited at 340 and 380 nm, was imaged for 1 min or so
A
0.8
(7)
(8)
intial R
0.6
0.4
0.2
0.0
WT
B
Rate constants, min-1
RESULTS
to establish a baseline of cytoplasmic Ca2⫹ before an addition
of CPA to the bathing solution. The addition of CPA (100 ␮M)
induced a Ca2⫹ transient in the cytoplasm of the fiber due to its
blockade of the SR Ca2⫹ pump, which led to a net loss of Ca2⫹
from SR via leak through the SR Ca2⫹ release channels/RyRs.
The rise of cytoplasmic Ca2⫹ was expected to induce binding
of Ca2⫹ to cytoplasmic buffers, troponin C and mitochondria.
In some fibers, a high peak of cytoplasmic Ca2⫹ was observed
following CPA addition (where R ⬎ 1, not shown) and a fast
decline in R was observed probably due to Ca2⫹ equilibration
with the cytoplasmic buffers and extrusion through active
PMCA. A later, continuing, slower decline in cytoplasmic
Ca2⫹ was considered to be largely due to PMCA activity. This
decline in R could not be due to SR Ca2⫹ pump (which was
blocked by CPA) or to Na⫹/Ca2⫹ exchange (because there was
no inward Na⫹ gradient across the surface membrane). Examples of this experiment in WT and mdx fibers are shown in Fig.
1. Note that the CPA-induced transient took ⬎1 h to decline
because of the large Ca2⫹ content of the skeletal muscle fibers
(14) and the small capacity of PMCA for Ca2⫹ translocation.
We also note that, consistent with the large amount of Ca2⫹
stored in skeletal muscle fibers, the release of this Ca2⫹ to the
cytoplasm in the presence of CPA induced a rapid loss of
membrane integrity and hypercontracture in a large (and unrecorded) number of isolated fibers.
From the experiments shown in Fig. 1 and similar ones, we
were able to determine the initial R in the 0 Ca2⫹, K⫹-based
0.05
mdx
(6)
0.04
(7)
0.03
0.02
0.01
0.00
WT
mdx
Fig. 2. Summary of the changes in cytoplasmic R in WT and mdx fibers during
the addition of CPA. A: initial R in the cytoplasm of WT and mdx fibers bathed
in 0 Ca2⫹, K⫹-based physiological solution for 10 min or more. B: rate
constants of the exponential functions fitting to the declining phase of CPAinduced transients from values of R that were 0.85 or lower in the cytoplasm
of WT and mdx fibers. In all cases, there was no statistical different between
the parameters (P ⬎ 0.1 in all cases).
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
achieved by sequential line-interleaving of two excitation wavelengths (488 and 543 nm) and collecting emitted light in two emission
ranges (500 –540 and 562– 666 nm) at 2.64 ms/line. F1 was the image
of the fluo-5N Ca2⫹-dependent fluorescence from the cytoplasm of
intact fibers. The initial F1 or F2 in the resting fiber was assigned as
F0. For skinned fiber experiments, the experimental chamber with a
mounted skinned fiber was imaged on the same confocal microscope
and wavelengths but at 2 ms/line. F1 and F2 of fluo-5N and rhod-2,
respectively, were normalized by dividing the average resting value
F1,0 or F2,0 within the borders of the preparation for each xy image in
a series.
Western blotting. EDL muscles from WT and mdx mice that were
used in our previous study (11) are shown here for STIM1L protein
expression. In that study, STIM1 was analysed using the antibody
commercially available from BD Biosciences (mouse monoclonal,
catalog no. 610955; BD Biosciences, Lexington, KY). It has now been
shown that the higher molecular mass band detected at ⬃110 kDa
with this antibody is in fact STIM1L and the band we have already
described, which migrates at ⬃90 kDa, is what we now will refer to
as STIM1S (6). For STIM1, increasing amounts of total protein
(ranges between 1.7 and 20 ␮g) from mdx and WT EDL skeletal
muscle were separated on 8% SDS-PAGE and probed for STIM1. For
determination of PMCA protein, the procedure was similar to that
described for STIM1 including being mixed 1:3 with solubilizing
buffer (0.125 M Tris·Cl pH 6.8, 4% SDS, 10% glycerol, 4 M urea,
10% mercaptoethanol, and 0.001% bromophenol blue; Ref. 11).
Whole tissue samples were prepared to ensure no loss of sample, and
it is assumed that the PMCA identified constituted a membrane
associated protein. Of note, samples were not heated at any stage of
preparation. Samples were stored at ⫺20°C until analyzed by Western
blotting as previously described (11), and 4 –15% Stain Free TGX gels
were used for protein separation (27), which allows the total protein
in a sample to be seen on a gel image before transferring to nitrocellulose. The amounts of the abundant muscle protein myosin heavy
chain (MHC) in the gel are then used as an indicator of the total
protein in a sample. Following probing with relevant primary (antiPMCA, Abcam ab 2825, clone 5F10; Abcam, Cambridge, UK) and
secondary antibodies and washes, the relative positions of molecular
mass markers were visualized and image captured under white light
before chemiluminescent imaging that was taken without moving the
membrane; the separate images were superimposed (see Fig. 6).
Bands were visualized using West Femto chemiluminescent substrate
(ThermoScientific), and densitometry was performed using ImageLab
Software (Bio-Rad) where the background was excluded from the
density values obtained. The PMCA signal was expressed relative to
the amount of MHC observed in the Stain Free gel for a given sample.
Statistics. Results are presented as means ⫾ SE. Unpaired, twotailed Student’s t-tests were used to determine statistical significance
between groups, where P ⬍ 0.05.
C570
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
(F2/F2,0) during SOCE because the input of Ca2⫹ from the
t-system is very small compared with the amount of Ca2⫹
already in the cytoplasm (21). As the first cell-wide transient
ended, there was an abrupt end of the Ca2⫹ loss from the
t-system and a net recovery. This sudden change in direction of
the t-system Ca2⫹-dependent fluorescence signal from a gradual decrease to an abrupt increase has been shown to be due to
SOCE deactivation causing Orai1 Ca2⫹ channels on the t-system to shut, allowing a net reaccumulation of Ca2⫹ into the
t-system immediately following the resequestration of Ca2⫹ by
the SR during the declining phase of the cell-wide transient
(22, 23). At the end of the spatially averaged profile in Fig. 3C
(after the 60-s mark), there are two briefer Ca2⫹ transients.
These transients are Ca2⫹ waves that pass the length of the
fiber. The spatially averaged profiles of the two Ca2⫹ waves
are shown on an expanded time base in Fig. 4A, and images of
the first wave are shown in Fig. 5. The fronts of the cytoplasmic Ca2⫹ waves are due to SR Ca2⫹-release and are closely
followed by a depletion of [Ca2⫹]t-sys. Conversely, the decrease of the cytoplasmic Ca2⫹ waves is due to net SR
Ca2⫹-uptake and is similarly closely followed by the rise of
t-system Ca2⫹-dependent fluorescence. The close association
of cytoplasmic Ca2⫹ changes and t-system Ca2⫹ changes are
due to the tight coupling of [Ca2⫹]SR with SOCE activation
and deactivation (10, 11, 13). SOCE activation is observed
within milliseconds of the rise of the Ca2⫹ wave because of a
A
[Ca2+]cyto
B
20 µm
200 ms
A
58 s
B
64 s
F1/F1,0 ([Ca2+]t-sys)
C
low Mg2+
1.1
3.0
1.0
2.5
0.9
0.8
2.0
0.7
1.5
0.6
0.5
F2/F2,0 ([Ca2+]cyto)
[Ca2+]t-sys
1.0
0.4
0.3
0.5
time, s
2⫹
Fig. 3. Store-dependent Ca entry in mdx-skinned fibers. Selected, simultaneously acquired confocal xyt images of cytoplasmic rhod-2 fluorescence [F2(x, y);
A] and tubular (t-) system fluo-5N fluorescence [F1(x, y); B] from a skinned muscle fiber from mdx muscle immediately following changing the internal bathing
solution from one containing 1 to 0.01 mM Mg2⫹ to induce Ca2⫹ release. Note that time stamps on A correspond to points of the spatially averaged profile in
C. C: spatially averaged cytoplasmic rhod-2 fluorescence (F2/F2,0; ) and t-system fluo-5N fluorescence (F1/F1,0; Œ) during the change of the internal solution
from 1 to 0.01 mM Mg2⫹ to induce Ca2⫹ release. The solution change is indicated in line profile above the plot and by the vertical, light grey bar. Note there
are 3 Ca2⫹ release events: a cell-wide Ca2⫹ transient, followed by 2 propagating Ca2⫹ waves. [Ca2⫹]cyto and [Ca2⫹]t-sys, Ca2⫹ concentration in cytoplasm and
t-system, respectively.
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
physiological solution and the rate of PMCA activity by fitting
exponential functions to the declining phase of the transient.
These results are shown in Fig. 2. Each parameter for WT and
mdx fibers was found not to be statistically different (P ⬎ 0.1
in each case).
Store-dependent Ca2⫹-influx in WT and mdx muscle during
Ca2⫹ release. We are able to track the store-dependent Ca2⫹
influx during SR Ca2⫹ release by imaging fluo-5N trapped in
the t-system of mechanically skinned fibers (20) while simultaneously tracking rhod-2 in the cytoplasm (11, 13, 22). An
example of such an experiment is shown in Fig. 3. Figure 3, A
and B, shows selected confocal images from the xyt series of
fluo-5N fluorescence emitted from the t-system (F1) and the
rhod-2 fluorescence signal emitted from the cytoplasm (F2) of
a mechanically skinned fiber from mdx mouse during low
Mg2⫹-induced SR Ca2⫹ release. In Fig. 3C, the spatially
averaged profile derived from the full complement of fluorescence images emitted from t-system fluo-5N and cytoplasmic
rhod-2 is shown. The rhod-2 fluorescence signal indicates that
upon lowering of cytoplasmic [Mg2⫹], there was long-lasting
cell-wide release of SR Ca2⫹. The simultaneous fluo-5N fluorescence signal from the sealed t-system indicated a [Ca2⫹]t-sys
that was stable for ⬃20 s after which it began to decline. The
decrease of [Ca2⫹]t-sys indicates the activation of SOCE. The
t-system fluo-5N fluorescence signal continued to decline during the cell-wide release of Ca2⫹ for ⬃20 s. Note that there
was no observable increase in the cytoplasmic Ca2⫹ transient
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
2.5
0.85
0.80
2.0
0.75
0.70
1.5
F2/F2,0 ([Ca2+]cyto)
F1/F1,0 ([Ca2+]t-sys )
A
0.65
1.0
0.60
time, s
SOCE flux, au
0.15
(5)*
0.10
(7)
0.05
0.00
WT
mdx
2⫹
Fig. 4. Store-dependent Ca entry is increased in mdx-skinned fibers at low
SR Ca2⫹ release amplitude. A: the Ca2⫹ waves from Fig 3A shown on an
expanded time base (F1/F1,0; Œ; F2/F2,0; ). B: histogram of the amplitude of
the t-system Ca2⫹ depletions [store-operated Ca2⫹ entry (SOCE) flux] at
cytoplasmic Ca2⫹ amplitudes (F2/F2,0) values of 1 or less (*P ⬍ 0.05; au,
arbitrary units).
low SR Ca2⫹ content following the large loss of Ca2⫹ to the
bathing solution during the cell-wide Ca2⫹ release (21).
At the time that Ca2⫹ waves are propagating along the fiber,
the Ca-buffering power of calsequestrin has decreased significantly due to its depolymerization in the lowered calcium
environment of the fiber following the large release and loss of
Ca2⫹ to the low Mg2⫹ solution (22, 23). The decreased
Ca-buffering power of the SR allows [Ca2⫹]SR to change more
rapidly during SR Ca2⫹ release (23). The result of this on
store-dependent Ca2⫹ entry is clear from a comparison of the
rates of [Ca2⫹]t-sys decline during the cell-wide release and
during the Ca2⫹ waves (Fig. 3). This phenomenon is consistent
across WT and mdx fibers, suggesting calsequestrin is functional as a dynamic Ca-buffer and an indirect regulator of
store-dependent Ca2⫹ entry (13) in both cases.
Figure 4A shows that the amplitude of each cytoplasmic
Ca2⫹ wave to be roughly proportional to the decrease in
t-system Ca2⫹ (measured at nadir) that it is associated with.
This would indicate that the amplitude of the Ca2⫹ wave is
proportional to the depletion of SR Ca2⫹, which activates
SOCE in proportion and leads to the observed decrease in
t-system Ca2⫹, as we have shown before (10). The relationship
is sigmoidal over the range of Ca2⫹ levels inside SR (10, 25)
and thus equivalent levels of SR Ca2⫹ release (the cytoplasmic
Ca2⫹ wave amplitude as reference) can be used to compare
SOCE influx in mdx and WT muscles. For a fair comparison of
the store-dependent Ca2⫹ influx (SOCE flux) in WT and mdx
fibers, we only used Ca2⫹ wave-t-system depletion (SOCE
influx) pairs within an overlapping range of SR Ca2⫹ release,
where Ca2⫹ waves had amplitudes of 1 or less. From analysis
of the Ca2⫹ wave-t-system depletion pairs, SOCE influx was
found to be significantly increased in mdx compared with WT
for the same degree of SR Ca2⫹ release (Fig. 4B; P ⬍ 0.05,
t-test). This means that the SOCE flux is greater in mdx fibers
for the same level of SR Ca2⫹ depletion. Note that WT fibers
displayed a broader range of SR Ca2⫹ wave amplitudes than
the mdx fibers when exposed to low Mg2⫹. This is most likely
because Ca2⫹ release is generally depressed in mdx fibers
compared with WT fibers (11). We observed that the maximal
SOCE flux in WT to be greater than that observed in mdx
fibers, as the amplitude of Ca2⫹ waves in WT fibers increased
above that achievable in mdx fibers (not shown).
PMCA, STIM1, Orai1, and RyR are located throughout the
junctional membranes. The series of images in Fig. 5 represent
⌬Fx(x,y) during a Ca2⫹ wave. The images immediately preceding the wave [which were quiescent, Fx,0(x,y)] have been
subtracted from Fx(x,y) to produce ⌬Fx(x,y) where inhomogeneities along the images should be reduced. These images are
derived from the first Ca2⫹ wave displayed as a spatially
averaged profile in Fig. 3C. From these images, we can acquire
high spatiotemporal information in regards to the movements
of Ca2⫹ across the junctional membranes (13). Ca2⫹ release
and resequestration of Ca2⫹ from and to SR can be derived
from the images in A [⌬F2(x,y)]. These events involve the
RyRs and the SR Ca2⫹ pumps. In the simultaneously acquired
images in B [⌬F1(x,y)], Ca2⫹ entry to the cytoplasm via SOCE
and reaccumulation from cytoplasm to the t-system are observed. These events involve the STIM1, Orai1, and PMCA
proteins, as discussed below.
The advent of a cytoplasmic Ca2⫹ wave was observed in the
image at 1 s, and this wave propagates from the right-to-left in
the image at 2 s and continued in this manner for the duration
of the series. There is a tight association of the cytoplasmic
Ca2⫹ wave and the depression of the Ca2⫹ signal derived from
the t-system. Note that the amplitude of the cytoplasmic Ca2⫹
wave was smaller in the image at 1 s and this was rapidly
followed by a relatively small depression of the t-system
signal. The increase in Ca2⫹ release from SR in the next image
at 2 s is associated with a greater depression of the t-system
signal. Throughout the series of images, the front of the Ca2⫹
wave was associated with the beginning of the depression of
the t-system Ca2⫹ signal. At the end of the cytoplasmic Ca2⫹
wave, the SR starts to refill with Ca2⫹ (23) causing SOCE
deactivation and net movement of Ca2⫹ into the sealed t-system by PMCA. The tight association between the concomitant
rise and fall of these signals indicates that the RyR, STIM1,
Orai1 and PMCA are arranged throughout the junctional membranes of skeletal muscle in a uniform manner.
PMCA density in WT and mdx muscle. PMCA exists as a
number of different isoforms and multiple isoforms are detected by the antibody used here (Abcam ab 2825, clone 5F10).
The PMCA isoform seen in total brain homogenates migrated
as a single band (Fig. 6A) at an apparent molecular mass of
⬃140 kDa and was used as a positive control for migration of
the ⬃140 kDa band seen in the skeletal muscle samples.
Hence, the band that migrated at the same distance as a brain
total homogenate was taken to represent the PMCA isoform of
interest. Following the convention of molecular mass used in
the literature and the differences attributable to molecular mass
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
B
C571
C572
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
A
[Ca2+ ] cyto
1s
2s
3s
B
20 µm
[Ca2+]t-sys
200 ms
A
4s
5s
6s
B
markers and specific gels used, the band will be referred to as
⬃140 kDa. Western blotting revealed that EDL muscle from
mdx mice had more than twice the amount of ⬃140-kDa
PMCA protein compared with MHC than that seen in EDL
muscle from WT animals (Fig. 6).
STIM1L density in WT and mdx muscle. To understand the
regulation of the store-dependent influx in WT and mdx mouse
muscle, it was important to know whether the recently identified STIM1 isoform STIM1L (6) was also upregulated in mdx
muscle. While the functional role(s) of the two STIM isoforms
remain unclear, STIM1L may be specifically responsible for
the rapid activation of SOCE in skeletal muscle and act
independently from the ubiquitously expressed STIM1 (to now
be defined as STIM1S; “L” and “S” stand for “long” and
“short,” respectively). We (11) have previously shown Western
blots for the ⬃90-kDa STIM1S protein. Reanalysis of the
larger molecular mass ⬃110-kDa band seen (but not previously reported) showed that, similar to STIM1S, there is
⬃1.8-fold more STIM1L in EDL muscle from mdx compared
with WT mice (Fig. 7). To evaluate the relative abundance of
each of the isoforms, the binding efficiency of the antibodies
would ideally be known to be equal. We can, however, using
the similar slopes for the linear regressions for STIM1L and
STIM1S between muscle from mdx and WT (Fig. 7B), suggest
a similar abundance of the isoforms in the sample examined. It
was also possible to examine the ratio of STIM1L to STIM1S
between mdx and WT mice samples run on the same gel.
Examination of the STIM1L:STIM1S revealed no difference
between EDL muscles obtained from mdx and WT animals
(1.1 ⫾ 0.3; n ⫽ 4; P ⫽ 0.72, unpaired, two-tailed t-test).
DISCUSSION
In this study, we show that there are increased levels of
PMCA and STIM1L proteins (Figs. 6 and 7), and previously
we (11) have shown increased STIM1S and Orai1 protein,
expressed in mdx compared with WT muscle. These differences in the respective protein expression levels between WT
and mdx fibers have the potential of changing the landscape of
Ca2⫹ fluxes that occur within the junctional membranes of the
two muscles. In turn, differences in the [Ca2⫹] within microdomains of the WT and mdx muscle fibers (Fig. 8) may influence
the expression levels of Ca2⫹ regulatory proteins within the
junctional membranes of skeletal muscle.
Importantly, imaging of Ca2⫹ movements across the t-system and SR of intact and skinned fibers from WT and mdx
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 5. RyR, PMCA, Orai1, and stromal
interacting molecule 1 (STIM1) are arranged
uniformly throughout the junctional membranes. Images derived from confocal images of simultaneously acquired cytoplasmic
rhod-2 fluorescence (A) and t-system fluo5N fluorescence (B; see text) during a Ca2⫹
wave are shown (this is the first wave shown
in profile in Fig 3A) in a mechanically
skinned mdx fiber. Below each pair of images are spatially averaged profiles from
within the borders of the preparation in x of
the signals from the cytoplasm (red) and
from the t-system (green). These lines are
spatiotemporal profiles, as each line of the
image in y is also a dimension of t (t is
determined by the rate of sequentially acquired confocal line scans in y). The wave
originates in the image at 1 s and then
propagates from right to left. Note that the
wave remains tightly associated with a depression in the fluo-5N signal indicating that
Ca2⫹ moves from t-system to cytoplasm
closely following the rise of the Ca2⫹ wave
and the t-system reaccumulates Ca2⫹ during
the decline of the wave. The tight association
between SR Ca2⫹ release and sequestration
with SOCE activation and deactivation (allowing PMCA activity to be observed) along the
entire length and width of the fiber indicates
that RyR, PMCA, STIM1, and Orai1 must be
positioned uniformly and evenly throughout
the junctional membranes (see text).
C573
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
muscle (Figs. 1–5) permitted the observed changes in protein
expression to be related to changes in the Ca2⫹ movements
associated with these proteins. Thus, consistent with an increase in STIM1, there was an increased store-dependent Ca2⫹
influx in mdx muscle for a given SR Ca2⫹ depletion. However,
the increased PMCA expression was not associated with an
increased activity of cytoplasmic Ca2⫹ removal. As detailed
below, our view is that the increased store-dependent Ca2⫹
influx in mdx fibers is necessary to compensate for the increased leak through the RyRs (2). An increased store-dependent Ca2⫹ influx is achieved by increasing STIM1 and Orai1
expression, but increased STIM1 expression reduces PMCA
activity. Therefore, PMCA expression in the mdx fiber appears
to be increased to maintain an overall PMCA activity similar to
that in the WT muscle.
STIM1, Orai1, PMCA, and RyR are throughout the junctional membranes of skeletal muscle, affecting Ca2⫹ signaling.
By simultaneously imaging cytoplasmic Ca2⫹ waves and t-system Ca2⫹ in skinned fibers, we are able to track the coupling of
Ca2⫹ release from and recovery to SR with the store-dependent
Ca2⫹ influx and net accumulation of Ca2⫹ by the t-system with
high spatiotemporal resolution. The tight association of SR
Ca2⫹ release with the depression and recovery of t-system
Ca2⫹ consistently across the radial aspect of the fiber and the
longitudinal aspect of the fiber as the Ca2⫹ wave propagated
can only be interpreted as having the Ca2⫹ signaling proteins
A
B
100
wt - Stim1S
Relative amount Stim1L
(relative to wt)
density STIM1 - L or S
(arbitrary units)
C
mdx - Stim1S
mdx - Stim1L
wt - Stim1L
80
60
40
20
2.0
1.5
1.0
0.5
5
5
wt
mdx
0.0
0
0
5
10
15
µg muscle
20
25
Fig. 7. STIM1L is upregulated in skeletal
muscle from mdx mice. A: decreasing amounts
of total protein (20 to 3.4 ␮g) from mdx (left)
and WT (right) EDL skeletal muscle was separated on 8% SDS-PAGE and probed for
STIM1. It has been shown that the higher
molecular weight band detected at ⬃110 kDa
with this antibody is in fact STIM1L and the
band we have already described, which migrates at ⬃90 kDa, is what we now refer to as
STIM1S (see METHODS, 90-kDa STIM1S data
shown in Ref. 11). MHC in the posttransferred
Coomassie blue gel is also shown (top) and is
indicative of the relative amount of protein
loaded. B: Linear regressions for arbitrary
units of densities where the 20 ␮g sample
from mdx mice was given a value of 100 for
each isoform and the amount of STIM1L and
STIM1S protein for the samples shown in A.
Slopes of the regressions and the r2 values are
as follows: STIM1L: mdx, y ⫽ 6.0x-18, r2 ⫽
0.99; WT, y ⫽ 2.5x-15, r2 ⫽ 0.99. STIM1S:
mdx, y ⫽ 5.3x-2.3, r2 ⫽ 0.98; WT, y ⫽
3.3x-19, r2 ⫽ 1.0. C: plot showing amount of
STIM1L in WT and mdx EDL muscle, expressed relative to the amount of STIM1L
protein in the WT muscle on a given gel.
Number of samples analysed indicated in parentheses. *P ⬍ 0.05, unpaired Student’s twotailed t-test.
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 6. PMCA protein is upregulated in skeletal muscle from mdx mice. A: whole homogenates from rat brain (⬃1 ␮g) and extensor digitorum longus (EDL)
muscle (⬃10 ␮g) from WT and mdx mice were separated on a 4 –15% Criterion Stain Free TGX gel and probed with anti-PMCA. Molecular mass markers
indicated from the white light image (see METHODS). In brain a single band at ⬃140 –150 kDa is seen, which corresponds with a band in skeletal muscle (arrows).
Myosin heavy chain (MHC) in the stain-free gel (top) indicates the relative amount of protein loaded in the skeletal muscle lanes. B: plot of the density of PMCA
protein (⬃140 kDa) normalized to MHC for WT and mdx mice, expressed relative to the amount of each protein in the WT muscle on a given gel. Number of
samples analysed indicated in parentheses. *P ⬍ 0.05, unpaired Student’s two-tailed t-test.
C574
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
responsible located throughout the junctional membranes in a
somewhat uniform fashion. This is the case in both WT and
mdx fibers (Fig. 5) (13). The RyR is responsible for SR Ca2⫹
release and is directly opposite the voltage sensor/dihydropyridine receptor in the t-system membrane. STIM1 and Orai1
must be opposite each other in the same immediate area for
activation of SOCE, and PMCA must also be in the same
location to account for the reaccumulation of Ca2⫹ into the
t-system following SOCE deactivation. We note that previously we (22) have shown that such reaccumulation of Ca2⫹
into the t-system occurs in the absence of a Na⫹ gradient
(removing the Na⫹/Ca2⫹ exchanger function), thus isolating
PMCA. The proximity of PMCA and STIM1 within the junctional membranes of skeletal muscle allows these proteins to
interact to shape the Ca2⫹ signals in this microdomain. The
higher level of expression of PMCA and STIM1 in mdx fibers
but similar levels of PMCA activity for the same [Ca2⫹]cyto as
WT (Figs. 1, 2, 6, and 7) strongly suggest that a (transient)
physical interaction must exist between these proteins, and it
follows therefore that STIM1 and PMCA must exist opposite
each other at some point across the junctional membranes. An
inhibitory action of a complex of STIM1-POST on PMCA
function as shown in other cells (18) appears to be at work in
skeletal muscle as well.
The higher density of STIM1 in mdx compared with WT
fibers was consistent with the observation of a higher storedependent Ca2⫹ influx for the same degree of SR Ca2⫹ release
(Figs. 4 and 7). We must point out that as Ca2⫹ release
amplitude increased in WT fibers beyond levels achievable by
mdx fibers so did the store-dependent Ca2⫹ influx in WT fibers
(not shown). Considering this and the fact that SOCE remained
tightly regulated by SR Ca2⫹ in mdx fibers, it is difficult to
draw a conclusion that the low amplitude store-dependent
Ca2⫹ influx is contributing to any Ca2⫹ overload in mdx fibers.
A more persistent entry of Ca2⫹ through stretch-activated
channels is probably responsible for such an effect (31).
STIM1 isoforms in muscle. Very similar SOCE models
based around a preformed or uniform prepositioning of STIM1
and Orai1 clusters along the junctional membranes of skeletal
muscle have been put forward (7, 13, 19, 21, 22) to account for
the rapid activation kinetics of SOCE in this tissue (9, 13, 22).
Recently, a STIM1 splice variant STIM1L has been described
that contains actin-binding sites, which promotes permanent
clustering of STIM1L to maintain its position within the SR
terminal cisternae regardless of [Ca2⫹]SR (6). Of further interest is that STIM1L coexists with the ubiquitously expressed
STIM1S (6) in skeletal muscle and both isoforms are upregulated in mdx muscle (Fig. 7) (11). Might there be different roles
for the respective STIM1 isoforms in muscle? The storedependent Ca2⫹-influx is certainly mediated by STIM1L in
adult muscle but a contribution by STIM1S cannot be ruled
out. Darbellay et al. (6) have indeed shown that both STIM1L
and STIM1S are capable of regulating SOCE.
The recent work of Krapivinsky et al. (18) shows that the
STIM1S isoform, in a complex with POST, interacts with
PMCA. Therefore, it is possible, for example, that STIM1L is
regulating store-dependent Ca2⫹ influx while STIM1S-POST is
regulating Ca2⫹ extrusion from the space between the junctional
membranes by interacting with PMCA. This may be a key element of
Ca2⫹ signaling within the junctional membranes and may be a major
part of STIM1 signaling in muscle development (30). In dystrophic
muscle, these changed levels of Ca2⫹ signaling proteins and
therefore the reshaped signals may be a critical link in attempts
at initiating fiber regeneration in dystrophic muscle (4, 17).
STIM1-POST also regulates other key proteins such as Na⫹K⫹-ATPase and SERCA (18). It is also relevant to mention our
(10) recent demonstration that the store-dependent Ca2⫹ current remains unaffected in the aging muscle despite an ⬃40%
reduction in STIM1 protein expression in these muscles. This
result indicates an excess of STIM1 is present in the young,
healthy muscle in regards to regulating SOCE. One might
envisage that the “excess” STIM1 is regulating non-SOCE
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 8. Model of PMCA and STIM1 upregulation in
mdx fibers in response to changes in microdomain
[Ca2⫹]. Cytoplasm and SR lumen are each modeled as
2 compartments: the triad and bulk cytoplasm; and
terminal cisternae and longitudinal SR, respectively.
[Ca2⫹] can be high in the triad of mdx fibers due to a
leaky RyR (2) while the resting bulk cytoplasmic
[Ca2⫹] is not different from that in WT. SR Ca2⫹ pump
are predominantly located on the longitudinal SR (lSR)
and pump Ca2⫹ into the lumen of this section of SR that
must diffuse to the terminal cisternae of the SR (tcSR)
to be available for release. Again the leaky RyR makes
the [Ca2⫹] of the tcSR low while the lSR is normal. The
changes in [Ca2⫹] on the triad and tcSR can account for
the upregulation of STIM1 and PMCA while the SR
Ca2⫹ pump maintains normal expression level against
unchanged resting levels of lSR and bulk cytoplasmic
[Ca2⫹] (1). DHPR, dihydropyridine receptor.
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
where dyes have been microinjected or the fibers have been
shown to have functional EC coupling (17, 24). We note that
while the [Ca2⫹] of the bulk cytoplasm of mdx compared with
WT muscle fibers has been controversial (15), these results
could be due to eccentric contractions causing changes in
membrane properties, inconsistent Ca2⫹ dye loading properties
in WT and mdx fibers, or weakened surface membrane in mdx
fibers following enzymatic dissociation inducing a nonphysiological net Ca2⫹ entry into mdx fibers.
Here we have shown that there are multiple changes in the
expression levels of Ca2⫹ regulatory proteins in mdx muscle
compared with the WT. On this background of Ca2⫹ regulatory
protein changes in mdx muscle, we observed changes in the
movements of Ca2⫹ across the t-system membrane. We have
shown an increased store-dependent Ca2⫹ influx for similar
levels of SR Ca2⫹ release but a similar level of PMCA activity
in mdx and WT muscle. These movements of Ca2⫹ can be
explained given recently described interactions among STIM1,
Orai1, PMCA, and other accessory proteins (18). The changed
Ca2⫹ signals in the triad of mdx fibers may be attempts by the
muscle to adapt to its diseased state.
GRANTS
This work was supported by National Health and Medical Research Council
(NHMRC; Australia) Project Grants (to B. S. Launikonis, D. G. Stephenson
and R. M. Murphy); the Leonie Stanley Memorial Fund of Muscular Dystrophy, Queensland (Australia; to B. S. Launikonis); a Duchenne Foundation
(Australia) Grant (to T. R. Cully); and an Australian Research Council
International Fellowship (to O. Friedrich and B. S. Launikonis). J. N. Edwards
was a CJ Martin Fellow of the NHMRC.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
AUTHOR CONTRIBUTIONS
Author contributions: T.R.C., R.M.M., and B.S.L. conception and design of
research; T.R.C., J.N.E., O.F., and R.M.M. performed experiments; T.R.C.,
R.M.M., and B.S.L. analyzed data; T.R.C., D.G.S., R.M.M., and B.S.L.
interpreted results of experiments; T.R.C., R.M.M., and B.S.L. prepared
figures; T.R.C., R.M.M., and B.S.L. drafted manuscript; T.R.C., D.G.S.,
R.M.M., and B.S.L. edited and revised manuscript; T.R.C., J.N.E., O.F.,
D.G.S., R.M.M., and B.S.L. approved final version of manuscript.
REFERENCES
1. Abell E, Ahrends R, Bandara S, Park BO, Teruel MN. Parallel
adaptive feedback enhances reliability of the Ca2⫹ signaling system. Proc
Natl Acad Sci USA 108: 14485–14490, 2011.
2. Bellinger AM, Reiken S, Carlson C, Mongillo M, Liu X, Rothman L,
Matecki S, Lacampagne A, Marks AR. Hypernitrosylated ryanodine
receptor calcium release channels are leaky in dystrophic muscle. Nat Med
15: 325–330, 2009.
3. Boyle PJ, Conway EJ. Potassium accumulation in muscle and associated
changes. J Physiol 100: 1–63, 1941.
4. Chan S, Head SI. The role of branched fibres in the pathogenesis of
Duchenne muscular dystrophy. Exp Physiol 96: 564 –571, 2011.
5. Cheung A, Dantzig JA, Hollingworth S, Baylor SM, Goldman YE,
Mitchison TJ, Straight AF. A small-molecule inhibitor of skeletal
muscle myosin II. Nat Cell Biol 4: 83–88, 2002.
6. Darbellay B, Arnaudeau S, Bader CR, Konig S, Bernheim L. STIM1L
is a new actin-binding splice variant involved in fast repetitive Ca2⫹
release. J Cell Biol 194: 335–346, 2011.
7. Dirksen RT. Checking your SOCCs and feet: the molecular mechanisms
of Ca2⫹ entry in skeletal muscle. J Physiol 587: 3139 –3147, 2009.
8. Dowling P, Doran P, Ohlendieck K. Drastic reduction of sarcalumenin in
Dp427 (dystrophin of 427 kDa)-deficient fibres indicates that abnormal
calcium handling plays a key role in muscular dystrophy. Biochem J 379:
479 –488, 2004.
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
processes in the muscle. Therefore, although SOCE is maintained in the aged muscle, other regulatory roles of STIM1 may
no longer be supported because of the STIM1 protein downregulation.
Parallel adaptive feedback. It has been demonstrated that
cells can continue to transmit signals correctly in the presence
of higher or lower than normal resting [Ca2⫹] and expression
levels of the major Ca2⫹ signaling proteins. Feedback mechanisms detect the [Ca2⫹] of the external environment, cytoplasm, and internal Ca2⫹ store, and protein expression levels of
STIM1, PMCA, and the SR Ca2⫹ pump are adapted in parallel
(1). Evidence for a similar situation involving the Na⫹-K⫹
ATPase, PMCA, SR Ca2⫹ pump, and Na⫹/Ca2⫹ exchanger has
been put forward from work in smooth muscle cells (28). To fit
the parallel adaptive feedback model in mdx muscle with the
changes in STIM1 and PMCA expression levels (Figs. 6 and 7;
Ref. 11; but no such change in SR Ca2⫹ pump expression; Ref.
8), we must model persistent changes in resting [Ca2⫹] of
microdomains of the mdx muscle. According to the work of
Abell et al. (1), an increase in cytoplasmic [Ca2⫹] will increase
the expression of PMCA (Fig. 6) but also the SR Ca2⫹ pump
protein levels. Furthermore, a low level of Ca2⫹ in the SR will
increase STIM protein expression levels (Fig. 7) but again also
those of the SR Ca2⫹ pump. To reconcile these apparent
discrepancies in the mdx with the parallel adaptive feedback
model (1), we must view the cytoplasm and SR lumen each as
two separate compartments, creating four compartments that
Ca2⫹ can move through within the fiber (Fig. 8).
In this model, the space between the junctional membranes
is segregated from the bulk cytoplasm, allowing each to have
a different [Ca2⫹]. The [Ca2⫹] in the junctional membrane
space (triad) can be high, leading to an increase in PMCA
expression levels while the bulk cytoplasmic [Ca2⫹] remains
similar to that in the WT muscle (17, 24). With the PMCA
located on the t-system membrane (22), it is exposed to higher
[Ca2⫹] at rest whereas the SR Ca2⫹ pump, which has its
highest density back along the longitudinal SR, is exposed to
the bulk cytoplasmic solution that does not have a different
resting [Ca2⫹] from that in WT muscle. The bulk cytoplasm
and triad can have a different [Ca2⫹] because the triad is a
diffusionally restricted space. The reported leaky RyR on the
SR terminal cisternae of mdx muscle would create a persistent
and higher [Ca2⫹] in this defined region (2).
To explain the increased expression level of STIM1 and
maintained levels of the SR Ca2⫹ pump in mdx compared with
WT muscle, we have modeled the SR as two compartments,
the terminal cisternae (tcSR) and the longitudinal SR (lSR)
(Fig. 8). To be consistent with parallel adaptive feedback, the
tcSR of mdx muscle must have a lower [Ca2⫹] than that in WT
to induce an increase in PMCA protein expression and the lSR
must maintain a [Ca2⫹] that is similar in mdx and WT muscle
to maintain a similar expression level of the SR Ca2⫹ pump (1,
8). A lower [Ca2⫹] in the tcSR of mdx muscle is probably the
result of the leaky Ca2⫹ release channel on the tcSR (2). This
low [Ca2⫹] within the tcSR and high [Ca2⫹] within the triad is
not in direct contact with the majority of the SR Ca2⫹ pumps,
as they are positioned along the lSR. The SR Ca2⫹ pumps span
the lSR membrane that directly faces the bulk cytoplasm and
the [Ca2⫹] in the lumen of the lSR. The [Ca2⫹] in these two
domains must be similar in WT and mdx muscle and has been
shown with ratiometric Ca2⫹ indicators under conditions
C575
C576
PMCA AND SOCE ACTIVITY IN DYSTROPHIC MUSCLE
22. Launikonis BS, Rios E. Store-operated Ca2⫹ entry during intracellular
Ca2⫹ release in mammalian skeletal muscle. J Physiol 583: 81–97, 2007.
23. Launikonis BS, Zhou J, Royer L, Shannon TR, Brum G, Rios E.
Depletion “skraps” and dynamic buffering inside the cellular calcium
store. Proc Natl Acad Sci USA 103: 2982–2987, 2006.
24. Lovering RM, Michaelson L, Ward CW. Malformed mdx myofibers
have normal cytoskeletal architecture yet altered EC coupling and stressinduced Ca2⫹ signaling. Am J Physiol Cell Physiol 297: C571–C580,
2009.
25. Luik RM, Wang B, Prakriya M, Wu MM, Lewis RS. Oligomerization
of STIM1 couples ER calcium depletion to CRAC channel activation.
Nature 454: 538 –542, 2008.
26. Melzer W, Herrmann-Frank A, Luttgau HC. The role of Ca2⫹ ions in
excitation-contraction coupling of skeletal muscle fibres. Biochim Biophys
Acta 1241: 59 –116, 1995.
27. Murphy RM, Mollica JP, Beard NA, Knollmann BC, Lamb GD.
Quantification of calsequestrin 2 (CSQ2) in sheep cardiac muscle and
Ca2⫹-binding protein changes in CSQ2 knockout mice. Am J Physiol
Heart Circ Physiol 300: H595–H604, 2011.
28. Pritchard TJ, Bowman PS, Jefferson A, Tosun M, Lynch RM, Paul
RJ. Na⫹-K⫹-ATPase and Ca2⫹ clearance proteins in smooth muscle: a
functional unit. Am J Physiol Heart Circ Physiol 299: H548 –H556, 2010.
29. Royer L, Sztretye M, Manno C, Pouvreau S, Zhou J, Knollmann BC,
Protasi F, Allen PD, Rios E. Paradoxical buffering of calcium by
calsequestrin demonstrated for the calcium store of skeletal muscle. J Gen
Physiol 136: 325–338, 2010.
30. Stiber J, Hawkins A, Zhang ZS, Wang S, Burch J, Graham V, Ward
CC, Seth M, Finch E, Malouf N, Williams RS, Eu JP, Rosenberg P.
STIM1 signalling controls store-operated calcium entry required for development and contractile function in skeletal muscle. Nat Cell Biol 10:
688 –697, 2008.
31. Whitehead NP, Streamer M, Lusambili LI, Sachs F, Allen DG.
Streptomycin reduces stretch-induced membrane permeability in muscles
from mdx mice. Neuromuscul Disord 16: 845–854, 2006.
32. Whitehead NP, Yeung EW, Allen DG. Muscle damage in mdx (dystrophic) mice: role of calcium and reactive oxygen species. Clin Exp
Pharmacol Physiol 33: 657–662, 2006.
33. Whitehead NP, Yeung EW, Froehner SC, Allen DG. Skeletal muscle
NADPH oxidase is increased and triggers stretch-induced damage in the
mdx mouse. PLos One 5: e15354, 2010.
AJP-Cell Physiol • doi:10.1152/ajpcell.00144.2012 • www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on June 16, 2017
9. Duke AM, Hopkins PM, Calaghan SC, Halsall JP, Steele DS. Storeoperated Ca2⫹ entry in malignant hyperthermia-susceptible human skeletal muscle. J Biol Chem 285: 25645–25653.
10. Edwards JN, Blackmore DG, Gilbert DF, Murphy RM, Launikonis
BS. Store-operated calcium entry remains fully functional in aged mouse
skeletal muscle despite a decline in STIM1 protein expression. Aging Cell
10: 675–685, 2011.
11. Edwards JN, Friedrich O, Cully TR, von Wegner F, Murphy RM,
Launikonis BS. Upregulation of store-operated Ca2⫹ entry in dystrophic
mdx mouse muscle. Am J Physiol Cell Physiol 299: C42–C50, 2010.
12. Edwards JN, Macdonald WA, van der Poel C, Stephenson DG. O2·⫺
production at 37 degrees C plays a critical role in depressing tetanic force
of isolated rat and mouse skeletal muscle. Am J Physiol Cell Physiol 293:
C650 –C660, 2007.
13. Edwards JN, Murphy RM, Cully TR, von Wegner F, Friedrich O,
Launikonis BS. Ultra-rapid activation and deactivation of store-operated
Ca2⫹ entry in skeletal muscle. Cell Calcium 47: 458 –467, 2010.
14. Fryer MW, Stephenson DG. Total and sarcoplasmic reticulum calcium
contents of skinned fibres from rat skeletal muscle. J Physiol 493:
357–370, 1996.
15. Gailly P. New aspects of calcium signaling in skeletal muscle cells:
implications in Duchenne muscular dystrophy. Biochim Biophys Acta
1600: 38 –44, 2002.
16. Gehrig SM, van der Poel C, Sayer TA, Schertzer JD, Henstridge DC,
Church JE, Lamon S, Russell AP, Davies KE, Febbraio MA, Lynch
GS. Hsp72 preserves muscle function and slows progression of severe
muscular dystrophy. Nature 484: 394 –398, 2012.
17. Head SI. Branched fibres in old dystrophic mdx muscle are associated
with mechanical weakening of the sarcolemma, abnormal Ca2⫹ transients
and a breakdown of Ca2⫹ homeostasis during fatigue. Exp Physiol 95:
641–656, 2010.
18. Krapivinsky G, Krapivinsky L, Stotz SC, Manasian Y, Clapham DE.
POST, partner of stromal interaction molecule 1 (STIM1), targets STIM1
to multiple transporters. Proc Natl Acad Sci USA 108: 19234 –19239,
2011.
19. Lamb GD. Store-operated Ca2⫹ release in skeletal muscle: tailored for a
specialized system. J Physiol 583: 1, 2007.
20. Launikonis BS, Barnes M, Stephenson DG. Identification of the coupling between skeletal muscle store-operated Ca2⫹ entry and the inositol
trisphosphate receptor. Proc Natl Acad Sci USA 100: 2941–2944, 2003.
21. Launikonis BS, Murphy RM, Edwards JN. Toward the roles of storeoperated Ca2⫹ entry in skeletal muscle. Pflügers Arch 460: 813–823, 2010.