Am J Physiol Cell Physiol 293: C1906–C1915, 2007.
First published October 10, 2007; doi:10.1152/ajpcell.00241.2007.
SR Ca2⫹ refilling upon depletion and SR Ca2⫹ uptake rates during
development in rabbit ventricular myocytes
Jingbo Huang,1,2 Leif Hove-Madsen,3* and Glen F. Tibbits1,2*
1
Cardiac Membrane Research Lab, Simon Fraser University, Burnaby and 2Cardiovascular Sciences, Child and Family
Research Institute, Vancouver, British Columbia, Canada; and 3Laboratorio
de Fisiologı́a Celular, Cardiologı́a, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain
Submitted 7 June 2007; accepted in final form 3 October 2007
sarco(endo)plasmic reticulum Ca2⫹ pump; sarcoplasmic reticulum
Ca2⫹ loading; cytosolic Ca2⫹ concentration; L-type Ca2⫹ channel;
Na⫹/Ca2⫹ exchanger; neonate cardiomyocytes; store-operated Ca2⫹
channel
that a relatively small amount of Ca2⫹
that enters through L-type Ca2⫹ channels during the action
potential triggers Ca2⫹ release from the sarcoplasmic reticulum
(SR) in a process known as Ca2⫹-induced Ca2⫹ release (CICR)
in adult mammalian myocytes (6, 21). In adult rabbit ventricular myocytes, it has been reported that the Ca2⫹ released from
the SR and transmembrane Ca2⫹ influx provide ⬃70% and
30% of the total Ca2⫹ during cell contraction, respectively (8,
17). Accordingly, ⬃28% of the Ca2⫹ removed from the
cytosol during cell relaxation is extruded by the Na⫹/Ca2⫹
exchanger (NCX), and the sarco(endo)plasmic reticulum
Ca2⫹ (SERCA)2a pump transports ⬃70% of total Ca2⫹ back
into the SR (3). The sarcolemmal Ca2⫹-ATPase (plasma
membrane Ca2⫹-ATPase, PMCA) and mitochondrial Ca2⫹
IT IS WELL DOCUMENTED
* L. Hove-Madsen and G. F. Tibbits are co-senior authors of this article.
Address for reprint requests and other correspondence: Glen F. Tibbits,
Kinesiology, Simon Fraser Univ., 8888 University Drive, Burnaby, BC,
Canada V5A 1S6 (e-mail: tibbits@sfu.ca).
C1906
uniporter, collectively referred to as the slow Ca2⫹ removal
system, remove the remaining ⬃2% (2).
Ultrastructural data from the neonatal heart have led to the
suggestion that the SR is relatively sparse at birth and develops
gradually from neonate to adult through the first few weeks of
life (11, 25). Several studies have also indicated that SERCA
mRNA and protein expression levels are ⬃50% of adult levels
at birth and gradually increase to adult levels by postpartum
day 15 (9, 11), leading further support to the notion that
SERCA2a function is less important in neonates than in adults
(5, 10). Consequently, it has been assumed that the SR in
neonatal ventricular myocytes is unable to store amounts of
Ca2⫹ comparable to those of adult myocytes. However, recent
studies have challenged this assumption. Observations such as
robust and spatially homogeneous caffeine (Caf)-induced Ca2⫹
transients and contractures have been reported in neonate
hearts (1, 10, 16). Moreover, integration of the Caf-induced
inward NCX current (INCX) in neonatal rabbit ventricular
myocytes revealed that SR Ca2⫹ loading (loadSR) at rest was
significantly larger in early neonatal than in late neonatal and
adult ventricular myocytes (14), suggesting that store-operated
Ca2⫹ entry (SOCE) may contribute significantly to loadSR in
the neonate rabbit heart. The aim of the present study was
therefore to elucidate the cellular mechanisms contributing to
refilling of the SR on a beat-to-beat basis during postnatal
development.
To achieve this, we used a whole cell perforated patchclamp technique and Ca2⫹ transient measurements combined
with pharmacological manipulation of the L-type Ca2⫹ channel
and the NCX, the main sarcolemmal Ca2⫹ sources in the adult
mammalian ventricular myocytes (3, 12, 14). Our results show
that the SR Ca2⫹ pump is capable of efficiently loading the SR
with Ca2⫹ even at the earliest neonatal stage, but that the
principal Ca2⫹ sources contributing to loadSR change during
ontogeny.
MATERIALS AND METHODS
Isolation of ventricular myocytes. Animals were cared for in
accordance with the principles established by the Canadian Council
on Animal Care (CCAC). The Simon Fraser University Animal Care
Committee approved the use of animals and the experimental protocol
used in this study in accordance with CCAC regulations. Ventricular
myocytes were isolated from the hearts of New Zealand White rabbits
(of either sex) from four distinct age groups, 3 (3d), 10 (10d), 20
(20d), and 56 (56d) days postpartum, by methods previously described (12, 13, 17).
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.
0363-6143/07 $8.00 Copyright © 2007 the American Physiological Society
http://www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Huang J, Hove-Madsen L, Tibbits GF. SR Ca2⫹ refilling upon
depletion and SR Ca2⫹ uptake rates during development in rabbit
ventricular myocytes. Am J Physiol Cell Physiol 293: C1906–C1915,
2007. First published October 10, 2007; doi:10.1152/ajpcell.00241.2007.—
While it has been reported that a sparse sarcoplasmic reticulum (SR) and
a low SR Ca2⫹ pump density exist at birth, we and others have recently
shown that significant amounts of Ca2⫹ are stored in the neonatal rabbit
heart SR. Here we try to determine developmental changes in SR Ca2⫹
loading mechanisms and Ca2⫹ pump efficacy in rabbit ventricular myocytes. SR Ca2⫹ loading (loadSR) and k0.5 (Ca2⫹ concentration at halfmaximal SR Ca2⫹ uptake) were higher and lower, respectively, in
younger age groups. Inhibition of the L-type calcium current (ICa) with
15 ␮M nifedipine dramatically reduced loadSR in older but not in younger
age groups. In contrast, subsequent inhibition of the Na⫹/Ca2⫹ exchanger
(NCX) with 10 ␮M KB-R7943 strongly reduced loadSR in the younger
but not the older age groups. Accordingly, the time integral of the inward
NCX current (tail INCX) elicited on repolarization was highly sensitive to
nifedipine in the older groups and sensitive to KB-R7943 in the younger
groups. Interestingly, slow SR loading took place in the presence of both
nifedipine and KB-R7943 in all age groups, although it was less prominent in the older groups. We conclude that the SR loading capacity at the
earliest postnatal stages is at least as large as that of adult myocytes.
However, reverse-mode NCX plays a prominent role in SR Ca2⫹ loading
at early postnatal stages while ICa is the main source of SR Ca2⫹ loading
at late postnatal and adult stages.
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
Whole cell perforated-patch voltage clamp. Whole cell amphotericin-perforated voltage-clamp technique was used at room temperature as described previously (12, 13, 17). The internal pipette solution
contained (in mM) 110 CsCl, 5 MgATP, 1 MgCl2, 20 tetraethylammonium, 5 Na2 phosphocreatine, and 10 HEPES, pH 7.1 (adjusted
with CsOH). The standard external solution contained (in mM) 130
NaCl, 5 CsCl, 1 MgCl2, 2.0 CaCl2, 5 Na-pyruvate, 10 glucose, and 10
C1907
HEPES, pH 7.4 (adjusted with NaOH). All drugs were purchased
from Sigma (St. Louis, MO). Because nifedipine (Nif) is very sensitive to light, particular precautions were taken. A fresh working
solution of 15 ␮M Nif was made by diluting a fresh 15 mM stock
solution (dissolved in DMSO), resulting in a final DMSO concentration of 0.1%. The entire Nif delivery pathway including the micromanifold was light tight.
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Fig. 1. General protocol and representative traces from a 3-day-old (3d) myocyte. A: general experimental protocol. The sarcoplasmic reticulum (SR) Ca2⫹ was
cleared by the 1st application of caffeine (Caf) and then followed by a train of 20 repetitive depolarizations at 0.2 Hz. The 2nd Caf (dark gray bar) was applied
to evaluate the SR Ca2⫹ content. B: representative traces of Ca2⫹ transients (intracellular Ca2⫹ concentration, [Ca2⫹]i), the depolarization-induced Ca2⫹ transient
at left in contrast to the Caf-induced Ca2⫹ transient (Caf [Ca2⫹]i) at right with the same scale. C: representative traces of L-type Ca2⫹ current (ICa). D: tail
Na⫹/Ca2⫹ exchanger (NCX) current (INCX; tail INCX at left and Caf INCX at right with a different scale). E: time integral of the corresponding INCX: 兰tail INCX
at left and 兰Caf INCX at right with a different scale. F and G: net Ca2⫹ entry and cumulative net Ca2⫹ entry, respectively, as a function of depolarization. The
depolarization dependence was observed in Ca2⫹ transient, tail INCX, and net Ca2⫹ entry; steady state (SS) was achieved at ⬃10th depolarization.
AJP-Cell Physiol • VOL
293 • DECEMBER 2007 •
www.ajpcell.org
C1908
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
Age-dependent differences in sarcolemmal Ca2⫹ entry,
Ca2⫹ transients, and SR Ca2⫹ loading. Figure 2A shows that
the cumulative net Ca2⫹ entry significantly decreased with age
while the number of depolarizations required to achieve steady
state was comparable in all age groups. Figure 2B shows the
maximum amplitude of [Ca2⫹]i as a function of number of
depolarizations. The data were well fit with a Boltzmann
function (R2 0.98). The peak [Ca2⫹]i increased with subsequent
depolarizations and reached steady state near the 10th depo-
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Measurement of Ca2⫹ fluorescence. Intracellular Ca2⫹ concentration ([Ca2⫹]i) was measured with the fluorescent Ca2⫹ indicator
fluo-3 AM as described previously (12, 13, 17). The [Ca2⫹]i was
calculated from the formula [Ca2⫹]i ⫽ Kd 䡠 (F ⫺ Fmin)/(Fmax ⫺ F),
where Fmin is the background fluorescence determined from a cellfree area and Fmax is the fluorescence acquired after the cell was
depolarized to ⫹150 mV for 10 –20 s to maximize [Ca2⫹]i and kill the
cell at the end of each experiment. F0 was taken as the difference
between the background fluorescence determined in the absence and
presence of a cell in the area of measurement. ⌬F is the incremental
fluorescence measured from baseline or the background fluorescence
in the presence of a cell. Kd is the fluo-3 Ca2⫹ dissociation constant,
and a value of 400 nM was used for all age groups (20).
General protocol. Figure 1A shows the SR Ca2⫹ loading experimental protocol applied in Fig. 1, B and C, as well as Figs. 3 and 4.
The SR Ca2⫹ was first cleared by a brief Caf application. A train of
20 repetitive depolarizations at 0.2 Hz (first depolarized to ⫺40 mV
for 50 ms in order to inactivate Na⫹ channels and T-type Ca2⫹
channels, and then depolarized to ⫹10 mV for 400 ms, at a holding
potential of ⫺80 mV) was then initiated 5 s after Caf removal.
Immediately after the 20th depolarization, Caf was applied again to
evaluate the SR Ca2⫹ content.
Figure 6A shows the experimental protocol used to measure the SR
Ca2⫹ uptake rate (VSR) and [Ca2⫹]i. The first Caf application was
used to clear the SR Ca2⫹ content, and the cell was then depolarized
to various voltages (from ⫺30 mV to ⫹50 mV, in increments of 20
mV) for 3 s to load the SR. Long depolarizations were used to induce
a relative spatial and temporal homogeneity of [Ca2⫹]i. Pulses longer
than 3 s were also tried, but these frequently resulted in cell death,
particularly in the younger age groups. More positive depolarizations
were also tried, but these often resulted in saturation of loadSR and as
a consequence frequently generated spontaneous release. Therefore,
these data were not included in the analysis. The time integral of the
INCX (兰INCX) induced by the second Caf application was used as a
measure of loadSR during the preceding 3-s depolarization (12). The
average VSR (amol 䡠 s⫺1 䡠 pF⫺1) was obtained by dividing loadSR
(兰INCX, amol/pF) by the duration of the loading period (3 s).
Data analysis. Data are presented as means ⫾ SE. Statistical
significance of the results was tested with a one-way ANOVA (SPSS
11.0) or Student’s t-test for paired or unpaired samples. Post hoc tests
were taken with Tukey multiple comparisons. A P value of ⱕ0.05 was
taken to be significant.
RESULTS
Depolarization dependence of Ca2⫹ transient, tail INCX, as
well as sarcolemmal Ca2⫹ entry in a 3d myocyte. Figure 1B
shows representative Ca2⫹ transient recordings ([Ca2⫹]i, including the second Caf-induced Ca2⫹ transient, Caf [Ca2⫹]i),
Fig. 1C shows the corresponding ICa, and Fig. 1D shows the
inward INCX on repolarization (tail INCX) and during the second
Caf application (Caf INCX) in a 3d myocyte. The corresponding
time integrals of tail INCX (兰tail INCX) and Caf INCX (兰Caf
INCX) are shown in Fig. 1E. Peak ICa in the 3d group did not
show a significant decrease in magnitude with depolarization
number, although this phenomenon was observed in older age
groups (data not shown). Both the Ca2⫹ transient and tail INCX
showed a depolarization-dependent increase that reached a
steady state near the 10th depolarization. Assuming that sarcolemmal Ca2⫹ entry and extrusion are equal in magnitude at
steady state, subtraction of the average value of the last 5 兰tail
INCX from each individual 兰tail INCX should give the net Ca2⫹
entry during each depolarization. Figure 1F shows the net
Ca2⫹ entry for each depolarization, and the cumulative net
Ca2⫹ entry during the 20 depolarizations is shown in Fig. 1G.
AJP-Cell Physiol • VOL
Fig. 2. Age-dependent differences in sarcolemmal Ca2⫹ entry, Ca2⫹, transients and SR Ca2⫹ loading (loadSR). A and B: cumulative net Ca2⫹ entry
normalized by cell membrane capacitance (pC/pF) and amplitude of [Ca2⫹]i as
a function of depolarization in 3d, 10d, 20d, and 56d myocytes. Cumulative net
Ca2⫹ entry significantly decreased with age, although the amplitude of [Ca2⫹]i
did not show significant age dependence except that the 1st [Ca2⫹]i was
significantly greater in 3d compared with 56d myocytes (*P ⬍ 0.05). Numbers
of depolarizations required to reach half-maximal load (N0.5) and steady-state
levels (NSS) were not significantly different with age. C: cumulative net Ca2⫹
entry and amount of loadSR in control solution (loadSR-Con) as a function of
age. Significant decrease with age was observed both in cumulative net Ca2⫹
entry (**P ⬍ 0.01 for 3d vs. 20d, ***P ⬍ 0.001 for 3d vs. 56d, and *P ⬍ 0.05
for 10d vs. 56d) and loadSR-Con (†††P ⬍ 0.001 for 3d vs. either 20d or 56d
and ††P ⬍ 0.01 for 10d vs. either 20d or 56d). However, the fraction of
cumulative net Ca2⫹ entry to SR Ca2⫹ loading did not reach a level of
significance between all age groups. n ⫽ 15.
293 • DECEMBER 2007 •
www.ajpcell.org
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
C1909
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Fig. 3. Age-dependent differences in nifedipine (Nif)-sensitive Ca2⫹ transients and sarcolemmal Ca2⫹ entry. A and B: representative traces from a 3d myocyte
(A) and a 56d myocyte (B) in the presence of Nif; the [Ca2⫹]i (depolarization-induced [Ca2⫹]i at left and Caf [Ca2⫹]i at right with the same scale), inward INCX
(tail INCX at left and Caf INCX at right with different scales), and time integral of corresponding INCX (兰tail INCX at left and 兰Caf INCX at right with different scales)
are shown from top to bottom, respectively. The depolarization-dependent Ca2⫹ transient and the tail INCX were abolished by the addition of Nif in 56d but not
in 3d myocytes. C and D: cumulative net Ca2⫹ entry normalized by cell membrane capacitance (pC/pF; C) and amplitude of [Ca2⫹]i [incremental fluorescence
(⌬F)/difference in background fluorescence between absence and presence of cell (F0); D] as a function of depolarization in the presence of Nif in 3d, 10d, 20d,
and 56d myocytes. Nif significantly abolished sarcolemmal Ca2⫹ entry as well as the Ca2⫹ transient in older age groups but not in younger age groups. n ⫽ 12.
AJP-Cell Physiol • VOL
293 • DECEMBER 2007 •
www.ajpcell.org
C1910
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
larization. The response was comparable for all age groups,
which was neither proportional to cumulative Ca2⫹ entry nor to
loadSR. Figure 2C compares the cumulative net Ca2⫹ entry and
the loadSR (loadSR-Con) as a function of age. Both decreased
significantly with age, but loadSR was larger than cumulative
Ca2⫹ entry in all age groups, suggesting that an additional
source of loadSR exists.
Age-dependent difference in nifedipine-insensitive Ca2⫹
transients and sarcolemmal Ca2⫹ entry. Figure 3, A and B, are
representative traces from a 3d myocyte and a 56d myocyte,
respectively, using the same protocol as shown in Fig. 1A but
in the presence of 15 ␮M Nif, a selective inhibitor of L-type
Ca2⫹ channels that completely blocked ICa in all age groups
(data not shown). [Ca2⫹]i, inward INCX (tail INCX and Caf
INCX), and their time integrals (兰INCX and 兰Caf INCX) are
shown from top to bottom, respectively, in Fig. 3, A and B. The
depolarization-dependent Ca2⫹ transient and the tail INCX were
abolished by the addition of Nif in 56d but not in 3d myocytes.
Both Caf [Ca2⫹]i and Caf INCX were still observed in 56d
myocytes, although tail INCX and Ca2⫹ transients were abolished. Figure 3, C and D, show the cumulative net Ca2⫹ entry
and the [Ca2⫹]i amplitude in the presence of Nif for the
different age groups. Note that Nif abolished sarcolemmal
Ca2⫹ entry and Ca2⫹ transients in older but not in younger age
groups.
Age-dependent difference in nifedipine and KB-R7943-insensitive SR Ca2⫹ loading. Figure 4, A and B, show representative traces from a 3d myocyte and a 56d myocyte in the
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Fig. 4. Age-dependent differences in loadSR in the presence of Nif and KB-R7943 (KB-R). A and B: representative traces from a 3d myocyte (A) and a 56d
myocyte (B) in the presence of Nif⫹KB-R. [Ca2⫹]i (depolarization induced at left and Caf [Ca2⫹]i at right with the same scale), INCX (tail INCX at left and Caf
INCX at right with different scales), and time integral of INCX (兰INCX at left and 兰Caf INCX at right with different scales) are shown from top to bottom,
respectively. The remaining depolarization-dependent Ca2⫹ transient and the tail INCX observed in 3d myocytes in the presence of Nif were abolished by the
addition of KB-R. C: loadSR in the presence of both Nif and KB-R (NIF⫹KB-R insensitive) as a function of age, which significantly decreased with age (***P ⬍
0.005 for both 3d and 10d vs. 56d and **P ⬍ 0.01 for 3d vs. 20d). n ⫽ 12.
AJP-Cell Physiol • VOL
293 • DECEMBER 2007 •
www.ajpcell.org
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
different postnatal stages by measuring the SR Ca2⫹ uptake
rate as a function of bulk phase [Ca2⫹]i. Figure 6B shows
representative traces of the Ca2⫹ transient and membrane
current elicited by the protocol shown in Fig. 6A (on depolarization to ⫹30 mV) in a 3d cell. The magnitude of the Ca2⫹
transient after it reached a plateau state ([Ca2⫹]PS) was used as
the [Ca2⫹]i corresponding to its average VSR. Figure 6C shows
the average Ca2⫹ transient traces and its corresponding second
Caf-induced INCX from five each of 3d and 56d cells at
depolarization steps from ⫺30 mV to ⫹50 mV. It is clear that
the second Caf-induced INCX is greater in 3d than 56d cells
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
presence of both 15 ␮M Nif and 10 ␮M KB-R7943 (KB-R).
[Ca2⫹]i, inward INCX (tail INCX and Caf INCX), and their time
integrals (兰INCX and 兰Caf INCX) are shown from top to bottom,
respectively, in Fig. 4, A and B. The depolarization-dependent
Ca2⫹ transient and tail INCX observed in 3d myocytes with Nif
were abolished by the subsequent addition of KB-R. However,
a significant Caf-induced Ca2⫹ transient and a considerable
amount of Caf INCX were still observed in both 3d and 56d
myocytes, although the values were significantly greater for 3d
than for 56d myocytes, confirming that an additional loadSR
source exists and is insensitive to Nif and KB-R. Figure 4C
shows that Nif- and KB-R-insensitive loadSR was significantly
smaller in older compared with younger age groups.
Age-dependent changes in relative contributions of different
sarcolemmal Ca2⫹ sources. Figure 5A shows the cumulative
Ca2⫹ entry in control and Nif conditions for the different age
groups (there is no cumulative Ca2⫹ entry in the presence of
NIF⫹KB-R). The relative contributions of Nif-sensitive and
KB-R-sensitive (but Nif insensitive) Ca2⫹ entry are shown in
Fig. 5B. KB-R-sensitive but Nif-insensitive Ca2⫹ entry was
dominant at the earliest developmental stage and significantly
decreased with age; in contrast, Nif-sensitive Ca2⫹ entry significantly increased with age and became dominant at the latest
developmental stage examined. Figure 5C shows the corresponding control, Nif, and Nif⫹KB-R or Nif⫹KB-R-insensitive loadSR as shown in Fig. 4C. The relative contributions
(Con ⫽ 1) of Nif-sensitive, KB-R-sensitive, and Nif⫹KB-Rinsensitive loadSR are shown in Fig. 5D for the different age
groups. The calculation of the contribution of Nif⫹KB-Rinsensitive loadSR is predicated on two observations: 1) steady
state in control solution was achieved at ⬃10th depolarization
(⬃50 s) and 2) Nif⫹KB-R-insensitive loadSR is linear with
time for up to 100 s (13). Therefore the contribution of
Nif⫹KB-R-insensitive loadSR relative to Con is derived from
the assumption that its loading is the half-value of Nif⫹KBR-insensitive loadSR (⬃100 s). In accordance with Fig. 5B,
Nif-sensitive loadSR significantly increased with age, while
KB-R-sensitive loadSR and Nif⫹K-BR-insensitive loadSR significantly decreased with age.
Age-dependent changes in k0.5 of SR Ca2⫹ uptake. Since
age-dependent changes in loadSR may be due to changes in the
SERCA2a Ca2⫹ affinity, we estimated this parameter at the
Fig. 5. Age-dependent changes in the relative contributions of different
sarcolemmal Ca2⫹ sources. A: amount of cumulative net Ca2⫹ entry as a
function of age in Con and Nif conditions. A significant increase of cumulative
net Ca2⫹ entry with age was present in both Con (significance between groups
shown in Fig. 2C) and Nif (†††P ⬍ 0.001 for either 3d or 10d vs. either 20d
or 56d). B: relative Nif-sensitive and KB-R sensitive contributions to cumulative net Ca2⫹ entry (Con ⫽ 1) as a function of age. Nif-sensitive Ca2⫹
contribution increased with age (†††P ⬍ 0.001 for either 3d or 10d vs. either
20d or 56d); in contrast, KB-R-sensitive Ca2⫹ contribution decreased with age
(***P ⬍ 0.001 for either 3d or 10d vs. either 20d or 56d). C: loadSR in Con,
Nif, and Nif⫹KB-R-insensitive as a function of age. loadSR significantly
decreased with age in the presence of Con (significance between age groups
shown in Fig. 2C), Nif (**P ⬍ 0.01 for both 3d and 10d vs. 20d and 56d), and
Nif⫹KB-R insensitive (significance between age groups shown in Fig. 4C).
D: relative Nif-sensitive, KB-R-sensitive, and Nif⫹KB-R-insensitive contributions to loadSR (Con ⫽ 1) as a function of age. Nif-sensitive loadSR
significantly increased with age (***P ⬍ 0.001 for either 3d or 10d vs. either
20d or 56d); KB-R-sensitive and NIF⫹KB-R-insensitive loadSR significantly
decreased with age (†††P ⬍ 0.001 for either 3d or 10d vs. either 20d or 56d
in KB-R sensitive; ‡P ⬍ 0.05 for 3d vs. either 20d or 56d in Nif⫹KB-R
insensitive). n ⫽ 12.
AJP-Cell Physiol • VOL
C1911
293 • DECEMBER 2007 •
www.ajpcell.org
C1912
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
despite the fact that the amplitudes of the Ca2⫹ transients were
comparable in the two groups. The representative average VSR
as a function of [Ca2⫹]PS derived from Fig. 6C (including data
point of VSR at ⫺80 mV) is shown in Fig. 6D. Data points were
fit with the Hill equation: VSR ⫽ Vmax 䡠[Ca2⫹]PSnH/(KnH ⫹
[Ca2⫹]nH) (R2 0.98) to obtain the asymptotic value of VSR
(Vpeak), k0.5 (the value of [Ca2⫹]PS at half of Vpeak) and Hill
coefficient nH (the slope of the sigmoid curve). From Fig. 6D,
it is clear that k0.5 is smaller in 3d than 56d cells; k0.5 increased
significantly after 10 days of age. In contrast, there were no
significant differences in either Vpeak or nH between groups
(Table 1).
AJP-Cell Physiol • VOL
DISCUSSION
It has been suggested that neonatal mammalian ventricular
myocytes have a relatively sparse and immature SR compared
with that in adults (11, 25), but recent studies indicate that a
prominent SR Ca2⫹ accumulation can take place in neonatal
myocytes (12, 13, 21). Here we have attempted to identify,
quantify, and compare the mechanisms contributing to loadSR
in neonatal and adult hearts. In accordance with our previous
report (12), loadSR after SR Ca2⫹ depletion was significantly
greater in the neonatal compared with the adult heart when
normalized per unit membrane capacitance. Moreover, our
293 • DECEMBER 2007 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
Fig. 6. Apparent age-dependent changes in sarco(endo)plasmic reticulum (SERCA)2a Ca2⫹ affinity. A: experimental protocol used in the measurement of SR
Ca2⫹ uptake rate (VSR) and [Ca2⫹]i. B: representative traces of the resultant Ca2⫹ transient (black) and membrane current (gray) at a depolarization of ⫹30 mV
in a 3d myocyte. C: top: average Ca2⫹ transient trace from 5 cells each for 3d and 56d myocytes at depolarization steps of ⫺30, ⫺10, ⫹10, ⫹30, and ⫹50 mV.
Bottom: 2nd Caf-induced INCX traces corresponding to the Ca2⫹ transients at top. D: corresponding fitting traces of the average VSR as a function of Ca2⫹ transient
after it reached plateau stage ([Ca2⫹]PS) derived from C (including data point of VSR at ⫺80 mV) in 56d and 3d myocytes.
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
Table 1. Parameters measured in SR Ca2⫹ uptake
Age, days
3
10
20
56
No. of myocytes
25
25
25
25
0.25⫾0.01* 0.26⫾0.04 0.32⫾0.02 0.38⫾0.04
k0.5, ␮M
Vpeak, amol 䡠 s⫺1 䡠 pF⫺1 12.44⫾0.82 10.49⫾1.22 11.62⫾0.79 12.79⫾1.03
nH
1.90⫾0.13
1.75⫾0.12 2.31⫾0.16
2.2⫾0.23
Values are means ⫾ SE. k0.5, Ca2⫹ concentration at half-maximal sarcoplasmic reticulum (SR) Ca2⫹ uptake; Vpeak, asymptotic value of SR uptake
rate; nH, slope of the sigmoid curve. *Significantly different from 20-day and
56-day groups (P ⬍ 0.05). Differences in Vpeak and nH within age groups did
not reach a level of significance.
AJP-Cell Physiol • VOL
study from our laboratory (13) showing that this phenomenon
can likely be ascribed to SOCE in rabbit cardiac myocytes. In
agreement with this study, we observed that the contribution of
SOCE to total SR Ca2 refilling significantly decreased with
age. The average rate of sarcolemmal Ca2⫹ influx induced by
depolarization was 1.1 and 0.5 pC䡠pF⫺1 䡠s⫺1 for 3d and 56d
myocytes, respectively, which is 15- to 50-fold greater than our
previous estimate of the average Ca2⫹ influx rate via SOCE
during the first 10 s after SR Ca2⫹ depletion (13). Thus,
although SOCE may contribute to SR refilling after Ca2⫹
depletion, this mechanism is unlikely to make a significant
contribution to E-C coupling on a beat-to-beat basis even in the
neonate myocyte. The remarkable loadSR via SOCE (40%)
after SR Ca2⫹ depletion does, however, suggest that it might
contribute to regulate loadSR at the earliest developmental
stages. Our previous results indicated that SOCE loadSR was
intimately modulated by NCX (13), which we suggested resulted from the SOCE influx pathway and NCX being in the
same microdomain. More recent evidence from two different
laboratories has strongly supported the concept that TRP-C3
channels and NCX colocalize in heart tissue (7, 8). Although
not the focus of this study, it is tempting to speculate that
TRP-C3 channels may contribute to the observed SOCE.
The cellular mechanisms underlying loadSR provide important insight into the intact working heart. With an increase in
the number of depolarizations, [Ca2⫹]i and loadSR dramatically
increase, providing the bases for the phenomenon of the
positive force-frequency relationship observed in rabbit ventricular muscle strip and intact heart (19, 31) preparations.
Although the intact heart might be less dependent on Ca2⫹
influx via depolarization compared with isolated muscle (31),
the developmental changes in the sources of Ca2⫹ for SR
refilling in an intact heart are likely to be similar to those
observed for the single myocyte, and these observations are
clinically significant. For example, a prolonged whole heart
arrest during open-heart surgery in the newborn will likely be
more prone to cause SR Ca2⫹ overload and arrhythmogenesis
as a result of a greater SOCE; therefore, a different approach
for arresting the newborn heart during cardiopulmonary bypass
may be beneficial.
SR Ca2⫹ pump activity during development. Our findings
that 10 consecutive depolarizations were sufficient to reload
the SR independently of the age stage although loadSR (normalized by membrane capacitance) was more than threefold
greater in 3d than in 56d myocytes (Fig. 2C) suggest that at the
early developmental stages there is an increase in 1) SERCA2a
Vmax, 2) affinity of SERCA2a for calcium, and/or 3) [Ca2⫹]
in the microdomain in which SERCA2a resides. Our estimates of Vpeak (12.8 amol Ca2⫹ 䡠 pF⫺1 䡠 s⫺1 or 82 ␮M 䡠 liter
cytosol⫺1 䡠 s⫺1) and k0.5 (0.38 ␮M) in the 56d group agree well
with values reported by Bassani et al. (2) in intact myocytes
from adult rabbits. The nH value observed from our data is
close to the theoretical maximum of 2, but smaller than that
observed by other groups (2) using intact cardiomyocytes,
which might be explained by differences in the measurement
and analysis techniques. The data shown in Table 1 indicate
that Vpeak, as measured in our study, is not different between
the age groups. Studies of SR vesicles have shown that the
maximum SR Ca2⫹ uptake rates increase during development
in a variety of species (23, 24) and are consistent with previous
observations of sparse SR and a lower density of SERCA2a in
293 • DECEMBER 2007 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
results show that NCX plays a prominent role in loadSR in the
3d heart, whereas ICa is the main sarcolemmal Ca2⫹ source for
loadSR in late neonatal and adult rabbit ventricular myocytes.
Age-dependent changes in relative contributions of ICa and
reverse-mode NCX. The present study shows that 10 consecutive depolarizations are sufficient to fully reload the SR after
transient exposure to 10 mM Caf. This was true for all rabbit
ventricular myocytes examined independent of the age stage.
In the older age groups, the calcium reloading was primarily
due to an enhanced ICa and a decreased tail INCX during the first
10 depolarizations. This is in accordance with previous reports
on adult rat and ferret hearts (28) and is also consistent with a
previous study from our laboratory (14). It is generally agreed
that the ICa is the main pathway for sarcolemmal Ca2⫹ entry in
the adult rabbit heart (8, 17), and in agreement with this notion,
we find that loadSR as well as the Ca2⫹ transient were sensitive
to nifedipine in older age groups (Fig. 3, C and D, and Fig. 5).
In contrast, in the early developmental stages, loadSR as well as
the Ca2⫹ transient were insensitive to Nif but were sensitive to
subsequent addition of KB-R (Fig. 4A and Fig. 5). The findings
indicate that reverse-mode NCX plays an important role in
both excitation-contraction (E-C) coupling and SR Ca2⫹ refilling at early neonatal stages. Moreover, total sarcolemmal Ca2⫹
efflux estimated via tail INCX at steady state in 3d hearts was
about twice the magnitude of that in 56d hearts (0.4 vs. 0.2
pC/pF), and the first [Ca2⫹]i transient in younger age groups
was larger than that in 56d hearts and sensitive to KB-R (Fig.
2B and Fig. 3D), confirming previous findings of a more
prominent role of reverse-mode NCX in neonatal E-C coupling
(14, 30).
It should be noted that the slow Ca2⫹ removal system has
been suggested to play a greater role in neonatal compared with
adult heart (3, 16, 29). In rat ventricular myocytes, the slow
Ca2⫹ removal system has been reported to be approximately
two times greater in immature than in adult cardiac myocytes
(3). Therefore, it is possible that the contributions of sarcolemmal Ca2⫹ efflux and SR Ca2⫹ content determined by the
integral of INCX are underestimated in younger age groups.
However, the contribution of the slow Ca2⫹ removal system in
the early developmental stage is still too small (5% of total
Ca2⫹) to be considered physiologically significant (3).
Age-dependent changes in Nif⫹KB-R-insensitive SR Ca2⫹
loading. Interestingly, a considerable amount of loadSR was
observed at all age stages despite the fact that the Ca2⫹
transient was abolished by the addition of Nif and KB-R (Fig.
4). This is in accordance with observations of Trafford et al.
(28) in quiescent adult ferret and rat myocytes and a previous
C1913
C1914
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
AJP-Cell Physiol • VOL
we recently showed that a narrow cleft (20 nm) between the
sarcolemma and SR observed in both 3d and 56d myocytes was
delimited by a threefold longer SR in 3d than 56d myocytes
(300 vs. 100 nm), resulting in a much more restricted microdomain in the early developmental stages. It is therefore possible
that an apparent greater Ca2⫹ affinity of the SR Ca2⫹ pump
observed in neonate myocytes may at least partly result from
measurement of VSR as a function of the average bulk [Ca2⫹]i.
Conclusions. In conclusion, there is a switch in the sarcolemmal calcium fluxes contributing to SR Ca2⫹ refilling
from a predominance of NCX and SOCE at the earliest stage
to a predominance of ICa at later stages. Moreover, the number
of depolarizations required to achieve steady state did not vary
during development, although steady-state loadSR was threefold larger in the neonatal heart. This may be explained by
either a higher Ca2⫹ affinity of the SERCA2a in neonatal
myocytes or a higher local [Ca2⫹] around SERCA2a for a
given level of [Ca2⫹] in the bulk phase. Together, this suggests
that age-dependent downregulation of SR calcium sequestration
and increased dependence on calcium entry through L-type calcium channels during the first 20 days postpartum in rabbit
ventricular myocytes is due to a selective downregulation of
NCX-dependent SR Ca2⫹ refilling.
GRANTS
The authors acknowledge the generous grant support of the Canadian
Institutes of Health Research and the Heart and Stroke Foundation of British
Columbia and Yukon (G. F. Tibbits). J. Huang is the recipient of a Canada
Research Scholarship from the Canadian Institutes of Health Research. L.
Hove-Madsen holds a “Ramon y Cajal” grant from the Spanish Ministry of
Science and Technology, and G. F. Tibbits is the recipient of a Tier I Canada
Research Chair.
REFERENCES
1. Balaguru D, Haddock PS, Puglisi JL, Bers DM, Coetzee WA, Artman
M. Role of the sarcoplasmic reticulum in contraction and relaxation of
immature rabbit ventricular myocytes. J Mol Cell Cardiol 29: 2747–2757,
1997.
2. Bassani JW, Bassani RA, Bers DM. Relaxation in rabbit and rat cardiac
cells: species-dependent differences in cellular mechanisms. J Physiol
476: 279 –293, 1994.
3. Bassani RA, Bassani JW. Contribution of Ca2⫹ transporters to relaxation
in intact ventricular myocytes from developing rats. Am J Physiol Heart
Circ Physiol 282: H2406 –H2413, 2002.
4. Bers DM. Excitation-Contraction Coupling and Cardiac Contractile
Force (2nd ed.). Dordrecht, The Netherlands: Kluwer Academic, 2001.
5. Bhogal MS, Colyer J. Depletion of Ca2⫹ from the sarcoplasmic reticulum
of cardiac muscle prompts phosphorylation of phospholamban to stimulate
store refilling. Proc Natl Acad Sci USA 95: 1484 –1489, 1998.
6. Chu G, Ferguson DG, Edes I, Kiss E, Sato Y, Kranias EG. Phospholamban ablation and compensatory responses in the mammalian heart. Ann
NY Acad Sci 853: 49 – 62, 1998.
7. Eder P, Probst D, Rosker C, Poteser M, Wolinski H, Kohlwein SD,
Romanin C, Groschner K. Phospholipase C-dependent control of cardiac
calcium homeostasis involves a TRPC3-NCX1 signaling complex. Cardiovasc Res 73: 111–119, 2007.
8. Goel M, Zuo CD, Sinkins WG, Schilling WP. TRPC3 channels colocalize with Na⫹/Ca2⫹ exchanger and Na⫹ pump in axial component of
transverse-axial tubular system of rat ventricle. Am J Physiol Heart Circ
Physiol 292: H874 –H883, 2007.
9. Harrer JM, Haghighi K, Kim HW, Ferguson DG, Kranias EG.
Coordinate regulation of SR Ca2⫹-ATPase and phospholamban expression in developing murine heart. Am J Physiol Heart Circ Physiol 272:
H57–H66, 1997.
10. Hicks MJ, Shigekawa M, Katz AM. Mechanism by which cyclic
adenosine 3⬘:5⬘-monophosphate-dependent protein kinase stimulates calcium transport in cardiac sarcoplasmic reticulum. Circ Res 44: 384 –391,
1979.
293 • DECEMBER 2007 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
neonates (25, 27). The discrepancies between the results of the
present study and those presented above are probably related to
differences in technique. The advantages of in vitro preparations such as microsomal vesicles include control over “cytosolic” [Ca2⫹], measurement of initial rates for a more accurate
determination of Vmax and affinity, as well as regulation of
intravesicular [Ca2⫹] with oxalate or similar compounds. However, the major disadvantage is that the SERCA2a has been
removed from its native environment with the attendant potential loss of regulatory cofactors [e.g., Gi proteins, ␤-receptors,
SR-associated phosphatases, and phospholamban (PLB)] and
destruction of the microdomain in which SERCA2a resides.
Despite these differences, however, it is useful to compare our
data in the 56d intact myocyte with data from SR vesicles from
the adult rabbit heart. In one such a study by Xu and Narayanan
(33) in which uptake was determined at 37°C, they found a
Vmax of 489 nmol Ca2⫹ 䡠mg protein⫺1 䡠min⫺1, which is comparable to that determined by others in different species (10,
15). In addition, they found a k0.5 value of 0.6 ␮M and an nH
of 1.4, which compare favorably to the k0.5 of 0.4 ␮M and nH
of 2.2 determined in the present study. To convert the microsomal Vmax (in nmol Ca2⫹ 䡠mg protein⫺1 䡠min⫺1) to units
comparable to the Vpeak (in amol Ca2⫹ 䡠pF⫺1 䡠s⫺1) observed in
the present study, we used the following assumptions: 10-fold
microsomal purification factor, 120 mg homogenate protein/g
wet wt, 2.43 g wet wt/ml cytosol, 4.58 pF/pl cytosol (4, 26),
and a Q10 of 2 to derive a value of 12.1 amol Ca2⫹ 䡠pF⫺1 䡠s⫺1
in vesicles vs. 12.8 amol Ca2⫹ 䡠pF⫺1 䡠s⫺1 determined in intact
myocytes in the present study. Because we cannot unequivocally state that our VSR reflects initial rates because of the
complexity of the preparation, we chose to refer to the maximal
VSR in our study as Vpeak instead of Vmax. However, the fact
that the Vpeak values are comparable to the Vmax determined
under the simpler and more controlled in vitro conditions
serves to allay many of these concerns.
Consistent with the notion that affinity of SERCA2a for
Ca2⫹ was altered, the k0.5 for SR loading significantly increased with age (Table 1). However, studies using electrical
field stimulation of immature ventricular myocytes have suggested a diminished contribution of the SR Ca2⫹ pump to
cytosolic Ca2⫹ removal (1, 3, 34), which would appear to
contradict our findings. On the other hand, another study using
electrical field stimulation recently demonstrated that a functional SR is present long before birth in a linear heart tube (22),
and the different results are likely due to the specific experimental conditions and/or species differences.
Indeed, one possible explanation for a higher SR Ca2⫹
affinity without a change of Vmax in neonate rabbit ventricular
myocytes may be a lower PLB expression relative to that of
SERCA2a (9, 18). In agreement with this notion, it has been
demonstrated that PLB-deficient myocytes exhibit a higher SR
Ca2⫹ load (6), and it was recently demonstrated that the degree
of PLB phosphorylation per SERCA2a was greater in the fetus
and newborn compared with adult (32). It has been reported
that the SR Ca2⫹ depletion prompts the phosphorylation of
PLB to stimulate store refilling (5). An alternative explanation
to the apparently higher Ca2⫹ affinity of the SR Ca2⫹ pump in
neonate myocytes is that there are age-dependent differences in
the subsarcolemmal microdomain (12, 13) resulting in a higher
[Ca2⫹] in the immediate vicinity of the SERCA2a in the
neonate heart for a given level of bulk phase [Ca2⫹]. Indeed,
SR Ca2⫹ UPTAKE RATES IN NEONATE CARDIOMYOCYTES
AJP-Cell Physiol • VOL
23. Nakanishi T, Jarmakani JM. Developmental changes in myocardial
mechanical function and subcellular organelles. Am J Physiol Heart Circ
Physiol 246: H615–H625, 1984.
24. Nayler WG, Fassold E. Calcium accumulating and ATPase activity of
cardiac sarcoplasmic reticulum before and after birth. Cardiovasc Res 11:
231–237, 1977.
25. Page E, Buecker JL. Development of dyadic junctional complexes
between sarcoplasmic reticulum and plasmalemma in rabbit left ventricular myocardial cells. Morphometric analysis. Circ Res 48: 519 –522,
1981.
26. Satoh H, Delbridge LM, Blatter LA, Bers DM. Surface:volume relationship in cardiac myocytes studied with confocal microscopy and membrane capacitance measurements: species-dependence and developmental
effects. Biophys J 70: 1494 –1504, 1996.
27. Shibata Y, Nakata K, Page E. Ultrastructural changes during development of gap junctions in rabbit left ventricular myocardial cells. J Ultrastruct Res 71: 258 –271, 1980.
28. Trafford AW, Diaz ME, Negretti N, Eisner DA. Enhanced Ca2⫹ current
and decreased Ca2⫹ efflux restore sarcoplasmic reticulum Ca2⫹ content
after depletion. Circ Res 81: 477– 484, 1997.
29. Vornanen M. Activation of contractility and sarcolemmal Ca2⫹-ATPase
by Ca2⫹ during postnatal development of the rat heart. Comp Biochem
Physiol A 78: 691– 695, 1984.
30. Wetzel GT, Chen F, Klitzner TS. Na⫹/Ca2⫹ exchange and cell contraction in isolated neonatal and adult rabbit cardiac myocytes. Am J Physiol
Heart Circ Physiol 268: H1723–H1733, 1995.
31. Wohlfart B, Elzinga G. Electrical and mechanical responses of the intact
rabbit heart in relation to the excitation interval. A comparison with the
isolated papillary muscle preparation. Acta Physiol Scand 115: 331–340,
1982.
32. Xu A, Hawkins C, Narayanan N. Ontogeny of sarcoplasmic reticulum
protein phosphorylation by Ca2⫹-calmodulin-dependent protein kinase. J
Mol Cell Cardiol 29: 405– 418, 1997.
33. Xu A, Narayanan N. Reversible inhibition of the calcium-pumping
ATPase in native cardiac sarcoplasmic reticulum by a calmodulin-binding
peptide. Evidence for calmodulin-dependent regulation of the Vmax of
calcium transport. J Biol Chem 275: 4407– 4416, 2000.
34. Yao A, Matsui H, Spitzer KW, Bridge JH, Barry WH. Sarcoplasmic
reticulum and Na⫹/Ca2⫹ exchanger function during early and late relaxation in ventricular myocytes. Am J Physiol Heart Circ Physiol 273:
H2765–H2773, 1997.
293 • DECEMBER 2007 •
www.ajpcell.org
Downloaded from http://ajpcell.physiology.org/ by 10.220.32.246 on July 31, 2017
11. Hoerter J, Mazet F, Vassort G. Perinatal growth of the rabbit cardiac
cell: possible implications for the mechanism of relaxation. J Mol Cell
Cardiol 13: 725–740, 1981.
12. Huang J, Hove-Madsen L, Tibbits GF. Na⫹/Ca2⫹ exchange activity in
neonatal rabbit ventricular myocytes. Am J Physiol Cell Physiol 288:
C195–C203, 2005.
13. Huang J, van Breemen C, Kuo KH, Hove-Madsen L, Tibbits GF.
Store-operated Ca2⫹ entry modulates sarcoplasmic reticulum Ca2⫹ loading in neonatal rabbit cardiac ventricular myocytes. Am J Physiol Cell
Physiol 290: C1572–C1582, 2006.
14. Huang J, Xu L, Thomas M, Whitaker K, Hove-Madsen L, Tibbits
GF. L-type Ca2⫹ channel function and expression in neonatal rabbit
ventricular myocytes. Am J Physiol Heart Circ Physiol 290: H2267–
H2276, 2006.
15. Inui M, Chamberlain BK, Saito A, Fleischer S. The nature of the
modulation of Ca2⫹ transport as studied by reconstitution of cardiac
sarcoplasmic reticulum. J Biol Chem 261: 1794 –1800, 1986.
16. Karttunen P, Vornanen M. Sarcolemmal ATPase activities of the rat
heart ventricle— dependence on age and sodium ion. Comp Biochem
Physiol B 86: 815– 820, 1987.
17. Katsube Y, Yokoshiki H, Nguyen L, Yamamoto M, Sperelakis N.
L-type Ca2⫹ currents in ventricular myocytes from neonatal and adult rats.
Can J Physiol Pharmacol 76: 873– 881, 1998.
18. Mahony L, Jones LR. Developmental changes in cardiac sarcoplasmic
reticulum in sheep. J Biol Chem 261: 15257–15265, 1986.
19. Maier LS, Bers DM, Pieske B. Differences in Ca2⫹-handling and
sarcoplasmic reticulum Ca2⫹-content in isolated rat and rabbit myocardium. J Mol Cell Cardiol 32: 2249 –2258, 2000.
20. Merritt JE, McCarthy SA, Davies MP, Moores KE. Use of fluo-3 to
measure cytosolic Ca2⫹ in platelets and neutrophils. Loading cells with the
dye, calibration of traces, measurements in the presence of plasma, and
buffering of cytosolic Ca2⫹. Biochem J 269: 513–519, 1990.
21. Miller MS, Friedman WF, Wetzel GT. Caffeine-induced contractions in
developing rabbit heart. Pediatr Res 42: 287–292, 1997.
22. Moorman AF, Schumacher CA, de Boer PA, Hagoort J, Bezstarosti
K, van den Hoff MJ, Wagenaar GT, Lamers JM, Wuytack F, Christoffels VM, Fiolet JW. Presence of functional sarcoplasmic reticulum in
the developing heart and its confinement to chamber myocardium. Dev
Biol 223: 279 –290, 2000.
C1915