Functional Roles of Cav1.3 (α1D) Calcium Channel in Sinoatrial Nodes
Insight Gained Using Gene-Targeted Null Mutant Mice
We directly examined the role of the Cav1.3 (α1D) Ca2+ channel in the sinoatrial (SA) node by using Cav1.3 Ca2+ channel-deficient mice. A previous report has shown that the null mutant (Cav1.3−/−) mice have sinus bradycardia with a prolonged PR interval. In the present study, we show that spontaneous action potentials recorded from the SA nodes show a significant decrease in the beating frequency and rate of diastolic depolarization in Cav1.3−/− mice compared with their heterozygous (Cav1.3+/−) or wild-type (WT, Cav1.3+/+) littermates, suggesting that the deficit is intrinsic to the SA node. Whole-cell L-type Ca2+ currents (ICa,Ls) recorded in single isolated SA node cells from Cav1.3−/− mice show a significant depolarization shift in the activation threshold. The voltage-dependent activation of Cav1.2 (α1C) versus Cav1.3 Ca2+ channel subunits was directly compared by using a heterologous expression system without β coexpression. Similar to the ICa,L recorded in the SA node of Cav1.3−/− mutant mice, the Cav1.2 Ca2+ channel shows a depolarization shift in the voltage-dependent activation compared with that in the Cav1.3 Ca2+ channel. In summary, using gene-targeted deletion of the Cav1.3 Ca2+ channel, we were able to establish a role for Cav1.3 Ca2+ channels in the generation of the spontaneous action potential in SA node cells.
Spontaneous activity in sinoatrial (SA) node cells results from a characteristic phase of their action potentials (APs), ie, slow diastolic depolarization.1,2⇓ During this phase, the membrane slowly depolarizes until the threshold for a new AP is reached. The slow diastolic depolarization results from a number of ionic currents, including the hyperpolarization-activated inward current (If), an inward T-type Ca2+ current (ICa,T), an inward L-type Ca2+ current (ICa,L), and possibly a net inward background conductance.3–5⇓⇓ It is postulated that the initial phase of the pacemaker depolarization is carried by If. At a membrane potential positive to −60 mV, ICa,T is activated; ICa,L, with its more positive activation threshold close to −30 mV, would be responsible for the upstroke of the AP.2–5⇓⇓⇓
Recently, the molecular basis of the Ca2+ current (ICa) in SA nodes was investigated. By use of an in situ hybridization technique, it was found that the most prominently expressed low-voltage activated Ca2+ channel in the SA node was Cav3.1 (α1G), whereas Cav3.2 (α1H) is present at moderate levels. The dominant high-voltage activated Ca2+ channel was Cav1.2 (α1C), but a small amount of Cav1.3 (α1D) mRNA was detected in the SA node myocytes of mice.6 Indeed, a recent report of Cav1.3 Ca2+ channel-deficient mice, similar to ours,7 showed evidence of SA nodal dysfunction in the homozygous mutant mice.8 Important questions remain regarding the exact cellular mechanisms for the observed phenotype in the mutant animals. We theorize that a detailed examination of the biophysical properties of the SA node and the two L-type Ca2+ channel isoforms using mutant mice with a robust phenotype of SA nodal dysfunction should yield important information regarding the distinct functional roles of the Cav1.2 versus Cav1.3 Ca2+ channel subunits in pacemaking cells.
Materials and Methods
The present investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1985) and was performed in accordance with the guidelines of the Animal Care and Use Committee of the University of California, Davis.
Generation of Cav1.3 null mutant (Cav1.3−/−) mice has previously been described.7 Single SA node cells were isolated from wild-type (WT) and mutant mice.9,10⇓ Whole-cell ICa was recorded at room temperature by using patch-clamp techniques.11,12⇓ The external solution contained (mmol/L) N-methyl-d-glucamine 140, CsCl 5, MgCl2 0.5, CaCl2 2, 4-aminopyridine 2, glucose 10, and HEPES 10, pH 7.4 with HCl, and the internal solution contained (mmol/L) N-methyl-d-glucamine 135, tetraethylammonium chloride 20, Mg-ATP 4, EGTA 1, and HEPES 10, pH 7.3 with HCl. All chemicals were purchased from Sigma Chemical Co unless stated otherwise. The cell capacitance was calculated by integrating the area under an uncompensated capacitive transient elicited by a 20-mV hyperpolarizing pulse from a holding potential of −40 mV. Whole-cell current records were filtered at 2 kHz and sampled at 10 kHz. Spontaneous APs were recorded from isolated SA nodal preparations by using microelectrode techniques with 3 mol/L KCl microelectrodes at 34°C.13 The external solution contained (mmol/L) NaCl 138, KCl 4, MgCl2 1, CaCl2 2, NaH2PO4 0.33, glucose 10, and HEPES 10, pH 7.4 with NaOH. Cav1.214 or Cav1.315 Ca2+ channel subunits were transiently expressed in human embryonic kidney (HEK293) cells.12 Green fluorescence protein (Invitrogen) was used as a transfection reporter. For whole-cell ICa recordings from HEK293 cells, the external solution contained (mmol/L) CaCl2 40, CsCl 80, glucose 10, and HEPES 10 (pH 7.4 with CsOH), and the pipette solution contained (mmol/L) CsCl 120, EGTA 10, Mg-ATP 3, and HEPES 10 (pH 7.4 with CsOH). Liquid junction potentials were measured as previously described,16 and all data were corrected for the liquid junction potentials.
Curve fits and data analysis were performed by using Origin software (MicroCal Inc). Where appropriate, pooled data are presented as mean±SEM. Statistical comparison was performed by using the Student t test, with a value of P<0.05 considered significant. The rate of diastolic depolarization (DDR, in mV/s) was determined from recordings of the spontaneous AP by determining the first derivative of the diastolic depolarization at three different voltages (−50, −45, and −40 mV).
Cav1.3−/− Mice Show Evidence of SA and AV Nodal Dysfunction
First, we were able to document a robust phenotype of SA and atrioventricular (AV) nodal dysfunction in the Cav1.3−/− mice. ECG recordings from anesthetized Cav1.3−/− mice show significant sinus bradycardia with significant prolongation of the RR intervals compared with their heterozygous (Cav1.3+/−) and WT (Cav1.3+/+) littermates (Figure 1A). Summary data (n=3 from each group) are shown in Figure 1B (left panel). Significant prolongation of the RR and PR intervals was observed in the Cav1.3−/− mice (P<0.05). These differences in the PR and RR intervals could represent intrinsic abnormalities in the pacemaking activities in SA and AV node cells in mutant mice. However, alternative possibilities include altered autonomic input into the SA and AV nodes resulting from the deletion of the Cav1.3 Ca2+ channel. We recorded ECG in the WT and mutant mice by using combined intraperitoneal injection of atropine (1 mg/kg) and propranolol (20 mg/kg), concentrations that are known to abolish autonomic control of the heart.17,18⇓ Results are presented in Figure 1B (right panel). Administration of atropine and propranolol resulted in prolongation of the RR intervals in the Cav1.3−/− as well as Cav1.3+/+ littermates; however, the effects were much more pronounced in the Cav1.3−/− mice. Importantly, the Cav1.3−/− mice showed significant abnormalities in the SA and AV nodes after treatment with atropine and propranolol, suggesting that the defects are intrinsic to the pacemaking tissues.
SA Nodes Isolated From Cav1.3−/− Mice Show Decreases in Rate of Firing and DDR
Spontaneous APs were recorded from intact SA nodes by using microelectrode techniques. Figure 1C shows the characteristic spontaneous APs recorded from the regions within and around the SA nodes at 34°C. APs recorded from within the SA node can be identified by the presence of slow diastolic depolarization and a very slow upstroke of phase 0. Cav1.3−/− mice show a significant slowing of the spontaneous activities of the SA nodes compared with the Cav1.3+/+ or Cav1.3+/− mice (Figure 2A). Analysis of the AP (Figures 2B and Figure 3⇓) shows a significant decrease in DDR at −40 and −45 mV in Cav1.3−/− mice compared with Cav1.3+/+ or Cav1.3+/− mice, suggesting a critical role of the α1D Ca2+ channel in the spontaneous diastolic depolarization of the SA node cells. In addition, Cav1.3−/− mice showed evidence of second-degree AV block (intermittent nonconducted P wave, Figure 2C, left panel). Recordings from SA nodal preparations also showed intermittent loss of firing in the SA node (Figure 2C, right panel).
Figure 3 shows summary data of cycle length (CL) in milliseconds, the maximum diastolic potential (MDP), AP amplitude (APA), maximum upstroke velocity (Vmax), DDR, and AP duration (APD) at 50% and 80% repolarization (APD50 and APD80, respectively). There was a significant decrease in the rate of firing in the SA nodal preparations isolated from Cav1.3−/− mice that was associated with a significant decrease in DDR at −45 and −40 mV compared with Cav1.3+/+ or Cav1.3+/− mice (P<0.05, Figure 3). In contrast, DDRs at −50 mV were not significantly different between the three groups of animals. The APD50 was significantly shorter in the mutant animals (most likely secondary to the faster decay of ICa,L in the mutant mice, see results in next section; Figures 4D and 4E). However, APD80 was slightly more prolonged in the mutant mice (see Discussion).
There were no significance differences in the Vmax between the three groups of animals. Our data suggest that once the threshold for activation of the AP is reached, the Cav1.3 Ca2+ channel plays minimum roles in the rapid upstroke of the AP. In addition, the MDP and APA did not differ between the mutant mice and WT controls. Of note, the frequency of firing of isolated SA nodal preparations was nearly 2-fold different between the Cav1.3+/+ and Cav1.3−/− groups. The differences in the ECG recordings were less marked (Figure 1B, left panel), which was likely due to the presence of autonomic input or compensatory changes in the whole animals. Indeed, after treatment with atropine and propranolol, the null mutant mice, compared with the WT littermates, showed an exaggerated increase in the RR interval, and the differences between the mutant and WT mice became more marked, consistent with the data on CL obtained with the use of in vitro preparations of isolated SA nodes.
Whole-Cell ICa, L Recorded From Isolated SA Node Cells
Previous data provide important clues that the differences in the biophysical properties of Cav1.2 versus Cav1.3 Ca2+ channels may be directly responsible for the observed phenotypes.19,20⇓ To directly confirm this possibility, we isolated single SA node cells by using the various landmarks noted in Figure 1C. Cells were identified after occurrence of spontaneous beating upon rewarming of the external solution in the bath to 34°C (Figure 4A). Whole-cell ICa,L was recorded from single isolated SA node cells by using a holding potential of −55 mV to inactivate ICa,T, which is known to be present in SA node cells. ICa,L activated at more depolarizing potentials in Cav1.3−/− animals compared with Cav1.3+/+ or Cav1.3+/− animals, which expressed both Cav1.2 and Cav1.3 Ca2+ channels (Figure 4B). We further confirmed this initial impression by generating activation curves from WT and mutant animals (Figure 4C). ICa, L from Cav1.3−/− mice, compared with Cav1.3+/+ mice, showed an ≈5-mV depolarizing shift in the midpoint of activation. We next examined the rate of inactivation of ICa.L recorded from Cav1.3+/+ compared with Cav1.3−/− animals (Figure 4D). The inactivation profile could be best fit by using two exponential time constants. As illustrated in Figure 4E, both the fast and slow time constants at a test potential of 0 mV were significantly abbreviated in ICa,Ls recorded from the Cav1.3−/− mutant mice (P<0.05).
Because the relative contribution of Cav1.3 Ca2+ channel to total ICa,L in the SA node is not known, we directly compared the maximum ICa,L conductances (gmax) between mutant mice and their WT littermates (Figure 4F). The conductance was normalized to the cell capacity. Cell capacitance of single isolated SA node cells from the three groups of animals was 53.9±2.2, 41.9±2.4, and 52.3±5.0 pF for Cav1.3+/+, Cav1.3+/−, and Cav1.3−/−, respectively (n=7, P=NS). There was an ≈5-mV depolarization shift in the peak ICa,L in the Cav1.3−/− mice; however, gmax was not significantly different between Cav1.3+/+ and the Cav1.3−/− animals. This may represent compensatory changes with upregulation of the Cav1.2 ICa,L in the mutant animals.
The increase in the inactivation rate of ICa,L in the Cav1.3−/− mice may be secondary to an increase in the voltage-dependent inactivation or Ca2+-dependent inactivation.21 The dominant Ca2+ sensor for Ca2+-dependent inactivation of the Cav1.2 Ca2+ channel has recently been identified as calmodulin, which appears to be constitutively tethered to the channel complex.22–25⇓⇓⇓ We further examined the contribution of Ca2+-dependent inactivation to the inactivation profile. Figure 4G shows data obtained by using a 2-pulse protocol to examine the voltage- and Ca2+-dependent inactivation of ICa,L in WT compared with mutant animals. The two curves were nearly superimposed, with no significant differences in the half-inactivation voltages. In addition, the curves show the typical U-shaped configuration for Ca2+-dependent inactivation of ICa,L. Prepulses more positive to +20 mV elicited a progressively smaller inward current as the command voltages approached the reversal potential, leading to partial recovery of ICa,L elicited with the test pulse due to a decrease in the Ca2+-dependent inactivation. No significant differences were observed between the WT and the mutant animals in the Ca2+-dependent inactivation.
Heterologous Expression of Cav1.2 Versus Cav1.3 Ca2+ Channels
To further crosscheck that the observed shift in the activation curve is secondary to the ablation of the Cav1.3 subunit in the mutant animals, we expressed Cav1.2 or Cav1.3 subunits heterologously in HEK293 cells. Previous data compared the properties of the two isoforms with β-subunit coexpression.19 However, to directly examine the intrinsic properties of each subunit, auxiliary subunits were not used in our experiments (Figure 5). Figure 5B shows a significant depolarization shift in the steady-state activation in Cav1.2 compared with Cav1.3 Ca2+ currents that is consistent with findings in the Cav1.3−/− mice, which express only the Cav1.2 subunit. In contrast, the increase in the rate of inactivation of ICa,L in the Cav1.3 mutant mice (Figures 4C and 4D) cannot be recapitulated by the expression of Cav1.3 alone in the expression system (Figures 5C and 5D).
In the present study, we directly tested the role of the Cav1.3 Ca2+ channel in SA nodal function by using Cav1.3−/− mice with phenotypes of marked sinus bradycardia. Recordings from isolated SA nodal preparations confirmed the presence of a significant decrease in the firing frequency of the SA node, localizing the defect to within the SA nodes. Close examination of the spontaneous AP identified an isolated depression in the slope of the spontaneous diastolic depolarization principally at −40 and −45 mV. Whole-cell ICa,L recordings showed a depolarization shift in the voltage-dependent activation in the null mutant mice compared with the WT mice. Heterologous expression of Cav1.2 versus Cav1.3 Ca2+ channel subunits alone in HEK293 cells resulted in ICa, L, with a similar shift in the voltage-dependent activation. Together, our data suggest that the Cav1.3 Ca2+ channel is one of the key determinants of the slope of spontaneous diastolic depolarization in SA node cells because of the hyperpolarization shift of its activation curve compared with that of the Cav1.2 Ca2+ channel.
Role of Cav1.3 Ca2+ Channel in SA Nodes
Our data are consistent with the previous notion that L-type Ca2+ channels contribute mainly to the terminal phase of the spontaneous diastolic depolarization.26,27⇓ On the basis of biophysical and electrophysiological data, it was previously suggested that the initial phase of the spontaneous diastolic depolarization results from the activation of If. At a membrane potential positive to −60 mV, ICa,T was activated, whereas ICa,L would be responsible for the later phase of the diastolic depolarization.2–5,26,27⇓⇓⇓⇓⇓ In the present study, using gene-targeted deletion of the Cav1.3 isoform, we were able to document that the single most prominent abnormality observed in the SA node is a decrease in the rate of firing associated with a diminished rate of diastolic depolarization.
However, further inspection of the results in Figure 2B suggests that differences in the DDR are not solely responsible for the differences in the spontaneous beating rate. The rate of terminal repolarization was also delayed in the Cav1.3−/− mice, with prolongation of APD80. There are several possibilities for this prolongation. ICa,L showed a more rapid inactivation profile in the null mutant mice compared with the WT mice. This could account for the shorter APD50 in the null mutant mice. In addition, it is conceivable that the rate of deactivation of the ICas is different in knockout mice.
Compensatory Changes in the Mutant Mice
Much information has been gained in recent years using different transgenic mouse models. One common finding among different models has been compensatory changes in the animals in response to the targeted genetic manipulation.28 These compensatory changes most likely exert beneficial effects for the animal’s survival but may introduce complications in understanding the phenotypes of the animals. Nonetheless, our data showed that even though Cav1.2 ICa,L may be upregulated in the homozygous mutant animals, a depolarization shift remained in the mutant animals that was associated with in vitro and in vivo evidence of SA nodal dysfunction.
Heterologous Expression of Cav1.2 Versus Cav1.3 α Subunits of Ca2+ Channels
The coexpression of the Cav1.3 and β subunits of Ca2+ channels has previously been reported.19,20⇓ However, no data investigating the biophysical properties of Cav1.2 versus Cav1.3 subunits alone without β coexpression have previously been published. Indeed, the present study was the first to directly document the differences in the biophysical properties of the α1 subunits alone. Although the shift in the midpoint of the activation curves occurred to similar extents between Cav1.3 null mutant mice versus the WT mice and between Cav1.2 versus Cav1.3 Ca2+ channels in the expression systems, the absolute midpoints are notably different. This may be secondary to several possibilities. First is the lack of auxiliary subunits. The published data in Koschak et al19 showed half-activation voltage (V1/2) of −17.5±0.9 versus −3.9±1.3 mV for Cav1.3 versus Cav1.2 coexpressed with α2δ, β2a, and β3. These values are closer to our data of V1/2 recorded from Cav1.3+/+ mice (expressing both Cav1.2 and Cav1.3 α1 subunits) versus Cav1.3−/− mice (expressing only Cav1.3 α1 subunits) of −16.6 and −11.4 mV, respectively. In addition, the differences may also stem from differences in the recording conditions. When expressed alone, α1 subunits result in very small currents. The high external Ca2+ concentration required for the recordings would be predicted to lead to a depolarization shift in the activation curves as a result of the surface charge screening.21,29⇓ Nonetheless, the data obtained from the expression system agree well with our notion that the lack of a Cav1.3 Ca2+ channel in the null mutant mice would result in a depolarization shift in the voltage-dependent activation of ICa,L in SA node cells.
In contrast, the increase in the rate of inactivation of ICa,L in the Cav1.3 mutant mice cannot be recapitulated by the expression of Cav1.3 alone in the expression system. It appears that the coexpression with other auxiliary subunits, eg, β or α2δ subunits, is required to observe the differences in the inactivation kinetics.19 In addition, splice variants of the α1 subunits as well as channel regulation may contribute to differences in current kinetics.
Limitations of the Study
Close inspection of the records (Figures 2B and 3⇑A) shows that the decrease in DDR was observed at voltages between −40 and −45 mV (with some variation between cells). On the other hand, the activation threshold for ICa.L in wild-type animals occurs at voltages of ≈−40 mV. The reasons that the activation threshold of ICa, L may be slightly shifted to the depolarization direction compared with the diastolic depolarization potentials include differences in the experimental solutions and cell dialysis with whole-cell current recordings. In addition, ICa,L was recorded at room temperature, whereas the APs were recorded at 34°C. Indeed, a small hyperpolarization shift of ≈5 mV in the peak current on a change from 20°C to 30°C has previously been reported in ICas of bullfrog sympathetic neurons.30 In addition, at higher temperature, there is an increase in the current amplitude as well as in the rate of activation and inactivation of ICa.31 Nonetheless, our data clearly indicate much larger effects of Cav1.3 ablation on the slope at −40 mV compared with more hyperpolarized potentials, consistent with the expected role of Cav1.3 currents.
In summary, using gene-targeted deletion of Cav1.3 Ca2+ channels, we established the role of Cav1.3 Ca2+ channels in the generation of the spontaneous AP in SA node cells. The hyperpolarization shift in the activation threshold of Cav1.3 Ca2+ channels can be directly documented in vivo as well as in the expression system of the Cav1.3 subunit alone. Our data support distinct roles of Cav1.2 versus Cav1.3 Ca2+ channels in the generation of the pacemaking activities of the SA nodes; the Cav1.2 subunit cannot substitute for the Cav1.3 subunit. The phenotype of SA nodal dysfunction was observed in the null mutant mice even though no significant changes in ICa, L density could be documented. Similar to the expression of the neuronal systems, the expression of multiple Ca2+ channel subtypes is important in the coordination of different physiological functions in pacemaking cells.
This study was supported by the National Heart, Lung, and Blood Institute of the NIH (grants RO1 HL-67737 and HL-68507), by an American Heart Association Scientist Development Grant, and by a VA Merit Review Grant (Dr Chiamvimonvat). The authors are indebted to Dr E.N. Yamoah for helpful suggestions and comments. The α1D clone was a gift from Dr S. Seino (Chiba University, Chiba, Japan).
Original received January 9, 2002; resubmission received February 12, 2002; revised resubmission received April 1, 2002; accepted April 1, 2002.
- ↵Irisawa H, Brown HF, Giles W. Cardiac pacemaking in the sinoatrial node. Physiol Rev. 1993; 73: 197–227.
- ↵Boyett MR, Honjo H, Kodama I. The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res. 2000; 47: 658–687.
- ↵Mangoni ME, Nargeot J. Properties of the hyperpolarization-activated current (If) in isolated mouse sino-atrial cells. Cardiovasc Res. 2001; 52: 51–64.
- ↵Vinogradova TM, Zhou YY, Bogdanov KY, Yang D, Kuschel M, Cheng H, Xiao RP. Sinoatrial node pacemaker activity requires Ca2+/calmodulin-dependent protein kinase II activation. Circ Res. 2000; 87: 760–767.
- ↵Chiamvimonvat N, O’Rourke B, Kamp TJ, Kallen RG, Hofmann F, Flockerzi V, Marban E. Functional consequences of sulfhydryl modification in the pore-forming subunits of cardiovascular Ca2+ and Na+ channels. Circ Res. 1995; 76: 325–334.
- ↵Verheijck EE, van Kempen MJ, Veereschild M, Lurvink J, Jongsma HJ, Bouman LN. Electrophysiological features of the mouse sinoatrial node in relation to connexin distribution. Cardiovasc Res. 2001; 52: 40–50.
- ↵Seino S, Chen L, Seino M, Blondel O, Takeda J, Johnson JH, Bell GI. Cloning of the α1 subunit of a voltage-dependent calcium channel expressed in pancreatic β cells. Proc Natl Acad Sci U S A. 1992; 89: 584–588.
- ↵Jumrussirikul P, Dinerman J, Dawson TM, Dawson VL, Ekelund U, Georgakopoulos D, Schramm LP, Calkins H, Snyder SH, Hare JM, Berger RD. Interaction between neuronal nitric oxide synthase and inhibitory G protein activity in heart rate regulation in conscious mice. J Clin Invest. 1998; 102: 1279–1285.
- ↵Koschak A, Reimer D, Huber I, Grabner M, Glossmann H, Engel J, Striessnig J. α1D (Cav1.3) subunits can form l-type Ca2+ channels activating at negative voltages. J Biol Chem. 2001; 276: 22100–22106.
- ↵Bell DC, Butcher AJ, Berrow NS, Page KM, Brust PF, Nesterova A, Stauderman KA, Seabrook GR, Nurnberg B, Dolphin AC. Biophysical properties, pharmacology, and modulation of human, neuronal L-type (α1D), CaV1:3) voltage-dependent calcium currents. J Neurophysiol. 2001; 85: 816–827.
- ↵Hille B. Ion Channels of Excitable Membranes. 3rd ed. Sunderland, Mass: Sinauer Associates Inc; 2001.
- ↵Qin N, Olcese R, Bransby M, Lin T, Birnbaumer L. Ca2+-induced inhibition of the cardiac Ca2+ channel depends on calmodulin. Proc Natl Acad Sci U S A. 1999; 96: 2435–2438.
- ↵Zuhlke RD, Pitt GS, Tsien RW, Reuter H. Ca2+-sensitive inactivation and facilitation of L-type Ca2+ channels both depend on specific amino acid residues in a consensus calmodulin-binding motif in the α1C subunit. J Biol Chem. 2000; 275: 21121–21129.
- ↵Shibata EF, Giles WR. Ionic currents that generate the spontaneous diastolic depolarization in individual cardiac pacemaker cells. Proc Natl Acad Sci U S A. 1985; 82: 7796–7800.
- ↵Ji Y, Lalli MJ, Babu GJ, Xu Y, Kirkpatrick DL, Liu LH, Chiamvimonvat N, Walsh RA, Shull GE, Periasamy M. Disruption of a single copy of the SERCA2 gene results in altered Ca2+ homeostasis and cardiomyocyte function. J Biol Chem. 2000; 275: 38073–38080.
- ↵Zhou W, Jones SW. Surface charge and calcium channel saturation in bullfrog sympathetic neurons. J Gen Physiol. 1995; 105: 441–462.