Phosphorylation sites required for regulation of cardiac calcium channels in the fight-or-flight response

Significance

This work defines the in vivo role of phosphorylation of Ser1700 and Thr1704 in Ca_V1.2 channels in the fight-or-flight response. Mutation of both residues to Ala in STAA mice reduced basal L-type Ca²⁺ currents and markedly decreased β-adrenergic stimulation of Ca²⁺ currents and cell contraction. STAA ventricular myocytes exhibited arrhythmic contractions and often failed to sustain contractions when stimulated at 1 Hz. STAA mice have reduced exercise capacity, and cardiac hypertrophy is evident at 3 mo. Our results define the molecular mechanism for Ca²⁺ channel regulation in the fight-or-flight response and give insight into mechanisms that underlie cardiac homeostasis and hypertrophy.

Abstract

L-type Ca²⁺ currents conducted by Ca_V1.2 channels initiate excitation–contraction coupling in the heart. Their activity is increased by β-adrenergic/cAMP signaling via phosphorylation by PKA in the fight-or-flight response, but the sites of regulation are unknown. We describe the functional role of phosphorylation of Ser1700 and Thr1704—sites of phosphorylation by PKA and casein kinase II at the interface between the proximal and distal C-terminal regulatory domains. Mutation of both residues to Ala in STAA mice reduced basal L-type Ca²⁺ currents, due to a small decrease in expression and a substantial decrease in functional activity. The increase in L-type Ca²⁺ current caused by isoproterenol was markedly reduced at physiological levels of stimulation (3–10 nM). Maximal increases in calcium current at nearly saturating concentrations of isoproterenol (100 nM) were also significantly reduced, but the mutation effects were smaller, suggesting that alternative regulatory mechanisms are engaged at maximal levels of stimulation. The β-adrenergic increase in cell contraction was also diminished. STAA ventricular myocytes exhibited arrhythmic contractions in response to isoproterenol, and up to 20% of STAA cells failed to sustain contractions when stimulated at 1 Hz. STAA mice have reduced exercise capacity, and cardiac hypertrophy is evident at 3 mo. We conclude that phosphorylation of Ser1700 and Thr1704 is essential for regulation of basal activity of Ca_V1.2 channels and for up-regulation by β-adrenergic signaling at physiological levels of stimulation. Disruption of phosphorylation at those sites leads to impaired cardiac function in vivo, as indicated by reduced exercise capacity and cardiac hypertrophy.

In the heart, action potentials initiate excitation–contraction coupling by activation of voltage-gated Ca²⁺ channel Ca_V1.2. Ca²⁺ entering through these channels activates Ca²⁺-dependent Ca²⁺ release from the sarcoplasmic reticulum by activation of the ryanodine-sensitive Ca²⁺ release channels (1). The force of contraction is critically dependent on the amplitude, kinetics, and voltage dependence of the L-type Ca²⁺ current conducted by Ca_V1.2 channels (2). Under conditions of fear, stress, and exercise, the sympathetic nervous system activates the fight-or-flight response, in which the marked increase in contractile force of the heart is caused by epinephrine and norepinephrine acting through β-adrenergic receptors, activation of adenylate cyclase, increase in cAMP, activation of cAMP-dependent protein kinase (PKA), and phosphorylation of the Ca_V1.2 channel (3, 4). This pathway has been extensively studied because of the functional significance of β-adrenergic regulation in the normal heart and in cardiac hypertrophy and failure. However, the molecular mechanism remains unresolved.

Ca_V1.2 channels are composed of pore-forming α₁1.2 subunits (also designated α_1C) in association with β, α₂δ, and possibly γ subunits (5). Biochemical studies of the closely related Ca_V1.1 channel in skeletal muscle showed that it is proteolytically processed near the center of its C-terminal domain (6) and identified the precise point of truncation (7). The large C-terminal domain of Ca_V1.2 channels is also proteolytically processed in the corresponding position (8), and the distal C-terminal (dCT) associates noncovalently with the proximal C-terminal (pCT) and serves as a potent autoinhibitor (7, 9). Regulation of Ca_V1.2 channels by PKA has been reconstituted in nonmuscle cells with a dynamic range of 3.6-fold, similar to cardiomyocytes (10). Successful reconstitution required an A-kinase anchoring protein (AKAP), which recruits PKA to the dCT (10–13). Deletion of the dCT in vivo in mice results in loss of regulation of the L-type Ca²⁺ current by the β-adrenergic pathway and embryonic death from heart failure (14, 15). These results suggest that the signaling complex consisting of the truncated Ca_V1.2 channel with noncovalently bound dCT is the functional substrate for physiological regulation in the fight-or-flight response.

Extensive studies have demonstrated that the Ca_V1.2 channel is the primary target for PKA phosphorylation upon β-adrenergic stimulation of cardiac myocytes (16–20), and multiple PKA sites have been identified in both α₁ (8, 21) and β (22–24) subunits by in vitro phosphorylation. However, none of these sites has been shown to be required for regulation of Ca_V1.2 channels in vivo. For example, Ser1928 has been well characterized as a PKA phosphorylation site both in vitro and in vivo (8, 13, 21), but its phosphorylation is not required for PKA-dependent up-regulation of channel activity in transfected nonmuscle cells, virally transduced ventricular myocytes, or acutely dissociated ventricular myocytes from genetically modified mice (10, 25, 26). Exhaustive analysis of endogenous phosphorylation of Ca_V1.1 channels by mass spectrometry identified two previously undetected phosphorylation sites: Ser1575 and Thr1579 (27). Ser1575 is located in an exact consensus sequence for PKA phosphorylation and is phosphorylated in skeletal muscle in vivo in response to β-adrenergic stimulation (27). These two phosphorylation sites are conserved in Ca_V1.2 channels as Ser1700 and Thr1704, and they reside at the interface between the pCT and dCT in an ideal position to regulate channel function [Fig. 1A (10)]. The Ca_V1.2 signaling complex reconstituted in nonmuscle cells required phosphorylation of both of these sites for normal basal activity, whereas only Ser1700 phosphorylation was essential for PKA stimulation (10).

Fig. 1.

Alanine substitution at Ser1700 and Thr1704 in Ca_V1.2 channels. (A) Docking model of the proximal and distal C terminus regulatory domain. Amino acids surrounding Ser1700 and Thr1704 (shown in green ribbon) constitute the interface between pCT and dCT. (Adapted with permission from ref. 10.) (B) Mutations indicated by an asterisk (*) were introduced via homologous recombination into genomic DNA to generate a knockin mouse line expressing Ala in place of Ser1700 and Thr1704 (STAA) in the α₁ subunit of Ca_V1.2. The targeting construct includes genomic sequence from exon 39 to exon 43. The point mutation is in exon 41, immediately followed by the Neo cassette. Red, long arm; blue, short arm. (C) Representative confocal images from stacked Z-series of myocytes labeled with anti-Ca_V1.2 antibody (anti-CNC1). The average intensity of anti-Ca_V1.2 per myocyte was determined using the region of interest function in Igor (bar graph) and the mean transverse line scan [50 pixels wide (8 µm); lines, C, Left] from top surface (dashed line, Right) and middle plane (solid line, Right) of the myocytes from WT mice (black; n = 19 myocytes) and STAA mice (red; n = 12 myocytes). *P < 0.05. (D and E) Representative confocal images and mean intensities (bar graphs) from stacked Z-series of myocytes labeled with anti-dCT (anti-CH2) (D) and anti–phospho-Ser1928 (Anti-CH3P) (E) antibodies in absence and presence of 100 nM Iso. **P < 0.01, ***P < 0.001. (Scale bar: 25 µm; 40× magnification for all of the confocal images.)

To determine whether phosphorylation of Ser1700 and/or Thr1704 mediates Ca_V1.2 channel regulation by the β-adrenergic pathway in vivo, we generated a mutant mouse line in which Ala is substituted for both Ser1700 and Thr1704 (STAA mice). We report here that phosphorylation at Ser1700 and Thr1704 is essential for maintaining basal Ca_V1.2 channel activity and for normal β-adrenergic regulation. Disruption of this regulation results in impaired contractile function, decreased exercise capacity, and cardiac hypertrophy in STAA mice.

Results

Ser1700A and Thr1704A Mutations in Ca_V1.2 Channels in Mice.

STAA mice were produced with a standard gene-targeting procedure that introduced nucleotides encoding the mutations Ser1700Ala and Thr1704Ala in the gene encoding Ca_V1.2 channels in all cells (Fig. 1B). These mice are viable and fertile and do not show overt physiological deficits. In isolated ventricular myocytes from STAA mice, the total level of Ca_V1.2 channels is reduced by ∼12% compared with wild type (WT) (Fig. 1C), as shown by immunofluorescent staining with the anti-CNC1 antibody recognizing the intracellular linker connecting domains II and III of the α₁1.2 subunit (28, 29). In averaged line scans of Ca_V1.2 staining at the top surface of the cells, a similar small reduction in staining intensity is observed in STAA myocytes (Fig. 1C, dotted lines). These results indicate that there is no preferential loss of Ca_V1.2 channels from the plasma membrane at the cell surface. Averaged line scans across the center of the myocytes showed that the distribution of Ca_V1.2 channels between the cell surface and internal compartments such as the t-tubule/sarcoplasmic reticulum junction, endoplasmic reticulum, and Golgi apparatus is similar in WT and STAA myocytes (Fig. 1C). Overall, these immunocytochemical results indicate that STAA myocytes have a slightly reduced total number of Ca_V1.2 channels (∼88% of WT), but the subcellular distribution of Ca_V1.2 channels is unchanged.

In contrast to our results with the ant-CNC1 antibody against the linker connecting domains II and III, specific staining with the anti-CH2 antibody recognizing the dCT is increased in STAA myocytes relative to WT (Fig. 1D). These results show that the level of the proteolytically processed dCT is increased in STAA myocytes, whereas the level of the transmembrane core of the Ca_V1.2 channel is decreased. The increased ratio of the proteolytically processed dCT compared with the transmembrane core of the Ca_V1.2 channel implied by these results could reduce the level of basal activity of Ca_V1.2 channels through the autoinhibitory effect of the dCT (9, 10).

To assess the overall integrity of the β-adrenergic/PKA signaling pathway in STAA myocytes, we examined phosphorylation at Ser1928, a well-characterized PKA site located in the dCT. Basal immunostaining with the phosphospecific antibody anti-CH3P, which recognizes phosphorylated Ser1928, was lower in STAA myocytes than WT, but was significantly increased in both WT and STAA upon treatment with 100 nM isoproterenol (Iso) (Fig. 1E). The increase in phospho-Ser1928 after treatment with Iso is comparable between STAA and WT myocytes, suggesting that PKA phosphorylation of the Ca_V1.2 channel at Ser1928 remains intact.

Reduced Basal Ba²⁺ Currents and Reduced PKA Regulation of Ca_V1.2 Channels in Neonatal Cardiomyocytes.

We first examined regulation of Ca_V1.2 channel function in neonatal cardiomyocytes. Although β-adrenergic regulation is not fully developed at this age, it is important to examine the Ca_V1.2 channels in these developing myocytes before adverse effects of impaired cardiovascular performance caused by cardiac hypertrophy in older STAA mice (see below) could alter the regulatory process. In ventricular myocytes dissociated from hearts of mice on postnatal days 1–3, we found that basal L-type current measured with Ba²⁺ as charge carrier was decreased compared with WT (Fig. 2; 252 ± 26 vs. 479 ± 62 pA). Iso increased peak L-type Ba²⁺ currents 1.44 ± 0.1-fold for WT, whereas there was no detectable effect of Iso on L-type Ba²⁺ currents in STAA myocytes. These results indicate a fundamental impairment of basal and PKA regulation of Ca_V1.2 channels in neonatal ventricular myocytes before any compensatory effects induced by cardiac hypertrophy (see below) have occurred.

Fig. 2.

Reduced basal and Iso-induced Ba²⁺ current in STAA neonatal cardiomyocytes. (Left) Representative current traces during steps from −40 to +10 mV in absence (−) and presence (+) of Iso. (Right) Current–voltage relationship of I_Ba in unstimulated cardiomyocytes and after stimulation with 1 µM Iso. Peak I_Ba and half activation (V_a) values were obtained from fitting I–V curves. V_a values were not significantly affected by Iso treatment for either genotype (WT: V_a = −2.33 ± 1.2 mV before Iso vs. V_a = −5.49 ± 1.5 mV after Iso; STAA: V_a = −2.0 ± 0.7 mV before Iso vs. V_a = −3.5 ± 1.5 mV after Iso). One-way ANOVA with a Bonferroni posttest was used to determine statistical significance.

Reduced Basal L-Type Ca²⁺ Currents and Decreased Response to β-Adrenergic Stimulation in Adult STAA Myocytes.

To determine whether phosphorylation at Ser1700 and Thr1704 is functionally important in adult cardiomyocytes, we measured whole-cell Ca²⁺ currents at basal level (0 min) and during 5-min perfusion with 10 nM Iso. Representative current traces for depolarization to +10 mV from a holding potential of −40 mV at t = 0 reveal substantially lower basal L-type Ca²⁺ currents in STAA myocytes (Fig. 3A, Inset). Ca²⁺ currents increased over time after treatment with 10 nM Iso and reached a plateau after 4 min for WT myocytes (Fig. 3A, black). Although much smaller in magnitude, L-type Ca²⁺ currents in STAA myocytes also increased detectably during treatment with 10 nM Iso over a similar time course (Fig. 3A, red).

Fig. 3.

Reduced basal Ca current and decreased response to β-adrenergic stimulation in adult STAA cardiomyocytes. (A) Time course of I_Ca following application of 10 nM Iso. Cells were depolarized from holding potential of −40 to +10 mV. (Inset) Representative traces, 0 and 5 min. (Scale: 0.2 nA, 20 ms.) (B–D) I–V relationships of I_Ca in absence and presence of Iso at 3 nM (WT V_a = −7.82 ± 1.9 vs. −17.54 ± 1.8 mV, P < 0.001; STAA V_a = −7.34 ± 1.3 vs. −5.5 ± 0.2 mV), 10 nM (WT V_a = −8.27 ± 1.7 vs. −16.72 ± 1.65 mV, P < 0.01; STAA V_a = −7.34 ± 1.3 vs. −9.86 ± 1.9 mV), and 100 nM (WT V_a = −5.21 ± 0.6 vs. −16.12 ± 1.3 mV, P < 0.001; STAA V_a = −8.09 ± 0.8 vs. −17.99 ± 1.2 mV, P < 0.01), respectively. (E) Peak current density at 0, 3, 10, and 100 nM Iso in WT and STAA myocytes. (F) Concentration–response curve of Iso-elicited increase in current (baseline subtracted) (EC₅₀ = 7.2 nM for WT; 33 nM for STAA).

To quantify the extent of up-regulation of L-type Ca²⁺ currents by Iso, we determined current–voltage relationships for cells treated for 0 or 5 min with 3, 10, or 100 nM Iso (Fig. 3 B–D). Basal Ca²⁺ currents in young-adult STAA myocytes were markedly reduced in all three datasets (Fig. 3 B–D, red; 2.7 ± 0.2 pA/pF) compared with WT (Fig. 3 B–D, black; 9.1 ± 1.3 pA/pF). Treatment of WT myocytes with 3 nM Iso increased peak Ca_V1.2 currents by an increment of 5.3 ± 2.3 pA/pF, and the voltage dependence of the L-type Ca²⁺ current was shifted to more negative membrane potentials (Fig. 3B, black), whereas no response was detected in STAA myocytes (Fig. 3B, red). Following treatment with 10 nM Iso, the peak L-type Ca²⁺ current was increased by 9.2 ± 1.7 pA/pF over basal and also shifted to more negative potentials in WT (Fig. 3C, black), whereas the smaller current in STAA mutant myocytes was increased by only 1.6 ± 0.5 pA/pF. With 100 nM Iso, the L-type Ca²⁺ current was increased by 15.3 ± 0.8 pA/pF (P < 0.001; n = 9), an increase of 3.1 ± 0.3-fold, which approaches the maximal stimulation observed with young-adult mouse ventricular myocytes (Fig. 3D, black). The negative shift in the voltage dependence of activation was also prominent (Fig. 3D, black). At this nearly saturating concentration of Iso, we also observed a clear incremental increase in peak L-type Ca²⁺ currents for STAA myocytes (4.92 ± 0.6 pA/pF) and a clear negative shift in the voltage dependence of Ca²⁺ currents (Fig. 3D, red).

WT and STAA ventricular myocytes have nearly the same level of Ca_V1.2 channels (Fig. 1C), and phosphorylation of Ser1700 and Thr1704 regulates both basal and stimulated Ca_V1.2 channel activity in transfected nonmuscle cells (10). Therefore, the full impact of the mutations in these residues is best illustrated by concentration–response curves that plot the absolute magnitude of the L-type Ca²⁺ current vs. Iso concentration (Fig. 3E). Four effects are evident in these concentration–response curves. (i) The magnitude of the L-type Ca²⁺ current is much reduced at all concentrations of Iso in STAA myocytes, even though the number of Ca_V1.2 channels is only reduced ∼12%, indicating that each Ca_V1.2 channel has much lower basal activity. (ii) No response to Iso is detectable in STAA myocytes at the lowest concentration of Iso illustrated (3 nM). (iii) The concentration dependence of the increase in Ca²⁺ current caused by Iso is shifted from EC₅₀ = 7.2 nM to EC₅₀ = 33 nM. (iv) Even at the near-saturating concentration of 100 nM Iso, the magnitude of L-type Ca²⁺ current is much reduced in STAA myocytes [STAA, 7.6 ± 1.3 pA/pF, vs. WT, 22.6 ± 1.3 pA/pF (P < 0.001)], and the stimulation by Iso is also significantly less than WT.

Although it is typical to analyze the effects of Iso treatment on Ca²⁺ currents in cardiac myocytes by plotting the ratio of stimulated current to basal current, this simple metric is misleading here because the level of basal Ca²⁺ current conducted by approximately the same density of Ca_V1.2 channels in STAA myocytes (Fig. 1E) is much lower than in WT myocytes and greatly reduces the denominator in this ratio because of the effects of the mutations. To focus specifically on up-regulation of WT and STAA Ca_V1.2 channel activity by Iso stimulation of a constant number of Ca_V1.2 channels, we have plotted the basal-subtracted incremental increase in pA/pF vs. Iso concentration in Fig. 3F. This plot illustrates the amount of Ca²⁺ current that is recruited by treatment with Iso for approximately the same density of WT and STAA Ca_V1.2 channels in these two genotypes of myocytes and therefore captures the full effect of loss of β-adrenergic/PKA-dependent up-regulation of activity of a similar density of WT and STAA Ca_V1.2 channels. These results show that the EC₅₀ for increase in Ca²⁺ current by Iso is increased and the maximum Iso-stimulated Ca²⁺ current is markedly decreased in STAA myocytes.

Impaired β-Adrenergic Regulation of Contraction in STAA Myocytes.

The Ca_V1.2 channel controls the influx of Ca²⁺ that initiates Ca²⁺-induced Ca²⁺ release and cellular contraction. Therefore, phosphorylation of Ca_V1.2 channels by the β-adrenergic/PKA pathway mediates the increased force of contraction in fight-or-flight response. Because blocking phosphorylation at Ser1700 and Thr1704 results in reduced basal Ca²⁺ current and reduced sensitivity to β-adrenergic regulation in single dissociated ventricular myocytes, we measured shortening of similarly prepared single ventricular myocytes in response to electric field stimulation at 1 Hz under conditions similar to our electrophysiological measurements. All WT myocytes and most STAA myocytes responded to 1-Hz stimulation with a rhythmic series of contractions that reduced cell length significantly, and treatment with Iso increased the magnitude of contraction (Fig. 4A). Surprisingly, there was no significant difference in fractional shortening under basal conditions, despite the markedly reduced basal L-type Ca²⁺ current in STAA myocytes (Fig. 4A). However, the increase in cell shortening in response to treatment with Iso was significantly reduced in STAA myocytes at both 10 and 100 nM (Fig. 4A). These results show that the decreased up-regulation of Ca²⁺ currents in STAA myocytes (Figs. 2 and 3) leads to impaired up-regulation of cellular contractility (Fig. 4A).

Fig. 4.

Iso-induced increase in contractility was impaired in STAA ventricular myocytes. (A) (Upper) Representative trace of cell fractional shortening in response to 0, 10, and 100 nM Iso. Cells were field-stimulated at 1 Hz. (Lower) Mean percentage shortening (**P < 0.01). (B) (Upper) Representative traces of cells failing to follow field stimulation at 1 Hz or entering total cellular arrhythmia (lower trace). (Lower) Fraction of cells exhibiting contraction failure and arrhythmia.

Contractile Failure and Arrhythmia in STAA Myocytes.

Although the vast majority of dissociated myocytes showed rhythmic contractions during field stimulation, a portion of cells (∼20%) from STAA mice failed to sustain contractions during stimulation at 1 Hz (Fig. 4B, Failure). In addition, a smaller group of STAA myocytes exhibited arrhythmic contractions that did not follow the frequency of stimulation and did not generate measurable cell shortening (Fig. 4B, Arrhythmic contraction). The proportion of cells exhibiting contractile failure and arrhythmia was much greater for STAA myocytes than for WT myocytes (Fig. 4B). Interestingly, only two cells from STAA mice and one cell from WT showed failure to sustain contraction under basal conditions. The vast majority of contractile failures and arrhythmic contractions were observed during treatment with Iso (Fig. 4B). These results demonstrate that phosphorylation of Ser1700 and Thr1704 plays an essential role in regulation of Ca²⁺ current and cellular contractility by the β-adrenergic pathway and also is required to sustain the normal rhythm of electrical conduction and contraction in response to repetitive depolarizing stimuli during Iso stimulation.

Reduced Exercise Capacity in STAA Mice.

The substantial deficits in β-adrenergic regulation of L-type Ca²⁺ currents and contractility in STAA myocytes would be expected to have major effects on the cardiovascular system. To test the cardiac response to sympathetic stimulation in the fight-or-flight response, WT and STAA mice were subjected to treadmill running on an inclined plane to escape a mild foot shock, and the total distance run before exhaustion was measured. Total distance run by STAA mice was shorter compared with WT by ∼30% (Fig. 5A), indicating decreased maximal exercise capacity. This result supports the hypothesis that phosphorylation at Ser1700 and Thr1704 of Ca_V1.2 channel is essential for the fight-or-flight response in vivo.

Fig. 5.

Impaired exercise capacity and cardiac hypertrophy in STAA mice. (A) Total distance run on treadmill to reach exhaustion. **P < 0.01. (B) Cell capacitance (Left) measured in patch clamp and heart-to-body weight ratio (Right). *P < 0.05.

Cardiac Hypertrophy in STAA Mice.

Impaired β-adrenergic regulation of L-type Ca²⁺ currents and contractility in STAA myocytes would also be expected to have major effects on cardiac physiology and homeostasis. We observed increased heart-to-body weight ratio in STAA mice compared with their littermate controls, indicating cardiac hypertrophy (Fig. 5B). Cardiac hypertrophy results from increased myocyte size rather than from proliferation of myocytes. As expected, therefore, our electrophysiological recordings of STAA myocytes showed a corresponding increase in cell capacitance, which is proportional to their cell surface area (Fig. 5B). This result indicates that compensatory mechanisms have been engaged in vivo to overcome the deficit in contractile function caused by disruption of phosphorylation of Ca_V1.2 channel at Ser1700 and Thr1704. These compensatory mechanisms may lead to more advanced hypertrophy and eventually to heart failure in older mice.

Discussion

Altered Levels of Ca_V1.2 and dCT in STAA Myocytes.

STAA myocytes have ∼88% of the WT level of Ca_V1.2 channels (Fig. 1C). In previous work (14), deletion of the dCT reduced Ca_V1.2 channels to ∼10% of WT. These results demonstrated an important role for the dCT in assembly, insertion, or cell surface stabilization of Ca_V1.2 channels. The STAA mutation has much less effect than deletion of the dCT, suggesting that phosphorylation of Ser1700 and Thr1704 has a smaller impact on the number of Ca_V1.2 channels on the cell surface. The C-terminal domain of Ca_V1.2 channels is proteolytically processed in vivo, and <20% of the α₁1.2 subunits isolated from the heart are full length (8). The proteolytically processed dCT serves as a noncovalently associated autoinhibitor of Ca_V1.2 channel activity, and its autoinhibitory effects are relieved by phosphorylation of Ser1700 and Thr1704 (10). Phosphorylation may also weaken the association of the dCT with Ca_V1.2 channels and increase its dissociation and degradation, but there is no direct evidence for this mechanism from previous studies in transfected nonmuscle cells. Our results showing that the level of the dCT is increased in STAA myocytes (Fig. 1D) are consistent with the idea that block of phosphorylation of Ser1700 and Thr1704 does indeed stabilize the dCT in vivo, perhaps by stabilizing its association with Ca_V1.2 channels and thereby enhancing its autoinhibitory effects.

Phosphorylation of Ser1700 and Thr1704 Is Required for Normal Basal Activity of Ca_V1.2 Channels.

Although STAA myocytes have ∼88% of the number of Ca_V1.2 channels compared with WT, they have only ∼30% of the basal L-type Ca²⁺ current. This substantial reduction of basal Ca_V1.2 channel activity in STAA ventricular myocytes agrees with previous studies of these same Ca_V1.2 mutations expressed in nonmuscle cells (10). Many signaling pathways are thought to contribute to control of the basal activity of Ca_V1.2 channels in cardiac myocytes (30). Classic studies of PKA regulation of L-type Ca²⁺ current in guinea pig ventricular myocytes showed that inhibition of PKA with excess regulatory subunit or the peptide PKI reduced basal Ca²⁺ currents significantly. Likewise, inhibitors of other protein kinases reduce basal L-type Ca²⁺ currents (31). Regulation of basal L-type Ca²⁺ currents is reconstituted in nonmuscle cells expressing the Ca_V1.2 autoinhibitory signaling complex, and blocking the activity of PKA and other protein kinases significantly reduces basal Ca_V1.2 channel activity (10). In that cell system, mutation of either Ser1700 or Thr1704 reduced basal L-type Ca²⁺ currents, indicating that these two phosphorylation sites are both involved in control of basal Ca_V1.2 channel activity. Ser1700 is phosphorylated by PKA and calcium/calmodulin-dependent protein kinase II, so both of these kinases may regulate Ca_V1.2 channel activity through phosphorylation of this site (10, 27). Thr1704 is phosphorylated by casein kinase II, implicating this constitutively active protein kinase in control of basal Ca_V1.2 channel activity as well (10, 27). Our results extend these previous findings to native ventricular myocytes by showing that mutation of Ser1700 and Thr1704 in vivo greatly reduces basal L-type Ca²⁺ currents. These experiments demonstrate that multiple protein kinase pathways can potentially contribute to the multifaceted regulation of basal Ca_V1.2 channel activity through phosphorylation of these two sites. It is likely that control of basal L-type Ca²⁺ current by the β-adrenergic regulatory pathway contributes importantly to physiological homeostasis in cardiac myocytes and the heart as a whole.

Phosphorylation of Ser1700 and Thr1704 Is Required for Normal β-Adrenergic Regulation of Ca_V1.2 Channels.

Extensive research shows that β-adrenergic regulation of L-type Ca²⁺ current in the heart is caused by activation of PKA and phosphorylation of the Ca_V1.2 channel. Our results show that this response is dramatically impaired in STAA mice. However, the impairment is greatest at low levels of stimulation by Iso. After treatment with 3 nM Iso, the up-regulation of Ca_V1.2 channel activity is nearly one-third of maximal in WT ventricular myocytes, but no response is detected in STAA myocytes. As this range of stimulation of the β-adrenergic signaling cascade is likely to occur frequently during normal physiological regulation, the complete loss of Ca_V1.2 channel regulation at this level of stimulation would have profound physiological consequences. The greater up-regulation of Ca_V1.2 channel activity by treatment with 10 and 100 nM Iso is also substantially impaired in STAA mice. These results show that phosphorylation of Ser1700 and Thr1704 is important for regulation of Ca_V1.2 channels across the full range of physiological stimulation of the β-adrenergic regulatory pathway. However, recent studies of virally transduced ventricular myocytes did not detect a significant role for phosphorylation of Ser1700 (32). We consider it likely that the overexpression of Ca_V1.2 channels in this experimental system and use of only one high concentration of Iso (200 nM) are responsible for the different results (SI Discussion).

Our results indicate that the effect of the STAA mutation on β-adrenergic regulation is smaller when high concentrations of Iso (≥100 nM) are used to maximally stimulate the signaling cascade. These results suggest that an alternative regulatory mechanism is engaged at high stimulus levels in parallel with phosphorylation of Ser1700. Based on the extensive evidence indicating that PKA signaling is responsible for β-adrenergic regulation of Ca_V1.2 (3, 17, 18, 20, 33–37), it seems most likely that this alternative regulatory mechanism involves PKA phosphorylation of undetected sites in Ca_V1.2 channels or associated regulatory proteins. A recent genome-wide phosphoproteomic study (38) also detected Ser1700 as a major phosphorylation site on cardiac Ca_V1.2 channels in response to β-adrenergic stimulation. Further studies of STAA mice using such powerful phosphoproteomic methods may allow identification of new PKA phosphorylation sites that are engaged at high levels of stimulation by the β-adrenergic signaling cascade.

Phosphorylation of Ser1700 and Thr1704 Is Required for Regulation of Contraction.

Ca²⁺ entry through Ca_V1.2 channels initiates excitation–contraction coupling, but the relationship between Ca²⁺ entry and contraction is nonlinear (2, 39). For that reason, submaximal reduction of Ca²⁺ entry is expected to cause a proportionally smaller reduction in contractile force. Nevertheless, we were surprised that the large reduction in basal Ca²⁺ current in STAA myocytes did not cause a detectable reduction in cellular contraction under basal conditions. Unlike other species, the Ca²⁺ store in the sarcoplasmic reticulum in mouse or rat ventricular myocytes is relatively full at rest and at low stimulation frequency (40); therefore, Ca²⁺ release and subsequent contraction may be less dependent on Ca²⁺ influx as a trigger under basal physiological conditions. In addition, failure to sustain contractions during repetitive stimulation at 1 Hz may reflect the deficiency in Ca²⁺ reloading of the sarcoplasmic reticulum due to the large reduction in Ca²⁺ influx through STAA Ca_V1.2 channels. It also seems likely that compensatory mechanisms have been engaged to increase the response of the ryanodine-sensitive Ca²⁺ release channels to the lower L-type Ca²⁺ current in STAA myocytes and/or the sensitivity of the contractile apparatus to Ca²⁺ has been increased. Further studies will be required to dissect the downstream compensatory changes that may have developed in the excitation–contraction coupling pathway in STAA mice.

Even though we observed no reduction of contraction under basal conditions, STAA myocytes were substantially less responsive to β-adrenergic stimulation than WT myocytes. The reduction of approximately twofold in the contractile response to 10 nM Iso, near the middle of the concentration–response range of β-adrenergic regulation, would be expected to have major physiological consequences in the response of intact cardiovascular system to sympathetic stimulation.

Impaired Phosphorylation of Ser1700 and Thr1704 Leads to Reduced Exercise Capacity and Cardiac Hypertrophy in Vivo.

The substantial impairment of Ca_V1.2 channel regulation and myocyte contractility that we have observed at the cellular level leads to major adverse consequences for STAA mice. Their maximum exercise capacity in a simulated fight-or-flight setting is greatly reduced. A mild foot shock is a powerful fear-inducing stimulus in mice, so it is likely that STAA mice exerted maximum effort to avoid it. Despite the strong fear-related stimulus, they were unable to perform as well as WT mice in running up an inclined treadmill to escape. The impairment of Ca_V1.2 channel regulation and contractility that we have observed at the cellular level is also sufficient to cause substantial cardiac hypertrophy. The 25% increase in heart-to-body weight ratio in young adulthood is surprisingly large and presages major cardiovascular pathology at older ages. These results show clearly that the substantial reduction of β-adrenergic regulation of Ca_V1.2 channels at the lower end of the range of physiological activity of this signaling pathway is sufficient to cause major disruption of cardiac function and lead to serious cardiac pathology.

Potential Effects of Altered Ca_V1.2 Regulation in Other Tissues.

Our studies have focused on Ca_V1.2 channels in cardiac myocytes. However, Ca_V1.2 channels are broadly expressed in vascular smooth muscle, neurons, endocrine cells, and many other cell types (5). For example, impairment of voltage-dependent inactivation of Ca_V1.2 channels by mutations in Timothy syndrome causes cardiac arrhythmia, developmental abnormalities, autistic-like behaviors, and other pathologies (41, 42). Alterations of vascular regulation may contribute to the deficits of cardiovascular function and regulation in STAA mice. However, previous studies of mice with the dCT deleted showed far more serious loss of regulation of Ca_V1.2 channels and cardiac hypertrophy, without any observable change in the density of these channels in vascular smooth muscle cells (14). Therefore, our working hypothesis is that the milder pathological effects observed in these studies of STAA mice derive primarily or entirely from impaired cardiac function. Future studies of the impact of mutation of Ser1700 and Thr1704 on vascular, neuronal, and endocrine functions are likely to reveal impairments of β-adrenergic regulation in those cell types as well.

Materials and Methods

STAA mice were generated according to standard protocols (SI Text) by inGenious Targeting Laboratory. Neonatal cardiomyocytes were isolated on postnatal days 1–3 (14). Adult ventricular myocytes were isolated at 3–5 mo (43). L-type Ba²⁺ and Ca²⁺ currents in dissociated cardiomyocytes were recorded as described previously (14). For immunocytochemistry, ventricular myocytes were incubated for 2–3 h after isolation, changed to Tyrode’s solution containing 5 µM BAPTA-AM, and then incubated at room temperature for 20 min, followed by 10 min in absence or presence of 100 nM Iso. Cells were fixed and immunolabeled as described previously (13). For measurement of cell shortening, myocytes were field stimulated at 1 Hz. Cell length was measured from video frames and expressed as percentage of original length. To measure exercise tolerance, animals ran until exhaustion on a treadmill set at 5% incline. All exercise tolerance experiments were carried out on naive mice so that learned responses to fear would not alter the results. See SI Text for additional experimental details.