β-Adrenergic Stimulation of L-type Ca²⁺ Channels in Cardiac Myocytes Requires the Distal Carboxyl Terminus of α_1C but Not Serine 1928

Abstract

β-Adrenoceptor stimulation robustly increases cardiac L-type Ca²⁺ current (I_CaL); yet the molecular mechanism of this effect is still not well understood. Previous reports have shown in vitro phosphorylation of a consensus protein kinase A site at serine 1928 on the carboxyl terminus of the α_1C subunit; however, the functional role of this site has not been investigated in cardiac myocytes. Here, we examine the effects of truncating the distal carboxyl terminus of the α_1C subunit at amino acid residue 1905 or mutating the putative protein kinase A site at serine 1928 to alanine in adult guinea pig myocytes, using novel dihydropyridine-insensitive α_1C adenoviruses, coexpressed with β₂ subunits. Expression of α_1C truncated at 1905 dramatically attenuated the increase of peak I_CaL induced by isoproterenol. However, the point mutation S1928A did not significantly attenuate the β-adrenergic response. The findings indicate that the distal carboxyl-terminus of α_1C plays an important role in β-adrenergic upregulation of cardiac L-type Ca²⁺ channels, but that phosphorylation of serine 1928 is not required for this effect.

Voltage-gated L-type (Ca_v1.2) Ca²⁺ channels play a central role in controlling cardiac function, providing the trigger for intracellular Ca²⁺ release during excitation-contraction coupling, contributing to the plateau phase of the cardiac action potential, and modulating pacemaker activity in the sinoatrial node.^1-3 The L-type Ca²⁺ channel is a multiprotein complex, comprised of the pore-forming α₁ subunit and the auxiliary β₂ and α₂δ subunits.⁴ In the heart, the molecular identity of the β subunit remains a subject of ongoing investigation.^5,6

A number of receptor-mediated signal transduction pathways regulate Ca²⁺ influx via Ca_v1.2 channels, the most prominent of which is the β-adrenergic/cAMP signaling pathway. β-Adrenergic stimulation is the dominant mechanism of positive chronotropy, inotropy, and lusitropy in the heart.⁷ Activation of β-adrenergic receptors strongly enhances cardiac L-type Ca²⁺ current (I_CaL) via cAMP generation and the activation of protein kinase A (PKA), presumably by phosphorylating specific sites on the channel protein.^8,9 Nevertheless, the structural determinants of PKA action on the Ca²⁺ channel in native adult cardiomyocytes are unknown.

Previous biochemical studies have implicated the carboxyl terminus as a target of PKA-mediated phosphorylation: for example, a truncated form of the channel was reportedly not phosphorylated by PKA in vitro.¹⁰ The consensus phosphorylation site at serine 1928 was later identified as the main target of PKA.^11,12 On the other hand, phosphorylation sites have also been mapped to the β subunit of the channel and localized to serines 478 and 479 of β_2a.^13,14

Mixed results have been obtained when attempts have been made to confirm the functional role of these sites in the regulation of Ca²⁺ channel activity, in part, because of the difficulty of reconstituting PKA-mediated channel activation in heterologous expression systems. Whereas some studies have observed PKA-mediated activation of the channel in heterologous systems,^10,14 others have observed no such stimulation.^15,16 Other groups have used PKA inhibitors to argue that Ca_v1.2 channels are phosphorylated at baseline when expressed in cultured cell lines.^17-19

The failure to reproduce the strong upregulation of the L-type Ca²⁺ current in nonnative cells has been attributed to the lack of A-kinase anchoring proteins (AKAPs) in heterologous systems. By coexpressing AKAP 79, Gao et al²⁰ were able to reconstitute forskolin-mediated upregulation of Ca²⁺ channels expressed in HEK 293 cells, and the authors concluded that S1928 was necessary to produce PKA-mediated activation of the L-type current. In contrast, a subsequent study by the same group reported PKA-mediated activation of Ca²⁺ current through L-type channels with a mutant α_1C truncated at 1905 and provided evidence indicating that β-subunit phosphorylation was important.¹⁴ Additionally, other investigators have been unable to reproduce PKA-mediated upregulation of Ca²⁺ channels in HEK 293 cells, even in the presence of AKAP 79,²¹ and it has been suggested that the primary role of AKAP 79 is to facilitate α₁ subunit trafficking.²²

The equivocal findings in heterologous expression systems suggest that the actual site responsible for β-adrenergic regulation of L-type Ca²⁺ channel in cardiac myocytes remains undefined. In the present study, we use novel adenoviral gene transfer vectors to express mutant α_1C subunits in native guinea pig cardiomyocytes to address this fundamental question of L-type Ca²⁺ channel physiology. The evidence supports a role for the distal carboxyl terminus of the α₁ subunit in the β-adrenergic enhancement of I_CaL. However, serine 1928 phosphorylation is not required.

Materials and Methods

Vector Construction

A full-length, dihydropyridine (DHP)-insensitive rabbit Ca_v1.2 (X15539) α₁ subunit adenoviral shuttle vector pAdEcd-α_1C(D-) was prepared as previously described.²³ A serine to alanine substitution of α_1C(D-) at S1928 (α_{1C(D-)-S1928A}) was constructed by site-directed mutagenesis with overlapping primers, using the QuikChange XL kit (Stratagene). A carboxyl-terminal truncation mutant, pAdEcd-α_1C(D-)Δ1905, in which the carboxyl terminus of the α_1C was truncated after amino acid 1905, was also created. Hence, this construct also lacked the putative phosphorylation site at serine 1928. Mutations at Thr1066 (to Tyr) and Glu1089 (to Met) were also incorporated in this construct to make the channel DHP insensitive, as for α_1C(D-) and α_{1C(D-)-S1928A}, so that the current contributed by the mutated channels could be discriminated from that of native Ca_v1.2 channels.²³ Concentration-response experiments with nitrendipine were performed on α_1C(D-), α_1C(D-)S1928A, and α_1C(D-)Δ1905 in A549 cells as previously described²³ to verify that each construct produced DHP-insensitive current. A full-length wild-type α_1C fused to green fluorescent protein at its carboxyl terminus, α_1C-GFP, was also constructed for the Western blot experiments shown in Figure 2.

Figure 2.

Overexpression of DHP-insensitive α_1C enables pharmacological discrimination of native and exogenously expressed channels. Whole-cell current recordings were undertaken in adult myocytes transduced with α_1C adenoviruses and β_2a-AA adenoviruses. A repeated-pulsing protocol at 0.5 Hz to 0 mV was used to obtain a steady state, following which 10 μmol/L nitrendipine was added. Example current tracings are shown in myocytes transduced with β_2a-AA (A) or α_1C(D-)/β_2a-AA (B) adenoviruses, before (black) and after (gray) the application of 10 μmol/L nitrendipine. In C, the fraction of I_CaL left unblocked after the application of nitrendipine is shown following transduction with each of the DHP-insensitive α_1C adenoviruses. D, Representative Western blot. Left, Lane 1 was transduced with α_1C-GFP and AdVgRXR; lane 2, with AdVgRXR alone. Right, The same membrane was reprobed with anti-GFP antibody, confirming that the strong upper band was attributable to α_1C-GFP expression. The absence of the middle band in lane 1 is indicative of carboxyl-terminal processing of α_1C-GFP. In E, the anti-Ca_v1.2 signal intensity plot profile is shown for α_1C-GFP and control cells, demonstrating the presence of novel peaks in upper and middle bands caused by α_1C-GFP expression. F, Representative image of an α_1C-GFP/AdVgRXR-transduced myocyte, stained with anti-GFP primary antibody, and an Alexa-Fluor 568 secondary antibody. *P<0.05 vs β_2a-AA.

The adenoviral shuttle plasmid pAd-β_2a, was created by subcloning the rat β_2a cDNA (NM053851) into the EcoR1 site of the adenoviral shuttle vector pAdLox. To avoid potential confounding effects of β-subunit phosphorylation on the response, the S478A/S479A double mutation of β_2a (pAd-β_{2a-S478/S479A}; β_2a-AA) was also created via site-directed mutagenesis with overlapping primers, using the QuikChange XL kit. pAd-VgRXR, a bicistronic vector containing the elements of the ecdysone receptor, was created as previously described.^23,24

Adenoviral Construction

Adenoviruses were constructed by the Cre-lox technique, according to the methods of Hardy et al,²⁵ with variations as previously described.²³ The α_1C adenovirus shuttle vectors pAdEcd-α_1C(D-), pAdEcd-α_1C(D-)S1928A, or pAdEcd-α_1C(D-)Δ1905 were cotransfected onto CRE-FB cells, along with ψFB6 adenoviral DNA as previously described.²³ Resulting viruses were expanded in 3 rounds and purified by plaque assay on 911-FB cells. The plaque purified virus was further expanded before purification on CsCl gradients, and dialyzed against 10 mmol/L Tris-HCl/1 mmol/L MgCl₂. Viral titers were estimated by particle counts and calculated by plaque assay performed on 911-FB cells.

Ad-β_2a-AA and AdVgRXR were grown by cotransfection of their respective shuttle plasmids with ψ5 adenoviral DNA,²⁵ onto standard Cre-8 cells.²⁵ These viruses were expanded in 3 rounds and purified on CsCl gradients, as described previously.^23,26 Viral titers were estimated by particle counts and calculated by plaque assay performed on 911 cells.²⁷

Adult Cardiac Myocyte Expression Experiments

Animal protocols used were in accordance with the Guide for Care and Use of Laboratory Animals, published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996). Isolated guinea pig ventricular myocytes were obtained by enzymatic digestion as previously described.²⁸ Freshly isolated myocytes (20 to 40×10³/mL) were plated on laminin-coated coverslips and bathed in DMEM supplemented with 5% FBS and 1% penicillin-streptomycin. Virus was added after allowing cells to adhere to coverslips for 3 to 4 hours. For experiments involving overexpression of β_2a-AA, cells were transduced with Ad-β_2a-AA at ≈1×10⁹ plaque-forming units (PFU).

For the DHP-insensitive α_1C experiments, cells were transduced with Ad-VgRxR at 3×10⁸ PFU/mL and Ad-β_2a-AA at 1×10⁹ PFU/mL, along with either Ad-α_1C(D-), Ad-α_1C(D-)S1928A or Ad-α_1C(D-)Δ1905 at ≈5×10⁹ PFU/mL. As we have described previously, coexpression of the α_1C constructs with Ad-β_2a-AA was necessary to enhance I_CaL amplitude, which facilitates detection of the DHP-insensitive Ca²⁺ currents.²³

After incubation with the viruses, the medium was replaced with fresh DMEM containing 10 μmol/L ponasterone A (H101-01, Invitrogen). Electrophysiological recordings were made at 36 to 48 hours following isolation.

Western Blot Experiments

The anti-Ca_v1.2 antibody used in these studies was an affinity purified rabbit polyclonal antibody, raised against a peptide made from amino acids 1 to 46 of the rabbit α_1C subunit N terminus (Alomone Labs, Jerusalem, Israel). The anti-green fluorescence protein (GFP) antibody was a mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, Calif). The lysate was prepared by scraping cultured cardiac myocytes into ice-cold RIPA buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate in PBS containing 5 mmol/L ethylenediaminetetraacetic acid, 1 mmol/L β-glycerol phosphate, 1 mmol/L L-phenylalanine, 1 mmol/L sodium orthovanadate, 50 mmol/L sodium fluoride, and the following protease inhibitors: pepstatin A, 2 μg/mL; leupeptin, 10 μg/mL; aprotinin, 10 μg/mL; elastinal, 2 μg/mL; benzamidine, 0.5 mg/mL; calpain inhibitor peptide 10 μg/mL; and 1 mmol/L phenylmethane-sulfonyl fluoride, 0.4 mmol/L iodoacetic acid) and passed through a 25-gauge needle 15 times. This homogenate was lysed on ice for 1 hour, before centrifugation at 2000g for 20 minutes at 4°C.

The concentration of the supernatant protein was measured with the Bio-Rad DC protein assay, according to the instructions of the manufacturer, and samples were divided into aliquots and frozen at -20°C. Samples were mixed with 5:1 Laemmli sample buffer and incubated at room temperature for 60 minutes, before gel loading.

Following separation by 7.5% SDS-PAGE, proteins were transferred to polyvinylidene sulfonyl fluoride membranes (Bio-Rad) in Tris/glycine transfer buffer containing 5% methanol and 0.05% sodium dodecyl sulfate. Membranes were blocked with 5% nonfat milk in PBS supplemented with 0.05% Tween 20 (PBS-T), for 1 hour at room temperature, and then incubated with primary antibody (1:50 to 1:200) overnight at 4°C. The membrane was washed 6 times for 5 minutes with PBS-T and incubated with horseradish peroxidase-conjugated donkey anti-rabbit antibody diluted 1:50 000 in PBS-T. Following 6 further washes in PBS-T, blots were visualized with the Lumi-light Plus blotting substrate (Roche), according to the instructions of the manufacturer.

Immunofluorescence

Immunofluorescence labeling experiments were undertaken using myocytes cultured with α_1C-GFP using an anti-GFP antibody (Abcam-6556). Cultured myocytes were fixed for 20 minutes at room temperature in freshly prepared 3% paraformaldehyde in PBS. Fixed cells were permeabilized in 0.5% Triton X-100 in PBS/10 mmol/L glycine for 3 minutes. After 2 further washes in PBS/10 mmol/L glycine, myocytes were incubated for 1 hour in PBS/10% goat serum to block nonspecific binding. Following 2 washes in PBS/10 mmol/L glycine, the cells were incubated in primary antibody diluted 1:500 in PBS/10 mmol/L glycine supplemented with 4% goat serum, for 1 hour. The cells were then washed 2 times in PBS/10 mmol/L glycine, before a 1-hour incubation with secondary antibody, Alexa 568 goat anti-rabbit IgG (H+L) (Invitrogen), and diluted 1:500 in PBS/10 mmol/L glycine supplemented with 4% goat serum. After 2 final washes with PBS/10 mmol/L glycine, cells were mounted on coverslips. Experiments to determine nonspecific binding were performed with cells transduced with VgRXR alone, and control experiments were also carried out with secondary antibody alone.

Images were recorded using a 2-photon laser scanning microscope (Bio-Rad, MRC-1024MP) with excitation at 820 nm (Tsunami Ti:Sa laser, Spectra-Physics). The red emission of Alexa 568 was collected at 605±25 nm.

Electrophysiological Recordings

Electrophysiological recordings were performed using the whole-cell variation of the patch clamp technique. All experiments were performed at room temperature. Currents were recorded with an Axopatch 200B amplifier (Axon Instruments, Union City, Calif) and sampled at 5 kHz. The cells were superfused with an external solution comprised of (mmol/L): 140 NaCl, 5 CsCl, 1 MgCl₂, 10 HEPES, 2 CaCl₂, and 10 d-glucose. Patch pipettes (2.0 to 5.0 MΩ) were filled with an intracellular solution containing (mmol/L): 110 CsCl, 0.4 MgCl₂, 5 d-glucose, 10 HEPES, 20 tetraethylammonium, and 5 BAPTA (1,2-bis[2-aminophenoxy]ethane-N,N,N’,N’-tetraacetic acid). In experiments involving the use of DHP-insensitive α₁ subunits, 10 μmol/L nitrendipine (Sigma, St Louis, Mo) was washed in at the end of a subset of experiments to verify the expression of the exogenous α_1C. Voltage pulses were applied and data recorded with custom software. Leak currents were subtracted by a P/4 leak subtraction protocol. Total series resistance after establishing the whole-cell configuration was typically 10 MΩ and was electronically compensated by ≈70%. Thus we estimate the voltage error for a 1-nA current (the amplitude of isoproterenol [ISO]-stimulated peak inward I_CaL) to be <3 mV. I_CaL was elicited by 300-ms pulses to different test potentials (-80 to +50mV), following a 200-ms prepulse to -40 mV, from a holding potential of -80 mV. I_CaL measurements were normalized to cell capacitance, measured by integration of the area under an uncompensated -20 mV pulse from -80 mV. The current-density to voltage relationships (I-V) recorded were fitted to a modified Boltzmann equation:

$$I=G_{\max }(\mathrm{V}-{\mathrm{V}}_{\mathrm{rev}})∕(1+\exp (-(\mathrm{V}-V_{1∕2})∕k)$$

where G_max is the maximal conductance (nS/pF), V_rev is the reversal potential (mV), V_1/2 is the midpoint voltage of activation (mV), and k is the slope factor.

Data analysis was performed offline, with custom software (Ionview, B.O’R.). Statistical analysis was performed with Origin 6.0 (Microcal). All data are presented as the mean±SEM. Statistical significance was calculated using an unpaired Student’s t test, and differences were considered significant at P<0.05.

Results

β-Adrenergic Response in Adult Cardiomyocytes With β_2a-AA Expression

The primary objective of this study was to investigate the role of the distal carboxyl terminus of α_1C in the β-adrenergic response. In a prior study, we determined that replacement of the native Ca²⁺ channels with virally expressed mutants was most effectively achieved when β subunits were coexpressed,²³ consistent with earlier reports that β-subunit expression is a limiting factor for Ca²⁺ channel membrane trafficking in myocytes.⁵ However, our initial experiments indicated that the expression of wild-type β_2a subunits led to a marked hyperpolarizing shift (approximately -12.2 mV) in the midpoint of voltage-dependence for I_CaL activation (V_1/2) and a diminished ISO response, presumably attributable to the enhanced Ca²⁺ channel activity under basal conditions (data not shown). Mutation of the phosphorylation sites of β_2a at S478 and S479 to alanine lessened the hyperpolarizing shift in Ca²⁺ channel activation, but preserved the ability of the β subunit to facilitate Ca²⁺ channel trafficking. Expression of β_2a-AA subunits resulted in a marked increase in baseline L-type Ca²⁺ current density (uninfected myocytes -3.2±0.7 pA/pF versus β_2a-AA expressing -9.2±1.6 pA/pF; P<0.05) and a hyperpolarizing shift in V_1/2 of just -4.8 mV (Table). Importantly, a robust β-adrenergic response could be observed under these conditions. In Ad-β_2a-AA-transduced myocytes, ISO (100 nmol/L) caused a 95% increase in peak I_CaL (from -9.2±1.6 pA/pF to -17.9±1.4 pA/pF, P<0.05; Figure 1B and 1D) and a significant hyperpolarizing shift in V_1/2 (Table). This was comparable to the ISO-induced increase in peak inward L-type Ca²⁺ currents in uninfected cultured myocytes (Figure 1A and 1C; I_CaL baseline: -3.2±0.3 pA/pF versus 100 nmol/L ISO: -11.8±1.7 pA/pF). Thus, β_2a-AA coexpression was used for the subsequent experiments with α_1C mutants.

Voltage Dependence of Activation (V_1/2) of L-Type Ca²⁺ Channels in Cardiac Myocytes.

	V1/2	V1/2 (100 nmol/L ISO)
Control	-7.9±0.1	-20.4±2.2*
β2a-AA	-12.70±2.4	-24.20±2.6†
α1C(D-)/β2a-AA	-16.95±1.9	-25.21±1.9‡
α1C(D-)1905/β2a-AA	-14.52±2.4	-19.60±2.4§
α1C(D-)S1928A/β2a-AA	-13.5±3.3	-23.1±1.4¶

^*P<0.05 vs control

^†P<0.05 vs β_2a-AA

^‡P<0.05 vs α_1C(D-)/β_2a-AA

^§P=NS vs α_1C(D-)Δ1905/β_2a-AA

^¶P<0.05 vs α_1C(D-)S1928A/β_2a-AA

Figure 1.

Isoproterenol response in adult myocytes transduced with adenoviruses overexpressing β_2a-AA subunits. Whole-cell currents were recorded using voltage clamp pulses from a holding potential of -80 mV to test potentials, following a 200 ms prepulse to -40 mV, before and after the application of 100 nmol/L ISO. A and B, Example current tracings following pulses to 0 mV, before and after 100 nmol/L ISO. C and D, Current-voltage relationships: before (black) and after (gray) the application of 100 nmol/L ISO in uninfected myocytes (C) and myocytes transduced with β_2a-AA (D). β_2a-AA transduction caused a baseline increase in I_CaL and a small hyperpolarizing shift in the voltage dependence of activation; however, the β-adrenergic response was preserved. Values are mean±SEM.

Functional Expression of DHP-Insensitive α_1C Subunits in Cardiac Myocytes

Following a strategy similar to our previous study,²³ α_1C mutants were engineered to be dihydropyridine insensitive, to enable pharmacological discrimination of L-type Ca²⁺ current contributed by exogenously expressed channels from that of native Ca²⁺ channels. Consistent with our earlier results, a concentration of 10 μmol/L nitrendipine blocked more than 85% of I_CaL in myocytes transduced with Ad-β_2a-AA alone, whereas only ≈50% of the total I_CaL was blocked in α_1C(D-)+β_2a-AA-transduced cells (Figure 2). Similar fractions of DHP-insensitive current were observed for myocytes expressing β_2a-AA with either α_1C(D-)S1928A or α_1C(D-)Δ1905 (Figure 2C).

Significant expression of mutant α_1C subunits was also confirmed in biochemical studies. A full-length α_1C fused to green fluorescent protein at the carboxyl terminus was used to discriminate exogenous and native α_1C subunits by molecular weight on Western blots and to provide a specific tag for the heterologous channels. Two intense high molecular mass bands appeared (>220 kDa) in samples from myocytes infected with Ad-α_1C-GFP (Figure 2) using an anti-Ca_v1.2 antibody, in addition to the lowermost band representing native α_1C. Densitometry confirmed that the signal from the uppermost bands in α_1C-GFP cells represented 63±6% of total anti-Ca_v1.2 signal intensity (Figure 2). When reprobed with an anti-GFP antibody, only the uppermost band was evident, verifying the presence of the fusion protein and suggesting that the middle band was a carboxyl-terminal cleavage product of the Ca²⁺ channel.

Ad-α_1C-GFP was next used to examine the spatial localization of the expressed mutant channels. Immunofluorescence staining of fixed cardiac myocytes with an anti-GFP antibody conjugated to the fluorescent dye Alexa 568 revealed a membrane distribution of expressed α_1C subunits including strong transverse-bands consistent with t-tubular localization, in accord with the known distribution of α_1C subunits in myocytes. Hence, expression of mutant α_1C subunits was confirmed both functionally and biochemically.

Distal Carboxyl Terminus of α_1C Is Necessary for β-Adrenergic Regulation of Ca²⁺ Channels

To ensure that the observed responses could be exclusively attributed to the expressed α_1C mutants, we first examined the effects of β-adrenergic stimulation on I_CaL in the presence of 10 μmol/L nitrendipine. Under these conditions, 100 nmol/L ISO induced a 50±11% increase of peak I_CaL in cells expressing α_1C(D-)/β_2a-AA (Figure 3A and 3D). Remarkably, the ISO response was almost completely eliminated (I_CaL increased by only 3±7%) when the α_1C carboxyl terminus was truncated at amino acid residue 1905 (α_1C(D-)Δ1905/β_2a-AA; Figure 3B and 3D). To determine whether the loss of responsiveness involved impairment of the pathway between the β-adrenergic receptor and PKA, we also repeated the experiments using the membrane permeable cAMP analogue 8-bromo-adenosine 3′5′-cyclic monophosphate (8-Br-cAMP) to directly activate the kinase. Similar results were obtained, with a substantial reduction in 8-Br-cAMP induced Ca²⁺ current augmentation with α_1C(D-)Δ1905 compared with α_1C(D-) (Figure 3E).

Figure 3.

In the presence of nitrendipine, carboxyl-terminal truncation mutant shows diminished response to ISO, but S1928A remains responsive. Whole-cell current recordings were obtained in adult myocytes before and after 100 nmol/L ISO, for myocytes transduced with α_1C(D-) (A), α_1C(D-)Δ1905 (B), or α_1C(D-)S1928A (C) viruses. Current records are for repeated pulses to 0 mV in the continuous presence of 10 μmol/L nitrendipine, to block I_CaL carried by native channels. The fractional response to 100 nmol/L ISO is shown in D. ISO (100 nmol/L) elicited a robust increase in I_CaL in α_1C(D-) cells in the presence of the blocking dose of nitrendipine (A and D) but failed to elicit I_CaL upregulation in α_1C(D-)Δ1905 cells (B and D), indicating that the carboxyl terminus of α_1C is necessary for β-adrenergic modulation of Ca_v1.2 channels in cardiac myocytes. ISO also elicited a robust increase in I_CaL in α_1C(D-)S1928A cells, indicating that S1928 phosphorylation is not essential for β-adrenergic modulation of L-type Ca²⁺ current. E, 8-Br-cAMP (1 mmol/L) elicited an increase in I_CaL in α_1C(D-) cells in the presence of 10 μmol/L nitrendipine but failed to elicit I_CaL upregulation in α_1C(D-)Δ1905 cells. Values are mean±SEM. *P<0.05, **P<0.01.

The ISO response was also examined in the absence of nitrendipine, to eliminate any possibility that the results were somehow dependent on the presence of concomitant Ca²⁺ channel block. ISO increased peak I_CaL by 104±23% in myocytes expressing α_1C(D-) and by only 25±16% in myocytes expressing α_1C(D-)Δ1905. The small increase in the latter case is consistent with the presence of a small fraction of native L-type channels that would be available to be activated by ISO. In addition to the significant effect of the 1905 truncation to blunt the ISO effect on peak I_CaL, it is notable that the hyperpolarizing shift in V½ was also markedly reduced in myocytes expressing α_1C(D-)Δ1905 (Table).

S1928 Is Not Required for β-Adrenergic Regulation of Ca²⁺ Channels

In contrast to the effect of truncating the carboxyl terminus, the S1928A point mutation of the α_1C did not significantly diminish the ISO response, either in the presence or in the absence of nitrendipine (Figures 3 and 4). In myocytes expressing α_1C(D-)S1928A, ISO increased I_CaL by 82±16% in the absence of nitrendipine (Figure 4) and a prominent leftward shift in V_1/2 was still observed (Table).

Figure 4.

Carboxyl-terminal truncation mutant shows diminished response to ISO, but S1928A remains responsive. To ensure that the results were not a consequence of the presence of nitrendipine, whole-cell current recordings were obtained from myocytes before and after 100 nmol/L ISO in the absence of DHP. Current-voltage relationships before (black circles) and after (open triangles) 100 nmol/L ISO are plotted for myocytes transduced with α_{1C (D-)}/β_2a-AA (A), α_1C(D-)Δ1905/β_2a-AA (B), or α_1C(D-)S1928A/β_2a-AA (C) adenoviruses. The β-adrenergic response of I_CaL was markedly attenuated in α_1C(D-)Δ1905/β_2a-AA cells (B and D), as it was in the presence of nitrendipine (Figure 3). However, the β-adrenergic enhancement of I_CaL was preserved in α_1C(D-)S1928A/β_2a myocytes (C and D). The fractional increase in response to 100 nmol/L ISO of the peak I_CaL is shown in D. Values are mean±SEM. *P<0.05, **P<0.025.

Discussion

The 2 main findings of the study were (1) that the distal carboxyl terminus of α_1C is required for the β-adrenergic stimulation of L-type Ca²⁺ channels in cardiac cells and (2) that serine 1928 phosphorylation is not required. This was most evident in the failure of Ca²⁺ currents to be enhanced by ISO in α_1C(D-)Δ1905/β_2a-AA-transduced cells in the presence of nitrendipine. The β-adrenergic response was also significantly blunted in α_{1C(D-)Δ1905/}β_2a-AA-transduced cells in the absence of nitrendipine, but the presence of ISO-responsive native L-type Ca²⁺ channels made the distinction less prominent. The ISO response was not significantly diminished in myocytes expressing the S1928A α_1C point mutant.

The present study is the first to investigate the structural determinants of the α_1C subunit involved in the β-adrenergic regulation of L-type Ca²⁺ channels in native adult myocytes. This was enabled by the efficient expression of Ca_V1.2 mutants using novel adenoviral constructs. Discrimination between the heterologous channels and native channels was accomplished by engineering the former to be DHP-insensitive. Both functional and biochemical data indicated that the majority of the L-type Ca²⁺ channels expressed were derived from the viral construct, so that significant differences in function could be readily observed, even in the absence of the dihydropyridine.

An important technical issue was how to optimize Ca²⁺ channel expression using β-subunit coexpression without introducing confounding effects caused by alterations in baseline Ca²⁺ channel activity by the β subunit. We found that mutation of the consensus phosphorylation sites of β_2a prevented the pronounced hyperpolarization of the voltage dependence of L-type Ca²⁺ channel activation under basal conditions, permitting us to observe a robust ISO response. Beyond the technical advantage of this approach, the results with β_2a-AA indicate that β subunit phosphorylation is not obligatory for the β-adrenergic response, a conclusion that differs from previous studies of Ca²⁺ channel regulation for channels heterologously expressed in cultured cell lines.¹⁴

The present results also provide evidence against the current dogma that PKA-mediated Ser1928 phosphorylation is required for the β-adrenergic stimulation of I_CaL. Both the ISO-induced increase in peak inward I_CaL and the left shift in the V_1/2 of activation were preserved in cardiomyocytes expressing α_1C(D-)S1928A. In this regard, the functional data supporting S1928 as the site of modulation have been equivocal. Gao et al²⁰ reported that Ba²⁺ currents through expressed Ca²⁺ channels were upregulated by forskolin or 8-Br-cAMP in an AKAP 79 stably transfected HEK 293 cell line and that this response was absent for the S1928A α_1C point mutation. However, only half of the control cells tested were clearly responsive to the effects of the PKA activators, suggesting a high degree of variability in this model system. In addition, forskolin increased Ba²⁺ current by only ≈50%, and 8-Br-cAMP by ≈15%, much less than that typically observed for PKA-mediated I_CaL enhancement observed in cardiac myocytes. Moreover, a later study from the same laboratory¹⁴ reported that I_CaL mediated by the 1905 α_1C truncation mutant, lacking serine 1928, could be increased by PKA activation.

Other reports have also raised cautions about the conclusions reached in the earlier studies. Many investigators have failed to reproduce PKA modulation of L-type Ca²⁺ currents in heterologous expression systems. Zong et al¹⁶ tested 25 μmol/L PKA along with okadaic acid in CHO and HEK 293 cells expressing either α_1C-a or α_1C-b and were not able to observe PKA-mediated potentiation of Ba²⁺ currents. Mikala et al¹⁵ systematically substituted 5 potential PKA phosphorylation sites in human α_1C, including the S1928 homolog, and found that modulation of baseline L-type Ca²⁺ current behavior could not account for the failure to observe PKA modulation. Altier et al²² also failed to observe PKA-mediated activation of Ba²⁺ currents in HEK 293 cells, even in the presence of AKAP 79 overexpression.

Although the functional effects of S1928 may be questionable, there is agreement that S1928 can be readily phosphorylated by PKA in vitro.^11,12 However, according to a recent report, this site appears to be promiscuously phosphorylated by at least 3 isoforms of PKC (cPKCα, nPKCε, and aPKCζ) in addition to PKA.²⁹ This raises an important question about the physiological relevance of this in vitro biochemical evidence, considering that the considerably divergent physiological effects of PKA and PKC activation on channel activity.⁸

The present data suggest that although S1928 is not critical for the β-adrenergic effect on Ca²⁺ channel function, the distal α_1C carboxyl terminus beyond residue 1905 is required. One plausible explanation for this effect could be that the carboxyl terminus plays a central role in the assembly of a PKA signaling complex. For example, Hulme et al³⁰ recently demonstrated an interaction between AKAP 15 and a leucine zipper motif in the distal carboxyl terminus of α_1C using immunoprecipitation methods. Disruption of the interaction with competing AKAP 15 leucine zipper peptides substantially reduced ISO-induced upregulation of L-type Ca²⁺ currents in cardiac myocytes, suggesting that the AKAP 15-α_1C interaction was necessary for efficient β-adrenergic regulation of Ca²⁺ currents in cardiac myocytes.³⁰ Further investigation into the role of the leucine zipper motif in the response using the present strategy will be important.

The existing literature provides a number of potential alternative sites of PKA-mediated phosphorylation of α_1C. For example, Leach et al,³¹ using phosphopeptide antibodies, reported that serines 1627 and 1700 were the primary sites of PKA-mediated phosphorylation of the cardiac Ca²⁺ channel α_1C subunit, whereas no phosphopeptide products could be detected for sites S1575, S1848, or S1928 on the carboxyl terminus. Other sites at S1829³²and S1142³³ also have been proposed as possible sites of regulation. The present work reopens the issue and provides a new way to test potential sites.

There is also the possibility that phosphorylation of other Ca²⁺ channel-associated proteins could modify channel function. However, in contrast to previous reports,¹⁴ the present findings do not point toward the β subunit as the functionally important site of regulation in adult cardiac cells.

The results have demonstrated the potential for structure/function studies of the α_1C subunit in native cells to complement investigations in heterologous systems. However, the potential limitations of this strategy need to be acknowledged. In this case, overexpressed β₂ subunits may form complexes with native α_1C subunits, or, conversely, overexpressed α_1C may complex with native β subunits. However, although the possibilities for such channel subpopulations pose challenges for the interpretation of studies of molecular regulation, the experimental evidence in our study suggests that the overexpressed subunits dominated the phenotype of L-type Ca²⁺ currents. In the case of the β₂ subunit, this was apparent with the changes in current density and the voltage dependence of activation. With α_1C-subunit overexpression, this was evident with changes in DHP-sensitivity and the β-adrenergic response.

Conclusions

Our findings demonstrate the distal carboxyl terminus beyond amino acid 1905 of the α_1C subunit is required for β-adrenergic enhancement of L-type Ca²⁺ currents in adult cardiomyocytes. However, our results do not support a functional role for phosphorylation of S1928. The present study also demonstrates that structure/function studies of the α_1C subunit of the L-type Ca²⁺ channel are possible using adenoviral gene transfer methods.