Role of the Ca_V1.2 distal carboxy terminus in the regulation of L-type current

ABSTRACT

L-type calcium channels are essential for the excitation-contraction coupling in cardiac muscle. The Ca_V1.2 channel is the most predominant isoform in the ventricle which consists of a multi-subunit membrane complex that includes the Ca_V1.2 pore-forming subunit and auxiliary subunits like Ca_Vα₂δ and Ca_Vβ_2b. The Ca_V1.2 channel’s C-terminus undergoes proteolytic cleavage, and the distal C-terminal domain (DC_termD) associates with the channel core through two domains known as proximal and distal C-terminal regulatory domain (PCRD and DCRD, respectively). The interaction between the DC_termD and the remaining C-terminus reduces the channel activity and modifies voltage- and calcium-dependent inactivation mechanisms, leading to an autoinhibitory effect. In this study, we investigate how the interaction between DCRD and PCRD affects the inactivation processes and Ca_V1.2 activity. We expressed a 14-amino acid peptide miming the DCRD-PCRD interaction sequence in both heterologous systems and cardiomyocytes. Our results show that overexpression of this small peptide can displace the DC_termD and replicate the effects of the entire DC_termD on voltage-dependent inactivation and channel inhibition. However, the effect on calcium-dependent inactivation requires the full DC_termD and is prevented by overexpression of calmodulin. In conclusion, our results suggest that the interaction between DCRD and PCRD is sufficient to bring about the current inhibition and alter the voltage-dependent inactivation, possibly in an allosteric manner. Additionally, our data suggest that the DC_termD competitively modifies the calcium-dependent mechanism. The identified peptide sequence provides a valuable tool for further dissecting the molecular mechanisms that regulate L-type calcium channels’ basal activity in cardiomyocytes.

KEYWORDS:

• Ca²⁺-dependent inactivation
• voltage-dependent inactivation
• L-type calcium current inactivation
• carboxyl terminal

Introduction

L-type calcium channels represent one of mammals’ three voltage-dependent calcium channel families. They include four different isoforms of the Ca_Vα₁ pore-forming subunit (Ca_V1.1 to Ca_V1.4). In the heart ventricle, L-type Ca²⁺-currents are carried by a multi-subunit membrane complex that includes Ca_V1.2 as the pore-forming subunit and the auxiliary subunits Ca_Vα₂δ and Ca_Vβ₂. These channels represent the main Ca²⁺-influx pathway involved in excitation-contraction coupling in cardiac muscle [1]; in fact, Ca²⁺ influx through L-type calcium channels promotes the opening of ryanodine receptors and the subsequent calcium release from the sarcoplasmic reticulum, thus determining the magnitude for cardiac muscle contraction [2]. The activity of Ca_V1.2 in cardiomyocytes is tightly controlled through many mechanisms, among them, the voltage-dependent inactivation (VDI), modulated by intracellular Mg⁺² [3], the calcium-dependent inactivation (CDI), regulated by calmodulin (CaM) [4], and the inhibition exerted by Rad, a monomeric G-protein from the RGK family [5].

Ca_V1.2 comprises four homologous transmembrane domains (I–IV), each with six transmembrane helices (S1–S6) where the S1 to S4 form the voltage sensor domain (VSD) and S5-S6 contribute to the pore domain (PD) structure. The four domains are linked by intracellular loops [6]. The N- and C-terminus are cytoplasmic and are subjected to many post-translational modifications, including the proteolytic cleavage of the C-terminus [7].

It has been observed that even though Ca_V1.2 mRNA predicts a protein of 240 kDa, Western blots of heart tissue show an extra band at 210 kDa as the result of the proteolytic cleavage of the C-terminus [8]. Moreover, Co-IP assays [9] and pull-down experiments [8,10] demonstrated that the distal C-terminal domain (DC_termD) remains non-covalently associated with the channel core through the remaining C-terminus domain [9], at least in heterologous systems.

The interaction of these regions is thought to involve two groups of α-helices (proximal and a distal C-terminal regulatory domain, PCRD, and DCRD, respectively), which bind to each other through two Arg residues located at the channel core and three Glu residues located at the DC_termD. Neutralization of these Arg and Glu residues prevents these interactions and the effect of the DC_termD over truncated channels [3,9]. Alternative binding sites between the DC_termD and the remaining C-terminus have also been described involving the end of the C-terminus and the IQ domain located at the beginning of the C-terminus [10].

Functionally, the interaction between the channel core and its DC_termD wields an autoinhibitory effect, reducing channel activity [8,9]. Overexpression in heterologous systems of a Ca_V1.2 partially truncated at the C-terminus increases the calcium current, independently of the number of channels [11,12] and the co-expression of the truncated region as a separate protein [3,9] or its inclusion into the patch pipette as a recombinant peptide [8,10], sharply reduces the ionic current through the truncated channel.

This interaction also modifies the voltage- and the calcium-dependent inactivation, by mechanisms that is still under debate. There is evidence to suggest that the DC_termD prevents calmodulin binding to the pre-IQ and IQ regions, thus preventing CDI [10]. Furthermore, the interaction between the DC_termD and the channel core modify VDI in an Mg²⁺-dependent manner [3].

In this study, we explore the mechanism by which the interaction of the DC_termD and the remaining C-terminus of Ca_V1.2 modifies the inactivation processes and find that the overexpression of a small peptide that mimics the DCRD-PCRD interaction sequence is sufficient to inhibit the truncated version of Ca_V1.2 channel just as effectively as the entire C-terminal distal region when overexpressed. The expressed peptide displaces the DC_termD from the channel core and reproduces the effect over VDI, but not CDI. Our data suggests that the interaction of PCRD with DCRD is enough to inhibit the L-type channel and regulate VDI in an allosteric manner. However, the modification over CDI seems to be competitive and involves other regions of the DC_termD.

Methods

Constructs

cDNAs for calcium channel subunits (GenBank™ accession numbers: X15539 (Ca_V1.2) and AF286488 (Ca_Vα₂δ₁) were kindly provided by Dr. T. Snutch. Ca_Vβ₂b channel subunit (AF423193) was cloned as described in Moreno et al. [13]. Ca_V1.2-YFP and DC_termD-RLuc were PCR amplified and subcloned into the N1-YFP vector or the N1-RLuc (Clontech Laboratories Inc., Mountain View, CA, USA) between the NheI and BamHI or HindIII and BamHI restriction sites, respectively. The DCRD peptide sequence was subcloned into the pAD-RFP (Agilent, Santa Clara, CA, USA) adenoviral vector between BamHI and XbaI restriction sites.

AD-293 culture and transfection

Tissue culture of AD-293 cells (Agilent) was performed as recommended by the manufacturer. Transfection solutions for individual culture dishes (35-mm diameter) contained a mixture of cDNA expression vectors (500 ng for each L-type calcium channel subunits (Ca_Vα, Ca_Vβ and Ca_Vα₂δ subunits), 500 ng of DC_termD or DCRD peptide and 250 ng of CaM when corresponding and were transfected into cells with lipofectamine 2000 (Life Technologies, Waltham, MA, USA) following manufacturer’s instructions. Experiments were conducted at room temperature 2–3 days after transfection.

Electrophysiological recordings

Membrane currents were recorded at room temperature (20–23 °C) using the Patch-Clamp technique in whole-cell recording mode. Borosilicate glass pipettes (World Precision Instrument, Sarasota, FL, USA) were pulled to have 2–4-MΩ resistance using a P97 horizontal puller (Sutter Instrument, Novato, CA, USA) and filled with internal solution containing (in mM): 120 CsCl, 4 MgCl₂, 10 HEPES, 10 EGTA, 5 MgATP and 0.3 LiGTP (pH 7.2 adjusted with CsOH); the bath solution contained 100 NaCl, 2 CaCl₂, 1 MgCl₂, 10 HEPES, 100 sorbitol (pH 7.4 adjusted with NaOH). After the whole-cell configuration was achieved, the bath solution was changed by gravity perfusion depending on the current under study. For barium currents recording the bath solution contained 90 CsCl, 20 BaCl₂, 1 MgCl₂, 10 HEPES, 10 D-glucose, 40 TEA-Cl (pH 7.4 adjusted with CsOH). For calcium currents the bath solution contained (in mM) 90 CsCl, 10 CaCl₂, 1 MgCl2, 10 HEPES, 20 D-glucose, 40 TEA-Cl (pH 7.4 adjusted with CsOH). Osmolarity was checked for every solution and ranged between 310 and 320 mOsm.

Membrane potentials were sampled at 10 kHz using a 1322a Digidata (Axon Instruments, Foster City, CA, USA). Cell capacitance and series resistance were calculated offline by fitting a single exponential to the capacitive current induced by a 5-mV hyperpolarization step (5 ms) at the end of every pulse. Data acquisition, storage, analysis, current fitting, and offline subtraction were performed using Clampfit 9.2 (Axon Instruments), and all curves were fitted with Prism 8 (GraphPad, Boston, MA, USA).

Bioluminescence resonance energy transfer (BRET)

AD-293 cells were co-transfected with Ca_V1.2-YFP and DC_termD-Rluc fusion proteins (+Ca_Vβ_2b and Ca_Vα₂δ). After 48 hours, cells were divided into a black 96-well plate for YFP fluorescence measurement (EX 485/20, EM 528/20) or a white 96-well plate for BRET assays [14]. Coelenterazine H was added 5 min before light emission measurements, and BRET signal was calculated as the ratio of YFP emission (530 nm) to RLuc emission (460 nm). Every BRET experiment was performed at a controlled temperature of 23°C using a Tecan Infinite F200 Pro fluorometer (Männedorf, Switzerland).

Primary culture of neonatal rat cardiomyocytes

Rats were bred in the Animal Breeding Facility of the Facultad de Medicina, Universidad de Chile (Santiago, Chile). All studies were done in accordance and with the approval of the Institutional Bioethical Committee of Universidad de Chile (CBA 1171 FMUCH) and reported in compliance with the ARRIVE guidelines. Cardiomyocytes were isolated enzymatically from neonatal Sprague-Dawley rats (P0–1), as previously described [15]. Briefly, animals were euthanized, and the hearts were immediately removed and minced in Hank’s solution (in mM): 116 NaCl, 5.4 KCl, 0.8 NaH₂PO₄, 0.8 MgS0₄, 5.6 D-glucose, 20 HEPES (pH 7.4, adjusted with NaOH). The tissue was then digested with pancreatin (1.2 mg/mL, Sigma-Aldrich, St. Louis, MO, USA) and collagenase type-II (0.2 mg/mL, Life Technologies) for 15 min at 37°C under constant agitation. Thereafter, the supernatant was discarded and replaced with fresh digestion solution (15 min at 37°C, by agitation). The supernatant was then withdrawn and centrifuged (5 min at 900×g) and the cells resuspended in 1.5 mL of horse serum. This last step was repeated five times, and the collected cells were resuspended in DMEM (Life Technologies) supplemented with 15% fetal bovine serum (FBS) and 1% antibiotic (Penicillin/Streptomycin), seeded on a culture dish, and incubated for 2 h at 37°C in a 95% O₂–5% CO₂ atmosphere to allow fibroblast adhesion. Non-adherent cells (cardiomyocytes) were recovered from the supernatant and centrifuged (5 min at 900×g), to be seeded onto gelatin-coated culture plates (1%). Sixteen hours later, the medium was replaced with DMEM 0.5% FBS.

Adenovirus production

Recombinant adenoviral particles were generated using the AdEasy system (Agilent). Briefly, cDNAs of the DCRD peptide or the DC_termD fused to CFP were subcloned into the commercial adenoviral vector pAD/RFP or pShuttle-CMV, and homologous recombination was done by PmeI linearized DNA transformation of BJ5183-AD-1 cells (Agilent). According to manufacturer guidelines, recombinant adenoviral plasmids were digested with PacI and transfected in AD293 cells with Lipofectamine 2000. Following observation of cytopathic effects (CPEs; 14–21 days), cells were collected and subjected to three freeze – thaw cycles in a dry-ice methanol bath to purify the recombinant viruses [16]. The resulting supernatant infects a 10-cm dish of 90% confluent AD293 cells. Following observation of CPEs after 2–3 days, viral particles were purified and expanded by infecting 10 plates of AD293 cells.

Cardiomyocyte infection

The number of viral particles for each adenovirus was determined by 260-nm absorbance [17] and was in the order of 10¹¹ particles per milliliter; the effective virus titer was determined at 90% infection efficiency (RFP fluorescence) in the absence of cytopathic effects. Infection was performed in the moment of seeding, in gelatin pre-treated plates by adding 10⁹ particles of the adenovirus per dish. After 16 h, the medium was replaced with DMEM media with 0.5% FBS. Cells were used 48 h after infection.

Statistics

All values are reported as mean ± SEM (n). Statistical analysis of the data was performed with Prism 8 using the one-way ANOVA or t-test and was considered significant at p < 0.05.

Results

We first aimed to standardize the effect of the DC_termD over the Ca_V1.2 channel and its inactivation processes in our experimental setting. To achieve this, we overexpress the full-length channel as a control or a truncated version of Ca_V1.2 at aa 1820 (Ca_V1.2^Δ1820) in AD293 cells with auxiliary Ca_Vβ_2b and Ca_Vα₂δ₁ subunits. In agreement with previous work and as is shown in Figure 1, overexpressing the DC_termD fused to CFP reduces the barium current through the truncated Ca_V1.2 channel; this reduction achieves a 63 ± 8% of the maximal current (Figure 1(a–c)). Additionally, an approximately 10 mV rightward shift in the activation curve is observed upon DC_termD overexpression (Figure 1d), as determined by fitting the IV curves to a modified Boltzmann function. In contrast, no effect over the maximal current nor the voltage-dependence is observed when Ca_V1.2^Δ1820 is replaced by the full-length version of the channel (Supplementary Figure S1).

Figure 1. DC_termD effect over Ca_V1.2^Δ1820: (a) Representative whole-cell L-type Ba²⁺ current traces from AD293 cells overexpressing the truncated version of Ca_V1.2 channel (Ca_V1.2^Δ1820) with or without the DC_termD. Currents elicited by a voltage step protocol from − 60 to +50 mV in 10-mV increments, V_h = −80 mV. (b) Summary peak current I/V plots (mean ± SEM) obtained from currents family as shown in (A); black lines represent the best fit to a Boltzmann equation. c) Graph shows the mean ± SEM of the peak Ba²⁺ current density. d) Graph shows the mean ± SEM of the midpoint of activation. In every panel, control cells are represented with filled circles and cells co-expressing the DC_termD with empty circles. *p < 0.05 with respect to control, n = 8 for control and n = 7 for DC_termD.

Description of the effect of DCtermD on CaV1.2Δ1820: A) Visual representation of whole-cell L-type Ba2+ current traces from AD293 cells overexpressing CaV1.2Δ1820 with or without the DCtermD. B) Summary peak current I/V plots (mean ± SEM) derived from the current traces shown in (A). Black lines indicate the best fit to a Boltzmann equation. C) Graph illustrating the mean ± SEM of the peak Ba2+ current density. D) Graph presenting the mean ± SEM of the midpoint of activation. In each panel, filled circles denote cells overexpressing the truncated version of CaV1.2 channel, while empty circles represent cells co-expressing DCtermD and CaV1.2Δ1820. *P < 0.05 compared to control.

Next, the impact of DC_termD overexpression on L-type channel inactivation was explored. To evaluate the kinetics of voltage-dependent inactivation of L-type channel current, we measured the residual barium current after 200-ms of a 250-ms pulse (R₂₀₀) at various voltage steps and plotted it against the command potential. As shown in Figure 2, when Ca_V1.2^Δ1820 is expressed, the residual barium current percentage after 200 ms is higher in the −10 to +20 mV range, indicating that VDI is impaired at this voltage range. Nevertheless, no apparent change in the voltage dependence of this inactivation is observed (Figure 2a), as determined by fitting a single exponential decay to the data (τ_ctrl = 0.053 ± 0.011 and τ_CtermD = 0.040 ± 0.08, p = 0.32; Figure 2(a)). In addition, the difference of R₂₀₀ values between barium and calcium as charge carriers taken at 0 mV was used to quantify the strength of Ca²⁺-dependent inactivation (Figure 2(b,c)). Despite the weak CDI recorded, the CDI fraction was further reduced by the presence of DC_termD when the truncated version of Ca_V1.2 was expressed (0.11 ± 0.04 vs 0.03 ± 0.01, respectively). No significant difference in the VDI nor the CDI was observed between the full-length channel expressed with or without DC_termD (Supplementary Figure S2).

Figure 2. DC_termD modifies VDI and CDI of Ca_V1.2^Δ1820: (a) Graph showing the residual of current after 200 ms during a 250-ms depolarization pulse (R₂₀₀, mean ± SEM, n = 8 for control and n = 7 for DC_termD) versus command voltage of L-type Ba²⁺ currents AD293 cells overexpressing the Ca_V1.2^Δ1820 channel with (empty circles) or without (filled circles) the DC_termD. Slower inactivation rates result in higher R₂₀₀ values. (b) Representative trace of the I_Ba (black line) and I_Ca evoked by a 0 mV depolarizing pulse. Both currents were scaled to the same amplitude for comparison. (c) Graph shows the mean ± SEM of CDI fraction (n = 5 for control and n = 6 for DC_termD). *p < 0.05 with respect to control.

With these results, we validated previous observations about the impact of the interaction between the proximal and distal regions of the C-terminus of Ca_V1.2. As a next step, we decided to investigate whether the entire DC_termD is necessary to produce the previously described effects. For this, we cloned the sequence corresponding only to the interaction domain, which consists of 14 amino acids (LTIEEMENAADDNIL) and expressed this small peptide mimicking the DCRD (DCRD motif) in cells overexpressing Ca_V1.2^Δ1800 with the respective auxiliary subunits.

First, to demonstrate that the peptide codified by the DCRD motif interacts with the Ca_V1.2 channel, we analyzed its ability to disrupt the interaction between DC_termD and the channel core. For this, we fused the Renilla luciferase (RLuc) with the DC_termD to use it as an energy donor, and Ca_V1.2^Δ1820 fused to YFP at its N-terminus as the acceptor. After transfecting these proteins along with the auxiliary subunits, we determined their interaction by the bioluminescence energy transfer (BRET) assay measuring the light intensity at 535 and 475 nm and calculating the net BRET value (BRET signal minus background BRET from DC_termD-Rluc alone).

The net BRET value was plotted against the YFP signal, resulting in a saturation curve that indicates the expected interaction between the two proteins (Figure 3). When replacing the donor with the DC_termD alone, the BRET signal was absent. Additionally, including the DCRD motif during transfection significantly reduced the net BRET signal (Figure 3), indicating that the overexpression of this motif disrupts the DC_termD-Ca_V1.2 interaction, likely by competing for the binding domain.

Figure 3. DCRD peptide displaces DC_termD from Ca_V1.2^Δ1820 BRET titration curve from AD293 cells co-transfected with increasing amounts of Ca_V1.2^Δ1820-YFP with (gray circles) or without (black circles) the DCRD peptide coding sequence, and a constant amount of DC_termD-RLuc as the energy donor. White circles denote experiments were the DC_termD without the RLuc was used. Saturation of the BRET curve indicates specific interaction between the DC_termD and the channel pore, whereas the low linear BRET signal denoted absence of interaction. Dashed lines represent the best fit to a hyperbolic curve or a linear fit respectively. Data from five experiments are included in the graph.

Analysis of DCRD peptide impact on DCtermD displacement from CaV1.2Δ1820 Graph showing the net BRET versus the ratio of YFP and Cltz emitted light form cells overexpressing CaV1.2Δ1820-YFP and the DCtermD with or without the DRCD peptide. A saturation curve is observed only in the absence of the DRCD peptide.

Afterward, we explore the impact of DCRD motif expression on barium currents in a heterologous system by co-expressing the DCRD cDNA with Ca_V1.2^Δ1820 and the corresponding auxiliary subunits. Remarkably, the overexpression of the DCRD motif led to a reduction in barium currents to almost the same extent as when the entire DC_termD is expressed (55 ± 8% of the maximal current). Likewise, the voltage dependence curve of the activation was shifted to the right at approximately 8 mV (Figure 4d). The data, including the impact of DC_termD, was subjected to a one-way ANOVA test, which revealed that the DCRD motif alters Ca_V1.2^Δ1820 barium currents to the same extent as the entire DC_termD.

Figure 4. Effect of the DCRD peptide over Ca_V1.2^Δ1820. (a) Representative whole-cell L-type Ba²⁺ current traces from AD293 cells overexpressing the Ca_V1.2^Δ1820 channel with or without the DCRD peptide. Currents elicited by a voltage step protocol from − 60 to +50 mV in 10-mV increments, V_h = −80 mV. (b) Summary peak current I/V plots (mean ± SEM) obtained from currents family as shown in (A); black lines represent the best fit to a Boltzmann equation. (c) Graph shows the mean ± SEM of the peak Ba²⁺ current density. (d) Graph shows the mean ± SEM of the midpoint of activation. In every panel, control cells are represented with filled circles and cells co-expressing the DCRD peptide with empty circles. Control data set is the same used in Figure 1. *p < 0.05 with respect to control, n = 8 for control and n = 7 for DCRD peptide.

Description of the effect of DCRD peptide overexpression on CaV1.2Δ1820: A) Visual representation of whole-cell L-type Ba2+ current traces from AD293 cells. B) Summary peak current I/V plots (mean ± SEM) derived from the current traces shown in (A). Black lines indicate the best fit to a Boltzmann equation. C) Graph illustrating the mean ± SEM of the peak Ba2+ current density. D) Graph presenting the mean ± SEM of the midpoint of activation. In each panel, filled circles denote cells overexpressing the truncated version of CaV1.2 channel, while empty circles represent cells co-expressing DCtermD and CaV1.2Δ1820. Control data set is the same used in Figure 1. *P < 0.05 compared to control.

We next explored how the DCRD motif affects the inactivation processes of the L-type channel and compared its effect with that previously observed with DC_termD. As seen in Figure 5, unlike the DC_termD effect (Figure 2(b,c)), overexpression of the DCRD motif does not reduce the CDI (Figure 5(a,b)). Therefore, to gain a deeper understanding of the modulation mechanism of CDI by DC_termD, which is thought to compete with CaM [10], and the lack of modulation by the DCRD motif, we performed experiments with and without CaM overexpression.

Figure 5. No effect of DCRD peptide over Ca_V1.2^Δ1820 calcium-dependent inactivation: (a) and (c) Representative trace of Ca_V1.2^Δ1820 barium current (black line) and calcium current evoked by a 0 mV depolarizing pulses from control AD293 cells (a) or AD293 cells overexpressing calmodulin (c). I_Ba and I_Ca currents were scaled to the same amplitude for comparison. (b) and (d) Graph shows the mean ± SEM of CDI fraction from AD293 control cells (n = 5) (b) or AD293 cells overexpressing calmodulin (n = 8 for control, n = 6 for DC_termD, and n = 6 for DCRD peptide). The control data set of panel B is the same used in Figure 2c. *p < 0.05 with respect to control.

Impact of DCRD peptide on CaV1.2Δ1820 voltage-dependent inactivation: A) Graph depicting the residual current after 200 ms during a 250-ms depolarization pulse (R200) versus command voltage of L-type Ba2+ currents. B) and C) Graphs displaying the mean ± SEM of R200 in cells overexpressing CaM (panel C) or with endogenous CaM (panel B). Data from cells overexpressing CaV1.2Δ1820 alone or with the DCtermD, or the DCRD peptide is shown. The control dataset of panel A and B is identical to that used in Figure 2. *P < 0.05 compared to control.

As expected, the fraction of CDI was increased almost three times when CaM is co-expressed with the Ca_V1.2^Δ1800 (0.11 ± 0.04 to 0.32 ± 0.05), but importantly, CaM overexpression impedes the reduction of CDI observed upon DC_termD overexpression (compare Figure 2(b,c) with Figure 5(c,d)). Likewise, the DCRD motif does not modify the fraction of inactivation that depends on calcium when the truncated version of Ca_V1.2 is co-expressed with CaM (Figure 5(c,d)).

On the other hand, similar to the impact over VDI observed when the entire DC_termD is expressed, co-expression of the DCRD motif increments the residual barium current percentage after 200 ms in the range measured, indicating that VDI is reduced at this voltage range (Figure 6a). Additionally, as seen with the entire DC_termD (Figure 2), no significant change was observed in the voltage dependence of VDI when fitted to a single exponential decay function (τ_ctrl = 0.053 ± 0.011 and τ_DCRD = 0.034 ± 0.09, p = 0.17; Figure 6a).

Figure 6. The DCRD peptide modify Ca_V1.2^Δ1820 voltage-dependent inactivation: (a) Graph showing the residual of current after a 200-ms during a 250 ms depolarization pulse (R₂₀₀, mean ± SEM. n = 8 for control and n = 7 for DCRD peptide) versus command voltage of L-type Ba²⁺ currents from AD293 cells overexpressing the Ca_V1.2^Δ1820 channel with (empty circles) or without (filled circles) the DCRD peptide. Slower inactivation rates result in higher R₂₀₀ values. (b) Graph shows the mean ± SEM of R₂₀₀ from AD293 cells overexpressing the Ca_V1.2^Δ1820 channel alone (n = 5), or with the DC_termD (n = 6) or the DCRD peptide (n = 5). (c) Graph shows the mean ± SEM of R₂₀₀ from AD293 cells overexpressing calmodulin and the Ca_V1.2^Δ1820 channel alone (n = 8), or with the DC_termD (n = 7), or the DCRD peptide (n = 6). The control data set of panel a and B is the same used in Figure 2. *p < 0.05 with respect to control.

Impact of DCRD peptide on CaV1.2Δ1820 voltage-dependent inactivation: A) Graph depicting the residual current after 200 ms during a 250-ms depolarization pulse (R200) versus command voltage of L-type Ba2+ currents. B) and C) Graphs displaying the mean ± SEM of R200 in cells overexpressing CaM (panel C) or with endogenous CaM (panel B). Data from cells overexpressing CaV1.2Δ1820 alone or with the DCtermD, or the DCRD peptide is shown. The control dataset of panel A and B is identical to that used in Figure 2. *P*p <  compared to control.

Interestingly, co-expression of CaM with Ca_V1.2^Δ1800 increases the fraction of residual barium current evoked by a depolarization pulse at 0 mV after 200 ms from 0.32 ± 0.03 to 0.58 ± 0.05 indicating a reduction of VDI (Figure 6(b,c)). Nevertheless, even with this increase in R₂₀₀, the effect of the DC_termD and the DCRD motif observed in the absence of CaM is preserved in this condition, with a further reduction of VDI (Figure 6b), indicating that, although calmodulin is able to modulate VDI, the effect of DCRD-PCRD interaction is independent.

Finally, to explore the effect of the overexpression of the DCRD motif in a more physiological setting, we prepare recombinant adenovirus particles carrying the coding sequence of the DC_termD fused to CFP, or the coding sequence of the DCRD motif subcloned in an adenoviral vector with the RFP reporter protein encoded by an independent CMV promoter and infect cardiomyocytes from newborn rats; barium currents were recorded only in bright fluorescent cardiomyocytes (CFP for DC_termD and RFP for DCRD motif) and the effect over peak current, VDI and CDI was determined.

The overexpression of the DCRD motif did not produce any statistically significant change (p = 0.35) in the maximal barium current at 0 mV or in the R₂₀₀, as shown in Figure 7(a,b), respectively. Likewise, the overexpression of the DCRD motif did not affect CDI. To investigate whether the lack of effect was due to the low efficiency of the coded peptide over endogenous L-type channel, the entire DC_termD fused to CFP was overexpressed. Cardiomyocytes that have been transduced with the DC_termD exhibit similar characteristics to those that have been transduced with the DCRD motif. No changes on the barium current at 0 mV, or in the residual current after 200 ms at that voltage were registered. Additionally, no change in the CDI fraction was observed (Figure 7(c,d)).

Figure 7. DC_termD and DCRD peptide effect over endogenous L-type currents in cardiomyocytes (a) Graph shows the mean ± SEM of the peak endogenous L-type Ca²⁺ current density from newborn rat cardiomyocytes (n = 10 for control, n = 7 for DC_termD, and n = 7 for DCRD peptide). (b) Graph shows the mean ± SEM of R₂₀₀ newborn rat cardiomyocytes infected with RFP as control, or with the DC_termD-CFP, or the DCRD peptide coding sequence (n = 10 for control, n = 7 for DC_termD, and n = 7 for DCRD peptide). (c) Representative trace of the I_Ba (black line) and I_Ca evoked by a 0-mV depolarizing pulse. Both currents were scaled to the same amplitude for comparison. (d) Graph shows the mean ± SEM of CDI fraction (n = 7). *p < 0.05 with respect to control.

Impact of DCtermD and DCRD peptide on endogenous L-type currents in cardiomyocytes: A) Graph displaying the mean ± SEM of peak endogenous L-type Ca2+ current density from newborn rat cardiomyocytes. B) Graph depicting the residual L-type current after 200 ms during a 250-ms depolarization pulse (R200) from newborn rat cardiomyocytes. C) Illustrative trace of IBa (black line) and ICa evoked by a 0-mV depolarizing pulse. Both currents were adjusted to the same amplitude for comparison. D) Graph showing the mean ± SEM of calcium-dependent inactivation (CDI) fraction. Data of cardiomyocytes overexpressing RFP (as control), the DCtermD or the coding sequence of the DCRD peptide is shown. *p < 0.05 compared to control.

Discussion

A singular feature of L-type calcium channels is the length of their C-terminal domain which involves almost 1/3 of the protein, suggesting that this region plays a crucial role in LTCCs regulation. In Ca_V1.4, the distal C-terminus harbors a domain that serves to abolish CDI (ICDI, inhibitor of CDI) by binding to the EF-hand motif in the proximal C-terminal, uncouples the molecular machinery responsible for CDI without preventing CaM binding [18–20]. In contrast, in Ca_V1.3, a C-terminal automodulatory domain (CTM), formed by the interaction of two α-helical domains has been described to compete with CaM and modulate, but not abolish, the CDI [21–23].

Less information is available on how the proteolytically processed C-terminus of Ca_V1.2 modifies the channel activity, where most of the studies have focused on the relationship between the autoinhibitory effect imposed by the DC_termD and the β-adrenergic response of this channel [9,24–27]. However, recent findings have challenged this model by uncovering that PKA-dependent phosphorylation relieves the inhibitory effect caused by the interaction between the Ca_Vβ subunit and Rad, a monomeric G-protein from the RGK family [28–31]. Nevertheless, the inhibitory effect of DC_termD may remain a relevant mechanism for setting up the low basal activity of endogenous L-type calcium channels in cardiomyocytes.

In this study, we have shown that the heterologous expression of the cDNA that encodes a small peptide that mimics the putative sequence responsible for the interaction between the DC_termD and the channel pore, which is named the distal C-terminal regulatory domain (DCRD), can reduce barium currents in cells that express a version of Ca_V1.2 channel truncated at aa 1820 (Figure 4) at the same extent as the entire DC_termD, including the rightward shift of channel voltage-dependence (Figure 1). This result discards the possibility that the occlusion of the pore by a portion of the DC_termD, similar to the ball and chain, is the mechanism for channel inhibition.

For the Ca_V1.2 channel to open, the pore-lining S6 helices that obstruct ion permeation in the closed state must move, diverging from each other and allowing ions to pass [32]. Therefore, our findings suggest that the PCRD-DCRD interaction could modulate channel gating by modifying the remaining C-terminus flexibility and restricting IVS6 helix movement. However, further corroboration of this idea is necessary.

Regarding the inactivation processes, previous research has demonstrated that CaM inhibits the interaction between the C-terminal and DC_termD [10]; also, overexpression of CaM prevents the current inhibition induced by DC_termD and CaV1.2^Δ1801 co-expression [33], indicating that CaM and DC_termD compete with each other to modulate the CDI fraction. This competition is supposed to be achieved through the interaction between the pre-IQ and IQ regions located on the proximal C-terminus, and a region on the DC_termD distal to the DCRD domain (aa 2116–2169 of guinea pig clone) known as the “CaM-competitive domain” (CCD) [34], in fact, the inclusion of a peptide that lacks the DRCD motif but includes the CCD region is able to inhibit the L-type current from guinea pig ventricular myocytes supposedly by competing with CaM for the CaM-binding domains [10].

Our findings provide evidence to support this idea. We show here that the increased expression of DC_termD hinders CDI when the truncated version of the channel is expressed (Figure 2) and that the overexpression of the DCRD motif, which lacks the putative CCD, does not affect the CDI (Figure 5), thus supporting the idea of the existence of independent regulatory modules in the DC_termD.

Additionally, CDI modulation by the DC_termD is not observed when CaM is co-expressed (Figure 5). The increase in the CDI fraction of Ca_V1.2^Δ1820 of almost three times upon CaM overexpression (Figure 5) suggests that, in the basal condition, not every channel is bound to CaM, which could be due to a low CaM expression level in this cell line. Thus, overexpression of CaM would displace the CCD fragment, preventing the CDI regulation by the DC_termD. However, the maintenance of VDI regulation, even with CaM overexpression, indicates that the DC_termD remains bound to the channel core.

Concerning the effect of the DC_termD interaction with the channel pore on the voltage-dependent inactivation process, we demonstrate three important effects here. Firstly, the overexpression of DC_termD results in a slowdown of barium current, as shown in Figures 1 and 2. Secondly, the co-expression of CaM leads to an increase in the residual barium current, as depicted in Figure 5. And lastly, overexpression of the DCRD motif reproduces the effect of DC_termD, as illustrated in Figure 5. Our results differ from previously published data in terms that it was shown that the DC_termD enhances VDI [3], and we show a diminished VDI induced by the DC_termD or DCRD motif co-expression; however, since the binding of Ca_Vβ subunit has a significant impact on VDI [2] kinetics, it is possible that the choice of Ca_Vβ subunit used could be the reason for this difference.

On the other hand, it has been shown that the molecular determinants of VDI include multiple regions of Ca_V1.2: the cytosolic end of the S6 segments, the I-II linker, the amino-terminal, and the EF-motif located at the beginning of the C-terminus [35,36] which is consistent with our data that CaM modulate VDI (Figure 5). Moreover, it has been suggested that DC_termD-dependent VDI modulation requires the EF-motif as a downstream structural element [3], in this sense, we demonstrated here that overexpression of the DCRD motif is enough to modify the VDI, suggesting that the sole interaction between PCRD and DCRD allosterically induces the necessary conformational changes that are propagated to the DIVS6 segment.

The total endogenous L-type current is expected to be a mixture of at least three types of Ca_V1.2 channels: the full-length and truncated channels with or without the DC_termD attached. Overexpression of the DC_termD in developing mouse cardiomyocytes has been shown to reduce L-type barium currents and lower diastolic calcium [33], likewise, addition of peptides derived from the DC_termD to inside-out patches from guinea pig ventricular myocytes modifies L-type channel activity [10]. Since the overexpression of the DC_termD does not modify the full-length channel (Figures 1 and 2), this indicates that truncated channels not bound to their DC_termD make up a significant proportion of the total population of L-type channels in this model.

In contrast, transgenic mice expressing a proteolytic-resistant Ca_Vα_1C subunit showed that preventing C-terminal cleavage did not modify the L-type calcium current or the contractibility of cardiomyocytes from adult mice while preserving the β-adrenergic response [37]. Similarly, when we overexpressed either the DC_termD or the DCRD motif in cardiomyocytes from rat newborns (Figure 7), we did not observe a significant effect, either in the maximal current or the VDI fraction., results that are consistent with the small proportion of truncated channels detected in rat ventricles [38] and rabbit hearts [39].

Altogether, the data suggests that the amount of truncated Ca_V1.2 varies across species, developmental stages, or cell types. Given the experimental difficulties for biochemically establish the proportion of truncated channels in different preparations [40,41], a functional approach may help to shed light about the physiological role of this posttranslational modification in the regulation of the amount of basal L-type current. While the overexpression of the full DC_termD could help determine the amount of truncated Ca_V1.2 channel unbound to the channel core, it does not differentiate if a truncated channel is bound to its DC_termD. Conversely, a DCRD cell-penetrating peptide, based on the DRCD motif described in this study, would not only modify truncated channels unbound to the DC_termD but also displace any proteolyzed region tethered, offering the possibility of functionally determining the importance of truncated Ca_V1.2 channels in different cell types.

Author Contributions

Conceptualization and design, DV; Experiments, FA, DDG, FM, and TH. Data Analysis, FA, TH and DV; writing and editing, TH, FS and DV. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

The authors declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability statement

The data that support the findings of this study are available from the corresponding author, DV, upon reasonable request.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19336950.2024.2338782.

Additional information

Funding

This work was supported by a research grant from Fondo Nacional de Desarrollo Científico y Tecnológico (FONDECYT; grant 1210881 to D. Varela). The Millennium Nucleus of Ion Channels- Associated Diseases (MiNICAD) (NCN19_168) is a Millennium Nucleus supported by the Iniciativa Científica Milenio of the Ministry of Economy, Development and Tourism (Chile). Millennium Science Initiative Program – ICN09_016/ICN 2021_045: Millennium Institute on Immunology and Immunotherapy (ICN09_016/ICN 2021_045; former P09/016-F).