Functional properties of a newly identified C-terminal splice variant of Cav1.3 L-type Ca2+ channels

Abstract

An intramolecular interaction between a distal (DCRD) and a proximal regulatory domain (PCRD) within the C terminus of long Ca(v)1.3 L-type Ca(2+) channels (Ca(v)1.3(L)) is a major determinant of their voltage- and Ca(2+)-dependent gating kinetics. Removal of these regulatory domains by alternative splicing generates Ca(v)1.3(42A) channels that activate at a more negative voltage range and exhibit more pronounced Ca(2+)-dependent inactivation. Here we describe the discovery of a novel short splice variant (Ca(v)1.3(43S)) that is expressed at high levels in the brain but not in the heart. It lacks the DCRD but, in contrast to Ca(v)1.3(42A), still contains PCRD. When expressed together with α2δ1 and β3 subunits in tsA-201 cells, Ca(v)1.3(43S) also activated at more negative voltages like Ca(v)1.3(42A) but Ca(2+)-dependent inactivation was less pronounced. Single channel recordings revealed much higher channel open probabilities for both short splice variants as compared with Ca(v)1.3(L). The presence of the proximal C terminus in Ca(v)1.3(43S) channels preserved their modulation by distal C terminus-containing Ca(v)1.3- and Ca(v)1.2-derived C-terminal peptides. Removal of the C-terminal modulation by alternative splicing also induced a faster decay of Ca(2+) influx during electrical activities mimicking trains of neuronal action potentials. Our findings extend the spectrum of functionally diverse Ca(v)1.3 L-type channels produced by tissue-specific alternative splicing. This diversity may help to fine tune Ca(2+) channel signaling and, in the case of short variants lacking a functional C-terminal modulation, prevent excessive Ca(2+) accumulation during burst firing in neurons. This may be especially important in neurons that are affected by Ca(2+)-induced neurodegenerative processes.

FIGURE 1.

C-terminal regulatory domains and quantitative PCR using TaqMan gene expression assays. a, schematic presentation of alternative splicing in the C terminus of Ca_v1.3 channels generating multiple Ca_v1.3 α1 isoforms with different C-terminal lengths. Constitutive and alternative exons are shown as black and white squares, respectively. The schematic protein domain structure is always shown below. Ca_v1.3_L, expression of exons 39 to 49 (with/without exon 44 (8)) yields the full-length Ca_v1.3 C terminus. Ca_v1.3_43S, use of an alternative 3′ splice acceptor site in exon 43 yields a C-terminal truncated Ca_v1.3 splice variant that terminates in the amino acid sequence, KQEIRGVITIITIIP. Ca_v1.3_Δ41–1, Ca_v1.3_Δ41–1 channels arise from skipping exon 41 and combining with exon 42 (terminating with LDQVVPPAGGGIKDTA). Ca_v1.3_Δ41–2 combined with exon 42A (terminating with LDQVVPPAGDA). b, relative comparison of transcripts containing exon 42 (42, black), exon 42A (42A, dark gray), and exon 49 (49, light gray) in brain and heart regions of WT c57Bl6/N mice. Expression in percent (mean ± S.E.) relative to the sum of 42 and 42A transcripts was detected in 5 whole brains (WB), 20 ng (all others) of RNA equivalent from individuals (WB) or tissue pools (all others; 10 animals) are indicated. The number of experiments is given in parentheses. Statistical analysis was performed using one-way ANOVA followed by Bonferroni post hoc testing; *, p < 0.05; **, p < 0.01; ***, p < 0.001, 42 versus 49. VTA, ventral tegmental area; Cx, cortex; Ce, cerebellum; PFC, prefrontal cortex; OLB, olfactory bulb; HPC, hippocampus; AMY, amygdala.

FIGURE 2.

Detection by PCR of the C-terminal alternative splicing in human and mouse Ca_v1.3 transcripts. a, fragments containing exon 43S (502 bp) or 43l (656 bp) generated by primers specific for exon 39 (forward) and 43 (reverse) of human Ca_v1.3. WB, 500 ng of whole brain cDNA (positive control); H₂O, no template (negative control). One representative of 4 experiments is shown. b, fragments containing 43S (403 bp) or 43L (557 bp) were generated by primers specific for exons 42 (forward) and 45 (reverse) of mouse Ca_v1.3. Each sample contained 20 ng of RNA equivalent. VTA, ventral tegmental area; CP, caudate putamen; AMY, amygdale; OLB, olfactory bulb; HPC, hippocampus; Ce, cerebellum; FC, frontal cortex; PFC, prefrontal cortex. One representative of 3 experiments is shown. c, conditions are as described in b. WH, whole heart; Vtr, ventricle; rA, right atrium; lA, left atrium; Eye, whole eye preparation; Mock tsA, untransfected tsA-201 cells; tsA Ca_V1.3_L, tsA-201 cells transiently expressing mouse Ca_v1.3_L; all samples were tested for GAPDH expression and contamination. GAP, PCR with primers specific for human GAPDH. One representative of 3 experiments is shown. d, fragments containing exons 41 (39–42, 331 bp; 39–42A, 488 bp), Δ41–1 (39–42, 200 bp), or Δ41–2 (39–42A, 357 bp) generated by primers specific for exon 39 and exons 42 or 42A, respectively, of human Ca_v1.3. 1 μg of whole brain cDNA was used. One representative of 6 independent experiments is shown.

FIGURE 3.

Biophysical properties of C-terminal Ca_v1.3 α1 splice variants with 15 mm Ca²⁺ as charge carrier. a, mean normalized current-voltage (I-V) curves recorded in tsA-201 cells expressing Ca_v1.3_L (black), Ca_v1.3_43S (open circle), and Ca_v1.3_42A (gray). Activation parameters and statistics are given in Table 1. b, steady-state inactivation curves for Ca_v1.3_L, Ca_v1.3_43S. Parameters for the activation curve were obtained from parameters in a. Inactivation parameters were as follows in mV: V_0.5, Ca_v1.3_L, −25.6 ± 0.52; Ca_v1.3_43S, −30.9 ± 0.7, p < 0.001; k (slope): Ca_v1.3_L, −5.5 ± 0.26_, Ca_v1.3_43S, −4.2 ± 0.2, p = 0.027, Mann-Whitney test. c, voltage dependence of CDI for Ca_v1.3_L (left) and Ca_v1.3_43S (right): r₂₅₀ corresponds to the fraction of I_Ca or I_Ba remaining after 250 ms; f is the difference between r₂₅₀ values at 11.5 mV. d, left, normalized peak current traces of Ca_v1.3_L, Ca_v1.3_43S, and Ca_v1.3_42A evoked by 5-s depolarization to V_max. Right, percent I_Ca inactivation during 0.1-, 0.25-, 0.5-, 1-, and 5-s test pulses to V_max. Color code as described in a. Number of experiments is given in parentheses. Error bars reflect S.E. *, p < 0.05; ***, p < 0.001, one-way ANOVA analysis followed by Bonferroni post test.

FIGURE 4.

Differential coupling of ON-gating current to the opening of Ca_v1.3 splice variant α1-subunits at the reversal potential. a, ON-gating currents were measured at potentials at which no ionic inward and outward current was observed (V_rev). V_rev was determined individually in each cell by 20-ms pulses to voltages between +70 and +90 mV in 2-mV increments. Tail current was elicited during repolarization to −50 mV as indicated in the step protocol above. Representative currents are shown below (cells: Ca_v1.3_L, 091009_62; Ca_v1.3_43S, 081009_96; Ca_v1.3_42A, 151009_15). b, bar graphs show ON-gating current (Q_ON) for Ca_v1.3_L (black), Ca_v1.3_43S (gray), and Ca_v1.3_42A (white). Error bars reflect S.E. ***, p < 0.001 (Mann-Whitney test). Number of experiments is given in parenthesis. c, correlation of Q_ON to maximal tail current amplitude at V_rev. Color code according to b. Calculated slopes were: −0.0067 ± 0.0008 for Ca_v1.3_L, −0.0403 ± 0.0005 for Ca_v1.3_43S, and −0.0705 ± 0.012 for Ca_v1.3_42A, respectively.

FIGURE 5.

Single channel properties of Ca_v1.3 C-terminal α1 subunit splice variants. a–c, representative single channel recordings for Ca_v1.3_L, Ca_v1.3_42A, and Ca_v1.3_43S were obtained during step depolarization to voltage ranging from −30 to 0 mV from the holding potential of −100 mV with 15 mm Ba²⁺ as the charge carrier. The depicted single channel currents were elicited by 150-ms pulses to the indicated potentials and applied every 600 ms. Ten representative consecutive traces (of at least 300 recorded traces per experiment) are depicted (cells: Ca_v1.3_L, C9625; Ca_v1.3_43S, J0427; Ca_v1.3_42A, I9604). Below the corresponding ensemble average current is presented. d, open probability (P_open) within active sweeps. e, fraction of active sweeps containing at least one channel opening. One-way ANOVA followed by Dunnett's post test was performed among all test potentials. Significance levels were: *, p < 0.05; **, p < 0.01; and ***, p < 0.001 for comparing with Ca_v1.3_L; +, p < 0.05 and **, p < 0.01 for comparison with Ca_v1.3_43S. Data are presented as mean ± S.E. f, current-voltage relationship of unitary currents of Ca_v1.3_L (black, n = 6), Ca_v1.3_43S (light gray, n = 8), and Ca_v1.3_42A (gray, n = 6). Solid line represents the best-fit curve to data obtained from all-point histograms. Despite minor differences in unitary current level at −10 mV, single channel conductance was similar for all three isoforms (in pS: Ca_v1.3_L, 16.10 ± 1.97, n = 5; Ca_v1.3_43S, 15.94 ± 2.19, n = 7; and Ca_v1.3_42A, 14.86 ± 0.9, n = 6).

FIGURE 6.

Biophysical properties of C-terminal Ca_v1.3 α1 splice variants with 2 mm Ca²⁺ as charge carrier. a, current activation properties shown in representative of normalized I-V curves recorded in tsA-201 cells expressing Ca_v1.3_L (black), Ca_v1.3_43S (gray), and Ca_v1.3_42A (white). Activation parameters and statistics are given in Table 2. b, voltage dependence of inactivation elicited after 5-s conditioning prepulses using 20-ms test pulses to V_max. Inactivation parameters are given in mV: V_0.5,inact: Ca_v1.3_L, −43.4 ± 1.1, Ca_v1.3_43S, −43.7 ± 0.4, Ca_v1.3_42A, −46.4 ± 0.34; k (slope): Ca_v1.3_L, −5.1 ± 0.3_, Ca_v1.3_43S, −4.7 ± 0.15, Ca_v1.3_42A, −4.5 ± 0.3. c, percent I_Ca inactivation during 0.1-, 0.25-, 0.5-, 1-, and 5-s test pulses to V_max. Color code as described in a. d, voltage dependence of CDI: r₂₅₀ corresponds to the fraction of I_Ca or I_Ba remaining after 250 ms; f is the difference in r₂₅₀ of I_Ba and I_Ca at −19 mV. Number of experiments are given in parentheses. Error bars reflect S.E. *, p < 0.05; **, p < 0.01; ***, p < 0.001, one-way ANOVA followed by Bonferroni post test.

FIGURE 7.

Changes in Ca_V1.3_43S channel gating properties in the presence of distal Ca_V1.3 or Ca_V1.2 C-terminal peptides. a, mean normalized I-V curves recorded in tsA-201 cells expressing Ca_V1.3_43S GFP-labeled peptides 1.3_C158 (gray dots) and 1.2_C349 (black dots); I-V curves for Ca_V1.3_43S and Ca_V1.3_L are depicted as gray and black fitted lines, respectively. Activation parameters of Ca_V1.3_43S + C-terminal peptides were not statistically significant from Ca_V1.3_L. For statistically significant differences, see Table 1. b, normalized peak currents evoked by 5-s depolarizing steps to V_max. c, percent I_Ca inactivation was calculated during 0.1–0.25-s test pulses to V_max. Number of experiments is given in parentheses. Error bars reflect S.E. **, p < 0.01; ***, p < 0.001 compared with Ca_v1.3_43S, one-way ANOVA was followed by Bonferroni post test.

FIGURE 8.

Activity-dependent I_Ca through Ca_v1.3 α1 subunit splice variants during APW-like stimuli. All experiments shown in a-f were recorded using 2 mm Ca²⁺ as charge carrier. a, top, APW voltage protocol elicited from −70 mV HP composed of 3 voltage ramps: −70 to +40, +40 to −80, and −80 to −70 mV (afterhyperpolarization). Bottom, normalized I_Ca in response to first APW for Ca_v1.3_L, Ca_v1.3_43S, and Ca_v1.3_42A (cells: Ca_v1.3_L, 251010_95; Ca_v1.3_43S, 221110_133; Ca_v1.3_42A, 291010_78). Right: the I_Ca response to eight APWs elicited at 100-Hz is shown for representative experiments (cells: Ca_v1.3_L, 251010_128; Ca_v1.3_43S, 251010_5; Ca_v1.3_42A, 151110_187). b, decrease of the I_peak during 100- and 25-Hz trains expressed as the ratio of I_peak after the 8th stimulus divided by the I_peak of the first APW. c, normalized integrated I_Ca (AUC) throughout eight 100- and 25-Hz trains (I_Ca during first APW was normalized to 1). *, p < 0.05; **, p < 0.01; ***, p < 0.001 comparison to Ca_v1.3_L, + and ++ comparison between Ca_V1.3_43S and Ca_V1.3_42A; d, top, voltage protocol of (SNc)-like APWs. Bottom, I_Ca in response to first APW for Ca_v1.3_L, Ca_v1.3_43S, and Ca_v1.3_42A α1 subunits; curves reflect the mean from all experiments each normalized to I_peak. Right, I_Ca response to eight APWs elicited at 25 Hz is shown for representative experiments (cells: Ca_v1.3_L, 151210_43; Ca_v1.3_43S, 101210_250; Ca_v1.3_42A, 101210_140). e and f are as described in b and c for (SNc)-like APWs. g, recovery of I_Ca inactivation in 2 mm (left) and 15 mm Ca²⁺ (right) for Ca_v1.3_L, Ca_v1.3_43S, and Ca_v1.3_42A. Recovery curves were best fitted with a bi- (Ca_v1.3_L) or monoexponential function (short variants) yielding the following time constants (τ, in ms): 15 mm Ca²⁺, Ca_v1.3_L, τ_fast, 47.4 ± 3.4; τ_slow, 385.5 ± 105; 78% τ_fast, Ca_v1.3_43S, τ = 300 ± 14.2; Ca_v1.3_42A, τ = 284 ± 7.5; 2 mm Ca²⁺, Ca_v1.3_L, τ_fast, 150 ± 102.5, τ_slow, 336 ± 367.6, 57% τ_fast, Ca_v1.3_43S: τ = 448.3 ± 14.2, Ca_v1.3_42A, τ = 519.3 ± 8.2. Data are given as mean ± S.E. All statistical comparisons were made using one-way ANOVA and Bonferroni post test.