Erbin enhances voltage-dependent facilitation of Ca(v)1.3 Ca2+ channels through relief of an autoinhibitory domain in the Ca(v)1.3 alpha1 subunit

Abstract

Ca(v)1.3 (L-type) voltage-gated Ca2+ channels have emerged as key players controlling Ca2+ signals at excitatory synapses. Compared with the more widely expressed Ca(v)1.2 L-type channel, relatively little is known about the mechanisms that regulate Ca(v)1.3 channels. Here, we describe a new role for the PSD-95 (postsynaptic density-95)/Discs large/ZO-1 (zona occludens-1) (PDZ) domain-containing protein, erbin, in directly potentiating Ca(v)1.3. Erbin specifically forms a complex with Ca(v)1.3, but not Ca(v)1.2, in transfected cells. The significance of erbin/Ca(v)1.3 interactions is supported by colocalization in somatodendritic domains of cortical neurons in culture and coimmunoprecipitation from rat brain lysates. In electrophysiological recordings, erbin augments facilitation of Ca(v)1.3 currents by a conditioning prepulse, a process known as voltage-dependent facilitation (VDF). This effect requires a direct interaction of the erbin PDZ domain with a PDZ recognition site in the C-terminal domain (CT) of the long variant of the Ca(v)1.3 alpha1 subunit (alpha1 1.3). Compared with Ca(v)1.3, the Ca(v)1.3b splice variant, which lacks a large fraction of the alpha1 1.3 CT, shows robust VDF that is not further affected by erbin. When coexpressed as an independent entity with Ca(v)1.3b or Ca(v)1.3 plus erbin, the alpha1 1.3 CT strongly suppresses VDF, signifying an autoinhibitory function of this part of the channel. These modulatory effects of erbin, but not alpha1 1.3 CT, depend on the identity of the auxiliary Ca2+ channel beta subunit. Our findings reveal a novel mechanism by which PDZ interactions and alternative splicing of alpha1 1.3 may influence activity-dependent regulation of Ca(v)1.3 channels at the synapse.

Figure 1.

The PDZ domain of erbin binds to the C-terminal domain of α₁1.3. A, Schematic of rat α₁1.3 showing class I PDZ-binding consensus sequence (ITTL) in the cytoplasmic C-terminal domain and the region used for GST-α₁1.3 CT in pull-down assays (amino acids 1962–2155). B, Schematic of mouse erbin indicating position of LRR (amino acids 392–428) and the PDZ domain (amino acids 1280–1371). C, Binding of erbin to α₁1.3 CT. GST-α₁1.3 CT (lane 3) or GST alone (lane 2) was immobilized on glutathione-agarose beads and incubated with lysates from cells transfected with GFP-erbin₉₆₅ (amino acids 965–1371). Bound GFP-erbin₉₆₅ was detected by Western blot with anti-GFP antibodies. Lane 1 shows ∼15% of GFP-erbin₉₆₅ input used in the pull-down assay. D, Binding of erbinPDZ but not MAGI1PDZ1 to α₁1.3 CT. GST-α₁1.3 CT was incubated with purified his-S-erbinPDZ (lanes 3–5) or his-S-MAGI1PDZ1 (lanes 6–8). Input his-S-tagged protein was the following (in μg): 1 (lanes 3, 6), 2 (lanes 4, 7), and 4 (lanes 5, 8). Input lanes 1 and 2 show his-S-tagged proteins used in the assay (1 μg). E, Impaired binding of erbinPDZ to α₁1.3 CT_L–A. His-S-erbinPDZ protein (lanes 1, 4, 1.5 μg; lanes 2, 5, 3 μg; lanes 3, 6, 6 μg) was incubated with GST-α₁1.3 CT (lanes 1–3) or GST-α₁1.3 CT_L–A (lanes 4–6). In D and E, bound proteins were detected by Western blot (top) with anti-S-protein antibodies, and Ponceau staining (bottom) shows equal levels of GST-α₁1.3 CT or GST-α₁1.3 CT_L–A in each group.

Figure 2.

Coimmunoprecipitation of erbin with Ca_v1.3 from transfected cells and brain. A, Schematic of α₁1.3 and α₁1.2 showing C-terminal PDZ-binding motifs. B, Coimmunoprecipitation of erbin and Ca_v1.3 from transfected cells. HEK293T cells were cotransfected with myc-erbin and Ca_v1.3 (FLAG-α₁1.3, β_1b, and α₂δ; lanes 1, 4, 6) or Ca_v1.2 (FLAG-α₁1.2, β_1b, and α₂δ; lanes 2, 3, 5) and subjected to lysis and immunoprecipitation using rabbit antibodies against α₁1.3 (lanes 1, 2) or α₁1.2 (lanes 3, 4). Immunoprecipitated proteins (I.P.; lanes 1–4) were detected by Western blotting (WB) with antibodies recognizing FLAG (top) or myc (bottom) epitopes. Lanes 5 and 6 represent 5% of the cell lysate input used for coimmunoprecipitation. C, Coimmunoprecipitation of erbin and Ca_v1.3 from the brain. Rat brain lysates were incubated with goat α₁1.3 antibodies or control goat IgG for immunoprecipitation, and α₁1.3 and erbin were detected by Western blotting with rabbit α₁1.3 and erbin antibodies, respectively. D, Lack of α₁1.3 immunoprecipitation in Ca_v1.3^−/− mouse brain. α₁1.3 immunoprecipitation protocol used in C was applied to brain lysates from Ca_v1.3^+/+ or Ca_v1.3^−/− mice. An ∼200 kDa protein corresponding to α₁1.3 was detected in the brain lysate, and samples were immunoprecipitated with α₁1.3 antibodies from Ca_v1.3^+/+ but not Ca_v1.3^−/− mice. Lysate lanes represent 3% of the input used for coimmunoprecipitation.

Figure 3.

Ca_v1.3 and erbin colocalize in somatodendritic domains of cortical neurons in culture. A–I, Confocal images of primary cortical neuron cultures (A–F, 14 d in culture; G–I, 8 d in culture) sequentially double labeled with antibodies against α₁1.3 (to detect Ca_v1.3) and erbin are shown. Immunofluorescence was viewed under optics for rhodamine for Ca_v1.3 (A, D, G) or fluorescein for erbin (B, E, H). Regions of colocalization appear yellow in the merged images (C, F, I). Extensive colocalization of erbin and Ca_v1.3 was detected in cell bodies (A–C, G–I) and dendrites (D–F; higher-magnification view of boxed region in panels A–C). In D–I, elements immunoreactive for Ca_v1.3 alone (double arrowheads), erbin alone (arrowheads), or Ca_v1.3 plus erbin (arrows) are indicated. Scale bars: A (for A–C), G (for G–I), 20 μm; (in D) D–F, 10 μm.

Figure 4.

Erbin does not affect voltage-dependent activation or mamplitude of Ca_v1.3 currents. A, B, Normalized tail current (Norm. I_tail) activation curves for HEK293T cells transfected with Ca_v1.3 alone (n=9) (A) or cotransfected with Ca_v1.3 plus erbin (n=6) (B). Test pulses (10 ms) were applied from a holding voltage of −90 mV to various voltages, and peak tail currents were measured after repolarization to −70 mV for 2 ms. Tail currents were normalized to the largest in the series and plotted against test voltage. Representative current traces are shown at top. C, D, Current–voltage relationships for Ca_v1.3 alone (C) and Ca_v1.3 plus erbin (D). Data were from same voltage protocol as in A and B, except that peak current amplitudes during the test pulse were plotted against test voltage. Error bars represent SEM.

Figure 5.

Erbin augments VDF of Ca_v1.3 Ba²⁺ currents. A, Representative Ba²⁺ current traces and double-pulse voltage protocol for measuring VDF in cells transfected with Ca_v1.3 alone or cotransfected with erbin. Currents were evoked by 10 ms test pulses from −90 to −20 mV before (P1) and after (P2) a conditioning 20 ms prepulse (Pre) to +60 mV. B, No effect of erbin on prepulse voltage dependence of facilitation. Percentage of facilitation was calculated as [(I_P2/I_P1) − 1] × 100 for cells for different prepulse voltages, normalized to that for a +60 mV prepulse, and plotted as percentage of maximum (Max.) facilitation against prepulse voltage for cells transfected with Ca_v1.3 alone (n=11) or cotransfected with erbin (n=10). C, Increased net facilitation attributable to erbin. F_ratio represents the ratio of the P2 and P1 test currents and is plotted against prepulse voltage for Ca_v1.3 alone (n = 12) or Ca_v1.3 plus erbin (n = 10). *p < 0.05. Error bars represent SEM.

Figure 6.

Erbin increases VDF of Ca_v1.3 Ca²⁺ currents. A, Representative Ca²⁺ current traces and double-pulse voltage protocol for measuring VDF in cells transfected with Ca_v1.3 alone (n=6) or cotransfected with erbin (n=7). Currents were evoked by 5 ms test pulses from −90 to −10 mV before (P1) and after (P2) a conditioning 10 ms prepulse (Pre) to varying voltages. B, Increased net facilitation for prepulse voltages greater than +40 mV in cells cotransfected with erbin. F_ratio was determined and plotted against prepulse voltage as in Figure 5C. The dotted line indicates F_ratio=1. Points falling below the line result from Ca²⁺-dependent inactivation with prepulse voltages evoking significant Ca²⁺ entry. *p < 0.05 by two-factor ANOVA and Tukey's post-test. Error bars represent SEM.

Figure 7.

Increased VDF depends on PDZ domain of erbin. A, Representative current traces obtained with the same voltage protocol as in Figure 5A for cells cotransfected with Ca_v1.3 plus erbin, erbin_ΔPDZ, or erbin965 (amino acids 965–1371). B, Deletion of erbinPDZ prevents effects on VDF. F_ratio is plotted against prepulse V for cells cotransfected with Ca_v1.3 plus erbin (n=10) or erbin_ΔPDZ (n=8). *p < 0.05. C, Deletion of LRR of erbin does not prevent effects of erbin on VDF. F_ratio was plotted against prepulse voltage for cells cotransfected with Ca_v1.3 plus erbin (n=10) or erbin965 (n=5). The dashed line represents data replotted from Figure 5C for cells transfected with Ca_v1.3 alone. *p < 0.05 for Ca_v1.3 plus erbin965 compared with Ca_v1.3 alone. Error bars represent SEM.

Figure 8.

Erbin does not cause increased VDF of Ca_v1.3b channels. A, Schematic showing differences in C-terminal domain of α₁1.3 and α₁1.3b. Parentheses indicate unique sequence including the PDZ-binding site (ITTL) in α₁1.3 but not α₁1.3b. Splicing of this region in α₁1.3b results in a truncated C-terminal domain with a distinct C-terminal sequence (ERML). B, ErbinPDZ interacts weakly with α₁1.3b CT. GST-α₁1.3 CT or GST-α₁1.3b CT was used in pull-down assay with his-S-erbinPDZ (1.5 or 3.0 μg). Top, Western blot detection of erbinPDZ. Bottom, Ponceau staining showing equal levels of immobilized GST-α₁1.3 CT (left) and GST-α₁1.3b CT (right) used in assay. C, No effect of erbin on VDF of Ca_v1.3b. F_ratio was plotted against prepulse voltage for cells transfected with Ca_v1.3b alone (n = 14) or cotransfected with erbin (n=8). The dashed line represents data replotted from Figure 5C for cells transfected with Ca_v1.3 alone. Representative current traces obtained with a +60 mV prepulse are shown above. Error bars represent SEM.

Figure 9.

Autoinhibition of VDF by the CT of α₁1.3. A–C, Cotransfection of α₁1.3 CT fragment (CT₅₀₀; amino acids 1655–2155) suppresses VDF. Effect of CT₅₀₀ on F_ratio is plotted against prepulse V in cells with Ca_v1.3b (n=6) (A), Ca_v1.3 (n=4; B), or Ca_v1.3 plus erbin (n=10; C). Representative current traces obtained with a +60 mV prepulse are shown above. Vertical scale bars, 0.8 nA; horizontal scale bars, 40 ms. *p < 0.01. D, Voltage-dependent inhibition of VDF by CT₅₀₀. Percentage inhibition was measured as [1 − (F_ratio+CT500/F_ratio)] × 100, where F_ratio is for Ca_v1.3b, Ca_v1.3, or Ca_v1.3 plus erbin, and F_ratio+CT500 represents the mean F_ratio for these groups cotransfected with CT₅₀₀. Stronger prepulse voltage dependence of CT₅₀₀ inhibition is observed for Ca_v1.3b and Ca_v1.3 plus erbin compared with Ca_v1.3. *p < 0.001. Error bars represent SEM.

Figure 10.

Effect of erbin but not α₁1.3 CT depends on auxiliary Ca²⁺ channel β subunit. A, Erbin does not increase VDF of Ca_v1.3 channels containing the β₄ subunit (Ca_v1.3-β₄). F_ratio was determined from double-pulse protocol as described in Figure 5, except that P1 and P2 currents were evoked by 7 ms test pulses from −90 to −20 mV in cells transfected with Ca_v1.3-β₄ (n=7) or cotransfected with erbin (n=8). B, CT₅₀₀ inhibits VDF of β₄-containing Ca_v1.3b channels (Ca_v1.3b-β₄). F_ratio was determined and plotted as in A for cells transfected with Ca_v1.3b-β₄ (n=6) or cotransfected with CT₅₀₀ (n=7). *p < 0.001. For A and B, representative current traces obtained with +80 mV prepulse are shown above. Error bars represent SEM.