|
Figure 1. CaVβ–AID complex and sequence alignment of AID motif. (A) Ribbon diagram of CaVβ2a–AID complex structure (Protein Data Bank ID no.: 1T0J) showing the side chain of the residues facing the hydrophilic environment and the conserved tryptophan that were mutated in this study (side chain shown in green, W470; in yellow, E462 and D469; and in blue, K465 and Q473). α-helixes encompassing the hydrophobic pocket of CaVβ2a are displayed in light blue, and AID α-helix is shown in red. (B) Alignment of AID sequences from several isoforms of rat CaVα1 showing positions of residues that were mutated (indicated by *).
|
|
Figure 2. Deletion of AID sequence or replacement of the conserved tryptophan abolished modulation by CaVβ2a. (A) Superimposed macroscopic current traces from oocytes coexpressing CaVβ2a either with CaV1.2 W470S or CaV1.2 ΔAID. Each trace was obtained during a 70-ms pulse of increasing amplitude, starting at −40 mV and ending at 150 mV in 10-mV increments. Membrane was held at −80 mV until the beginning of the pulse and returned to −40 mV for the remaining of the trace (shown at the top). Currents were sampled at 2.5 kHz until 3 ms before the end of the pulse, and then at 50 kHz. Traces were filtered at 10 kHz, and a P/−4 prepulse protocol was used to subtract linear components. Calibration bars correspond to 20 ms and 200 nA. (B) Conductance–voltage relationship (GV curve) for the different subunit combinations shown in A. The peak amplitude of the tail currents for each test voltage was normalized by the largest tail current (I/Imax) to generate the GV curves. Open and filled symbols correspond to oocytes recorded with or without CaVβ2a, respectively. The sums of two Boltzmann distributions that best described each set of data are shown as continuous lines.
|
|
Figure 3. Mutations of AID-exposed residues reduce I/Q only in the presence of CaVβ2a. (A) Macroscopic currents from oocytes coexpressing CaVβ2a with CaV1.2 E462D, CaV1.2 E462R, or CaV1.2 K465N during an IV stimulation protocol that consisted of 70-ms depolarizing pulses ranging from −50 mV to +60 mV in 10-mV increments from a holding voltage of −80 mV. Calibration bars correspond to 20 ms and 200 nA. (B) Average I/Q versus voltage plots for the indicated CaV1.2 subunits in the absence of CaVβ. (C) Same as B but in the presence of CaVβ2a subunit. For comparison, data from CaV1.2 WT with and without CaVβ2a are shown as dashed lines in B and C.
|
|
Figure 4. CaVβ2a retains its ability to shift GV curves, but maximal conductances are reduced in channels bearing mutations of AID-exposed residues. (A) Macroscopic currents from oocytes coexpressing CaVβ2a with either CaV1.2 E462R or CaV1.2 K465N during the same stimulation protocol used in Fig. 2 A (shown at the top), with calibration bars corresponding to 20 ms and 200 nA. (B) GV curves in the presence (filled symbol) or absence (open symbol) of CaVβ2a. (C) Plots of tail current amplitudes normalized by Qon (Itail/Qon) for CaV1.2 WT (▪), CaV1.2 E462R (•), and CaV1.2 K465N (▴). Itail/Qon (mean ± SEM) versus voltage plots were fitted to the sum of two Boltzmann distributions. The maximal Itail/Qon was 17.8 ± 2.5 nA/pC (n = 24) for CaV1.2 WT + CaVβ2a, 7.4 ± 1.74 nA/pC (n = 11) for CaV1.2 K465N + CaVβ2a, and 3.5 ± 0.8 nA/pC pC (n = 10) for CaV1.2 E462R + CaVβ2a.
|
|
Figure 5. Inactivation in the presence of CaVβ2a is not altered by E462R or K465N mutations. (A) Macroscopic currents from the indicated CaV1.2 variants during a 10-s depolarization pulse to 0 mV, with calibration bars corresponding to 5 s and 20 nA. (B) Bar plot of the percentage of currents remaining after a 10-s depolarization pulse to 0 mV for the CaV1.2 variants shown in A. Values, expressed as mean ± SEM, are 81 ± 3% (n = 7) for CaV1.2 WT, 80 ± 3% (n = 7) for CaV1.2 E462R, and 80 ± 2% (n = 8) for CaV1.2 K465N. (C) Superimposed macroscopic currents from the same oocytes shown in A during a 200-ms pulse to 0 mV after a 10-s prepulse to either −120 mV (black) or −60 mV (red), with calibration bars corresponding to 200 ms and 20 nA.
|
|
Figure 6. CaV1.2 E462R and CaV1.2 K465N bind to CaVβ2a with similar apparent affinities. (A) Macroscopic current traces (right) and I/Q versus voltage from oocytes expressing CaV1.2 E462R after the injection of purified CaVβ2a protein at the indicated concentrations. Traces correspond to superimposed responses to three 60-ms depolarizing pulses to −30 mV, 0 mV, and +30 mV from a holding voltage of −90 mV. Calibration bars correspond to 20 ms and 200 nA. Experimental I/Q values (8) were fitted to the equation (blue line):\documentclass[10pt]{article}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{pmc}
\usepackage[Euler]{upgreek}
\pagestyle{empty}
\oddsidemargin -1.0in
\begin{document}
\begin{equation*}I/Q={\mathrm{{\beta}}}2a\_like{\cdot}\left[\frac{G_{MAX}(V-V_{rev})}{1+{\mathrm{exp}}^{\left(\displaystyle\frac{z{\cdot}(V_{1/2}-V)}{25.4}\right)}}\right]_{+{\mathrm{{\beta}}}2a}+(1-{\mathrm{{\beta}}}2a\_like){\cdot}\left[\frac{G_{MAX}{\cdot}(V-V_{rev})}{1+{\mathrm{exp}}^{\left(\displaystyle\frac{z{\cdot}(V_{1/2}-V)}{25.4}\right)}}\right]_{-{\mathrm{{\beta}}}2a}.\end{equation*}\end{document}Each member of the equation corresponds to templates in absence (−β2a) or presence (+β2a) of saturating concentration of CaVβ2a protein (2.0 μM). Variables defining each template were obtained from the fit to average I/Q plot from each condition. The contribution of +β2a and −β2a templates are shown as green and red lines, respectively. (B) As A) but for CaV1.2 K465N. (C) Mean ± SE of β2a-like versus protein concentration ([CaVβ2a]) in μM. Continuous lines show the fit to a standard Hill equation:\documentclass[10pt]{article}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{pmc}
\usepackage[Euler]{upgreek}
\pagestyle{empty}
\oddsidemargin -1.0in
\begin{document}
\begin{equation*}{\mathrm{{\beta}}}2a\_like=\frac{100}{1+\left(\displaystyle\frac{K_{{\mathrm{d}}}}{[Ca_{V}{\mathrm{{\beta}}}_{2a}]}\right)^{n}}.\end{equation*}\end{document}Where Kd is the apparent dissociation constant and n is the Hill coefficient. n ranged between 1.4 and 1.6, whereas Kd for WT, E462R, and K465N was 0.20, 0.22, and 0.25 μM, respectively. The number of averaged experiments ranged from three to six for every concentration and calcium channel variant.
|
|
Figure 7. Single channel activity for different variants of CaV1.2 in the presence of CaVβ2a. (A) Representative traces from four patches containing CaV1.2 WT, CaV1.2 E462R, CaV1.2 K465N, and CaV1.2 W470S coexpressed with CaVβ2a during 200-ms depolarizations to 0 mV repeated at 1 Hz from a holding potential of −70 mV. Calibration bar corresponds to 50 ms and 2 pA. (B) Diary plots showing the calculated Po versus trace number for CaV1.2 variant shown in A. (C) Po histograms in logarithmically binned histograms for CaV1.2 variants shown in A. Note that CaV1.2 W470S yields a mono-modal Po distribution.
|
|
Figure 8. Relative frequency of Po for CaV1.2 WT, CaV1.2 E462R, CaV1.2 K465N, CaV1.2 D469S, and CaV1.2 Q473K in the presence of CaVβ2a. (A) Log-binned Po histograms compiled from several patches in the presence of CaVβ2a for the different variant of CaV1.2 as indicated in each plot. (B) Bar plot showing percentage of null traces (dark gray), traces with Po ≤ 0.1 (white), and traces with Po > 0.1 (gray) for the different CaV1.2 variants. The percentage of traces in the different categories were similar in CaV1.2 WT (34.3 ± 8.8, 25.6 ± 6.9, and 40.1 ± 9.2% for null traces, Po ≤ 0.1, and high Po > 0.1, respectively; n = 6), whereas the percentage of traces with Po > 0.1 was significantly smaller (t test; P < 0.05) for CaV1.2 E462R (6.5 ± 1.5%; n = 7), CaV1.2 K465N (4.3 ± 2.2%; n = 6), CaV1.2 D469S (8.9 ± 8.0%; n = 4), and CaV1.2 Q473 (11.3 ± 6.0%; n = 4). The fraction of low Po traces (Po ≤ 0.1) was significantly higher for CaV1.2 K465N (40.3 ± 12.0%), CaV1.2 D469S (37.8 ± 13.8%), and CaV1.2 Q473Q (41.9 ± 10.1%) than for CaV1.2 E462R (16.6 ± 2.5%). Comparing the percentage of null sweeps, only CaV1.2 E462R (78.9 ± 3.6%) is significantly higher than CaV1.2 WT (34.3 ± 8.8%). For CaV1.2 K465N, 55.4 ± 12.2% of the traces are nulls, which compares to 53.3 ± 14.4 and 46.8 ± 13.7% for CaV1.2 D469S and CaV1.2 Q473, respectively.
|
|
Figure 9. Single channel mean currents and I/Q plots from macroscopic currents from different CaV1.2 variants recorded in high Ba2+ and S(-)Bay K8644. (A) Mean current traces for six patches containing single CaV1.2 WT/CaVβ2a channels (black) from seven patches with CaV1.2 E462R/CaVβ2a channels (blue), and from 6 with CaV1.2 K465N/CaVβ2a channels (red). The number of traces averaged in each case was 4,032 for CaV1.2 WT/CaVβ2a, 7,504 for CaV1.2 E462R/CaVβ2a, and 6,104 for CaV1.2 K465N/CaVβ2a. Voltage protocol and recording condition were as described in Fig 7. Calibration bars correspond to 50 ms and 100 fA. (B) I/Q versus voltage plot for CaV1.2 WT (n = 12), CaV1.2 E462R (n = 12), and CaV1.2 K465N (n = 13) coexpressed with CaVβ2a and recorded in external 76 mM Ba2+ and 0.1 μM of S(-) Bay K 8644 as used for single channel.
|
|
Figure 10. Dwell-time histograms for CaV1.2 WT, CaV1.2 E462R, CaV1.2 K465N, and CaV1.2 W470S in the presence of CaVβ2a. (A) Representative open-time histograms in Sine-Sigworth coordinates for the indicated CaV1.2 variants. The sum of the two exponential distributions that best fit the data are shown in red, and individual components are shown in blue and green. (B) Similar to A, but for shut-time histograms. The recordings used for Fig. 7 were used for the analysis shown here.
|
|
Figure 11. Burst-duration histograms for CaV1.2 WT, CaV1.2 E462R, CaV1.2 K465N, and CaV1.2 W470S in the presence of CaVβ2a. Openings separated by brief closings of less than 1 ms were included in the same burst. To build histograms, bursts coming from all traces were used in A, traces with Po ≤ 0.1 were chosen in B, and traces with Po > 0.1 were used in C. In A and B, the sum of two exponential distributions was necessary to describe the data (red line). Individual exponential components are shown in blue and green. A single exponential distribution was sufficient to describe burst duration histogram from traces with Po > 0.1. For W470S, an exponential fit was not attempted due to the limited number of events. The same recordings shown in Fig. 7 were used here.
|
|
Figure 12. Single channel activity and burst duration histograms for CaV1.2 WT, CaV1.2 E462R, CaV1.2 K465N, and CaV1.2 W470S in the absence of CaVβ2a. (A) Representative traces of patches containing CaV1.2 WT, CaV1.2 E462R, or CaV1.2 K465N in the absence of CaVβ2a in identical recording condition as in Fig. 7. Calibration bar corresponds to 50 ms and 2 pA. (B) Burst-duration histograms for CaV1.2 WT, E462R, and K465N in the absence of CaVβ. As in Fig. 11, openings separated by brief closings of less than 1 ms were included in the same burst. All traces were used to build these histograms. A sum of two exponential distributions was necessary to describe the data.
|
