Supramolecular clustering of the cardiac sodium channel Nav1.5 in HEK293F cells, with and without the auxiliary β3‐subunit

Figures

• 

  Figure 1. Oligomerization of β3. A, Co‐IP of overexpressed β3‐EGFP and β3‐myc from singly or co‐transfected HEK293F cell lysates. β3‐myc only appeared in the bound fraction when co‐expressed with β3‐EGFP. P = “pellet,” that is, bound fraction; FT = “flow through,” that is, non‐bound fraction. B, Cross‐linking of singly expressed β3‐EGFP and β3‐myc subunits. In the absence of the cross‐linking agent (BS3), only bands consistent with the monomer (lower arrow‐head), ~56‐60 and ~36‐38 kDa for β3‐EGFP and β3‐myc, respectively, and glycosylation states were observed. Chemical cross‐linking resulted in multiple bands, consistent with the monomeric, dimeric (~120 and ~72 kDa, middle arrowhead) and trimeric (~180 and ~110 kDa, upper arrowhead) forms, and glycosylation states. C, Proximity ligation assay (PLA) of β3‐myc and β3‐EGFP. Oligomerization of β3 in situ is demonstrated by the amplification of PLA signals, seen throughout the cell and plasma membrane through a series of z‐stacks (left‐hand panel: top to bottom at 1.8 µm intervals), when β3‐myc is co‐expressed with β3‐EGFP. D, Expression of β3‐EGFP alone (right‐hand panels) did not produce the characteristic punctate red signal typical of PLA…

• 

  Figure 2. Super‐resolution STORM imaging and cluster and nearest‐neighbor analysis of Nav1.5 α‐subunits with and without β3‐subunit. A, Representative reconstructed images of fluorophore AF‐647 blinking in HEK293F cells co‐transfected with Nav1.5‐HA/EGFP and Nav1.5‐HA/β3‐EGFP. Bars = 5 µm B, The percentage of total blinking events that were observed in the same area (cluster) as other blinking events (ie, reflecting co‐localized Nav1.5 channels) in the presence and absence of β3. C, The density of blinking events observed per unit area (µm2) for cells expressing Nav1.5‐HA, plus and minus β3. D, Normalized cluster radii distribution density for cells expressing Nav1.5‐HA plus and minus β3. E, Normalized nearest‐neighbor distance density between Nav1.5‐HA molecules for cells expressing Nav1.5‐HA, plus and minus β3, both individually and superimposed. In D, differences between the histogram distributions with and without β3 were tested using the two‐sample Kolmogorov‐Smirnov test, giving a P value of 4 × 10−9. In E, differences between the histogram distributions with and without β3 were tested using the two‐sample Kolmogorov‐Smirnov test, giving a P value of ~0

• 

  Figure 3. Enhanced oligomerization of Nav1.5 α‐subunit by β3. A, Representative PLA images of Nav1.5 EGFP and Nav1.5 HA in the absence (left‐hand panels) and presence (right‐hand panels) of β3‐myc. Although not visible to the naked eye, a number of PLA signals can be detected by ImageJ and blobfinder in the absence of β3 and is recorded in the histograms. B, Quantification of PLA “dots” within the EGFP volume and PLA signal intensity (mean ± SEM). C, Orthogonal views from sequential images of the PLA signal (red) through the z‐stack in cells expressing Nav1.5 EGFP, Nav1.5 HA, and β3‐myc, highlighting a predominant, but not exclusive, localization of PLA signal at the plasma membrane. To compare the expression levels of Nav1.5 EGFP and Nav1.5 HA in these cells, see Supplementary Figure S1

• 

  Figure 4. Steady‐state activation in HEK293F‐Nav1.5 cells with and without β3 co‐expression. A, Representative whole cell Na+ currents of HEK295F‐Nav1.5 cells transfected with EGFP or β3‐EGFP in response to depolarizing steps (50 ms duration) from a holding potential of −120 mV to test pulses between −90 and +35 mV in 5 mV increments. Cell capacitances of these patches were 10.4 pF and 13.1 pF for Nav1.5 + EGFP and Nav1.5 + B3EGFP, respectively. B, I/V curves with currents normalized to cell capacitance for EGFP (open circles) and β3‐EGFP (closed squares) show no effect of β3. C, Channel conductance as a function of voltage fit with a Boltzmann function (Equation 2), showing no shift in V ½ or k with β3. Means ± SEM, n = 8 and 10 for Nav1.5 + EGFP and Nav1.5 β3‐EGFP, respectively

• 

  Figure 5. Steady‐state inactivation in HEK293F‐Nav1.5 cells with and without β3 co‐expression. A, Representative whole cell Na+ currents of HEK293F‐Nav1.5 cells transfected with EGFP or β3‐EGFP in response to a test pulse of −40 mV (50 ms duration) immediately subsequent to a pre‐pulse step (100 ms duration) to voltages ranging from −140 to −50 mV in 5 mV increments. Cell capacitances of these patches were 10.4 pF and 13.1 pF for Nav1.5 + EGFP and Nav1.5 + β3‐EGFP, respectively. B, Inactivation curves reflecting channel availability (G/G max) as a function of the pre‐pulse voltage step, means ± SEM, n = 27 and 20 for Nav1.5‐EGFP and Nav1.5 + β3‐EGFP, respectively. Solid lines are the fits to a Boltzmann function (Equation 2) highlighting a depolarizing shift in the voltage dependence of inactivation with the presence of β3

• 

  Figure 6. Hodgkin‐Huxley kinetics; time constants of the activation, m, and inactivation, h, gates, from individual experiments with (left panels) and without β3 co‐expression (center panels), and comparison of their mean (+/‐ SEM) results (right panels). A, Time constants of activation (τm). B,C, Time constants of inactivation (τh). B, before and C, following correction of the τh of the β3 containing HEK293F‐Nav1.5 cells for the observed shift (5.5 mV) in steady‐state inactivation

• 

  Figure 7. Time constants of inactivation with (open symbols) and without β3 co‐expression (filled symbols). Current decays from the steady‐state activation protocol fitted with a double exponential to give both the fast (τ1) (left panels) and slow (τ2) time constants of inactivation (right panels) in the presence and absence of the β3‐subunit. A, τ1 plotted as a function of test voltage. This demonstrates voltage dependences that are similar whether the β3‐subunit is present or absent (left panel). In contrast τ2 is relatively voltage insensitive whether or not β3 is present (right panel). B, Voltage dependences of τ1 and τ2 following shifting the results obtained in the presence of β3 by −5 mV to allow for the associated depolarizing shift in the steady‐state inactivation

• 

  Figure 8. Recovery from inactivation with and without β3 co‐expression. A, Typical whole cell Na+ currents of HEK293F‐Nav1.5 cells transfected with EGFP or β3‐EGFP in response to voltage pulse (−40 mV, 50 ms) at varying time intervals (0‐72 ms, in 3 ms increments). B, Peak currents of the test pulse normalized to the conditioning pulse (I P2/I P1) plotted as a function of recovery time. Solid lines are mono‐exponential fits highlighting an accelerated recovery time in the β3‐EGFP containing cells. n = 7, Nav1.5 + EGFP; n = 10, Nav1.5 + β3‐EGFP