The trafficking protein, EHD2, positively regulates cardiac sarcolemmal KATP channel surface expression: role in cardioprotection

Abstract

ATP-sensitive K⁺ (K_ATP) channels uniquely link cellular energy metabolism to membrane excitability and are expressed in diverse cell types that range from the endocrine pancreas to neurons and smooth, skeletal, and cardiac muscle. A decrease in the surface expression of K_ATP channels has been linked to various disorders, including dysregulated insulin secretion, abnormal blood pressure, and impaired resistance to cardiac injury. In contrast, up-regulation of K_ATP channel surface expression may be protective, for example, by mediating the beneficial effect of ischemic preconditioning. Molecular mechanisms that regulate K_ATP channel trafficking are poorly understood. Here, we used cellular assays with immunofluorescence, surface biotinylation, and patch clamping to demonstrate that Eps15 homology domain-containing protein 2 (EHD2) is a novel positive regulator of K_ATP channel trafficking to increase surface K_ATP channel density. EHD2 had no effect on cardiac Na⁺ channels (Nav1.5). The effect is specific to EHD2 as other members of the EHD family-EHD1, EHD3, and EHD4-had no effect on K_ATP channel surface expression. EHD2 did not directly affect K_ATP channel properties as unitary conductance and ATP sensitivity were unchanged. Instead, we observed that the mechanism by which EHD2 increases surface expression is by stabilizing K_ATP channel-containing caveolar structures, which results in a reduced rate of endocytosis. EHD2 also regulated K_ATP channel trafficking in isolated cardiomyocytes, which validated the physiologic relevance of these observations. Pathophysiologically, EHD2 may be cardioprotective as a dominant-negative EHD2 mutant sensitized cardiomyocytes to ischemic damage. Our findings highlight EHD2 as a potential pharmacologic target in the treatment of diseases with K_ATP channel trafficking defects.-Yang, H. Q., Jana, K., Rindler, M. J., Coetzee, W. A. The trafficking protein, EHD2, positively regulates cardiac sarcolemmal K_ATP channel surface expression: role in cardioprotection.

Keywords: caveolae; endocytic recycling; endocytosis; ischemic preconditioning.

Figure 1.

Of EHDx family members, only EHD2 regulates K_ATP current density. A) Representative inside-out current recordings obtained from HEK293 cells that stably expressed K_ir6.2/SUR2A in control, EHD2-G65R, and wild-type (WT) EHD2 groups. ATP concentrations were switched as indicated. B) Summary data of mean patch currents for all WT (black) and dominant-negative (mut; gray) EHDs, shown as percentage changes relative to control group (n > 8 patches in each group). *P < 0.05 vs. the control group.

Figure 2.

EHD2 does not directly affect K_ATP channel properties. A) Fractional currents—normalized by the mean patch current in the absence of ATP—was plotted as a function of ATP concentration. B, C) These data were fitted to a curve of a modified Hill equation to obtain IC₅₀ values and Hill constants for each trace. Aggregate data are depicted as bar graphs (means ± sem) for the control group (white, n = 11), EHD2-G65R (gray, n = 7), and wild-type EHD2 (black, n = 19). D–F) Representative recordings of residual single K_ATP channel openings in the presence of 1 mM ATP. All-points histograms were constructed from these events and fitted to a curve fitting of a gaussian distribution to determine the unitary current amplitude (n > 3 patches in each group).

Figure 3.

EHD2 increases K_ATP channel surface expression. A) HEK293 cells were transfected with Avi-K_ir6.2/SUR2A (extracellular Avi tag) and the surface channel was biotinylated. Biotinylated proteins were enriched with neutravidin beads and subjected with immunoblotting with anti-K_ir6.2 Abs. Shown are representative blots of total (T) and biotinylated surface (S) Avi-K_ir6.2 expression. B) Ratios of total K_ir6.2 to GAPDH and biotinylated to total Avi-K_ir6.2 are shown for control (white), EHD2-G65R (gray), and wild-type EHD2 (black; n = 3 blots/group). *P < 0.05 vs. control.

Figure 4.

K_ATP channels cycle through the ERC and accumulate in the ERC in the presence of EHD2. A) Representative images of COS7 cells that were transfected with myc-tagged K_ir6.2 (magenta) and GFP-tagged wild-type and dominant-negative Rab11-S25N (green). Note the intense labeling of K_ir6.2 in a perinuclear region, especially in the Rab11-S25N–expressing cells (arrows). B) Manders M2 coefficient of colocalization was calculated as an index of the presence of K_ir6.2-myc in various subcellular compartments. Scale bar, 5 μm. *P < 0.05 vs. control or Rab11 groups.

Figure 5.

K_ATP channel endocytosis is minimized in the presence of EHD2. A) Live HEK293 cells—transfected with Avi-K_ir6.2/SUR2A—were surface biotinylated and the rate of biotinylated Avi-K_ir6.2 was observed over time by immunoblot analysis—with anti-K_ir6.2 Abs—of internalized biotinylated Avi-K_ir6.2. B) Band intensities were quantified and plotted in the right panel (n = 3/group). C) Ab (anti-HA) uptake experiments with HEK293 cells that were transfected with K_ir6.2-HA (extracellular tag) and SUR2A. Endocytosed K_ir6.2-HA in vesicles was labeled and visualized with DyLight488-conjugated secondary Ab. D) Quantification of the endocytosed K_ir6.2 in control (open circle), EHD2-G65R (closed triangle), and wild-type EHD2 (closed square) groups. E) Mean patch current was recorded in excised patches from HEK293 cells that were transfected with K_ir6.2/SUR2A. Cells were pretreated with a dynamin inhibitor, Dynasore (80 µM for 2 h), or with DMSO as control. Scale bar, 10 μm. *P < 0.05 vs. control groups.

Figure 6.

EHD2 stabilizes K_ATP channels in caveolae. A) Representative images obtained with Abs against K_ir6.2 (green) and caveolin3 (magenta) showing colocalization on the cell periphery (arrowheads). Scale bar, 10 μm. B) Fractions after OptiPrep gradient ultracentrifugation collected from top to bottom were analyzed with Abs against K_ir6.2 and caveolin3 in control, EHD2-G65R, and myocytes that were pretreated with 10 mM MβCD for 1 h. C–E) Representative images showing the tracking routes of surface K_ATP channel vesicles (C), histogram distribution of vesicle mean velocity (D), and average median velocities in control, EHD2-G65R, and wild-type EHD2 groups (E; n > 132 vesicles in each group). *P < 0.05 vs. control group.

Figure 7.

Effect of dominant-negative EHD2 on K_ATP surface expression is dependent on clathrin-medicated endocytosis. A) HEK293 cells were transfected with K_ir6.2-HA (extracellular tag) and SUR2A, and surface channels were labeled with anti-HA Abs at room temperature. Ab uptake was stimulated by warming cells to 37°C (as in Fig. 5C). Shown are groups of cells that were cotransfected with EHD2-G65R (shaded bars) or with empty vector as a control (ctrl; open bars). Experiments were performed in the presence of pitstop2 (preincubation with 30 µM for 2 h) or with solvent (DMSO) alone as control. *P < 0.05 vs. no treatment groups (n > 7 cells in each group). B) HEK-293 cells transfected with Kir6.2/SUR2A, co-transfected with EHD2-G65R or empty vector, were pretreated with pitstop 2 (30 µM for 2 h) before being patch clamped. Mean patch current was measured in inside-out patches immediately after patch excision and in the absence of ATP. *P < 0.05 vs. control group; ^#P < 0.05 vs. no pitstop 2 groups (n > 11 patches in each group).

Figure 8.

EHD2 positively regulates the density of native K_ATP channels in ventricular cardiomyocytes. A) Representative images obtained with antibodies against Kir6.2 (green) and EHD2 (magenta) showing co-localization on cell periphery (arrowheads). Scale bar, 10 μm. B–D) K_ATP channel mean patch currents (B), fractional currents (normalized by the mean patch current in the absence of ATP) plotted as a function of ATP concentration (C), and unitary currents (D) in cardiomyocytes infected with mCherry (ctrl, white, open circle), EHD2-G65R (gray, closed triangle) or wild type EHD2 adenoviruses (black, closed square). *P < 0.05 vs. control group.

Figure 9.

EHD2-G65R abolished the cardioprotective effect of ischemic preconditioning. A) K_ATP channel mean patch currents in Ad-EHD2-G65R and Ad-mCherry groups (n = 4 patches in each group). B) Comparation of survival rates on rat cardiomyocytes infected with Ad-EHD2-G65R or Ad-mCherry in conditions without ischemia, with ischemia, and phenylephrine preconditioning before ischemia (n = 3 rats/group). *P < 0.05 vs. mCherry group.

Figure 10.

Proposed mechanism by which EHD2 affects K_ATP channel subcellular trafficking. K_ATP channels are internalized via dynamin and clathrin-mediated endocytosis uptake mechanism. Our data demonstrate that K_ATP channels recycle through the ERC (or recycling endosome) en route to the surface membrane. Dominant-negative constructs of Rab11 and EHD2 have opposite effects on K_ATP channel distribution and surface density, which demonstrates that EHD2 acts upstream of this step. Although it is possible that EHD2 may influence endocytosis by promoting actin nucleation, the expected result—enhanced endocytosis by EHD2—is not supported by our data. Instead, we found that EHD2 stabilizes K_ATP channel–containing caveolae, thereby limiting surface channel mobility. The result is a reduced rate of endocytosis and an increased density of surface K_ATP channels.