Caveolin-3 associates with and affects the function of hyperpolarization-activated cyclic nucleotide-gated channel 4

Abstract

Targeting of ion channels to caveolae, a subset of lipid rafts, allow cells to respond efficiently to extracellular signals. Hyperpolarization-activated cyclic nucleotide-gated channel (HCN) 4 is a major subunit for the cardiac pacemaker. Caveolin-3 (Cav3), abundantly expressed in muscle cells, is responsible for forming caveolae. P104L, a Cav3 mutant, has a dominant negative effect on wild type (WT) Cav3 and associates with limb-girdle muscular dystrophy and cardiomyopathy. HCN4 was previously shown to localize to lipid rafts, but how caveolae regulate the function of HCN4 is unknown. We hypothesize that Cav3 associates with HCN4 and regulates the function of HCN4 channel. In this study, we applied whole-cell patch clamp analysis, immunostaining, biotinylation, and immunoprecipitation methods to investigate this hypothesis. The immunoprecipitation results indicated an association of HCN4 and Cav3 in the heart and in HEK293 cells. Our immunostaining results showed that HCN4 colocalized with Cav3 but only partially colocalized with P104L in HEK293 cells. Transient expression of Cav3, but not P104L, in HEK 293 cells stably expressing HCN4 caused a 45% increase in HCN4 current (IHCN4) density. Transient expression of P104L caused a two-fold increase in the activation time constant for IHCN4 and shifted the voltage of the steady-state inactivation to a more negative potential. We conclude that HCN4 associates with Cav3 to form a HCN4 macromolecular complex. Our results indicated that disruption of caveolae using P104L alters HCN4 function and could cause a reduction of cardiac pacemaker activity.

Figure 1

HCN4 stably expressed in HEK293 cells. (Panel A) Detecting HCN4 protein from a HCN4 stable cell line (right) but not from HEK293 cell protein (left). (Panel B) Confocal image of HCN4 stable cell lines immunostained with anti-HCN4 antibody. (Panel C) Representative current traces recorded from HCN4 stable cell lines before (black) and after (red) perfusion of 2 mM Cs⁺.

Figure 2

Representative images from Co-Ip and Western blot analysis of HCN4 and Cav3 from the mouse heart (panels A and B) and HEK293 cells transiently expressed with HCN4 and/or Cav3 (panels C, D, E, and F). Panel A: mouse heart lysate was Co-IPed with anti-Cav3 antibody or mouse IgG, and then the Co-IP samples were analyzed by Western blotting with anti-HCN4 antibody. Panel B: mouse heart lysate was Co-IPed with anti-HCN4 antibody or rabbit IgG, and then the Co-Ip samples were analyzed by Western blotting with anti-Cav3 antibody. Panel C: lysates of HEK293 cells expressing HCN4 and Cav3 were co-IPed with anti-Cav3 antibody or mouse IgG, and then the Co-IP samples were analyzed by Western blotting with anti-HCN4 antibody. Panel D: lysates of HEK293 cells expressing HCN4 and Cav3 were co-IPed with anti-HCN4 antibody or mouse IgG, and then the Co-IP samples were analyzed by Western blotting with anti-Cav3 antibody. Panel E and F: control IP experiments using lysates from HCN4 transfected HEK293 cells IPed with anti-Cav3 antibody (panel E), or Cav3 transfected HEK293 cells IPed with anti-HCN4 antibody (panel F).

Figure 3

Confocal images from HCN4 stably expressed HEK293 cells transiently expressed with pcDNA3 (A), WTCav3 (B), or Cav3P104L(C). The cells were immunostained with rabbit anti-HCN4 polyclonal antibody and mouse anti-Cav3 monoclonal antibody. Cell surface biotinylation of HCN4 channels was examined in HEK293 cells transiently transfected with HCN4, HCN4/Cav3, HCN4/P104L, or pcDNA3. Biotin-labeled HEK293 cell lysate was immunoprecipitated with anti-HCN4 antibody and subjected to Western blot analysis with antibiotin antibody (D), then stripped and reprobed with anti-HCN4 antibody (E); the 160 kDa band for HCN4 is indicated.

Figure 4

I_HCN recorded from HCN4 stably expressed HEK293 cells. Panel A: representative traces were recorded with a test potential of 3-s duration from −130 to −60 mV from a holding potential of −40 mV. Panel B: summary data for peak I_HCN densities for pcDNA3, Cav3, or P104L transiently expressed in the HCN4 stably expressed HEK293 cell. I_HCN were elicited by a step hyperpolarization to −120 mV from a holding potential of −40 mV and normalized to membrane capacitance. Bars depict the mean and standard error of the mean derived from the number (n) of experiments with n shown above the bar.

Figure 5

Activation kinetics of I_HCN from HCN4 stably expressed HEK293 cells transiently expressed with pcDNA3, Cav3, or P104L. Panel A: current–voltage relationship. The y-axis is peak I_HCN normalized to the maximal I_HCN obtained in the protocol as depicted in Figure 4. Solid symbols represent the mean data from 9 experiments, and the bars represent stand error. Individual curves were fitted to the Boltzmann equation. Panel B: summary data for an activation voltage (V_1/2) when 50% of I_HCN activated. Panel C: summary data for activation time constant (τ) at the testing potential of −120 mV. I_HCN was elicited by a step hyperpolarization to −120 mV from a holding potential of −40 mV and normalized to membrane capacitance. Bars depict the mean and standard error of the mean derived from the number of experiments (n) with the means and stand errors of the mean shown above the bar.