The ε subtype of protein kinase C is required for cardiomyocyte connexin-43 phosphorylation

BW Doble, P Ping, E Kardami - Circulation research, 2000 - Am Heart Assoc
Circulation research, 2000Am Heart Assoc
Gap junctions (GJs), composed of connexins, are intercellular channels ensuring electric
and metabolic coupling between cardiomyocytes. We have shown previously that an
endogenous mitogenic and cardioprotective protein, fibroblast growth factor-2 (FGF-2),
decreases cardiomyocyte GJ permeability by stimulating phosphorylation of connexin-43
(Cx43). Identifying the kinase (s) phosphorylating cardiac Cx43 may thus provide a way of
modulating cardiac intercellular communication. Because FGF-2 activates receptors linked …
Abstract
—Gap junctions (GJs), composed of connexins, are intercellular channels ensuring electric and metabolic coupling between cardiomyocytes. We have shown previously that an endogenous mitogenic and cardioprotective protein, fibroblast growth factor-2 (FGF-2), decreases cardiomyocyte GJ permeability by stimulating phosphorylation of connexin-43 (Cx43). Identifying the kinase(s) phosphorylating cardiac Cx43 may thus provide a way of modulating cardiac intercellular communication. Because FGF-2 activates receptors linked to protein kinase C (PKC) and mitogen-activated protein kinase, we first investigated participation of these enzymatic systems in Cx43 phosphorylation. The inhibitor PD98059 blocked activation of mitogen-activated protein kinase, but it did not prevent the FGF-2 effects on GJs. In contrast, the PKC inhibitor chelerythrine blocked the effects of FGF-2 on Cx43 phosphorylation and permeability. Because the ε-isoform of PKC localizes to plasma membrane sites, we examined whether it is directly involved in the FGF-2–induced Cx43 phosphorylation. In nonstimulated myocytes, PKCε displayed a discontinuous pattern of localization at intercellular contact sites and partial colocalization with Cx43. Treatment with FGF-2 or phorbol 12-myristate 13-acetate induced a more continuous pattern of PKCε distribution, whereas the anti-Cx43 staining appeared to overlap extensively with that of PKCε. In immunoprecipitation experiments using specific anti-Cx43 antibodies, PKCε but not PKCα coprecipitated with Cx43. FGF-2 increased levels of coprecipitated PKCε, suggesting increased association between PKCε and Cx43 on stimulation. Transient gene transfer and overexpression of cDNAs coding for truncated or mutated dominant-negative forms of PKCε decreased cardiomyocyte Cx43 phosphorylation significantly. We conclude that PKC mediates the FGF-2–induced effects on cardiac GJs and that PKCε likely interacts with and phosphorylates cardiac Cx43 at sites of intercellular contact.
Am Heart Assoc