Distinct roles of myofibroblast-specific Smad2 and Smad3 signaling in repair and remodeling of the infarcted heart.

TGF-βs regulate fibroblast responses, by activating Smad2 or Smad3 signaling, or via Smad-independent pathways. We have previously demonstrated that myofibroblast-specific Smad3 is critically implicated in repair of the infarcted heart. However, the role of fibroblast Smad2 in myocardial infarction remains unknown. This study investigates the role of myofibroblast-specific Smad2 signaling in myocardial infarction, and explores the mechanisms responsible for the distinct effects of Smad2 and Smad3. In a mouse model of non-reperfused myocardial infarction, Smad2 activation in infarct myofibroblasts peaked 7 days after coronary occlusion. In vitro, TGF-β1, -β2 and -β3, but not angiotensin 2 and bone morphogenetic proteins-2, -4 and -7, activated fibroblast Smad2. Myofibroblast-specific Smad2 and Smad3 knockout mice (FS2KO, FS3KO) and corresponding control littermates underwent non-reperfused infarction. In contrast to the increase in rupture rates and adverse remodeling in FS3KO mice, FS2KO animals had mortality comparable to Smad2 fl/fl controls, and exhibited a modest but transient improvement in dysfunction after 7 days of coronary occlusion. At the 28 day timepoint, FS2KO and Smad2 fl/fl mice had comparable adverse remodeling. Although both FS3KO and FS2KO animals had increased myofibroblast density in the infarct, only FS3KO mice exhibited impaired scar organization, associated with perturbed alignment of infarct myofibroblasts. In vitro, Smad3 but not Smad2 knockdown downmodulated fibroblast α2 and α5 integrin expression. Moreover, Smad3 knockdown reduced expression of the GTPase RhoA, whereas Smad2 knockdown markedly increased fibroblast RhoA levels. Smad3-dependent integrin expression may be important for fibroblast activation, whereas RhoA may transduce planar cell polarity pathway signals, essential for fibroblast alignment. Myofibroblast-specific Smad3, but not Smad2 is required for formation of aligned myofibroblast arrays in the infarct. The distinct in vivo effects of myofibroblast Smad2 and Smad3 may involve Smad3-dependent integrin synthesis, and contrasting effects of Smad2 and Smad3 on RhoA expression.