Cardiac fibroblast GSK-3α mediates adverse myocardial fibrosis via IL-11 and ERK pathway

BackgroundHeart failure is the leading cause of mortality, morbidity, and healthcare expenditures worldwide. Numerous studies have implicated Glycogen Synthase Kinase-3 (GSK-3) as a promising therapeutic target for cardiovascular diseases. GSK-3 isoforms appear to play overlapping, unique, and even opposing functions in the heart. Recently our group has identified cardiac fibroblast (CF) GSK-3{beta} as a negative regulator of fibrotic remodeling in the ischemic heart. However, the role of CF-GSK-3 in myocardial fibrosis is unknown. Methods and ResultsHerein, we employed two entirely novel conditional fibroblast-specific and tamoxifen-inducible mouse models to define the role of CF-GSK-3 in fibroblast activation and myocardial fibrosis. Specifically, GSK-3 was deleted from cardiac fibroblasts or myofibroblasts with tamoxifen-inducible Tcf21- or periostin-promoter-driven Cre recombinase. At 2 months of age, WT and KO mice were subjected to cardiac injury, and heart functions were monitored by serial echocardiography. Histological analysis and morphometric studies were performed at 8 weeks post-injury. In both settings, GSK-3 deletion restricted fibrotic remodeling and improved cardiac function. To investigate underlying mechanisms, we examined the effect of GSK-3 deletion on myofibroblast transformation and pro-fibrotic TGF{beta}1-SMAD3 signaling in vitro. A significant reduction in cell migration, collagen gel contraction, and -SMA expression in TGF{beta}1 treated GSK-3 KO MEFs confirmed that GSK-3 is required for myofibroblast transformation. Surprisingly, GSK-3 deletion did not affect SMAD3 activation, indicating the pro-fibrotic role of GSK-3 is SMAD3 independent. To further delineate the underlying mechanism, total proteins were isolated from CFs of WT and KO animals at 4 weeks post-injury, and kinome profiling was performed by utilizing PamStation(R)12 high throughput microarray platform. The kinome analysis identified the downregulation of RAF family kinase activity in GSK3-KO-CFs. Moreover, mapping of significantly altered kinases against literature annotated interactions generated ERK-centric networks. Importantly, flow cytometric analysis of CFs confirmed a significant decrease in pERK levels in KO mice. Additionally, our in vitro studies demonstrated that GSK-3 deletion prevented TGF{beta}1 induced ERK activation thereby validating our findings from kinome analysis. Interestingly, IL-11, a fibroblast specific downstream effector of TGF{beta}1, was very low in GSK-3 KO MEFs as compared to WT and ERK inhibition further reduced IL-11 expression in them. All these results indicate that GSK-3 mediates pro-fibrotic response in the injured heart through IL-11 and ERK pathway. ConclusionCF-GSK-3 plays a causal role in myocardial fibrosis that could be therapeutically targeted for future clinical applications.