Previously, Liu et al. demonstrated that Smad3 inhibits MyoD transcriptional activity through disruption of its binding to E-box sites of muscle genes. We thus asked whether Smad3 repression on miR-29 promoter could be executed in a similar fashion as MyoD has been implicated as an activator of miR-29 at the onset of myogenic differentiation. Four putative MyoD binding E-boxes were identified. As shown in Figure 5B, an association of MyoD with these sites was detected in differentiated myotubes without TGF-b treatment. However, the binding was largely suppressed by TGF-b. In addition to MyoD regulation, we have previously demonstrated that miR-29 promoter is epigenetically silenced in undifferentiated myoblasts by an YY1/Polycomb LEE011 CDK inhibitor repressive complex through recruitment to an YY1 binding CCAT box, and removal of this complex is necessary for the myogenic program to occur. This promoted us to ask whether TGF-b silencing miR-29 can be mediated by YY1/Polycomb complex. A search for putative YY1 binding sites uncovered a total of six sites. According to our previous findings, Y6 was competent for YY1 binding in undifferentiated myoblasts whereas Y3, Y4, Y5 were not. Y1 and Y2 represent two new sites previously untested. Subsequent ChIP-PCR assays revealed no enrichment of YY1 on any site in differentiated cells without TGF-b treatment, which is in agreement with the activation status of miR-29. However, an increase of enrichment was found at Y1, Y2, Y3 and Y6 after TGF-b treatment, indicating that TGF-b indeed enhanced YY1 binding on multiple locations. Yet, no binding on Y4 and Y5 was detected in both untreated and treated cells. GSK2118436 Additional ChIP-PCR assays showed marked increase of Ezh2 binding at all four YY1 sites ; consequently, increased levels of H3K27me3 were detected, suggesting that TGF-b treatment stabilizes YY1 binding and recruitment of Ezh2 and subsequent histone modification on multiple regions, which leads to silencing of miR-29 promoter. To substantiate the above findings from ChIP assays, reporter assays using miR-29-promoter-Luc plasmid were performed. As shown in Figure 5F, ectopic expression of YY1 repressed miR-29 reporter activities and the repression is enhanced with co-transfection of Smad3 at a dose-dependent manner, suggesting a repressive synergy between YY1 and Smad3. Ectopic expression of MyoD, on the other hand, strongly trans-activated the reporter, and this activation was repressed by Smad3 co-expression at a dosedependent manner, suggesting Smad3 inhibits MyoD activation. Moreover, addition of YY1 further abrogated MyoD activation, indicating that the two mechanisms probably co-act. Collectively, the above results suggest the inhibitory action of TGF-b/Smad3 on miR-29 transcription is exerted through dual mechanisms by blocking MyoD binding and enhancing YY1/ Ezh2 association.