SMAD1 and SMAD5 Expression Is Coordinately Regulated by FLI1 and GATA2 during Endothelial Development
Marella F. T. R. de Bruijn
The bone morphogenetic protein (BMP)/SMAD signaling pathway is a critical regulator of angiogenic sprouting and is involved in vascular development in the embryo. SMAD1 and SMAD5, the core mediators of BMP signaling, are vital for this activity, yet little is known about their transcriptional regulation in endothelial cells. Here, we have integrated multispecies sequence conservation, tissue-specific chromatin, in vitro reporter assay, and in vivo transgenic data to identify and validate Smad1+63 and the Smad5 promoter as tissue-specific cis-regulatory elements that are active in the developing endothelium. The activity of these elements in the endothelium was dependent on highly conserved ETS, GATA, and E-box motifs, and chromatin immunoprecipitation showed high levels of enrichment of FLI1, GATA2, and SCL at these sites in endothelial cell lines and E11 dorsal aortas in vivo. Knockdown of FLI1 and GATA2 but not SCL reduced the expression of SMAD1 and SMAD5 in endothelial cells in vitro. In contrast, CD31+ cKit− endothelial cells harvested from embryonic day 9 (E9) aorta-gonad-mesonephros (AGM) regions of GATA2 null embryos showed reduced Smad1 but not Smad5 transcript levels. This is suggestive of a degree of in vivo selection where, in the case of reduced SMAD1 levels, endothelial cells with more robust SMAD5 expression have a selective advantage.Ethics approval for animal work was granted by the Animal Care and Ethics Committee of the University of New South Wales.To determine where canonical BMP/SMAD signaling was active in the developing embryo, we used the BMP response element (BRE)-lacZ reporter mouse line that expresses the lacZ reporter in response to Smad binding (21). The expression pattern observed showed reporter activity in the endothelium, including the dorsal aorta (Fig. 1A). To complement these data, qPCR was performed on sorted endothelial cell, hematopoietic stem and progenitor cell (HSPC), and mesenchymal cell fractions from the E11 aorta-gonad-mesonephros (AGM) region (Fig. 1B) (22). Confocal microscopy confirmed the presence of SMAD1 and SMAD5 protein expression in endothelial cells in E11 dorsal aortas (Fig. 1C). Together, these data demonstrate that canonical BMP/SMAD signaling is active in the developing endothelium of mouse embryos (13, 27).Potential regulatory regions in the Smad1 and Smad5 loci were screened by aligning sequence conservation plots with chromatin accessibility data for human umbilical vein endothelial cells (HUVECs) from the ENCODE project (Fig. 2A). Four putative enhancer sites were identified in the Smad1 locus and termed the +1, +19, +25, and +63 sites according to kilobase distance from the Smad1 transcriptional start site (Fig. 2A). One putative site was also identified in the Smad5 promoter. No sites were observed in the intergenic regions between Smad1 and Smad5 and their respective flanking genes (data not shown). ChIP-qPCR performed with MS1 and SEND endothelial mouse cell lines showed strong enrichment of H3K9ac at the corresponding mouse regions, indicating that they are likely to be conserved (Fig. 2B).Human DNA sequences for the four potential Smad1 cis-regulatory regions and the Smad5 promoter were cloned into luciferase vectors and tested in stable-transfection assays in endothelial cell lines (Fig. 2C). All regulatory regions showed some level of activity over the empty vector control, with the Smad5 promoter and Smad1+63 regions consistently showing the highest level of activity across both cell lines. The three most active Smad1 elements, Smad1+19, Smad1+25, and Smad1+63, as measured by stable-transfection data for endothelial cells, and the Smad5 promoter were cloned into lacZ vectors and tested in F0 transgenic embryo assays. The Smad1+63 (5/5) and Smad5 promoter (6/6) transgenics showed staining in endothelial and mesenchymal cells consistent with Smad1 and Smad5 expression in the E11 AGM region (Fig. 2D). The Smad1+19 (3/3) and Smad1+25 (3/4) transgenics showed variable staining only in subendothelial mesenchymal cells in the dorsal aorta (data not shown).This study was funded by the National Health and Medical Research Council of Australia, the Translational Cancer Research Network, the Cancer Institute of New South Wales, and the Tom Bee Research Fund (J.E.P.); the Medical Research Council, United Kingdom (M.F.T.R.D.B.); the Kay Kendall Leukemia Fund and Leukemia and Lymphoma Research (K.O.); and a National Institutes of Health grant (R37DK054077) and a ZonMw TOP award (40-00812-98-11068) (E.D.).J.M.-B., A.K., V.C., T.B., C.E., K.K., S.F., Q.Q., J.T., and Y.C.K. performed research and analyzed data. L.O. and K.O. provided essential reagents and supervision. M.F.T.R.D.B. and E.D. provided essential supervision and resources and helped in project design and data interpretation. J.E.P. designed the study, interpreted data, and wrote the paper with J.M.-B.We have no conflicts of interest to declare.Supplemental material for this article may be found at http://dx.doi.org/10.1128/MCB.00239-15.