Supplementary Figure 4: The regulation of miR25 and miR93 by TET1 protein-mediated epigenetic change in response to hypoxia.

(a) The binding of HIF-1α to the VEGF gene promoter and miR25/93genome regions. VEGF gene promoter was used a positive control for HIF-1α binding upon hypoxia. Four sets of primers targeting genome regions of miR25/93 were used, as indicated. MCF7 and MDA-MB-231 cells were used, as indicated. (b) Levels of 5hmC were locally increased in miR25/93 genomic regions in response to hypoxia. (c) hMeDIP-qPCR validation of the 5hmC change in the miR25/93 genomic regions upon hypoxia. (d) ChIP analysis showed an increase in binding of TET1, but not TET3, to the miR25/93 genomic regions in response to hypoxia (left). TNF genome region was used as a positive control for either TET1 or TET3 (right). (e) qPCR analysis of TET1 knockdown cells for the expression of miR25-3p, miR93-5p, and miR93-3p upon hypoxia. (f) Western bolt analysis for TET1 level in TET1 knockdown cells. (g) The deficiency of TET3 did not alter the ability of hypoxia to induce miR25 and miR93 expression. (h) Western bolt analysis for TET3 in TET3 knockdown cells (i) In TET1 knockdown cells, hMeDIP-qPCR analysis showed that 5hmC levelson the miR25/93 genomic regions were restored in response to hypoxia by re-expressing wild type TET1, but not a TET1 inactive mutant. As representation, the third primer set out of four primer sets was used in qPCR analysis. The cells expressing shRNA targeting EGFP were used as control in panels d to h. N: normoxia H: hypoxia. All data are presented as means ± SD (n = 3 independent experiments). Samples were compared using two-tailed Student’s t test. Asterisk indicates P < 0.05, compared with controls. The image shown in panel f,h is representative of three independent experiments. Unprocessed scans of western blot analysis is available in Supplementary Figure 9. Statistics source data is available in Supplementary Table 4.