HIF-inducible miR-191 promotes migration in breast cancer through complex regulation of TGFβ-signaling in hypoxic microenvironment.

The molecular mechanisms of hypoxia induced breast cell migration remain incompletely understood. Our results show that hypoxia through hypoxia-inducible factor (HIF) brings about a time-dependent increase in the level of an oncogenic microRNA, miR-191 in various breast cancer cell lines. miR-191 enhances breast cancer aggressiveness by promoting cell proliferation, migration and survival under hypoxia. We further established that miR-191 is a critical regulator of transforming growth factor beta (TGFβ)-signaling and promotes cell migration by inducing TGFβ2 expression under hypoxia through direct binding and indirectly by regulating levels of a RNA binding protein, human antigen R (HuR). The levels of several TGFβ pathway genes (like VEGFA, SMAD3, CTGF and BMP4) were found to be higher in miR-191 overexpressing cells. Lastly, anti-miR-191 treatment given to breast tumor spheroids led to drastic reduction in spheroid tumor volume. This stands as a first report of identification of a microRNA mediator that links hypoxia and the TGFβ signaling pathways, both of which are involved in regulation of breast cancer metastasis. Together, our results show a critical role of miR-191 in hypoxia-induced cancer progression and suggest that miR-191 inhibition may offer a novel therapy for hypoxic breast tumors.


Supplementary data 3.-Confirmation of overexpression/ inhibition of specific molecules achieved by transient transfection a,b.
Level of miR-191 was transiently up/ downregulated by transfecting with pre-191 (Ambion) & anti-191 oligos (Exiqon) and exposed to hypoxic microenvironment for 24 hrs. The corresponding effect on the expression of miR-191 was then checked by qRT-PCR compared to that of their respective controls in MCF-7 cells. c-e. MCF-7 cells were transfected with pcdnA3.1-HuR or with esiRNA for HuR inhibition and exposed to hypoxic microenvironment for 24 hrs. The graph shows relative expression levels as determined by qRT-PCR (c,d) and western blotting (e) of HuR compared to that of respective control transfected cells. f-h. MCF-7 cells were transfected with TGFβ2 esiRNA or treated with recombinant TGFβ2 and the efficiency of esiRNA mediated inhibition or overexpression was confirmed through measuring TGFβ2 levels through qRT-PCR (f,g) or luciferase activity of pTPlux reporter construct (h). i. qRT-PCR data showing effect of ctrl or SMAD3 siRNA on SMAD3 levels in MCF7 cells. The graphical data points represent mean + S.D of at least three independent experiments. (**P<0.01). Error bars denote + SD.
Supplementary data 4.-Effect of miR-191 on TGFβ2 & HuR levels in the presence/absence of hypoxic microenvironment a,b. Effect of miR-191 on TGFβ2 and HuR at protein levels in hypoxic microenvironment. The MCF7 cells with differential level of miR-191 were exposed to hypoxic microenvironment and analyzed for expression of TGFβ2 (a) and HuR (b) protein by western blotting and the relative protein levels were then quantified using imagej software for band densitometery. c. Level of miR-191 was differentially modulated and cells (MCF7) were exposed to hypoxic microenvironment, Immunofluorescence was done by using an antibody specific for TGFβ2 to compare the difference in fluorescence intensity observed. d. Cells were transfected with esictrl or esiHuR oligos in normoxia and qRT-PCR was performed. The results show that under normoxic conditions HuR is unable to downregulate TGFβ2 levels in breast cancer cell lines. e-g qRT-PCR was done to find out the effect of miR-191 on TGFβ2 & HuR levels in both normoxic and hypoxic microenvironment in a panel of breast cancer cell lines MCF7 (e), T47D (f) & MM231 (g). We found that the effect was minimal in the absence of hypoxic microenvironment. The graphical data points in a-g represent mean + S.D of at least three independent experiments. (*P<0.05, **P<0.01, *^P>0.05<0.1). Error bars denote + SD.

Supplementary data 5.-HuR is a direct target of miR-191. a.
Diagram showing sequence and position of miR-191 binding sites (wild type: HuR B1 & B2, mutant type HuR Mut B1) in HuR 3'UTR. b,c. 3'UTR luciferase activity of lucifease constructs (individually cloned: HuR B1, HuR B2, Both together: HuR B1&2) in response to differential miR-191 levels. It was found that miR-191 overexpression led to decrease in luciferase activity with HuR B1 and HuR B1&2 constructs (b) while lesser effect was observed with HuR B2 (b) or mutated HuR B1 sites (c). Therefore, miR-191 mediated downregulation of HuR is mediated mainly through binding to the HuR B1 site present in the HuR 3'UTR. The graphical data points in b & c represent mean + S.D of at least three independent experiments. (*P<0.05, **P<0.01, *^P>0.05<0.1). Error bars denote + SD.

Supplementary data 6.-Diagram showing putative miR-191 and HuR binding sites in TGFβ2 3'UTR
Diagram showing in-silico analysis data of the 3'UTR region of TGFβ2 for the putative HuR consensus sequences and miR-191 binding sites using RNAhybrid, ARE score and catRAPID softwares. The 3'UTR region containing the miR-191 and HuR binding sites of TGFβ2-3'UTR was cloned and checked for luciferase activity. Primer details are given in Supplementary data 1.

Supplementary data 7.-Validation of hypoxic microenvironment and transfection efficiency in the tumor spheroids a.
To recapitulate the endogenous hypoxic microenvironment 3D tumor spheroids were manifested and the level of VEGFA was sought as a marker of hypoxic microenvironment. b. The 3D tumor spheroids of MCF7 were transfected with anti-miR91/NCtrl and the effectiveness of transfection was confirmed by qRT-PCR. The graph shows miR-191 levels in antimiR-191/NCtrl treated MCF7 spheroids on day 1 and 7. The readings were normalized using U6. . The graphical data points in a and b represent mean + S.D of at least three independent experiments. (*P<0.05, **P<0.01). Error bars denote + SD.