Functional study of microRNAs miR-450a and miR-450b-5p in tumorigenesis

Ovary cancer is the fifth largest cause of cancer-related deaths for women worldwide, with patients presenting extremely low 5-year survival rates. MicroRNAs (miRNAs), non-coding RNAs around 22 ribonucleotides long, play a critical role in gene expression regulation in virtually all biological processes. miRNAs induce gene silencing of their targets through translation repression or mRNA cleavage. We showed that miR-450a and miR-450b-5p, two functionally underexplored miRNAs in gynecologic tumors, were downregulated in ovarian cancer and cervical cancer cell lines. The aim of this work was to verify the hypothesis that both miR-450a and miR- 450b-5p display tumor suppressor functions in reproductive tissue cancers. For this purpose, we utilized in vitro assays and an in vivo xenographic model. Using a lentiviral expression vector, miR-450a and miR-450b-5p were stably overexpressed in A2780, OVCAR-3 and SKOV-3 (ovarian cancer cell lines) and C33 and SiHA (cervical cancer cell lines). The anoikis rate was analyzed in those cells submitted to low adherence plates for 24 hours: miR-450a overexpression resulted in lower viability in SiHa, OVCAR-3, A2780 and SKOV-3 cells. Regarding to cell proliferation, A2780 cells overexpressing miR-450a proliferated at higher rate compared to the control counterpart. However, these cells presented fewer but noticeable larger colonies at the clonogenic assay. On the other hand, miR-450b decreased the proliferation rate in SKOV-3, what was corroborated in the clonogenic assay, wherein the number of colonies were reduced at the same cell line. We also measured cell migration and invasion rates using transwell assays. Overexpression of miR-450a or miR-450b-5p resulted in higher migration potential of C33 cell line but lower migration and invasion rates in A2780 and SKOV3 cell lines. In OVCAR-3 cell line, miR-450a overexpression caused an increase in number of invaded cells and miR-450b-5p overexpressed cells migrated in higher number than control group. The overexpression of miR-450b-5p in SiHa cell line resulted in lower invasion potential compared to correspondent control group. After these in vitro experiments, we decided to continue our analysis only with A2780 and SKOV-3 cells, which had the most similar pattern. In the in vivo assay, transduced A2780 cells were injected intraperitoneally in 9-11 weeks immunodeficient mice and the weight of dissected tumors were compared after 4 weeks. In this in vivo xenographic model of ovarian cancer, we verified that both miRNAs impaired tumor growth, which corroborates our in vitro assays. We also verified that most of the direct targets found through PARCLIP technique validated in this work for miR-450a are related to energetic metabolism in mitochondria (TIMMDC1, ACO2, ATP5B) besides PAPSS1, CITED2, MAML1 and VIM, the canonical marker of EMT. The direct target for miR-450b, ME1, is also related to energetic metabolism. To date, our analysis points to the scenario of miR-450a and miR-450b-5p playing functions like tumor suppressors proven to be relevant to the tumorigenic process of ovarian epithelial tumors. Additionally, they do that mostly by regulating genes responsible by the energetic metabolism. (AU)