Molecular profiling uncovers a p53-associated role for microRNA-31 in inhibiting the proliferation of serous ovarian carcinomas and other cancers

Abstract

MicroRNAs (miRNA) regulate complex patterns of gene expression, and the relevance of altered miRNA expression to ovarian cancer remains to be elucidated. By comprehensively profiling expression of miRNAs and mRNAs in serous ovarian tumors and cell lines and normal ovarian surface epithelium, we identified hundreds of potential miRNA-mRNA targeting associations underlying cancer. Functional overexpression of miR-31, the most underexpressed miRNA in serous ovarian cancer, repressed predicted miR-31 gene targets including the cell cycle regulator E2F2. MIR31 and CDKN2A, which encode p14(ARF) and p16(INK4A), are located at 9p21.3, a genomic region commonly deleted in ovarian and other cancers. p14(ARF) promotes p53 activity, and E2F2 overexpression in p53 wild-type cells normally leads via p14(ARF) to an induction of p53-dependent apoptosis. In a number of serous cancer cell lines with a dysfunctional p53 pathway (i.e., OVCAR8, OVCA433, and SKOV3), miR-31 overexpression inhibited proliferation and induced apoptosis; however, in other lines (i.e., HEY and OVSAYO) with functional p53, miR-31 had no effect. Additionally, the osteosarcoma cell line U2OS and the prostate cancer cell line PC3 (p14(ARF)-deficient and p53-deficient, respectively) were also sensitive to miR-31. Furthermore, miR-31 overexpression induced a global gene expression pattern in OVCAR8 associated with better prognosis in tumors from patients with advanced stage serous ovarian cancer, potentially affecting many genes underlying disease progression. Our findings reveal that loss of miR-31 is associated with defects in the p53 pathway and functions in serous ovarian cancer and other cancers, suggesting that patients with cancers deficient in p53 activity might benefit from therapeutic delivery of miR-31.

Figure 1

Profiling of miRNA, mRNA, and DNA in human serous ovarian tumors and cell lines. (A) Heat map representation of miRNAs (left panel) and genes (right panel) overexpressed (yellow) and underexpressed (blue) in both cell lines and tumors compared to NOSE. Rows, miRNAs or gene transcripts; columns, profiled samples. (B) Numbers of predicted miRNA-mRNA functional pairs for each algorithm and intersection of algorithms based on anti-correlated expression in ovarian cancer. (C) Percentages of genes overexpressed or underexpressed (from part A) that were located in a cytoband regions showing consistent copy number gain (orange) or loss (blue). (D) DNA copy alterations in serous ovarian cancer. Top panel: Frequency plot of DNA copy number gains or losses in a panel of 14 serous tumors, where locations of microRNAs from part A are indicated. Bottom panel: DNA loss or gain in regions flanking miR-31 in 178 serous ovarian tumors from the TCGA.

Figure 2

Modulation of miR-31 impacts predicted gene targets. A) Cells were transfected with miR-31 and profiled for gene expression. Genes represented in the profile dataset were ranked by fold change (overexpression/control). GSEA evaluated enrichment within the miR-31-expressing cells for predicted miR-31 targets, as determined by the given algorithm (Miranda, green; PicTar, blue; TargetScan, red). Vertical bars along the x-axis of the GSEA plot denote the positions, within the ranked list, of genes in the given set. Negative GSEA curve denotes anti-enrichment. (B) QPCR analysis showing relative quantity of miR-31 predicted targets (each normally underexpressed in cancer) after miR-31 overexpression in OVCAR-8 cells. For each gene, “miR-31 mimic” group (with overexpression of miR-31) is lower than either of the two control groups (p<0.05, two-sided t-test, each comparison). Bars indicate standard error. (C) Anti-enrichment of E2F2 transcriptional targets within miR-31 over-expressing cells, as determined by GSEA.

Figure 3

Model for the synergistic interactions of miR-31, p16^INK4A, p14^ARF, p53, and E2F pathways in cancer. Gray lines with arrowheads denote stimulatory pathways, while black lines that end with a flat line are inhibitory pathways.

Figure 4

Gene expression signature of miR-31 overexpression is correlated with progression in advanced stage ovarian tumors. (A) The expression patterns of the miR-31 gene signature in a panel of 243 advanced stage human serous ovarian tumors from Tothill et al. Tumors are ordered by the “miR-31 signature t-score” which measures the similarity of the tumor profile with the miR-31-inducible profile. (B) Kaplan-Meier analysis comparing the differences in risk of disease relapse between tumors showing activation (red line, t-score>0 at P<0.05) of the miR-31 signature, tumors showing deactivation (blue line, t-score<0 and P<0.05) of the signature, and tumor showing intermediate patterns (yellow line). Log rank test evaluates whether there are significant differences between any of the three arms. Univariate Cox test evaluates the association of the miR-31 signature t-score with patient outcome, treating the coefficient as a continuous variable.

Figure 5

Overexpression of miR-31 inhibits cancer cell proliferation in p53 pathway-inactivated ovarian cancer cell lines. A) Expression patterns of E2F2, CDKN2A (p14^ARF/p16^INK4A gene), and well-known p53-inducible targets (e.g., p21 (CDKN1A)) in ovarian cancer cell lines and NOSE controls. The p53 and CDKN2A gene status for these cell lines as described in the literature is indicated. B) ATP quantitation-based CellTiter-Glo assay to examine the effect of miR-31 overexpression on proliferation of OVCAR8 cells (left panel), and Caspase 3/7 activity assay for effect of miR-31 on caspase-mediated apoptosis (right panel). C) MTS assay for cell lines SKOV3 and OVCA433. D) MTS assay for HEY and ATP quantitation-based CellTiter-Glo assay for OVSAYO. For each time point in parts B–D delineated by asterisk (“*”), differences are significant with P<0.05 (two-sided t-test). Error bars reflect results from three independent cultures.

Figure 6

Overexpression of miR-31 inhibits cancer cell proliferation in osteosarcoma and prostate cancer cell lines. An MTS assay was used to examine the effect of miR-31 overexpression on proliferation of U2OS cells (osteosarcoma, panel A) and PC3 cells (prostate, panel B).