miR-769-5p is associated with prostate cancer recurrence and modulates proliferation and apoptosis of cancer cells

Abstract

MicroRNAs (miRs) are relevant in biological processes, including human prostate cancer. In the present study, the role of miR-769-5p and its targets in prostate cancer were explored. Publicly available data on expression of genes, miRs and disease-free survival of patients with prostate cancer were analyzed along with RNAseq of transfected cell lines. miR-769-5p expression was inversely associated with patient survival and in vitro assays indicated that its inhibition reduced the proliferation and increased apoptosis of prostate cancer cells. miR-769-5p was revealed to target Rho GTPase activating protein 10 (ARHGAP10) and increased expression of ARHGAP10 in tumors was determined to be associated with a favorable prognosis regarding disease-free survival. Of note, ARHGAP10 is a purported tumor suppressor in ovarian cancer, where it inhibits cell division cycle 42 (CDC42) activity and increases apoptosis. Similar effects were observed in prostate cancer cells, where miR-769-5p inhibition increased ARHGAP10 and led to reduced CDC42 activity. Furthermore, miR-769-5p inhibition increased apoptosis, which was partly reversed by additional knockdown of ARHGAP10. These results suggested that miR-769-5p is an oncogene targeting ARHGAP10, which in turn is a candidate tumor suppressor in prostate cancer.

Keywords: Rho GTPase activating protein 10; miR-769-5p; prostate cancer recurrence.

Figure 1

Analyses of miR-769-5p expression and its impact on survival. (A) Kaplan-Meier survival analysis suggested that miR-769-5p expression was inversely associated with disease-free survival. Data were extracted from Taylor et al (5) ^*P<0.05 according to the log-rank test with n=107. The patients were stratified into groups of high and low expression of miR-769-5p by assigning the bottom 50% (n=54) to the ‘Low’ and the top 50% (n=53) to the ‘High’ group according to their rank of miR-769-5p expression. Low: 46 censored, 8 events. High: 34 censored, 19 events. (B) miR-769-5p expression was elevated in tumor vs. normal tissues among all cases. In an analysis of matched pairs (n=28), the difference in miR-769-5p expression was not significant. However, after dividing the patients into groups with low (gray arrow) or high expression (black arrow) in the tumors, the difference in miR-769-5p expression between tumor and adjacent non-cancerous tissue was significant in the latter group. (C) When subdivided into 3 groups, miR-769-5p expression was the highest in metastasis, followed by tumor and then normal tissues. For B and C, normalized miR-769-5p expression data were extracted from Taylor et al (5) and fold differences were calculated using 2^-∆∆Cq. (D) Analysis of all TCGA samples indicated that the difference in miR-769-5p expression between tumor and normal tissue did not reach statistical significance, but in matched pairs (n=52), miR-769-5p expression was significantly elevated in tumor vs. normal tissues. Normalized miR-769-5p expression data were extracted from TCGA and fold differences were calculated using 2^-∆∆Cq. ^*P<0.05, ^**P<0.01 according to an unpaired t-test with Welch's correction or Welch's ANOVA with Dunnett's test in C. miR, microRNA; TCGA, The Cancer Genome Atlas.

Figure 2

miR-769-5p expression in prostate cancer cells and scheme for identifying tumor suppressors targeted by the miR. (A) miR-769-5p expression in the six human prostate cell lines as assessed by quantitative PCR. (B) miR-769-5p expression in cell lines grouped by origin; comparison of 2 cell lines from normal and primary tumors (RWPE-1 and 22Rv1) vs. 4 cell lines from metastases (LNCaP, MDA PCa 2b, DU145 and PC-3). miR-769-5p expression fold differences between the 2 groups derived from the average miR expression in the 2 groups and calculated using the 2^-∆∆Cq method. Values are expressed as the mean ± standard deviation (n=3). ^*P<0.05 according to Welch's ANOVA with Dunnett's test in A and an unpaired t-test with Welch's correction in B. (C) The strengths of RNA sequencing, TargetScan, Oncomine cohorts and literature search were combined to identify potential tumor suppressors targeted by miR-769-5p. The 322 (DU145) and 280 (PC-3) candidate genes that overlapped with predicted TargetScan targets were compared to the genes differentially expressed in the 8 prostate cancer cohorts. The genes with a higher expression in normal tissue vs. tumor tissues in ≥2 cohorts were selected and searched using PubMed. miR, microRNA; TCGA, The Cancer Genome Atlas.

Figure 3

RT-qPCR analysis of predicted candidate targets of miR-769-5p. Top ranked candidate tumor suppressors were selected from the gene lists obtained from RNAseq data after transfection of DU145 and PC-3 cells with the miR-769-5p mimics. Their fold changes and P-values from RNAseq are listed in the embedded table. miR-769-5p inhibitor, negative control or miR-769-5p mimics were transfected into DU145 cells, followed by RT-qPCR to measure the expression of the predicted targets (n=3). Fold differences compared to the negative control miR group were calculated using the 2^-∆∆Cq method. ^*P<0.05, ^**P<0.01, ^***P<0.001 and ^****P<0.0001 according to Welch's ANOVA with Dunnett's test. AGPAT1, 1-acyl-sn-glycerol-3-phosphate acyltransferase α; KREMEN1, kremen protein 1; CCNDBP1, cyclin-D1-binding protein 1; ARHGAP10, Rho GTPase-activating protein 10; FHL3, four and a half LIM domains protein 3; KCTD11, potassium channel tetramerisation domain containing 11; miR, microRNA; RNAseq, RNA sequencing; RT-qPCR, reverse-transcription quantitative PCR.

Figure 4

ARHGAP10 is a target of miR-769-5p and a tumor suppressor. (A) ARHGAP10 is at the intersection of the predicted targets of miR-769-5p and recurrence-associated transcripts (P<0.05 for continuous variables in the Cox proportional hazards regression). ARHGAP10 is a tumor suppressor using hazard ratio ≤1 as a criterion. Predicted targets are from TargetScan and recurrence-associated transcripts are from Taylor et al (5). (B) Patients with low expression of ARHGAP10 had reduced disease-free survival compared to those with high expression. (C) Patients with high expression of miR-769-5p and low expression of ARHGAP10 exhibited reduced disease-free survival compared to all others, ^**P<0.01, ^***P<0.001 as per the log-rank test (n=140 in B and 105 in C. B- Low: 46 censored, 24 events. High: 58 censored, 12 events. C- miR-769-5p high/ARHGAP10 low: 19 censored, 14 events. All others: 61 censored, 11 events). (D) Inverse correlation between ARHGAP10 and miR-769-5p expression in prostate tumors (n=98) determined using Spearman correlation analysis; ^**P<0.01. Data were extracted from Taylor et al (5). ARHGAP10, Rho GTPase activating protein 10; miR, microRNA.

Figure 5

miR-769-5p targets ARHGAP10 and affects proliferation and apoptosis. (A) Sequence-verified WT 3'UTR, MUT1 3'UTR, MUT2 3'UTR or MUT1+2 3'UTR of ARHGAP10 was cloned into a luciferase reporter plasmid and transfected into DU145 human prostate cancer cells together with a Renilla plasmid for normalization and either negative control miR or miR-769-5p mimics. The Y-axis indicates the ratio of luciferase to Renilla, relative to the miRNA mimics, when the negative control was set at 1.0. (B) Reduced proliferation following inhibition of R-769-5p in three prostate cancer cell lines. (C) miR-769-5p inhibition significantly increases apoptosis in two human prostate cancer cell lines, as measured by caspase-3 activity, most noticeably in PC-3 cells. Values are expressed as the mean ± standard deviation (n=5 in A, 3 in B and 9 in C). ^*P<0.05, ^****P<0.0001 according to unpaired t-test with Welch's correction. UTR, untranslated region; ARHGAP10, Rho GTPase activating protein 10; miR/miRNA, microRNA; MUT, mutant; WT, wild-type; Neg, negative control.

Figure 6

CDC42 activity and the effect of double transfection of miR-769-5p inhibitor and ARHGAP10 siRNA on apoptosis. (A) CDC42 activity was inversely associated with ARHGAP10, consistent with the literature; CDC42 activity was lower with miR-769-5p inhibitor vs. control miR. Combined miR-769-5p inhibition and ARHGAP10 knockdown increased CDC42 activity vs. miR-769-5p inhibition alone; thus, ARHGAP10 knockdown reversed the effect of miR-769-5p inhibition. (B) miR-769-5p inhibition increased apoptosis, which was slightly inhibited by ARHGAP10 knockdown (however, this effect was not remarkable). Values are expressed as the mean ± standard deviation (n=6 in A and 7 in B). ^***P<0.001, ^****P<0.0001 according to Welch's ANOVA with Dunnett's test. Scr, scrambled; CDC, cell division cycle; ARHGAP10, Rho GTPase activating protein 10; miR, microRNA; siRNA, small interfering RNA; OD, optical density; Neg, negative control.