Novel role of lncRNA CHRF in cisplatin resistance of ovarian cancer is mediated by miR-10b induced EMT and STAT3 signaling

Abstract

Ovarian Cancer (OC) is a highly lethal gynecological cancer which often progresses through acquired resistance against the administered therapy. Cisplatin is a common therapeutic for the treatment of OC patients and therefore it is critical to understand the mechanisms of resistance against this drug. We studied a paired cell line consisting of parental and cisplatin resistant (CR) derivative ES2 OC cells, and found a number of dysregulated lncRNAs, with CHRF being the most significantly upregulated lncRNA in CR ES2 cells. The findings corroborated in human patient samples and CHRF was significantly elevated in OC patients with resistant disease. CHRF was also found to be elevated in patients with liver metastasis. miR-10b was found to be mechanistically involved in CHRF mediated cisplatin resistance. It induced resistance in not only ES2 but also OVCAR and SKOV3 OC cells. Induction of epithelial-to-mesenchymal-transition (EMT) and activation of STAT3 signaling were determined to be the mechanisms underlying the CHRF-miR-10b axis-mediated cisplatin resistance. Down-regulation of CHRF reversed EMT, STAT3 activation and the resulting cisplatin resistance, which could be attenuated by miR-10b. The results were also validated in an in vivo cisplatin resistance model wherein CR cells were associated with increased tumor burden, CHRF downregulation associated with decreased tumor burden and miR-10b again attenuated the CHRF downregulation effects. Our results support a novel role of lncRNA CHRF in cisplatin resistance of OC and establish CHRF-miR-10b signaling as a putative therapeutic target for sensitizing resistant OC cells.

Figure 1

Cisplatin Resistant (CR) ES2 cells and their lncRNA expression profile. (A) Proliferation of parental vs CR ES2 cells was measured by MTT assay as described in Methods after their culture in cisplatin containing media (indicated concentrations of cisplatin on X-axis) for 96 h. (B) Expression of various lncRNAs in CR ES2 cells as quantitated by quantitative RT-PCR, relative to the parental ES2 cells (control). For each lncRNA, the expression in control parental cells was set to ‘1’ and the relative expression in CR ES2 cells is shown. *p < 0.01, relative to control.

Figure 2

lncRNA validation in human samples. (A) The levels of top 3 lncRNAs were quantitated by quantitative RT-PCR in ovarian tumor samples from patients with diagnosed cisplatin resistance (n = 8). The levels shown are relative to levels in adjacent non-tumor tissue. (B) Levels of CHRF were evaluated in OC patients with liver metastasis (n = 6) by quantitative RT-PCR and compared to those in OC patients without liver metastasis (n = 6). *p < 0.01, relative to ‘no metastasis’.

Figure 3

miRNAs in cisplatin resistant cells and human samples. Expression of various (A) tumor suppressor and (B) oncogenic miRNAs in CR ES2 cells as quantitated by by quantitative RT-PCR, relative to the parental ES2 cells (control). For each miRNA, the expression in control parental cells was set to ‘1’ and the relative expression in CR ES2 cells is shown. *p < 0.01, relative to control. (C) The levels of top 3 miRNAs were quantitated in ovarian tumor samples from patients with diagnosed cisplatin resistance (n = 8) by quantitative RT-PCR. The levels shown are relative to levels in adjacent non-tumor tissue. (D) Levels of miR-10b were evaluated in OC patients with liver metastasis (n = 6) by quantitative RT-PCR and compared to those in OC patients without liver metastasis (n = 6). *p < 0.01, relative to ‘no metastasis’.

Figure 4

CHRF-miR-10b axis in multiple OC cell lines. (A) miR-10b induces cisplatin resistance. miR-10b was transfected in 3 different OC cell lines (ES2, OVCAR3, SKOV3) and effect of increasing doses of cisplatin was evaluated by MTT assay after 96 h of cisplatin treatment. (B) Expression of lncRNA CHRF in different OC cells as quantitated by quantitative RT-PCR, relative to the respective parental cells (control). The expression of CHRF in control parental cells was set to ‘1’ and the relative expression in miR-10b transfected cells is shown. (C) Expression of miR-10b in different OC cells as quantitated by quantitative RT-PCR, relative to the respective parental cells (control). The expression of miR-10b in control parental cells was set to ‘1’ and the relative expression in cells with down-regulated CHRF is shown. (D) miR-10b attenuates the effects of CHRF down-regulation in cisplatin resistant (CR) ES2 cells. CR ES2 cells (CR-control) and CR ES2 cells with down-regulated CHRF (without and with miR-10b transfections) were subjected to 96 h of cisplatin treatment (increasing doses, as indicated on X-axis) and then the proliferation evaluated by MTT assay. *p < 0.01, relative to control, ^#p < 0.01, relative to CR-CHRF-Down.

Figure 5

EMT and STAT3 phosphorylation in CR cells and role of CHRF-miR-10b. EMT markers E-cadherin and Vimentin were quantitated using quantitative RT-PCR in (A) CR and (B) miR-10b transfected ES2 cells, relative to parental cells (control). For each EMT marker, the expression in control parental cells was set to ‘1’ and the relative expression in CR/miR-10b transfected ES2 cells is shown. (C) miR-10b attenuates the effects of CHRF down-regulation on EMT markers in cisplatin resistant ES2 cells. CR ES2 cells (control) and CR ES2 cells with down-regulated CHRF (without and with miR-10b transfections: CHRF-Down and CHRF-down + miR-10b, respectively) were subjected to quantitative RT-PCR for the evaluation of EMT markers E-cadherin and Vimentin. The expression levels of EMT markers in control group were set as ‘1’ and the relative expressions in other groups are reported. STAT3 phosphorylation was quantitated using ELISA in (D) CR and (E) miR-10b transfected ES2 cells, relative to parental cells (control), as described in “Materials and methods” (F) miR-10b attenuates the effects of CHRF down-regulation on STAT3 phosphorylation in cisplatin resistant ES2 cells. CR ES2 cells (control) and CR ES2 cells with down-regulated CHRF (without and with miR-10b transfections: CHRF-Down and CHRF-Down + miR-10b, respectively) were subjected to ELISA for the evaluation of STAT3 phosphorylation. *p < 0.01, relative to control, ^#p < 0.01, relative to CHRF-Down.

Figure 6

Cisplatin resistance model in vivo and the effects of CHRF-miR-10b axis. (A) ES2 cells xenografts were studied in mice in vivo, as described in “Materials and methods”. Mice (n = 10/group) with control ES2 xenografts or CR-ES2 xenografts were challenged with a single dose of 3.0 mg/kg cisplatin at 14 day time point. A control group was administered sub-optimum 0.75 mg/kg dose at 7 day time point to develop resistance against cisplatin. *p < 0.01, relative to control. (B) The growth of tumor in CR-ES2 xenografts was further compared to that in CR-ES2 with down-regulated CHRF xenografts (without and with miR-10b: CR-CHRF Down and CR-CHRF-Down + miR-10b, respectively). n = 10 mice per group. *p < 0.01, relative to CR, ^#p < 0.01, relative to CR-CHRF-Down.