FSH induces EMT in ovarian cancer via ALKBH5-regulated Snail m6A demethylation

Abstract

Background: The therapeutic benefits of targeting follicle-stimulating hormone (FSH) receptor in treatment of ovarian cancer are significant, whereas the role of FSH in ovarian cancer progresses and the underlying mechanism remains to be developed. Methods: Tissue microarray of human ovarian cancer, tumor xenograft mouse model, and in vitro cell culture were used to investigate the role of FSH in ovarian carcinogenesis. siRNA, lentivirus and inhibitors were used to trigger the inactivation of genes, and plasmids were used to increase transcription of genes. Specifically, pathological characteristic was assessed by histology and immunohistochemistry (IHC), while signaling pathway was studied using western blot, quantitative RT-PCR, and immunofluorescence. Results: Histology and IHC of human normal ovarian and tumor tissue confirmed the association between FSH and Snail in ovarian cancer metastasis. Moreover, in epithelial ovarian cancer cells and xenograft mice, FSH was showed to promote epithelial mesenchymal transition (EMT) progress and metastasis of ovarian cancer via prolonging the half-life of Snail mRNA in a N6-methyladenine methylation (m6A) dependent manner, which was mechanistically through the CREB/ALKBH5 signaling pathway. Conclusions: These findings indicated that FSH induces EMT progression and ovarian cancer metastasis via CREB/ALKBH5/Snail pathway. Thus, this study provided new insight into the therapeutic strategy of ovarian cancer patients with high level of FSH.

Keywords: ALKBH5; Epithelial ovarian cancer; Epithelial-mesenchymal transition; Follicle-stimulating hormone; Snail.

Figure 1

FSH induced EMT in epithelial ovarian cancer cells via demethylase ALKBH5. (A, B) Immunofluorescence and bright filed images of SK-OV3 cells treated with FSH (75 mIU) or TGF-β (10 ng/μl) for 48 hours, scale bar = 20 µm. (C-H) SK-OV3 cells were treated with different concentrations of FSH for various time points. (C, D) Western blots for E-cadherin and N-cadherin. (E, F) Western blots for METTL3, METTL14, ALKBH5, FTO and YTHDF1/2/3. (G, H) Western blot quantification of ALKBH5 intensity measurements (n = 3), data was analyzed by Mann-Whitney U test. *P < 0.05, **P < 0.01 versus CON. (I) Western blots of E-cadherin, N-cadherin and ALKBH5 in SK-OV3 cells treated with FSH (75 mIU, 48 h) after transfection of ALKBH5 siRNA (siALKBH5). (J-L) SK-OV3 cells transfected with ALKBH5 plasmid followed by FSH treatment (75 mIU, 48 h), n = 3 - 4 per group. (J) Immunofluorescence of E-cadherin- and N-cadherin-stained SK-OV3 cells. Scale bar = 20 µm. (K, L) Wound healing and Transwell assays were performed in SK-OV3 cells, data was analyzed by one-way ANOVA test. *P < 0.05, **P < 0.01, ***P < 0.001 versus CON transfected with Vector.

Figure 2

Snail is a critical transcription factor in FSH-induced EMT. (A-D) SK-OV3 cells were treated with different concentrations of FSH for various time points. (A, B) Western blots for Snail, Slug, ZEB1/2, Twist and Vimentin. (C, D) Western blot quantification of Snail intensity measurements (n = 3), data was analyzed by Mann-Whitney U test. *P < 0.05 versus CON. (E-H) SK-OV3 cells were treated with FSH (75 mIU, 48 h) after transfection of scramble or Snail siRNA (siSnail). (E) Western blots of E-cadherin, N-cadherin and Snail in SK-OV3 cells. (F, G) Wound healing and Transwell assays were performed. Quantitative data of healing area (n = 3) and cell invasion (n = 5) were analyzed by one-way ANOVA test. ***P < 0.001 versus CON group transfected with scramble siRNA, ###P < 0.001 versus FSH group transfected with scramble siRNA. (H) Immunofluorescence of SK-OV3 cells stained with E-cadherin and N-cadherin. Scale bar = 20 µm.

Figure 3

m6A modification of Snail mRNA is essential for FSH-induced EMT via ALKBH5. (A) Snail promoter activity was measured by luciferase assay in 293T cells transfected with FSHR plasmid before PBS or FSH (75 mIU, 48 h) treatment (n = 5), data was analyzed for statistical difference by two-tailed unpaired t test, ns, no significant. (B) qRT-PCR of Snail mRNA in SK-OV3 cells with pretreatment of FSH (75 mIU, 48 h) and treatment of Actinomycin D (ActD) for various time points (n = 3). (C) N6-methyladenosine (m6A) content of total RNA in SK-OV3 cells with FSH treatment (75 mIU, 48 h), n = 3 per group, data was analyzed by two-tailed unpaired t test. **P < 0.01. (D) m6A RIP-qPCR analysis of Snail mRNA in SK-OV3 cells treated with FSH (75 mIU, 48 h), n = 5 per group, data was analyzed by two-tailed unpaired t test. *P < 0.05. (E) m6A RIP-qPCR analysis of Snail mRNA in SK-OV3 cells transfected with ALKBH5 siRNA (siALKBH5) (n = 3), data was analyzed by two-tailed unpaired t test. *P < 0.05. (F) qRT-PCR of Snail mRNA in SK-OV3 cells pretreated with Actinomycin D (ActD) for various time points (n = 7), data was analyzed by two-tailed unpaired t test, *P < 0.05, **P < 0.01, ***P < 0.001. (G) Scheme showed locations of the m6A motifs within the Snail mRNA, and A to T mutation within m6A consensus site in Snail mRNA. (H-K) SK-OV3 cells were transfected with empty vector (Vector), Snail-CDs (CDs) and Snail-mut (Mut) plasmid. Data was analyzed by two-tailed unpaired t test, **P < 0.01, ***P < 0.001. versus Vector. #P < 0.05, ###P < 0.001 versus CDs. (H) qRT-PCR of Snail mRNA. (I) qRT-PCR of E-cadherin mRNA. (J) Western blots for E-cadherin and Snail. (K) RT-PCR of Snail mRNA in SK-OV3 cells transfected with Snail-CDs (CDs) and Snail-mut (Mut) plasmid and treatment of Actinomycin D (ActD) for various time points (n = 3). (L) Western blots of E-cadherin, N-cadherin, and Snail in SK-OV3 cells treated with FSH (75 mIU, 48 h), after transfection of Snail-CDs or Snail-mut plasmid. (M) Western blots for E-cadherin, N-cadherin and Snail in SK-OV3 cells co-transfected with ALKBH5 plasmid and wild type Snail (Snail-CDs) or CDS region mutated (Snail-mut) plasmid.

Figure 4

FSH activates CREB to increase ALKBH5 expression. (A, B) Western blots of p-CREB and CREB in SK-OV3 cells treated with different concentrations of FSH for various time points. (C, D) Wound healing and invasion assays were performed on SK-OV3 cells with pretreatment of CREB inhibitor 666-15 (10 μM, 1 h), then with FSH treatment (75 mIU, 48 h), n = 4 per group, data was analyzed for statistical difference by one-way ANOVA test. *P < 0.05, **P < 0.01 versus CON+PBS, ###P < 0.001 versus FSH+PBS. (E) Western blots of ALKBH5, Snail, E-cadherin and N-cadherin expression in SK-OV3 cells with pretreatment of 666-15 (10 μM, 1 h) and treatment of FSH. (F) Immunofluorescence for CREB and ALKBH5 in SK-OV3 cells treated with FSH (75 mIU, 48 h). Scale bar = 20 µm. (G) ALKBH5 promoter activity was measured by luciferase assay in SK-OV3 cells transfected with Vector or CREB plasmid (n = 15), data was analyzed by two-tailed unpaired t test. ***P < 0.001.

Figure 5

ALKBH5 is essential for FSH-induced EMT and metastasis of ovarian cancer in mouse xenografts model. (A-F) SK-OV3 cells were transfected with a control (Lv-Empty) or shALKBH5 (Lv-shALKBH5) luciferase-tagged lentivirus. SK-OV3 cells were then implanted to bilateral ovariectomized (OVX) nude mice to develop a xenografts model for FSH treatment (3 IU/day per mouse for 14 weeks). (A) Body increasement was calculated from the body weight of 8 weeks (n = 4). (B) In vivo bioluminescence imaging was performed to monitor tumor burden, and the bioluminescent signal was quantified and analyzed by Mann-Whitney U test (n = 4). *P < 0.05 versus CON group with Lv-Empty transfection. #P < 0.05 versus FSH group with Lv-Empty transfection. (C-E) Tumors of the SK-OV3 tumor-bearing mice were excised, imaged and weighed (n = 4). Scale bar = 1 cm. Data was analyzed by Mann-Whitney U test. *P < 0.05, **P < 0.01 versus CON group with Lv-Empty transfection. #P < 0.05, ###P < 0.001 versus FSH group with Lv-Empty transfection. (F) SK-OV3 xenografts were sectioned and stained for E-cadherin, N-cadherin, ALKBH5 and Snail expression by immunohistochemistry. Scale bar = 200 µm.

Figure 6

FSHR and Snail expression are positively associated with ALKBH5 expression. (A) IHC of ALKBH5 and Snail on tissue microarray of ovarian tumor. Scale bar = 250 μm. (B) Correlation of ALKBH5 to Snail expression in ovarian tumor. Cramér's V was used for the association analysis. V = 0.711, P < 0.001. (C) IHC of Snail, ALKBH5 and FSHR in tissue sections of normal and tumor ovarian tissues. Scale bar = 200 μm. (D) Association between Snail and FSHR mRNA expression in The Cancer Genome Atlas Program (TCGA, https://tcga-data.nci.nih.gov/tcga/tcgaHome2.jsp). Cramér's V was used for the association analysis. (E) Association between ALKBH5 and FSHR mRNA expression in TCGA database. Cramér's V was used for the association analysis. (F) Western blots of FSHR, ALKBH5 and Snail protein in normal ovarian tissues (N) and ovarian tumors (T). (G) m6A content of total RNA in normal ovarian tissue and ovarian tumor (n = 5), data was analyzed by two-tailed unpaired t test. *P < 0.05.