Distinct Roles of m5C RNA Methyltransferase NSUN2 in Major Gynecologic Cancers

Abstract

RNA methylation has recently emerged as an important category of epigenetic modifications, which plays diverse physiopathological roles in various cancers. Recent studies have confirmed the presence of 5-methylcytosine (m⁵C) modification on mammalian mRNAs, mainly modified by NOP2/Sun RNA methyltransferase family member 2 (NSUN2), but little is known about the underlying functions of m⁵C. Gynecologic cancers are malignancies starting from women's reproductive organs. The prevalence of gynecologic cancers leads to a massive economic burden and public health concern. In this study, we investigated the potential biological functions of NSUN2 in common gynecologic cancers including cervical cancer, ovarian cancer, and endometrial cancer. Remarkably, distinct scenarios were found. The levels of NSUN2 did not show alteration in endometrial cancer, and in ovarian cancer, depletion of upregulated NSUN2 did not reduce carcinogenesis in cancer cells, suggesting that the upregulated NSUN2 might be an incidental effect. On the contrary, NSUN2 played a role in tumorigenesis of cervical cancer; depletion of upregulated NSUN2 notably inhibited migration and invasion of cancer cells, and only wild-type but not catalytically inactive NSUN2 rescued these malignant phenotypes of cancer cells. Mechanistically, NSUN2 promoted migration and invasion by leading to m⁵C methylation on keratin 13 (KRT13) transcripts, and methylated KRT13 transcripts would be recognized and stabilized by an m⁵C reader, Y-box binding protein 1 (YBX1). Collectively, these results not only displayed the nature of diversity among human malignancies, but also demonstrated a novel NSUN2-dependent m⁵C-YBX1-KRT13 oncogenic regulatory pathway.

Keywords: KRT13; NSUN2; YBX1; cervical cancer; endometrial cancer; m5C; ovarian cancer.

Figure 1

Bioinformatics analysis regarding expression status of NSUN2 gene in gynecological cancer. (A) Dot plot of the expression level of NSUN2 mRNA in uterine corpus endometrial cancer (UCEC) tissues from GSE17025 and GSE106191. (B) Dot plot of the expression level of NSUN2 in ovarian cancer (OV) tissues from GSE12470 and GSE66957. (C) Dot plot of the expression level of NSUN2 in cervical squamous cell carcinoma (CESC) tissues from GSE138080 and GSE36514. (D) Pan-cancer expression status of NSUN2 mRNA (normal vs. tumor) in TCGA database assessed by UALCAN website. (E) OncoPrint of c-BioPortal showing the proportions of each mutation type of NSUN2 in three gynecological cancers from TCGA samples. (F–H) Association of NSUN2 RNA expression with 5-year overall survival (OS) in microarray data in all UCEC patients. (F) OV patients (G) and CESC patients (H) determined using KM-plotter online survival analysis tool. (Data shown are mean ± SEM, n = 3, Student’s unpaired t-test was used for statistical analysis, the log-rank test was used for Kaplan–Meier curves. The HR was performed using the Cox model. **p < 0.01; ***p < 0.001).

Figure 2

Role of NSUN2 in cervical cancer. (A) The expression of NSUN2 mRNA in 20 cervical cancer tissues and 20 normal cervical tissues from clinical samples as assessed by RT-qPCR. (B) The expression of NSUN2 protein in 5 clinical samples with paired adjacent normal tissues in cervical cancer by Western blot. (C, D) Representative images (C) and quantification (D) of NSUN2 immunohistochemistry from tissue chip consisting of 29 normal cervix and 30 cervical cancer paraffin sections. (E, G) Validation of NSUN2 knockdown efficacy of in CaSki (E) and SiHa cells (G) by RT-qPCR and Western blot. (F, H) The effects of NSUN2 knockdown on migration and invasion as measured by transwell assays in CaSki (F) and SiHa (H) cells. Representative images (left panel) and quantification (right panel) of transwell assays showed the migration and invasion capability of CaSki and SiHa cells. (Data shown are mean ± SEM, n = 3; Student’s unpaired t-test was used for statistical analysis, ***p < 0.001).

Figure 3

NSUN2 promotes migration and invasion of cervical cancer cells in an m⁵C-dependent manner. (A, B) Effect of wild-type or mutant NSUN2 on cell migration in NSUN2 knockdown CaSki cells. Representative images (A) and quantification (B) of transwell assays showed the migration capability of CaSki cells. (C, D) Effect of wild-type or mutant NSUN2 on cell invasion in NSUN2 knockdown CaSki cells. Representative images (C) and quantification (D) of transwell assays showed the invasion capability of CaSki cells. (EV, empty vector; data shown are mean ± SEM, n = 3, Student’s unpaired t-test was used for statistical analysis, ***p < 0.001).

Figure 4

NSUN2 promotes cell migration and invasion in cervical cancer via KRT13. (A) Venn diagram showing 6 overlapping genes between RNA-BisSeq and RNA-Seq data. (B–E) RT-qPCR and Western blot showing KRT13 mRNA and protein level in NSUN2-knockdown CaSki (B, C) and SiHa cells (D, E). (F, G) Effect of KRT13 overexpression on cell migration (upper panel) and invasion (lower panel) in NSUN2 knockdown CaSki cells. Representative images (F) and quantification (G) of transwell assays showed the migration and invasion capability of CaSki. (H, I) Effect of KRT13 overexpression on cell migration (upper panel) and invasion (lower panel) in NSUN2 knockdown SiHa cells. Representative images (H) and quantification (I) of transwell assays showed the migration and invasion capability of SiHa. (EV, empty vector; data shown are mean ± SEM, n = 3, Student’s unpaired t-test for statistical analysis, ***p < 0.001).

Figure 5

KRT13 mRNA is enriched of m⁵C modification. (A) m⁵C modification on KRT13 mRNA analyzed by m⁵C-MeRIP-qPCR in CaSki cells. NAPRT1 was used as a positive control. (B–D) RIP analysis of NSUN2-KRT13 interaction (B) and m⁵C-MeRIP-qPCR detection of m⁵C level on KRT13 (C) along with RT-qPCR analysis of KRT13 mRNA level (D) in NSUN2 knockdown and control CaSki cells. GAPDH and Actin were used as a negative control. (E–G) RIP analysis of NSUN2-KRT13 interaction (E) and m⁵C-MeRIP-qPCR detection of m⁵C level on KRT13 (F) along with RT-qPCR analysis of KRT13 mRNA level (G) in wild-type and mutant NSUN2 overexpressed CaSki cells. (Data shown are mean ± SEM, n = 3; Student’s unpaired t-test was used for statistical analysis, ***p < 0.001).

Figure 6

NSUN2 stabilizes KRT13 mRNA by recruiting YBX1. (A) Effect of NSUN2 on the stability of KRT13 mRNA. CaSki cells transfected with two NSUN2-targeting shRNAs were treated with 10 μg/ml RNA synthesis inhibitor Actinomycin D (Actd) for indicated time. Level of KRT13 mRNA was analyzed by RT-qPCR. (B, C) Effect of YBX1 (B) and ALYREF (C) knockdown on KRT13 mRNA and protein level in CaSki cells as assessed by RT-qPCR and Western blot. (D, E) Effect of YBX1 (D) and ALYREF (E) overexpression on KRT13 mRNA and protein level in CaSki cells as assessed by RT-qPCR and Western blot. (F) RIP-qPCR analysis of YBX1-KRT13 interaction. IgG was used as a negative control. (G, H) Effect of YBX1 on m⁵C modification occurred on KRT13 mRNA. YBX1 was silenced (G) or overexpressed (H) in CaSki cells, and the m⁵C modification on KRT13 mRNA was measured by m⁵C-MeRIP-qPCR. NAPRT1 was used as a positive control. KRT13-Ctrl was used as a negative control. (I) A working model of the mechanism by which NSUN2 promotes cervical cancer migration and invasion. (Data shown are mean ± SEM, n = 3; Student’s unpaired t-test was used for statistical analysis, **p < 0.01, ***p < 0.001).