Common genetic variants in pre-microRNAs are associated with cervical cancer susceptibility in southern Chinese women

Cervical cancer is a commonly diagnosed cancer among females. Polymorphisms in pre-microRNAs have been demonstrated to play critical roles in cancer. However, the roles of pre-microRNA polymorphisms in the aetiology of cervical cancer have not been well documented. We genotyped eight pre-microRNA polymorphisms in 290 cervical cancer patients and 445 cancer-free female controls using quantitative polymerase chain reaction with TaqMan probes. To estimate the association between pre-microRNA polymorphisms and the risk of cervical cancer, an unconditional logistic regression model was used to calculate the odds ratio (OR) and 95% confidence interval (CI), adjusting for age, menopause, delivery, and abortion. We found that the pre-miR-137 rs1625579 T > G polymorphism was associated with a significant decrease in cervical cancer risk (TG/GG versus TT: adjusted OR (AOR) = 0.47, 95% CI = 0.27-0.81; TG versus TT: AOR = 0.56, 95% CI = 0.34-0.91). We also observed a significant association between the pre-miR-27a rs895819 T > C polymorphism and decreased cervical cancer risk (TC/CC versus TT: AOR = 0.65, 95% CI = 0.44-0.96). Stratified analysis further demonstrated that the pre-miR-137 rs1625579 T > C and pre-miR-27a rs895819 T > C polymorphisms significantly reduced the risk of cervical cancer susceptibility in patients younger than 49 years, those who experienced fewer abortions, and clinical stage I patients. Moreover, the pre-miR-137 rs1625579 T > G polymorphism showed protective effects in premenopausal women, squamous cell carcinoma patients, and patients with unclassified types of pathologies; the pre-miR-27a rs895819 T > C polymorphism was also associated with a decreased risk in patients older than 49 years, menopausal women, and women who had experienced vaginal pregnancies. The pre-miR-137 rs1625579 T > G and pre-miR-27a rs895819 T > C polymorphisms may provide protective effects against susceptibility to cervical cancer risk.


Introduction
Cervical cancer was the fourth leading cause of cancer death and the fourth most frequently diagnosed cancer among females worldwide in 2018 [1]. In China, 26,400 and 30,500 estimated deaths from cervical cancer were reported in 2013 and 2015, respectively [2,3]. High-risk strains of human papillomavirus (HPV) are now well established as the causative agents responsible for invasive cervical cancer and its precursor, cervical intraepithelial neoplasia [4]. HPV infection is strongly associated Ivyspring International Publisher with the risk of cervical cancer. Only a small number of women exhibit persistent infection with oncogenic HPV types that eventually lead to cervical cancer, and most HPV infections regress spontaneously [5]. Increasing evidence indicates that host genetic variations play critical roles in cervical carcinogenesis independent of HPV infection [6][7][8][9].
MicroRNAs (miRNAs) are a class of small regulatory RNAs that are processed from endogenous hairpin transcripts and function as regulators of gene expression in eukaryotes [10,11]. In the nucleus, miRNAs are transcribed from genomic DNA into long primary transcripts (pri-miRNAs) and can be cleaved into 60-70-nucleotide precursor miRNAs (pre-miRNAs) by nuclear Drosha [12,13]. Next, the pre-miRNAs are further processed into mature miRNAs of approximately 21-25 nucleotides in length [14,15]. Polymorphisms or mutations in the promoter or the miRNA sequence itself may result in changes in the structure or expression of miRNAs, which affect the expression of hundreds of target genes. Cancer susceptibility and prognosis cancer may be affected by polymorphisms in miRNAs [16]. Recent studies have shown that the genetic variants in pre-miRNAs are associated with the risks and clinical outcomes of many types of cancers [17][18][19][20], including cervical cancer [7,21]. In this study, we sought to identify common genetic polymorphisms of pre-miRNAs associated with the risk of cervical cancer susceptibility in southern Chinese women.

Patients and controls
In the present case-control study, 290 cervical cancer patients (mean ± standard deviation (SD), 50.10 ± 10.51) were enrolled at the Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University (WMU) from February 2007 to February 2017. The diagnosis of cervical cancer was confirmed in all cases by histological examination of tissues from biopsies or resected specimens. Cancer-free controls (n=535, mean ± SD, 57.84 ± 15.47) were recruited from the same hospital during routine physical examinations, and we excluded participants who had been diagnosed with malignant neoplasm or a family history of cancers. This study was approved by the Second Affiliated Hospital and Yuying Children's Hospital of WMU (Ethics Reference No: LCKY2019-274), and written informed consent was obtained from all patients.
The TIAN quick FFPE DNA Kit (Qiagen NV, Venlo, the Netherlands) was applied to extract genomic DNA from all patients from paraffinembedded tissues, while genomic DNA from the controls was extracted from peripheral blood specimens using the TIANamp Blood DNA Kit (TianGen-Biotech, Beijing, China) as we described previously [23,24]. A UV absorption spectrophotometer was used to detect DNA purity and concentrations.
Genotyping analysis was performed by real-time PCR with TaqMan PCR master mix and the ABI Prism 7900HT genetic detection system. The details of the analysis procedures have been previously described [25]. In addition, approximately 5% of the samples were randomly selected as positive and negative controls for assessing the accuracy of genotyping results.

Statistical analysis
The goodness-of-fit chi-square test was adopted to evaluate departure from Hardy-Weinberg equilibrium (HWE) for the selected polymorphisms in control subjects. The heterogeneity of the genotypes and ages between patients and controls was evaluated by using a 2-sided chi-square test. The association between pre-miRNA polymorphisms and cervical cancer risk was assessed with an unconditional logistic regression model, calculated as the crude and adjusted odds ratios (OR) and 95% confidence intervals (95% CI). Additionally, stratified analyses were performed by age, menopause, the number of vaginal pregnancies, the number of abortions, clinical stages and pathological stages. All statistical tests were carried out by using SAS software (Version 9.4; SAS Institute, Cary, NC, USA), with a two-sided P-value < 0.05 considering significant controls.

Characteristics of the study participants
In the present study, we enrolled 290 cervical cancer patients and 445 cancer-free controls. The frequency distributions of the selected demographic characteristics of the cases and controls are shown in Table 1. The incidence of abortion in the patients (43.5 %) was much higher than that in the control subjects (4.9%), implying that a high incidence of abortion was a critical risk factor for cervical cancer development in our study population. In addition, compared with the controls, the distribution of age in the cervical cancer cases was much younger. Among the 290 cases, 105 (36.2%) were postmenopausal women. However, there was no significant difference in the number of vaginal deliveries between the cases and controls. There were 208 patients (71.7%) who had squamous cell carcinomas; other patients who had adenocarcinoma or adenosquamous carcinoma or were unclassified were represented 28.3% of the study population. Among all 290 cases, 70.0% of patients were classified as stage I and 21.0%, 1.7%, and 0.7% as stages II, III, and IV, respectively, while 6.6% of the patients were classified as being in other stages (Table 1).

Association between selected pre-miRNA polymorphisms and cervical cancer risk
To investigate the association with cervical cancer risk of each of the eight selected pre-miRNA polymorphisms, the HWE test was first used to identify genotyping errors in the cancer-free controls. As shown in Table 2, most of the genotype distributions were in accordance with HWE (P values ranged from 0.215 to 0.953), except that the pre-miR-149 rs2292832 T > C and pre-miR-218 rs11134527 A > G polymorphisms exhibited significant deviation from HWE in the controls (P<0.001). Next, through allelic association analysis, we found that the pre-miR-137 rs1625579 T > G polymorphism was significantly associated with decreased cervical cancer risk (TG/GG versus TT: adjusted odds ratio (AOR) = 0.47, 95% CI = 0.27-0.81, P=0.007; TG versus TT: AOR = 0.56, 95% CI = 0.34-0.91, P=0.018). The pre-miR-27a rs895819 T > C polymorphism was also shown to significantly decrease cervical cancer susceptibility (TC/CC versus TT: AOR = 0.65, 95% CI = 0.44-0.96, P=0.030). These findings suggest that these two polymorphisms may be common pre-miRNA polymorphisms that influence the risk of cervical cancer susceptibility in southern Chinese women.

Stratified analysis
We further explored the association of the pre-miR-137 rs1625579 T > G and pre-miR-27a rs895819 T > C polymorphisms with cervical cancer susceptibility through stratified analysis. We found that the protective effects of the pre-miR-137 rs1625579 T > G and pre-miR-27a rs895819 T > C polymorphisms were significant in patients younger than 49 years, those exhibiting fewer abortions, and those in clinical stage I (Table 3). Moreover, the protective effect of the pre-miR-137 rs1625579 T > G polymorphism was also observed in premenopausal women, patients with squamous cell carcinoma, and patients with unclassified types of pathologies, while the pre-miR-27a rs895819 T > C polymorphism was also associated a low risk of cervical cancer susceptibility in patients older than 49 years, those who were menopausal, and those who had experienced vaginal pregnancies (Table 3).

Discussion
In the current case-control study, we explored the associations of eight common polymorphisms in pre-miRNAs with cervical cancer susceptibility in southern Chinese women. We found that the pre-miR-137 rs1625579 T>G and pre-miR-27a rs895819 T > C polymorphisms were significantly associated with decreased cervical cancer risk. Our findings suggest that the pre-miR-137 rs1625579 T>G and pre-miR-27a rs895819 T > C polymorphisms may play critical roles in the aetiology of cervical cancer.
Tumour-suppressive functions of miR-137 in cervical cancer have been recently reported [26]. Endogenous miR-137 expression is downregulated in both cervical cancer cell lines and tissues, and overexpression of miR-137 in cervical cancer cells substantially inhibits cell proliferation, migration and transplantation in nude mice [26]. In our study, the results showed that the pre-miR-137 rs1625579 T>G polymorphism exhibits a protective effect against cervical cancer, suggesting that the pre-miR-137 rs1625579 T>G polymorphism may present a positive correlation with miR-137 expression. Further studies are needed to confirm this hypothesis.
Remarkably, we also found that the pre-miR-27a rs895819 T>C polymorphism exerted a protective effect on cervical cancer risk. This finding is consistent with a previous study conducted in cervical cancer [7]. It is worth noting that our results demonstrated that the pre-miR-27a rs895819 T>C polymorphism exerted a protective effect against cervical cancer risk in dominant genetic model, while Xiong et al. found that the pre-miR-27a rs895819 T>C polymorphism played a protective role in a recessive genetic model but found no associations in other genetic models [7]. Although the participants of Xiong et al.' study and our study were both selected from among southern Chinese women, the sample size in the current study was larger than the sample size in the Xiong et al. study (including 103 cervical cancer patients and 417 cancer-free controls). Thus, the differences may result from the baseline characteristics and the larger number of samples.
Additionally, previous studies have suggested that the pre-miR-149 rs2292832 T > C polymorphism is associated with an increased risk of cervical cancer susceptibility [27] and that the pre-miR-218 rs11134527 A > G polymorphism is associated with a decreased risk of cervical cancer susceptibility [28,29]. However, our study did not identify a significant association of these two polymorphisms with cervical cancer risk. This inconsistency with previous studies may be due to the genotype distributions of the pre-miR-149 rs2292832 T > C and pre-miR-218 rs11134527 A > G polymorphisms significantly deviating from HWE in the controls. For the other pre-miRNA polymorphisms, no associations were found with the risk of cervical cancer susceptibility in the current study, so further studies are required for confirmation.  Although the current case-control study provides evidence that common pre-miRNA polymorphisms are associated with the risk of cervical cancer susceptibility in southern Chinese women, several limitations should be addressed. First, only eight common pre-miRNA polymorphisms were genotyped in the current study; hence, more common pre-miRNA polymorphisms should be investigated to fully illuminate the contribution of polymorphisms in pre-miRNAs to cervical cancer susceptibility. Second, the sample size in the current study was relatively small. Additionally, the selected samples came from a single hospital in the current hospital-based case-control study, which may result in selection bias or even a decreased or increased risk assessment. Third, in addition to polymorphisms, many that confounders may also influence susceptibility to cervical cancer, such as gene-gene interactions, gene-environment interactions, and the specific tumour pathologic classification, were not taken into consideration due to the lack of individual information. Fourth, although our findings suggested that the pre-miR-137 rs1625579 T > G and pre-miR-27a rs895819 T > C polymorphisms show a statistically significant association with the risk of cervical cancer, much more research is needed to confirm this result and to detect the potential related mechanisms and functions in the future.