Association study between the polymorphisms of angiogenesis‐related genes and cervical cancer susceptibility in Chinese Uygur population

Abstract Background Cervical cancer is the second most common malignant tumor in women, and its invasion and metastasis are regulated by tumor angiogenic growth factors and their cognate receptors. In this study, we explored the relationship between genetic polymorphisms of angiogenesis‐related genes (VEGF‐C, VEGFR‐2, and VEGFR‐3) and the risk of cervical cancer in Chinese Uygur population. Methods We investigated four single‐nucleotide polymorphisms (SNPs) in 342 cervical cancer cases and 498 controls to evaluate their association with the risk of cervical cancer. Their correlations were evaluated by chi‐squared test, Fisher's exact test, t test, and genetic model analyses. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated using unconditional logistic regression. Results We observed that rs12646659 in VEGF‐C was associated with a lower cervical cancer risk in allele, dominant, and log‐additive models (allele: p = .017; dominant: p = .018; log‐additive: p = .018). For the individuals older than 43, rs4604006 (VEGF‐C) was related to an increased cervical cancer risk under codominant model (p = .035), and rs12646659 was significantly associated with a reduced cervical cancer risk in allele, dominant, log‐additive models (allele: p = .028; codominant: p = .037; log‐additive: p = .037) However, there were no significant correlation of rs1000611 (VEGFR‐2) and rs1195571 (VEGFR‐3) with cervical cancer risk in Chinese Uygur population. Conclusion Our study firstly provided evidence that rs4604006 and rs12646659 of VEGF‐C gene were related to the susceptibility of cervical cancer in Chinese Uygur population.


| INTRODUCTION
Cervical cancer is one of the common malignant tumor in women, with 528,000 cases and 266,000 deaths in 2012 (Fang et al., 2017;Niu et al., 2019). The high incidence and mortality of cervical cancer among Uygur women in Xinjiang has become the most important health issue (Lin, Huang, Shen, & Yiming, 2015;Ma, Hong, Lu, Chen, & Ma, 2015). In recent years, the incidence of cervical cancer among women in Xinjiang is on the rise. The chances of the second cure for cervical cancer are very low, and the early detection has a significant impact on the survival of cervical cancer patients (Waggoner, 2003). Increasing growth factors and their homologous receptors have been reported and can regulate the invasion and metastasis of cervical cancer (Tomao et al., 2014). Therefore, the research of these growth factors and their receptors is undoubtedly a great benefit for the treatment of cervical cancer.
Angiogenesis is a pivotal step in tumor formation, growth and metastasis (Wu, 2014). VEGF (vascular endothelial growth factor) is a key angiogenic stimulator, and VEGF signaling pathway has been identified as an important part of angiogenesis (Li, Xu, Gao, Bi, & Huo, 2018). VEGF-C (Vascular Endothelial Growth Factor C; OMIM: 601,528) is an important member of the VEGF family. It encodes proteins that affect angiogenesis, endothelial cell growth, and vascular permeability (Chen et al., 2014). Recent studies have shown that elevated levels of vascular VEGF-C have in many invasive tumors and it is strongly associated with poor prognosis in cancer patients (Cheng, Jiang, Yuan, Liu, & Simoncini, 2018). For example, the level of VEGF-C increases in women with lung carcinoma, and it is significantly associated with lymph node metastasis (Tamura & Ohta, 2003). The high expression of VEGF-C was also observed in cervical cancer (Mitsuhashi et al., 2005). At the same time, VEGF binds to the extracellular receptor domain and promotes the activation of tyrosine kinase in the intracellular receptor domain, thereby phosphorylating tyrosine residues and activating several intracellular signaling pathways (Robinson & Stringer, 2001).
In this case-control study, we genotyped four SNPs (rs10006115 [VEGFR-2], rs4604006 [VEGF-C], rs12646659 [VEGF-C], and rs11955717 ) and performed a comprehensive association analysis to identify whether SNPs were associated with cervical cancer risk in Chinese Uygur population.

| Ethics statement
The study was approved by the ethics committee of People's Hospital of Xinjiang Uygur Autonomous Region, in accordance with the principles of the Helsinki Declaration. Each participant was informed of the procedures and purpose of our research and signed a written informed consent before donating 5 ml venous blood for further analyses.

| Research participates
In this case-control study, 342 cervical cancer patients (mean age, 43.27 ± 11.78 years) were recruited from the Department of Gynecology, People's Hospital of Xinjiang Uygur Autonomous Region, between 2016 and 2019. All patients were diagnosed with cervical cancer by histopathological examination according to the International Federation of Gynecology and Obstetrics (FIGO; Du, Wang, Richards, & Wang, 2019).
The healthy control group consisted of 498 individuals (43.46 ± 13.03 years), who were recruited from the Health Examination Center of the People's Hospital of Xinjiang Uygur Autonomous Region during the same time with cases. And these control subjects were comparable to the cervical cancer subjects in terms of age and race. The selection criteria for the control group were as follows: no history of cancer or a family history of cancer, no known history of infectious HPV. Written informed consent was obtained from each participant prior to enrollment in the study.

| SNP selection and genotyping
Genomic DNA was extracted from participant's peripheral venous blood by a Gold Mag Mini Whole Blood Genomic DNA Purification Kit (GoldMag Ltd) following the manufacturer's protocol and then stored at −80°C before genotyping. The concentration and purity of DNAs were determined by the NanoDrop 2000 (Thermo Scientific).
We established the following criteria to identify the target SNPs: (a) MAF (minor allele frequency) of Han Chinese in Beijing (HCB) > 0.05 and disease relevance in 1,000 genome (http://www.inter natio nalge nome.org/); (b) a linkage disequilibrium value of r 2 < .8 for each target SNPs. Agena MassARRAY Assay Design 4.0 software was used to design the primers for amplification and extension reactions. Agena MassARRAY RS1000 was used to perform SNP genotyping according to the standard protocol. Two staffs independently operated genotyping assay and randomly selected more than 10% samples for verification, and the results were exactly same in two sets of assays. Then, Agena Typer 4.0 software was applied to analyze and manage our data. The PCR primers for each SNP are shown in Table 1.

| Statistical analysis
All statistical analysis was done with SPSS version 19.0 software (SPSS) and Microsoft Excel. All analyses were two sided, and statistical significance was set at p ＜ .05. SNP genotype frequencies in the case and controls were calculated by chi-squared test (Hu, Wang, Hu, & Li, 2018;Yang et al., 2018). Deviation from Hardy-Weinberg equilibrium (HWE) was assessed using the chi-squared test to compare the observed and expected genotype frequencies among the control subjects. Logistic regression analysis was used to examine the odds ratios (ORs) and 95% confidence intervals (95% CIs) in order to assess the association between SNPs and cervical cancer (Bland & Altman, 2000;Liu et al., 2017). Four models (codominant, dominant, recessive, and log-additive) were used to test the association between SNPs and Cervical cancer (Jin et al., 2016;Sole, Guino, Valls, & Iniesta, 2006).

| Demographic characteristics
The general characteristics were listed in Table 2. Among the 840 participants, 342 were patients with cervical cancer and 498 were healthy controls. The mean age and standard deviation were 43.27 ± 11.78 for cases and 43.46 ± 13.03 for control subjects. There were no significant differences between the cases and controls in terms of age.

SNPs alleles
Basic information containing SNP ID, alleles, role, MAF distribution, p-HWE value, ORs, 95% CIs of all candidate SNPs were presented in Table 3. The call rate for all SNPs was above 95% in cases and controls, which was considered as high quality to perform association analyses. None of the candidate SNPs significantly deviated from HWE. OR = 1 indicates that the factor had no effect on the disease; OR > 1 means it is a risk factor; and OR < 1 means it is a protective factor. The comparison of allele distributions between the cervical cancer patients and the control subjects with the χ 2test revealed that there was a statistical correlation between the rs12646659 polymorphism of VEGF-C and the risk reduction of cervical cancer (OR = 0.41, 95% CI = 0.20-0.87; p = .017).

| Associations between genotype frequencies and cervical cancer risk
Furthermore, we analyzed the association between the SNPs and the risk of cervical cancer under multiple inheritance models (codominant, dominant, recessive, log-additive models; Table 4). Our analyses showed that rs12646659 in VEGF-C gene was correlated with a decrease the risk of cervical cancer in the dominant model (OR = 0.40, 95% CI = 0.19-0.86, p = .018 for the "C/G-G/G" genotype) and log-additive model (OR = 0.40, 95% CI = 0.19-0.86, p = .018) before and after adjustment for age, respectively. In addition, we failed to find any significant association between other polymorphisms and the risk of cervical cancer.

| Stratification analysis by age
To further explore the potential effect of age on the relationship of VEGF-C, VEGFR-2, and VEGFR-3 gene polymorphisms with the susceptibility to cervical cancer, we performed the same statistical analysis on the recruited population after stratification of age (Table 5). Among the individuals older than 43, rs4604006 of VEGF-C gene was correlated with a 1.55-fold increased the risk of cervical cancer in the codominant model (adjusted, OR = 1.55, 95%  CI = 1.03-2.34, p = .035 for the "C/T" genotype). While rs1264665 of VEGF-C was related to a decreased the risk of cervical cancer under the allele model (OR = 0.27, 95% CI, 0.08-0.94; p = .028 for the "C" allele) and dominant model (adjusted, OR = 0.26, 95% CI = 0.08-0.92, p = .037 for the "C/G-G/G" genotype). The variant rs1264665 was also observed to decrease the risk of cervical cancer in the logadditive model (adjusted, OR = 0.26, 95% CI = 0.08-0.92, p = .037). However, no significant association between candidate polymorphisms and cervical cancer risk was found in populations at age ≤43.

| DISCUSSION
In this hospital-based case-control study, we investigated the association of four important polymorphisms (rs10006115, rs4604006, rs12646659, and rs11955717) with the risk of cervical cancer in Chinese Uygur population, and we observed a significant association between the VEGF-C rs12646659 polymorphism and the risk of cervical cancer. The presence of the VEGF-C rs12646659 conferred a lower risk of developing cervical cancer. Further stratified analysis revealed that rs4604006 of VEGF-C gene was related to a higher risk of cervical cancer at age >43, while rs12646659 of VEGF-C was associated with a lower risk of cervical cancer at age >43. Our present study is the first to provide substantial basic evidence that gene polymorphisms in VEGF-C were corrected with cervical cancer risk in Chinese Uygur population.
Due to tumor growth and metastasis require the formation of new blood vessels, blocking tumor angiogenesis can be used to treat cancer. Meanwhile, genes related to angiogenesis also become potential target molecules for tumor treatment (Hajari Taheri et al., 2019). VEGF family proteins are specific and potent angiogenic factors that increase vessel permeability, endothelial cell growth, proliferation, migration, and differentiation (Keck et al., 1989). VEGF-C, a member of the VEGF family, is a protein-coding gene (Olofsson et al., 1996). VEGF-C associated with Diseases, including Lymphedema, Hereditary, and Lymphedema. Recently, several reports have confirmed the correlation between VEGF-C expression in tumor tissue specimens and lymph node metastasis. In cervical cancer, VEGF-C mRNA expression in tumor tissue samples has been shown to be associated with lymph node metastasis (Niki et al., 2000). VEGF-C tissue status also was an important independent factor for poor prognosis, and serum VEGF and VEGF-C levels can also be used as biomarkers for cervical SCC (Mitsuhashi et al., 2005). In addition, VEGF-C can binds to receptor Flt-4 and promotes angiogenesis and/or lymphangiogenesis, thus accelerating cancer metastasis Lohela, Bry, Tammela, & Alitalo, 2009 p-values were calculated by unconditional logistic regression analysis with adjustments for age.
T A B L E 5 (Continued) protein expression (Liu, Cheng, He, & Yao, 2014). VEGF-C can also reduce the expression of miR-326 and increase the expression of cortactin through c-Src signaling, leading to enhanced cervical cancer invasiveness (Cheng et al., 2018).
VEGF binds to the receptor tyrosine kinase (VEGFR) by transphosphatase, and then activates VEGFR. VEGFR-2 and VEGFR-3 are belongs to the tyrosine kinase receptor family (Masabumi, 2006). VEGFR-3 is activated by two proteins (VEGF-C and VEGF-D), and plays an essential role in the morphogenesis of the lymphatic vessel network during embryonic development, being involved in formation of new lymphatic vessels in the life. Binding of VEGF-C to VEGFR-3 is responsible for mostly biological effects of VEGFR-3 (Olsson, Dimberg, Kreuger, & Claesson-Welsh, 2006). Some studies have reported that VEGFR-2 may also interfere with lymphangiogenesis by binding VEGF-C and VEGF-D, which was crucial for the normal process of vasculogenesis during embryonic development (Ferrara, 2004;Vokes & Krieg, 2015). VEGFR-2 and VEGFR-3 are involved in normal and pathological angiogenesis through different mechanisms, such as: activation of MAPK extracellular signal-regulated kinases (ERK1/2) through the PKC and Ras pathways (Shibuya, 2011; important pathways in cell proliferation), as well as the PI3K Akt/PKB pathway (involved especially in survival of lymphatic endothelial cells; Olsson et al., 2006). Moreover, it activates certain integrins, which disrupt cell to cell cohesion and initiate cellular migration (Takahashi & Shibuya, 2005).
In this study, we explored the relationship of VEGF-C, VEGFR-2, and VEGFR-3 polymorphisms with cervical cancer risk. There were significant associations between their polymorphisms and the risk of various diseases (hepatocellular carcinoma, oral cancer, gastric cancer, lymphedema, etc.; Chien et al., 2013;Debrah et al., 2017;Hsieh et al., 2014;Li, Yu, et al., 2018), except cervical cancer. We found that the rs4604006 of VEGF-C gene was associated with an increased risk of cervical cancer, while rs12646659 was significantly associated with a decreased risk of cervical cancer for the first time. Meanwhile, there are few reports focused on the polymorphisms of rs4604006 and rs12646659. Only one study aimed to assess the role of VEGF and VEGFR polymorphisms in clinical outcomes of HCC patients receiving sorafenib therapy, and it found that rs4604006 (VEGF-C) T allele was significantly associated with patients' progressionfree survival and overall survival (Scartozzi et al., 2014). In addition, rs4604006 and rs12646659 were located in the intron region of VEGF-C gene and may be involved in mRNA processing, and hence regulate posttranscriptional modification, protein translation, or promoter/enhancer cluster activity (Rose, 2008;Zhang et al., 2018). Besides, we will expand the sample size to verify our results and conduct further functional studies so as to provide more evidence for the effect of VEGF-C polymorphism on cervical cancer risk.
Inevitably, this study had some limitations that should be accounted for when interpreting the results. First of all, the inherent selection bias and information bias were inevitable problems, because all participants were recruited from the identical hospitals. Second, the number of cases in our study was limited and our study population was all Chinese Uygur population, which cannot preclude false-negative results and extrapolated to other populations. Hence, larger prospective studies are necessary to fully elucidate the role of these polymorphisms in cervical cancer. Despite these limitations, our current findings provide scientific evidence of VEGFR-2, VEGF-C, and VEGFR-3 with the risk of cervical cancer in the future studies.

| CONCLUSION
To sum up, our study firstly provided evidence that the variants of VEGF-C gene had a significant effect on the risk of cervical cancer in Chinese Uygur population, especially individuals older than 43. These results may contribute to further elucidate the potential role of VEGF-C in cervical cancer susceptibility among Chinese Uygur population.