Status of Gene Methylation and Polymorphism in Different Courses of Ulcerative Colitis and Their Comparison with Sporadic Colorectal Cancer

Abstract

Background

The objective of this study is to explore the common genetic and epigenetic mechanism of ulcerative colitis (UC) and sporadic colorectal cancer (SCRC) by observing genes methylation level and single nucleotide polymorphisms (SNPs) of different disease courses in UC and SCRC.

Methods

Two hundred subjects were enrolled, including 40 in the healthy control (HC) group, 50 in the short disease course UC group (SUC), 52 in the long disease course UC group (LUC), and 58 in the SCRC group. Methylation-specific polymerase chain reaction was used to detect the methylation of MINT1 and cyclooxygenase 2 (COX-2) gene. Single nucleotide polymorphisms of interleukin (IL)-23R rs10889677 and IL-1β rs1143627 were detected by Sanger sequencing.

Results

Compared with HCs (32.5%), methylation level of MINT1 was significantly increased in SCRC (67.2%; P = 0.001) and was a risk factor for CRC (odds ratio, [OR] 4.26). The methylation ratios of COX-2 were 95.0%, 58.0%, 23.1%, and 24.1% in HC, SUC, LUC, and SCRC, respectively, which were negatively correlated with the disease course of UC (r = −0.290). Hypermethylation of COX-2 was a protective factor for SUC (OR, 0.11), LUC (OR, 0.02), and SCRC (OR, 0.03; P < 0.05). Compared with HCs, rs10889677 allele A was a risk factor for SUC and LUC, and rs1143627 allele T was a protective factor for SUC and LUC. Genotype TT was a protective factor for SUC.

Conclusion

The hypomethylation of COX-2 gene was a common risk factor and epigenetic modification for UC and SCRC, which might be one of the mechanisms through which UC patients were susceptible to CRC. The hypermethylation of MINT1 was a risk factor for SCRC but not for UC; alleles of IL-23Rrs10889677 and IL-1βrs1143627 were related to UC but not to SCRC.

INTRODUCTION

Chronic inflammation can promote the occurrence and development of tumors,¹ which is reflected by the fact that the incidence of colorectal cancer (CRC) in ulcerative colitis (UC) patients is higher than healthy people, and the incidence rate of CRC increases as the duration of UC prolongs.² Therefore, researchers have been focusing on clarifying the mechanisms by which UC develops and increases the risk of cancer. The abnormal methylation of oncogenes and tumor suppressor genes is part of the pathogenesis of malignant tumors.³

Many inflammatory factors and their gene polymorphisms are also related to the pathogenesis and progression of UC and CRC.^{4, 5} Methylated in tumor 1 is a hypermethylated gene (MINT1) in CRC tissue. Some studies have shown that the chromosome localization of MINT1 gene was also associated with inflammatory bowel disease (IBD).⁶ The yclooxygenase 2 gene (COX-2) is not only related to inflammation but also participates in tumorigenesis, progression, and metastasis.¹ Currently, there have been studies on the association of MINT1 and COX-2 gene methylation with UC and CRC, but these studies still lack adequate data and exact conclusions.

Interleukin (IL)-23R and IL-1 are closely related to colorectal cancer; ^7–9 meanwhile, IL-23R is also a UC-related gene.¹⁰ However, studies on the relevance of gene polymorphisms of IL-23R, IL-1 with UC and CRC susceptibility are still needed.

Some studies have investigated the risk factors for UC progressing to UC-related CRC (UC-CRC). For example, Garrity-Park and colleagues^{11, 12} detected and compared the methylation and polymorphism status of target genes in the corresponding nonadjacent and nondysplastic mucosa of UC and UC-CRC and found there were significant differences between them that could predict the potential risk of UC-CRC in UC patients. UC progressed to UC-CRC due to the persistence of chronic inflammation. In addition to the factor of chronic inflammation, is there another factor, such as a common genetic and epigenetic basis that both UC and SCRC may have, which leads to a higher incidence of CRC in UC patients than in healthy people? From this perspective, our study observed the status of gene polymorphism and the change of methylation rate with the disease course of UC of target genes MINT1, COX-2, IL-23R rs10889677, and IL-1βrs1143627 and compared them with SCRC, so as to explore the possible genetic and epigenetic mechanisms for higher incidence of CRC in UC than in healthy people and for the increased incidence of CRC as the course of UC prolongs.

MATERIAL AND METHODS

Subjects

Patients who were diagnosed with UC or SCRC by histopathological results of endoscopic biopsies or surgical resected tissues from 2017 to 2019 in the Department of Gastroenterology at the Affiliated Hospital of Qingdao University were enrolled in this study. Fifty patients with a disease course of no more than 1 year were classified into the short disease course of UC (SUC) group. Fifty-two patients with a disease course of more than 1 year (average course: 5.77 ± 2.09) were sorted into the long disease course of UC (LUC) group. The severity of ulcerative colitis was assessed by endoscopic Mayo score. All UC patients in this study were treated with mesalazine, and all SCRC patients had no history of radiotherapy and chemotherapy before surgery. Fifty-eight patients were diagnosed with SCRC, and we also randomly selected 40 healthy individuals who had undergone colonoscopy with negative findings as healthy controls (HCs). The genomic DNA was extracted from the colonic or SCRC tissue specimens, and relevant clinical data of the patients were collected (Table 1).

Exclusion criteria included 1) familial or hereditary colorectal adenoma or tumor; 2) tumors of other systems or organs or severe liver and kidney dysfunction; 3) history of radiation therapy or chemotherapy before surgery; 4) pregnant women; 5) and secondary colorectal cancer.

DNA Extraction

The tissue samples were collected in a clean environment and stored at −80 °C. The samples were ground to powder with liquid nitrogen, then the genomic DNA was extracted using Tiangen Genome Extraction Kit (DP304, Beijing, China), and the concentration and purity of the DNA were accurately measured using SpectraMax QuickDrop Micro-Volume (Molecular Devices, Silicon Valley, CA, USA).

Polymerase Chain Reaction Primers Design

The primers were synthesized by Ruiboxingke Biotechnology Co., Ltd. (Beijing, China) according to the reference gene sequences of COX-2, MINT1, IL-23R, and IL-1β in the National Center for Biotechnology Information database. Detailed information is shown in Table 2.

Target Gene Methylation Status Detection

Qualified DNA was treated with bisulfite according to the instruction of EZ DNA Methylation-Gold Kit (Zymo Research, Orange, CA, USA), and its concentration was determined. The target gene was amplified by methylation-specific (MSP) polymerase chain reaction (PCR). After staining with nucleic acid dyes and agarose gel electrophoresis, the DNA were analyzed under a Vilber Lourmat gel imager to evaluate the methylation status of the MINT1 and COX-2 genes. If a band occurred after amplification of the target gene by methylation primers, the target gene was considered positive for methylation and vice versa. The typical results of electrophoresis are shown in Figures 1 and 2.

Determination of Single Nucleotide Polymorphism (SNP) of Target Genes

Polymerase chain reaction was performed to amplify target genes. After nucleic acid dye staining and agarose gel electrophoresis, the target genes were sequenced to detect the IL-23Rrs10889677 and IL-1βrs1143627 polymorphisms. The major genotypes and alleles of rs10889677 and rs1143627 were retrieved from the dbSNP database of NCBI. The major genotype CC and allele C of rs10889677 and the major genotype CC and allele C of rs1143627 were set as baselines for the comparison analysis.

Statistical Analysis

Categorical variables were tested by χ ² test or Fisher exact test, and continuous variables were analysed using t test. Logistic regression analysis was used to analyze the relevance between genetic methylation, polymorphisms, and disease risk. Spearman rank correlation analysis was used to analyze the correlation between gene methylation status and different disease courses of UC. Bonferroni method was applied for post hoc comparisons to control the probability of type 1 error (P = 0.05/6 = 0.008). P < 0.05 indicates statistical significance for other variables. SPSS19.0 (SPSS Inc., Chicago, USA) was used for all the statistical analysis.

ETHICAL CONSIDERATIONS

The research plan was approved by the ethics committee of affiliated hospital of Qingdao university. Each participant signed an informed consent form.

RESULTS

Baseline Characteristics of Subjects

There was no statistical difference in gender, age, sampling location, or severity of UC among the groups of subjects (P > 0.05), as is shown in Table 1. The genotype distribution frequency of IL-23R rs10889677 and IL-1βrs1143627 in the HC group was in accordance with Hardy-Weinberg equilibrium (P > 0.05), which was representative of the population. The methylation status of COX-2 and MINT1 Gene, and genotype and allele distribution frequency of IL-23R rs10889677 and IL-1β rs1143627 in different groups were shown in Table 3.

Relationship Between Gene Methylation Status and Diseases

MINT1 gene

The χ ² test showed the methylation level of MINT1 gene in SCRC group was significantly higher than that of the HC group (P = 0.001, Table 4). Compared with HCs (odds ratio [OR], 4.26; 95% CI, 1.80–10.07; P = 0.001), the SUC group (OR, 2.61; 95% CI, 1.19–5.71; P = 0.016), and the LUC group (OR, 2.40; 95% CI, 1.11–5.19; P = 0.027), logistic regression analysis indicated that MINT1 gene hypermethylation was a risk factor for SCRC (Table 6). No statistical difference was found in the gene methylation level among other groups. MINT1 methylation level was not correlated to the disease course of UC by Spearman rank correlation test (r = 0.070, P = 0.486), which suggested that the methylation level of MINT1 gene did not change as UC occurred and progressed (Table 5).

COX-2 gene

The methylation rate of COX-2 in the SUC (P = 0.000), LUC (P = 0.000), and SCRC (P = 0.000) groups was lower than in the HC group; the LUC group was lower than the SUC group (P = 0.000); the SCRC group was lower than the SUC group (P = 0.000; Table 4). Therefore, the methylation status of the COX-2 gene appeared to change in the early stage of UC and was further reduced with the extension of disease course. There was no statistical difference in COX-2 gene methylation level between the LUC and SCRC groups (P = 0.896), indicating that methylation of LUC had reached the same level as SCRC.

Compared with the HC group, COX-2 gene hypermethylation was a protective factor in the SUC group (OR, 0.11; 95% CI, 0.03–0.41; P = 0.001), LUC group (OR, 0.02; 95% CI, 0.01–0.09; P = 0.000), and SCRC group (OR, 0.03; 95% CI, 0.01–0.10; P = 0.000). Compared with the SUC group, COX-2 gene hypermethylation was a protective factor for the LUC group (OR, 0.22; 95% CI, 0.09–0.51; P = 0.000). Additionally, hypermethylation of COX-2 gene was also a protective factor for SUC against progression to SCRC (OR, 0.23; 95% CI, 0.10–0.53; P = 0.000). There was no statistical difference between the SCRC group and LUC group by logistic regression analysis (OR, 1.06; 95% CI, 0.44–2.56; P = 0.896; Table 6), which was consistent with the results by χ ² test(Table 4).

Spearman rank correlation test indicated that the COX-2 gene methylation level was negatively correlated with the disease duration of UC (r = −0.290; P = 0.003; Table 5); that is, during the process from healthy colorectal mucosa to the development of UC, the methylation ratio of COX-2 gene decreased. The result was also consistent with that of χ ² test and logistic regression analysis.

Relationship Between Alleles and Genotypes of IL-23R rs10889677 and IL-1β rs1143627 With Diseases

IL-23R rs10889677

Compared with HCs, allele A was a risk factor for SUC (OR, 3.13; 95% CI, 1.51–6.50; P = 0.002) and LUC (OR, 2.61; 95% CI, 1.30–5.23; P = 0.007) by logistic regression analysis. In SUC and LUC, allele A increased the risk for UC by 2.13 and 1.61 times than allele C, respectively. There were no significant differences between the SCRC group and other groups. Thus, allele A was a risk factor for UC but not for SCRC. Genotypes AA and CA had no correlation with SUC, LUC, and SCRC (P > 0.05; Table 7)

IL-1β rs1143627

Compared with the HC group, allele T was a protective factor in the SUC (OR, 0.50; 95% CI, 0.28–0.92; P = 0.026) and LUC (OR, 0.51; 95% CI, 0.28–0.92; P = 0.025) groups by logistic regression analysis, but no statistical difference between any other 2 groups was found. As a result, the allele T was a protective factor for SUC and LUC but not for SCRC. Compared with the HC group, TT genotype was a protective factor for SUC (OR, 0.28; 95% CI, 0.08–0.96; P = 0.043). The TT genotype was a risk factor for SCRC compared with SUC (OR, 4.80; 95% CI, 1.25–18.42; P = 0.022) and LUC (OR, 4.57; 95% CI, 1.14–18.41; P = 0.033). Genotype CT had no correlation with SUC, LUC, and SCRC (P > 0.05; Table 7).

DISCUSSION

It has been verified by a large number of studies that chronic inflammation promotes tumorigenesis and development.^{13, 14} The pathogenesis of many types of tumors also strongly supports this notion, such as Correa’s cascade from chronic gastritis caused by Helicobacter pylori to gastric cancer, and chronic hepatitis caused by hepatitis B virus can eventually lead to liver cancer. The formation and development of inflammation-related tumors is a complex process. A milieu rich in growth factors, inflammatory cells, and DNA damage factors promotes normal cells to proliferate, invade, and metastasize. Chronic inflammatory cells and mediators function as promoters for tumor development. In addition to the factor of chronic inflammation, is there another factor such as a common genetic and epigenetic basis that both UC and sporadic CRC may have that leads to a higher incidence of CRC in UC patients than healthy people? We explored this issue.

Both hypomethylation of oncogenes and hypermethylation of tumor suppressor genes occur in the early stages of tumorigenesis, and they can coexist in the same tumor. The hypermethylation of tumor suppressor genes is relatively more important, and the hypermethylation of CpG island is considered to be a mechanism for inactivation of tumor suppressor genes. It is generally believed that genome-wide hypomethylation can activate oncogenes, whereas hypermethylation mainly occurs in tumor suppressor genes, which inhibits the expression of genes and eventually accelerates tumorigenesis.

Toyato et al isolated 33 hypermethylated gene sequences using methylated CpG island amplification in CRC tissues and named them methylated in tumor (MINT)1–33,¹⁵ among which MINT1 was one of the methylation sites of CRC-specific tumor suppressor gene promoter. Previous studies have found that the position of MINT1 on chromosome was also related to inflammatory bowel disease.⁶ The methylation level of MINT1 in UC-associated colorectal cancer (UC-CRC) was significantly higher than that in UC patients; ¹¹ additionally, MINT1 was also significantly methylated in UC-CRC tissues and adjacent or distant tissue from cancer. Some studies showed that there was no difference in the methylation ratio of MINT1gene between UC-CRC and SCRC.¹⁶The univariate and multivariate analysis indicated that MINT1 was significantly associated with UC-CRC, and the methylation level of MINT1 was significantly higher in UC-CRC than in other tumors.^{12, 17} Some studies demonstrated that elevated levels of MINT1 methylation were associated with gastric cancer and gingival squamous cell carcinoma. Our study shows that the methylation level of MINT1 gene increased significantly in the SCRC group, which was also a risk factor for SCRC but not for UC.

Macrophages can induce the production of COX-2 by the NF-κB signaling pathway through the secretion of IL-1β.¹⁸ The hypomethylation of the COX-2 gene promoter region can promote the expression of COX-2.^{19, 20} COX-2 plays an important role in the process of inflammation and tumorigenesis and is involved in the formation and development of various types of tumors.¹⁸ The COX-2 gene is highly expressed in colorectal cancer, gastric cancer, esophageal cancer, breast cancer, prostate cancer, and nonsmall cell lung cance, and participates in tumor development.^21–23 The methylation level of COX-2 gene strongly affects the expression of COX-2 gene and COX-2 protein, and so does its pro-inflammatory and cancer promoting effects. That was proved by a previous study, which also showed that the methylation level of COX-2 gene in UC-CRC was significantly lower than that of UC patients.¹¹ Another study found that in addition to the lower level of COX-2 gene methylation in UC-CRC tissue than in UC, the level was also lower in nontumor colonic mucosa in UC-CRC patients than in UC patients and healthy people. Notably, unmethylated COX-2 gene significantly increased the risk for cancer, suggesting that overexpression of COX-2 may be a risk factor for the procession from UC to UC-CRC.^{12, 17}The present study shows that from HC to different stages of UC, the methylation level of COX-2 gene gradually decreased and was negatively correlated with the disease course duration. The methylation level of the COX-2 gene in the LUC group had already decreased to the same level as the CRC group.

COX-2 hypomethylation was a risk factor for both UC and SCRC. These results suggested that COX-2 hypomethylation was a common epigenetic basis for UC and SCRC and a possible mechanism for the susceptibility of UC, pariticularly long-course UC patients to get CRC.

Interleukin-23 plays a crucial part in inflammatory response, autoimmune diseases, and tumors. The role of IL-23 in inflammatory bowel disease and inflammation-related colorectal cancer has also received widespread attention. The main function of IL-23R is to mediate the IL-23 signaling pathway transduction in cells. The IL-23-IL-23R pathway is an important part in inflammatory response. It is also a potential pathway to promote the development of malignant tumors. Genetic polymorphisms of IL-23R have also been linked to certain inflammatory diseases, autoimmune diseases, and malignancies. Duerr et al found that specific variation in the IL-23R gene (rs11209026, c.1142G>A, p.Arg381Gln) was proved to be a strong protective factor against Crohn’s disease, and noncoding IL-23R variants were also a protective factor.⁹ Silverberg et al²⁴ identified 2 polymorphic sites of the IL-23R gene, rs10889677, and rs11209026, which were related to IBD. But some studies indicated that IL-23R rs10889677 had nothing to do with the risk of IBD.²⁵ It has been demonstrated that IL-23R rs10889677 was also related to cancer. A study found that the distribution of the IL-23R rs10889677 genotype and allele frequency in Chinese Han women with ovarian cancer was significantly different from that of female healthy controls. The frequency of the C allele of rs10889677 increased significantly in ovarian cancer patients, and Chinese women who carried the IL-23R rs10889677 CC + CA genotype were at higher risk of developing ovarian cancer and even more malignant endometrial cancer.²⁶ However, there was still research reporting that IL-23R rs10889677and rs1209032 had no correlation with the risk of colorectal cancer.²⁷

This study found that the polymorphism of IL-23R rs10889677 was related to the development of UC. Allele A of IL-23R gene was a risk factor for SUC and LUC but not for SCRC. The results were consistent with some previous studies^{9, 27} but not with the result in Iranian population.²⁵ The conflicting results might be due to the different races, geographical environments, and sample sizes in different studies.

Interleukin-1β has pro-inflammatory properties, which are central mediators of inflammation and involved in the pathogenesis of chronic inflammatory diseases including IBD. Additionally, IL-1β is also expressed in most inflammation-related tumors²⁸ and can enhance tumor aggressiveness,²⁹ thereby promoting tumorigenesis, progression and metastasis. Studies reported that IL-1β activity was related to genetic polymorphism, which showed different distribution frequencies in different races. It is indicated³⁰ that the IL-1β gene polymorphism determined the disease duration and severity of IBD and could be the mechanism to explain the heterogeneity of IBD. There have been some studies on the relationship among IL-1β rs1143627, inflammation, and tumors. Results show that the white population carrying rs1143627genotype CC had a higher risk of gastritis, whereas the genotype of IL-1β rs1143627 had no correlation with the risk for gastritis in Asian individuals.³¹The genotype CT of IL-1β rs1143627could lower the risk for hepatocellular carcinoma³² and prostate cancer.³³ No correlation was found between IL-1β rs1143627 and nonsmall cell lung cancer.³⁴ These results suggest that IL-1β rs1143627 may play different roles in various inflammatory diseases and tumors in different races.

The results of our study indicate that the allele T of IL-1β rs1143627 was a protective factor for SUC and LUC but not for SCRC. Genotype TT is a protective factor for SU but not for LUC. We then merged SUC and LUC groups as one and found the genotype TT was still a protective factor for UC (OR, 0.281; 95% CI, 0.094–0.842; P = 0.023) by logistic regression analysis. Meanwhile, our results also showed that allele T had nothing to do with SCRC, but genotype TT was a risk factor for SCRC compared with the UC group. We speculated that those results might be attributed to grouping and sample size, which need to be determined by further research.

CONCLUSION

The hypomethylation of COX-2 gene was a common risk factor and epigenetic modification for UC and SCRC, which might be one of the mechanisms through which UC patients are susceptible to CRC. The hypermethylation of MINT1 was a risk factor for SCRC but not for UC; allele status of IL-23R rs10889677 and IL-1β rs1143627 were related to UC but had nothing to do with SCRC, suggesting the hypermethylation of MINT1 gene and allele status of the 2 genes might not participate in the process from UC to CRC.

Supported by: This work was supported by the Health Commission of Shandong Province (No. 2016WS0262) and (No. 2016WS0680).

Conflicts of interest: All authors declare no conflict of interest.

Author Contribution: CL collected experimental specimens, conducted experiments, performed the statistical analysis, and revised the paper. ZYY interpreted the data, wrote the article, and helped in conception and design of the research. HY designed the research, guided and supervised the experiment, interpreted the data, and critically revised the article. KXW, BC, KYR, MJC, JHL, HXC, and YWP helped in sample collection and experiment and revised the article. All authors read and approved the final manuscript.

REFERENCES

Author notes