Matrix metalloproteinase-8 rs11225395 polymorphism correlates with colorectal cancer risk and survival in a Chinese Han population: a case-control study

Matrix metalloproteinase-8 (MMP-8) is a gene associated with inflammation and prognosis in colorectal cancer (CRC). Here, we studied the link between the rs11225395 polymorphism of MMP-8 gene and CRC risk. We recruited 551 CRC cases and 623 controls from among a subpopulation of Han Chinese patients. Data found that this variant was connected to an increased risk of CRC (TT versus CC: OR, 1.76; 95%CI, 1.09–2.84; P = 0.021; T versus C: OR, 1.29; 95%CI, 1.07–1.56; P = 0.007). Stratified analyses indicated a positive association among smokers (TT versus CC: OR, 2.31; 95%CI, 1.12–4.79; P = 0.024), males, and patients ≥ 60 years old. Crossover analysis showed that the potential interaction between smoking or drinking and the MMP-8 rs11225395 polymorphism was related to elevated risk for CRC. The rs11225395 polymorphism was also connected with lymph node metastasis and TNM stage. Moreover, the CRC cases carrying a TT genotype of MMP-8 rs11225395 presented had poorer overall survival than the CC genotype carriers. These findings show that MMP-8 rs11225395 correlates with an elevated risk of CRC and poor patient prognosis in a subpopulation of the Han Chinese subpopulation. Thus, the MMP-8 rs11225395 polymorphism could potentially function as a biomarker predictive of CRC susceptibility.


INTRODUCTION
Colorectal cancer (CRC) is one of the most deadly cancers, causing about 900,000 cancer related deaths every year [1,2]. The number of CRC patients is projected to increase to more than 2.2 million by 2030 [3]. CRC ranks as the fifth and fourth most prevalent cancer among men and women in China, respectively [4]. The incidence and mortality among women are ~25% lower than those in men [1]. Nationwide screening programs, improved lifestyle and diet, and increased colonoscopy screenings have contributed to the decreasing trend in the incidence of this disease [5]. CRC patients show classical signs and symptoms including altered bowel habits, occult or overt rectal bleeding, anemia, or abdominal pain [6]. Unfortunately, CRC pathogenesis is poorly understood and only becomes symptomatic at advanced stages, with surgery currently being the most effective treatment. Genetic predisposition, unhealthy lifestyles, obesity, and other environmental factors such as smoking and drinking corelate with the development of CRC [1,6,7]. Genome wide association studies (GWASs) revealed a host of novel risk loci for CRC patients [8][9][10][11].
Matrix metalloproteinases (MMPs), zinc and calciumdependent proteolytic enzymes could degrade extracellular matrix proteins and components. Some MMPs are related with cancer progression, metastasis and invasion [12,13]. MMP-8 is an enzyme in the connective tissue that is primarily produced by neutrophils and that cleaves collagens, growth factors, cell adhesion proteins, and cytokines [14]. Increased MMP-8 levels in CRC correlated with disease progression and inflammation [15]. Böckelman et al. showed that MMP-8 levels served as prognostic AGING biomarker for CRC [16]. In addition, Sirniö et al. showed that high MMP-8 serum levels corelated with decreased survival in CRC [17]. Elevated MMP-8 levels were also observed in CRC patients who developed anastomotic leakage after surgery [18].
The MMP-8 gene is shown to locate on chromosome 11q22.2. Rs11225395 polymorphism is in the promoter region of MMP-8 gene. The T allele of the rs11225395 polymorphism resulted in increased protein expression of MMP-8 [19]. Several studies addressed the link between the rs11225395 polymorphism of MMP-8 gene and different cancer risk, but with conflicting findings [20][21][22][23][24][25][26][27][28][29][30][31]. Here, we aimed to explore the relationships between both CRC risk and survival prognosis and the MMP-8 rs11225395 polymorphism. Table 1 summarizes the clinicopathological characteristics and data of all included patients. The frequencies of gender, age, smoking and drinking showed no obvious differences between these two groups. In this study, 551 CRC patients and 623 controls were recruited. In terms of tumor site, CRC patients were divided into two groups: the rectum (66.9%) and colon (33.1%). Adenocarcinoma accounted for the most common pathology subtype (95.1%). Other relevant data are presented in Table 1.

Association of MMP-8 gene rs11225395 polymorphism with the risk of CRC
The relationship between this locus and CRC risk is presented in Table 2. The genotype distributions in control groups were in accordance with HWE test (P > 0.05). The minor allele frequency (MAF) of the rs11225395 polymorphism for CRC cases was 0.27. In addition, we summarized the MAF for different populations around the world in Figure 1, which was obtained from the 1000 Genomes repository [33]. The TT or CT+TT genotypes were shown to increase the susceptibility to CRC (TT vs. CC: OR, 1.76; 95%CI, 1.09-2.84; P = 0.021). The associations remained statistically significant even after adjusting for gender and age. Additionally, the T allele was shown to increase the risk of CRC.

Stratified analysis of the rs11225395 polymorphism and CRC risk
Stratified analyses of gender, age, smoking, BMI, ESR, CRP, and drinking were conducted to measure the connection between the rs11225395 polymorphism and the risk of CRC (Table 3). We observed an increased risk for CRC patients among smokers, males, and those individuals aged ≥ 60 years. However, no positive findings were reported in the stratified analysis of drinking, BMI, ESR, and CRP.

Combined and interactive effects of the rs11225395 polymorphism and drinking or smoking on CRC risk
Due to the association of the rs11225395 polymorphism with environmental factors observed in Table 3, we next used cross-over analysis to further evaluate the impact of the interactions between the MMP-8 gene rs11225395 polymorphism and environmental factors on CRC susceptibility. Smokers with CT or TT genotype increased the risk of CRC when compared with non-smokers with CC genotype (Table 4). Similarly, CRC patients who carried the CT genotype and drank alcohol presented a higher risk of developing CRC than individuals who did not carry the CC genotype and drink alcohol (drinking + CT vs. nondrinking +CC: OR, 1.44, 95% CI, 1.02-2.05; P = 0.040). In summary, the data showed potential associations between the rs11225395 polymorphism and contributions from smoking or drinking to an increased risk of developing CRC.

MMP-8 gene rs11225395 polymorphism and clinicopathological characteristics of CRC patients
Next, we assessed whether the MMP-8 gene rs11225395 polymorphism was connected to CRC clinicopathologic features ( Table 5). The CRC patients with TT genotype were linked with TNM III+IV stage and lymph node metastasis. No positive association was observed for histological grade, tumor size, family history of CRC, histology, and location of CRC.

Survival analysis of CRC patients with the MMP-8 gene rs11225395 polymorphism
Last, Kaplan-Meier single factor analysis showed that CRC patients with TT genotype exhibited worse overall AGING  Figure 4).  rs11225395 polymorphism elevated the risk of breast cancer in a subpopulation of Han Chinese [30]. Other Taiwanese studies demonstrated no associations between the MMP-8 gene rs11225395 polymorphism and childhood leukemia [25], lung cancer [27], oral cancer [22], and bladder cancer risk [29]. Hashim et al. suggested that MMP-8 gene rs11225395 polymorphism was a protective factor associated with nasopharyngeal carcinoma susceptibility in a Malaysian population [24], and Arechavaleta-Velasco et al. detected that this locus increased the risk of ovarian cancer in Mexican women [31]. All such studies yielded conflicting findings regarding the association with this variant and cancer risk. There may be many reasons that may explain such contradictory findings: One, diverse exposure factors and dietary habits; two, different populations have genetic heterogeneity; three, the sample size among the aforementioned studies differed; four, the studies exhibited clinical heterogeneity among different cancers. To address these inconsistent results, a metaanalysis by Feng et al. to explore the connection between this SNP and cancer susceptibility, and found no association of the rs11225395 polymorphism and overall cancer risk [36]. However, elevated cancer risk was observed in non-Asian populations, and no relationship was detected in Asian populations [36]. Up to date, no studies investigated the link between CRC risk and the MMP-8 rs11225395 polymorphism. Therefore, we conducted this study to explore such association. Our study revealed that the MMP-8 rs11225395 polymorphism elevated the risk of CRC.

DISCUSSION
In addition, we uncovered an increased risk for developing CRC in males, smokers, and those subjects aged ≥ 60 years. However, Qiu et al. did not find any associations between hepatocellular carcinoma susceptibility and rs11225395 polymorphism when the analyses were stratified by age, gender, and drinking and smoking statuses [26]. Cross-over analysis showed that the integrated effects of the MMP-8 rs11225395 polymorphism and smoking or drinking were related to an increased risk of CRC. Our data also showed that the MMP-8 rs11225395 polymorphism related to TNM III+IV stage and lymph node metastasis among CRC patients.
Last, we explored the association between CRC prognosis and the MMP-8 rs11225395 polymorphism. We found that patients with TT genotype showed worse overall survival compared with patients with CC

Subjects
We

Genotyping
We collected 2 ml of peripheral blood from all CRC patients and matched controls, which were stored at -80° C until used. We used the TIANamp Blood DNA kit to obtain DNA (Tiangen Biotech, Beijing, China). Genotyping of the studied SNP was analyzed by RFLP-PCR. The primers used for this polymorphism were as following: GCCAGAGACTCAAGTGGGAGACTACC ATGCAGATC (forward) and reverse primer TTATGATTGCCCAGACATTTG (reverse). Approximately 10% of all enrolled individuals were re-genotyped, and the concordance was 100% [32,33].

Genotype and gene expression correlation analysis
The data from the GTEx Portal database (https://www.gtexportal.org/home/) were utilized to evaluate the link between genotypes of the MMP-8 rs11225395 polymorphism and mRNA expression-level alteration [34].

Statistical analysis
We used student's t-test and chi-square (χ 2 ) test to evaluate continuous and categorical variables, respectively. The Hardy-Weinberg equilibrium (HWE) test among controls was calculated by a goodness-of-fit χ 2 test. Logistic regression analysis was utilized to estimate the odds ratios (ORs) and their 95% confidence intervals (CIs). The Kaplan-Meier method was used to calculate overall survival [35]. P value < 0.05 indicated a statistical difference. All statistical analyses were addressed by SPSS 22.0 (SPSS Inc., Chicago, USA).

Availability of data and materials
The data can be made available by the corresponding author upon reasonable request.

AUTHORS CONTRIBUTIONS
Zhenhua Kang conceived of the study and participated in its design. Di Sun and Xu Wang conducted a systematic literature review. Zhenhua Kang performed data analyses. Jiandong Tai drafted the manuscript. All authors gave final approval and agreed to be accountable for all aspects of the work, ensuring integrity and accuracy.