Abstract
Background/Aim: In gastric cancer (GCa) tissues, the mRNA and protein levels of tissue inhibitor of metalloproteinase-1 (TIMP-1) are significantly elevated compared to adjacent non-cancerous tissues. Moreover, the abnormal up-regulation of TIMP-1 has been associated with a poor prognosis. However, the role of TIMP-1 genotypes in susceptibility to GCa has seldom been investigated. This study aimed to evaluate the influence of TIMP-1 genotypes on GCa susceptibility and their potential interactions with clinico-pathological factors, including age, sex, body mass index, smoking, alcohol consumption, Helicobacter pylori (H. pylori) infection, and metastasis status. Materials and Methods: TIMP-1 rs4898, rs6609533, and rs2070584 genotypes were analyzed in 161 patients with GCa and 483 non-cancer control subjects from a Taiwanese population using PCR-RFLP methodology and direct sequencing. Results: The genotypic (p for trend=0.1987) and allelic (p=0.0733) frequencies of TIMP-1 rs4898 did not differ significantly between GCa cases and controls. Under the dominant model, combined CT+CC genotypes were not associated with GCa risk [odds ratio (OR)=0.74, 95% confidence interval (95%CI)=0.51-1.07, p=0.1272]. Similarly, no significant association was found for TIMP-1 rs6609533 or rs2070584 polymorphisms. Importantly, patients with GCa carrying the TIMP-1 rs4898 TT genotype exhibited a significantly enhanced risk of GCa when they had smoking (p=0.0140) and alcohol drinking habits (p=0.0011). Furthermore, the CC genotype of TIMP-1 rs4898 was linked to a lower risk of distant metastasis. Conclusion: The TIMP-1 rs4898 CC genotype may serve as a prognostic biomarker and could inform lifestyle modifications aimed at GCa prevention. Validation of TIMP-1 genotypic profile in diverse populations is warranted.
- Gastric cancer
- genotype
- single nucleotide polymorphism
- Taiwan
- tissue inhibitor of metalloproteinase-1 (TIMP-1)
Gastric cancer (GCa) is the fifth most common malignancy and the third leading cause of cancer-related deaths worldwide (1, 2). Despite recent advances, GCa continues to pose a major global public health challenge due to its complex pathogenesis, poor prognosis, and the absence of precise predictive markers (3, 4). The incidence and mortality rates of GCa are particularly high in Asia, especially in Eastern Asia, which accounts for more than 60% of recent cases (5, 6). Various risk factors have been identified, including Helicobacter pylori (H. pylori) infection (7, 8), smoking (9, 10), alcohol drinking (11, 12), obesity (13, 14), and high salt intake (15, 16). Additionally, the interaction between environmental risk factors and genetic predispositions plays a critical but still poorly understood role in the multifactorial etiology of GCa (17-19). Nevertheless, the identification of genetic markers, whether diagnostic or prognostic, for assessing GCa susceptibility remains largely unexplored.
Tissue inhibitor of metalloproteinase-1 (TIMP-1) is a glycoprotein with a molecular weight of 28 kDa (20). TIMP-1 functions as an inhibitory protein that modulates the activity of matrix metalloproteinases (MMPs) as well as disintegrin-metalloproteinases (21, 22). By regulating MMP expression, TIMP-1 plays a pivotal role in maintaining extracellular matrix (ECM) integrity (23, 24). Elevated TIMP-1 levels have been documented in various tumor types, including oral cancer (25), laryngeal cancer (26), thyroid cancer (27), colorectal cancer (28), breast cancer (29), skin cancer (30), renal cell carcinoma (31), ovarian cancer (32), in addition to prostate cancer (33), and hepatocellular carcinoma in animal models (34). Regarding GCa, TIMP-1 over-expression in tumor tissues was first reported as early as 1996 (35) and 1998 (36). Subsequent studies identified TIMP-1 as a potent prognostic marker associated with poorer survival outcomes (37, 38), and higher recurrence rates (39). In 2009, Mroczko et al. further demonstrated that elevated TIMP-1 levels in GCa tissues correlated with advanced tumor stages and reduced two-year survival rates (40). In 2006, Wang and his team found that serum TIMP-1 levels were significantly higher in patients with GCa than in healthy controls, with elevated serum TIMP-1 associated with lower survival rates (41). This was the first study to examine serum TIMP-1 expression in GCa. In 2011, Kemik and his team expanded on these findings, showing that high serum TIMP-1 levels were indicative of larger tumor size, greater depth of wall invasion, lymph node metastasis, liver metastasis, perineural invasion, and advanced pathological stage, but were not associated with age, sex, tumor location, or histological type (42). Despite numerous studies highlighting TIMP-1 over-expression in GCa tissues and serum, the contribution of TIMP-1 genotypes to GCa risk remains largely unexplored.
This study aimed at achieving two primary objectives. The first is to characterize the genotypic distribution of TIMP-1 rs4898, rs6609533, and rs2070584 in a well-defined Taiwanese cohort consisting of 161 patients with GCa and 483 cancer-free controls. The second objective is to investigate how TIMP-1 genotypes interact with age, sex, body mass index (BMI), smoking, alcohol consumption, H. pylori infection, and metastasis status on influencing GCa susceptibility.
Materials and Methods
Recruiting of GCa cases and non-cancer controls. A hospital-based cohort of 161 patients with GCa was recruited from the general surgery outpatient clinics at China Medical University Hospital (CMUH), as documented in previous studies (43-45). Each participant voluntarily provided a 5 ml sample of peripheral blood. For comparison, 483 healthy, cancer-free individuals with matched age and sex were selected from the CMUH Health Examination Cohort. The study design and protocols were approved by the hospital’s Institutional Review Board (IRB number: DMR100-IRB-107), and written informed consent was obtained with the assistance of the Tissue Bank of CMUH. Table I provides a summary of the demographic characteristics of the study population, including age, sex, BMI, smoking and alcohol consumption habits, H. pylori infection status, and histological classifications.
Selected characteristics of the control and gastric cancer groups.
Methodology of TIMP-1 genotype identification. Genomic DNA was extracted from the peripheral blood leukocytes of each participant using the QIAamp Blood Mini Kit (Blossom, Taipei, Taiwan, ROC) and processed according to methods described in our previous publications (46, 47). The polymerase chain reaction (PCR) was conducted under the following conditions: an initial denaturation at 94°C for 5 min; followed by 35 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s; with a final extension at 72°C for 10 min. The sequences of the forward and reverse primers, as well as the restriction enzymes used for each SNP, were listed in Table II. Genotyping was independently conducted by two translational researchers in a blinded manner. For quality control, 5% of the total samples for TIMP-1 rs6609533 were randomly selected for direct sequencing, and the results showed 100% concordance with the PCR-RFLP findings. Additionally, direct sequencing results for rs4898 and rs2070584 were fully concordant between the forward and reverse primers.
The summary of primer sequences, restriction enzyme and DNA fragment sizes for tissue inhibitor of metalloproteinase-1 (TIMP-1) rs4898, rs6609533 and rs2070584 polymorphic sites.
Statistical analyses. The unpaired Student’s t-test was employed to compare the ages between the GCa case and the control group. Pearson’s chi-square test, with Yates’ correction, was utilized to assess the distributions of sex, personal habits, BMI, and the various TIMP-1 genotypes and alleles across the subgroups. The association between different TIMP-1 genotypes and GCa risk was evaluated by calculating odds ratios (ORs) with corresponding 95% confidence intervals (CIs). A p-value of less than 0.05 was considered to indicate statistical significance.
Results
Table I presents the demographic characteristics of the 161 patients with GCa and 483 non-cancer control subjects. As the case and control groups were matched by age and sex, there were no statistically significant differences in age (p=0.3519) or sex (p=1.0000) between them. However, significant differences were observed in the distributions of smoking habits, alcohol consumption, and H. pylori infection status between the patients with GCa and the control group (all p<0.0001), indicating these factors as potential risk indicators for GCa in the Taiwanese population. In terms of tumor localization, the GCa cases had tumors located in the upper (14.2%), middle (42.9%), and lower (42.9%) regions of the stomach. Additionally, 91 (56.5%) of the patients with GCa experienced distant metastasis (Table I).
Table III provides a summary of the genotypic distributions for TIMP-1 rs4898, rs6609533, and rs2070584 among the participants in this study. The genotypic distributions for TIMP-1 rs4898, rs6609533, and rs2070584 in the non-cancer control group were consistent with Hardy-Weinberg equilibrium (p=0.1211, 0.2859, and 0.4174, respectively). Regarding TIMP-1 rs4898, no significant difference in genotypic frequencies was observed between the groups (p for trend=0.1987). Specifically, the variant CT and CC genotypes were present in 37.3% and 25.4% of patients with GCa, respectively, compared to 45.8% and 20.9% in the control group (OR=0.78 and 0.63, 95%CI=0.53-1.16 and 0.38-1.07, p=0.2681 and 0.1137, respectively). Analyzed under a dominant model, the combined CT+CC genotypes of TIMP-1 rs4898 were not significantly associated with GCa risk (OR=0.74, 95%CI=0.51-1.07, p=0.1272; Table III, top section). Similarly, for TIMP-1 rs6609533 and rs2070584, no significant differences in genotypic distributions were found between patients with GCa and non-cancer controls (Table III, middle and bottom sections).
Associations between tissue inhibitor of metalloproteinase-1 (TIMP-1) genotypes and gastric cancer risk.
To further explore the results outlined in Table III, allelic frequency analyses for TIMP-1 rs4898, rs6609533, and rs2070584 were performed. Initially, the C allele of TIMP-1 rs4898 was found to be weakly associated with a reduced risk of GCa (p=0.0733) (Table IV, top section). In contrast, no significant association was observed between the variant alleles of TIMP-1 rs6609533, rs2070584 and GCa risk (Table IV, middle and bottom sections).
Distributions of tissue inhibitor of metalloproteinase-1 (TIMP-1) rs4898, rs6609533 and rs2070584 allelic frequencies among the investigated patients with gastric cancer and non-cancer control subjects.
We examined the interactions between TIMP-1 rs4898 and various clinical and lifestyle factors to assess their combined impact on GCa risk (Table V). Initially, when patients were stratified by age, the distribution of TIMP-1 rs4898 genotypes did not vary significantly between those older and younger than 50 years (p=0.3594). Similarly, genotype distributions were consistent across sexes in both the GCa and control groups (p=0.3417) (Table V). Next, no significant differences in genotype distributions were observed based on body mass index, using a threshold of 25 kg/m2 (p=0.6818). For H. pylori infection status, the results remained non-significant across all comparisons (all p>0.05). Notably, cigarette smokers and alcohol drinkers with the CC genotype at TIMP-1 rs4898 exhibited a lower risk of GCa (p=0.0140 and 0.0011, respectively). Additionally, GCa patients with the variant CT or CC genotypes at TIMP-1 rs4898 had a reduced risk of metastasis (p=1.68E-05; Table V).
Combined effects of TIMP-1 rs4898 genotype with demographic and clinical features on gastric cancer risk.
Discussion
The association between TIMP-1 genotypes and GCa risk or prognosis has been infrequently studied. In 2006, Kubben et al. were the first to report that the TIMP-1 rs4898 genotype was not correlated with GCa risk or tumor-related survival outcomes (48). Subsequently, in 2009, Hung et al. demonstrated that the TIMP-1 rs4898 genotypes, in conjunction with MMP-9 rs3918242 genotypes, played a role in determining sex differences in the risk of gastric intestinal metaplasia following H. pylori infection (49). In the present study, we did not only explore the relationship between TIMP-1 genotypes and GCa risk but also examined the interactions between various clinicopathological parameters and TIMP-1 genotypes. Although no significant association was found between the TIMP-1 variant genotypes and overall GCa risk, our analysis suggested a borderline influence of TIMP-1 rs4898 genotypes on GCa susceptibility (Table III and Table IV). Further stratified analysis revealed that the TIMP-1 rs4898 variant genotypes were associated with an increased GCa risk in specific subgroups, particularly among smokers and alcohol consumers (Table V). Notably, GCa patients with TIMP-1 rs4898 variant genotypes were found to have a higher risk of distant metastasis compared to those with the wild-type genotype at TIMP-1 rs4898 (Table V). Early genotypic screening could be beneficial for individuals who smoke or consume alcohol, encouraging them to modify their risky behaviors. Additionally, it may be advisable for GCa patients with TIMP-1 rs4898 variant genotypes to undergo more rigorous whole-body follow-up examinations to monitor the risk of distant metastasis.
Genotype-phenotype correlations for TIMP-1 rs4898 are exceedingly rare. In literature, one study assessed TIMP-1 expression in a cohort of 275 patients with severe sepsis and found that carriers of the TIMP-1 rs4898 T allele exhibited higher serum TIMP-1 levels compared to C allele carriers (50). While the increased TIMP-1 expression in both serum and tumor tissues of patients with GCa is well-documented, it remains essential to stratify TIMP-1 expression levels according to the TIMP-1 rs4898 genotype in patients with GCa and their matched controls.
TIMP-1 has been identified very early to play an important role in carcinogenesis in multiple aspects. As early as 1998, Heppner-Goss et al. developed a mouse model of gastric intestinal neoplasia with TIMP-1 over-expression, demonstrating that TIMP-1 increased tumor multiplicity (51). However, when these mice were concurrently treated with batimastat, a metalloprotease inhibitor, throughout the progression of neoplasia, there was a noticeable reduction in tumorigenicity (51). This discrepancy may be attributed to the biphasic effects of TIMP-1. Specifically, TIMP-1 acts as a multifunctional protein, inhibiting metalloprotease activity at high concentrations while exhibiting mitogenic properties at lower concentrations (52, 53). Moreover, TIMP-1 is recognized for its role in promoting angiogenesis (31, 54), inducing or suppressing apoptosis (55, 56), enhancing inflammation (57, 58), and regulating metastasis (59, 60). Early studies demonstrated that either over-expression of TIMPs or intraperitoneal injection of recombinant TIMP-1 reduced experimental metastasis (61-64). Functional studies using transgenic models further highlighted the significance of the TIMP/MMP balance in carcinogenesis, showing inhibition of SV40 large T antigen-induced liver cancer (65), and MMP-3/stromelysin-induced mammary carcinogenesis (66, 67).
In the literature, numerous studies using experimental mouse models have shown that TIMP-1 can have both pro-neoplastic and anti-neoplastic effects during the processes of primary and metastatic carcinogenesis (67-71). Recently, Tan et al. investigated the relationship between the Vav Guanine Nucleotide Exchange Factor 2 (VAV2) gene and GCa. Their study demonstrated that silencing endogenous VAV2 significantly reduced the viability, migration, and invasion of BGC823 cells. This suppression was accompanied by down-regulation of Rac1, MMP-2, and MMP-9, along with up-regulation of TIMP-1. The authors concluded that VAV2 may promote GCa invasion and metastasis through pathways related to invasion and metastasis, with TIMP-1 potentially playing a crucial role (72). In our study, we further observed that TIMP-1 rs4898 variant genotypes appeared to be modestly associated with a reduced risk of GCa (Table II and Table III), while also contributing to an increased risk of distant metastasis among patients with GCa (Table V).
While the prevalence of GCa varies significantly by geography, it is important to note that the frequency of the variant C allele at TIMP-1 rs4898 remains consistent across different ethnic groups. In East Asians (n=878), the minor C allele frequency is 0.4580, which is comparable to the frequencies observed in Africans (0.4762, n=7,186), Europeans (0.4622, n=107,912), and Latin Americans (0.4355, n=6,246) (73). In our study, the C allele frequency in the control group was 0.4379, aligning closely with those reported in these ethnic groups, particularly East Asians. Therefore, larger, multi-center genotypic studies are meaningful to clarify the role of TIMP-1 genotypes, especially rs4898, in the risk of GCa across diverse populations.
GCa has been established as a smoking-related malignancy (9, 74). Our study not only confirmed that cigarette smoking is a significant risk factor for GCa in the Taiwanese population (Table I) but also expanded upon this by demonstrating that TIMP-1 rs4898 genotypes interact with smoking behavior to influence individual susceptibility to GCa. Our results show that smokers carrying the variant genotypes of TIMP-1 rs4898 have a significantly heightened risk of GCa (Table V). Previous research has shown that smoking can significantly increase TIMP-1 levels in saliva (75) and induce TIMP-1 production in peripheral blood mononuclear cells (76). A similar elevated risk was observed among patients with GCa who consume alcohol (Table V). Literature suggests that alcohol consumption can significantly elevate TIMP-1 serum levels in human adolescents (77) and in the hearts of Sprague–Dawley rats (78). Although the precise mechanisms remain unclear, this evidence underscores the importance of behavioral management in patients with GCa. Those with the TT genotype at TIMP-1 rs4898 should consider reducing or eliminating cigarette smoking and alcohol consumption to lower their risk and improve their prognosis. Further studies are required to clarify the role of TIMP-1 rs4898 genotypes and TIMP-1 expression in GCa.
Alterations in TIMP-1 expression are often experimentally linked with changes in MMPs, making it logical to explore the potential of MMP genotypes as biomarkers for GCa. For example, increased TIMP-1 expression in GCa cells is commonly associated with decreased MMP-1 levels (79), although findings regarding MMP-1 alterations in GCa serum or tumor tissues have been inconsistent (42, 80). In 2017, our team discovered that the variant 1G allele in MMP-1 rs1799750, a promoter polymorphism, was significantly associated with an increased risk of GCa, particularly when combined with cigarette smoking and H. pylori infection (45). Additionally, Zhang et al. reported in 2020 that while MMP-7 or TIMP-1 expression alone could not serve as reliable prognostic indicators for GCa, the co-expression of MMP-7 and TIMP-1 in tumor tissues could (81). The GG genotype at MMP-7 A-181G was identified as a risk factor for GCa among smokers (82). Both MMP-9 and TIMP-1 play crucial roles in regulating metastasis in GCa cells (83, 84). In 2021, we found that the TT genotype of MMP-9 rs3918242 was associated with an increased risk of GCa, and the CT/TT genotypes were linked to a higher risk of distant metastasis (85). It is worthwhile to examine the potential of other MMP genotypes as biomarkers for GCa.
In conclusion, our study provides strong evidence that the variant CC genotype of TIMP-1 rs4898 is associated with a reduced risk of GCa in Taiwanese individuals, particularly among smokers and alcohol consumers. Additionally, this genotype appears to be a favorable marker for predicting a lower risk of distant metastasis. The prognostic potential of the TIMP-1 rs4898 genotype should be further validated across different ethnic groups to confirm its clinical applicability.
Acknowledgements
The Authors are grateful to China Medical University Hospital the Tissue-bank and Dr. Yang’s leadership for their excellent sample collection and technical assistance. The technical assistance from Yun-Chi Wang and Hou-Yu Shih are appreciated. This study was supported by Taichung Armed Forces General Hospital (TCAFGH-D-112014), Taichung Veterans General Hospital (TCVGH-NCHU1127619) and China Medical University plus Asia University (CMU113-ASIA-02). None of the funders had taken part in the study design, data collection and analysis, decision to publish or preparation of the manuscript.
Footnotes
Authors’ Contributions
Conceptualization: CKF, DTB and JCC; Data curation: CKF, HTL, YCH and MDY; Formal analysis: MCM, JCC, HCC and CKF; Funding acquisition: CKF, HTL and MCM; Investigation: CWT and WSC; Methodology: CWT, WSC and DTB; Project administration: CKF, HTL and DTB; Resources: CKF and HTL; Supervision: DTB; Validation: CWT and WSC and DTB; Writing – original draft: CKF, HTL and YCH; Writing – review & editing: CWT and WSC and DTB.
Conflicts of Interest
All the Authors declare no conflicts of interest regarding this study.
- Received August 29, 2024.
- Revision received September 9, 2024.
- Accepted September 10, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
