Association between TIMP-2 gene polymorphism and breast cancer in Han Chinese women

As one of the most prevalent malignancies with highly invasive and metastatic potential, breast cancer continues to be a major global health concern that leads to increasing morbidity and mortality among women worldwide [1]. Domestic and foreign scholars believe that extracellular matrix (ECM) plays a vital role in the invasion and migration of breast cancer cells [2]. Additionally, these studies have demonstrated that degradation of the basement membrane ECM is critical for the progression of tumorigenesis and metastasis [3]. Matrix metalloproteinase-2 (MMP-2) degrades type IV collagen, which is one of the major structural components of the basement membrane ECM. Based on this function, MMP-2 is considered a crucial enzyme in the regulation of tumor proliferation and metastasis [4]. Previous studies have shown that MMP-2 expression is elevated in cancer patients compared with control subjects and is associated with advanced stages of disease and worse prognosis [5].

Tissue inhibitor of metalloproteinase-2 (TIMP-2) is an endogenous inhibitor of MMP-2 that has been implicated in the regulation of MMP-2 proteolytic activity through formation of a 1:1 stoichiometric inhibitory complex with the enzyme [6]. Genetic polymorphisms in the TIMP-2 gene, located on chromosome 17q25, may lead to an increase or decrease in TIMP-2 activity and subsequently disrupt the balance between the activity of TIMP-2 and MMP-2. This disrupted balance could then influence cancer development and progression [7]. More and more research have shown that TIMP-2 mutation influence the risk of the development and persistence of numerous carcinomas and diseases [8‐12]. The correlation between the genetic variants of TIMP-2 and susceptibility to stroke [13], oral squamous cell carcinoma [8], prostate cancer [9], abdominal aortic aneurysm [10], head and neck squamous cell carcinoma [11], and gastric cancer [12] have been identified in a number of studies worldwide. Taken together, these findings suggest that evaluation of TIMP-2 polymorphism in cancers may be useful as a prognostic indicator.

Very few studies have evaluated polymorphism of TIMP-2 in individuals with breast cancer. Combining with the existing literature reports, and minor allele frequencies (MAFs) of greater than 5% in the global population, we selected rs2277698, rs2009196, rs7342880, rs11654470, rs2003241, and rs4789936 six SNPs to research the effect of TIMP-2 gene polymorphisms on the susceptibility of breast cancer in a cohort of Han Chinese women. Genetic screening involving polymorphism of the TIMP-2 gene could provide valuable information for breast cancer susceptibility and identification of high risk patients.

Using RegulomeDB (Table 2), we found that rs4789936 was likely to affect transcription factor binding, motifs, DNase footprint, and DNase peaks. Additionally, rs2003241 was likely to affect transcription factor binding, motifs, and DNase peaks; whereas, the remaining genetic variants (rs2009196, rs7342880, and rs11654470) were only likely to affect transcription factor binding or DNase peaks. Consistent with these findings, HaploReg also predicted that rs2009196, rs7342880, rs1165447, rs2003241, and rs4789936 may result in motif changes (Table 2).

Table 3 shows the location, alleles of the TIMP-2 gene polymorphisms in the breast cancer group and the control group, and whether these sites satisfy the Hardy Weinberg equilibrium. Based on their deviation from HWE, rs11654470 and rs2003241 were excluded from the subsequent analyses. Importantly, the frequencies of the rs2277698 alleles were significantly different between breast cancer patients and control subjects, individuals carrying allele T have a 19% lower risk of breast cancer than individuals with allele C, providing protection (OR = 0.81, 95%CI = 0.67–0.99, p = 0.041).

The detailed findings of the logistic regression analysis for each genetic model are presented in Table 4. Of note, we observed that the frequency of the heterozygous variant C/T genotype of TIMP-2 rs2277698 was significantly reduced in breast cancer patients, when compared with healthy group. In the dominant model, after adjustment for age, the individuals with TIMP-2 rs2277698 CT + TT genotype have a 24% lower risk of developing breast cancer than CC genotype (OR = 0.76, 95%CI = 0.60–0.97, p = 0.025).

As shown in Table 5, in the breast cancer patients with c-erb positive and PR positive, when the TIMP-2 rs2277698 CC genotype was used as a reference, individuals carrying the TT genotype promoted the risk of breast cancer by 72 and 63% in allele model, respectively (c-erb positive: OR = 1.72, 95%CI: 1.08–2.74, p = 0.022; PR positive: OR = 1.63, 95%CI: 1.09–2.43, p = 0.017). When less than 49 years old, individuals with TT genotype had a 31% lower risk of breast cancer than the CC genotype individuals (OR = 0.69, 95%CI: 0.52–0.9, p = 0.007).

Linkage analysis indicated that rs2277698, rs2009196, and rs7342880 exhibit extremely significant linkage disequilibrium (Fig. 1). Therefore, the haplotype frequencies of these SNPs were further examined for association with breast cancer (Table 6). Indeed, when the haplotype “CGC” used as a reference, the haplotype “TCC” was associated with a reduce ed. risk of breast cancer (OR = 0.79, 95%CI = 0.63–0.97, p = 0.028).

To further validate our findings, we employed the use of two publically-available data sets. Examination of 1097 breast cancer tissues and 114 normal tissues from The Cancer Genome Atlas (TCGA) using the UALCAN database demonstrated that TIMP-2 was under-expressed in breast cancer tissues (Fig. 2a). The GTEx database shown that the expression level of carrying the TT genotype is higher than that of the individual carrying the CC genotype,the risk allele of rs4789936 was associated with increased expression of TIMP-2 (p = 4.1× 10^− 8) in peripheral blood samples (Fig. 2b).

In this study, we found that SNP rs2277698 and haplotype, “TCC” in TIMP-2 was significantly associated with an altered risk of breast cancer. Additionally, the UALCAN database demonstrated that the TIMP-2 gene was under-expressed in breast cancer tissues. Based on the GTEx portal, the rs4789936 risk allele “A” increased the expression of TIMP-2 in peripheral blood samples.

In the context of tumor invasion, TIMP-2 is expected to serve as an anti-invasive/anti-metastatic agent through inhibition of MMP-2. Changes in the level of TIMP-2 are known to directly affect the activity level of MMP-2 [25]. In addition, experimental evidence indicates that TIMP-2 has pleiotropic activities, including inhibition of endothelial cell growth induced by basic fibroblast growth factor, suppression of angiogenesis, and regulation of apoptosis [26]. Our analysis using the UALCAN database showed that the TIMP-2 gene was under-expressed in breast cancer tissues. A common polymorphism in the TIMP-2 gene is the C to T transition at position 303 (C303T, rs2277698), which results in a synonymous amino acid change at codon position 101 (Ser101Ser). TIMP-2 gene mutation is associated with the occurrence of multiple diseases, including alcohol induced osteonecrosis of the femoral head [27], emphysema and paraseptal emphysema [28], and gastric cancer [29]. One research explore the association between TIMP-2 and breast cancer, and found that TIMP-2 rs7501477 and rs8064344 mutation affects the genetic susceptibility of breast cancer; while, no effect of rs2277698 mutation on breast cancer was found [16]. In Korean women Primary ovarian insufficiency (POI), revealed that TIMP-2 rs817990 GC (OR = 0.581) genotype and rs2277698 AA-GA (OR = 1.559) genotype influence the risk of Primary ovarian insufficiency in Korean women [30]. However, in our study, we only observed that rs2277698 mutation was associated with genetic susceptibility to breast cancer, and in the breast cancer patients with c-erb positive and PR positive, individuals carrying the TT genotype increased the risk of breast cancer. No other significant results were found. Combined with existing reports, we believe that rs2277698 is a susceptibility site for breast cancer, even affecting gynecological diseases. Other people reported significant results which we did not find in this study, which may be due to the false negative results caused by our small sample size. rs2277698 AA-GA (OR = 1.559) genotype influence the risk of Primary ovarian insufficiency in Korean women, while in our research our, rs2277698“T” allele with decreased breast cancer risk, this may be due to different functions of the same locus in different diseases and genetic differences among populations. Linkage disequilibrium analysis shown that rs2277698 was strongly linked to rs9889410 and rs11654470 in the 1000 Genomes Project population (r² > 0.9), Bioinformatics analysis found that some of which (rs9889410 and rs11654470) reside in a region may be involved in changing transcriptional regulation [31]. Therefore, we speculate that rs2277698 may affect the transcription rate of TIMP-2. However, additional studies are necessary to validate these findings, and the protective mechanism of rs2277698 requires further investigation by biological means.

In our research, we found that rs7342880 and rs4789936 in TIMP-2 gene have no effect on the genetic susceptibility of breast cancer. Nevertheless, a previous study suggested that mutations in the rs4789936 locus not only affect the genetic susceptibility of breast cancer, but also affect the survival of breast cancer patients [16]. So, the role of rs4789936 mutation on the genetic susceptibility of breast cancer remains controversial. Bioinformatics analysis found that mutation of the rs4789936 locus affects the expression of the TIMP-2 gene in peripheral blood samples, the expression level of carrying the TT genotype is higher than that of the individual carrying the CC genotype. And, TIMP-2 was under-expressed in breast cancer tissues. So, we will first expand the sample size to verify whether mutations at this site will affect the risk of breast cancer, and further explore how mutations at this site affect breast cancer development through functional tests.

Although some clinical indicators were collected in this study and stratified analysis was performed, the sample size of complete clinical information was small, which made some indicators unable to be analyzed hierarchically, for example, obesity, smoking and drinking. We will continue to refine this information for in-depth analysis.