The Contribution of Tissue Inhibitor of Metalloproteinase-2 Genotypes to Breast Cancer Risk in Taiwan

Tissue inhibitor of metalloproteinase-2 (TIMP-2) is an endogenous inhibitor of matrix metalloproteinase-2 and is highly expressed in breast cancer (BC) cases at diagnosis. However, the genetic investigations for the association of TIMP-2 genotypes with BC risk are rather limited. In this study, contribution of TIMP-2 rs8179090, rs4789936, rs2009196 and rs7342880 genotypes to BC risk was examined among Taiwan’s BC population. TIMP-2 genotypic profiles were revealed among 1232 BC cases and 1232 controls about their contribution to BC using a PCR-based RFLP methodology. The TIMP-2 rs8179090 homozygous variant CC genotype was significantly higher in BC cases than controls (odds ratio (OR) = 2.76, 95% confidence interval (95%CI) = 1.78–4.28, p = 0.0001). Allelic analysis showed that C allele carriers have increased risk for BC (OR = 1.39, 95%CI = 1.20–1.62, p = 0.0001). Genotypic together with allelic analysis showed that TIMP-2 rs4789936, rs2009196 or rs7342880 were not associated with BC risk. Stratification analysis showed that TIMP-2 rs8179090 genotypes were significantly associated with BC risk among younger (≤55) aged women, not among those of an elder (>55) age. Last, rs8179090 genotypes were also associated with triple negative BC. This study sheds light into the etiology of BC in Taiwanese women. Rs8179090 may be incorporated into polygenic risk scores and risk prediction models, which could aid in stratifying individuals for targeted breast cancer screening.


Introduction
Breast cancer (BC) stands as the foremost malignancy among women globally Among its various subtypes, triple-negative breast cancer (TNBC) constitutes 15-20 BC cases [3].Clinically, TNBC demonstrates a high aggressiveness, with the capacit metastasis to diverse organs such as the brain, lungs, heart, liver, and bones [4].Not the 5-year survival rates for TNBC patients are notably lower compared to other BC types (77% versus 93%) [5].In Taiwan, BC has consistently held the highest incid among cancers [6,7].Moreover, BC in Taiwan is characterized by a marked surge in p alence and relatively early onset (diagnosis at 45-49 years) [6,8].Our findings pro evidence suggesting that MTHFR genotypes might contribute to BC susceptibility an tentially play a role in the early onset of BC in Taiwan, aligning with alternative hyp ses proposed by other research groups [9][10][11][12].Currently, one of the focal points for t lational scientists is the search for practical biomarkers for BC, particularly TNBC in wan [13][14][15].
Tissue inhibitor of metalloproteinase-2 (TIMP-2) plays a critical role in regulatin mor invasion by modulating the activity of MMP-2 [16].A study dating back to 200 Nakopoulou and his colleagues examined 136 BC samples, revealing that larger tu sizes in BC patients were associated with negative TIMP-2 expression [17].Conver elevated TIMP-2 expression was often observed in low-grade BC cases, correlating w better survival rate compared to those exhibiting normal TIMP-2 expression [17].A study by Peeney and his colleagues reported that TIMP-2 could suppress the growth metastasis of TNBC cells by modulating the epithelial-to-mesenchymal transition and naling pathways associated with metastatic outgrowth [18].Subsequently, a 2022 s found significantly higher TIMP-2 expression among 96 BC patients compared t healthy individuals [19].However, the precise contribution of elevated TIMP-2 to BC ology remains unclear.On the other side, genetic variations within the TIMP-2 gen cated on chromosome 17q25, may potentially alter its activity, disrupting the balanc tween TIMP-2 and MMP-2 activity.This disrupted equilibrium could significantly im the development and progression of tumor cells [20].Accumulated research indicated TIMP-2 genetic variants may influence the risk of various types of cancer, including and neck cancer, prostate cancer, and gastric cancer [21][22][23].Collectively, these find suggested that assessing TIMP-2 polymorphisms, in addition to its expression le could serve as a valuable biomarker for various types of cancer.
Only a handful of studies worldwide have investigated the single nucleotide p morphisms (SNPs) of TIMP-2 in individuals with BC [24][25][26][27].In this study, we us hospital-based case control study to compare the genotype frequencies of four SN TIMP-2 gene, rs8179090, rs4789936, rs2009196, and rs7342880, between BC patients healthy controls and determine the associations between these SNPs and BC risks.A tionally, we aim to provide evidence of the role of TIMP-2 genotypes in predicting T risk.The physical map for the investigated SNPs in this study is shown in Figure 1.

BC and Non-BC Control Population
A total of 1232 patients diagnosed with BC were enrolled from the outpatient clinics of the general surgery department at China Medical University Hospital in Taiwan for Life 2024, 14, 9 3 of 12 this study.All participants were of Taiwanese descent, and the detailed procedures, exclusion and inclusion criteria were previously documented [28][29][30].Clinical characteristics, including histological details, were defined by Dr. Su and his team.BC tissue slides were independently reviewed and scored by at least two pathologists.Positivity for ER, PR, and HER-2/neu immunoassaying was determined via nuclear staining in 10% of neoplastic cells.A Ki67-labeling index of >30% was considered positive.HER-2/neu results adhered to the guidelines of the American Society of Clinical Oncology and the College of American Pathologists [31].All patients voluntarily participated, completed a self-administered questionnaire, and provided peripheral blood samples.As controls for this study, 1232 age-matched healthy volunteers were selected through initial random sampling from the health examination cohort of the hospital.Exclusion criteria for the control group included previous malignancies, metastasized cancer of other or unknown origins, and any familial or genetic diseases.Both groups completed a brief questionnaire that included lifestyle habits.Our study received approval from the Institutional Review Board of China Medical University Hospital (DMR-96-IRB-240), and written informed consent was obtained from all participants.

TIMP-2 Genotyping Methodology
Peripheral blood was collected from all participants, and their DNA was extracted within 24 h and stored according to our routine protocol [32,33].Genotyping for TIMP-2 rs8179090, rs4789936, rs2009196, and rs7342880 was performed using the polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) methodology as we previously published [34].Briefly, the primer sequences for TIMP-2 genotyping were designed and optimized by the Terry Fox Cancer Research Lab.The forward and reverse primer sequences for TIMP-2 rs8179090 were TTCTAAGGCCTCCATTTGAA and GTTCTTCCAGGACACCAGGC, respectively.For TIMP-2 rs4789936, the forward and reverse primer sequences were CCATCTACAGAGATGCCAGT and TAAGCTGAGATCG-CACCACT, respectively.The forward and reverse primer sequences for TIMP-2 rs2009196 were GACTGAAGCTCATCTGTTGA and CGCGAGACTCCATCTCAATA, and for TIMP-2 rs7342880, the sequences were GCCACAGTTGTTCACACCTA and GGACCCTGAA-GAATCTGAAT.PCR was conducted using a PCR thermocycler (Bio-RAD, Hercules, CA, USA) under the following conditions, initial denaturation at 94 • C for 5 min, followed by denaturation at 94 • C for 30 s, annealing at 64 • C for 40 s, and extension at 72 • C for 45 s.Last, after 35 PCR cycles, a final extension was performed at 72 • C for 10 min.The PCR products were analyzed using 3% agarose gel electrophoresis.For TIMP-2 rs8179090, the G allele was digested by Mnl I, resulting in two fragments of 117 and 119 base pairs (bps), while the undigested T allele remained at 236 bps.Similarly, for TIMP-2 rs4789936, the C allele was digested by BtsIMut I, yielding two fragments of 179 and 292 bps, while the undigested A allele remained at 471 bps.For TIMP-2 rs2009196, the C allele was digested by Mwo I, resulting in two fragments of 116 and 251 bps, while the undigested G allele remained at 367 bps.For TIMP-2 rs7342880, the A allele was digested by BssS I, resulting in two fragments of 135 and 418 bps, while the undigested C allele remained at 553 bps.We sent ten DNA samples with representative genotypes and the results of PCR-RFLP and sequencing were 100% concordant.Genotyping was independently and blindly repeated by two researchers, and all procedures yielded consistent results with 100% concordance.The genotype frequencies of all the four SNPs were in the Hardy-Weinberg equilibrium.

Analyzing Methodology
The age distribution difference between the case and control groups was evaluated using the typical Student's t-test.Pearson's chi-square test was employed to assess the differential distribution of the TIMP-2 genotypes.The associations between the TIMP-2 genotypes and BC risk were analyzed using multivariable logistic regression.Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were calculated in overall population and in various stratification analyses.To increase statistical power, we recruited as many cases and controls as possible.Our sample size of 1232 pairs of BC patients and healthy controls gives us 90% power to detect an OR of 1.37 for a SNP like rs8179090 with a risk allele frequency of 18.9%.A p-value below 0.05 was deemed statistically significant for all outcomes.

The Demographic Comparisons of the Taiwan BC Population
The comparison of age, age at menarche, age at first childbirth, age at menopause, personal habits, family history, tumor sites, and TNBC status between the 1232 BC cases and 1232 controls is presented in Table 1.Initially, no significant differences were observed between the case and control groups in terms of age, age at menarche, age at first childbirth, and age at menopause (all p > 0.05) (Table 1).Secondly, the prevalence of smokers and alcohol consumers was notably higher in the BC patient group than in the control group (both p < 0.0001) (Table 1).Lastly, among the 1232 BC cases, 194 were identified as TNBC cases, and the majority (97.2%) of BC cases were unilateral (Table 1).

The Genotypic Patterns of TIMP-2 in Taiwan BC Population
The genotypic distributions of TIMP-2 among the 1232 controls and 1232 BC cases are presented in Table 2. Initially, the frequencies of TIMP-2 rs8179090, rs4789936, rs2009196, and rs7342880 genotypes among the controls all conformed to the Hardy-Weinberg equilibrium (p = 0.3886, 0.1735, 0.3904, and 0.8422).Subsequently, significant differences were observed in the distribution of TIMP-2 rs8179090 genotypes between the BC and control groups (p for trend = 1.00 × 10 −5 ).In detail, carriers of the TIMP-2 rs8179090 heterozygous variant CG and homozygous variant CC genotypes exhibited 1.16-and 2.76-fold increased ORs for BC risk, respectively, compared to individuals with the wild-type GG genotype (95%CI = 0.97-1.39 and 1.78-4.28,p = 0.1249 and 0.0001, respectively).The latter association was considered statistically significant.Furthermore, individuals with the homozygous variant CC genotype showed a significantly higher risk for BC than those carrying GG + CG genotypes in the recessive model (OR = 2.65, 95%CI = 1.71-4.10,p = 0.0001).Additionally, carriers of the TIMP-2 rs8179090 CG + CC genotypes exhibited a significantly higher risk for BC than GG carriers in the dominant model (OR = 1.30, 95%CI = 1.09-1.55,p = 0.0034).Notably, carriers of variant CG and CC genotypes at TIMP-2 rs8179090 displayed a de-Life 2024, 14, 9 5 of 12 creased risk for BC.Regarding TIMP-2 rs4789936, rs2009196, and rs7342880, no variant genotype was found to be associated with altered BC risk in any of the examined models (Tables 3-5 for each SNP).

The Patterns of TIMP-2 Allelic Frequencies in Taiwan BC Population
The allelic frequency distribution analyses for the four TIMP-2 SNPs were performed to confirm the findings presented in Tables 2-5.Consistently, the sole significant observation was that the variant C allele of TIMP-2 rs8179090 exhibited a higher frequency (18.9%) in the BC group compared to the control group (14.3%) (OR = 1.39, 95%CI = 1.20-1.62,p = 0.0001, as shown in Table 6).

TIMP-2 Rs8179090 Genotypes Correlated with Onset Ages in Determining BC Risk
The genotyping results for TIMP-2 rs8179090 were further stratified by age among both cases and controls (as presented in Table 7) to investigate the interaction between TIMP-2 rs8179090 genotype and age concerning BC risk.Intriguingly, the homozygous variant CC genotype at TIMP-2 rs8179090 showed an increased association with BC risk among individuals aged 55 years or younger (OR = 3.67, 95%CI = 2.11-6.38,p = 0.0001).Conversely, the CC genotypes at TIMP-2 rs8179090 did not exhibit an altered BC risk among individuals aged over 55 years (OR = 1.47, 95%CI = 0.69-3.14,p = 0.4236) (Table 7).

TIMP-2 Rs8179090 Genotypes Were Associated with TNBC Risk
We sought to investigate whether the TIMP-2 rs8179090 genotype could serve as a biomarker for predicting TNBC risk.Therefore, the BC patients were further stratified into TNBC and non-TNBC subgroups.The findings revealed significant associations between the TIMP-2 rs8179090 homozygous variant CC genotype and both BC and TNBC.Among both TNBC and non-TNBC cases, the presence of the TIMP-2 rs8179090 homozygous variant CC genotype showed significant associations with BC and TNBC (OR = 2.63 and 3.48, 95%CI = 1.67-4.14and 1.79-6.76,p = 0.0001 and 0.0003) (Table 8).

Discussion
The major finding of this study is that the TIMP-2 SNP rs8179090 homozygous variant CC genotype was associated with a 2.76-fold increased risk, and the C allele associated with a 1.39-fold increased risk of breast cancer in Taiwanese women.The other three SNPs, rs4789936, rs2009196 or rs7342880, were not associated with BC risk.Moreover, rs8179090 genotypes can also serve as a biomarker for triple negative BC.
Consistent with our results, a previous study evaluated 19 TIMP-2 SNPs in a Chinese population consisting of 1062 BC cases and 1069 healthy controls and found that rs8179090 variant genotypes were significantly associated with BC risk.They also found that another nearby SNP, rs7501477, exhibited a 3-fold higher likelihood of BC with the TT variant genotype compared to women with the wild-type CC genotype [26].We were the first to examine the contribution of TIMP-2 genotypes to the risk of BC in Taiwan.Moreover, we found that the genotypes of rs8179090 can serve as predictors for the occurrence of TNBC (Table 8).Furthermore, we discovered that the genotypes of rs8179090 were selectively associated with BC susceptibility among women younger than 55 years old (Table 7).Consequently, we propose applying this novel marker for early detection of BC risk in young women.There were some reports examining the expression level of TIMP-2.In 2019, Wang and his colleagues reported significantly elevated transcriptional expression levels of TIMP-2 among tumor sites from 1097 BC cases compared to 114 normal controls, which conflicts with, yet holds greater persuasiveness in terms of sample size, than the findings presented by Ozdemir, which involved only 96 BC patients and 35 healthy individuals [19].Furthermore, in 2022, Krasnikova and co-authors observed a notable decrease in serum TIMP-2 levels post-BC chemo-treatments among 67 BC patients compared with 25 healthy subjects [35].The authors suggested that reduced serum TIMP-2 could potentially serve as an indicator for endothelial dysfunction resulting from anti-tumor therapy.Nakopoulou [17] and Jones [36] observed comparable immuno-activity of TIMP-2 in breast primary cancer cells and fibroblasts, while Garbett and colleagues reported heightened TIMP-2 expression in tumor cells compared to fibroblasts and inflammatory cells [36].Nevertheless, these studies were constrained by small sample sizes and varied types of chemo-treatments.In 2020, Peeney and his colleagues provided evidence from an orthotopic mice model demonstrating the effective suppression of growth and metastasis of TNBC cells by TIMP-2 [18].
In the current study, for rs4789936, rs2009196, and rs7342880 we did not find significant associations (Tables 3-5), while rs8179090 can serve as a novel marker for BC risk (Table 2).In the literature, there were several reports revealing the genotypes of TIMP-2 and then evaluated their contributions to BC.In 2009, the BC study by Peterson and colleagues evaluated the contribution of TIMP-2 genotypes to BC risk using 19 TIMP-2 polymorphic sites in a population consisting of 1062 BC cases and 1069 healthy controls.Their data supported our finding that promoter polymorphic rs8179090 genotypes were significantly associated with BC risk (Tables 2 and 6).They also found that a polymorphic site nearby, rs7501477, exhibited a 3-fold higher likelihood of BC cases with the TT variant genotype compared to women with the wild-type CC genotype [26].The samples collected in our current study are both genetically and geographically consistent and notably larger.Additionally, we were the first to examine the contribution of TIMP-2 genotypes to the risk of TNBC, revealing that the genotypes of TIMP-2 rs8179090 can serve as predictors for the occurrence of TNBC (Table 8).Furthermore, we discovered that the genotypes of TIMP-2 rs8179090 were selectively associated with BC susceptibility among women younger than 55 years old (Table 7).Liu and his colleagues reported that the variant genotypes of the rs4789936 were associated with an increased risk of BC in a study of 480 Chinese BC patients and 530 healthy controls [25].Wang and his colleagues investigated four SNPs, including rs2277698, rs2009196, rs7342880, and rs4789936 in another Chinese population comprising 566 BC patients and 578 healthy controls.Consistent with our results, they did not find significant associations for the rs2009196 and rs7342880 SNPs.TIMP-2 is an endogenous inhibitor of matrix metalloproteinases that are involved in cancer development and progression.There is ample literature supporting that TIMP-2 exhibits antitumor activities, inhibiting tumor cell growth, angiogenesis, epithelialmesenchymal transition (EMT), and metastasis [18,37,38].TIMP-2 is downregulated or silenced in a variety of cancers [39,40].For example, TIMP-2 is hypermethylated and silenced in prostate cancer cell lines and primary tumors, and silenced TIMP-2 gene expression is associated with cancer progression during the invasive and metastatic stages of the disease.Furthermore, re-expression of TIMP-2 in metastatic prostate cancer cells significantly inhibited cell invasion [40].Nakopoulou et al. found that increased tumor volumes often correlated with negative TIMP-2 expression [17].Additionally, higher TIMP-2 levels were detected in most cases of low-grade BC patients, correlating with longer survival periods [17].Furthermore, Krasnikova et al. observed a notable decrease in serum TIMP-2 levels post-BC chemo-treatments [35].Notably, Peeney and colleagues demonstrated in an orthotopic mouse model that TIMP-2 can suppress the proliferation and metastasis of TNBC tumor cells [18].All these data support a tumor-suppressive role of TIMP-2 in BC.
Rs8179090 is located at position −418 in the consensus sequence for the Sp1 binding site within the core promoter region of the TIMP-2 gene [41].Sp1 protein binds to the consensus sequence and stimulates transcriptional activity.Therefore, it is hypothesized Life 2024, 14, 9 9 of 12 that a G/C transition at −418 position results in downregulation of the transcriptional activity of the promoter, leading to reduced TIMP-2 expression.Reduced TIMP-2 would increase BC risk.It is therefore biologically plausible for the significant associations between the CC genotype and C allele and increased BC risk.
Giving that TIMP-2 serves as an endogenous inhibitor of MMPs, the precise reason for how variant genotypes may influence individual susceptibility to BC remains unclear.Multiple studies have suggested that, alongside its inhibitory effects on MMP-2/9, TIMP-2 can foster tumor cell proliferation, facilitate invasiveness/metastasis, and impede tumor cell apoptosis [42][43][44].Notably, with regards to TNBC, Peeney and colleagues demonstrated in an orthotopic mouse model that TIMP-2 can suppress the proliferation and metastasis of TNBC tumor cells [18].Multiple pieces of evidence have linked elevated levels of TIMP-2 with the proliferation, invasion, and/or metastasis of BC [17,45,46], as well as other cancer types such as oral cancer [47], laryngeal cancer [48], colorectal cancer [49], renal cell carcinoma [50], bladder cancer [51], and prostate cancer [52].It is also noteworthy that other members of the TIMP family, such as TIMP-1 and TIMP-4, have been observed to exert activating influences on the proliferation of BC cells [53,54].Moreover, high levels of serum TIMP-1 have been associated with a poor prognosis for TNBC [55].
This study has a few limitations.First, we do not have breast cancer tissues and could not compare the expression of TIMP-2 between tumor and normal tissues as well as between tumor tissues of difference stages.Second, this is a single center study, and the results need to be validated in an independent population.Finally, we only studied one gene and four SNPs.The predictive accuracy of one significant SNP is modest.Future studies should perform whole-genome SNP genotyping to identity a large panel of BC susceptibility SNPs in Taiwanese women.Then polygenic risk scores (PRS) can be developed and incorporated into risk prediction models.A personalized risk prediction model incorporating PRS can help identify women at the highest risk of developing BC, which would allow the implementation of risk stratified, targeted screening and prevention [56].

Conclusions
This is the first study to investigate the TIMP-2 SNPs and BC risk in Taiwan.We found that the homozygous variant CC genotype and the variant C allele of TIMP-2 rs8179090 SNP were associated with significantly increased BC risks in Taiwanese women.Furthermore, the association appeared to be stronger in TNBC and in younger women (≤55 years old).The other three SNPs, rs4789936, rs2009196 and rs7342880, were not associated with BC risk.This study sheds light into the etiology of BC in Taiwanese women.Rs8179090 may be incorporated into PRS and risk prediction models, which could aid in stratifying individuals for targeted breast cancer screening.Funding: This study received significant support from China Medical University and Asia University (CMU112-ASIA-01) and Taichung Veterans General Hospital (TCVGH-VHCY1128606).The funders had no involvement in the study design, data collection, statistical analysis, decision to publish, or manuscript preparation.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of China Medical University Hospital (DMR-96-IRB-240: 9 April 2007).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Table 1 .
Demographics of the 1232 BC patients and the 1232 healthy controls.
a chi-square or b unpaired Student's t-test; * statistically significant; TNBC, triple negative breast cancer.

Table 2 .
TIMP-2 rs8179090 genotypes among the 1232 patients with BC and 1232 healthy controls.
a , based on chi-square test with Yates' correction; OR, odds ratio; CI, confidence interval; p trend , p-value for trend analysis; p HWE , p-value for Hardy-Weinberg equilibrium analysis; for significant p-values, they are shown as bold and a * behind.

Table 3 .
TIMP-2 rs4789936 genotypes among the 1232 patients with BC and 1232 healthy controls.
a , based on chi-square test with Yates' correction; OR, odds ratio; CI, confidence interval; p trend , p-value for trend analysis; p HWE , p-value for Hardy-Weinberg equilibrium analysis.

Table 4 .
TIMP-2 rs2009196 genotypes among the 1232 patients with BC and 1232 healthy controls.
a , based on chi-square test with Yates' correction; OR, odds ratio; CI, confidence interval; p trend , p-value for trend analysis; p HWE , p-value for Hardy-Weinberg equilibrium analysis.

Table 5 .
TIMP-2 rs7342880 genotypes among the 1232 patients with BC and 1232 healthy controls., based on chi-square test with Yates' correction; OR, odds ratio; CI, confidence interval; p trend , p-value for trend analysis; p HWE , p-value for Hardy-Weinberg equilibrium analysis. a

Table 6 .
Distribution of allelic frequencies for TIMP-2 rs8179090 among the 1232 patients with BC and 1232 healthy controls.
a , based on chi-square test with Yates' correction; OR, odds ratio; CI, confidence interval; for significant p-value, it is shown as bold and a * behind.

Table 7 .
TIMP-2 rs8179090 genotypes in BC risk after stratification by age., by multivariate logistic regression analysis; b , by multivariate logistic regression analysis after adjusting for gender, smoking and alcohol drinking status; p trend , p-value for rend analysis; *, statistically significant; CI, confidence interval; aOR, adjusted odds ratio; * and bolded, statistically significant. a
a , based on chi-square test without Yates' correction; OR, odds ratio; CI, confidence interval, TNBC, triple negative breast cancer; p trend , p-value for trend analysis; * and bolded, statistically significant.