Genotypes and haplotypes of matrix metalloproteinase 1, 3 and 12 genes and the risk of lung cancer

Abstract

The MMPs (matrix metalloproteinases) are a family of secreted zinc metalloproteases that degrade the collagens of the extracellular matrix important in tissue remodeling and repair during development and inflammation. We investigated the associations between polymorphisms of MMP-1 ( -1607 1G/2G , rs1799750), MMP-3 ( -1171 5A/6A , rs3025058), and MMP-12 ( -82AG , rs2276109, and 1082A/G , rs652438) and the risk of lung cancer in 2014 Caucasian lung cancer patients and 1323 healthy controls. The results were analyzed using logistic regression models, adjusting for covariates. The four polymorphisms were in Hardy-Weinberg disequilibrium. Except for the 1G-1082A , the other linkage disequilibrium tests between the four MMP polymorphisms were statistically significant ( P < 0.001). There was no overall association between individual MMP polymorphism and the risk of lung cancer. The MMP polymorphisms jointly were associated with a non-statistically significant higher risk of lung cancer, with the adjusted odds ratio (AOR) of subjects with 5+ variant alleles versus zero variant allele of 1.31 [95% confidence interval (CI), 0.92–1.88]. Stronger associations were observed in never-smokers and males, with the corresponding AORs of 2.44 (95%CI, 1.10–5.43, P_trend = 0.04) in never smokers and 1.35 (95%CI, 0.79–2.30, P_trend = 0.04) in men. In haplotype analysis, the 1G-6A-82A-1082G haplotype was associated with higher risk of lung cancer among never smokers, with the AOR of 3.65 (95%CI, 1.62–8.20) when compared with the most common 1G-5A-82A-1082A haplotype. In conclusion, the combined MMP genotypes and associated haplotypes may be associated with higher risk of lung cancer, particularly among never smokers and men.

Introduction

Matrix metalloproteinases (MMPs) are a pivotal family of zinc enzymes responsible for degradation of the extracellular matrix components including basement membrane collagen, interstitial collagen, fibronectin and various proteoglycans, during normal remodeling and repair processes. MMPs play a central role in the enhancement of tumor-induced angiogenesis, cell migration, proliferation, apoptosis and connective tissue degradation ( 1 ). Among the 20+ MMPs, MMP-1 (collagenase) may degrade the interstitial types I, II and III collagens and contribute to tumor initiation and development by altering the cellular microenvironment that facilitates tumor formation. MMP-3 (stromelysin-1, also known as STR1, STMY1) is capable of degrading proteoglycan, fibronectin, laminin, type IV collagen, and may activate other MMPs including MMP-1 ( 2 ). MMP-12 (macrophage metalloelastase, MME) shares the highly conserved exon size and intron-exon borders with other MMPs ( 3 ), and participates in aortic elastin degradation ( 4 ).

MMP-1 , MMP-3 and MMP-12 are located in the same chromosome region ( 11q22-q23 ), with functional polymorphisms suggested in in vitro studies. The 2G allele of the MMP-1-1607 1G/2G polymorphism (rs1799750) has been associated with higher MMP-1 expression levels ( 5 ) and higher risk of lung cancer ( 6 , 7 ). The 6A allele of the MMP-3-1171 5A/6A polymorphism (rs3025058) is in linkage disequilibrium with the MMP-1 2G allele ( 8 ), while it was associated with lower promoter activity in a transient expression experiment ( 9 ), and lower risk of lung cancer among smokers in a Chinese study ( 8 ). Moreover, the MMP 2G/6A haplotype was associated with lower risk of lymphatic metastasis of lung cancer when compared with the 1G/5A haplotype ( 8 ). The A allele of the MMP-12 -82AG polymorphism (rs2276109) shows a higher affinity for the transcription factor activator protein-1 (AP-1) and higher gene expression in reporter gene assays ( 10 ). The MMP-12 1082A/G ( 357Asn/Ser , rs652438) polymorphism is located in the coding region of the hemopexin domain that is responsible for MMP-12 activity, while the function of this polymorphism remains unknown. MMP-12 expression levels were found to be upregulated in recurrent versus non-recurrent stage IB lung cancer ( 11 ) and correlated with local recurrence and metastasis ( 12 ). Currently, there are no reports on the MMP-12 polymorphisms in the risk of lung cancer, although haplotyes of the 1G/1082A polymorphism have been associated with higher rate of decline of lung function ( 13 ).

The MMP-1 2G allele has been associated with higher risk of lung cancer in never smokers and males in our previous analysis ( 7 ). We hypothesized that the 6A allele of MMP-3 polymorphism, the G allele of the MMP-12-82AG polymorphism, and the G allele of the MMP-12 1082A/G polymorphism, which have been shown to be in linkage disequilibrium with the MMP-1 2G allele ( 8 , 13 ), are also associated with higher risk of lung cancer, specially among never smokers and men. In addition, we hypothesized that the joint effects or haplotypes of these polymorphisms are stronger than the individual effect of each polymorphism. In this ongoing study with an expanded sample size, we investigated the associations between these MMP polymorphisms and the risk of lung cancer, using gene–gene joint effects as well as haplotype analyses.

Materials and methods

Study population

This is a hospital based case–control study with details described previously ( 14 – 16 ). Briefly, all eligible cases (patients with histologically confirmed lung cancers) at Massachusetts General Hospital were recruited between December 1992 and December 2004. Controls were recruited among healthy friends and non-blood-related family members (usually spouses) of several groups of hospital patients: (i) patients with cancer, whether related or not related to a case; or (ii) patients with a cardiothoracic condition undergoing surgery. No matching was performed. Importantly, none of the controls were themselves patients. Potential controls who carried a previous diagnosis of any cancer (other than non-melanoma skin cancer) were excluded from participation. Over 85% eligible cases, and over 90% controls participated in this study and provided blood samples. Interviewer-administered questionnaires collected information on demographic and detailed smoking histories from each subject. The study was approved by the Human Subjects Committees of Massachusetts General Hospital and the Harvard School of Public Health, Boston, MA.

MMPs genotyping

DNA was extracted from peripheral blood samples using the Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, MN). The MMP-1 , MMP-3 and MMP-12 polymorphisms were genotyped by the 5′ nuclease assay (TaqMan) using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster city, CA). The primers, probes and reaction conditions are available upon request. Genotyping was performed by laboratory personnel blinded to case–control status, and a random 5% of the samples were repeated to validate genotyping procedures. Two authors reviewed independently all genotyping results.

Statistical analysis

Although individuals of all races were recruited for this study, we restricted our analyses to Caucasians (97%) to minimize confounding due to allele frequency variation by race. We analyzed all Caucasians with complete information on age, gender, smoking status (never smokers, ex-smokers and current smokers), pack-years of smoking, and years since smoking cessation (for ex-smokers).

Hardy–Weinberg Disequilibrium of each polymorphism in cases and controls were tested using chi-square test; detection of linkage disequlibrium between the four polymorphisms was based on Lewontin's D′ in controls. Haplotype frequencies and individual haplotypes were generated using the Expectation Maximization (EM) algorithm, which reconstruct individual probabilities for individual phasing accuracy based on unphased genotype data, as well as estimates on the overall haplotype frequencies and their standard errors ( 17 – 20 ).

Analyses of all genotype and haplotype associations with lung cancer risk were performed using logistic regression models. A total of three types of models were fit to examine the relationship between the log odds of lung cancer and each covariate, after adjusting for possible confounding factors such as age, gender, smoking status, pack-years of smoking and years since smoking cessation (if ex-smoker). First, we investigated the associations between individual MMP polymorphism and the risk of lung cancer in separate logistic regression models, overall and in different strata of gender and smoking status (Model type I). Second, we investigated the joint effects of all of the four MMP polymorphisms, based on the results of linkage disquilibrium and the number of variant alleles, where the wild genotype ( 1G/1G of MMP-1 , 5A/5A of MMP-3 , A/A of MMP-12 -82A/G , and A/A of the MMP-12 1082A/G polymorphism) has zero variant allele, heterozygous genotype has one variant allele, and homozygous variant genotype has two variant alleles (Model type II). Last, we investigated the associations between MMP haplotypes and the risk of lung cancer using the ‘expectation substitution’ approach in SAS macro of ‘HAPPY’ by Dr Peter Kraft ( 17 , 18 , 20 ), which treats expected haplotype scores (calculated under additive model) as observed covariates in a standard unconditional logistic analysis, instead of assigning each subject with the most likely haplotype pair (Model type III). A lack of fit test, as described in Hosmer and Lemeshow ( 21 ), was performed to summarize the goodness-of-fit for each logistic regression model. Where appropriate, the odds ratios (OR) and 95% confidence intervals (CI) for the risk of lung cancer were calculated from these models. All reported P -values are from two-sided tests. P -values <0.05 were considered statistically significant. All analyses were performed using SAS software version 9 (SAS Institute, Cary, NC).

Results

Population characteristics

There were no significant demographic differences (age and gender) between enrolled and unenrolled eligible cases and controls. Genotyping success rate was 98% for all of the four MMP polymorphisms. We restricted our analysis to the 3337 Caucasians with complete data (excluding 116 non-Caucasian subjects). Of these, there were 2014 lung cancer cases and 1323 controls. There was 100% concordance of the randomly repeated samples.

The distributions of demographic characteristics for cases and controls are summarized in Table I . Compared to the controls, cases were older, had a higher proportion of males, more likely to be current smokers or heavy smokers, and had a shorter time since smoking cessation (if an ex-smoker) and larger pack-years of smoking. The distribution of smoking variables in our controls was similar to the general Massachusetts population over the age of 45 years ( 15 , 22 ). Adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and small-cell carcinoma represented 51, 21, 8 and 9% of cases, respectively. Eleven percent of patients were of mixed histological subtype or had more than one primary tumor. Clinical AJCC stage data were available for 1987 cases, 45% were early stage (I or II).

Distribution of MMP polymorphisms among cases and controls

All MMP polymorphisms in the control and case populations were consistent with Hardy–Weinberg equilibrium ( P > 0.05, chi-squared goodness-of-fit). Genotype frequencies of MMP polymorphisms ( Table I ) were comparable with previous studies ( 6 , 13 ). Except for the linkage of 1G-1082A ( D′ = 0.16), the other linkages disequilibrium tests between the four MMP polymorphisms were statistically significant ( P < 0.001), with the D′ of 1G-5A of 0.46, 1G-82A of 0.80, 5A-82A of 0.81, 5A-1082A of 0.98 and 82A-1082G of 1.00.

Associations between MMP genotypes and lung cancer risk

There were no overall crude or adjusted associations between individual MMP polymorphism and the risk of lung cancer ( Table II , Model type I). In the subgroup analysis, although the variant genotypes of all of the MMP polymorphisms were associated with higher risk of lung cancer in never smokers than in ever smokers, in men than in women, the majority of the results were not statistically significant ( Table II ). Specifically, the 6A/6A genotype of the MMP-3 polymorphism was associated with higher risk of lung cancer in never smokers [adjusted odds ratio (AOR), 1.76, 95%CI, 1.04–2.97; 6A/6A versus 5A/5A ], and the G allele of the MMP-12 1082 A/G polymorphism was associated with higher risk of lung cancer in men (AOR, 1.51, 95%CI, 1.04–2.09; A/G + G/G versus A/A ).

In Model type II, the combined MMP polymorphisms were dichotomized into six groups based on the number of variant alleles and sample sizes of cases and controls. The joint MMP genotypes were associated with a non-statistically significant higher risk of lung cancer ( Table III ): the AOR of subjects with 4 and 5+ variant alleles versus zero variant allele were 1.23 (95%CI, 0.92–1.68) and 1.31 (95%CI, 0.92–1.88), respectively ( P_trend = 0.29). By different strata of smoking status and gender, the joint MMP polymorphisms were associated with higher risk of lung cancer in never smokers and in men: the AORs of subjects with 4 and 5+ variant alleles versus zero variant allele were 2.37 (95%CI, 1.15–4.90) and 2.44 (95%CI, 1.10–5.43), respectively ( P_trend = 0.04), in never smokers; and 1.61 (95%CI, 1.02–2.54) and 1.35 (95%CI, 0.79–2.30), respectively ( P_trend = 0.04), in males.

Associations between MMP haplotypes and lung cancer risk

There are total of seven common haplotypes (>2%) among both cases and control. The distributions of different haplotypes were similar between cases and controls ( Table IV ). The most common haplotype was the 1G-5A-82A-1082A haplotype, with the frequencies of 36% in both cases and controls. In the haplotype analysis where the expected haplotype scores were fitted as observed covariates in the logistic regression model (Model type III), the most common 1G-5A-82A-1082A haplotype was treated as reference group. In the overall analysis, although the 1G-6A-82A-1082G haplotype was associated with borderline significantly higher risk of lung cancer with the AOR of 1.52 (95%CI, 1.00–2.30), the global test of haplotype association was not statistically significant ( P = 0.18 in likelihood test). Stratified by smoking status, the higher risk of the 1G-6A-82A-1082G haplotype was observed in never smokers (AOR of 3.65, 95%CI, 1.62–8.20), while not in ex-smokers (AOR of 0.91, 95%CI, 0.53–1.58) or current smokers (AOR of 2.21, 95%CI, 0.80–6.10). Similar associations between MMP haplotypes and the risk of lung cancer were found among women and men, with the AORs ( 1G-6A-82A-1082G versus 1G-5A-82A-1082A ) of 1.49 (95%CI, 0.81–2.74) for women and 1.53 (95%CI, 0.86–2.71) for men.

Discussion

MMPs are secreted zinc metalloproteases that degrade the collagens of the extracellular matrix important in tissue remodeling and repair during development and inflammation. MMPs may alter cell cycle checkpoint controls, promote genomic instability conceivably by affecting cell adhesion ( 23 ), and contribute to tumor initiation and development by altering the cellular microenvironment that facilitates tumor formation ( 6 ). Excessive or inappropriate expression of MMP may contribute to the pathogenesis of tissue destructive processes in a wide variety of diseases including lung cancer ( 24 ). There are currently more than 20 MMPs reported that can be categorized by substrate specificity, with functional polymorphisms reported for MMP-1 , MMP-2 , MMP-3 , MMP-9 and MMP-12 . In this study, we investigated the associations between genotypes and haplotypes of functional polymorphisms of MMP-1 ( -1607 1G/2G ), MMP-3 ( -1171 5A/6A ) and MMP-12 ( -82AG and 1082A/G ), which are located in the same chromosome region and in linkage disequilibrium with each other, and the risk of lung cancer. Although we did not observe the overall associations between individual polymorphism and lung cancer risk, we found that the joint MMP polymorphisms and specific MMP haplotype ( 1G-6A-82A-1082G ) may be associated with higher risk of lung cancer, specifically among never smokers or among men, consistent with our previous findings for the MMP-1 -1607 1G/2G polymorphism ( 7 ).

In the genotype analysis where each MMP polymorphism was analyzed individually, the MMP-3 6A/6A genotype was associated with higher risk of lung cancer among never smokers ( Table II ), which is similar to the findings of coronary heart disease ( 25 ). The 6A/6A genotype was associated with lower promoter activity in in vitro assay ( 9 ), and this lower level of proteolytic activity would favor extracellular matrix deposition in lung lesions, since MMP-3 is capable of degrading proteoglycan, fibronectin, laminin and type IV collagen. We did not observe the effect of MMP-3 polymorphism (or other MMP genotypes) in ever smokers. One possible explanation is that the effect of MMP polymorphisms on lung cancer risk may be overwhelmed by the effect of cigarette smoking among smokers. Alternatively, cigarettes smoking is a major source of extracellular matrix and may induce mRNA levels of MMPs and tissue inhibitors of metalloproteases ( 26 ). Therefore, the effect of MMP polymorphisms in smokers may depend upon the balance between MMPs and tissue inhibitors of metalloprotease. Our results are not consistent with the Chinese study, where the MMP-3 6A/6A genotype was associated with a lower risk of lung cancer among smokers ( 8 ).

We did not observe an association between the MMP-12 -82A/G polymorphism and risk of lung cancer overall, or in different strata of smoking or gender. Instead, the G allele ( A/G + G/G ) of the MMP-12 1082A/G polymorphism was associated with higher risk of lung cancer among men, and not among women ( Table II ). Another study suggested that the MMP-12 polymorphisms were not associated with breast cancer risk in women ( 27 ). Although the function of the MMP-12 1082A/G polymorphism remains unknown, the MMP-12 1082A/G polymorphism is located in the coding region of the hemopexin domain and is in high linkage disequilibrium with the MMP-12 -82A/G polymorphism, which has been shown to affect MMP-12 gene expression levels ( 10 ). Although there is no report of an association between MMP-12 polymorphisms and lung cancer risk, studies have shown that MMP-12 expression levels are upregulated in lung cancer tissues ( 11 , 12 , 28 ).

In the joint effect analysis of MMP polymorphisms, where the four polymorphisms were combined based on the number of variant alleles, subjects with 4 or 5+ variant alleles had higher risk of lung cancer, especially among never smokers and men ( Table III ). Carcinogenesis is a multicellular and multistage process, and different genes that metabolize different types of collagens and stromelysins may be involved in different stages of carcinogenesis. Therefore, gene–gene joint effects may provide more complete and reliable information than the single polymorphism analysis, which may only contribute partially to the MMPs pathway.

In the haplotype analysis, the 1G-6A-82A-1082G haplotype was associated with higher risk of lung cancer among never smokers (when compared with the 1G-5A-82A-1082A haplotype). The results of haplotype analysis were consistent with the genotype analysis, where the associations were driven by the MMP-3 6A and MMP-12 1082 G alleles. Previous studies have suggested that the MMP 1G/5A haplotype was associated with increased risks of lung cancer lymphatic metastasis ( 8 ) and head and neck squamous cell carcinoma ( 29 ), when compared with the 2G/6A haplotype. In our analysis, we did not observe a significant difference between the 2G-6A-82A-1082A haplotype and the 1G-5A-82A-1082A haplotype ( Table IV ).

We acknowledge several limitations to our study. First, this is a hospital-based case–control study. Second, we only evaluated the four functional polymorphisms of MMPs located in the same chromosome region of 11q22–q23, in order to perform haplotype analysis. It is possible that functional polymorphisms of other MMP genes such as MMP-2 and MMP-9 may affect the association between these four polymorphisms and lung cancer risk. In addition, the potent proteolytic activities of MMPs are mainly regulated by the balance with specific tissue inhibitors of MMPs ( 24 ). However, adding other polymorphisms into the analysis will require a larger sample size, especially in the subgroup analysis by smoking status. Third, although we adjusted for various smoking variables in our analysis, second hand smoke exposure, alcohol consumption, diet, and environmental and occupational exposure data were not adjusted in our logistic regression models because of incomplete and missing information. Given the consistent results between gene–gene joint effects and haplotype analysis, these confounders will only probably have had mild effects on the results, if any. Fouth, the inferred haplotype frequencies were based on the selected functional polymorphisms instead of haplotype tagging SNPs, which may introduce inherited error in the analysis. Lastly, the significant associations between MMP polymorphisms and lung cancer risk were observed only among never smokers. Although the sample size of this case–control population is large, results from stratified analysis are based on relatively small sample sizes. We cannot exclude the possibility of ‘false positive’ results, especially in light of the multiple comparisons performed. However, we did observe consistent results in genotype and haplotype analyses.

In conclusion, this is the first study on the associations between the joint polymorphisms and haplotypes of MMP-1 , MMP-3 , and MMP-12 and the risk of lung cancer. Our results suggested that the joint MMP polymorphisms are associated with higher risk of lung cancer, specifically among never smokers and men. In addition, the 1G-6A-82A-1082G haplotype is associated with higher risk of lung cancer among never smokers. These results need to be confirmed by other independent studies, and further studies are needed to investigate the gene–environment interactions between MMP polymorphisms and cigarettes smoking.

We thank the following staff members of the Lung Cancer Susceptibility Group: Barbara Bean, Jessica Shin, Andrea Solomon, Thomas Van Geel, Lucy Ann Principe, Salvatore Mucci, Richard Rivera-Massa, David P.Miller; and the generous support of Dr Panos Fidias and the physicians and surgeons of the Massachusetts General Hospital Cancer Center. This study was supported by National Institute of Health Grants CA74386, CA90578, ES/CA 06409, and ES00002, Flight Attendants Medical Research Institute Young Clinical Scientist Award, and Doris Duke Charaticable Foundation and Sue's Fund.

Conflict of Interest Statement : None declared.

References

Author notes

1Department of Environmental Health, 2Department of Biostatistics, Harvard School of Public Health, 3Thoracic Surgery Unit, Department of Surgery, 4Department of Hematology-Oncology, and 5Pulmonary and Critical Care Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA