Matrix Metalloproteinase-9 (MMP-9) polymorphisms in patients with cutaneous malignant melanoma

Cutaneous Malignant Melanoma (CMM) causes over 75% of skin cancer-related deaths. Melanoma incidence has been rising steadily, with 62,190 new cases and 7,910 deaths estimated for 2006 in the United States [1]. It is becoming clear that genetic factors contribute to the development and/or progression of the disease, as well as the exposure to ultraviolet light (UV) and tumor microenvironment.

Invasive primary melanomas are aggressive and prone to metastasis. Proteolytic degradation of the extracellular matrix (ECM) constituents may release and activate ECM-bound cytokines and ECM fragments (matrikines) that modulate cell growth, migration and angiogenesis [2, 3]. Under physiological conditions, matrix metalloproteinase (MMP) expression is low in most cells; however, many tumor types show a dramatically overexpression of these enzymes. This, in turn, produces an overactive proteolysis of the ECM and basement membrane, promoting cancer invasion through a variety of biochemical, biophysical, and signal transduction mechanisms.

Liu et al found that the MMP-2 and MMP-9 protein levels were higher in breast tumor tissues when compared to the corresponding normal tissue (p < 0.01) and that MMP-2 was significantly increased in larger tumors (p < 0.0001), and in metastatic lesions (p < 0.05) [4]. MMP-9 expression was significantly higher in colorectal tumors when compared to the adjacent normal tissue (p < 0.05) [5]. In melanoma, there is some evidence suggesting a role of MMPs in the progression of the disease. It has been demonstrated that invasive melanoma cell lines show higher MMP-9 expression and higher activity when compared to non-invasive cell lines [6, 7]. In primary melanomas, MMP-9 is variably expressed in radial but not in the vertical growth phase and the de novo expression seems associated with early invasion [8]. MMP-2 was evident by immunohistochemistry in malignant melanoma lesions, but not in benign and atypical nevi [9]. Corte et al found an association between MMP-13 expression with mitotic index (p = 0.002) in CMM [10]. Circulating blood levels of MMP-9 may be useful in predicting progression in patients with melanoma although it has been suggested that serum levels contain higher levels of proteases released during the clotting process, and therefore, the results obtained by Nikkola et al should be validated using plasma samples [11–14].

Some polymorphisms are known to influence gene expression, protein activity, protein stability and protein-protein interactions, and some of them are associated with an increased risk for cancer and other diseases. To date, and to the best of our knowledge, there are no reports on associations between polymorphisms in MMP-9 and melanoma.

Polymorphisms in the promoter of MMP-9 have been implicated in the regulation of gene expression and susceptibility to various diseases. Two out of five identified sequence variants are functionally relevant: a single nucleotide polymorphism at (-)1562 bp (C → T) and a (CA)_n microsatellite at position (-)131 bp. The C to T substitution at position (-)1562 results in a loss of binding of a nuclear protein and an increase in transcriptional activity in macrophages [15]. Similarly, the highest promoter activity has been observed in reporter constructs containing 21 or 23 (CA) tandem repeats, suggesting that the number of repeats modulates the transcriptional activity [15].

To test the hypothesis that polymorphisms in the MMP-9 gene [GenBank: AF538844, OMIM: 120361] may influence melanoma progression, we examined seven different polymorphisms in a cohort of 1002 patients with melanoma in order to evaluate germline genetic variants in the MMP-9 gene and their association with progression and known risk factors of melanoma.

We included 1002 patients with melanoma stages 0 (in situ) to IV in the study. Ninety-five percent of these were cutaneous malignant melanoma patients; the rest included mucosal melanomas, other non-cutaneous sites, or unknown primary sites. Ninety-six percent were Caucasians followed by Hispanic (1.1%), black non-Hispanic (1.1%), and Asian/Indian (0.3%); fifteen had missing information on ethnicity and one refused to answer the question about race (1.5%). The age at diagnosis ranged from 5 to 89 years old (mean = 54 and median = 55).

Evaluation of non-synonymous SNPs

There are 10 non-synonymous MMP-9 variations in the dbSNP (build 123). These variations are located in the activation peptide domain (A20V, E82K), catalytic domain (N127K, D165N, L447P, L449P), fibronectin type II domain (R279Q), and hemopexin domain (P574R, R621K, R668Q). Currently, there is no complete 3D structure of MMP-9 in any known biologically important state (precursor, active, bound to TIMP inhibitor). However, there are 3D structures of different fragments of MMP-9 and its homologs, including the complexes with TIMPs.

Therefore, we derived a synthetic structure of MMP-9 by 3D superposition of different structural fragments of MMP-9 and its close homolog MMP-2. Although this synthetic structure was not optimized for residue-residue contacts, it showed the MMP-9/TIMP binding site. Structure-based analysis of residue-residue interactions and analysis of protein domain family alignments suggested that five of the 10 non-synonymous SNPs -N127K, D165N, R279Q, P574R, and R668Q- would be appropriate candidates that might affect MMP-9 function and/or interaction with its inhibitors. None of the 5 polymorphic residues studied localizes to the MMP-9/TIMP binding interface (Figure 1), however all of these residues have specific interactions between their side-chain and surrounding residues, which are abolished as a result of an amino acid change. Our analysis is consistent with the results obtained by PolyPhen (Polymorphism Phenotyping [29]).

There are several reports showing different MMP expression patterns in melanoma [6, 33–35] but, to the best of our knowledge, this is the first genetic study to examine the role of polymorphisms in MMP-9 in melanoma progression and other melanoma risk factors. We hypothesized that the MMP-9 polymorphisms might alter the expression and activity of the enzyme, increasing ECM degradation and invasion, leading to melanoma progression.

Gene transcription is the primary point of regulation of MMPs; therefore, sequence changes in the promoter may have important implications for the transcription, and in turn, for the protein levels and cell physiology. This led us to assess previously described functional polymorphisms within the promoter region of MMP-9: (-)1562 C > T and (-)131 (CA)_n.

The polymorphism at position (-)1562 changes the promoter activity of MMP-9 because the T allele abolishes a binding site for a transcription repressor [36]. Although the less active C allele was expected to be present in less aggressive lesions, we found the opposite results. A similar association between this SNP and breast cancer was found by Grieu et al, where the T allele correlated with non-ductal histology, positive estrogen receptor and absence of TP53 mutations in breast cancer [37]. However, Matsumura et al found significant associations between this allele and the invasive phenotype of gastric cancer [38]. These divergent results indicate that there might be other factors that play a role in the regulation of the MMP-9 transcription and/or activity, such as other regulatory elements, promoter methylation or even other steps such as secretion and type of cells under study [39, 40].

The dinucleotide repeat shows a bimodal distribution (reviewed by Van den Steen et al [15]), with the most prevalent allele being (CA)₁₄ and a second peak at (CA)_21–23 in American white, Finnish, Swedish, Belgian, African-American and southern English population. On the other hand, the Japanese population show the highest incidence of (CA)₂₁ followed by (CA)_{> 21}. Similar to the SNP at position (-)1562, the length of the microsatellite may influence the transcriptional activity of the gene due to its close localization to the transcriptional start site and several transcription factor binding sites and its length-dependent interaction with nuclear proteins [31]. However, data are inconsistent with regard to the relationship between the length of the microsatellite and the promoter activity [30–32]. Our analysis showed the same distribution and that short alleles were present more frequently in patients at higher risk for melanoma.

There are no published data on the effect of the coding polymorphisms in the MMP-9 activity therefore; it is not possible to hypothesize which variants could be associated with advanced stages of the disease and/or shorter progression-free survival. In this analysis, we found that the reference alleles were more frequent among patients with higher risk for melanoma development (tendency to sunburn, family history of melanoma), more advanced disease (higher melanoma stage, presence of intransit metastasis), and desmoplastic melanoma. However, after adjustment for age, sex, phenotypic index, moles, and freckles only four of these associations remained significant. The 279QQ genotype was associated with the presence of intransit metastasis (p_adjusted = 0.02) and tendency to sunburn (p_adjusted< 0.01); the 574R allele was more common among patients with melanomas in the extremities (p_adjusted = 0.02); and 668RR was associated with the tendency to sunburn (p_adjusted = 0.02). Our genotyping results suggest that the "reference" MMP-9 enzyme is more active than the "variant" product. These observations agree with the results of a very recent case-control study conducted in Japanese that suggests a link between the R279Q polymorphism and malignant potential of renal cell carcinoma [41].

Our study subjects included ten patients who were younger than 18 years old at diagnosis. In this particular group, the etiology of the disease may differ from the etiology of melanoma in adults. However, other than the presence of spitzoid melanomas in 4 of 10 of these patients, we did not find any additional withstanding clinicopathological or genotypic characteristic and therefore, they were not excluded from the analysis.

Interestingly, variant alleles (-)1562T and 668Q were more frequent among patients with primary melanomas localized to non-cutaneous sites, mainly mucosal melanomas (90%). Whether there is a significant association between these SNPs and mucosal melanoma remains to be determined in a larger group of patients with this type of the disease.

We examined 437 correlations between clinicopathological variables and 5 MMP-9 polymorphisms. Although we found 39 statistically significant associations, after adjustment, only 4 of these associations remained significant and displayed no clear pattern, consequently the correlations may simply be due to chance. Thus, we conclude that this study does not provide strong evidence for further investigation into the role of the MMP-9 variants in melanoma progression.