Mmp-1 gene polymorphism, g6pd phenotype and their interaction in lymphoma risk

Keywords: lymphoma; MMP-1 polymorphism: G6PD polymorphism. Introduction. Over-expression of the matrix metalloproteinase (MMP) genes is involved in growth and metastasis of several solid tumors by degrading the extracellular matrix. A single guanine insertion polymorphism (2G) in the MMP-1 promoter region is associated with an increase in MMP-1 transcription and local expression of MMP-1. G6PD is a key enzyme in the cholesterol synthesis, a basic component of cell membranes, and G6PD gene polymorphisms associated with deficient enzyme activity are specially prevalent among the Sardinian male population. We inquired into the association between the 2G insertion type MMP-1 polymorphism and risk of lymphoma, as well as with its interaction with the G6PD polymorphism in Sardinia, Italy. Methods. We used polymerase chain reaction (PCR) multiplex to identify the MMP-1 -1607 1G/2G polymorphism in 154 male lymphoma cases identified in 1998 – 2004 in two hematology units in Sardinia, Italy, and 182 male population controls, frequency matched to cases by residence area and 5-year age-group. GdMed+, the most widespread G6PD variant associated with enzyme deficiency in Sardinia, was tested with PCR, according to Kurdi Haidar in a subsample of 49 cases and 32 controls. Due to its strong agreement to the genotyping results (kappa =0.88; p = 0.0000)), the self reported G6PD phenotype was used in the analysis. The OR associated with the MMP-1 polymorphism under the dominant and co-dominant models and that associated with the G6PD phenotype, along with the respective 95% confidence intervals, was calculated with unconditional logistic regression adjusting by age and residence area. The gene-gene interaction was formally tested with the likelihood ratio test for interaction. Results. Risk for all lymphoma, and the B-cell lymphoma and non Hodgkin lymphoma groupings were very similar and not related to the MMP-1 gene polymorphism (OR = 0.9; 95% CI 0.5, 1.9 under the dominant model), while risk associated with the G6PD deficient phenotype was not significantly decreased, particularly when the MMP-1 covariate was included in the regression model (OR = 0.5; 95% CI 0.2,1.0). Introducing the interaction term in the regression model did not show a significant reduction in its residual deviance (c2 = 1.40; GL 2; NS). However, it might be worth reporting that absence of the homozygosis for guanine insertion in the MMP-1gene among subjects with the G6PD deficient phenotype was associated with a low risk of developing lymphoma (OR =0.6; 95% CI 0.1,2.2). Conclusion. Our study did not have enough statistical power to detect an association between the MMP-1 -1607 1G/2G polymorphism and risk of lymphoma, nor a significant interaction with the G6PD phenotype in modifying lymphoma risk. Due to the known role of the metalloproteases in growth and metastatic spread of various tumors, survival of lymphoma patients might be more likely affected. Studies are in progress in this regard.