bs_bs_banner

International Journal of Urology (2014) 21, 744–752

doi: 10.1111/iju.12431

Review Article

Functional polymorphisms in the matrix metalloproteinase genes and their association with bladder cancer risk and recurrence: A mini-review Edyta Wieczorek, Wojciech Wasowicz, Jolanta Gromadzinska and Edyta Reszka Department of Toxicology and Carcinogenesis, Nofer Institute of Occupational Medicine, Lodz, Poland

Abbreviations & Acronyms BC = bladder cancer BCG = bacillus Calmette–Guérin CA = cytosine–adenine CI = confidence interval DIV = deletion/insertion variations ECM = extracellular matrix GWAS = Genome Wide Association Studies HR = hazard ratio MIBC = muscle invasive bladder cancer MMP = matrix metalloproteinases MT-MMP = membrane-type matrix metalloproteinases NMIBC = non-muscle invasive bladder cancer OR = odds ratio PM = polymorphism of the microsatellite SNP = single nucleotide polymorphisms VEGF = vascular endothelial growth factor Correspondence: Edyta Wieczorek M.Sc., Department of Toxicology and Carcinogenesis, Nofer Institute of Occupational Medicine, 91-348 Lodz, Poland. Email: [email protected] Received 4 November 2013; accepted 4 February 2014. Online publication 17 March 2014

Abstract: Molecular pathogenesis of muscle invasive bladder cancer and non-muscle invasive bladder cancer is incompletely elucidated. It is believed that matrix metalloproteinases, which are involved in the processes of uncontrolled extracellular matrix substrates degradation and participate in modulating the activity of a variety of non-matrix proteins, can contribute to carcinogenesis. Polymorphisms in the MMP genes associated with unique genomic changes in bladder cancer patients are still being investigated to discover direct links with pathophysiological mechanisms. Because of the functional polymorphisms in the MMP genes, which have a proven or likely effect on their protein expression, they could possibly affect the tumor process. The current mini-review synthesizes findings regarding the association of genetic polymorphisms in the MMP genes with bladder cancer risk and recurrence in patients. We discuss the current views on the feasibility of genetic polymorphisms in the MMP1, 2, 3, 7, 8, 9 and 12 genes as a risk, and prognostic markers for patients with bladder cancer. The majority of the research described in the present mini-review proves that the genetic polymorphism in the MMP1 (rs1799750) is the most widely studied, and suggests that the rare genotype, 2G2G, of that gene might show increased susceptibility for bladder cancer, especially among smokers. However, existing statistically significant associations between the genetic polymorphisms in the MMP genes and bladder cancer risk have not been clearly shown, and further studies are necessary in order to positively confirm them or dispel potential false hopes. Key words: bladder cancer, genetic polymorphisms, matrix metalloproteinases, recurrence, risk.

Introduction Among neoplastic diseases of the urinary system, the most common is BC. In 2008, BC was ranked as the ninth most common cancer worldwide, among both sexes, although it is approximately fourfold more frequent in men.1 The development of BC is a multistep disease associated with individual susceptibility dominated by genetic factors and heterogeneity in environmental risk factors. The major BC risk factors include: exposure to tobacco smoke, occupational exposure to industrial chemicals and schistosomiasis infection.2 Research shows that approximately 25% of BC cases represent MIBC, the other approximately 75% represent NMIBC. It is noted that MIBC with a high risk of metastasis is the most frequent cause of mortality among BC patients, regardless of the therapy applied. Most cases of NMIBC, characterized by frequent recurrences but excellent survival rates, rarely progress to MIBC. Usually just 10–15% cases of high-grade NMIBC are characterized by frequent invasion and metastasis, and might complicate the therapeutic efficacy and economical aspects.3 With regard to survival among BC patients, it has been reported that the 5-year survival rate is at a level below 10% for MIBC, and over 90% for NMIBC.4,5

MMP and their roles in pathological processes MMP, in addition to playing major roles in the physiological processes, might also participate in the pathological processes by uncontrolled resorption, remodeling and degradation of connective tissue, and modulation of the activity of a variety of non-matrix proteins. Thus, MMP are also involved in chronic inflammatory diseases (arthritis, rheumatic, ulcerative colitis, periodontitis, neurodegenerative diseases) and common diseases more dangerous to human life, such as cardiovascular diseases and cancer.6–10 744

© 2014 The Japanese Urological Association

MMP polymorphisms and bladder cancer

The activity of proteolytic enzymes was observed for the first time in 1962 by Gross and Lapiere, during a test of collagenolytic activity of different tadpole tissues.11 In 1980, Liotta et al. found that the ability of cancer cells to degrade ECM collagen affects the cancer metastasis. This effect correlated with the cancer cell metastatic potential and showed the importance of the MMP role in carcinogenesis.12 Currently, the human family of MMP is composed of 25 extracellular zinc endopeptidases, and they constitute the most important family of proteases.13 MMP, because of differences in their domain structure and preferred substrates of the ECM, have been divided into six groups and named: collagenases, gelatinases, stromelysins, matrilysins, MT-MMP and other MMP. Many MMP, except for the group of MT-MMP, are synthesized as zymogens (proMMP). ProMMP are synthesized in fibroblasts, leukocytes, monocytes, macrophages, neutrophils and endothelial cells in normal conditions, but also by tumor stromal cells in pathological states. The processes of activation and inhibition of MMP might be carried out in a few stages, such as induction of gene transcription, proenzyme activation and inhibition of active MMP.14 At the transcriptional level, the MMP gene expression is regulated by AP-1, ETS transcription factors, and a variety of extracellular factors, such as cytokines and growth factors.15 At the protein level in tissue samples, the mechanism of the MMP activity is regulated by many factors, such as proteolytic enzymes (e.g. MMP), plasmin as activator and membrane-bound RECK, endogenous tissue inhibitors of metalloproteinases, α2-macroglobulin, thrombospondins-1, −2 as inhibitors.16–18 The multistep process of carcinogenesis is proposed to be based on the release of matrix-degrading enzymes, such as MMP. MMP can cleave most of the ECM components; for example, some native collagen types, gelatin, aggrecan, entactin, tenascin, perlecan, proteoglycan, vitronectin, fibronectin and others.13 The spectrum of many MMP substrates is actually fairly similar, but some substrates can degrade more efficiently than others.19 Years of research have proven that the degradation of matrix substrates is not the only function of these enzymes. MMP not only cause tissue remodeling, but they are also involved in modifying individual cell–cell and cell–ECM communication signals. Thus, MMP can promote mechanisms that support or inhibit tumor growth by modulating the activity of a variety of non-matrix proteins. They are able to release, process or degrade cytokines, chemokines, receptors and adhesion molecules, such as fibroblast growth factor, tumor necrosis factor, VEGF, insulin-like growth factor, E-cadherin, cell surface bound Fas-L and monocyte chemotactic protein. Alterations of the activity of bioactive substrates could contribute to tumor angiogenesis, epithelial–mesenchymal transition or apoptosis.13,20–23 Therefore, it is postulated that MMP could play various opposed roles in carcinogenesis, such as tumor supporting effects (including MMP1, 2 and 7), tumor inhibiting (MMP8, 12 and 14) or possess both tumor supporting and inhibiting roles (MMP3 and 9).8,24

Role of MMP genetic polymorphisms as a tumor biomarker The tools currently used to diagnose BC are unsatisfactory, because they fail to unambiguously detect the BC at high risk of © 2014 The Japanese Urological Association

metastasis formation in future years, which is extremely important in cancer mortality. Thus, the limited ability to predict biological aggressiveness of malignancies, disease recurrence and survival probabilities urges researchers to carry out further investigations intended to improve the staging system of BC that they currently use. The tumor-specific protein biomarkers can show the effects of the regulatory process of angiogenesis, cell proliferation, cell-growth fraction, cell cycle regulation, apoptosis, cell adhesion, epithelial–mesenchymal transition or other processes associated with carcinogenesis. Hence, biomarkers at the gene or protein level can be classified for their correlation with biological processes involved in cancerogenesis into the following categories: angiogenesis markers (VEGF, thrombospondin1), cell growth and signal transduction markers (Ki-67, human epidermal growth factor receptor 2, galectin-325,26), cell cycle markers (p53, p21, pRB, p27, and the cyclins D1 and E127–31), apoptosis markers (caspase-3, Bcl-2, Bax and survivin32–34), migration and invasion markers (E-cadherin and MMP35,36). Involvement in the breakdown of connective tissue barriers has suggested that the MMP at the gene and protein level might be related to the migration and invasion biomarkers. MMP belonging to six different groups of MMP were analyzed as tumor biomarkers among breast, pancreatic, lung, colorectal, ovarian, prostate, brain and also bladder cancer patients. Studies intended to find BC biomarkers, as a result of the location of the tumor, were mainly concerned with the differences between urine samples collected from BC patients and healthy controls.3,36–39 Interaction of MMP with other biomarkers, such as E-cadherin, VEGF in tissue environment and other non-genetic influences, are not yet completely understood, and might also limit its usefulness as a molecular biomarker. In general, molecular biomarkers may represent essential individual cancer susceptibility factors like common SNP, haplotypes, DNA modifications, cytogenetic rearrangements or RNA expression levels. However, searching for MMP as useful bladder cancer biomarkers is still an open problem.40–42 New regions associated with BC have been identified by GWAS, although none of them involved MMP genetic polymorphisms.43,44 The association reports from other platforms determined variant alleles that were in full linkage disequilibrium, but were not confirmed in GWAS. It should be kept in mind that GWAS have disadvantages, and important SNP might not be significant at a genome-wide level. Furthermore, GWAS are using total cases of cancer versus control population, so important cancer phenotypes are not taken into consideration. The approach applied in the GWAS has a limitation; it analyses a single SNP and does not take into account the association with other polymorphisms in the genome, or occupational exposures.45,46 Further studies are required to explain the clinical importance of the genetic variants involved in cancer. Additionally, future research should focus more on polymorphism problems, such as dependencies between SNP across different chromosomes to help explain the importance of MMP polymorphisms. It is thought that when information on human genes and genetics-related techniques becomes more applicable and available, the SNP information might be used in practice. 745

746

2G allele has greater transcriptional activity Possible influence on transcriptional activity C allele has greater transcriptional activity C allele has greater transcriptional activity 5A allele has greater transcriptional activity Glu45Lys replacement Ala362Ala replacement G allele has greater transcriptional activity Lys87Glu replacement T allele has greater transcriptional activity T allele has greater transcriptional activity Gln279Arg replacement Pro574Arg replacement Arg668Gln replacement Ala20Val replacement >24 CA repeats have greater MMP9 expression than ≤24 CA A allele has greater transcriptional activity rs1799750 rs514921 rs243865 rs2285053 rs3025058 rs6796720 rs520540 rs11568818 rs1940475 rs11225395 rs3918242 rs17576 rs2250889 rs17577 rs1805088 – rs2276109 1G/2G A/G C/T C/T 5A/6A A/G A/G A/G A/G C/T C/T A/G C/G G/A C/T ≤24/>24 CA A/G Supporting Inhibiting

Supporting and inhibiting

Inhibiting

MMP9

MMP12

Supporting and inhibiting MMP3

MMP7 MMP8

Supporting MMP2

SNP SNP SNP SNP SNP PM SNP

Promoter Promoter Promoter Promoter Promoter Exon 2 Exon 8 Promoter Exon 2 Promoter Promoter Exon 6 Exon 10 Exon 12 Exon 1 Promoter Promoter Supporting MMP1

DIV SNP SNP SNP DIV SNP SNP SNP SNP

Effect dbSNP ID Major/Minor Alleles Region Variation class MMPs exert cancer effects Gene

Genetic polymorphisms of MMP genes examined in case–control studies of BC Table 1

Our knowledge of the human genome, enhanced by technological advances in genotyping methodologies and statistical methods to analyse genotype and haplotype data, allows us to expand the field of exploratory biomarkers. The identification and characterization of functional genetic polymorphisms that have a proven or likely effect on gene and protein expression highlights their possible impact on the process of carcinogenesis. Furthermore, it is worth noting that the same genetic background of SNP in the MMP genes might have a different impact in patients depending on the specific cancer site. Among the studies describing MMP as molecular biomarkers in BC patients, in most cases the correlations between gene or protein expression profiles and pathological parameters or prognosis were assayed, followed by studying the association of genetic polymorphism profiles with BC risk, progression and survival.47 Functional and non-functional genetic polymorphism might be used as genetic marker associated with cancer disease. It has been reported that haplotypes of two or more SNP could be associated with phenotype, even if they are made up of SNP without any functional effect on the gene products. That association is the result of an interaction between the variety of polymorphisms and the collective influence of several polymorphisms on gene transcription regulation. Additionally, haplotype analysis provides a greater statistical power than analysis based on individual SNP.48,49 It appears that the combination of two forecasts, such as genetic polymorphism profiles and current status of RNA expression levels, could be useful in assessing the BC recurrence or BC risk. A systematic literature review was carried out in the PubMed Medline and Google Scholar databases up to December 2013 without limitation of year of publication or journal. The PubMed search was made using the followings keywords for articles in English: “matrix metalloproteinase” OR “matrix” AND “polymorphisms” OR “single nucleotide polymorphism” AND “bladder cancer” by which we found 237 articles. Of these articles, we selected those abstracts that specified data on the genetic polymorphisms of the MMP genes in bladder cancer; seven articles were selected. To date, only a few studies have attempted to establish the association between genetic polymorphisms in the MMP genes, and BC risk, invasiveness and recurrence. Investigations have been carried out for MMP belonging to five groups, such as collagenases (MMP1 and MMP8), gelatinases (MMP2 and MMP9), stromelysins (MMP3), matrilysins (MMP7) and group of other MMP (MMP12). Most of the examined genetic polymorphisms are SNP located in the gene promoter region, whereas DIV are in the minority. The following SNP in MMP genes present wellcharacterized functional significance, correlated with susceptibility, invasiveness or recurrence of various cancer site types.50–59 As shown in Table 1, the functional genetic polymorphism involving DIV in the promoter region of MMP1 1G/2G (rs1799750) or MMP3 5A/6A (rs3025058), lead to higher levels of gene transcription for 2G or 6A alleles, respectively.60–63

No. association studies

Genetic polymorphisms in the MMP genes as a modifier of BC risk, invasiveness and recurrence

4 1 3 1 2 2 1 2 1 2 2 2 2 2 1 1 2

E WIECZOREK ET AL.

© 2014 The Japanese Urological Association

MMP polymorphisms and bladder cancer

SNP in the promoter regions with specific rare alleles are associated with higher transcriptional activity compared with those with lower transcriptional activity: MMP2 T versus C allele (rs243865, rs2285053), MMP7 G versus A allele (rs11568818), MMP8 T versus C allele (rs11225395), MMP9 T versus C allele (rs3918242) and also with specific frequent allele MMP12 A versus G allele (rs2276109).64–70 There have been reports that the changes in transcriptional activity might be due to the facts that allele insertion/deletion affects the E-26 specific domain transcription factor binding site in the MMP1 gene,62 or allele substitution that abolishes the Sp1 and AP-1 transcription factor binding site, in the MMP2 and MMP12 gene, respectively.65,66 Additionally, there is speculation about the modulation role of CBG-02 protein in MMP8 C/T (11 225 395) transcription, the binding site of which is potentially contained in −799 oligonucleotide.68 Non-synonymous SNP due to allele substitution result in an amino acid changed in the codon. The investigation includes: substitution of A/G in MMP3 (rs679620 and rs520540), which leads to Glu45Lys and Ala362Ala replacement in exon 2 and 8, respectively. Substitution of A/G in MMP8 (rs1940475) results in change Lys87Glu in exon 2. Although substitution of A/G in MMP9 (rs17576) results in change Gln279Arg in exon 6, C/G (rs2250889) results in change Pro574Arg in exon 10, C/T (rs1805088) results in change Ala20Val in exon 1 and G/A (rs17577) results in change Arg668Gln in exon 12. The PM CA repeat in the promoter region of the MMP9 gene probably plays a role in the susceptibility to some diseases, such as carotid atherosclerosis, intracranial aneurysm and Behçet’s disease.71–73 Some authors have reported that PM CA repeat in the promoter region of MMP9 could affect its expression. Greater MMP9 expression has been observed for PM length 24 CA repeats relative to less than and equal to 20 CA repeats.74–76 The reviews of the in vivo and in vitro studies are often consistent with the importance of MMP SNP in regulation of gene expression, activity of MMP, but also susceptibility to development of cancer.77,78

Association of genetic polymorphisms in the MMP genes with BC risk and invasiveness As we have already stated, the transcriptional activity of MMP, which might depend on the genetic polymorphisms, could be required for the tumorigenic process; therefore, not only risk but also characteristics, such as tumor migration, invasion and metastatic spread, were included in the case–control studies (Table 2). To explore whether the effects of MMP polymorphisms were associated with BC risk, logistic regression models were used. OR and 95% CI were estimated. For all studies, a P-value less than 0.05 was considered statistically significant. Kader et al. observed no significant association between genetic polymorphisms in the MMP2 (rs243865), MMP3 (rs3025058, rs6796720), MMP8 (rs1940475), MMP9 (rs3918242, rs17576, rs2250889, rs17577) and MMP12 genes (rs2276109), and BC risk in a Caucasian population.79 Srivastava et al. also showed that genetic polymorphisms in the MMP3 (rs6796720) and MMP9 genes (rs17577) were not associated with BC risk in a Northern Indian population.85 However, in contrast to Kader et al., they observed a significant associa© 2014 The Japanese Urological Association

tion of genetic polymorphisms in the MMP2 (rs243865), MMP3 (rs3025058), MMP9 genes (rs17576 and rs2250889) with BC risk.79 Srivastava et al. also reported an association between genetic polymorphisms in the MMP2 (rs2285053), MMP7 (rs11568818) and MMP8 genes (rs11225395), and no association of genetic polymorphism in the MMP3 gene (rs520540) with BC risk.82,84,85 Our study confirmed no association of genetic polymorphisms in the MMP2 (rs243865), MMP7 (rs11568818), MMP8 (rs11225395), MMP9 (rs3918242) and MMP12 genes (rs2276109) with BC risk in Caucasian population.83,86 Kader et al. showed no significant correlation between 2G/2G genotype of MMP1 (rs1799750) and BC risk.79 However, the result of the studies carried out by Tasci et al., Srivastava et al. and Wieczorek et al. showed that MMP1 (rs1799750) is significantly associated with BC risk, namely the 2G2G genotype with OR 2.79 (95% CI 1.53–5.60), 2G2G genotype with OR 3.04 (95% CI 1.71–5.39) and 2G1G + 1G1G genotype with OR 0.62 (95% CI 0.39–0.98), respectively.82,83,86 The aforementioned researchers also categorized in their works the patients into subgroups, such as NMIBC versus MIBC, TaG1 versus TaG2–3 and T1G1–3, TaG1 versus T2, and G1 versus G2–3 to assess the association of polymorphisms in the MMP genes with BC invasiveness. Patients with low-risk NMIBC were the reference group. Only Kader et al. on stratified analysis found a statistically significant association of the MMP9 ≥ 24 CA repeats, OR 2.60 (95% CI 1.07–6.26) or for MMP12 (rs2276109) GG alleles, OR 4.59 (95% CI 1.21– 17.32) with MIBC risk.79–86

MMP gene haplotype associations with BC risk The associations of specific common haplotypes of MMP and BC risk were studied by Kader et al. and Srivastava et al. Furthermore, Kader et al. estimated MMP diplotype, combinations of the two highest transcriptional efficiency haplotypes. The researchers suggest that there are significant associations with the haplotype and risk of BC. Srivastava et al. investigated the MMP1 haplotypes (rs514921–rs1799750), and found significantly overrepresented G-1G and G-2G alleles in patients with BC with OR 1.7 (95% CI 1.20–2.41, P = 0.003) and OR 2.62 (95% CI 1.68–4.09, P < 0.001), respectively.82 Similarly, analysis of the MMP2 haplotype (rs2285053–rs243865) showed that C-T and T-C alleles were associated with BC risk, OR 1.94 (95% CI 1.35–2.80, P < 0.001) and OR 1.48 (95% CI 0.10–2.21, P = 0.051), respectively.84 Haplotypes of the MMP3 (rs3025058–rs6796720–rs520540) A-A-A alleles, MMP9 (rs17576–rs2250889) A-G and G-G alleles were also significantly associated with BC risk OR 1.99 (95% CI 1.11–3.60, P = 0.022), OR 2.00 (95% CI 1.35–2.97, P = 0.001) and OR 2.97 (95% CI, 1.65–5.37, P < 0.001), respectively.85 Kader et al. studied the MMP9 haplotype and diplotype. They analyzed six alleles (rs3918242, rs1805088, rs17576, rs17577, rs2250889 and PM of CA repeats). Patients (including NMIBC and MIBC) with the most frequent number of the rare alleles showed differences relative to controls, OR 2.76 (95% CI 1.14– 6.68). Associations were also confirmed by MMP9 diplotype analysis; the most frequent of the rare diplotype was significantly associated with risk of BC, OR 6.99 (95% 747

748

Co/Cab

NS

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

199/100/95e 37/111/52d 556/311/242 311/242 37/111/52d 537/294/232 294/232 37/111/52d 37/111/52d 37/111/52d 199/113/102e 546/307/236d 307/236 37/111/52d 199/100/95e 560/315/243 315/243 199/100/95e 521/298/221 560/315/243 298/221 531/299/229 299/229 37/111/52d 560/314/242 37/111/52d

545/311/234 311/234 37/111/52d 557/314/241 314/241

NA Yes NS NA Yes NS NA Yes Yes NA NA NS NA NA NA Yes NA NA Yes Yes NA Yes NA Yes Yes Yes

Yes NA NA NS NA

199/100/95e

NS NS NS

545/305/236 305/236 37/111/52d

NS NA Yes

NA

NS NS

199/100/95e 37/111/52d

NA Yes

– –

GG 4.59 (1.21–17.32)g GG 4.82 (0.97–23.90) AG+GG 1.50 (1.00–2.28) NS



– – 44/34

– – – – – – – – 44/34 – 44/34

– – 44/34 –

44/34 –

– – 44/34 – – 44/34 44/34

– 44/34

– – 44/34

– 44/34

– – – 44/34

T/Uf

NA

NA NA

NA NA NS

NA NA NA NA NA NA NA NA NS NA NS

NA NA NS NA

NS NA

NA NA 5A6A 2.01 (0.98–4.12), 0.055 NA NA NS NS

NA NS

NA NA CT 4.32 (1.51–12.39), 0.006 CT+TT 2.06 (1.01–4.18), 0.047

NA NA NA 1G2G 0.28 (0.10–0.80), 0.017 1G2G+2G2G 0.39 (0.16–0.95), 0.039 NA NS

Alleles with risk and HR (95% CI), P*

Association with recurrencea

NS NS NS

NS NS NS NS ≥24–2.60 (1.07–6.26)g

Functional polymorphisms in the matrix metalloproteinase genes and their association with bladder cancer risk and recurrence: a mini-review.

Molecular pathogenesis of muscle invasive bladder cancer and non-muscle invasive bladder cancer is incompletely elucidated. It is believed that matrix...
157KB Sizes 0 Downloads 3 Views