DNA methylation profiles delineate epigenetic heterogeneity in seminoma and non-seminoma

Testicular cancer is the most commonly diagnosed malignancy among young men aged 15 to 40 years, and its incidence has doubled in the past 40 years (Chia et al, 2010). An annual increase of 3–6% is reported for Caucasian populations. In the United States, approximately 8500 newly diagnosed testicular cancer cases and 350 deaths were expected in 2010 (Jemal et al, 2010). Testicular germ cell tumours (TGCTs) represent over 95% of the testicular cancers and histopathologically are classified into two major groups of seminomas (SE) or non-SEs (NSE), frequently occurring as mixed tumours (Holmes et al, 2008). Histologically, SE resembles primordial germ cells/gonocytes, whereas NSE shows somatic, primitive embryonal or extra-embryonal differentiation (Horwich et al, 2006). Carcinoma in situ or intra-tubular germ cell neoplasia unclassified are believed to be the origin of both SE and NSE (Rajpert-De Meyts, 2006). They have a primordial germ cell/gonocyte origin, and it is important to emphasise that epigenetic reprogramming is known to occur during germ cells development. The treatment of testicular cancer includes orchiectomy, and according to the metastatic condition surveillance, or adjuvant retroperitoneal surgery, radiation and chemotherapy are offered. The decision whether to choose the adjuvant treatment regimen is based only on clinical parameters, leading to around 25% development of metastasis in patients on surveillance or unnecessary adjuvant treatments in 20% of the patients. Genetic understanding of each tumour will enable to better tailor the treatment. The importance of epigenetic alterations has long been demonstrated in carcinogenesis. Several studies have shown that methylation-associated silencing inactivates certain tumour suppressor genes (TSGs) as effectively as mutations and is one of the cancer-predisposing hits described in Knudson's two hit hypothesis. Promoter methylation, the most studied epigenetic alteration, is also increasingly recognised as a major mechanism of gene inactivation during TGCT progression (Manton et al, 2005; Ellinger et al, 2009). CG dinucleotide-rich regions, also known as CpG islands, in or near the proximal promoter regions of genes are targets for DNA methylation, leading to effective transcriptional silencing (Herman, 1999). In normal cells, CpG methylation is an important mechanism for regulating gene expression, whereas in cancer cells, aberrant promoter methylation (hypermethylation) can lead to abnormal gene silencing, including repression of TSGs. Another relevant aspect is that environmental and endogenous conditions can influence the epigenetic processes. The exposure to certain environmental risk factors may be related to the onset of cancer or to participate in carcinogenesis (Brait et al, 2009). There are many evidences that endogenous factors (such as oestrogens or androgens inhibitors) exposure may lead to cancer (Godmann et al, 2009), and endogenous factors (like hormonal stimuli) induce methylation of promoter region of certain genes (Kutanzi et al, 2010). Even though TGCT has high survival rates due to good responses to therapy, significant consequences of multimodality therapies exist with regard to general health, secondary late malignancies, reproduction and economic productivity (Sokoloff et al, 2007). A great need for understanding TGCT biology still exists to help curb its increasing incidence and potentially adopt effective prevention strategies. Several genetic alterations have been shown in TGCTs. For example, abnormalities in the short arm of chromosome 12 (Mostert et al, 1998), as well as loss on chromosomes 1, 3, 5, 9, 11, 12q, 13q, 17p and 18q (Mathew et al, 1994; Lothe et al, 1995; al-Jehani et al, 1995; Honorio et al, 2003; Oosterhuis and Looijenga, 2005) have been reported. But major TSGs having a role in TGCTs are yet to be identified (Honorio et al, 2003). One of the most intriguing questions in the biology of TGCTs is how such distinct histological tumour subtypes (SE and NSE) can arise from the same cell type (cell of origin). Both subtypes exhibit similar cytogenetic abnormalities (Lutzker and Barnard, 1998). So far, molecular alterations that could distinct SEs from NSEs have not been clear yet. A pioneer study, performed in 1991, has shown that hypermethylation was abundant in NSEs, but not in SEs (Peltomaki, 1991). In 2002, Smiraglia et al (2002) demonstrated significant epigenetic differences between SEs and NSEs by a global methylation approach. In addition, studies of the X chromosome have demonstrated little or no methylation in SEs, and increased methylation in NSEs, particularly in more highly differentiated NSEs (Looijenga et al, 1997). Netto et al (2008) analysed global methylation status by immunohistochemistry and concluded that SEs cells generally retain the lack of methylation that occurs due to normal developmental erasure of methylation marks, whereas NSEs do not, thus showing more methylation. This was independently shown as well (Wermann et al, 2010). In the present study, we used a candidate gene approach to investigate the methylation profile of 57 primary TGCT (of which 43 were SEs and 14 were NSEs) and 23 normal testis (NT) by quantitative methylation-specific PCR (QMSP). QMSP has been successfully used in other tumour models and has the benefit of providing accurate and precise data regarding the level of methylation in the various tumours. Six of the genes we evaluated, including ARF, APC, MGMT, RAR-β2, CCNA1 and hMLH1, were previously shown to be aberrantly methylated in TGCT (Muller et al, 2000; Koul et al, 2002; Honorio et al, 2003; Olasz et al, 2005; Lind et al, 2007). The remaining nine genes we studied has been found to be methylated in other cancer types, but not yet tested in TGCT; these genes include AIM1, PGP9.5, S100A2, ER-α, ER-β, MCAM, VGF, FKBP4 and SSBP2. Among the later genes, four genes (MCAM, VGF, FKBP4 and SSBP2) were recently discovered by our group, using a pharmacological unmasking strategy in other cancer types (Hoque et al, 2008). We compared the promoter methylation profiles of SEs and NSEs along with NT to better understand the role of epigenetic silencing in testis tumourigenesis. Relationships between methylation values and clinicopathological parameters were further assessed.