DNA methylation is considered as an important contributor to the development of HCC. However in spite of the growing body of evidence which suggests that it is an early event in hepatocarcinogenesis [32–35], identification of important methylation events in diseased conditions of liver (CH & LC), which might predispose affected patients to the development of HCC later in life, has not been studied extensively. Recent cell culture studies point towards a possible direct role of HCV core protein in modulating host cell methylome [36, 37]. Therefore it will be important to investigate whether patients infected with HCV, yet without HCC, harbor any aberrations in DNA methylation patterns in their livers. Most of the studies aiming to identify early changes in methylation patterns make use of histologically non-malignant liver tissues obtained from HCC patients [32, 33, 38, 39]. However, as viral induced HCC develops after decades of infection, this strategy might not prove to be adequate because it will be difficult to differentiate between “driver” and “passenger” methylation  aberrations in such long infected hepatocytes, even if they appear histologically non-malignant. Therefore, in this study we set out to explore possible methylation aberrations in HCV infected non-cancerous patients (CH & LC) with the aim to identify DNA methylation markers which can be exploited for surveillance and carcinogenic risk estimation in high risk populations as well as development of novel drug targets.
Our analysis revealed that both of the Wnt inhibitors included in this study display an increased level of methylation in their promoter regions in HCV infected patients even at the very early stages of hepatic disease (Figures 4 & 2, Table 3). Chronic hepatitis tissue samples exhibited significantly higher levels of methylation as compared to normal controls (p = 0.0136 & 0.0084 for SFRP2 & DKK1 respectively, Mann–Whitney U test). CH patients included in our study had minimal liver damage and were ideal candidates for interferon therapy. Occurrence of promoter hypermethylation in such early stage liver disease, points to the fact that this epigenetic aberration might occur much earlier than previously thought . Similarly, the other non-cancerous disease cohort i.e. LC also showed significantly higher methylation levels as compared to normal controls (see Table 3 for more details). It is important to note here that although the methylation percentages in CH and LC groups differed significantly from normal controls, the maximum observed percent methylation did not exceed 16% for both the regions studied, in both of the diseased cohorts. This apparently “low” methylation can be attributed to the fact that sample DNA was obtained from a heterogeneous population of infected and non-infected hepatocytes. As previous studies have reported that only a very small percentage of hepatocytes (2-30%) are actually infected with HCV [41, 42], the said “low” methylation observed in our experiments appears to be in agreement with findings of these studies.
Interestingly SFRP2 promoter exhibited a pattern of progressive increase in DNA methylation that coincided with the various stages of liver disease, i.e. chronic hepatitis and cirrhotic samples exhibited methylation levels which were statistically higher than normal controls (N Vs CH p = 0.0136, N Vs LC p = 0.0084, Mann Whitney-U test). Similarly, HCC sample cohort showed significantly higher methylation as compared to chronic hepatitis as well as liver cirrhosis samples (CH Vs HCC p < 0.0001, LC Vs HCC p = 0.0013). This apparent increasing trend might be because promoter methylation mediated silencing of SFRP2 at the early stages of liver disease provides a definitive selective advantage to affected cells and hence helping in their clonal expansion .
One striking observation that emerged from our results was that although DKK1 promoter methylation was found to be significantly higher in all the three diseased groups as compared to normal controls (p = 0.001; Kruskal-Wallis test), the three said groups did not differ from each other considerably (See Table 3 for details). Similar results were obtained in an earlier study by Yang et al.,  which reported that DKK1 promoter methylation in HCC tissues did not differ significantly from surrounding non-tumor tissue or from independent cirrhotic liver samples. However, to the best of our knowledge, it is for the first time that “baseline” methylation in completely normal liver tissues has been reported and compared with that of various disease cohorts. Significance of this comparison is apparent from our own results where we observed a disease dependent hypermethylation as well as those of Yang et al. because the said study also found a certain level of hypermethylation in DKK1 promoter in HCC free LC patients (See Figure 3 in the corresponding reference) . Our results speak strongly in favor of use of infection free normal controls in order to put observed methylation percentages in true context.
In a recent study by Nishida et al.
, a subset of early HCC related TSGs (HIC1, GSTP1, SOCS1, RASSF1, CDKN2A, APC, RUNX3, and PRDM2) were also found to be hypermethylated in HCV infected CH patients. However, interestingly SFRP2 promoter methylation was only found in highly progressed HCCs and hence was not analyzed in non-cancerous HCV infected patients. In contrast, our results demonstrate that SFRP2 promoter exhibits hypermethylation in early HCCs as well as HCV infected non-cancerous (CH and LC) samples. These conflicting observations regarding SFRP2 promoter methylation indicate epigenetic as well as genetic heterogeneity that is characteristic of all cancers in general including HCC .
Previous in vitro
[36, 37] and in vivo studies  as well as our own results point out towards a possible direct role of HCV infection and/or intracellular expression of HCV genes in modulating host cell methylome. Although the exact mechanism underlying this possible effect of HCV infection on host epigenome is still poorly understood, evidence has started to emerge which can provide plausible explanations to this hitherto enigmatic phenomenon. An earlier study reported the promoter hypermethylation mediated down regulation of E-cadherin by HCV core protein in cultured hepatoma cells . Quan et al.  proposed that HCV core causes epigenetic silencing of Wnt inhibitor SFRP1 by modulating the expression and binding of histone deacetylase-1 (HDAC1) and DNA methyltransferase-1 (Dnmt1). In a recent study Okamoto et al.  demonstrated that HBV or HCV infected mice with humanized livers exhibit a time dependent genome wide hypermethylation at various gene promoters. The authors claimed that ROS production by innate immune system components like natural killer T-cells contributes to the development of this aberrant methylation profile. Further studies will be needed to explicitly elucidate the possible methylome modulating properties of various HCV proteins. Moreover, comprehensive genome wide analysis of HCC predisposing conditions (CH & LC) is also highly imperative in order to establish a detailed methylation map of such diseased tissues which could be further exploited as biomarker for cancer risk prediction and identification of novel drug targets for prevention of HCC development.