Deregulation of histone methyltransferases SETDB1 and G9a and their functional roles in liver cancer

Abstract

Liver cancer, primarily hepatocellular carcinoma (HCC), is the fifth most common cancers and the second leading cause of cancer mortality worldwide. Recently, increasing evidences suggested that epigenetic alternations play an important role in liver carcinogenesis. Using transcriptome sequencing, we examined the expression of 591 epigenetic regulators in hepatitis B-associated human HCC and found that deregulation of epigenetic regulators was a common event in HCC. We further identified SETDB1 (SET domain, bifurcated 1) and G9a (Euchromatic histone-lysine N-methyltransferase 2, EHMT2), two histone H3 lysine 9 (H3K9) specific histone methyltransferases, as two of the most significantly up-regulated epigenetic regulators in human HCCs. SETDB1and G9a arespecific for H3K9 tri-and di-methylation, respectively. Previous studies about epigenetic alternations have primarily focused on promoter DNA hypermethylation. However, the pathological implications of histone modifications, especially the roles of histone methyltransferases such as SETDB1 and G9a in human HCC remain elusive. This study aim to characterize the functional roles and underlying mechanisms of deregulated SETDB1 and G9a in human HCC.

The frequent up-regulation of SETDB1 was validated in various HCC sample cohorts and was significantly associated with HCC progression, cancer aggressiveness and poor survival rate of HCC patients. Functionally, knockdown of SETDB1 suppressed HCC cell proliferation and migration in vitro and inhibited orthotopic tumorigenicity and lung metastasis in vivo. Mechanistically, the frequent-upregulation of SETDB1 inhuman HCC was attributed to multiple molecular mechanisms indifferent levels including recurrent SETDB1 gene copy number gain at chromosome1q21; hyperactivation of SP1 transcription factor which enhanced STEDB1 promoter activity transcriptionally; loss ofmiR-29 which facilitated SETDB1 up-regulation by relieving its post-transcriptional repression; and loss of PTEN which increased SETDB1 protein level post-translationally by activating PI3K/Akt signaling.

Likewise, the frequent up-regulation of G9a was also validated in different HCC sample cohorts. Up-regulation of G9a was significantly associated with HCC disease progression, cancer aggressiveness, and more malignant tumor phenotypes. Functionally, we demonstrated that shRNA knockdown and CRISPR/Cas9 knockout of G9a suppressed HCC cell proliferation in vitro and inhibited subcutaneously xenograft HCC tumorigenicity in vivo. Depletion of G9a significantly reduced HCC cell migration ability and induced cell senescence.

Pharmacological inhibition of G9a by small molecule inhibitors, UNC0638 and BIX01294, also suppressed HCC cell growth and altered cell morphology. Mechanistically, we showed that the frequent up-regulation of G9a in human HCC was attributed to gene copy number gain at chromosome 6p21 and loss of miR-1. Furthermore, up-regulation of G9 also epigenetically repressed miR-1 expression and thus formed a feed forward regulation loop between them. By utilizing RNA-Seq and GSEA analysis, we identified a potential tumor suppressor RARRES that was epigenetically silenced by G9a and promoted tumor cells proliferation in human HCC.

Taken together, we showed that SETDB1 and G9a are novel oncogenes that are frequently up-regulated in human HCCs. Multiple mechanisms at chromosomal, transcriptional, post-transcriptional, and post-translational levels contributed to their up-regulations in human HCCs. Furthermore, their oncogenic functions may attributed to epigenetic silence of downstream tumor suppressors. Our findings suggested that SETDB1 and G9acould be novel therapeutic targets for HCC treatment.