Search

Background and Aims: Cholangiocarcinoma (CCA) is a devastating disease often detected at advanced stages when surgery cannot be performed. Conventional and targeted systemic therapies perform poorly, and therefore effective drugs are urgently needed. Different epigenetic modifications occur in CCA and contribute to malignancy. Targeting epigenetic mechanisms may thus open therapeutic opportunities. However, modifications such as DNA and histone methylation often coexist and cooperate in carcinogenesis. We tested the therapeutic efficacy and mechanism of action of a class of dual G9a histone-methyltransferase and DNA-methyltransferase 1 (DNMT1) inhibitors., Approach and Results: Expression of G9a, DNMT1, and their molecular adaptor, ubiquitin-like with PHD and RING finger domains-1 (UHRF1), was determined in human CCA. We evaluated the effect of individual and combined pharmacological inhibition of G9a and DNMT1 on CCA cell growth. Our lead G9a/DNMT1 inhibitor, CM272, was tested in human CCA cells, patient-derived tumoroids and xenograft, and a mouse model of cholangiocarcinogenesis with hepatocellular deletion of c-Jun-N-terminal-kinase (Jnk)-1/2 and diethyl-nitrosamine (DEN) plus CCl 4 treatment (Jnk Δhepa + DEN + CCl 4 mice). We found an increased and correlative expression of G9a, DNMT1, and UHRF1 in CCAs. Cotreatment with independent pharmacological inhibitors G9a and DNMT1 synergistically inhibited CCA cell growth. CM272 markedly reduced CCA cell proliferation and synergized with Cisplatin and the ERBB-targeted inhibitor, Lapatinib. CM272 inhibited CCA tumoroids and xenograft growth and significantly antagonized CCA progression in Jnk Δhepa + DEN + CCl 4 mice without apparent toxicity. Mechanistically, CM272 reprogrammed the tumoral metabolic transcriptome and phenotype toward a differentiated and quiescent status., Conclusions: Dual targeting of G9a and DNMT1 with epigenetic small molecule inhibitors such as CM272 is a potential strategy to treat CCA and/or enhance the efficacy of other systemic therapies., (© 2020 by the American Association for the Study of Liver Diseases.)

Tumor necrosis factor alpha (TNF-alpha) is one of the key cytokines of the acute phase response and of many inflammatory processes. This cytokine has several antifibrogenic actions and down-regulates the expression of the type I collagen genes and induces the expression of metalloproteinases. Because TNF-alpha directly antagonizes some fibrogenic actions of transforming growth factor beta(1) (TGF-beta(1)), we considered it important to map the cis-acting regulatory element of the alpha1(I) collagen (col1a1) promoter involved in TNF-alpha responsiveness in hepatic stellate cells (HSC), to investigate the transcription factors that bind to it, and to establish possible mechanisms by which TNF-alpha down-regulates its expression. In this article, we show the presence of a functional TNF-alpha-responsive element (TaRE) in the -378 to -345 region of the col1a1 promoter. This element colocalizes with a previously reported TGF-beta(1)-responsive element. We further demonstrate that TNF-alpha induces nuclear translocation and binding of transcriptional complexes containing p20C/EBPbeta, p35C/EBPbeta, and C/EBPdelta to this sequence of the promoter. Transient overexpression of C/EBPdelta or p20C/EBPbeta, the natural dominant negative form of C/EBPbeta in HSC, down-regulated activity of a CAT reporter vector driven by -412 to +110 of the col1a1 promoter. Taken together, these data suggest that the -378 to -340 region of the col1a1 promoter is the site of convergence of different stimuli that ultimately modulate col1a1 gene transcription.

Oxidative stress plays a key role in liver fibrosis. Both inflammatory cells and activated Kupffer cells produce H2O2, an oxidant involved in the activation of hepatic stellate cells (HSC). Increased production of reactive oxygen intermediates (ROIs) in fibrotic livers is associated in part with the up-regulation of transforming growth factor beta (TGF-beta), and this cytokine enhances collagen production by cultured HSC. However, the possible link between oxidative stress and the molecular mechanisms by which TGF-beta induces collagen gene expression in HSC remains to be elucidated. To address this question, we investigated whether H2O2 is a mediator of TGF-beta-elicited alpha1(I) collagen gene (col1a1) up-regulation. We demonstrated that TGF-beta induces the accumulation of H2O2, and that this oxidant is, in turn, directly involved in up-regulating the expression of the col1a1 gene. While the addition of H2O2 to HSC induced the expression of alpha1(I) procollagen mRNA, catalase, an H2O2 enzyme scavenger, abrogated TGF-beta-mediated col1a1 gene up-regulation. We transfected HSC with chimeric plasmids driven by different segments of the mouse col1a1 promoter and mapped a cis-acting element (-370 to -344) essential for TGF-beta responsiveness. We further showed that TGF-beta induced the activation and binding of a C/EBPbeta-containing transcriptional complex to this sequence, an effect that was also mimicked by the addition of H2O2. Taken together, these data demonstrate a direct connection between TGF-beta-mediated accumulation of H2O2 and the up-regulation of col1a1 gene in HSC.