Epigenetic modulation inhibits epithelial-mesenchymal transition-driven fibrogenesis and enhances characteristics of chemically-derived hepatic progenitors

Abstract

Purpose: One of the novel cell sources of cell-based liver regenerative medicine is human chemically-derived hepatic progenitors (hCdHs). We previously established this cell by direct hepatocyte reprogramming with a combination of small molecules (hepatocyte growth factor, A83-01, CHIR99021). However, there have been several issues concerning the cell's stability and maintenance, namely the occurrences of epithelial-mesenchymal transition (EMT) that develop fibrotic phenotypes, resulting in the loss of hepatic progenitor characteristics. These hepatic progenitor attributes are thought to be regulated by SOX9, a transcription factor essential for hepatic progenitor cells and cholangiocytes.

Methods: To suppress the fibrotic phenotype and improve our long-term hCdHs culture technology, we utilized the epigenetic modulating drugs DNA methyltransferase inhibitor (5-azacytidine) and histone deacetylase inhibitor (sodium butyrate) that have been reported to suppress and revert hepatic fibrosis. To confirm the essential role of SOX9 to our cell, we used clustered regularly interspaced short palindromic repeats-interference (CRISPRi) to repress the SOX9 expression.

Results: The treatment of only 5-azacytidine significantly reduces the fibrosis/mesenchymal marker and EMT-related transcription factor expression level in the early passages. Interestingly, this treatment also increased the hepatic progenitor markers expression, even during the reprogramming phase. Then, we confirmed the essential role of SOX9 by repressing the SOX9 expression with CRISPRi which resulted in the downregulation of several essential hepatic progenitor cell markers.

Conclusion: These results highlight the capacity of 5-azacytidine to inhibit EMT-driven hepatic fibrosis and the significance of SOX9 on hepatic progenitor cell stemness properties.

Keywords: CRISPR; Epigenomics; Fibrosis; Hepatocytes; Human chemically-derived hepatic progenitors.

Fig. 1. Effect of epigenetic modulating drugs (epidrugs) on the cellular reprogramming of human primary hepatocytes (hPHs) generation into human chemically-derived hepatic progenitors (hCdHs). (A) Scheme of the generation of hCdHs perfused hPHs with the treatment of epidrugs 5-azacytidine (5-Aza) and sodium butyrate (NaB). (B) Brightfield imaging of the reprogramming process of hCdHs with epidrugs treatment with different drug concentrations (5-Aza, 0.5 µM; NaB, 0–100 µM). Scale bars, 100 µm. (C–E) Relative messenger RNA (mRNA) expression level of hepatic progenitor markers (EpCAM, SOX9) (C), epithelial-mesenchymal transition-related transcription factor (TWIST1, SNAI1) (D), and mesenchymal marker (COL1A1) (E). Values are presented as mean ± standard error of mean in triplicates (n = 3). HAC, HGF, A83-01, CHIR-99021; HGF, hepatocyte growth factor; EpCAM, epithelial cell adhesion molecule; SOX9, SRY-box transcription factor 9; SNAI1, Snail family transcriptional repressor 1; TWIST1, twist family bHLH transcription factor 1; COL1A1, collagen type I alpha 1 chain; NS, not significant. ^*P < 0.05 and ^****P < 0.001, by 2-tailed t-test.

Fig. 2. Epigenetic modulating drugs (epidrugs) inhibit epithelial-mesenchymal transition (EMT)-driven fibrogenesis by regulating the EMT transcription factors SNAI1 and TWIST1. (A) Brightfield imaging of human chemically-derived hepatic progenitors (hCdHs) and fibrotic hCdHs. (B) Schematic image of the treatment of 5-azacytidine (5-Aza) and sodium butyrate (NaB) on the long-term maintenance and culture of hCdHs. (C) Brightfield imaging of the long-term culture and expansion of hCdHs treated with epidrugs with different drug concentrations on day 2 and day 5 after passaging. Scale bars, 100 µm. (D–F) Relative messenger RNA (mRNA) expression level of hepatic progenitor markers (EpCAM, SOX9) (D), EMT-related transcription factor (TWIST1, SNAI1) (E), and mesenchymal marker (COL1A1) (F). Values are presented as mean ± standard error of mean in triplicates (n = 3). HAC, HGF, A83-01, CHIR-99021; HGF, hepatocyte growth factor; EpCAM, epithelial cell adhesion molecule; SOX9, SRY-box transcription factor 9; SNAI1, Snail family transcriptional repressor 1; TWIST1, twist family bHLH transcription factor 1; COL1A1, collagen type I alpha 1 chain; NS, not significant. ^*P < 0.05 and ^****P < 0.001, by 2-tailed t-test.

Fig. 3. Clustered regularly interspaced short palindromic repeats-interference (CRISPRi)-mediated SOX9 knockdown demonstrates SOX9 essential role as a transcription factor of mouse chemically-derived hepatic progenitors (mCdHs). (A) Illustration of the SOX9 repression by CRISPRi electroporation. (B) SOX9 genome map shows the single guide RNA (sgRNA) target site (sgRNA#1, #4, and #5) relative to its transcription start site. (C) Brightfield imaging of the impact of SOX9 repression by CRISPRi with and without sgRNA transfection on mCdHs. Scale bars, 100 µm (a) and 500 µm (b). Relative messenger RNA (mRNA) expression level of hepatic progenitor markers (SOX9, AFP, KRT7, EpCAM, CD44) (D) and hepatic marker (ALB, HNF4A) (E). Values are presented as means ± standard error of mean in triplicates (n = 3). AFP, alpha-fetoprotein; KRT7, keratin 7; EpCAM, epithelial cell adhesion molecule; CD44, cluster of differentiation 44; ALB, albumin; HNF4A, hepatocyte nuclear factor 4 alpha. ^*P < 0.05 and ^****P < 0.001, by 2-tailed t-test.