Identification of an epigenetic signature of early mouse liver regeneration that is disrupted by Zn-HDAC inhibition

Abstract

Liver regeneration has been well studied with hope of discovering strategies to improve liver disease outcomes. Nevertheless, the signals that initiate such regeneration remain incompletely defined, and translation of mechanism-based pro-regenerative interventions into new treatments for hepatic diseases has not yet been achieved. We previously reported the isoform-specific regulation and essential function of zinc-dependent histone deacetylases (Zn-HDACs) during mouse liver regeneration. Those data suggest that epigenetically regulated anti-proliferative genes are deacetylated and transcriptionally suppressed by Zn-HDAC activity or that pro-regenerative factors are acetylated and induced by such activity in response to partial hepatectomy (PH). To investigate these possibilities, we conducted genome-wide interrogation of the liver histone acetylome during early PH-induced liver regeneration in mice using acetyL-histone chromatin immunoprecipitation and next generation DNA sequencing. We also compared the findings of that study to those seen during the impaired regenerative response that occurs with Zn-HDAC inhibition. The results reveal an epigenetic signature of early liver regeneration that includes both hyperacetylation of pro-regenerative factors and deacetylation of anti-proliferative and pro-apoptotic genes. Our data also show that administration of an anti-regenerative regimen of the Zn-HDAC inhibitor suberoylanilide hydroxamic acid (SAHA) not only disrupts gene-specific pro-regenerative changes in liver histone deacetylation but also reverses PH-induced effects on histone hyperacetylation. Taken together, these studies offer new insight into and suggest novel hypotheses about the epigenetic mechanisms that regulate liver regeneration.

Keywords: Ac-H3K9, histone H3 acetylated on lysine 9; CDKI, cyclin dependent kinase inhibitor; ChIP-Seq, chromatin immunoprecipitation-next generation DNA sequencing; GO, gene ontology; PH, partial hepatectomy; SAHA, suberoylanilide hydroxamic acid; TSS, transcription start sites; Zn-HDAC, zinc-dependent histone deacetylase; chromatin immunoprecipitation; histone acetylation; histone deacetylase; partial hepatectomy; qRT-PCR, semi-quantitative real-time reverse-transcription polymerase-chain-reaction; suberoylanilide hydroxamic acid.

Figure 1.

Patterns of Histone H3K9 Acetylation in Early Regenerating Liver. Examples of gene-specific patterns of histone acetylation in regenerating vs. sham-operated liver are shown for several specific genes identified as (A) hyperacetylated (Ccnd1, Myc, Ahr, Cdkn1a) or (B) deacetylated (Cebpa, Foxo3, Nr3c1, Gas1) 12 hours after PH. UCSC gene maps (transcription (*) and translation (**) start sites, exons (E), and introns (I) as designated) are aligned with abundance of immunoprecipitated sequence (sequence reads per bp with scale indicated to the right) integrated from livers of 3 replicates each after PH or sham surgery. The sequence abundance images were generated using the Integrative Genomics Viewer (IGV) genome browser.^40,41 The bars below the sequence abundance data indicate specific sequence(s) identified as differentially acetylated.

Figure 2.

Interactome Plots of Differentially Acetylated Genes in Early Regenerating Liver. Patterns of biological interaction between (A) hyper- or (B) de-acetylated genes in regenerating vs. sham-operated liver are illustrated (with interactions identified using MetaCore^TM from GeneGo and images generated using Cytoscape version 3.1.1 (http://cytoscape.org)). Functional interactions are indicated by lines between designated genes, with increased connectivity represented by increased node size. The most over-connected genes in each set are listed.

Figure 3.

The Influence of SAHA on Histone Acetylation in Early Regenerating Liver. Examples of gene-specific patterns of histone acetylation in regenerating liver from animals treated with SAHA or vehicle control are shown for specific genes identified as (A) deacetylated (Myc, Ahr) or (B) hyperacetylated (Foxo3, Nr3c1, Bik, Bmf) 12 hours after PH. UCSC gene maps (transcription (*) and translation (**) start sites, exons (E), and introns (I) as designated) are aligned with abundance of immunoprecipitated sequence (sequence reads per bp with scale indicated to the right) integrated from livers of 4 replicates each treated with SAHA or vehicle. Sequence abundance images were generated as in Figure 1, and bars below these data indicate specific sequence(s) identified as differentially acetylated.

Figure 4.

Hepatic mRNA Expression of Differentially Acetylated Genes during and Model of Epigenetic Regulation of Liver Regeneration. (A) Relative mRNA expression (± standard error) 12 hours after PH or sham surgery (indexed to sham; *P < 0 .05 for PH vs. sham; n = 6 replicates per experimental group) or 12 hours after PH in vehicle- (Veh) or SAHA-treated mice (indexed to Veh; *P < 0 .05 for SAHA vs. vehicle; n = 6 replicates per experimental group). (B) Proposed model of epigenetically regulated Myc/Foxo3 switch in regenerating liver (known positive (→) and negative (—|) regulation and hypothesized (?) regulation as indicated; see text for additional discussion).