Acute-on-chronic liver failure is mainly due to host immunity self-destruction. The histone H3 lysine 27 (H3K27) trimethylating enzyme, enhancer of zeste homolog 2 (EZH2) mediates epigenetic silencing of gene expression and regulates immunity, also involves pathogenesis of several liver diseases. The current study was to determine the role of methyltransferase EZH2 and its catalysed H3K27 trimethylation (H3K27me3) in liver failure, and to further investigate the potential target for liver failure treatment. EZH2 and its catalysed H3K27me3 were determined in peripheral blood mononuclear cells (PBMC) from liver failure patients and Kupffer cells from experimental mice. Furthermore, GSK126 (an inhibitor for EZH2 trimethylation function) was applied in liver failure mice in vivo, and lipopolysaccharide-stimulated mononuclear cells in vitro. EZH2 and H3K27me3 were significantly upregulated in human PBMC from liver failure patients or murine Kupffer cells from the liver failure animals, respectively. GSK126 ameliorated disease severity in liver failure mice, which maybe attribute to down-regulate circulating and hepatic proinflammatory cytokines, especially TNF via reducing H3K27me3. In-depth chromatin immunoprecipitation analysis unravelled that decreased enrichment of H3K27me3 on Tnf promotor, resulting in TNF elevation in Kupffer cells from liver failure mice. Nuclear factor kappa B (NF-κB) and protein kinase B (Akt) signalling pathways were activated upon lipopolysaccharide stimulation, but attenuated by using GSK126, accompanied with decreased TNF in vitro. In conclusion, EZH2 and H3K27me3 contributed to the pathogenesis of liver failure via triggering TNF and other indispensable proinflammatory cytokines. EZH2 was to modify H3K27me3 enrichment, as well as, activation of the downstream NF-κB and Akt signalling pathways.

ObjectivesT Follicular helper (Tfh) cells, recognized as a distinct CD4+ T cell subset, mediate the development of long-lived humoral immunity via B cell activation/differentiation. Tfh cells play an important role during hepatic viral infection, but its role in hepatitis B virus-related acute on chronic liver failure (HBV-ACLF) remains to be explored.Materials and MethodsThe frequency of Tfh cells, serum pro-inflammatory cytokine (IL-12, IL-21, IL-17 and TNF) levels and IgG/M levels were investigated in HBV-ACLF (n = 36), serious chronic hepatitis B (n = 21), moderate chronic hepatitis B patients (n = 32) and healthy control (HC) subjects (n = 10).ResultsCirculating Tfh cells were significantly increased in HBV-ACLF patients compared to other groups, correlating well with MELD score. However, the frequency of Tfh cells decreased in ameliorated HBV-ACLF patients. Furthermore, serum IL-12 and IL-21 levels were higher in HBV-ACLF patients, compared to other groups. Naïve CD4+ T cells from HC subjects differentiate into Tfh cells following treatment with HBV-ACLF patients’ serum, a process that can be blocked by IL-12/21 neutralizing antibodies. Tfh cells induced by HBV-ACLF patient’s serum promoted the proliferation and IgG production of B cells in vitro. Moreover, circulating CD19+ B cells, serum and liver IgG/M levels were significantly higher in HBV-ACLF patients, compared to other groups.ConclusionsOur data demonstrated that there was a high frequency of Tfh cells and high levels of serum IL-12/21 in HBV-ACLF patients. Naïve CD4+ T cells differentiate into Tfh cells in the presence of HBV-ACLF patients’ serum rich in IL-12/21, which can be blocked by neutralizing IL-12/21 antibodies. These data may provide useful insights for both clinical and basic research in the treatment of HBV-ACLF.

In response to comments on PubPeer, the author uploaded the raw data as Supplementary Material 2 to support Figures 4A-C in the original file. The authors apologize to the readership for any inconvenience caused.These experiments were completed from July to September 2012. The workspace can't be used due to the upgrade of the version of FlowJo. The data were reanalyzed using the same gating and labelling in FlowJo version 10.6. The results are presented in the format of Pseudocolour. The trend of the data has not changed, which does not affect the original results.The authors uploaded the FCS format raw data and Pseudocolour format graphical of all 6 patients, which mentioned in the original article (n=6, and original figure 4 shows the flow analysis data of patient #1).The documents are as follows: 1) Data 1: FACS raw files of patient (n=6. A patient's data is summarized in a folder. Each folder contains 23 FCS files, which are marked according to the processing.