Non-alcoholic fatty liver disease (NAFLD) leads to hepatocellular carcinoma (HCC). However, the underlying mechanism remains largely unclear. Here, we investigated the role of the tumor suppressor Zinc fingers and homeoboxes 2 (ZHX2) in the progression of NAFLD to HCC. ZHX2 expression was significantly decreased in fatty liver tissues, especially in the liver with NAFLD-HCC. ZHX2 overexpression disturbed lipid homeostasis of cultured HCC cells, and inhibited lipid deposition in hepatocytes both in vitro and in vivo. Moreover, ZHX2 inhibited uptake of exogenous lipids through transcriptional suppression of lipid lipase (LPL), leading to retarded proliferation of HCC cells. Importantly, LPL overexpression significantly reversed ZHX2-mediated inhibition of HCC cell proliferation, xenograft tumor growth, lipid deposition, and spontaneous liver tumor formation. Consistently, IHC staining demonstrated a negative correlation of ZHX2 with LPL in an HCC cohort. Collectively, ZHX2 protects hepatocytes from abnormal lipid deposition in NAFLD through transcriptional repression of LPL, which subsequently retards cell growth and NAFLD-HCC progression. These findings illustrate a novel mechanism of NAFLD progression into HCC.

Background: Liver cancer stem cells (CSCs) are critical determinants of HCC relapse and therapeutic resistance, but the mechanisms underlying the maintenance of CSCs are poorly understood. We aimed to explore the role of tumor repressor Zinc-fingers and homeoboxes 2 (ZHX2) in liver CSCs. Methods: CD133 + or EPCAM + stem-like liver cancer cells were sorted from tumor tissues of HCC patients and HCC cell lines by flow cytometry. In addition, sorafenib-resistant cells, tumor-sphere forming cells and side population (SP) cells were respectively cultured and isolated as hepatic CSCs. The tumor-initiating and chemoresistance properties of ZHX2-overexpressing and ZHX2-knockdown cells were analyzed in vivo and in vitro. Microarray, luciferase reporter assay, chromatin immunoprecipitation (ChIP) and ChIP-on-chip analyses were performed to explore ZHX2 target genes. The expression of ZHX2 and its target gene were determined by quantitative RT-PCR, western blot, immunofluorescence and immunohistochemical staining in hepatoma cells and tumor and adjacent tissues from HCC patients. Results: ZHX2 expression was significantly reduced in liver CSCs from different origins. ZHX2 deficiency led to enhanced liver tumor progression and expansion of CSC populations in vitro and in vivo. Re-expression of ZHX2 restricted capabilities of hepatic CSCs in supporting tumor initiation, self-renewal and sorafenib-resistance. Mechanically, ZHX2 suppressed liver CSCs via inhibiting KDM2A-mediated demethylation of histone H3 lysine 36 (H3K36) at the promoter regions of stemness-associated transcription factors, such as NANOG, SOX4 and OCT4. Moreover, patients with lower expression of ZHX2 and higher expression of KDM2A in tumor tissues showed significantly poorer survival. Conclusion: ZHX2 counteracts stem cell traits through transcriptionally repressing KDM2A in HCC. Our data will aid in a better understanding of molecular mechanisms underlying HCC relapse and drug resistance.

Cancer vaccines can amplify existing antitumor responses or prime naïve T cells to elicit effector T-cell functions in patients through immunization. Antigen-specific CD8+ T cells are crucial for the rejection of established tumors. We constructed XCL1-GPC3 fusion molecules as a liver cancer vaccine by linking the XCL1 chemokine to glypican-3 (GPC3), which is overexpressed in hepatocellular carcinoma (HCC). Cells expressing XCL1-GPC3 chemoattracted murine XCR1+CD8α+ dendritic cells (DC) and human XCR1+CD141+ DCs in vitro and promoted their IL12 production. After subcutaneous mXcl1-GPC3 plasmid injection, mXCL1-GPC3 was mainly detected in CD8α+ DCs of mouse draining lymph nodes. XCL1-GPC3–targeted DCs enhanced antigen-specific CD8+ T-cell proliferation and induced the de novo generation of GPC3-specific CD8+ T cells, which abolished GPC3-expressing tumor cells in mouse and human systems. We immunized a murine autochthonous liver cancer model, with a hepatitis B background, with the mXcl1-GPC3 plasmid starting at 6 weeks, when malignant hepatocyte clusters formed, or at 14 weeks, when liver tumor nodules developed, after diethylnitrosamine administration. mXcl1-GPC3–immunized mice displayed significantly inhibited tumor formation and growth compared with GPC3-immunized mice. After mXcl1-GPC3 immunization, mouse livers showed elevated production of IFNγ, granzyme B, IL18, CCL5, CXCL19, and Xcl1 and increased infiltration of GPC3-specific CD8+ T cells, activated natural killer (NK) cells, and NKT cells. The antitumor effects of these immune cells were further enhanced by the administration of anti–PD-1. Anti-HCC effects induced by hXCL1-GPC3 were confirmed in an HCC-PDX model from 3 patients. Thus, XCL1-GPC3 might be a promising cancer vaccine to compensate for the deficiency of the checkpoint blockades in HCC immunotherapy.

No abstract

BackgroundIt is estimated that 35% of gastric cancer patients appear with synchronous distant metastases—the vast majority of patients presenting with metastatic hepatic disease. How to choose the most appropriate drugs or regimens is crucial to improve the prognosis of patients. We conducted this retrospective cohort analysis to evaluate the efficacy of OncoVee™-MiniPDX-guided treatment for these patients.MethodsGastric cancer patients with liver metastases (GCLM) were enrolled. Patients were divided into MiniPDX and control group according to their wishes. In the observation group, the OncoVee™-MiniPDX model was conducted to screen the most sensitive drug or regimens to determine the clinical administration. Meanwhile, patients were treated with regular medications in the control group according to the guidelines without the MiniPDX model. The primary endpoint was overall survival (OS), and the secondary outcomes included objective response rate (ORR), disease control rate (DCR), and progression-free survival (PFS).ResultsA total of 68 patients with GCLM were included, with the observation and control groups of 21 and 47 patients, respectively. The baseline characteristics of patients were balanced between these two groups. MiniPDX drug sensitivity tests were associated with the increased use of targeted drugs when compared with the control group (33.3 vs. 0%, p=0.032). Median OS was estimated to be 9.4 (95% CI, 7.9–11.2) months and 7.9 (95% CI, 7.2–8.7) months in the observation and control group, respectively. Both univariate (control group vs. MiniPDX group: HR=2.586, 95% CI= 1.362–4.908, p=0.004) and multivariate regression analyses (Control group vs. MiniPDX group: adjusted HR (aHR)=4.288, 95% CI= 1.452–12.671, p=0.008) showed the superiority of the observation group on OS. Similarly, MiniPDX-based regiments significantly improve the PFS of these cases (median PFS 6.7 months vs. 4.2 months, aHR=2.773, 95% CI=1.532–3.983, p=0.029). ORR and DCR were also improved in MiniPDX group comparing with control group (ORR, 57.14 vs. 25.53%, p=0.029; DCR: 85.71 vs. 68.08%, p=0.035).ConclusionOncoVee™-MiniPDX model, which was used to select drugs to guide antitumor treatment, was promising to prolong survival and improve the response rate of patients with GCLM. Further well-designed studies are needed to confirm the clinical benefits of MiniPDX.