Non-obese non-alcoholic fatty liver disease (NAFLD) in Asia: an international registry study.

BACKGROUND: A significant proportion of the non-alcoholic fatty liver disease (NAFLD) population is non-obese. Prior studies reporting the severity of NAFLD amongst non-obese patients were heterogenous. Our study, using data from the largest biopsy-proven NAFLD international registry within Asia, aims to characterize the demographic, metabolic and histological differences between non-obese and obese NAFLD patients. METHODS: 1812 biopsy-proven NAFLD patients across nine countries in Asia assessed between 2006 and 2019 were pooled into a curated clinical registry. Demographic, metabolic and histological differences between non-obese and obese NAFLD patients were evaluated. The performance of Fibrosis-4 index for liver fibrosis (FIB-4) and NAFLD fibrosis score (NFS) to identify advanced liver disease across the varying obesity subgroups was compared. A random forest analysis was performed to identify novel predictors of fibrosis and steatohepatitis in non-obese patients. FINDINGS: One-fifth (21.6%) of NAFLD patients were non-obese. Non-obese NAFLD patients had lower proportions of NASH (50.5% vs 56.5%, p = 0.033) and advanced fibrosis (14.0% vs 18.7%, p = 0.033). Metabolic syndrome in non-obese individuals was associated with NASH (OR 1.59, 95% CI 1.01-2.54, p = 0.047) and advanced fibrosis (OR 1.88, 95% CI 0.99-3.54, p = 0.051). FIB-4 performed better than the NFS score (AUROC 81.5% vs 73.7%, p < 0.001) when classifying patients with F2-4 fibrosis amongst non-obese NAFLD patients. Haemoglobin, GGT, waist circumference and cholesterol are additional variables found on random forest analysis useful for identifying non-obese NAFLD patients with advanced liver disease. CONCLUSION: A substantial proportion of non-obese NAFLD patients has NASH or advanced fibrosis. FIB-4, compared to NFS better identifies non-obese NAFLD patients with advanced liver disease. Serum GGT, cholesterol, haemoglobin and waist circumference, which are neither components of NFS nor FIB-4, are important biomarkers for advanced liver disease in non-obese patients.

Human Pluripotent Stem Cell-Derived Organoids as Models of Liver Disease.

BACKGROUND & AIMS: There are few in vitro models for studying the 3-dimensional interactions among different liver cell types during organogenesis or disease development. We aimed to generate hepatic organoids that comprise different parenchymal liver cell types and have structural features of the liver, using human pluripotent stem cells. METHODS: We cultured H1 human embryonic stem cells (WA-01, passage 27-40) and induced pluripotent stem cells (GM23338) with a series of chemically defined and serum-free media to induce formation of posterior foregut cells, which were differentiated in 3 dimensions into hepatic endoderm spheroids and stepwise into hepatoblast spheroids. Hepatoblast spheroids were reseeded in a high-throughput format and induced to form hepatic organoids; development of functional bile canaliculi was imaged live. Levels of albumin and apolipoprotein B were measured in cell culture supernatants using an enzyme-linked immunosorbent assay. Levels of gamma glutamyl transferase and alkaline phosphatase were measured in cholangiocytes. Organoids were incubated with troglitazone for varying periods and bile transport and accumulation were visualized by live-imaging microscopy. Organoids were incubated with oleic and palmitic acid, and formation of lipid droplets was visualized by staining. We compared gene expression profiles of organoids incubated with free fatty acids or without. We also compared gene expression profiles between liver tissue samples from patients with nonalcoholic steatohepatitis (NASH) versus without. We quantified hepatocyte and cholangiocyte populations in organoids using immunostaining and flow cytometry; cholangiocyte proliferation of cholangiocytes was measured. We compared the bile canaliculi network in the organoids incubated with versus without free fatty acids by live imaging. RESULTS: Cells in organoids differentiated into hepatocytes and cholangiocytes, based on the expression of albumin and cytokeratin 7. Hepatocytes were functional, based on secretion of albumin and apolipoprotein B and cytochrome P450 activity; cholangiocytes were functional, based on gamma glutamyl transferase and alkaline phosphatase activity and proliferative responses to secretin. The organoids organized a functional bile canaliculi system, which was disrupted by cholestasis-inducing drugs such as troglitazone. Organoids incubated with free fatty acids had gene expression signatures similar to those of liver tissues from patients with NASH. Incubation of organoids with free fatty acid-enriched media resulted in structural changes associated with nonalcoholic fatty liver disease, such as decay of bile canaliculi network and ductular reactions. CONCLUSIONS: We developed a hepatic organoid platform with human cells that can be used to model complex liver diseases, including NASH.

A murine model demonstrating reversal of structural and functional correlates of cirrhosis with progenitor cell transplantation.

Development of cell transplantation for treating liver cirrhosis hinges critically on the availability of animal models for studying human stem cell transplantation. We report an immune-permissive murine model of liver cirrhosis with full clinical correlates of decompensated liver disease, and allows testing efficacy of stem cell transplantation. Liver cirrhosis was induced in Nod-scid gamma(NSG) mice with oral thioacetamide(TA) and compared to controls over 12 months. 4 month TA treated cirrhotic mice were then transplanted intrasplenically with 2million human fetal liver progenitor cells(HFH) and compared with cirrhotic controls 2 months after transplantation. NSG-TA mice developed shrunken and nodular livers with histological evidence of fibrosis as compared to controls. This was associated with evidence of worsening decompensated liver disease, with jaundice, hypoalbuminemia, coagulopathy, and encephalopathy in NSG-TA mice. Transplantation of HFH resulted in improvement in both fibrosis and markers of decompensated liver disease. We have demonstrated that NSG-TA mice can recapitulate the full clinical picture of structural and functional cirrhosis, both of which can be improved by transplantation of human fetal liver cells. This model serves as a valuable tool for validation of in vivo liver stem cell transplantation and opens up opportunities for studying the mechanism how stem cells reverse fibrosis.

Long-Term Fate of Human Fetal Liver Progenitor Cells Transplanted in Injured Mouse Livers.

Liver progenitor cells have the potential to repair and regenerate a diseased liver. The success of any translational efforts, however, hinges on thorough understanding of the fate of these cells after transplant, especially in terms of long-term safety and efficacy. Here, we report transplantation of a liver progenitor population isolated from human fetal livers into immune-permissive mice with follow-up up to 36 weeks after transplant. We found that human progenitor cells engraft and differentiate into functional human hepatocytes in the mouse, producing albumin, alpha-1-antitrypsin, and glycogen. They create tight junctions with mouse hepatocytes, with no evidence of cell fusion. Interestingly, they also differentiate into functional endothelial cell and bile duct cells. Transplantation of progenitor cells abrogated carbon tetrachloride-induced fibrosis in recipient mice, with downregulation of procollagen and anti-smooth muscle actin. Paradoxically, the degree of engraftment of human hepatocytes correlated negatively with the anti-fibrotic effect. Progenitor cell expansion was most prominent in cirrhotic animals, and correlated with transcript levels of pro-fibrotic genes. Animals that had resolution of fibrosis had quiescent native progenitor cells in their livers. No evidence of neoplasia was observed, even up to 9 months after transplantation. Human fetal liver progenitor cells successfully attenuate liver fibrosis in mice. They are activated in the setting of liver injury, but become quiescent when injury resolves, mimicking the behavior of de novo progenitor cells. Our data suggest that liver progenitor cells transplanted into injured livers maintain a functional role in the repair and regeneration of the liver. Stem Cells 2018;36:103-113.

Hepatic differentiation of human amniotic epithelial cells and in vivo therapeutic effect on animal model of cirrhosis.

BACKGROUND AND AIM: Human amniotic epithelial cells (hAECs) have been touted as an ideal stem cell candidate, being ethically neutral, immunologically naïve, plentiful in origin, and retaining plasticity in its fetal stage. We hypothesized that by applying natural physiological signals of the developing liver, hAECs can be coaxed into becoming functional immunopermissive hepatocyte-like cells. These cells would have tremendous potential for allogenic cellular transplantation in the treatment of chronic liver insufficiency. METHODS: hAECs were obtained from term placentas and subjected to hepatic trans-differentiation. Hepatic differentiated cells were analyzed with immunophenotyping, electron microscopy, reverse transcription-polymerase chain reaction as well as characterized for hepatic metabolic function. In vivo efficacy was tested using intrasplenic transplantation into non-obese diabetic (NOD) Scid Gamma mice with thioacetamide-induced chronic liver failure and analyzed for engraftment and improvement in liver indices. RESULTS: With hepatic differentiation, hAECs assumed a hepatocytic polygonal morphology with upregulation of transcription factors responsible for liver specification. These hepatic differentiated-hAECs (HD-AECs) demonstrated bile canaliculi formation, secreted albumin, eliminated indo-cyanine green, uptook low-density lipoprotein, and inducible CYP3A4 and CYP2C9 enzymatic activities. Transplantation of HD-AECs and de novo hAECs in mice model of cirrhosis showed successful in vivo engraftment and differentiation into functional hepatocytes positive for human-specific albumin. HD-AEC cells that had undergone hepatic differentiation showed the greatest improvement in albumin function while preserving human leukocyte antigen-G expression postdifferentiation. CONCLUSION: hAECs were able to differentiate into functional hepatocyte-like cells both in vivo and in vitro. They showed therapeutic efficacy after transplantation in mice model of cirrhosis, offering an exciting source of cells for generation of functionally useful hepatocytes.