Hepatocyte HIF‐1 and intermittent hypoxia independently increase liver fibrosis in a mouse model of nonalcoholic fatty liver disease

Background Obstructive sleep apnea (OSA) is associated with metabolic dysfunction, including insulin resistance, lipid dysregulation, and hepatic steatosis and fibrosis in nonalcoholic fatty liver disease (NAFLD). Mechanisms underlying the association between OSA and NAFLD are unclear, but we have previously shown that hepatocyte hypoxia inducible factor‐1 (HIF‐1) mediates the development of liver fibrosis in a mouse model of NAFLD. Moreover, chronic intermittent hypoxia (IH) as a model of OSA has been shown to increase hepatic steatosis and fibrosis in rodent models. Here, we aimed to query the interactive effect of IH and hepatocyte HIF‐1. We hypothesized that IH would increase liver steatosis and fibrosis in murine NAFLD, in a HIF‐1‐dependent manner. Methods Mice were generated to have a hepatocyte‐specific deletion of Hif1a ( Hif1a −/− hep ) by a Cre‐recombinase system. Cre‐negative Hif1a flox mice were used as wild‐type control. Eight‐week‐old male mice from each genotype were fed a high trans ‐fat diet to induce characteristics similar to human NAFLD. At week 20, the mice were exposed either to intermittent hypoxia (IH, FiO 2 ranging from 21% to 6%, once per minute, for 12 hours during the light phase) or intermittent air (IA). Intraperitoneal glucose tolerance test (IPGTT) was done at week 25, one week before sacrifice. At sacrifice, liver fibrosis, the primary outcome of interest, was determined by hydroxyproline assay. Mitochondrial function was assessed in fresh liver tissue, and the respiratory control ratio (RCR) was calculated. Malondialdehyde and triglycerides were quantified in liver samples using commercial assays. Aminotransferase levels were quantified using veterinary laboratory services. Data were analyzed by two‐way ANOVA. Results Liver fibrosis was increased in IH versus IA (9% increase in hydroxyproline in IH, p=0.022), and was decreased in Hif1a −/− hep mice relative to Hif1a flox (20% decrease in hydroxyproline in Hif1a −/− hep , p<0.001). There was no significant interaction between genotype and hypoxic exposure. Hif1a −/− hep mice had an improved metabolic profile, with lower fasting glucose and area under the IPGTT curve. However, Hif1a −/− hep mice had increased oxidative stress (malondialdehyde increased 55% in Hif1a −/− hep , p=0.004), and worsened mitochondrial efficiency (32% reduction in RCR in Hif1a −/− hep , p=0.002). Hepatic triglycerides and aminotransferase levels were similar between groups. Conclusions Hepatocyte HIF‐1 appears to be a more important mediator of metabolic outcomes, liver fibrosis, liver oxidative stress, and hepatic mitochondrial function than hypoxia in this model of NAFLD and superimposed IH. The impact of HIF‐1 may be hypoxia‐independent, or the effect of hypoxia may be mediated via non‐HIF‐1 pathways. Support or Funding Information This research was funded by grants from the American Academy of Sleep Medicine Foundation (177‐PA‐17), the American Thoracic Society Foundation (ATS‐2017‐19), the University of California San Diego (RS295R), and the National Institutes of Health (1K08HL143140).