Forebrain microglia involvement in non‐alcoholic fatty liver disease during obesity

Abstract

Non-alcoholic fatty liver disease (NAFLD), characterized by increased hepatic triglyceride (TG) content (i.e. steatosis), leads to an increased risk for type II diabetes and obesity-related mortality. Emerging evidence points to a role for microglia, resident brain immune cells, in obesity-related diseases, although the contribution of microglia to NAFLD remains unknown. Thus, we hypothesized that microglial activation contributes to hepatic steatosis during obesity. Six week old male C57Bl/6J mice were fed a high fat diet (HFD) or normal chow for 11 wks and were then instrumented with an intracerebroventricular (ICV) cannula. Following surgical recovery, the microglia inhibitor minocycline or vehicle control was administered ICV daily over 3 days (n=7-10/group). Microglia inhibition did not alter body weight (47.2±1.3 vs 49.0±0.7, HFD vehicle vs minocycline, p>0.05), food intake or locomotor activity in either normal chow or HFD animals. However, histological examination (Oil Red O staining) of the liver revealed widespread lipid accumulation in HFD fed mice, which was reduced by ~48% following short-term microglia inhibition (23.6±5.2 vs 12.2±3.5 a.u. fold normal chow vehicle, HFD vehicle vs minocycline, p=0.07). In parallel with this, 3 day ICV minocycline treatment in obese mice was associated with downregulation of a number of hepatic mRNA markers including genes involved in de novo lipogenesis (e.g. ChREBP: 6.5±1.2 vs 1.9±1.2; Fasn: 24.1±3.8 vs 2.9±2.5 fold normal chow vehicle, HFD vehicle vs minocycline, both p<0.05) and free fatty acid uptake (e.g. FATP5: 6.0±1.0 vs 1.4±0.9 fold normal chow vehicle, HFD vehicle vs minocycline, p<0.05). To gain insight into brain regions that may be involved in microglia-mediated NAFLD, we turned our attention to the forebrain subfornical organ (SFO), a circumventricular nucleus located outside of the blood-brain-barrier that senses circulating factors. We have recently shown a critical role for the SFO in NAFLD during obesity, but the mechanism(s) involved remain unclear. Immunohistochemistry using Iba1 as a marker revealed an increase in SFO microglia following HFD feeding (250±26 vs 342±13 microglia #, normal chow vs HFD, p<0.05, n=5). Comprehensive 3-dimensional anatomical analysis further indicated that HFD feeding was associated with a shift toward an activated SFO microglia morphology including a loss of dendritic processes and reduced branching complexity (e.g. dendrite length/microglia: 125±7 vs 90±3 µm; Sholl intersections/microglia: 97±9 vs 68±3, normal chow vs HFD, both p<0.05). Finally, mRNA expression of microglial activation genes confirmed HFD-induced SFO microglia activation, which interestingly was reduced with ICV minocycline administration (e.g. Iba1: 2.1±0.1 vs 1.1±0.2; CD68: 1.5±0.03 vs. 1.0±0.1 fold normal chow vehicle, HFD vehicle vs minocycline, both p<0.05). Collectively, these findings indicate that microglial activation contributes to hepatic steatosis during obesity. Furthermore, marked microglial activation in the SFO points to SFO microglia as a potential culprit in the pathogenesis of NAFLD.