Cellular mechanism of insulin resistance in nonalcoholic fatty liver disease

Naoki Kumashiro, Derek M Erion, Dongyan Zhang, Mario Kahn, Sara A Beddow, Xin Chu, Christopher D Still, Glenn S Gerhard, Xianlin Han, James Dziura, Kitt Falk Petersen, Varman T Samuel, Gerald I Shulman, Naoki Kumashiro, Derek M Erion, Dongyan Zhang, Mario Kahn, Sara A Beddow, Xin Chu, Christopher D Still, Glenn S Gerhard, Xianlin Han, James Dziura, Kitt Falk Petersen, Varman T Samuel, Gerald I Shulman

Abstract

Insulin resistance is associated with nonalcoholic fatty liver disease (NAFLD) and is a major factor in the pathogenesis of type 2 diabetes. The development of hepatic insulin resistance has been ascribed to multiple causes, including inflammation, endoplasmic reticulum (ER) stress, and accumulation of hepatocellular lipids in animal models of NAFLD. However, it is unknown whether these same cellular mechanisms link insulin resistance to hepatic steatosis in humans. To examine the cellular mechanisms that link hepatic steatosis to insulin resistance, we comprehensively assessed each of these pathways by using flash-frozen liver biopsies obtained from 37 obese, nondiabetic individuals and correlating key hepatic and plasma markers of inflammation, ER stress, and lipids with the homeostatic model assessment of insulin resistance index. We found that hepatic diacylglycerol (DAG) content in cytoplasmic lipid droplets was the best predictor of insulin resistance (R = 0.80, P < 0.001), and it was responsible for 64% of the variability in insulin sensitivity. Hepatic DAG content was also strongly correlated with activation of hepatic PKCε (R = 0.67, P < 0.001), which impairs insulin signaling. In contrast, there was no significant association between insulin resistance and other putative lipid metabolites or plasma or hepatic markers of inflammation. ER stress markers were only partly correlated with insulin resistance. In conclusion, these data show that hepatic DAG content in lipid droplets is the best predictor of insulin resistance in humans, and they support the hypothesis that NAFLD-associated hepatic insulin resistance is caused by an increase in hepatic DAG content, which results in activation of PKCε.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
DAG content in lipid droplets was the strongest predictor of insulin resistance. DAG, diacylglycerol; HOMA-IR, homeostatic model assessment of insulin resistance index; LCCoA, long-chain fatty acyl-CoA. Red dots and blue triangles show females and males, respectively. n = 35, 35, 28, 28, 28, and 32 for A, B, C, D, E, and F, respectively.
Fig. 2.
Fig. 2.
PKCε activation was strongly correlated with DAG content in lipid droplets. m/c, membrane/cytosol. Representative bands are labeled with colors and shown with corresponding colors on the graph (A). Circles and triangles show females and males, respectively (A). n = 30 for both A and B. Lipid droplet fraction was confirmed with Western blotting (C) and EM (D).
Fig. 3.
Fig. 3.
Correlation between ER stress markers and HOMA-IR. The relative expression level was expressed by setting the lowest expression level as one (A and B). Representative bands are labeled with colors and shown with corresponding colors on the graph (A and B). n = 30 and 25 for A and B, respectively.

Source: PubMed

3
Subskrybuj