Hepatic steatosis is strongly associated
Hepatic steatosis is strongly associated with insulin resistance. Numerous studies in humans and animals have shown that insulin-resistant states are often accompanied by hepatic steatosis (Angulo and Lindor, 2001, Marchesini et al., 1999, Marchesini et al., 2005, Petersen and Shulman, 2006, Seppala-Lindroos et al., 2002, Yki-Jarvinen, 2005). For example, leptin-deficient ob/ob mice, lipodystrophic mice, and mice fed a high-fat diet all develop severe insulin resistance and hepatic steatosis (Browning and Horton, 2004, den Boer et al., 2004, Halaas et al., 1995, Koteish and Mae Diehl, 2002, Shimomura et al., 1999a, Shimomura et al., 1999b).
However, the causal relationship between hepatic steatosis and insulin resistance is uncertain. On the one hand, it seems clear that systemic insulin resistance may lead to steatosis. Insulin resistance is often accompanied by hyperinsulinemia, which can activate hepatic signaling pathways and sterol regulatory element binding protein 1c (SREBP-1c), which in turn activates the transcription of genes involved in fatty cytochrome p450 inhibitors biosynthesis (Farese et al., 2005a, Matsumoto et al., 2003, Shimomura et al., 2000). On the other hand, whether hepatic steatosis is sufficient to cause hepatic insulin resistance is less clear. Acute intravenous infusions of lipids or chronic high-fat feeding can result in hepatic steatosis that is accompanied by insulin resistance (Boden et al., 1994, Boden et al., 2002, Collins et al., 2004, Gauthier et al., 2003, Winzell and Ahren, 2004). However, models of lipid administration are complex and may be accompanied by alterations not restricted to the liver, making it difficult to determine the contribution of steatosis to insulin resistance.
To directly test the hypothesis that hepatic steatosis leads to insulin resistance, we studied transgenic mice that overexpress human DGAT enzymes, which catalyze the final step of triacylglycerol (TG) biosynthesis (Buhman et al., 2001), in the liver. We focused on mice overexpressing DGAT2 (Liv-DGAT2 mice), the more potent (Stone et al., 2004) and specific (Yen et al., 2005) DGAT for increasing TG synthesis. We analyzed the phenotype of Liv-DGAT2 mice, focusing on hepatic lipid accumulation and the effects of steatosis on hepatic insulin signaling and whole-body glucose metabolism. To further test the hypothesis, we also analyzed glucose metabolism in a line of Liv-DGAT2 mice with higher expression levels and in transgenic mice that overexpress DGAT1 in the liver.
Discussion The goal of this study was to test the hypothesis that primary hepatic steatosis causes insulin resistance. We found that hepatic steatosis induced by overexpressing DGAT2 in the livers of transgenic mice did not cause insulin resistance. Similarly, hepatic steatosis induced by overexpressing DGAT1 in the livers of transgenic mice did not impair glucose or insulin tolerance. Thus, although hepatic steatosis is strongly associated with insulin resistance, hepatic lipid accumulation mediated by DGAT overexpression is insufficient to cause insulin resistance. Our aim was to generate a primary increase in the TG content of the liver by DGAT overexpression. This was accomplished in the Liv-DGAT2-low mice, which exhibited a 2-fold increase in DGAT2 mRNA levels and a 5-fold increase in liver TG content, a degree of steatosis that is physiologically relevant. Interestingly, DGAT2 overexpression also increased the content of DG, ceramides, and several species of long-chain fatty acyl-CoAs, all of which are lipid metabolites that have been implicated in insulin resistance (Kim et al., 2001, Neschen et al., 2005, Schaffer, 2003, Unger, 2002). It is unclear how DGAT overexpression increased the levels of these lipids. It may be that the livers were simply more fat-laden due to the accumulation of TG and its subsequent flux into degradation pathways. Another possibility is that DGAT overexpression upregulated de novo lipid synthesis in general, as suggested by the increases in the mRNAs of SREBP-1c and its target genes. How DGAT overexpression might induce de novo lipid synthesis is unclear, although it could reflect a homeostatic response to increased partitioning of fatty acids into TG.