AMP-activated protein kinase α2 subunit is required for the preservation of hepatic insulin sensitivity by n-3 polyunsaturated fatty acids

Abstract

Objective: The induction of obesity, dyslipidemia, and insulin resistance by high-fat diet in rodents can be prevented by n-3 long-chain polyunsaturated fatty acids (LC-PUFAs). We tested a hypothesis whether AMP-activated protein kinase (AMPK) has a role in the beneficial effects of n-3 LC-PUFAs.

Research design and methods: Mice with a whole-body deletion of the α2 catalytic subunit of AMPK (AMPKα2(-/-)) and their wild-type littermates were fed on either a low-fat chow, or a corn oil-based high-fat diet (cHF), or a cHF diet with 15% lipids replaced by n-3 LC-PUFA concentrate (cHF+F).

Results: Feeding a cHF diet induced obesity, dyslipidemia, hepatic steatosis, and whole-body insulin resistance in mice of both genotypes. Although cHF+F feeding increased hepatic AMPKα2 activity, the body weight gain, dyslipidemia, and the accumulation of hepatic triglycerides were prevented by the cHF+F diet to a similar degree in both AMPKα2(-/-) and wild-type mice in ad libitum-fed state. However, preservation of hepatic insulin sensitivity by n-3 LC-PUFAs required functional AMPKα2 and correlated with the induction of adiponectin and reduction in liver diacylglycerol content. Under hyperinsulinemic-euglycemic conditions, AMPKα2 was essential for preserving low levels of both hepatic and plasma triglycerides, as well as plasma free fatty acids, in response to the n-3 LC-PUFA treatment.

Conclusions: Our results show that n-3 LC-PUFAs prevent hepatic insulin resistance in an AMPKα2-dependent manner and support the role of adiponectin and hepatic diacylglycerols in the regulation of insulin sensitivity. AMPKα2 is also essential for hypolipidemic and antisteatotic effects of n-3 LC-PUFA under insulin-stimulated conditions.

FIG. 1.

Liver AMPKα1 (A) and AMPKα2 (B) activity in wild-type and AMPKα2^−/− mice fed either a Chow diet, cHF, or cHF+F for 9 weeks. The data are the means ± SE (n = 5–8). In the AMPKα2^−/− mice, AMPKα2 activity was below the detection limit. *P < 0.05 versus genotype Chow; †P < 0.05 versus genotype cHF.

FIG. 2.

Insulin sensitivity assessed by hyperinsulinemic-euglycemic clamp. GIR (A), GTO (B), HGP (C), whole-body glycolysis (GL-WB; D), whole-body glycogen synthesis (GS-WB; E); and glycogen synthesis in quadriceps muscle (GS-QM; F) were measured in wild-type and AMPKα2^−/− mice fed either a Chow diet, cHF, or cHF+F for 9 weeks. The data are the means ± SE (n = 5–8). *P < 0.05 versus genotype Chow; †P < 0.05 versus genotype cHF; ‡P < 0.05 versus wild-type on respective diet.

FIG. 3.

Triglyceride concentration in the livers of ad libitum-fed mice (A) and mice killed at the end of a 3-h infusion period of the hyperinsulinemic-euglycemic clamp (B). Wild-type and AMPKα2^−/− mice were fed either a Chow diet, cHF, or cHF+F for 9 weeks. The data are the means ± SE (A, n = 13–15; B, n = 8–14). *P < 0.05 versus genotype Chow; †P < 0.05 versus genotype cHF; ‡P < 0.05 versus wild-type on respective diet. For the detailed fatty acid composition of triglyceride fractions in the livers of ad libitum-fed mice, see supplementary Table 4.

FIG. 4.

The effect of differential dietary treatment on the regulation of metabolic fluxes in the liver. AICAR-stimulated fatty acid oxidation (A) and insulin-stimulated de novo fatty acid synthesis (B) in cultured hepatocytes isolated from wild-type and AMPKα2^−/− mice fed for 9 weeks either a Chow diet, cHF, or cHF+F. For basal nonstimulated rates of lipid metabolism, see supplementary Table 3. The expression of SCD-1 (C) and SREBP-1c (D) genes was quantified in total RNA isolated from the livers of mice subjected to hyperinsulinemic-euglycemic clamp following the differential dietary treatment for 9 weeks. The data are means ± SE (isolated hepatocytes, n = 3 in triplets; hepatic gene expression, n = 5–8). *P < 0.05 versus genotype Chow; †P < 0.05 versus genotype cHF; ‡P < 0.05 versus wild-type on respective diet. A.U., arbitrary units.

FIG. 5.

The composition of fatty acids in hepatic diacylglycerol fraction in ad libitum-fed wild-type and AMPKα2^−/− mice: total fatty acids (TFAs; A), PUFAs (B), monounsaturated fatty acids (MUFAs; C), and saturated fatty acids (SFAs; D). Animals were fed either a Chow diet, cHF, or cHF+F for 9 weeks. The data are the means ± SE (n = 13–15). *P < 0.05 versus genotype Chow; †P < 0.05 versus genotype cHF; ‡P < 0.05 versus wild-type on respective diet. For the detailed fatty acid composition of diacylglycerol fractions in the livers of ad libitum fed mice, see supplementary Table 4.

FIG. 6.

Putative involvement of AMPKα2 in antisteatotic action of n-3 LC-PUFAs in the liver. Dietary intake of n-3 LC-PUFAs increases the capacity of hepatocytes to oxidize fatty acids in wild-type (left panels) but not in AMPKα2^−/− mice (right panels). When insulin and glucose levels are high, such as during hyperinsulinemic-euglycemic clamp, wild-type mice fed n-3 LC-PUFAs exhibit improved hepatic insulin sensitivity and decreased plasma levels of NEFAs as compared with high-fat diet-fed controls. This is associated with increased expression of lipogenic genes such as SREBP-1c and SCD-1 and increased drive for de novo fatty acid synthesis. Despite the elevated lipogenic drive under clamp conditions, the livers of wild-type mice fed n-3 LC-PUFAs show reduced accumulation of triglycerides. However, in AMPKα2^−/− mice fed n-3 LC-PUFAs, hepatic triglyceride content is markedly elevated despite reduced rates of de novo fatty acid synthesis. This effect could be secondary to persisting elevated NEFA levels in circulation and thus better substrate availability in AMPKα2^−/− mice under clamp conditions. FA, fatty acids; TG, triglycerides.