Acute or chronic upregulation of mitochondrial fatty acid oxidation has no net effect on whole-body energy expenditure or adiposity

Abstract

Activation of AMP-activated protein kinase (AMPK) is thought to convey many of the beneficial effects of exercise via its inhibitory effect on acetyl-CoA carboxylase 2 (ACC2) and promotion of fatty acid oxidation. Hence, AMPK and ACC have become major drug targets for weight loss and improved insulin action. However, it remains unclear whether or how activation of the fatty acid oxidation pathway without a concomitant increase in energy expenditure could be beneficial. Here, we have used either pharmacological (administration of the AMPK agonist 5(') aminoimidazole-4-carboxamide-riboside) or genetic means (mutation of the ACC2 gene in mice) to manipulate fatty acid oxidation to determine whether this is sufficient to promote leanness. Both of these strategies increased whole-body fatty acid oxidation without altering energy expenditure or adiposity. We conclude that negative energy balance is a prerequisite for weight reduction, and increased fatty acid oxidation per se has little, if any, effect to reduce adiposity.

Figure 1. AMPK activation drives FAO but not energy expenditure

Rats were acclimatized overnight and at 10 am dosed with either vehicle (saline), AICAR (250mg/kg), or DNP (30mg/kg). (A) Avidin pulldown from skeletal muscle. ACC phosphorylation and expression 60 min after treatment. Pyruvate carboxylase (PC) is shown as a loading control for the pulldown. B) Malonyl-CoA levels in skeletal muscle described in (A) above. C–D) Indirect calorimetry measurement of RER and VO₂ before and after treatment with saline, DNP, or AICAR. Values indicated in the figure represent the average of the 10 h following dosing. E–H) EDL muscle strips were isolated and incubated ex vivo in KHB media with 5mM glucose and 0.5mM palmitate and either AICAR (2mM), or DNP (0.5mM). Rates of FA and glucose oxidation were measured as described in experimental procedures. I–J) Indirect calorimetry measurement of RER and VO₂ before and after treatment with saline, DNP, or AICAR in rats fed a high fat diet for 21 days. Values indicated in the figure represent the average of the 10 h following dosing. All results are displayed as means −/+ s.e.m. n>5, *P<0.05 for the 10 hr period following dosing.

Figure 2. Confirmation of ACC2 deletion and phenotype

A) Exon 12 encodes an essential region of the biotin carboxylase motif and was targeted for deletion. Exon 12 was flanked with loxP sites and subsequently bred with mice expressing Cre recombinase driven by the early acting PGK promoter. Cre-mediated recombination removes exon 12, the Neomycin selection cassette, and causes a frameshift mutation and an early stop codon. The PGK-Cre allele was subsequently bred out. The embryonic stem cells and all crosses were performed with pure C57BL/6 lineage. B) PCR confirmation of exon 12 deletion as described in experimental procedures. C) Protein confirmation of ACC2 deletion by monomeric avidin pulldown (see experimental procedures). Membranes were probed with streptavidin-800 and ACC2, ACC1, and propionyl-CoA Carboxylase (PCC) enzymes were identified based on molecular weight. D) Malonyl-CoA levels in skeletal and cardiac muscle of WT and ACC2^−/− mice. E) FAO rates in isolated soleus muscle from WT and ACC2^−/− mice. *p<0.05.

Figure 3. Chronic FAO increases glycogen storage but not energy expenditure

A–B) Indirect calorimetry measurements were used to determine RER and VO₂ in male WT and ACC2^−/− mice. *p<0.05 for ACC2^−/− vs. WT in the dark cycle. C–E) Body weight, body fat (by DEXA), and organ size (by dissection) were measured in male littermate mice 18–22 weeks of age, n>5. F) Glycogen content in hind limb skeletal muscle from male WT and ACC2^−/− mice, n=6. *p<0.05. G) 14C-glucose incorporation into lipid in liver and white adipose tissue in WT and ACC2^−/− mice following a glucose load (1.5 g/kg), n=4–7. *p<0.05. H) Triacylglycerol content per g of liver tissue in WT and ACC2^−/− mice, n=4.

Figure 4. Effects of chronic FAO on body weight gain and insulin sensitivity with 36 days of high fat feeding

A–C) Food intake, body weight gain, and fat pad weights were measured over 36 days of high fat feeding. D) Glucose tolerance test of 16 week old male WT and ACC2^−/− mice fed a chow diet or high fat diet, n=5–7 per group. E) Skeletal muscle insulin resistance was determined by measuring glucose disposal into skeletal muscle during the glucose tolerance test as described in experimental procedures. F) Insulin levels were not altered between WT and ACC2^−/− mice fed either a chow or high fat diet. G) Inhibition of ACC2 increases FAO without altering energy expenditure. These data suggest that other effects of AMPK agonists, possibly involving their effects on mitochondrial metabolism (Supplementary Fig S3) drive an increase in energy expenditure.