Manu V. Chakravarthy, Yimin Zhu, Miguel López, Li Yin, David F. Wozniak, Trey Coleman, Zhiyuan Hu, Michael Wolfgang, Antonio Vidal-Puig, M. Daniel Lane, Clay F. Semenkovich
J Clin Invest.
2007;
117(9):2539–2552
doi:10.1172/JCI31183
This article Copyright © 2007, The American Society for Clinical Investigation
Abstract
|
Full text
|
PDF
C
entral nervous system control of energy balance affects susceptibility to obesity and diabetes, but how fatty acids, malonyl-CoA, and other metabolites act at this site to alter metabolism is poorly understood. Pharmacological inhibition of fatty acid synthase (FAS), rate limiting for de novo lipogenesis, decreases appetite independently of leptin but also promotes weight loss through activities unrelated to FAS inhibition. Here we report that the conditional genetic inactivation of FAS in pancreatic β cells and hypothalamus produced lean, hypophagic mice with increased physical activity and impaired hypothalamic PPARα signaling. Administration of a PPARα agonist into the hypothalamus increased PPARα target genes and normalized food intake. Inactivation of β cell FAS enzyme activity had no effect on islet function in culture or in vivo. These results suggest a critical role for brain FAS in the regulation of not only feeding, but also physical activity, effects that appear to be mediated through the provision of ligands generated by FAS to PPARα. Thus, 2 diametrically opposed proteins, FAS (induced by feeding) and PPARα (induced by starvation), unexpectedly form an integrative sensory module in the central nervous system to orchestrate energy balance.
Figure 1
Generation of FASKO mice.
(A) Homozygous FAS-floxed mice with loxP sites (black arrowheads) flanking exons 4–8 (gray boxes) were mated with RIPCre animals to yield FASKO mice and their littermate controls (WT, floxed mice without Cre). (B and C) PCR analysis. DNA from FASKO islets (I) and hypothalamus (H) produced a 317-bp product (B, lane 8; C, lane 4) using primers A (forward arrow) and B (reverse arrow) in A, indicating appropriate deletion of exons 4–8 of the FAS gene. The product was absent in the islets and hypothalamus in WT (B, lane 4; C, left panel, lane 2), as well as in epididymal fat pads (FAT), kidney (KID), liver (L), cerebral cortex (C), and cerebellum (C, right panel) of WT and FASKO mice. (D) Quantitative RT-PCR of tissues from RIPCre and FASKO mice showed Cre expression only in islets and hypothalamus. (E) Dorsal and ventral views of whole brain from FASKO and WT mice mated with ROSA26 animals and stained for β-galactosidase (blue staining), indicative of Cre expression (upper panel). Robust expression of Cre was largely restricted to the hypothalamus of FASKO mice (middle and lower panels), with weak expression in the striatum. Original magnification, ×1.5 (middle panels), ×6.3 (lower panels), ×6.3 (striatum). Other brain regions had no blue staining. (F and G) In situ hybridization. ARC, arcuate nucleus; VMH, ventral medial hypothalamus; Amyg, amygdala; CP, caudate putamen (striatum); Hb, habenula; CA1,2,3, hippocampus areas CA1, CA2, CA3; DG, hippocampus dentate gyrus; MC, motor cortex; PC, pyriform cortex; SC, sensory cortex; Thal, thalamus (zona incerta); SN, substantia nigra; 3V, third ventricle. (H) FAS mRNA expression in the hypothalamus by quantitative RT-PCR. (I) FAS enzyme activity in isolated islets (left panel), epididymal fat pads, and liver (right panel). (J) Determination of malonyl-CoA (MalCoA) content in isolated islets. (K) Malonyl-CoA content in the hypothalamus of 24-hour fasted (fast), and 24-hour fasted–24-hour refed (refed) mice. For D and G–K, results are mean ± SEM of 6–8 animals per group. *P < 0.05; **P < 0.001 compared with corresponding WT mice.
