Leptin-mediated increases in catecholamine signaling reduce adipose tissue inflammation via activation of macrophage HDAC4

Abstract

Obesity promotes systemic insulin resistance through inflammatory changes that lead to the release of cytokines from activated macrophages. Although the mechanism is unclear, the second messenger cAMP has been found to attenuate macrophage activity in response to a variety of hormonal signals. We show that, in the setting of acute overnutrition, leptin triggers catecholamine-dependent increases in cAMP signaling that reduce inflammatory gene expression via the activation of the histone deacetylase HDAC4. cAMP stimulates HDAC4 activity through the PKA-dependent inhibition of the salt-inducible kinases (SIKs), which otherwise phosphorylate and sequester HDAC4 in the cytoplasm. Following its dephosphorylation, HDAC4 shuttles to the nucleus where it inhibits NF-κB activity over proinflammatory genes. As variants in the Hdac4 gene are associated with obesity in humans, our results indicate that the cAMP-HDAC4 pathway functions importantly in maintaining insulin sensitivity and energy balance via its effects on the innate immune system.

Figure 1

Anti-inflammatory effects of cAMP in macrophages. A. and B. Effect of LPS i.p. (30mg/kg) on survival (A) and circulating cytokine concentrations (B) in 12 week old C57Bl/6J mice. Co-injection of phospho-diesterase inhibitor rolipram (5mg/kg) indicated. (n = 8; for this and other figures *P<0.05; **P<0.01). C. and D. Effect of PGE2 on cytokine mRNA amounts (C) and protein secretion (D) from Bone Marrow Macrophages (BMMs) treated with LPS. E. Effect of PGE2 on LPS-induced increases in JNK or P38 activation, IκBα phosphorylation, and on p65 acetylation. Co-treatment with increasing concentrations of PGE2 indicated. F. Chromatin immunoprecipitation (ChIP) assay of p65 recruitment and histone H4K5 acetylation over TNFα and IL12β promoters in BMMs exposed to LPS and PGE2.

Figure 2

Class IIa HDACs inhibit cytokine gene expression in response to cAMP. A. and B. Effect of LPS and PGE2 on HDAC4 de-phosphorylation (A) and cellular localization (B). A. Phospho (Ser246) HDAC4 amounts in control (HDAC4 fl/fl) and HDAC4 MKO cells exposed to LPS and PGE2. Amounts of acetylated and total p65 in wild-type and HDAC4MKO BMMs exposed to LPS plus PGE2 indicated. B. HDAC4 subcellular localization in BMMs exposed to LPS and PGE2. C. ChIP assay showing effects of PGE2 and LPS on HDAC4 recruitment to TNFα or IL12β promoters. D. and E. Effect of LPS and PGE2 on p65 occupancy (D) and histone H4K5 acetylation over TNFα and IL12β promoters (E) in BMMs from HDAC4 MKO or control littermates. F. Effect of LPS and rolipram administration on survival (left) and circulating cytokine concentrations (right) in control or HDAC4 MKO mice (n = 12).

Figure 3

LKB1 regulates cytokine gene expression by modulating HDAC4 phosphorylation. A. and B. Effect of LPS and PGE2 on HDAC4 de-phosphorylation (A) and nuclear localization (B) in control (LKB1 fl/fl) and LKB1 KO BMMs. A. Phospho (S246) HDAC4 protein amounts in LKB1 KO cells. Amounts of acetylated and total p65 indicated. (C.) and (D.) ChIP assays showing HDAC4 (C) and p65 recruitment (D) to TNFα and IL12β promoters in control or LKB1 KO BMMs. E. and F. Cytokine mRNA amounts (E) and protein secretion (F) from BMMs exposed to LPS and PGE2 as indicated.

Figure 4

Insulin resistance and obesity in HDAC4 MKO mice. A. Circulating norepinephine concentrations (left) and cAMP content in epididymal fat pads (right) of lean and ob/ob mice following intra-peritoneal (i.p.) injection with PBS or leptin (3ug/g) for 2hr (n = 3). Effect of propranolol (1ug/g) i.p. on cAMP content in leptin-treated mice (n = 3). B. HDAC4 localization in adipose tissue macrophages of lean and ob/ob mice. Mice were pre-injected with propranolol (1ug/g) or vehicle i.p. for 1hr followed by leptin (3ug/g) for 2hr (n = 3). Macrophages identified by co-staining with F4/80 antiserum. C. Circulating glucose, insulin, and free fatty acids in HDAC4 MKO compared to control littermates on a HFD for 8 weeks (n = 8). D. Glucose and Insulin tolerance testing of HDAC4 MKO and control (HDAC4 fl/fl) mice on a HFD for 8 weeks (n = 8). E. Macrophage infiltration in WAT by immunohistochemical (left) and Q-PCR (right) analyses. F. Effect of rolipram administration for 7 days on glucose and insulin tolerance in wild-type and HDAC4 MKO mice on a HFD for 12 weeks (n = 6). G. and H. Weight gain (G) (n = 8) as well as fat mass (n = 4) and circulating leptin levels (n = 8) (H) in HDAC4 MKO and control littermates.