Inhibition of Ceramide Production Reverses TNF-Induced Insulin Resistance Ross J. Grigsby and Rick T. Dobrowsky 1 Department of Pharmacology and Toxicology, University of Kansas, Lawrence, Kansas 66045 Received August 28, 2001 Ceramide has been implicated as a mediator of in- sulin resistance induced by tumor necrosis factor- (TNF) in adipocytes. Adipocytes contain numerous caveolae, sphingolipid and cholesterol-enriched lipid microdomains, that are also enriched in insulin recep- tor (IR). Since caveolae may be important sites for crosstalk between tyrosine kinase and sphingolipid signaling pathways, we examined the role of increased caveolar pools of ceramide in regulating tyrosine phosphorylation of the IR and its main substrate, in- sulin receptor substrate-1 (IRS-1). Neither exogenous short-chain ceramide analogs nor pharmacologic in- creases in endogenous caveolar pools of ceramide in- hibited insulin-induced tyrosine phosphorylation of the IR and IRS-1. However, inhibition of TNF-induced caveolar ceramide production reversed the decrease in IR tyrosine phosphorylation in response to TNF. These results suggest that TNF-independent increases in caveolar pools of ceramide are not sufficient to in- hibit insulin signaling but that in conjunction with other TNF-dependent signals, caveolar pools of cer- amide are a critical component for insulin resistance by TNF. © 2001 Academic Press Key Words: ceramide; TNF; adipocytes; insulin re- ceptor; insulin resistance; PPMP. Tumor necrosis factor- (TNF) is implicated as a causal mediator in the development of obesity-related insulin resistance, a major component of non-insulin- dependent diabetes mellitus (NIDDM) (1, 2). TNF in- hibits insulin signaling, at least in part, by decreasing the tyrosine phosphorylation of the insulin receptor (IR) and the major substrate for its intrinsic tyrosine kinase activity, insulin receptor substrate-1 (IRS-1). The mechanism by which TNF inhibits insulin signal- ing remains unclear. However, previous work has im- plicated ceramide as an endogenous mediator since addition of exogenous short chain ceramide analogs mimicked TNF in inhibiting tyrosine phosphorylation of the IR and IRS-1 in adipocytes (3–5). Ceramide is a bioactive sphingolipid metabolite pro- duced via the hydrolysis of membrane sphingomyelin following TNF-induced activation of either acidic or neutral sphingomyelinases (6). Caveolae are plasma membrane microdomains that may serve as localized sites for ligand-dependent sphingomyelinase activa- tion and ceramide production (7). Moreover, since in- sulin receptors are enriched in the numerous caveolae that populate the plasma membrane of mature adipo- cytes, they may serve as sites for crosstalk between tyrosine kinase and sphingolipid signaling pathways (8). In this study, we examined the role of increases in caveolar ceramide levels in inhibiting insulin signaling in adipocytes. MATERIALS AND METHODS Adipocyte differentiation and insulin stimulation. 3T3-L1 preadi- pocytes were obtained from the ATCC or were a kind gift from Dr. P. Scherer. Preadipocytes were cultured in DMEM–10% fetal calf serum plus antibiotics until 2 days postconfluency at which time they were induced to differentiate in DMEM–10% fetal calf serum containing 0.5 mM isobutylmethylxanthine–1 M dexamethasone and 1 g/ml insulin (9). The medium was changed every 2 days with DMEM-10% fetal calf serum and by 7 days postinduction, 90 –95% of the cells accumulated numerous cytoplasmic triglyceride droplets as determined by oil red O staining (9). Mature adipocytes were placed in serum free medium overnight and treated with 75 ng/ml recombinant mouse TNF for 24 h. In some experiments, the cells were treated with the broad specificity caspase inhibitor, benzoyloxycarbonyl-VAD-fluoromethylketone (zVAD.fmk), for 1 h prior to the addition of TNF. The cells were then stimulated with 0.1 M insulin for 5 min, washed twice with ice-cold phosphate-buffered saline containing 0.1 mM sodium orthovanadate and scraped into lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.2 mM sodium orthovanadate, 50 mM NaF, 60 mM -octylglucoside, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluo- ride, and 10 g/ml each of leupeptin, aprotinin, and bestatin). The cells were broken by sonication and cell debris removed by centrifugation at 3000g for 5 min at 4°C. Protein content was measured using the Bio- Rad reagent and 50 g of protein was used for SDS–PAGE. Western blot analysis. Following SDS–PAGE, the proteins were transferred to nitrocellulose and the membranes blocked with 1 To whom correspondence and reprint requests should be ad- dressed at Department of Pharmacology and Toxicology, University of Kansas, 5064 Malott Hall, Lawrence, KS 66045. Fax: (785) 864- 5219. E-mail: dobrowsky@ku.edu. Biochemical and Biophysical Research Communications 287, 1121–1124 (2001) doi:10.1006/bbrc.2001.5694, available online at http://www.idealibrary.com on 1121 0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.