Liporegulation in Diet-induced Obesity
2001; Elsevier BV; Volume: 276; Issue: 8 Linguagem: Inglês
10.1074/jbc.m008553200
ISSN1083-351X
AutoresYoung Lee, May-Yun Wang, Tetsuya Kakuma, Zhuo-Wei Wang, Evelyn E. Babcock, Kay McCorkle, Moritake Higa, Yan-Ting Zhou, Roger H. Unger,
Tópico(s)Adipose Tissue and Metabolism
ResumoTo test the hypothesis that the physiologic liporegulatory role of hyperleptinemia is to prevent steatosis during caloric excess, we induced obesity by feeding normal Harlan Sprague-Dawley rats a 60% fat diet. Hyperleptinemia began within 24 h and increased progressively to 26 ng/ml after 10 weeks, correlating with an ∼150-fold increase in body fat (r = 0.91, p < 0.0001). During this time, the triacylglycerol (TG) content of nonadipose tissues rose only 1–2.7-fold implying antisteatotic activity. In rodents without leptin action (fa/fa rats and ob/ob anddb/db mice) receiving a 6% fat diet, nonadipose tissue TG was 4–100 times normal. In normal rats on a 60% fat diet, peroxisome proliferator-activated receptor α protein and liver-carnitine palmitoyltransferase-1 (l-CPT-1) mRNA increased in liver. In their pancreatic islets, fatty-acid oxidation increased 30% without detectable increase in the expression of peroxisome proliferator-activated receptor-α or oxidative enzymes, whereas lipogenesis from [14C]glucose was slightly below that of the 4% fat-fed rats (p < 0.05). Tissue-specific overexpression of wild-type leptin receptors in the livers offa/fa rats, in which marked steatosis is uniformly present, reduced TG accumulation in liver but nowhere else. We conclude that a physiologic role of the hyperleptinemia of caloric excess is to protect nonadipocytes from steatosis and lipotoxicity by preventing the up-regulation of lipogenesis and increasing fatty-acid oxidation. To test the hypothesis that the physiologic liporegulatory role of hyperleptinemia is to prevent steatosis during caloric excess, we induced obesity by feeding normal Harlan Sprague-Dawley rats a 60% fat diet. Hyperleptinemia began within 24 h and increased progressively to 26 ng/ml after 10 weeks, correlating with an ∼150-fold increase in body fat (r = 0.91, p < 0.0001). During this time, the triacylglycerol (TG) content of nonadipose tissues rose only 1–2.7-fold implying antisteatotic activity. In rodents without leptin action (fa/fa rats and ob/ob anddb/db mice) receiving a 6% fat diet, nonadipose tissue TG was 4–100 times normal. In normal rats on a 60% fat diet, peroxisome proliferator-activated receptor α protein and liver-carnitine palmitoyltransferase-1 (l-CPT-1) mRNA increased in liver. In their pancreatic islets, fatty-acid oxidation increased 30% without detectable increase in the expression of peroxisome proliferator-activated receptor-α or oxidative enzymes, whereas lipogenesis from [14C]glucose was slightly below that of the 4% fat-fed rats (p < 0.05). Tissue-specific overexpression of wild-type leptin receptors in the livers offa/fa rats, in which marked steatosis is uniformly present, reduced TG accumulation in liver but nowhere else. We conclude that a physiologic role of the hyperleptinemia of caloric excess is to protect nonadipocytes from steatosis and lipotoxicity by preventing the up-regulation of lipogenesis and increasing fatty-acid oxidation. triacylglycerol fatty acid Zucker Diabetic Fatty polymerase chain reaction adenocytomegalovirus acyl-CoA oxidase liver-carnitine palmitoyltransferase-1 peroxisome proliferator-activated receptor magnetic nuclear resonance spectroscopy Compelling theoretical considerations coupled with corroborating experimental evidence argue against the conventional view that the physiologic role of leptin is to prevent obesity. First, plasma leptin levels of rodents and humans are low in the lean and high in the obese (1Maffei M. Halaas J. Ravussin E. Pratley R.E. Lee G.H. Zhang Y. Fei H. Kim S. Lallone R. Ranganathan S. Kern P.A. Friedman J.M. Nat. Med. 1995; 1: 1155-1161Crossref PubMed Scopus (3327) Google Scholar), hardly the credentials of an antiobesity hormone. Second, diet-induced obesity is not prevented in hypoleptinemic mice by restoring their plasma leptin levels to normal with recombinant leptin (2Surwit R.S. Edwards C.L. Murthy S. Petro A.E. Diabetes. 2000; 49: 1203-1208Crossref PubMed Scopus (32) Google Scholar). Third, there is no evidence that overnutrition and obesity have ever posed a serious survival threat in evolution. On the contrary, the principal survival threat throughout evolution has been famine, against which obesity provides a measure of protection as the "thrifty gene" hypothesis maintains (3Neel J.V. Nutr. Rev. 1997; 57: S2-S9Crossref Scopus (268) Google Scholar). Finally, it seems implausible to suggest that hormones evolve for the purpose of preventing the clinical consequences of their own deficiency. Just as insulin evolved to confer advantages in nutrient metabolism rather than to prevent diabetic ketoacidosis, leptin must have evolved, not to prevent its deficiency syndrome, obesity (4Halaas J.L. Gajiwaia K.S. Maffei M. Cohen S.L. Chalt B.T. Rabinowitz D. Lallone R.L. Burley S.K. Friedman J.M. Science. 1995; 269: 543-546Crossref PubMed Scopus (4230) Google Scholar), but rather to confer a metabolic advantage that has not as yet been identified. We previously had suggested that the metabolic advantage conferred by the hyperleptinemia of obesity might be the prevention of overaccumulation of triacylglycerols (TG)1 in nonadipose tissues (5Unger R.H. Zhou Y.-T. Orci L. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 2327-2332Crossref PubMed Scopus (376) Google Scholar). Clearly, leptin does have powerful antilipogenic activity in some such tissues (6Shimabukuro M. Koyama K. Chen G. Wang M.-Y. Trieu F. Lee Y. Newgard C.B. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4637-4641Crossref PubMed Scopus (618) Google Scholar). In the absence of leptin action, lipogenesis is increased and fatty-acid (FA) oxidation is reduced (7Lee Y. Hirose H. Zhou Y.-T. Esser V. McGarry J.D. Unger R.H. Diabetes. 1997; 46: 408-413Crossref PubMed Scopus (175) Google Scholar), accounting for the steatosis and lipotoxicity that occur in such circumstances (7Lee Y. Hirose H. Zhou Y.-T. Esser V. McGarry J.D. Unger R.H. Diabetes. 1997; 46: 408-413Crossref PubMed Scopus (175) Google Scholar, 8Lee Y. Hirose H. Ohneda M. Johnson J.H. McGarry J.D. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10878-10882Crossref PubMed Scopus (719) Google Scholar, 9Zhou Y.-T. Grayburn P. Karim A. Shimabukuro M. Higa M. Baetens D. Orci L. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1784-1789Crossref PubMed Scopus (1077) Google Scholar). For example, in Zucker Diabetic Fatty (ZDF) rats with a loss-of-function mutation in the leptin receptors (10Iida M. Murakami T. Ishida K. Mizuno A. Kuwajima M. Shima K. Biochem. Biophys. Res. Commun. 1996; 224: 597-604Crossref PubMed Scopus (168) Google Scholar, 11Phillips M.S. Liu O. Hammond H. Dugan V. Hey P. Caskey C.T. Hess J.F. Nat. Genet. 1996; 13: 18-19Crossref PubMed Scopus (759) Google Scholar), tissue TG ranges from 10 to 50 times the normal content (8Lee Y. Hirose H. Ohneda M. Johnson J.H. McGarry J.D. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 10878-10882Crossref PubMed Scopus (719) Google Scholar) and is associated with functional impairment of pancreatic β-cells (12Hirose H. Lee Y.H. Inman L.R. Nagasawa Y. Johnson J.H. Unger R.H. J. Biol. Chem. 1996; 271: 5633-5637Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar, 13Wang M.-Y. Koyama K. Shimabukuro M. Newgard C. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 714-718Crossref PubMed Scopus (90) Google Scholar) and myocardium (9Zhou Y.-T. Grayburn P. Karim A. Shimabukuro M. Higa M. Baetens D. Orci L. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1784-1789Crossref PubMed Scopus (1077) Google Scholar) and insulin resistance (14McGarry J.D. Science. 1992; 258: 766-770Crossref PubMed Scopus (573) Google Scholar). Ultimately, the progressive overaccumulation of lipids causes death of cells in pancreatic islets and myocardium, resulting in diabetes and myocardial failure, which are the most serious complications of obesity. It has been proposed that the lipid overaccumulation enlarges the intracellular pool of fatty acyl-CoA beyond the oxidative requirements of the cell (15Unger R.H. Trends Endocrinol. Metab. 1998; 7: 276-282Google Scholar), thereby providing substrate for potentially destructive nonoxidative pathways, such as de novo ceramide formation (16Shimabukuro M. Zhou Y.-T. Levi M. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 2498-2502Crossref PubMed Scopus (1016) Google Scholar) and lipid peroxidation (17Obeid L.M. Linardic C.M. Karolak L.A. Hannun Y.A. Science. 1993; 259: 1769-1771Crossref PubMed Scopus (1612) Google Scholar, 18Vincent H.K. Powers S.K. Stewart D.J. Shanely R.A. Demirel H. Naito H. Int. J. Obes. Relat. Metab. Disord. 1999; 23: 67-74Crossref PubMed Scopus (118) Google Scholar). If the foregoing abnormalities develop in the absence of leptin action, it follows that leptin must be able to prevent them. Certainly hyperleptinemia induced by adenoviral transfer of the leptin gene has remarkable lipopenic and antilipogenic activity in tissues of normal rats, down-regulating the expression of genes involved in lipogenesis while up-regulating those genes involved in β-oxidation and thermogenesis (19Zhou Y.-T. Wang Z.-W. Higa M. Newgard C.B. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 2391-2395Crossref PubMed Scopus (208) Google Scholar). Although they are consistent with putative antisteatotic activity of hyperleptinemia, such studies do not prove that the actual physiologic role of adipocyte-derived hyperleptinemia in obesity is to prevent the ectopic accumulation of TG in nonadipose tissues. This study was designed to test this premise. Three groups of rodents were employed. Obese homozygous (fa/fa) ZDF-Drt rats, which are unresponsive to leptin because of a loss-of-function mutation in their leptin receptor (10Iida M. Murakami T. Ishida K. Mizuno A. Kuwajima M. Shima K. Biochem. Biophys. Res. Commun. 1996; 224: 597-604Crossref PubMed Scopus (168) Google Scholar, 11Phillips M.S. Liu O. Hammond H. Dugan V. Hey P. Caskey C.T. Hess J.F. Nat. Genet. 1996; 13: 18-19Crossref PubMed Scopus (759) Google Scholar), and lean wild-type (+/+) ZDF controls were bred in our laboratory from ZDF/Drt-fa (F10) rats purchased from Dr. R. Peterson (University of Indiana School of Medicine, Indianapolis, IN). Two groups of mice, C57BL/6J-ob/ob and C57BL/KS-J-db/db, and their wild-type controls, C57BL/6J +/+ and C57BL/KS-J +/+ mice, were purchased from the Jackson Laboratory (Bar Harbor, ME). To achieve diet-induced obesity in normal rats, Harlan Sprague-Dawley rats, purchased from Charles River Laboratories (Raleigh, NC) were employed. They were housed in individual metabolic cages (Nalgene, Rochester, NY) with a constant temperature and 12 h of light altering with 12 h of darkness. Body weight and food intake were measured weekly. Initially, all rats were fed standard chow (Teklad 4% mouse/rat diet, Teklad Madison, WI) ad libitum and had free access to water. At 4 weeks of age they either continued on this diet, which contains 24.8% protein, 4% fat, and 3.94 Kcal/g, or they were switched to a high fat diet (Purina Test Diet, Purina Mills, Inc., Richmond, IN) containing 60% fat, 7.5% carbohydrate, 24.5% protein, and 6.7 Kcal/g to produce diet-induced obesity. In in vivo experiments containing a total of 1 × 1012 plaque-forming units of recombinant adenovirus containing the cDNA of the leptin receptor OB-Rb (AdCMV-OB-Rb) or as a control β-galactosidase (AdCMV-β-galactosidase), prepared as described previously (20Chen G.X. Koyama K. Yuan X. Lee Y. Zhou Y.-T. O'Doherty R. Newgard C.B. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14795-14799Crossref PubMed Scopus (285) Google Scholar), were infused into conscious animals over a 10-min period through polyethylene tubing (PE-50, Becton Dickinson) previously anchored in the left jugular vein of 9-week-old ZDF fa/fa rats under sodium pentobarbital anesthesia (20Chen G.X. Koyama K. Yuan X. Lee Y. Zhou Y.-T. O'Doherty R. Newgard C.B. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14795-14799Crossref PubMed Scopus (285) Google Scholar). To compare the expression of wild-type OB-Rb in fa/fa rats with mutated OB-Rb, total RNA of rat liver and hypothalamus were extracted using TRIzol reagent (Life Technologies, Inc.). Reverse transcription of total RNA was carried out after treating RNA samples with RNase-free DNase I. The first strand cDNA was then used to PCR-amplify an OB-RbcDNA fragment with OB-Rb-specific primers encompassing the region with thefa/fa mutation as described previously (11Phillips M.S. Liu O. Hammond H. Dugan V. Hey P. Caskey C.T. Hess J.F. Nat. Genet. 1996; 13: 18-19Crossref PubMed Scopus (759) Google Scholar). The conditions of the PCR were as follows: denaturation for 45 s at 92 °C, annealing for 45 s at 55 °C, and elongation for 1 min at 72 °C. The amplified PCR products were digested with MspI at 37 °C 1 h and then run on a 1.2% agarose gel. Total RNA was extracted by the TRIzol isolation method, and Northern blot analysis was carried out as described previously (21Kakuma T. Lee Y. Higa M. Wang Z.-W. Pan W. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 8536-8541Crossref PubMed Scopus (230) Google Scholar). cDNA probes for the oxidative enzymes, acyl-CoA oxidase (ACO) and liver-carnitine palmitoyl- transferase-1 (liver-CPT-1), were prepared by reverse transciptase-PCR using the following primers: ACO-sense (amino acids 2891–2910), 5′-GCCCTCAGCTATGGTATTAC-3′ and ACO-antisense (amino acids 3505–3524), 5′-AGGAACTGCTCTCACAATGC-3′ (GenBankTMaccession number J02752); and liver-CPT-1-sense (amino acids 3094–3113), 5′-TATGTGAGGATGCTGCTTCC-3′ andliver-CPT-1-antisense (amino acids 3703–3722), 5′-CTCGGAGAGCTAAGCTTGTC-3′ (GenBankTM accession numberL07736). The DNA fragment excised after digesting pAC CMV-OB-Rb (13Wang M.-Y. Koyama K. Shimabukuro M. Newgard C. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 714-718Crossref PubMed Scopus (90) Google Scholar) with KpnI/HindIII restriction enzymes hybridizes only the intracellular domain of OB-Rb was also used as a probe of OB-Rb. The hybridization signals were analyzed by Molecular Imager GS-363 (Bio-Rad). Values were normalized to the signal generated with an 18 S ribosomal RNA (rRNA) gene probe. The procedure used was based on methods described by Jensen et al. (22Jensen J. Serup P. Karlsen C. Nielsen T.F. Madsen O.D. J. Biol. Chem. 1996; 271: 18749-18758Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar) and O'Doherty et al. (23O'Doherty R.M. Jensen P.B. Anderson P. Jones J.G. Berman H.K. Kearney D. Newgard C.B. J. Clin. Invest. 2000; 105: 479-488Crossref PubMed Scopus (69) Google Scholar). Total RNA (1 μg) was treated with RNase-free DNase (Promega), and first-strand cDNA was generated with the oligo(dT) primer in the first-strand cDNA synthesis kit (CLONTECH). Multiplex reverse transcriptase-PCR was carried out in 25-μl reactions with 1.5 μl of the diluted cDNA reaction as template mixed with 23.5 μl of PCR mix containing 1.25 units of Taq polymerase and buffer (Roche Molecular Biochemicals) containing 25 μm dATP, dTTP, and dGTP, 2.5 μCi of 2500 Ci/mmol [α-33P]dCTP (1 CI = 37 GBq) (Amersham Pharmacia Biotech), and 5 pmol of each primer (Table I). The standard thermal cycle profile was as follows for lipogenic enzyme mRNA (FAS, ACC, and GPAT) and β-oxidative enzyme gene mRNA (liver-CPT-1 and ACO): denaturation of 94 °C for 1 min, annealing at 55 °C for 1 min, and extension at 72 °C for 1 min for 24 cycles in liver and for 26 cycles in islets.Table IPrimers employed for multiplex reverse transcriptase-PCRGeneSense primer (5′-3′)Antisense primer (5′-3′)Predicted sizeSpeciesGenBank accession no.base pairsACCACAGTGAAGGCTTACGTCTGAGGATCCTTACAACCTCTGC242RatJ03808FASTGCTGTGGACCTCATCACTATGGATGATGTTGATGATAGAC297RatM76767GPATCCTCTGAACTGGAGAAGTGAAGACAGTATGTGGCACTCTC287RatAF021348TBPACCCTTCACCAATGACTCCTATGTGACTGCAGCAAATCGCTTGG186MouseD01034 Open table in a new tab Reaction products were separated on 7 m urea, 1× TBE (0.1m Tris base, 83 mm boric acid, 1 mmEDTA) and 6% polyacrylamide gels. Dried and PhosphorImager screens were scanned by a Molecular Imager System (GS-363). TATA box-binding protein mRNA was coamplified as an internal control, and data were expressed as ratios to its signal. To avoid biased results caused by potential interference between individual amplicons, we analyzed the amplification kinetics of individual amplicons in reactions where several products were coamplified. Representative experiments, in which mRNA encoding lipogenic enzymes and TATA box-binding protein in pancreatic islets were simultaneously amplified, show the noncompetitive amplification of individual products and their almost identical rate of amplification as indicated by the slopes within the exponential phase observed from a linear regression analysis. 50 mg of liver from the rats were homogenized in 2 ml of lysate buffer with proteinase inhibitors. A total of 100 μg of protein in 0.5 ml of buffer were used for precipitation with 1:500 goat anti-PPARα (C-20, Santa Cruz Biotechnology, Inc. Santa Cruz, CA). Protein A beads from Amersham Pharmacia Biotech were used for binding. Immunoblotting was carried out with rabbit-anti-PPARα from Calbiochem at 1:1500. Using the method of Stein et al.(24Stein D.T. Babcock E.E. Malloy C.R. McGarry J.D. Int. J. Obes. 1995; 19: 804-810Google Scholar), proton MRS and magnetic resonance imaging data were obtained with a 4.7-T 40-cm-bore system (Omega chemical shift imaging model, Bruker Instruments, Fremont, CA) with a 6-inch-diameter birdcage coil. Anesthetized rats were placed supine within the coil and positioned in the center of the magnet. Proton spectra of the rat were resolved into water and fat resonances, the areas of which were quantified using the nuclear magnetic resonance software program NRM-1 (Tripos Associates, St. Louis, MO) assuming equal line widths for both resonances. Proton images were obtained from the abdominal region of each rat. Spin-echo transaxial images were acquired with the following parameters: two transients, recycle time = 500 ms, echo time = 16 ms, 2-mm slice thickness, 2-mm interslice gap, eight slices, a 140-mm field of vision, and a 128 × 256 matrix. Images were analyzed using NIH Image software (National Institute of Mental Health, Bethesda, MD). Animals were sacrificed under sodium pentobarbital anesthesia. Tissues were dissected and placed in liquid nitrogen. Total lipids were extracted from about 100 mg of tissue by the method of Folch et al. (25Folch J. Lees M. Stanley G.H.S. J. Biol. Chem. 1957; 226: 497-509Abstract Full Text PDF PubMed Google Scholar) and dried under N2 gas. TG was assayed by the method of Danno et al. (26Danno H. Jicho Y. Budiyanto S. Furukawa Y. Kimura S. J. Nutr. Sci. Vitaminol. 1992; 38: 517-521Crossref PubMed Scopus (69) Google Scholar). Tail vein blood was collected in capillary tubes coated with EDTA. Plasma was stored at −20 °C. Plasma leptin was assayed using the Linco leptin assay kit (Linco Research, St. Charles, MO). Plasma glucose was measured by the glucose oxidase method using the glucose analyzer II (Beckman, Brea, CA). Plasma free fatty acids were determined using the Roche Molecular Biochemicals kit. Plasma TG levels were measured by the glycerol phosphate oxidase-Trinder triglyceride kit (Sigma). Oxidation of [3H]palmitate by islets was determined as described previously. Groups of 100–200 islets were incubated in duplicate with 1 mm9,10-[3H]palmitate for 3 days. Palmitate oxidation was assessed by measuring tritiated water in the medium. Excess [3H]palmitate was removed by precipitating twice with an equal volume of 10% trichloroacetic acid and 2% bovine serum albumin. Supernatants in a microcentrifuge tube were placed in a scintillation vial containing unlabeled water and incubated at 50 °C for 18 h. Tritiated water was measured as described for use of [3H]glucose (27Milburn J.L. Hirose H. Lee Y.H. Nagasawa Y. Ogawa A. Ohneda M. BeltrandelRio H. Newgard C.B. Johnson J.H. Unger R.H. J. Biol. Chem. 1995; 270: 1295-1299Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar). Incorporation of [U-14C]glucose (14.6 mmol/liter, PerkinElmer Life Sciences) into lipids was measured in islets as described previously in detail (28Chen S. Ogawa A. Ohneda M. Unger R.H. McGarry J.D. Diabetes. 1994; 43: 878-883Crossref PubMed Scopus (167) Google Scholar). About 200 islets were cultured for 3 days in medium containing 8 mmol/liter of glucose. After 3 days in culture, lipids were extracted from the islets according to the method of Bligh and Dyer (29Bligh E.G. Dyer W.J. Can. J. Biochem. Physiol. 1959; 37: 911-917Crossref PubMed Scopus (42865) Google Scholar), and counts incorporated into total lipid were determined. All values shown are expressed as mean ± S.E. Statistical analysis was performed by two-tailed unpaired Student's t test by one-way analysis of variance. If the function of leptin during caloric excess is to minimize the accumulation of lipids in nonadipose tissues, hyperleptinemia should begin promptly at the start of overnutrition and increase progressively as the overnutrition continues. To test this theory, a group of 10 normal male Harlan Sprague-Dawley rats was fed a diet in which 60% of the calories were derived from fat. Age-matched control rats received a 4% fat diet. Both groups were observed for 70 days. Plasma leptin levels in control rats were relatively unchanged, rising by only 0.04 ± 0.002 ng/ml/day to a level of only 2.80 ± 0.77 ng/ml on the final day of the 70-day study. By contrast, in rats on a 60% fat diet, plasma leptin rose to 4.3 ± 0.2 ng/ml (p < 0.001) within 24 h and increased progressively thereafter by 0.37 ± 0.07 ng/ml/day to a level of 26 ng/ml at 70 days (Fig. 1 A). In this group the rise in plasma leptin levels paralleled the expansion in body fat mass quantified by MRS (Fig. 1 B); there was a highly significant correlation between body fat and the plasma leptin level (r = 0.91, p < 0.0001) (Fig.1 C). Thus, leptin levels appear to respond promptly to a caloric excess, and they increase in proportion to enlargement of the adipose mass, which is consistent with the postulated role. If the hyperleptinemia induced by high fat feeding does in fact protect nonadipose tissues of normal rats from overaccumulation of lipids, their tissue TG content should remain low during the development of obesity despite the expansion of the adipose tissue mass and the concomitant rise in plasma lipid levels. To test this theory, we measured tissue TG content of nonadipose tissues 70 days after the start of the high fat diet at which point the total body fat measured by MRS had increased an ∼150-fold above the pre-diet base line (Fig. 1 B), and plasma TG and free fatty-acid levels were significantly higher (Fig.2 A). However, TG content in nonadipose tissues increased only 1–2.7-fold above the base line (Fig.2 B). Thus, nonadipose tissues of leptin-responsive hyperleptinemic rats accumulated only a small fraction of the total increase in body fat acquired over 70 days of excessive fat intake, during which time the animals had became grossly obese (Fig.1 B). In the liver, protection against steatosis might involve not only increased secretion of very low density liprotein but also enhanced FA oxidation. In the latter case, an increase in the expression of PPARα and its target enzymes liver-CPT-1 and ACO might be expected (30Zhou Y.-T. Shimabukuro M. Wang M.-Y. Lee Y. Higa M. Milburn J.L. Newgard C.B. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 8898-8903Crossref PubMed Scopus (163) Google Scholar). To determine whether the in vivo protection against hepatic overaccumulation of TG in normal rats on a high fat diet is mediated by this mechanism, we semiquantified PPARα protein and ACO and liver-CPT-1 mRNAs in livers of normal rats receiving either a 60 or a 4% fat intake. PPARα protein and liver-CPT-1 mRNA were both significantly greater in the former group, but ACO mRNA was not different (Fig. 3, A andB). Unlike liver, islets cannot export excess FA, which may account for their vulnerability in obesity. To determine the mechanism of the protection against lipid overaccumulation that prevails early in the course of obesity, we measured the rate of oxidation of [3H]palmitate in isolated islets of Harlan Sprague-Dawley rats receiving either a 4 or 60% fat diet. Oxidation was 30% greater in pancreatic islets of rats on the 60% fat diet than in controls that were on the 4% fat diet (Fig.4 A). However, unlike in liver, no change in ACO or liver-CPT-1 could be detected by multiplex-PCR (data not shown). These findings suggest that the preexisting oxidative machinery of the islets was able to accommodate this increase in oxidation without an increase in expression of genes encoding the enzymes. We had previously reported that in the absence of leptin activity as infa/fa ZDF rats, increased lipogenesis was the most important single factor in the ectopic overaccumulation of lipids in islets (7Lee Y. Hirose H. Zhou Y.-T. Esser V. McGarry J.D. Unger R.H. Diabetes. 1997; 46: 408-413Crossref PubMed Scopus (175) Google Scholar,31Cook W.S. Yelandi B.V. Ras M.S. Hashimoto T. Reddy K. Biochem. Biophys. Res. Commun. 2000; 278: 250-257Crossref PubMed Scopus (71) Google Scholar). Accordingly, in normal rats the high fat diet should not induce the increase in lipogenesis and lipogenic enzymes that had been observed in fat-laden islets of the leptin-insensitive fa/farats. As shown in Fig. 4, B and C, there was no increase in the incorporation of [14C]glucose to lipids or in the expression of lipogenic enzymes. In fact, a small but significant decrease in lipogenesis and in fatty acid synthase mRNA was evident (Fig. 4, B and C). This was in sharp contrast to the ZDF fa/fa rats in which lipogenesis was 2.5 times greater. If the antilipogenic protection observed in normal rats during caloric excess did in fact require the action of the accompanying hyperleptinemia, rodent models with either a leptin deficiency (ob/ob mice) or a loss-of-function mutation in their leptin receptors (db/db mice and ZDF fa/fa rats) would be unprotected from lipid overaccumulation. We, therefore, measured the plasma leptin levels (Fig. 5 A) and the TG content of islets, skeletal muscle, heart, and liver of these "unleptinized" rodents (Fig. 5 B). Although their diet contained only 6% fat, the TG content of their nonadipose tissues ranged from ∼4 to ∼100-fold above normal controls on the same diet. Thus, when leptin action is lacking, protection from lipid overaccumulation in nonadipocytes is also lacking, even when the dietary fat intake is normal. If the marked hepatic steatosis and hypertriglyceridemia of obese ZDF fa/fa rats are the result of a lack of direct leptin action on the liver, transgenic overexpression of the wild-type leptin receptor in the liver of these leptin receptor-defective animals should protect them. Therefore, we infused into 9-week-old ZDF fa/fa rats 1012plaque-forming units of recombinant adenovirus containing the cDNA of wild-type OB-Rb, the full-length isoform of the leptin receptor (AdCMV-OB-Rb). AdCMV-β-galactosidase was infused into age-matched ZDFfa/fa rats as a control. The wild-type OB-Rb transgene introduced in vivo with an adenovirus vector was expressed exclusively in the steatotic liver of the ZDF fa/fa rats (Fig. 6 A). None was detected in any other tissues including the hypothalamus. One week after treatment with AdCMV-OB-Rb plasma, TG levels of ZDFfa/fa rats declined slightly below pretreatment levels and remained significantly below the controls for 3 weeks after AdCMV-OB-Rb treatment (Fig. 5 B). Liver TG content was significantly less than that of β-galactosidase controls and untreated controls (Fig.5 C), the result of a delay in the increase in liver TG compared with the controls. TG contents of heart and skeletal muscle were unaffected (Fig. 6 B). Food intake was identical in the two adenovirus-treated groups (29.8 ± 1.4 g/dayversus 29.8 ± 1.5 g/day). Because the liver was the only site of expression of the normal OB-Rb in these ZDFfa/fa rats and the only site of antisteatotic action, we must assume that the elevated endogenous leptin levels, which averaged 24 ± 2 ng/ml in AdCMV-OB-Rb-treated rats and 28 ± 2 ng/ml in controls, exerted a direct antisteatotic action on the liver. This strongly implies that the function of hyperleptinemia of obesity is to prevent steatosis in tissues with functioning OB-Rb. These findings suggest that a physiologic role of leptin during overnutrition is to protect nonadipocytes from the adverse consequences of lipid overaccumulation. This protection begins promptly at the start of overfeeding as the result of progressively increasing hyperleptinemia that continues to rise for the duration of hypernutrition. This appeared to minimize overaccumulation of lipids both by preventing the increase in lipogenesis that occurs in the absence of leptin action (31Cook W.S. Yelandi B.V. Ras M.S. Hashimoto T. Reddy K. Biochem. Biophys. Res. Commun. 2000; 278: 250-257Crossref PubMed Scopus (71) Google Scholar) and through the up-regulation of β-oxidative metabolism of the surplus fatty acids (7Lee Y. Hirose H. Zhou Y.-T. Esser V. McGarry J.D. Unger R.H. Diabetes. 1997; 46: 408-413Crossref PubMed Scopus (175) Google Scholar). Whereas in the liver there was an increase in PPARα protein and CPT-1 mRNA, no such changes could be detected in pancreatic islets despite a 30% increase in the rate of [3H]palmitate oxidation. The greater induction of FA β-oxidative enzymes in liver than in extrahepatic tissues confirms a recent observation by Cook et al. (31Cook W.S. Yelandi B.V. Ras M.S. Hashimoto T. Reddy K. Biochem. Biophys. Res. Commun. 2000; 278: 250-257Crossref PubMed Scopus (71) Google Scholar). In islets the antilipogenic action of hyperleptinemia appears to be at least as important as the increase in FA oxidation in protecting islets from the lipid overload. When leptin action is lacking as in hyperphagic fa/fa ZDF rats, the fat-laden islets have a high rate of [14C]glucose incorporation into lipids in association with increased PPARγ, acetyl CoA carboxylase, and fatty acid synthase expression on a 6% fat intake (32Zhou Y.-T. Shimabukuro M. Lee Y. Koyama K. Higa M. Ferguson T. Unger R.H. Diabetes. 1998; 49: 1904-1908Crossref Scopus (69) Google Scholar). By contrast, in normal rats receiving the 60% fat diet, these rates remained in the low normal range and the lipogenic rate declined. When the antilipogenic effect of leptin is lacking, lipogenesis is excessive and is not reduced by lipid excess, as in normal islets (30,32). The most compelling evidence in support of the antisteatotic role for leptin was the in vivo demonstration in leptin-unresponsivefa/fa ZDF rats that transgenic overexpression of the wild-type receptor in their livers prevented the severe hepatic steatosis and hypertriglyceridemia that otherwise occurred. These findings are congruent with earlier evidence of the antisteatotic action of recombinant leptin (33Shimomura I. Hamner R.E. Ikemoto S. Brown M.S. Goldstein J.L. Nature. 1999; 401: 73-76Crossref PubMed Scopus (859) Google Scholar) and of transplanted fat tissue in "fatless" mice with congenital lipodystrophy (34Gavrilova O. Marcus-Samuels B. Graham D. Kim J.K. Shulman G.I. Castle A.L. Venson C. Eckhaus M. Reitman M.L. J. Clin. Invest. 2000; 105: 271-278Crossref PubMed Scopus (521) Google Scholar). Furthermore, in our experiments the wild-type leptin receptors were expressed only in the liver and not in the hypothalamus or anywhere else. Therefore, it follows that the endogenous hyperleptinemia of those obesefa/fa rats must have acted directly via the transgenic OB-Rb to prevent the lipid overaccumulation. The prompt rise of plasma leptin levels on the very first day of the high fat diet and their high degree of correlation with the expanding body fat are all consistent with the response of an antilipogenic hormone with a physiologic liporegulatory mission, namely to maintain FA homeostasis in nonadipocytes during overnutrition. This protection may account for the fact that in hyperleptinemic rats and humans the lipotoxic complications of diet-induced obesity do not appear until late in life when leptin effectiveness wanes (35Qian H. Azain M.J. Hartzell D.L. Baile C.A. Proc. Soc. Exp. Biol. Med. 1998; 219: 160-165Crossref PubMed Scopus (45) Google Scholar, 36Wang Z.W. Pan W.T. Lee Y. Kakuma T. Zhou Y.-T. Unger R.H. FASEB J. 2001; 13: 105-114Google Scholar). When leptin is absent as in congenital generalized lipodystrophy (33Shimomura I. Hamner R.E. Ikemoto S. Brown M.S. Goldstein J.L. Nature. 1999; 401: 73-76Crossref PubMed Scopus (859) Google Scholar) or when leptin receptors are congenitally defective as in ZDF rats, these complications appear in severe form early in life. It should be emphasized that we do not suggest that the direct antisteatotic activity ascribed to the endogenous hyperleptinemia of obesity occurs in normal lean animals. It appears to be a factor only during overnutrition when plasma leptin levels approach or exceed the threshold for transport across the blood-brain barrier, which is probably in the vicinity of 10 ng/ml (37Wang Z.-W. Zhou Y.-T. Kakuma T. Lee Y. Higa M. Kalra S.P. Dube M.G. Kalra P.S. Unger R.H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 10373-10378Crossref PubMed Scopus (79) Google Scholar). In the absence of overnutrition, plasma levels are below 5 ng/ml, and leptin action is presumed to be largely on the hypothalamic centers for control of food intake and thermoregulation (38Friedman J.M. Nature. 2000; 404: 632-634Crossref PubMed Scopus (629) Google Scholar). We thank Susan Kennedy for superb secretarial services. We also thank Drs. Cai Li and Daniel Foster for critical review of this manuscript. We thank Dr. Per Bo Jensen for expert help with multiplex reverse trnscriptase-PCR.
Referência(s)