Interaction between Altered Insulin and Lipid Metabolism in CEACAM1-inactive Transgenic Mice
2004; Elsevier BV; Volume: 279; Issue: 43 Linguagem: Inglês
10.1074/jbc.m404764200
ISSN1083-351X
AutoresTong Dai, George A. Abou-Rjaily, Qusai Y. Al–Share, Yan Yang, Mats A. Fernström, Anthony M. DeAngelis, Abraham D. Lee, Lawrence Sweetman, Antonino Amato, Marzia Pasquali, Gary D. Lopaschuk, Sandra K. Erickson, Sonia M. Najjar,
Tópico(s)Glycosylation and Glycoproteins Research
ResumoInactivation of CEACAM1 in L-SACC1 mice by a dominant-negative transgene in liver impairs insulin clearance and increases serum free fatty acid (FFA) levels, resulting in insulin resistance. The contribution of elevated FFAs in the pathogenesis of insulin resistance is herein investigated. Treatment of L-SACC1 female mice with carnitine restored plasma FFA content. Concomitantly, it normalized insulin levels without directly regulating receptor-mediated insulin internalization and prevented glucose tolerance in these mice. Similarly, treatment with nicotinic acid, a lipolysis inhibitor, restored insulin-stimulated receptor uptake in L-SACC1 mice. Taken together, these data suggest that chronic elevation in plasma FFAs levels contributes to the regulation of insulin metabolism and action in L-SACC1 mice. Inactivation of CEACAM1 in L-SACC1 mice by a dominant-negative transgene in liver impairs insulin clearance and increases serum free fatty acid (FFA) levels, resulting in insulin resistance. The contribution of elevated FFAs in the pathogenesis of insulin resistance is herein investigated. Treatment of L-SACC1 female mice with carnitine restored plasma FFA content. Concomitantly, it normalized insulin levels without directly regulating receptor-mediated insulin internalization and prevented glucose tolerance in these mice. Similarly, treatment with nicotinic acid, a lipolysis inhibitor, restored insulin-stimulated receptor uptake in L-SACC1 mice. Taken together, these data suggest that chronic elevation in plasma FFAs levels contributes to the regulation of insulin metabolism and action in L-SACC1 mice. Insulin action is mediated by its binding to and activation of the insulin receptor tyrosine kinase to phosphorylate itself and other substrates (1Saltiel A.R. Kahn C.R. Nature. 2001; 414: 799-806Crossref PubMed Scopus (4063) Google Scholar). CEACAM1, an insulin receptor substrate in liver, but not in muscle or adipose tissue, regulates insulin action by promoting its receptor-mediated uptake and degradation in a phosphorylation-dependent manner (2Formisano P. Najjar S.M. Gross C.N. Philippe N. Oriente F. Kern B.C.L. Accili D. Gorden P. J. Biol. Chem. 1995; 270: 24073-24077Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar, 3Choice C.V. Howard M.J. Poy M.N. Hankin M.H. Najjar S.M. J. Biol. Chem. 1998; 273: 22194-22200Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar, 4Najjar S.M. Choice C.V. Soni P. Whitman C.M. Poy M.N. J. Biol. Chem. 1998; 273: 12923-12928Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). Overexpressing the dominant negative, phosphorylation-defective S503A CEACAM1 mutant in liver impaired insulin clearance and produced hyperinsulinemia in L-SACC1 1The abbreviations used are: L-SACC1, liver-specific S503A CEACAM1 mutant; IRα, α-subunit of the insulin receptor; TG, triacylglycerol; ACC1, acetyl-coenzyme A carboxylase 1; Glc-6-P, glucose-6-phosphate; FA, fatty acid; FFA, free fatty acid; WT, wild type. transgenic mice (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). Hyperinsulinemia caused insulin resistance in the L-SACC1 male mice at 2 months of age, the earliest age examined. These mice also developed altered fat metabolism with increased visceral adiposity, increased fasting plasma free fatty acids (FFAs), and triacylglycerols (TG) and increased hepatic TG content (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). The L-SACC1 mouse highlights the notion that intra-abdominal visceral adiposity and elevated plasma FFAs are commonly associated with impaired insulin clearance (6Svedberg J. Stromblad G. Wirth A. Smith U. Bjorntorp P. J. Clin. Invest. 1991; 88: 2054-2058Crossref PubMed Scopus (113) Google Scholar, 7Escobar O. Mizuma H. Sothern M.S. Blecker U. Udall Jr., J.N. Suskind R.M. Hilton C. Vargas A. Am. J. Med. Sci. 1999; 317: 282-286Crossref PubMed Scopus (27) Google Scholar, 8Bosello O. Zamboni M. Armellini F. Zocca I. Bergamo Andreis I.A. Smacchia C. Milani M.P. Cominacini L. Ann. Nutr. Metab. 1990; 34: 359-365Crossref PubMed Scopus (24) Google Scholar, 9Peiris A.N. Mueller R.A. Smith G.A. Struve M.F. Kissebah A.H. J. Clin. Invest. 1986; 78: 1648-1657Crossref PubMed Scopus (344) Google Scholar, 10Svedberg J. Bjorntorp P. Smith U. Lonnroth P. Diabetes. 1990; 39: 570-574Crossref PubMed Scopus (0) Google Scholar, 11Hennes M.M. Shrago E. Kissebah A.H. Int. J. Obes. 1990; 14: 831-841PubMed Google Scholar). It also emphasizes the important role of a CEACAM1-dependent insulin signaling downstream of the insulin receptor to directly regulate insulin clearance and sensitivity in liver and further regulate insulin action in extrahepatic tissues. Thus far, models of hepatic insulin resistance, including the LIRKO mouse with liver-specific insulin receptor ablation, which developed impaired insulin clearance, have primarily demonstrated that insulin signaling in hepatocytes is required to mediate insulin sensitivity in liver and extrahepatic tissues (12Michael M.D. Kulkarni R.N. Postic C. Previs S.F. Shulman G.I. Magnuson M.A. Kahn C.R. Mol. Cell. 2000; 6: 87-97Abstract Full Text Full Text PDF PubMed Scopus (1049) Google Scholar, 13Fisher S.J. Kahn C.R. J. Clin. Invest. 2003; 111: 463-468Crossref PubMed Scopus (173) Google Scholar). It is interesting, however, that the extent of impairment of fat metabolism is disproportionate to the extent of insulin resistance. For instance, LIRKO and other mice with primary hepatic insulin resistance did not develop elevated FFAs despite higher insulin levels than l-SACC1 transgenics (12Michael M.D. Kulkarni R.N. Postic C. Previs S.F. Shulman G.I. Magnuson M.A. Kahn C.R. Mol. Cell. 2000; 6: 87-97Abstract Full Text Full Text PDF PubMed Scopus (1049) Google Scholar, 14Kido Y. Burks D.J. Withers D. Bruning J.C. Kahn C.R. White M.F. Accili D. J. Clin. Invest. 2000; 105: 199-205Crossref PubMed Scopus (429) Google Scholar). This suggests that additional mechanisms may underlie insulin resistance in these models. The association of hepatic insulin resistance with elevation in FFA levels and increased visceral adiposity in L-SACC1 mice provides a potential mechanism to explain this apparent discrepancy. Chronic hyperinsulinemia caused by impaired insulin clearance in L-SACC1 mice may lead to increased hepatic triglyceride content and output (15Semenkovich C.F. Prog. Lipid Res. 1997; 36: 43-53Crossref PubMed Scopus (212) Google Scholar, 16Parks E.J. J. Nutr. 2001; 131: 2772S-2774SCrossref PubMed Google Scholar), thus contributing to hepatic insulin resistance that is brought about by the transgene. With normal pancreatic β-cell function in L-SACC1 mice, elevation in plasma triglycerides may, in turn, promote insulin secretion and proliferation of visceral adipose tissue (17McGarry J.D. Diabetes. 2002; 51: 7-18Crossref PubMed Scopus (1245) Google Scholar). This eventually increases plasma FFAs output even in the absence of lipolysis (18Robinson C. Tamborlane W.V. Maggs D.G. Enoksson S. Sherwin R.S. Silver D. Shulman G.I. Caprio S. Am. J. Physiol. 1998; 274: E737-E743PubMed Google Scholar). Exogenous plasma FFAs are preferentially removed by re-esterification in liver and by oxidation in muscle, heart, liver, and other tissues (19Lewis G.F. Carpentier A. Adeli K. Giacca A. Endocrinol. Rev. 2002; 23: 201-229Crossref PubMed Scopus (829) Google Scholar). When the uptake of FFAs is exceedingly high, it may interfere with glucose uptake. It may also elevate the level of long chain fatty acyl-CoA, reducing the inhibition of carnitine palmitoyltransferase 1 by malonyl-CoA (20Zammit V.A. Ann. N. Y. Acad. Sci. 2002; 967: 52-65Crossref PubMed Scopus (36) Google Scholar). This leads to increased transport of long chain fatty acyl-CoAs to the mitochondrial matrix to undergo β-oxidation (21Ruderman N.B. Saha A.K. Vavvas D. Witters L.A. Am. J. Physiol. 1999; 276: E1-E18Crossref PubMed Google Scholar). Conditions that partition the β-oxidation product, acetyl-CoA, to the citric cycle interfere with glucose metabolism and promote insulin resistance (22Randle P.J. Diabetes Metab. Rev. 1998; 14: 263-283Crossref PubMed Scopus (697) Google Scholar). The tight correlation between high fasting plasma FFAs and insulin resistance has been supported by the observation that fasting plasma FFAs are commonly elevated in obese and insulin-resistant individuals (23Baldeweg S.E. Golay A. Natali A. Balkau B. Del Prato S. Coppack S.W. Eur. J. Clin. Invest. 2000; 30: 45-52Crossref PubMed Scopus (107) Google Scholar). Given the regulatory role of plasma FFAs in insulin sensitivity, we investigated the role of FFAs in the pathogenesis of insulin resistance in L-SACC1 mice. To this end, we treated L-SACC1 females, which, unlike their male counterparts, did not develop hyperglycemia until 8 months of age, following increased visceral adiposity and impaired insulin clearance, with carnitine to normalize FFAs levels (24Maccari F. Arseni A. Chiodi P. Ramacci M.T. Angelucci L. Hulsmann W.C. Lipids. 1987; 22: 1005-1008Crossref PubMed Scopus (36) Google Scholar). We report that normalization of FFAs levels by carnitine in 6-month-old L-SACC1 mice restored insulin levels and prevented hyperglycemia. By decreasing lipolysis, nicotinic acid also restored receptor-mediated insulin uptake. Because carnitine does not modulate insulin internalization directly, these data suggest that increased visceral adiposity contributes to the pathogenesis of insulin resistance in L-SACC1 females. Animal Maintenance and Treatment—Animals were kept in a 12-h dark/light cycle and fed standard chow ad libitum. All procedures were approved by the relevant Institutional Animal Care and Utilization Committees at the Medical College of Ohio and Veterans Affairs Medical Center, San Francisco. When carnitine was used, 6-month-old wild type (WT) and L-SACC1 mice were treated at 1600 h for 1 or 2 weeks with a daily intraperitoneal injection of saline (vehicle-treated) or 0.2–1.5 g/kg body weight of L-carnitine, 0.4 g/ml saline (L-carnitine; Inner Salt Raw; Sigma). 6-Month-old mice were treated with two daily intraperitoneal injections of 200 mol/kg body weight of nicotinic acid (Sigma) or saline for 2 weeks (25Tunaru S. Kero J. Schaub A. Wufka C. Blaukat A. Pfeffer K. Offermanns S. Nat. Med. 2003; 9: 352-355Crossref PubMed Scopus (676) Google Scholar). Phenotypic Analysis—Following an overnight fast (with food being removed at 1700 h on the day prior to the experiment), mice were anesthetized with sodium pentobarbital (30 μg/g body weight) between 1100 and 1200 h. Whole venous blood was drawn from the retro-orbital sinuses to measure fasting glucose levels using a glucometer (Accu-chek; Roche Applied Science), plasma insulin, C-peptide, and leptin levels by radioimmunoassays (Linco Research), plasma FFAs using the NEFA C kit (Wako), triglycerides using the Infinity Triglycerides reagent (Sigma), and cholesterol using the Infinity Cholesterol reagent (Sigma). For liver and kidney functions, serum ALT (Sigma) and blood urea nitrogen (Infinity BUN reagent; Sigma) were measured, respectively. Visceral adipose tissues were weighed, and visceral adiposity was expressed as a percentage of total body weight. Liver and muscle triacylglycerols were determined as described previously (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). Serum β-Hydroxybutyrate—Following an overnight fast, mice were anesthetized, and blood was drawn to determine serum 3-hydroxybutyric acid levels by gas chromatography/mass spectroscopy following ethyl acetate/diethyl ether extraction of plasma deproteinized with 7% perchloric acid (26Tanaka K. West-Dull A.D.G.H. Lynn T.B. Lowe T. Clin. Chem. 1980; 26: 1847-1853Crossref PubMed Scopus (150) Google Scholar). Serum Carnitine Levels—Free carnitine and acylcarnitines were quantified in serum from fasted mice by tandem mass spectrometry (27Rashed M.S. Bucknall M.P. Little D. Awad A. Jacob M. Alamoudi M. Alwatter M. Ozand P.T. Clin. Chem. 1997; 43: 1129-1141Crossref PubMed Scopus (258) Google Scholar) with the following modifications. 15 μl of serum was extracted with 200 μl of methanol containing stable isotopically labeled internal standards. Dried extract was derivatized with butanolic HCl to form butyl esters that were in turn analyzed by electrospray tandem mass spectrometry of the precursor ions of m/z 85. CoA Ester Levels—Tissues were removed from fasted mice and snap frozen in liquid N2. Approximately 100–200 mg were added to an ice-cold mortar and pestle containing 1.5 ml of 6% perchloric acid and homogenized. The homogenate was centrifuged (2000 × g for 10 min at 4 °C), and 50 μl of 0.32 m dithiothreitol was added to the supernatant. CoA esters were measured using a modified HPLC procedure (28King M.T. Reiss P.D. Anal. Biochem. 1985; 146: 173-179Crossref PubMed Scopus (81) Google Scholar). Intracellular Glucose 6-Phosphate—The liver was removed from fasted 4-month-old WT and l-SACC1 mice and snap frozen. 1 g was homogenized, resuspended in 5 ml of 6 n perchloric acid, and centrifuged (3000 × g for 10 min at 4 °C). The pH of the supernatant was adjusted to 3.5, and it was placed on ice and mixed with 0.2 m triethanolamine buffer, 0.2 mm NADP, and 5 mm MgCl2 before adding 170 units/liter Glc-6-P-dehydrogenase. Absorbance was measured at 340 nm before and after the addition of enzyme, and Glc-6-P content was calculated in μmol/g of tissue (29Bergmeyer H.U. Bergmeyer N.U. Methods of Enzymatic Analysis 3. VCH Verlagsgesellschaft mbH, Weinheim, Germany1974: 1238-1242Google Scholar). Insulin and Glucose Tolerance Tests—These tests were performed on anesthetized fasted mice, as described previously (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). Glucose Uptake in Isolated Muscle—Glucose uptake in response to 200 microunits/ml of insulin by soleus muscle removed from hind limbs of fasted mice was determined as described previously (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). Insulin Binding and Internalization—Primary hepatocytes, isolated as described previously (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar), were grown for 24 h prior to [125I]insulin (10 pm) binding at 4 °C (30Li Calzi S. Choice C.V. Najjar S.M. Am. J. Physiol. 1997; 273: E801-E808PubMed Google Scholar). Unbound insulin was removed, and cells were incubated at 37 °C for 0–60 min before incubating in 0.1% bovine serum albumin/phosphate-buffered saline (pH 3.5) for 10 min. The acid wash was counted as surface-bound, noninternalized insulin, and NaOH-solubilized cells were counted as internalized cell-associated ligand. Internalized insulin was measured as a percentage of cell-associated per specifically bound ligand (the sum of surface-bound plus cell-associated ligand). Biotin Labeling of Surface Membrane Proteins—Following incubation of primary hepatocytes in the absence or presence of 100 nm insulin at 37 °C for 5 min, cells were incubated with biotin as previously described (3Choice C.V. Howard M.J. Poy M.N. Hankin M.H. Najjar S.M. J. Biol. Chem. 1998; 273: 22194-22200Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar). Following lysis in 1% Triton-X, proteins were immunoprecipitated with a polyclonal antibody against the α-subunit of the insulin receptor (α-IRα) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Proteins were then analyzed by electrophoresis and sequential immunoblotting with horseradish peroxidase-conjugated streptavidin and α-IRα, followed by detection with ECL. Insulin Receptor Quantification—Primary hepatocytes from 6-month-old mice were grown at 37 °C in triplicate plates before incubating at 4 °C for 5 h with insulin (0–1,000 ng/ml) and 125I-labeled insulin (20 pm; 5.0 × 104 cpm/ml) (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). NaOH-lysed cells were counted, and the insulin receptor number was determined by Scatchard plot analysis using the SCAFIT version 4.7 program (National Institutes of Health). Fatty Acid Synthesis in Primary Hepatocytes—Primary hepatocytes were grown in Dulbecco's modified Eagle's medium at a density of 2 × 106 cells/25 cm3, serum-starved overnight, and incubated with 3 ml of prewarmed fresh serum-free medium containing 0.5 μCi/ml [14C]acetate (Amersham Biosciences) in the absence or presence of 100 nm insulin at 37 °C for 30 min. Cells were harvested in 0.5 n KOH to determine protein concentration (Bio-Rad) and fatty acid synthesis in cpm of incorporated 14C/μg of proteins (31Maltese W.A. Reitz B.A. Volpe J.J. Biochim. Biophys. Acta. 1981; 663: 645-652Crossref PubMed Scopus (11) Google Scholar). Fatty Acid Synthase Activity—Livers were removed and homogenized in buffer containing 20 mm Tris, pH 7.5, 1 mm EDTA, 1 mm dithiothreitol, and phosphatase and protease inhibitors (32Najjar S.M. Philippe N. Suzuki Y. Ignacio G.A. Formisano P. Accili D. Taylor S.I. Biochemistry. 1995; 34: 9341-9349Crossref PubMed Scopus (74) Google Scholar). The homogenate was centrifuged at 12,500 × g for 30 min at 4 °C, and fatty acid synthase activity was assayed in the postmitochondrial supernatant (33Hennigar R.A. Pochet M. Hunt D.A. Lukacher A.E. Venema V.J. Seal E. Marrero M.B. Biochim. Biophys. Acta. 1998; 1392: 85-100Crossref PubMed Scopus (18) Google Scholar). Fatty Acid Synthesis in Vivo—Synthesis of fatty acids was determined in vivo by the 3HOH method in 7-month-old nonfasted mice at 1400–1500 h (34Erickson S.K. Lear S.R. Dean S. Dubrac S. Huling S.L. Nguyen L. Bollineni J.S. Shefer S. Hyogo H. Cohen D.E. Shneider B. Sehayek E. Ananthanarayanan M. Balasubramaniyan N. Suchy F.J. Batta A.K. Salen G. J. Lipid Res. 2003; 44: 1001-1009Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). Acetyl-CoA Carboxylase 1 Phosphorylation—The liver was removed from fasted 6-month-old mice and homogenized in buffer containing 20 mm Tris, pH 7.5, 1 mm EDTA, 1 mm dithiothreitol, and phosphatase and protease inhibitors (32Najjar S.M. Philippe N. Suzuki Y. Ignacio G.A. Formisano P. Accili D. Taylor S.I. Biochemistry. 1995; 34: 9341-9349Crossref PubMed Scopus (74) Google Scholar). Equal amounts of protein were immunoprecipitated with agarose-streptavidin, resolved by 6–15% gradient SDS-PAGE, and immunoblotted with α-phospho-acetyl-CoA carboxylase 1 (pACC1) antibody (Upstate Biotechnology, Inc., Lake Placid, NY) to detect phosphorylated ACC1 followed by blotting with horseradish peroxidase-streptavidin to determine the total amount of ACC1 in the immunopellet. The pACC1/ACC1 ratio was used as a measure of ACC1 activation (35Abu-Elheiga L. Almarza-Ortega D.B. Baldini A. Wakil S.J. J. Biol. Chem. 1997; 272: 10669-10677Abstract Full Text Full Text PDF PubMed Scopus (204) Google Scholar). mRNA Levels—Liver mRNA was purified using the MicroPoly(A) Pure kit (Ambion), analyzed by Northern blot, and probed with cDNAs for phosphoenolpyruvate carboxykinase, glucose-6-phosphatase, and glucokinase mRNA levels were normalized with β-actin. Statistics—Data were analyzed with Statview software (Abacus Concepts) using one-factor analysis of variance analysis. p values less than 0.05 were considered to be statistically significant. Abnormal Metabolism in L-SACC1 Female Transgenics—Similar to L-SACC1 males (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar), L-SACC1 females developed a 2–5-fold increase in plasma insulin levels by 2 months of age (Table I). They maintained normal glucose uptake in soleus muscle at submaximal insulin concentrations (200 microunits/ml) (Fig. 1A) and glucose transporter-4 mRNA levels (Fig. 1B). Glucose levels decreased to a similar extent in L-SACC1 and WT mice following insulin injection (Fig. 1C). This suggests that peripheral tissues in L-SACC1 mice, especially skeletal muscle, maintained intrinsic insulin sensitivity to glucose uptake. However, in control mice, glucose levels returned to normal within 3 h, whereas in L-SACC1 mice, they remained suppressed (Fig. 1C), consistent with the observation that L-SACC1 removed injected insulin less efficiently than WT mice.Table IMetabolic parameters in L-SACC1 female mice Phenotype characterization of age-matched 2-, 4-, and 6-month-old WT and L-SACC1 mice was performed as described under "Experimental Procedures." 10–12 mice per category were tested. Values are expressed as mean ± S.E.2 months4 months6 monthsWTL-SACC1WTL-SACC1WTL-SACC1Visceral fat/body weight (%)0.10 ± 0.002.60 ± 0.50ap < 0.05 L-SACC1 versus WT control mice0.10 ± 0.002.50 ± 0.30ap < 0.05 L-SACC1 versus WT control mice0.50 ± 0.104.00 ± 0.70ap < 0.05 L-SACC1 versus WT control miceFed glucose (mg/dl)136. ± 3.20131. ± 4.10114. ± 6.00122. ± 7.00112. ± 5.20107. ± 4.00Fasting glucose111. ± 9.5085.7 ± 4.00ap < 0.05 L-SACC1 versus WT control mice120. ± 10.5105. ± 5.30117. ± 6.4097.0 ± 5.10Plasma insulin (pm)43.3 ± 19.5155. ± 36.0ap < 0.05 L-SACC1 versus WT control mice25.4 ± 4.351.0 ± 12.6ap < 0.05 L-SACC1 versus WT control mice74.0 ± 9.494.0 ± 53.0ap < 0.05 L-SACC1 versus WT control micePlasma FFA (mm)0.40 ± 0.100.90 ± 0.10ap < 0.05 L-SACC1 versus WT control mice0.30 ± 0.000.60 ± 0.00ap < 0.05 L-SACC1 versus WT control mice0.30 ± 0.000.60 ± 0.10ap < 0.05 L-SACC1 versus WT control micePlasma TG (mg/dl)58.2 ± 9.80118. ± 8.90ap < 0.05 L-SACC1 versus WT control mice18.8 ± 3.0056.3 ± 5.00ap < 0.05 L-SACC1 versus WT control mice47.0 ± 3.1064.3 ± 6.00ap < 0.05 L-SACC1 versus WT control micea p < 0.05 L-SACC1 versus WT control mice Open table in a new tab Intraperitoneal glucose tolerance tests indicated that L-SACC1 females became increasingly glucose-intolerant with age (Fig. 2A). In view of the fact that muscle glucose uptake is normal, glucose intolerance is likely to arise from hepatic insulin resistance. This is supported by the ∼50% reduction (p < 0.05) in insulin receptor number in hepatocytes isolated from l-SACC1 females (Fig. 2B). Unlike l-SACC1 males, which developed random hyperglycemia by 2 months of age (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar), l-SACC1 females did not develop fed hyperglycemia until 8 months of age (107.0 ± 4.00 mg/dl versus 112.0 ± 5.20 in WT at 6 months of age and 135.4 ± 4.4 mg/dl versus 113.9 ± 6.6 in WT; p < 0.05 at 8 months of age). Accordingly, mRNA levels of phosphoenolpyruvate carboxykinase and glucose-6-phosphatase were significantly higher (∼3-fold) in l-SACC1 female livers at 8 but not at 6 months of age (Fig. 2C). In contrast, mRNA levels of glucokinase were not significantly altered at 8 months of age (Fig. 2C). Like males (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar), fasting glucose was normal in l-SACC1 females even at 8 months of age (110.3 ± 4.1 mg/dl versus 112.2 ± 4.1 in WT), consistent with normal β-cell insulin secretory function in l-SACC1 mice (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar). Altered Hepatic FFA Metabolism in L-SACC1 Females—Like their male counterparts (5Poy M.N. Yang Y. Rezaei K. Fernström M.A. Lee A.D. Kido Y. Erickson S.K. Najjar S.M. Nat. Genet. 2002; 30: 270-276Crossref PubMed Scopus (202) Google Scholar), L-SACC1 females exhibited increased body weight (not shown) and visceral adiposity at all ages examined (p < 0.05) (Fig. 3A). However, plasma leptin levels were normal (Fig. 3B, p > 0.05). This suggests that the food intake in L-SACC1 was normal and that visceral obesity is not attributed to changes in food intake. L-SACC1 mice also showed elevated fasting plasma FFAs and TG starting at 2 months of age (Table I). Hepatic acetyl- and malonyl-CoA levels were normal (Table II). Together with normal ACC1 activity in fasted L-SACC1 mice (pACC/ACC = 0.9–1.1 versus 1.1–1.3 in WT mice), this suggests that fatty acid synthesis is normal in L-SACC1 females under fasting conditions. Under nonfasting conditions, fatty acid synthase activity in liver extracts of 2-month-old L-SACC1 females was normal (174.3 ± 6.6 cpm/μg versus 177.7 ± 6.5 in WT mice); so was hepatic de novo fatty acid (FA) synthesis measured in vivo in 7-month-old L-SACC1 females (Table III). Similarly, fatty acid synthesis in whole body, skeletal muscle, heart, and small intestine of nonfasted L-SACC1 females was normal (Table III).Table IIEffects of L-carnitine on lipid and insulin metabolism of 6-month-old L-SACC1 females Vehicle- or carnitine-treated 6-month-old WT and L-SACC1 female mice (5–9 mice/group) were fasted overnight, and the whole venous blood was drawn to determine serum β-hydroxybutyrate carnitine, FFA, TG, insulin, and C-peptide levels. Liver and gastrocnemius muscle were removed and frozen in liquid nitrogen to measure CoA metabolites and TG. Visceral adipose tissues were collected, weighed, and expressed as a percentage of total body weight. Values are expressed as mean ± S.E.WT vehicleWT carnitineL-SACC1 vehicleL-SACC1 carnitineLipid metabolismVisceral fat/body weight (%)0.46 ± 0.110.33 ± 0.034.00 ± 0.75ap < 0.05 versus vehicle-treated WT1.62 ± 0.28bp < 0.05 carnitine versus vehicle-treated,cp < 0.05 carnitine-treated L-SACC1 versus vehicle-treated WTPlasma FFA (mmol/liter)0.58 ± 0.040.43 ± 0.03bp < 0.05 carnitine versus vehicle-treated1.14 ± 0.10ap < 0.05 versus vehicle-treated WT0.63 ± 0.04bp < 0.05 carnitine versus vehicle-treatedPlasma TG (mg/dl)35.9 ± 2.8332.4 ± 2.5560.6 ± 5.16ap < 0.05 versus vehicle-treated WT48.0 ± 5.09bp < 0.05 carnitine versus vehicle-treated,cp < 0.05 carnitine-treated L-SACC1 versus vehicle-treated WTLiver TG (μg/mg protein)40.2 ± 4.20104. ± 22.1bp < 0.05 carnitine versus vehicle-treated95.8 ± 12.1ap < 0.05 versus vehicle-treated WT74.6 ± 13.0cp < 0.05 carnitine-treated L-SACC1 versus vehicle-treated WTFatty acid metabolites in liverMalonyl-CoA (nmol/g dry weight)2.90 ± 0.805.30 ± 0.60bp < 0.05 carnitine versus vehicle-treated4.70 ± 0.904.80 ± 0.90Acetyl-CoA (nmol/g dry weight)144. ± 16.915.5 ± 1.40bp < 0.05 carnitine versus vehicle-treated149. ± 16.9167. ± 30.6CoA (nmol/g dry weight)650. ± 75.3708. ± 96.8645. ± 34.8622. ± 76.2Acetyl-CoA/CoA0.26 ± 0.050.04 ± 0.02bp < 0.05 carnitine versus vehicle-treated0.23 ± 0.020.27 ± 0.04LCFA-CoA (nmol/g dry weight)213.0 ± 53.0138.0 ± 23.8212.0 ± 91.8150.0 ± 17.2Fatty acid metabolites in plasmaβ-Hydroxybutyrate (mm)0.06 ± 0.010.07 ± 0.020.18 ± 0.04ap < 0.05 versus vehicle-treated WT0.17 ± 0.06cp < 0.05 carnitine-treated L-SACC1 versus vehicle-treated WTFree carnitine (μm)44.0 ± 14.023.8 ± 2.6032.7 ± 4.9049.6 ± 9.50Acetylcarnitine/total carnitine0.22 ± 0.040.35 ± 0.03bp < 0.05 carnitine versus vehicle-treated0.33 ± 0.03ap < 0.05 versus vehicle-treated WT0.24 ± 0.04Sum acylcarnitine/total carnitine0.25 ± 0.050.40 ± 0.03bp < 0.05 carnitine versus vehicle-treated0.38 ± 0.03ap < 0.05 versus vehicle-treated WT0.27 ± 0.05Insulin MetabolismPlasma Insulin (pm)38.5 ± 11.737.7 ± 11.382.2 ± 25.1ap < 0.05 versus vehicle-treated WT42.2 ± 5.72bp < 0.05 carnitine versus vehicle-treatedPlasma C-peptide (pm)199. ± 68.0200. ± 89.1245. ± 61.5245. ± 62.2C-peptide/insulin4.89 ± 0.294.76 ± 0.713.11 ± 0.28ap < 0.05 versus vehicle-treated WT5.65 ± 1.21bp < 0.05 carnitine versus vehicle-treateda p < 0.05 versus vehicle-treated WTb p < 0.05 carnitine versus vehicle-treatedc p < 0.05 carnitine-treated L-SACC1 versus vehicle-treated WT Open table in a new tab Table IIIIn vivo fatty acid synthesis (μmol of 3H2O incorporation/h/g) The experiment was performed on 7-month-old WT and L-SACC1 females (5–8 each). Values are expressed as mean ± S.E.Fatty acidsWTL-SACC1Whole body56.7 ± 8.1043.9 ± 3.10Liver31.5 ± 3.0641.6 ± 5.98Skeletal muscle1.95 ± 0.191.61 ± 0.22Heart5.39 ± 0.635.23 ± 0.37Small intestine10.6 ± 2.0610.0 ± 0.47a p < 0.01 versus WT. Open table in a new tab a p < 0.01 versus WT. Hepatic TG content was elevated in L-SACC1 females (Table II). Elevation in Glc-6-P levels (0.043 ± 0.006 μmol/g versus 0.015 ± 0.003 in WT; p < 0.05) suggests that this is in part due to increased esterification and TG synthesis in L-SACC1 liver. Because fatty acid synthesis is normal, FA substrates of TG synthesis are likely to derive from adipose tissue. Serum acyl- and acetylcarnitine/total carnitine ratios were high (Table II). Because carnitine esters that are released from the liver equilibrate with plasma more rapidly than those released from muscle and brain (36Leiter E. Herberg L. Diabetes Rev. 1997; 5: 131-148Google Scholar, 37Hokland B.M. Biochim. Biophys. Acta. 1988; 961: 234-241Crossref PubMed Scopus (17) Google Scholar), this suggests increased FA uptake into L-SACC1 hepatic mitochondria. This is supported by the observation that hepatic malonyl-CoA levels were not significantly elevated (Table II) to reduce FA uptake into the mitochondria. Moreover, the slight (∼3-fold) but significant elevati
Referência(s)