Increased insulin receptor substrate-1 and enhanced skeletal muscle insulin sensitivity in mice lacking CCAAT/enhancer-binding protein beta.

CCAAT/enhancer-binding protein beta (C/EBPbeta) controls gene transcription and metabolic processes in a variety of insulin-sensitive tissues; however, its role in regulating insulin responsiveness in vivo has not been investigated. We performed hyperinsulinemic-euglycemic clamps in awake, non-stressed, chronically catheterized adult mice homozygous for a deletion in the gene for C/EBPbeta (C/EBPbeta(-/-)). Fasting plasma insulin, glucose, and free fatty acid (FFA) levels were significantly lower in C/EBPbeta(-/-) mice compared with wild-type (WT) controls. Acute hyperinsulinemia (4 h) suppressed hepatic glucose production, phosphoenolpyruvate carboxykinase mRNA, and plasma FFA to a similar extent in WT and C/EBPbeta(-/-) mice, suggesting that C/EBPbeta deletion does not alter the metabolic and gene regulatory response to insulin in liver and adipose tissue. In contrast, using submaximal (5 milliunits/kg/min) and maximal (20 milliunits/kg/min) insulin infusions, whole-body glucose disposal was 77% (p < 0.01) and 33% (p < 0.05) higher in C/EBPbeta(-/-) mice, respectively, compared with WT mice. Maximal insulin-stimulated 3-O-methylglucose uptake in isolated soleus muscle was 54% greater in C/EBPbeta(-/-) mice (p < 0.05). Furthermore, insulin-stimulated insulin receptor and Akt Ser(473) phosphorylation and phosphatidylinositol 3-kinase activity were 1.6-2.5-fold greater in skeletal muscle from C/EBPbeta(-/-) mice compared with WT mice. The level of insulin receptor substrate-1 protein was increased 2-fold in skeletal muscle from C/EBPbeta(-/-) mice. These results demonstrate that C/EBPbeta deletion decreases plasma FFA levels and increases insulin signal transduction specifically in skeletal muscle, and both contribute to increased whole-body insulin sensitivity.

The C/EBP 1 family of nuclear transcription factors includes C/EBP␣, -␤, -␦, and -⑀ and D-binding protein, and these are encoded by separate genes located on different chromosomes (1)(2)(3). The C/EBPs are members of the bZIP family of proteins characterized by a basic DNA-binding domain and a region containing evenly spaced leucines (the zipper) that mediates homo-and heterodimerization (4). C/EBP␣ was the first member of the family to be isolated and characterized in 1989 (5). C/EBP␣ mRNA is present in a variety of cells, but the protein is detected only in differentiated lung, liver, white and brown adipose tissue, and placenta (5). C/EBP␤ (also known as NF-IL6, IL-6DBP, LAP, CRP2, and NF-M; see Ref. 6) was cloned in 1990, and the protein preferentially accumulates (from highest to lowest) in liver, adipose tissue, the reproductive tract, and mammary gland (2,7).
Members of the C/EBP family control the transcription of genes involved in a broad range of physiological processes, ranging from the acute-phase response to the control of glucose and lipid homeostasis (8). C/EBP␣ is essential for normal liver and adipose cell development. Elimination of C/EBP␣ results in a metabolic derangement characterized by lethal fetal hypoglycemia and a failure to accumulate lipids (9). C/EBP␤ Ϫ/Ϫ mice are born in the expected numbers and appear indistinguishable from their littermates; however, a subset of the homozygous mice die during the perinatal period (phenotype A) (10,11). The surviving adult mice lacking C/EBP␤ demonstrate significantly decreased epididymal fat pad weight (phenotype B). In addition, adult C/EBP␤ Ϫ/Ϫ mice have decreased gluconeogenesis and lipolysis during fasting and diabetes (12,13). The C/EBP␤ knockout mice are refractory to glucagon, and cAMP levels are lower in their livers and adipose tissue (12). These results suggest that deleting the gene for C/EBP␤ affects signal transduction by counter-regulatory hormones, resulting in an animal model with lower fasting blood glucose and decreased fatty acid mobilization.
Insulin coordinately up-regulates C/EBP␤ in adipose tissue while down-regulating C/EBP␣ expression and phosphorylation, which, in turn, controls the expression of adipocyte-specific genes (14). In C/EBP␤ null mice, there is ϳ35% less epididymal white adipose tissue mass at 8 weeks of age; however, mRNAs coding for major markers of terminal adipocyte differentiation (LPL, aP2, C/EBP␣, and PPAR␥) are not affected (10). Although fewer in number, the adipocytes appear normal in morphology and size, suggesting that C/EBP␤ may be more important for regulating adipose tissue differentiation than proliferation. In hepatocytes, C/EBP␤ expression is down-regulated by insulin (15), resulting in decreased PEPCK gene transcription. Although C/EBP␤ gene deletion does not affect basal PEPCK gene expression, in streptozotocin-diabetic C/EBP␤ Ϫ/Ϫ mice, PEPCK and glucose-6-phosphatase mRNA levels are lower by 35-40% compared with diabetic wild-type mice, consistent with decreased gluconeogenesis in these animals (13).
The involvement of C/EBP␤ in insulin-regulated metabolic processes has not been extensively investigated. C/EBP␤ controls the expression of numerous insulin-responsive genes that regulate glucose transport and metabolism in muscle and adipose tissue (insulin receptor, IRS-1, and GLUT4) as well as regulates glucose output from the liver (16 -18). The actions of insulin are mediated by autophosphorylation of the insulin receptor, followed by activation of receptor tyrosine kinase, which phosphorylates insulin receptor substrates, particularly IRS-1 and IRS-2 (19,20). IRS-1 phosphorylation is required for activation of the enzyme PI3K, a necessary step for several effects of insulin, including glucose transport (21)(22)(23)(24)(25). The downstream elements of PI3K signaling that regulate glucose transport and gene expression have not been well defined. However, activation of the downstream serine/threonine kinase Akt (protein kinase B) has been suggested to be involved in several metabolic aspects of insulin signaling, including regulation of glycogen synthesis, activation of phosphodiesterase 3B, and stimulation of glucose transporter translocation (26 -28).
Our previous observation (12) of decreased adiposity and reduced hepatic glucose production in C/EBP␤ Ϫ/Ϫ mice suggested that C/EBP␤ could be involved in the regulation of insulin signaling in liver and peripheral tissues. The main aim of this study was to determine whether C/EBP␤ deletion alters whole-body insulin sensitivity using the insulin clamp technique in combination with tracer infusions to measure the effect of insulin on hepatic glucose production. To determine whether C/EBP␤ regulates the cellular response to insulin, we also investigated the effect of insulin on liver gene expression, lipolysis, and insulin signaling intermediates in tissues from C/EBP␤ Ϫ/Ϫ mice. Surprisingly, C/EBP␤ deletion did not affect insulin signaling or metabolism in liver and adipose tissue. Rather, C/EBP␤ deletion led to an improvement in whole-body insulin sensitivity; this may be exerted through decreased FFA and enhanced insulin signaling in skeletal muscle.

MATERIALS AND METHODS
Experimental Animals-Mice used in this study were obtained by cross-breeding female mice heterozygous for a null mutation of the C/EBP␤ gene with homozygous male mice. The generation and genetic background as well as the methods used for genotyping have been previously described by Screpanti et al. (29). Age-matched (10 -14 weeks old) wild-type littermates were used as controls. Wild-type and C/EBP␤ Ϫ/Ϫ mice were housed in microisolater cages and were maintained on a fixed 12-h light/dark cycle. Animals had free access to water and were fed regular animal chow (Harlan Teklad, Madison, WI) ad libitum. The normal mouse chow diet used in these studies was Teklad F6 8664, containing 24% protein, 6% fat, 4.5% crude fiber, and the remainder carbohydrate. Screening for C/EBP␤ Ϫ/Ϫ mice was carried out by Southern blot analysis as outlined previously (12).
Glucose and Insulin Challenge Tests-Glucose tolerance was measured in mice fasted for 6 h and injected intraperitoneally with 2 g of glucose/kg of body weight. Blood was sampled from the tail vein and assayed for glucose and insulin at 0, 30, and 60 min. For the insulin challenge test, mice were fasted overnight and injected with insulin (1 milliunit/kg of body weight), and blood glucose was measured subsequently at the times shown, without re-clipping of the tail.
Hyperinsulinemic-Euglycemic Clamp-Hyperinsulinemic-euglycemic clamp studies were performed according to Ren et al. (30). Three days prior to study, under ketamine anesthesia, a jugular catheter was surgically implanted in the right jugular vein and externalized in the interscapulum. On the morning after a 10-h fast, the mice were weighed and placed in a plexiglass cylinder, and the cannula was removed from under the skin. The jugular cannula was attached to a 30-mm length of PE-50 tubing entering a three-way connector. The second port was attached to a 150-mm length of PE-50 leading to the glucose pump (Harvard Apparatus Ltd.) containing a 1-ml syringe filled with 20% glucose. The third port was connected via a 35-mm length of tubing to a second three-way connector. The second port of this connector was attached to 75 mm of tubing leading to the insulin pump, and the third port was attached to tubing leading to the D-[3-3 H]glucose (14.3 Ci/ml) pump. The experimental protocol consisted of a 100-min equilibration period, followed by up to 240 min of insulin infusion. In total, ϳ600 l of blood/animal was sampled, and ϳ300 -500-l solutions were infused. The animals were primed with 0.5 Ci of [3-3 H]glucose during the first 3 min (in 30 l of saline), followed by 0.045 Ci (in 3.2 l of saline)/min of constant infusion throughout the remainder of the experiment. For the determination of basal glucose production, three blood samples (20 l) were obtained from the tail vein at 60, 75, and 90 min. These samples were centrifuged immediately (10 min at 10,000 rpm), and plasma was frozen for determination of unlabeled and 3 H-labeled glucose as detailed previously (12). At 90 min, Step 1 insulin infusion (5 milliunits/kg/min; regular Humulin, Lilly) was started at a rate of 10 l/min for 2 min (bolus), followed by constant infusion at 4.3 l/min for the experiment. Blood was obtained from the tail vein (2 l) every 5 min and analyzed using a One-Touch glucose meter (Lifescan). Once the blood glucose level reached 100 mg/dl, the 20% glucose pump was started. The speed of the glucose pump varied according to the insulin sensitivity. When the plasma glucose reached a steady state between 90 and 110 mg/dl for three consecutive 10-min intervals, 70 l of whole blood was obtained for glucose, insulin, FFA, and radioactivity determination. At the end of 100 -130 min of insulin infusion, Step 2 insulin infusion (20 milliunits/kg/min) was begun. The glucose infusion was adjusted to obtain a steady-state blood glucose level as close to 100 mg/dl as possible. At the end of the second phase of the clamp (200 -230 min), a 70-l blood sample was obtained; the animal was killed by overdose of anesthetic (ketamine); and the muscles, liver, and fat were removed, frozen immediately in liquid nitrogen, and stored at Ϫ70°C.
Calculations-The rate of glucose appearance (mg/min/kg) was calculated as the ratio of the rate of infusion of [3-3 H]glucose (dpm/min) and the steady-state plasma [ 3 H]glucose-specific activity (dpm/mol). The rate of hepatic glucose production (HGP; mg/kg/min) during the insulin clamp was calculated as the difference between the tracerderived rate of appearance (disposal) and the rate of glucose infusion. The rate of glucose disposal (mg/kg/min) was calculated by subtracting hepatic glucose production from the rate of glucose infusion required to maintain steady-state plasma glucose levels during the insulin clamp.
Measurement of Glucose Transport Activity-Maximal glucose transport was measured in muscles from fasted WT and C/EBP␤ Ϫ/Ϫ mice as described previously (31). Mice were anesthetized with ketamine (150 mg/kg) and acepromazine (5 mg/kg). The soleus muscle was removed intact from both hind limbs and incubated in vitro. The muscles were preincubated at 29°C for 30 min in 2 ml of Krebs-Henseleit bicarbonate buffer containing 1% bovine serum albumin, 32 mM mannitol, 8 mM D-glucose, and either 0 or 20 milliunits/ml bovine insulin and gassed continuously with 95% O 2 and 5% CO 2 . After preincubation, the muscles were rinsed for 10 min in fresh buffer containing 1% bovine serum albumin, 40 mM mannitol, and 100 nM insulin. The muscles were then transferred to fresh buffer containing 8 mM 3-O-methylglucose, 250 Ci/mmol 3-O-[ 3 H]methylglucose, 30 mM mannitol, 10 Ci/mmol [1-14 C]mannitol, and 2 mM sodium pyruvate with or without insulin for 10 min. After incubation, muscles were removed, trimmed of connective tissue, quickly blotted on gauze, and immediately freeze-clamped. Frozen muscles were weighed and digested in 0.5 ml of 1 M KOH for 30 min at 70°C and neutralized with 0.5 ml of 1 M HCl. A 0.3-ml aliquot of the supernatant was added to 5 ml of Cryoscint liquid scintillation fluid (ICN, Costa Mesa, CA). The specific activity of the incubation medium was obtained using 50-l samples obtained from each well. The incubation media samples were added to 950 l of 1 M KOH-HCl solution similar to the muscle digest, and all samples were counted for radioactivity in a Beckman LS 8100 liquid scintillation counter with dualquench correction. The rate of 3-O-[ 3 H]methylglucose transport was expressed as nmol/mg (wet weight)/10 min, after correction for extracellular 3-O-[ 3 H]methylglucose, and the results were analyzed by analysis of variance.
Western Blot Analysis-Western blot analysis was carried out in samples of gastrocnemius muscle and adipose tissue. To determine GLUT4 levels, membranes were prepared by methods described previously (31). The samples were homogenized in 10ϫ solubilization buffer containing 25 mM HEPES, pH 7.5, 1 mM EDTA, 0.8 g/ml aprotinin, 0.6 g/ml leupeptin, 1 g/ml pepstatin, and 50 g/ml phenylmethylsulfonyl fluoride, and the sample was centrifuged at 38,000 ϫ g for 60 min. The pellet was resuspended in solubilization buffer; and 10 -40 g of protein was treated with Laemmli sample buffer, boiled for 5 min, and resolved on 8% SDS-polyacrylamide gel using a Bio-Rad Mini-PROTEAN gel apparatus at 100 V for 1 h. To ensure that the proteins were in a linear range of detection, each sample was run in an average of three distinct assays on separate minigels using a concentration between 10 and 40 g of protein. Each gel contained an internal standard of a rat heart protein (20-g aliquot) prepared similarly to the skeletal muscle. For insulin receptor ␤, IRS-1, IRS-2, and p85␣ analysis, frozen samples were homogenized in 10 volumes of solubilization buffer containing 50 mM HEPES, pH 7.5, 137 mM NaCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 2 mM Na 3 VO 4 , 10 mM Na 2 P 2 O 7 , 10 mM NaF, 2 mM EDTA, 1% Nonidet P-40, 10% glycerol, 2 g/ml aprotinin, 10 g/ml antipain, 5 g/ml leupeptin, 0.5 g/ml pepstatin, 1.5 mg/ml benzamidine, and 34 g/ml phenylmethylsulfonyl fluoride using a Polytron PTA 20S generator at maximal speed for 30 s. The homogenate was then centrifuged at 65,000 rpm at 4°C in a Model Ti-70 rotor (Beckman Instruments) for 60 min to remove insoluble material, and the supernatant was used for analysis. Proteins were electrotransferred from the gel to nitrocellulose at 90 V (constant) for 1 h using a mini-transfer apparatus. Nonspecific protein binding to the filter was blocked using 5% milk, 10 mM Tris, 150 mM NaCl, and 0.02% Tween 20. The polyvinylidene difluoride filter was incubated with antibodies to insulin receptor ␤, IRS-1, IRS-2, p85␣, or GLUT4 (1.5 g/ml) diluted in blocking buffer for 4 h at 22°C, followed by extensive washing with Tris-buffered saline (150 mM NaCl, 10 mM Tris, and 0.05% Tween 20). The blots were incubated with secondary antibody linked to horseradish peroxidase in 10 ml of blocking buffer for 1 h at 22°C and washed again before exposing the membranes to the ECL reagent (Amersham Pharmacia Biotech) according to the manufacturer's instructions. Autoradiography was carried out using Kodak XAR x-ray film, with exposure times varied from 30 s to 3 min, and the average specific band intensities from each exposure were quantified by optical density using a Digiscan scanner (U. S. Biochemical Corp.) for integrating the autoradiographic signals. The results were expressed as arbitrary units relative to an internal standard sample (rat heart membrane) run together with each blot.
Insulin Receptor, PI3K, and Akt Phosphorylation-To observe the effects of C/EBP␤ deletion on insulin signaling, insulin (1 milliunit/kg of body weight) was administered to anesthetized mice via the portal vein as outlined previously (32). Muscle and fat biopsies were removed before and 5 min after maximal insulin injection. The level of insulin receptor tyrosine phosphorylation was determined after homogenization and immunoprecipitation with anti-phosphotyrosine antibodies, followed by Western blot analysis with SDS-polyacrylamide gel electrophoresis using anti-insulin receptor antibody as described above. The levels of total and Ser 473 -phosphorylated Akt were determined by Western blot analysis using specific antibodies (New England Biolabs Inc.) as described above. The results were expressed as arbitrary units above basal levels prior to insulin injection, after correction for total insulin receptor or Akt levels. IRS-1-associated PI3K activity was measured in muscle extracts after immunoprecipitation with IRS-1 overnight at 4°C (400 g of muscle protein/4 g of anti-IRS-1 antibody), followed by incubation with protein A-Sepharose overnight. The immunoprecipitation complex was spun at 14,000 ϫ g for 10 min, followed by washing three times with isotonic phosphate-buffered saline containing 1% Nonidet P-40; two times with 0.5 M LiCl 2 and 100 mM Tris-HCl, pH 7.6; and two times with 10 mM Tris-HCl, pH 7.4, 100 mM NaCl, and 1 mM EDTA. The pellets were resuspended in 50 l of the final wash buffer; 10 l of 100 mM MgCl 2 was added along with 10 l of a phosphatidylinositol mixture containing 0.5 mg/liter ␣-phosphatidylinositol (Sigma) in 10 mM Tris and 1 mM EGTA; and the tube was sonicated for 20 s. To start the PI3K reaction, 10 l of ATP mixture containing 100 mM MgCl 2 , 10 mM Tris, pH 7.5, 0.55 mM ATP, and 1 mCi/ml [␥-32 P]ATP was added for 10 min at room temperature. The reaction was stopped with 20 l of 8 N HCl and, 5 min later, with 160 l of CHCl 3 /MeOH (1:1). The phases were separated by centrifugation; and the lower organic phase was removed, lyophilized to dryness, and resuspended in 15 l of ethanol. 5 l of the product was then loaded onto a silica gel TLC plate precoated with 1% potassium oxalate. The lipids were resolved in CHCl 3 /MeOH/ H 2 O/NH 4 OH (60:47:11.3:2), dried, and visualized by autoradiography. The images were quantified using a Kodak Dynamic phosphoimager, and the results (in duplicate) were expressed as percent stimulation over basal (arbitrary units) relative to the WT control.
RNA Extraction and Northern Blot Analysis-Total RNA was extracted from the livers of control and insulin-clamped animals using the guanidine thiocyanate procedure as described previously (13). Solutions were made in diethyl pyrocarbonate-treated water, and materials were soaked overnight in diethyl pyrocarbonate-treated water. RNA concentration was determined by absorbance at 260 nm, and purity was checked by the A 260 /A 280 ratio and by minigel electrophoresis. 20 g of total RNA/sample was loaded onto a 0.66 M formaldehyde-containing 1.4% agarose gel in 1ϫ MOPS buffer and size-fractionated by electrophoresis. RNA was transferred overnight to a GeneScreen Plus membrane (NEN Life Science Products) and UV-cross-linked in a Stratalinker (Stratagene). Prehybridization was done at 65°C for 2 h in Church buffer. Random-primed labeling (kit from Life Technologies, Inc.) was used to generate 32 P-labeled cDNA (10 6 dpm/g). The probes used were PEPCK, a 1.1-kb fragment (kind gift of Dr. Richard W. Hanson); GLUT2, a 1.7-kb EcoRI-EcoRI fragment of the 3Ј-end of the rat cDNA (provided by Christopher B. Newgard, Southwestern Medical Center, Dallas, TX); and glucose-6-phosphatase, a 1.1-kb XbaI-PstI fragment of the human cDNA (provided by Dr. Janice Yang Chou, Genetics Branch, NICHD, National Institutes of Health, Bethesda, MD). The glucokinase cDNA probe was a 3.3-kb fragment obtained from Dr. Mark Magnuson (Vanderbilt University). The cDNA for C/EBP␤ was a 1.5-kb fragment obtained from Dr. Valeria Poli (University of Dundee, Dundee, Scotland). After hybridization overnight at 65°C, the membrane was washed 3 ϫ 5 min with 2ϫ SSC and 0.1% SDS at 65°C and 2 ϫ 30 min with 0.2ϫ SSC and 0.1% SDS at 65°C. The membrane was then exposed to a PhosphorImager cassette overnight. For reprobing, the blots were stripped at 80°C in 1% glycerol for 5-15 min or until no counts remained. For quantitative analysis, the intensity of the hybridized band was scanned using a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA), and densitometric values were quantitated using ImageQuant software. Densitometric values were normalized to mouse ribosomal RNA (28 S) to account for loading differences.
Analytical Procedures-Blood was taken from the tails in the morning and centrifuged, and plasma was separated and frozen. Plasma concentrations of non-esterified fatty acids and glucose were measured with diagnostic reagent kits from Wako Bioproducts and Sigma, respectively. Insulin and leptin levels in plasma were determined using radioimmunoassay kits from LINCO Research (St. Charles, MO). Percent body fat was measured by solubilizing the entire carcass in ethanolic potassium hydroxide (35% KOH (3 M) and 65% ethanol) overnight at 70°C. The triglyceride content of the solution was then measured using a semi-enzymatic quantitative triglyceride assay kit (Sigma). Statistical comparisons between groups were made using Student's t test or analysis of variance.

RESULTS
Reduced Adiposity in C/EBP␤ Ϫ/Ϫ Mice-C/EBP␤ has been suggested to play an integral role in hormonal control of adipogenesis (33). To quantify how C/EBP␤ deletion affected body composition in vivo, we compared the body weight and fat content of 16 -20-week-old C/EBP␤ Ϫ/Ϫ and age-and sexmatched wild-type littermates (Table I). Their average body weight was not significantly different; however, the total body lipid content was 38% lower in C/EBP␤ Ϫ/Ϫ mice (p Ͻ 0.05). A similar difference was noted for gonadal fat pad weight. The decreased adipose tissue mass was associated with reduced plasma leptin concentration in C/EBP␤ Ϫ/Ϫ mice (p Ͻ 0.05). Fasting (6 h) plasma glucose levels were significantly less in C/EBP␤ Ϫ/Ϫ mice by 25% (p Ͻ 0.05), whereas insulin was 35% lower (p Ͻ 0.05).
Increased Insulin-stimulated Glucose Disposal in C/EBP␤ Ϫ/Ϫ Mice-Glucose tolerance tests were carried out on 6-h fasted mice (Fig. 1). The overall levels of glucose and insulin were lower in C/EBP␤ Ϫ/Ϫ mice during glucose tolerance (p Ͻ 0.05), suggestive of increased insulin sensitivity. The insulin tolerance test was performed as a more direct measure of insulin sensitivity. After an intraperitoneal insulin injection, the percent decline in plasma glucose at 15, 30, and 60 min was significantly greater in C/EBP␤ Ϫ/Ϫ mice compared with wildtype mice (p Ͻ 0.05). The decrease in heterozygous mice was not significant compared with wild-type mice, however. These results suggest that C/EBP␤ Ϫ/Ϫ mice have both a reduction in the base-line plasma glucose and insulin concentrations and a significant enhancement of glucose disposal in response to insulin.
To more rigorously test the effects of insulin at the same glucose and insulin concentrations on hepatic and peripheral insulin sensitivity, glucose turnover rates and HGP were assessed during a two-step hyperinsulinemic-euglycemic clamp. Mice were chronically catheterized via the jugular vein as described under "Materials and Methods." After a 10-h fast, blood glucose was clamped between 90 and 110 mg/dl for up to 4 h. Steady-state conditions for plasma glucose concentration and specific activity were achieved at 60 and 200 min of the clamp studies. The levels of insulin achieved during the hyperinsulinemic clamp conditions were nearly identical in C/EBP␤ Ϫ/Ϫ and wild-type mice (Table II). The rate of steadystate whole-body glucose disposal (mg/kg/min) at the same plasma insulin concentration was 77% higher (p Ͻ 0.01) at 5 milliunits of insulin/ml/kg and 33% higher at 20 milliunits/ ml/kg (p Ͻ 0.05) in C/EBP␤ Ϫ/Ϫ mice compared with wild-type controls (Fig. 2).
Effects of Insulin on Hepatic Glucose Production in C/EBP␤ Ϫ/Ϫ Mice-The increase in insulin-mediated glucose disposal in C/EBP␤ Ϫ/Ϫ mice could be due to increased glucose uptake and/or to enhanced suppression of HGP by insulin. HGP was determined in the basal state and during insulin infusion using D-[3-3 H]glucose as outlined under "Materials and Methods." Prior to insulin infusion, fasting HGP was significantly lower by 35% in C/EBP␤ Ϫ/Ϫ mice (p Ͻ 0.05). During insulin infusion, HGP was similarly inhibited by insulin during the hyperinsulinemic-euglycemic clamp conditions at 5 milliunits/kg/min of infusion (wild-type, 14.2 Ϯ 3.4 mg/kg/min; and C/EBP␤ Ϫ/Ϫ , 11.6 Ϯ 2.2 mg/kg/min; p ϭ not significant). Insulin nearly completely suppressed HGP at 20 milliunits/kg/min of infusion (wild-type, 1.7 Ϯ 0.8 mg/kg/min; and C/EBP␤ Ϫ/Ϫ , 1.9 Ϯ 0.7 mg/kg/min), suggesting that the greater glucose disposal in C/EBP␤ Ϫ/Ϫ mice was largely due to an increase in glucose uptake in peripheral tissues rather than a change in hepatic sensitivity to insulin.
Effects of Insulin on Hepatic Gene Expression and FFA Levels in C/EBP␤ Ϫ/Ϫ Mice-To investigate the downstream effects of insulin on the livers from C/EBP␤ Ϫ/Ϫ mice, we determined the levels of mRNA coding for several insulin-regulated genes before and after hyperinsulinemic clamp conditions (Fig. 3). In the fasted state prior to the hyperinsulinemic clamp, the levels of glucose-6-phosphatase, PEPCK, glucokinase, and GLUT2 were not significantly different. At the end of 240 min of insulin infusion, the levels of PEPCK, glucose-6-phosphatase, and GLUT2 mRNA expression were similarly reduced by ϳ70 -85% (p Ͻ 0.05) in C/EBP␤ Ϫ/Ϫ and wild-type mice. Similarly, insulin increased the relative levels of glucokinase mRNA by 2-3-fold

Steady-state plasma insulin and non-esterified fatty acid concentrations at base line and during 5 and 20 milliunits/kg/min
hyperinsulinemic clamp conditions Mice were chronically catheterized and glucose-maintained between 90 and 110 mg/dl during the insulin clamp by determining glucose every 10 min and adjusting the rate of unlabeled glucose infusion (20% glucose) to maintain euglycemia. Insulin was infused at a rate of 5 milliunits/kg/min for up to 90 min, followed by 20 milliunits/kg/min for a total of 1.5-3 h, depending on the ability to achieve a steady-state glucose concentration. Values are means Ϯ S.E. (n ϭ 8 C/EBP␤ Ϫ/Ϫ mice and 10 wild-type mice).

FIG. 1. Glucose and insulin tolerance tests carried out on wild-type C/EBP␤ ؉/؉ (OE), C/EBP␤ ؉/؊ (q), and C/EBP␤
؊/؊ (f) mice. Glucose tolerance tests were carried out on mice fasted for 6 h and injected intraperitoneally with 2 mg of glucose/kilogram of body weight. For insulin tolerance, porcine insulin (1 milliunit/kg of body weight) was injected intraperitoneally into conscious 10 -14week-old female mice. Plasma glucose levels in samples obtained from the tail vein were measured as described under "Materials and Methods." Results are expressed as percent of the starting fasting glucose level and are means Ϯ S.E. from 6 to 12 animals/group. *, significantly reduced compared with wild-type controls (p Ͻ 0.05).
(p Ͻ 0.05) during the insulin clamp in both wild-type and C/EBP␤ Ϫ/Ϫ mice, indicating that the response to insulin was similar in C/EBP␤ Ϫ/Ϫ mice. This was reflected at the protein level by no change in GLUT2 expression levels and similar PEPCK activity (data not shown).
To determine the effects of insulin on the regulation of adipose tissue metabolism, we measured plasma FFA levels before and during the insulin clamp. In C/EBP␤ Ϫ/Ϫ mice, fasting plasma FFA levels were lower before the hyperinsulinemic clamp and during the first 100 min of insulin infusion (Table  II). Insulin acutely decreases the release of FFA, primarily as a result of the effect of insulin on inhibiting hormone-sensitive lipase (34,35). After 100 min of hyperinsulinemia, FFA levels fell by a similar absolute amount in wild-type and C/EBP␤ Ϫ/Ϫ mice (0.39 versus 0.41 mM for wild-type and C/EBP␤ Ϫ/Ϫ , respectively). At the end of 240 min of hyperinsulinemia, the levels of FFA were not different in wild-type and C/EBP␤ Ϫ/Ϫ mice (p ϭ not significant). Overall, the level of suppression of plasma FFA was identical in controls and C/EBP␤ Ϫ/Ϫ mice (wild-type, 84 Ϯ 14%; and C/EBP␤ Ϫ/Ϫ , 84 Ϯ 2%), suggesting that C/EBP␤ deletion had no effect on the ability of insulin to suppress lipolysis.
Increased Glucose Transport Activity in Skeletal Muscle from C/EBP␤ Ϫ/Ϫ Mice-Because C/EBP␤ deletion increased insulin-stimulated glucose disposal in vivo, we wanted to determine whether skeletal muscle was more sensitive to insulin in vitro.
In intact soleus muscle from wild-type mice, 100 nM insulin increased 3-O-methylglucose transport by 210% (Fig. 4). In C/EBP␤ Ϫ/Ϫ mice, insulin-stimulated glucose transport was 54% greater (p Ͻ 0.05), with no change in basal glucose transport activity, indicating an enhanced response to insulin in isolated skeletal muscle. Thus, skeletal muscle from C/EBP␤ Ϫ/Ϫ mice appeared to be more responsive to insulin both in vivo and in vitro. Increased Insulin Receptor and Akt-1 Phosphorylation and PI3K Activity in Skeletal Muscle from C/EBP␤ Ϫ/Ϫ Mice-Given the increase in glucose transport activity in skeletal muscle from C/EBP␤ Ϫ/Ϫ mice, we wanted to determine whether this was associated with an increase in the insulin receptor signal transduction cascade. We measured the phosphorylation of the insulin receptor, PI3K activity, and the activation of the downstream serine/threonine kinase Akt-1 in response to insulin. Mice were treated with insulin; and after 5 min, skeletal muscle and adipose tissue were removed, and proteins were analyzed by Western blotting using anti-phosphotyrosine or serine-specific antibodies. In muscle from C/EBP␤ Ϫ/Ϫ mice, insulin-stimulated insulin receptor ␤ phosphorylation was increased 1.9-fold (Fig. 5, upper panel), whereas IRS-1-associated PI3K activity was significantly greater by 1.7-fold compared with WT mice (p Ͻ 0.05) (middle panel). Similarly, the level of insulin-stimulated Akt Ser 473 phosphorylation was 2.5-fold greater than in wild-type mice (p Ͻ 0.05) (lower panel). The total amount of Akt was not different in the muscle from C/EBP␤ Ϫ/Ϫ mice, suggesting that this did not contribute to higher levels of phosphorylated Akt. In adipose tissue, the level of Akt and the increase in phosphorylation were similar in WT and C/EBP␤ Ϫ/Ϫ mice after insulin stimulation (data not shown). These results indicate that the increase in insulin signal transduction was specific to skeletal muscle and included increases in specific kinases involved in glucose transport activity.
Expression of Insulin Signaling Proteins in Tissues from C/EBP␤ Ϫ/Ϫ Mice-We performed blot analysis to determine if C/EBP␤ deletion affected the levels of insulin signaling proteins in skeletal muscle and adipose tissue. There was no change in GLUT4 or the levels of the insulin receptor (␤ subunit) and the p85␣ subunit of PI3K in skeletal muscle. However, the levels of IRS-1 were increased 2-fold in C/EBP␤ Ϫ/Ϫ mice compared with WT mice (p Ͻ 0.01) (Fig. 6). The levels of IRS-2 were not different (data not shown). In adipose tissue, the levels of the insulin receptor and p85␣ protein were slightly lower by 25-40% in C/EBP␤ Ϫ/Ϫ mice compared with wild-type mice (p Ͻ 0.05); however, the levels of IRS-1 and GLUT4 were not affected. The increased IRS-1 in skeletal muscle from C/EBP␤ Ϫ/Ϫ mice was consistent with greater insulin-stimulated activation of p85␣ and Akt phosphorylation and glucose transport activity.

DISCUSSION
The physiological importance of C/EBP␤ to whole-body glucose disposal and to insulin regulation of metabolism in vitro has not been examined previously. Despite a significant decrease in basal hepatic glucose production and less adiposity, the results of this study demonstrate that C/EBP␤ deletion does not significantly alter the normal metabolic and gene regulatory response to insulin in liver and adipose tissue. In contrast, we found that deletion of C/EBP␤ is associated with enhanced insulin action in skeletal muscle. These findings suggest a mechanism whereby an altered metabolic environment produced by C/EBP␤ deletion can affect whole-body in-FIG. 2. Rates of glucose disposal and HGP during the basal period and during two-step hyperinsulinemic-euglycemic clamp studies in wild-type (closed bars) and C/EBP␤ ؊/؊ (shaded bars) conscious mice. Mice were chronically catheterized via the jugular vein, and glucose turnover was measured with [3-3 H]glucose in the basal state (time 0), followed by a two-step hyperinsulinemic-euglycemic infusion to determine insulin sensitivity and suppression of hepatic glucose production. Insulin was infused at a rate of 5 milliunits/ kg/min for up to 90 min, followed by 20 milliunits/kg/min for a total of 1.5-3 h, depending on the ability to achieve a steady-state glucose concentration. Blood glucose was determined every 10 min, and the rate of unlabeled glucose infusion (20% glucose) was adjusted to maintain euglycemia. Steady state-specific glucose radioactivity and blood glucose concentrations were achieved during the last 30 min of each step of the clamp and used to calculate glucose disposal rates and rates of hepatic glucose production. Glucose was maintained between 90 and 110 mg/dl. The rates of glucose disposal and HGP were calculated as described under "Materials and Methods." Data are means Ϯ S.E. from 8 to 10 mice/group. *, p Ͻ 0.05 versus wild-type mice; **, p Ͻ 0.01 versus wild-type mice. Ra, rate of glucose appearance; Rd, rate of glucose disposal.
sulin sensitivity associated with changes in the insulin signaling cascade, specifically in skeletal muscle.
This study demonstrates that mice with a C/EBP␤ deletion have a marked increase in whole-body insulin-stimulated glucose disposal (Fig. 2), consistent with accelerated glucose metabolism in skeletal muscle. In vitro, skeletal muscle from C/EBP␤ Ϫ/Ϫ mice has greater insulin-stimulated glucose transport activity (Fig. 4) and increased IRS-1 protein levels (Fig. 6). Furthermore, in the absence of C/EBP␤, the ability of insulin to stimulate insulin receptor tyrosine phosphorylation, PI3K activity, and Akt serine phosphorylation is increased in skeletal muscle (Fig. 5). The serine/threonine kinases Akt-1 and Akt-2 are phosphorylated and activated by phosphatidylinositol 3-kinases and mediate cellular events such as cell growth, protein synthesis, and regulation of glycogen metabolism by insulin (26 -28, 38). Overexpression of constitutively active Akt stimulates glucose uptake and GLUT4 translocation in the absence of insulin in 3T3L1 cells or in primary adipocytes (27,39,40); however, the detailed mechanism of this signaling pathway downstream from Akt remains to be defined. The present results showing increased insulin-stimulated Akt phosphorylation and glucose transport activity with no change in GLUT4 protein are consistent with increased insulin signal transduction and suggest that either glucose transporter translocation and/or activation is enhanced in skeletal muscle from C/EBP␤ Ϫ/Ϫ mice.
A major effect of C/EBP␤ deletion on increased insulin-stimulated glucose transport activity may be related to lower circulating FFA levels. In fasted C/EBP␤ Ϫ/Ϫ mice, plasma insulin and FFA levels were lower prior to hyperinsulinemic-euglycemic clamp (Table I), and FFA remained lower during the first 100 min of insulin infusion (Table II). It is known that increases in circulating FFAs as well as other fat-derived factors such as tumor necrosis factor-␣ reduce insulin-stimulated glucose disposal as well as impair skeletal muscle insulin signal transduction through PI3K (41,42). Although we did not specifically address whether decreased FFA from adipose tissue increased glucose disposal, there is substantial evidence to suggest that skeletal muscle glucose utilization is reciprocally related to fatty acid oxidation (43). During the transition from fasting to feeding, there is a substrate competition between plasma FFA and glucose levels for meeting energy demands in skeletal muscle. Since cAMP levels are lower in liver and adipose tissue from fasting C/EBP␤ Ϫ/Ϫ mice (12), it is possible that C/EBP␤ deletion may have decreased FFA availability and oxidation, which, in turn, could affect skeletal muscle glucose utilization via the classic Randle cycle effect (43). We also hypothesize that chronically lower plasma FFA levels may have altered insulin signal transduction through insulin receptor, PI3K, and Akt phosphorylation (Fig. 6), contributing to greater glucose disposal in C/EBP␤ Ϫ/Ϫ mice. The mechanism for this effect is under investigation.
We also found an increase in IRS-1 protein levels in skeletal muscle. A potential C/EBP-binding site has been identified on the IRS-1 gene promoter (16), suggesting that IRS-1 mRNA levels may vary due to regulation by C/EBPs. However, the levels of IRS-1 mRNA were not significantly different in muscle from C/EBP␤ Ϫ/Ϫ mice (data not shown). This suggests that the mechanism for C/EBP␤ deletion on IRS-1 protein may be due to an alteration in protein turnover rather than an effect on IRS-1 gene transcription. Previous studies have shown a tissue-spe- FIG. 4. Increased insulin-stimulated glucose transport activity in isolated soleus muscle from C/EBP␤ ؊/؊ mice. 3-O-Methylglucose uptake was measured in the absence or presence of 100 nM insulin as described under "Materials and Methods." Data are means Ϯ S.E. *, p Ͻ 0.05 versus wild-type mice (n ϭ 9 -12 muscles/group).
FIG. 3. Altered mRNA expression levels of hepatic glucose-6-phosphatase, PEPCK, glucokinase, and GLUT2 in response to the hyperinsulinemic-euglycemic clamp conditions in wild-type (shaded bars) and C/EBP␤ ؊/؊ (closed bars) mice. Total cellular RNA was isolated from the livers of mice under overnight fasting conditions (Insulin (Ϫ)) or following 4 h of insulin-clamped conditions (Insulin (ϩ)). Levels of glucose-6-phosphatase (G6-Pase), PEPCK, glucokinase, GLUT2, and ribosomal mRNA were detected as described under "Materials and Methods." The relative levels of mRNA were determined by densitometry and are expressed as a percentage of mRNA hybridization (arbitrary units) from wild-type (Insulin (Ϫ)) livers detected on the same Northern blot. All values were corrected for ribosomal mRNA to account for loading differences. Results are means Ϯ S.E. from four to six animals/group. *, p Ͻ 0.05 versus wild-type mice (Insulin (Ϫ)); #, p Ͻ 0.05 versus C/EBP␤ Ϫ/Ϫ mice (Insulin (Ϫ)).
cific increase in IRS-1 protein in animal models of hypoinsulinemic states such as during fasting and streptozotocin-induced diabetes (45) and a decrease in IRS-1 protein in hyperinsulinemic obese animals (31,46). Increased IRS-1 levels in skeletal muscle from C/EBP␤ Ϫ/Ϫ mice could be related to their lower insulin levels associated with fasting hypoglycemia ( Table I). The factors responsible for these alterations remain obscure since relatively little is known about the expression of genes that regulate IRS-1 protein turnover.
We also examined the tissue-specific effects of C/EBP␤ deletion on insulin sensitivity in C/EBP␤ Ϫ/Ϫ mice. Despite a slight decrease in insulin receptor and p85␣ protein concentrations, the ability of insulin to stimulate Akt serine phosphorylation was unaffected in adipose tissue from C/EBP␤ Ϫ/Ϫ mice. In addition, the rate of suppression of lipolysis, a highly sensitive indicator of insulin responsiveness in adipose tissue, was similar in WT and C/EBP␤ Ϫ/Ϫ mice (Table II). The levels of C/EBP␣, PPAR␥, GLUT4, LPL, and IRS-1 expression were all normal in C/EBP␤ Ϫ/Ϫ mice, suggesting that regulation of adipocyte insulin sensitivity was likely unaltered in the absence of C/EBP␤. These data are consistent with recent reports that C/EBP␣ expression promotes terminal adipocyte differentiation and is responsible for full insulin sensitivity to glucose transport in mouse NIH-3T3L1 adipocytes (47,48). An intriguing finding is that adipocytes develop in the C/EBP␤ Ϫ/Ϫ mouse, albeit in reduced numbers, despite the absence of C/EBP␤, a key initiator of adipogenesis (10). This suggests that C/EBP␣ and PPAR␥ expression in vivo may be induced by other, as yet unknown, transcription factors. Recent evidence suggests that in 3T3L1 adipocytes, the expression of the transcription factor FIG. 6. Representative immunoblots and quantification of insulin signaling proteins in skeletal muscle and adipose tissue from wild-type (shaded bars) and C/EBP␤ ؊/؊ (open bars) mice. Mice were killed after 6 h of fasting. The gastrocnemius muscle and gonadal fat pat were used for analyses. 10 -40 g of protein was electrophoresed, blotted, and probed with specific antisera. Results are expressed as arbitrary units of a reference standard for comparisons between blots. Data are means Ϯ S.E. *, p Ͻ 0.05 between WT and C/EBP␤ Ϫ/Ϫ mice (n ϭ eight samples/group). A, muscle tissue; B, adipose tissue. IR, insulin receptor.

FIG. 5. Effect of insulin signal transduction on skeletal muscle from wildtype (closed bars) and C/EBP␤ ؊/؊ (open bars) mice in vivo.
Mice were fasted overnight, anesthetized, and injected with insulin (1 milliunit/kg of body weight). Muscle samples were removed prior to insulin injection and 5 min later and were analyzed for phosphorylation by immunoprecipitation with an anti-phosphotyrosine antibody, followed by insulin receptor (IR␤) immunoblotting. IRS-1-associated PI3K (PI-3 kinase) activity was determined in muscle protein extracts after immunoprecipitation with anti-IRS-1 antibodies. The levels of total and Ser 473phosphorylated Akt were by determined Western blot analysis using specific antibodies as described under "Materials and Methods." Results are expressed as arbitrary units above basal levels prior to insulin injection. Data are means Ϯ S.E. *, p Ͻ 0.05 between WT and C/EBP␤ Ϫ/Ϫ mice (n ϭ four to six samples/group). CREB precedes the expression of C/EBPs and can initiate adipogenesis (49). CREB is expressed at normal levels in adipose tissues from C/EBP␤ Ϫ/Ϫ mice, 2 suggesting that CREB could be an important initiator of adipose differentiation in mice lacking C/EBP␤ in vivo. However, the adipocytes that develop in C/EBP␤ Ϫ/Ϫ mice are not totally normal; they release less fatty acids in response to catecholamines, consistent with decreased levels of cAMP (12). The decreased fasting plasma FFA in C/EBP␤ Ϫ/Ϫ mice may be due to a reduced size of the overall adipose tissue mass. However, even accounting for reduced cell number, there is a less ␤-adrenergic stimulation of FFA release (12).
We also examined the ability of insulin to inhibit hepatic glucose production and liver gene expression. Hyperinsulinemia inhibited PEPCK and other insulin-regulated genes in liver in the absence of C/EBP␤ (Fig. 2). These findings are consistent with normal suppression of HGP during the hyperinsulinemic-euglycemic clamp procedure in C/EBP␤ Ϫ/Ϫ mice. Our data suggest that C/EBP␤ is not required for down-regulation of PEPCK gene expression by insulin or perhaps that other C/EBP isoforms can functionally replace C/EBP␤ in liver. Previously, it had been shown that C/EBP␤ is phosphorylated in liver and binds in vitro to the insulin response sequence as well as to the cAMP response element and P3I sites in the PEPCK promoter, sites responsible for glucocorticoid and cAMP activation (50 -52). Although insulin can fully inhibit PEPCK gene expression in the absence of C/EBP␤, mice with a C/EBP␤ deletion have reduced glucagon-stimulated cAMP production (12) and lower PEPCK expression in response to diabetes, consistent with decreased glucocorticoid activation (13). The mechanism(s) responsible for decreased cAMP levels appear to involve an increase in phosphodiesterase 3B expression in livers from C/EBP␤ Ϫ/Ϫ mice. 3 In summary, experiments in vivo and in vitro indicate that C/EBP␤ plays a highly integrative role in the regulation of whole-body carbohydrate and lipid metabolism. Our previous studies indicated that C/EBP␤ regulates hepatic glucose production and FFA levels during fasting and diabetes (12,13). The molecular mechanism for reduced HGP and FFA levels in the absence of C/EBP␤ is currently unknown, but may be due to a reduced response to counter-regulatory hormones (12,13). The present data indicate that C/EBP␤ deletion does not alter insulin signaling in liver and adipose tissue. However, the fact that deleting C/EBP␤ can alter metabolic signals (insulin and FFA) that control insulin sensitivity in skeletal muscle suggests that inhibiting C/EBP␤ in liver or adipose tissue might be an important therapeutic tool in the treatment of obesity and diabetes. Muscle insulin resistance as a consequence of metabolic fuel derangements is an established phenomenon in animal and human models of insulin resistance and diabetes. In tissues from obese humans, there is a significant decrease in insulin-stimulated tyrosine phosphorylation of IRS-1 and glucose transport activity, suggesting that this may be a major factor in their post-receptor insulin resistance (44). We are currently in the process of rescuing C/EBP␤ Ϫ/Ϫ mice with tissue-specific genes to investigate the role of C/EBP␤ in interorgan relationships that regulate insulin sensitivity, hepatic glucose production, and whole-body energy metabolism.