0888-8809/04/$15.00/0 Printed in U.S.A.

Molecular Endocrinology 18(11):2751–2764 Copyright © 2004 by The Endocrine Society doi: 10.1210/me.2004-0213

Hepatic CCAAT/Enhancer Binding Protein ␣ Mediates Induction of Lipogenesis and Regulation of Glucose Homeostasis in Leptin-Deficient Mice KIMIHIKO MATSUSUE, OKSANA GAVRILOVA, GILLES LAMBERT, H. BRYAN BREWER, JR., JERROLD M. WARD, YUSUKE INOUE, DEREK LEROITH, AND FRANK J. GONZALEZ Laboratory of Metabolism, National Cancer Institute (K.M., Y.I., F.J.G.), and Comparative Medicine Branch, National Institutes of Allergy and Infectious Diseases (J.M.W.), Molecular Disease Branch, National Heart, Lung, and Blood Institute (H.B.B.) and Diabetes Branch, National Institute of Diabetes and Digestive and Kidney Diseases (D.L.), National Institutes of Health, Bethesda, Maryland 20892; and Institut National de la Sante´ et de la Recherche Me´dicale, Unite´ 539 (G.L.), Hotel Dieu, 44000 Nantes, France CCAAT/enhancer binding protein ␣ (C/EBP␣) is a critical factor in glucose metabolism in the neonate as revealed by conventional C/EBP␣-null mice that do not survive beyond the first day after birth because of severe hypoglycemia and a deficiency in hepatic glycogen accumulation. To elucidate the function of C/EBP␣ in leptin-deficient mouse (ob/ ob) liver, a C/EBP␣-liver null mouse on an ob/ob background (ob/ob-C/EBP␣/Creⴙ) was produced using a floxed C/EBP␣ allele and Cre recombinase under control of the albumin promoter (AlbCre). The C/EBP␣-deficient liver in ob/ob mice had significantly decreased triglyceride content compared with equivalent mice lacking the AlbCre transgene (ob/ob-C/EBP␣/Creⴚ). Expression of genes involved in lipogenesis including fatty acid synthase, acetyl-coenzyme A carboxylase, stearoyl-

coenzyme A desaturase 1 and ATP-citrate lyase dramatically decreased in ob/ob-C/EBP␣/Creⴙ mouse liver. Induction of these lipogenic genes by a high-carbohydrate diet caused an exacerbation in the development of fatty liver and an increase in liver size, hepatic triglyceride, and cholesterol contents in ob/ob-C/EBP␣/Creⴚ mice but not in ob/obC/EBP␣/Creⴙ mice. Deficiency in hepatic C/EBP␣ expression caused an exacerbation of hyperglycemia because of decreased insulin secretion. Taken together, these results indicate that hepatic C/EBP␣ plays a critical role in the acceleration of lipogenesis in ob/ob mice and in glucose homeostasis by the indirect regulation of insulin secretion. (Molecular Endocrinology 18: 2751–2764, 2004)

CAAT/ENHANCER BINDING PROTEIN (C/EBP␣) belongs to the basic leucine zipper class of transcription factors (1). All members of the C/EBP family have a C-terminal basic leucine zipper domain that is responsible for DNA binding and dimerization (2, 3). C/EBP␣ is expressed in adipose tissue, liver, intestine,

lung, adrenal gland, peripheral-blood mononuclear cells, and placenta (4, 5). In liver and adipose tissue, the highest levels of C/EBP␣ mRNA are present in terminally differentiated cells (4, 6). The developmental and physiological roles of C/EBP␣ have been investigated with the C/EBP␣-null mouse (7, 8). These mice did not survive beyond the first day after birth because of severe hypoglycemia and a deficiency of hepatic glycogen and gluconeogenesis (7, 8). Furthermore, analysis of homozygous newborn mice showed that hepatocytes and adipocytes failed to accumulate lipid and had a defect in control of hepatocyte growth accompanied by a marked induction of the c-myc and c-jun genes, and altered lung development with unusual development of airways. These results suggest the involvement of C/EBP␣ in energy homeostasis during development. To study the role of C/EBP␣ in energy metabolism in liver at later stages of postnatal development, a conditional knockout allele of c/ebp␣ including loxp site was generated (9). C/EBP␣ expression in the mice was deleted specifically in liver by infusion of a recombinant adenovirus carrying the cre gene and resulting in a more than 90% recombination and loss of C/EBP␣

C

Abbreviations: ACC, Acetyl-CoA carboxylase; AlbCre, albumin-Cre transgene; Apo, apolipoprotein; C/EBP␣, CCAAT/ enhancer binding protein ␣; ChREBP, carbohydrate-responsive element-binding protein; CoA, coenzyme A; 7DCR, 7-dehydrocholesterol reductase; FAS, fatty acid synthase; FFA, free fatty acid; FPP, farnesyl diphosphate synthetase; FXR, farnesoid X receptor; G6Pase, glucose-6-phosphatase; GTT, glucose tolerance test; HC, high-carbohydrate; HDL, high-density lipoprotein; HMG, 3-hydroxy-3-methylglutaryl; LDL, low-density lipoprotein; LXR, liver X receptor; NF-Y, nuclear factor Y; OB/OB, normal leptin mouse; ob/ob, leptin mutated mouse; PEPCK, phosphoenolpyruvate carboxykinase; PGD, phosphogluconate dehydrogenase; SHP, small heterodimer partner; SREBP, sterol regulatory elementbinding protein; TC, total cholesterol; TG, triglyceride; VLDL, very LDL. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

2751 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2752 Mol Endocrinol, November 2004, 18(11):2751–2764

expression. These mice showed a decrease in phosphoenolpyruvate carboxykinase and glycogen synthase mRNA and a diabetic phenotype suggesting that C/EBP␣ also has an important role in glycogenesis and gluconeogenesis in mature adult liver thus indicating a role for this factor in control of energy homeostasis Type 2 diabetes has become the most common metabolic disorder in the developed world. Type 2 diabetes results in an abnormal energy balance associated with severe insulin resistance including hyperglycemia and hyperinsulinemia. There is considerable interest in the biochemical or molecular mechanisms that contribute to the development and progression of type 2 diabetes. Obesity is a major component of the metabolic syndrome that includes hyperlipidemia and hypertension and commonly precedes the development of type 2 diabetes in genetically predisposed individuals. C/EBP␤ was found to be involved in hyperglycemia in streptozotocin-induced type I diabetes model by regulating gluconeogenesis (10, 11). In the present study, to elucidate role of C/EBP␣ in metabolic syndrome of type 2 diabetes, a liver-specific C/EBP␣-null mouse on an ob/ob background was generated. These mice exhibited an improvement in fatty liver by impaired induction of lipogenic gene expression. Surprisingly, they also had lower insulin levels and altered glucose homeostasis.

RESULTS A Liver-Specific C/EBP␣-Null Mouse on an ob/ob Background The conditional floxed allele of the C/EBP␣ gene consists of two loxP sites flanking the entire C/EBP␣ coding region (9). The loss of C/EBP␣ mRNA was found only in ob/ob-C/EBP␣/Cre⫹ liver and not in brain, fat and lung, other sites of C/EBP␣ expression (Fig. 1A). The C/EBP␣ mRNA can give rise to two polypeptides of 42 and 30 kDa by alternative use of translation initiation codons from a single mRNA (12). In support of the mRNA analysis, both proteins were completely lost in the ob/ob-C/EBP␣/Cre⫹ liver (Fig. 1B). The C/EBP␣ mRNA and proteins in OB/OB-C/EBP␣/Cre⫹ mice were also lost in liver (data not shown). In contrast to C/EBP␣, C/EBP␤ mRNA and protein were unchanged in ob/ob-C/EBP␣/Cre⫹ liver. To assess the potential effects of liver-specific C/EBP␣ deficiency, body and tissue weights were measured. For a period of 12 wk after birth, no significant difference in body, white adipose and liver weight was observed between ob/ob-C/EBP␣/Cre⫺ and Cre⫹ (Table 1). Contrary to the result of conventional C/EBP␣-null mice described in earlier reports (7, 8), these results demonstrate that deficiency of adult hepatic C/EBP␣ is not lethal.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Fig. 1. Liver-Specific Disruption of C/EBP␣ mRNA and Protein in ob/ob Mice A, Northern blot of RNA from each tissue of ob/ob-C/ EBP␣/Cre⫹ or Cre⫺ mice. RNA (10 ␮g) was denatured and electrophoresed in formaldehyde-containing 1% agarose gels, blotted to nylon membranes and hybridized to the indicated 32P-labeled probes. B, Western blot of nuclear extracts from liver of ob/ob-C/EBP␣/Cre⫹ or Cre⫺ mice. Total liver nuclear extract protein (20 ␮g) were subjected to electrophoresis on a 10% acrylamide gel, transferred to Immobilon-P membranes and stained with antibody to C/EBP␣ and ␤.

Altered Expression of Lipogenic and Apolipoprotein (Apo) A4 Genes in Liver-Specific C/EBP␣-Null Mice Histologically, hepatocytes of ob/ob-C/EBP␣/Cre⫹ mice had considerably smaller and fewer cytoplasmic vacuoles than those in ob/ob-C/EBP␣/Cre⫺ (data not shown). The hepatic triglyceride (TG) content of ob/ ob-C/EBP␣/Cre⫹ mice was significantly lower (60% of ob/ob-C/EBP␣/Cre⫺) than that in ob/ob-C/EBP␣/ Cre⫺ mice (Fig. 2B-a), although total cholesterol (TC) content was not different between Cre⫹ and Cre⫺ mice (Fig. 2B-b). Earlier studies showed that the C/EBP␣-null mouse had markedly decreased hepatic glycogen content (7, 8). However, under our conditions, glycogen content in ob/ob-C/EBP␣/Cre⫹ mice was higher than that found in ob/ob-C/EBP␣/Cre⫺ mice (Fig. 2B-c). An earlier study revealed that expression of bilirubin uridinediphosphate-glucuronosyltransferase,glucose6-phosphatase (G6Pase), glycogen synthase and phosphoenolpyruvate carboxykinase (PEPCK) were influenced by loss of hepatic C/EBP␣ (7). Surprisingly, G6Pase and PEPCK mRNA levels in liver were not different between OB/OB and ob/ob-C/EBP␣/Cre⫺ and Cre⫹ mice (Fig. 2A). However, in contrast to genes involved in gluconeogenesis, the expression of ApoA4 was dramatically decreased in the absence of C/EBP␣ in both OB/OB and ob/ob backgrounds (Fig. 2A). The different levels of expression between ob/ob-C/EBP␣/

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Mol Endocrinol, November 2004, 18(11):2751–2764 2753

Table 1. Tissue Weight and Blood Parameters in Hepatic C/EBP␣-Deficient ob/ob Mice OB/OB Cre

Age (wk) Body weight (g) Liver weight (g) Liver (%) Epididymal fat (%) Inguinal fat (%) Serum FFA (mM) Serum TG (mg/dl) Serum TC (mg/dl) Glucose (mg/dl)b Glucose (mg/dl)c

ob/ob

⫺

⫹

⫺

⫹

10.9 ⫾ 0.5 28.1 ⫾ 1.4 (9) 1.0 ⫾ 0.1 (9) 3.9 ⫾ 0.2 (9) 1.5 ⫾ 0.2 (9) 0.53 ⫾ 0.0 (9) 0.76 ⫾ 0.1 (6) 43.0 ⫾ 5.8 (6) 241 ⫾ 20 (6) 153 ⫾ 7.8 (24) 137 ⫾ 4.5 (20)

11.9 ⫾ 0.3 30.1 ⫾ 1.0 (5) 1.2 ⫾ 0.1 (5) 4.1 ⫾ 0.1 (5) 2.1 ⫾ 0.1 (5) 0.65 ⫾ 0.1 (5) 0.75 ⫾ 0.1 (6) 28.1 ⫾ 4.0 (6)a 215 ⫾ 15 (6) 154 ⫾ 5.5 (25) 147 ⫾ 5.3 (18)

11.0 ⫾ 0.5 52.1 ⫾ 1.1 (12) 2.9 ⫾ 0.1 (12) 5.7 ⫾ 0.3 (12) 5.1 ⫾ 0.4 (12) 1.3 ⫾ 0.1 (12) 1.3 ⫾ 0.0 (12) 84.3 ⫾ 11 (12) 559 ⫾ 44 (6) 330 ⫾ 26 (27) 304 ⫾ 24 (30)

11.0 ⫾ 0.4 50.4 ⫾ 1.0 (11) 3.1 ⫾ 0.2 (11) 6.2 ⫾ 0.3 (11) 6.1 ⫾ 0.5 (11) 1.4 ⫾ 0.1 (11) 1.7 ⫾ 0.3 (11)a 128 ⫾ 23 (11) 598 ⫾ 39 (6) 408 ⫾ 27 (23)a 392 ⫾ 30 (28)a

Values are mean ⫾ SE. Numbers of animals are indicated in parentheses. Animals were killed in nonfasting condition. a , P ⬍ 0.05, for difference between Cre⫺ and ⫹. b , 6 wk old; c, 12 wk old.

Cre⫺ and Cre⫹ mice were also observed in typical lipogenic genes, fatty acid synthase (FAS), acetyl-CoA carboxylase (ACC), S14, ATP-citrate lyase, malic enzyme, glycerol-3-phosphate acyltransferase, and stearoyl-CoA desaturase 1 genes. The ob/ob leptin deficiency results in elevated expression of lipogenic genes in liver (13). However, these genes were clearly expressed at lower levels in ob/ob-C/EBP␣/Cre⫹ mice; the levels were almost as low as that found in the wild-type OB/OB mice (Fig. 2A). Expression of sterol regulatory element-binding protein 1 (SREBP1), an essential factor for regulation of these genes (14), was also lower in ob/ob-C/EBP␣/Cre⫹ liver but not in OB/ OB-C/EBP␣/Cre⫹ (Fig. 2A). These results suggest that C/EBP␣ influences lipid accumulation in ob/ob mice liver by mediating the regulation of expression of the lipogenic genes. The Liver-Specific C/EBP␣-Null Mouse Is Resistant to Acceleration of Fatty Liver by HighCarbohydrate (HC) Diet HC diet is a robust method of lipogenic gene induction (15). Because the expression of lipogenic genes in ob/ob-C/EBP␣/Cre⫹ mice liver was dramatically decreased (Fig. 2A), a HC diet was used to examine the relationship between hepatic C/EBP␣ and lipogenesis. Livers in the ob/ob-C/EBP␣/Cre⫺ mice fed a HC diet for 12 d were significantly enlarged relative to those of Cre⫹ mice and were yellowish in appearance, typical of fatty liver (Fig. 3, A and F). However, ob/ob-C/EBP␣/ Cre⫹ mice exhibited a dramatically improved fatty liver (Fig. 3, A and F). Histological analysis of liver revealed the presence of numerous and large intracytoplasmic vacuoles in hepatocytes from the ob/ob-C/EBP␣Cre⫺ mice (Fig. 3B-1), whereas hepatocytes in the ob/obC/EBP␣/Cre⫹ liver (Fig. 3B-2) were much smaller and less numerous than those seen in ob/ob-C/EBP␣/ Cre⫺. These vacuoles in ob/ob-C/EBP␣/Cre⫺ were positive for the presence of lipid as revealed by Oil Red

O staining (Fig. 3B-5), but staining in ob/ob-C/EBP␣/ Cre⫹ liver was clearly less intense (Fig. 3B-6). Interestingly, a significant difference in weight and weight gained after feeding HC for 12 d was observed between ob/ob-C/EBP␣/Cre⫺ and Cre⫹ (Fig. 3, C and D). The hepatic TG and TC contents of ob/ob-C/EBP␣/ Cre⫹ were also significantly lower (41% TG and 47% TC of ob/ob-C/EBP␣/Cre⫺) than that for ob/ob-C/ EBP␣/Cre⫺ (Fig. 3, G and H). These results strongly suggest that hepatic C/EBP␣ positively regulates lipogenesis in the ob/ob liver. Hepatic C/EBP␣ Is Involved in Induction of Lipogenic Genes in ob/ob Mice Because deficiency of hepatic C/EBP␣ results in a marked decrease in the expression of lipogenic genes in ob/ob mice but not in OB/OB mice, the possibility exists that hepatic C/EBP␣ may be involved in the induction of lipogenic genes under leptin deficiency. To examine this possibility and to determine mechanism of fatty liver aggravated after HC feeding, mRNA levels of genes involved in hepatic lipogenesis were examined by HC feeding (Fig. 4A). All of the lipogenic genes in livers of OB/OB and ob/ob-C/EBP␣/Cre⫺ mice were dramatically induced by HC feeding, as compared with a normal diet, whereas the induction in ob/ob-C/EBP␣/Cre⫹ mice was clearly impaired. The induction in OB/OB-C/EBP␣/Cre⫹ mice was unchanged or slightly impaired as compared with OB/OB-C/EBP␣/Cre⫺ mice, except for FAS gene. SREBP1 induction was slightly impaired in ob/ob-C/ EBP␣/Cre⫹ mice, but other transcription factors showed no change (Fig. 4, A and B). In ob/ob-C/EBP␣/Cre⫹ mice fed a HC diet, a significant decrease in hepatic TC content was found (Fig. 3H) that did not occur with a normal diet (Fig. 2B-b). Therefore, genes involved in TC synthesis or excretion pathways were examined (Fig. 4B). Induction of farnesyl diphosphate synthase (FPP) and 7-dehydrocholesterol reductase (7DCR) by HC feeding was strikingly impaired in ob/ob-C/EBP␣/Cre⫹ mice,

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2754 Mol Endocrinol, November 2004, 18(11):2751–2764

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Fig. 2. Effect of Hepatic C/EBP␣ Deficiency in ob/ob Mice Fed a Normal Diet on Gene Expression Patterns A, Northern blot of RNA isolated from OB/OB or ob/ob mouse liver. Total RNA was isolated from nonfasting male mice and subjected to electrophoresis as described in Fig. 1. Quantitation of the hybridization signals was performed using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and are expressed as the fold change, after correction for 36B4 levels, relative to OB/OB-C/EBP␣/Cre⫺ mice. Values are the average obtained from two animals. B, Measurement of hepatic triglyceride (a), cholesterol (b), and glycogen (c) contents. To prevent effect of foods, total lipids on TG and cholesterol (TC) was extracted from 24 h fasting mouse liver. Glycogen (Gly) was measured from nonfasting mouse liver. The mouse number for each genotype is as follows: TG and TC assay—OB/OB-C/EBP␣/Cre⫺, seven (four males, three females); OB/OB-C/EBP␣/Cre⫹, seven (four males, three females); ob/ob-C/EBP␣/Cre⫺, nine (five males, four females); ob/ob-C/EBP␣/Cre⫹, nine (six males, three females); Gly assay—OB/OB-C/EBP␣/Cre⫺, nine (six males, three females); OB/OB-C/EBP␣/Cre⫹, five (three males, two females); ob/ob-C/ EBP␣/Cre⫺, 12 (9 males, three females); ob/ob-C/EBP␣/Cre⫹, 10 (six males, four females). Data are mean ⫾ SE. Significant differences compared with Cre ⫺ mice: *, P ⬍ 0.05. PPAR␥, Peroxisome proliferator-activated receptor ␥; MAL, malic enzyme; SCD1, stearoyl-CoA desaturase 1; ACL, ATP-citrate lyase; GPAT, glycerol-3-phosphate acyltransferase.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Mol Endocrinol, November 2004, 18(11):2751–2764 2755

Fig. 3. Effect of Feeding HC Diet to C/EBP␣ Deficiency on Hepatic Lipids Mice from each genotype were fed by HC diets for 12 d. A, Liver was obtained from ob/ob-C/EBP␣/Cre⫺ and Cre⫹ after 12 d of HC feeding. Note, Large and more yellow liver of ob/ob-C/EBP␣/Cre⫺ mice, compared with Cre⫹ mice. B, Histology of livers from each mouse genotype after 12 d of HC feeding. Hematoxylin and eosin staining of liver sections (original magnification, ⫻100) from ob/ob-C/EBP␣/Cre⫺ (1) and Cre⫹ (2). The ob/ob-C/EBP␣/Cre⫺ mouse has large and multi-vacuolated hepatocytes, whereas the Cre⫹ are have small and less vacuolated hepatocytes. Oil Red O stain shows liver sections (original magnification, ⫻300) from OB/OB-C/EBP␣/Cre⫺ (3) and Cre⫹ (4) or ob/ob-C/EBP␣/Cre⫺ (5) and Cre⫹ (6). Liver and fat weight and hepatic lipid content in HC-fed mice. Body weight before and after 12 d of HC feeding (C), epididymal fat weight (D), inguinal fat weight (E), liver weight (F), hepatic triglyceride (G) and cholesterol content (H). The mouse number for each genotype is as follows: OB/OB-C/EBP␣/Cre⫺, 15 (seven males, eight females); Cre⫹, 12 (six males, six females); ob/ob-C/EBP␣/Cre⫺, 25 (nine males, 16 females); Cre⫹, 11 (three males, eight females). Data are mean ⫾ SE. Significant differences compared with Cre ⫺ mice: *, P ⬍ 0.05; **, P ⬍ 0.01; ***, P ⬍ 0.001.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2756 Mol Endocrinol, November 2004, 18(11):2751–2764

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Fig. 4. Effect of Feeding HC Diet to C/EBP␣-Deficient ob/ob Mice on Hepatic Gene Expression Patterns Northern blot analysis was performed on RNA (10 ␮g) that was pooled from equal aliquots of total RNA derived from four mice in each group. Lipogenic genes in liver were induced by two methods as follows. A and B, Quantification of expression level of lipogenic genes induced by a HC diet. Mice were fed by a HC diet for 12 d. C, Quantification of expression levels of lipogenic genes induced by fasting-refeeding. Mice were fasted for 24 h to designate the fasting group and then refed with a HC for 24 h to designate the refeeding group. D, Western blot analysis of mature SREBP1 proteins by nuclear extracts from liver of OB/OBand ob/ob-C/EBP␣/Cre⫹ or Cre ⫺ mice (a). The conditions of Western blot analysis were described in Fig. 1. Nuclear proteins were extracted from livers used in Fig. 2. Northern blot analysis of genes regulated translocation of SREBP1 (b). The conditions of Northern blot analysis were described in Fig. 2. The membranes in Fig. 2 were used in this experiment after probe stripping. GPAT, Glycerol-3-phosphate acyltransferase; PPAR␣, peroxisome proliferator-activated receptor ␣; INSIG, insulin-induced gene; CYP, cytochrome P450; ABC, ATP-binding cassette.

whereas 3-hydroxy-3-methylglutaryl (HMG)-coenzyme A (CoA) synthase (Fig. 4B) and HMG-CoA reductase (data not shown) mRNA levels were not changed. No differ-

ence in the expression levels of ABCA1 and ABCG8 (data not shown) was observed between ob/ob-C/EBP␣/ Cre⫹ and Cre⫺ mice, whereas the expression of ABCG5

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

decreased in ob/ob-C/EBP␣/Cre⫹ mice fed a HC diet. Farnesoid X receptor (FXR) was specifically decreased on all Cre⫹ mice fed HC diets. Among potential FXR target genes, small heterodimer partner (SHP) was decreased but only in the ob/ob-C/EBP␣/Cre⫹ mice. It is known that SHP is regulated by FXR and represses the pathway of bile production from cholesterol by decreasing Cyp7a1 expression. Although SHP expression was decreased in ob/ob-C/EBP␣/Cre⫹ mice fed a HC diet, Cyp7a1 expression was already decreased in both OB/OB and ob/ob fed the HC diets indicating that this is because of diet alone and not genotype (Fig. 4B). Refeeding after fasting treatment also results in induction of lipogenic genes (15). To further examine the role of C/EBP␣ for in the induction of lipogenic genes a fastingrefeeding regimen was carried out. The mRNA levels of all lipogenic genes were suppressed in livers of fasted mice and were markedly elevated by refeeding (Fig. 4C). The fold induction of all lipogenic genes in ob/ob-C/ EBP␣/Cre⫺ mice by refeeding was higher than that of OB/OB-C/EBP␣/Cre⫺ mice. However, genes that were induced by refeeding showed lower induction in ob/obC/EBP␣/Cre⫹ mice. The fold induction in ob/ob-C/ EBP␣/Cre⫹ mice was almost same level with that of OB/OB-C/EBP␣/Cre⫺ mice, suggesting that hepatic C/EBP␣ is involved in signaling the induction of lipogenic genes in ob/ob mice. It is known that SREBP1 is critical for the induction of lipogenic genes by HC-feeding and fasting-refeeding (15). Indeed, SREBP1 mRNA in ob/ob-C/EBP␣/Cre⫹ mice was decreased under the normal dietary conditions (Fig. 2A), although it was only slightly decreased under the HC-feeding and fasting-refeeding (Fig. 4, A and C. Because SREBP1 proteins are known to translocate to the nucleus in a form competent to activate gene transcription, the levels of SREBP1-active form in the nucleus is a more reliable gauge of target gene induction than the mRNA levels. Interestingly, protein levels of SREBP1-active form in ob/ob-C/EBP␣/Cre⫹ mice were higher than that of ob/ob-C/EBP␣/Cre⫺ mice (Fig. 4D-a). These high levels in ob/ob-C/EBP␣/Cre⫹ mice is possibly because of the decrease in insulin-induced gene 1 (INSIG1) (16), a known regulator of nuclear translocation (Fig. 4D-b). These results indicate that the impaired induction of lipogenic genes in ob/ob-C/EBP␣/Cre⫹ mice is not because of decreased SREBP1 mRNA. Deficiency of Hepatic C/EBP␣ Leads to Lower Insulin and Cholesterol Levels in ob/ob Mouse Blood To assess the effects of deficiency of liver-specific C/EBP␣ on diabetic phenotypes, the level of glucose was measured under the nonfasting and normal dietary conditions (Table 1). Glucose levels in 6- and 12-wk-old ob/ob-C/EBP␣/Cre⫹ mice were significantly higher than that of ob/ob-C/EBP␣/Cre⫺ mice. Because the HC diet used in this study includes 37% sucrose, blood glucose levels during HC feeding was assessed by monitoring glucose levels every 2 d for

Mol Endocrinol, November 2004, 18(11):2751–2764 2757

12 d (Fig. 5A). Surprisingly, blood glucose levels in ob/ob-C/EBP␣/Cre⫹ were dramatically increased on the first day of HC feeding. To further characterize glucose metabolism, glucose tolerance tests (GTTs) were performed before and after HC feeding after an exogenous load of glucose (Fig. 5B). The glucose levels in ob/ob-C/EBP␣/Cre⫹ were significantly elevated at all time points compared with ob/ob-C/EBP␣/Cre⫺ (Fig. 5B-b). No significant difference before and after HC feeding was observed. Next, by using same mice, insulin levels were measured 0, 7, and 12 d after HC feeding (Fig. 5C). Insulin levels of ob/ob-C/EBP␣/Cre⫹ mice at all time points were significantly lower than that of ob/ob-C/EBP␣/ Cre⫺ (⬃50% of ob/ob-C/EBP␣/Cre⫺). Further, to asses insulin sensitivity in ob/ob-C/EBP␣/Cre⫹, insulin tolerance test is performed under the nonfasting and normal diet condition (Fig. 5D). The glucose levels in ob/ob-C/EBP␣/Cre⫹ after insulin injection is significantly higher than in the ob/ob-C/EBP␣/Cre⫺ mice at all time points, suggesting that the unusually high glucose levels found after HC diets and the GTT result from lower insulin levels and worsen insulin resistance in ob/ob-C/EBP␣/Cre⫹. The serum lipid contents of ob/ob-C/EBP␣/Cre⫺ and Cre⫹ are summarized in Table 1. All lipid classes, TG, free fatty acids (FFA), TC of ob/ob-C/EBP␣/Cre⫹ mice showed a tendency toward higher levels as compared with those of nonfasting ob/ob-C/EBP␣/Cre⫺ mice. Significant difference in serum lipids was observed only with FFA levels. Although HC feeding caused an elevation of serum TG levels in ob/ob-C/ EBP/Cre⫹ and Cre⫺ mice in a time-dependent manner, TG levels were not elevated in OB/OB-C/EBP/ Cre⫹ and Cre⫺ mice (Fig. 5E-a). Serum TC levels in ob/ob-C/EBP/Cre⫹ mice dramatically decreased from 7 d after HC-feeding as compared with ob/ob-C/EBP/ Cre⫺ mice (Fig. 5E-b) as reflected in hepatic TC levels (Fig. 3H). FFA levels in ob/ob-C/EBP␣/Cre⫹ mice at all time points showed a higher tendency than those of ob/ob-C/EBP␣/Cre⫺ mice although this difference at 7 and 12 d did not reach statistical significance (Fig. 5E-c). The difference in FFA levels was observed not only in ob/ob-C/EBP␣/Cre⫹ mice but also in OB/OBC/EBP␣/Cre⫹ mice. To elucidate mechanisms for increased serum FFA in ob/ob-C/EBP␣/Cre⫹ mice, lipase activities were measured under normal dietary conditions. However, no significant difference in the total and hepatic lipase activity was observed between ob/ob-C/EBP␣/Cre⫹ and Cre⫺ mice (Fig. 5G), consistent with similar amounts of plasma TG in both lines. We next characterized the TC levels decreased in ob/ob-C/EBP␣/Cre⫹ mice by FPLC analysis of pooled serum samples collected 7 d after the initiation of the HC challenge. The FPLC profile (Fig. 5F) of ob/obC/EBP␣/Cre⫹ mice shows a sharp decrease in plasma high-density lipoproteins (HDL) as well as low-density lipoproteins (LDL) TC levels, and a parallel decrease in the levels of the major HDL and LDL Apo (AI/AII, E, B100/B48, Fig. 5F, inset), compared

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2758 Mol Endocrinol, November 2004, 18(11):2751–2764

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

Fig. 5. Effect of Hepatic C/EBP␣ Deficiency on Diabetic Phenotypes A, Blood glucose concentrations during HC feeding. Blood glucose concentrations were monitored during HC-feeding. B, Glucose tolerance test. Glucose tolerance tests were performed on OB/OB (a) and ob/ob (b) groups before and after HC feeding. After a 6 h fast, the mice were injected with glucose (2 mg/g). C, Blood insulin concentrations during HC feeding. Blood insulin was measured in blood samples that were taken on 0, 7, and 12 d from OB/OB (a) and ob/ob (b) groups after HC feeding. D, Insulin tolerance test in ob/ob mice fed a normal diet. Insulin (2 U/kg) was injected to each mice under nonfasting condition. The mouse number is as follows: ob/ob-C/EBP␣/Cre⫺, nine (four males, five females); Cre⫹, eight (four males, four females). E, Blood lipid concentrations during HC-feeding. Plasma TGs (a), TC (b) and FFAs (c) were measured for blood samples that were taken on 0, 7, and 12 d after HC feeding. F, The lipoproteins from ob/ob-C/EBP␣/Cre⫹ and ob/ob-C/EBP␣/Cre⫺ mice fed HC for 7 d were separated from 60 ␮l of pooled (n ⫽ 4 for each group) plasma samples by FPLC. The concentration of cholesterol in each eluted fraction is indicated in the y-axis. Inset, Immunoblot analysis of Apo B, E, A-I, and A-II contained within the VLDL (V), LDL (L), HDL1

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

with ob/ob-C/EBP␣/Cre⫺ mice. In addition, ob/obC/EBP␣/Cre⫹ mice had slightly increased very LDL (VLDL) TC and associated ApoB48 vs. ob/ob-C/ EBP␣/Cre⫺ animals.

DISCUSSION Liver-Specific Disruption of C/EBP␣ in ob/ob Mice Mice lacking hepatic C/EBP␣ in the adult were viable, whereas conventional C/EBP␣-null mice do not survive beyond the first day after birth because of severe hypoglycemia by a deficiency of hepatic glycogen (7, 8). Furthermore, no significant difference in PEPCK, G6Pase was observed between Cre⫹ and Cre⫺ for both backgrounds (OB/OB and ob/ob). This result revealed that hepatic C/EBP␣ is critical for glyconeogenic or gluconeogenic pathways in only neonate and not in adult. In an earlier report, deficiency of C/EBP␣ induced by Cre introduced by acute adenovirus infection of adult C/EBP␣(fl/fl) mice resulted in a decrease in PEPCK and G6Pase mRNA similar to what was reported with conventional C/EBP␣-null mice (9). This suggests that acute or short loss of C/EBP␣ as produced by adenovirus-Cre infusion into adult liver leads to different phenotypes from mice made using the albumin-Cre. It should be noted that the albumin promoter is expressed in liver at low levels in fetal development (d 19) with promoter activity increasing gradually until adult levels are reached at 1–2 wk postnatally (17, 18). This could allow for compensation by other transcription factors that control the PEPCK and G6Pase genes. In this connection C/EBP␤, when expressed from the C/EBP␣ gene locus, can functionally replace C/EBP␣ in liver (19). Furthermore, C/EBP␤ has been shown to compensate for loss of C/EBP␣ in the regulation of PEPCK gene expression (20). Thus, compensation for loss of C/EBP␣ by C/EBP␤ for control of genes involved in glyconeogenic and gluconeogenic pathways in adult liver cannot be excluded (11) although C/EBP␤ mRNA and protein levels were not different in Cre⫹ and Cre⫺ mice. Hepatic C/EBP␣ Promotes Lipogenesis in ob/ob Mice Hepatic fatty acid synthesis from glucose is elevated by 4 wk of age in ob/ob mice. Further, hepatic activities of lipogenic genes in ob/ob mice are dramatically increased by 7 wk of age (21). In the present study,

Mol Endocrinol, November 2004, 18(11):2751–2764 2759

hepatic C/EBP␣ was found to have a role in the acceleration of lipogenesis in ob/ob mice. Hepatic TG content significantly decreased in ob/ob-C/EBP␣/ Cre⫹ mice. Furthermore, HC feeding, an inducer of lipogenic genes in ob/ob-C/EBP␣/Cre⫺ mice, caused an exacerbation in the development of fatty liver and an increase of liver size and hepatic TG content that were not observed in Cre⫹ mice. These results clearly indicate that hepatic C/EBP␣ promotes the development of fatty liver by mediating the induction of expression of lipogenic genes. The accumulation of hepatic TG is not only because of the acceleration of de novo lipogenesis. The uptake from dietary triglycerides or fatty acids to liver also contributes to the lipid accumulation in liver. However, the HC diet used in present study is a lipid-free diet. Therefore, decreased hepatic TG content in ob/ob-C/EBP␣/Cre⫹ mice by HC feeding is likely because of impairment of de novo lipogenesis. Interestingly, we also observed a marked decrease in hepatic TC content in ob/ob-C/EBP␣/Cre⫹ mice fed a HC diet. This is likely because of the levels of FPP and 7DCR that are inducible during HC feeding in ob/ob-C/EBP␣/Cre⫺ mouse liver, but not induced in ob/ob-C/EBP␣/Cre⫹ mouse liver. FPP is a key enzyme in the biosynthesis of sterols, ubiquinone, dolichol, and heme (22), whereas 7DCR catalyzes the reduction of the ⌬ (7)-double bond of sterol intermediates, which is the terminal reaction in the pathway of cholesterol biosynthesis from lanosterol (23). FPP and 7DCR are not known to be rate-limiting enzymes in cholesterol biosynthesis. However, mutations in the human 7DCR cause Smith-Lemli-Opitz Syndrome and among the symptoms in this syndrome is typical low serum and tissue cholesterol (24, 25). Indeed, it was also reported that 7-dehydrocholesterol, which accumulates from posttranslational inhibition of 7DCR, suppresses sterol biosynthesis by feedback inhibition of HMG-CoA reductase activity (26). Thus, these results suggest that even though FPP and 7DCR are not rate-limiting enzymes, the reduction in expression of these genes could result in decreased cholesterol levels in liver or blood both directly and indirectly. No difference in the expression levels of ABCA1 and ABCG8 (data not shown) was observed between ob/ ob-C/EBP␣/Cre⫹ and Cre⫺ mice. The decrease of ABCG5 expression in ob/ob-C/EBP␣/Cre⫹ mice fed a HC diet does not appear to be of sufficient magnitude to contribute to the decreased hepatic cholesterol levels. It is known that SHP expression is regulated by FXR and elevated SHP represses Cyp7a1 expression by forming a transcriptionally inert heterodimer with

(H1), and HDL2/3 (H2/3) top fractions. G, Post heparin serum lipase activity. Total (TL) and hepatic (HL) lipase activity in ob/ob mice fed a normal diet. The mouse number for each genotype is as follows: ob/ob-C/EBP␣/Cre⫺, six (three males, three females); Cre⫹, six (three males, three females). Experiments, A–C, E, and F were performed by using same mice. The mouse number for each genotype was described in Fig. 3C. Data are mean ⫾ SE. Significant differences compared with Cre ⫺ mice: *, P ⬍ 0.05; **, P ⬍ 0.01; ***, P ⬍ 0.001.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2760 Mol Endocrinol, November 2004, 18(11):2751–2764

LRH-1. Our results revealed that SHP expression that is elevated in the HC-fed ob/ob-C/EBP␣/Cre⫺ mice, was decreased in ob/ob-C/EBP␣/Cre⫹ mice. This decrease may be because of lower FXR expression in the ob/ob-C/EBP␣/Cre⫹ mice on the HC diet. However, Cyp7a1 expression was decreased in all mice fed a HC diet independent of SHP levels. The low Cyp7a1 mRNA may because of the HC diet used in this study that lacks fat. Therefore, these results indicate that the reduction of hepatic cholesterol levels in ob/ob-C/ EBP␣/Cre⫹ mice fed a HC diet is not because of increased cholesterol transport from liver or from the increased bile acid biosynthesis from cholesterol through increased Cyp7a1 expression. Liver is the main tissue for synthesis of TC (27). Therefore, the decrease in the expression of these genes involved in TC synthesis likely explains the decrease in hepatic TC levels (28). The decrease in hepatic TC further leads to a decrease in LDL and HDL TC levels in ob/ob-C/EBP␣/Cre⫹ mice. Hepatic C/EBP␣ Is Involved in Induction Pathway of Lipogenic Genes in ob/ob Mouse Liver Under a normal diet, the expression of lipogenic genes in ob/ob-C/EBP␣/Cre⫹ mice returned to the basal levels found in OB/OB-C/EBP␣/Cre⫺ mice. To induce the expression of these genes, two methods were used, HC feeding and fasting-refeeding. These treatments strongly induced the expression of lipogenic genes in the ob/ob-C/EBP␣/Cre⫺ mouse liver, but their induction in ob/ob-C/EBP␣/Cre⫹ mice was clearly impaired, suggesting that hepatic C/EBP␣ is involved in induction pathway of lipogenic genes in the ob/ob mouse liver. Interestingly, a larger contribution for C/EBP␣ was found in ob/ob mice as compared with OB/OB wild-type mice, suggesting that the possibility of the existence of diabetes- or deficient functional leptindependent signals that are mediated or potentiated by C/EBP␣. Furthermore, it is noteworthy that all genes having impaired induction by deficiency of C/EBP␣ are known SREBP1 target genes (29, 30). Thus, the deficiency of hepatic C/EBP␣ in ob/ob mice leads to impaired SREBP signaling. Several genes observed having different expression levels in between Cre⫺ and Cre⫹ contain typical C/EBP binding sites. By searching the gene database (MOTIF; http://motif.genome.ad.jp/, cutoff score; 85), C/EBP binding sites were revealed at position ⫺852 to ⫺840 and ⫺558 to ⫺ 546 bp of the ATP-citrate lyase gene and ⫺1498 to ⫺1486 and ⫺720 to ⫺707 bp of the GAPT gene, suggesting that C/EBP␣ could potentially directly regulate these genes. Whether C/EBP␣ regulates these genes by direct binding to cis-acting elements remains to be determined. However, it should be noted that constitutive expression of lipogenic genes were not different between wild-type OB/ OB-C/EBP␣/Cre⫹ and OB/OB-C/EBP␣/Cre⫺ mice, indicating that this transcription factor only affects inducible expression of lipogenic genes. In contrast,

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

the ApoA4 gene appears to be directly regulated by C/EBP␣ and indeed C/EBP binding sites were found 5⬘ upstream of the ApoA4 gene at position ⫺1653 to ⫺1641, ⫺1045 to ⫺1033 and ⫺1002 to ⫺989 bp. Studies using SREBP1-null mice demonstrated that SREBP1 is essential for the induction of lipogenic genes by HC feeding and fasting-refeeding cycles (15). Therefore, C/EBP␣ may have an indirect role in potentiating SREBP1 signaling. In this regard, it is noteworthy that C/EBPs and SREBPs have basic helix-loop-helix leucine zipper motifs and directly interact with other transcription factors (31–33). Thus, the possibility exists that an interaction occurs between C/EBP␣ and SREBP1c. However, coimmunoprecipitation assays were not able to reveal a direct physical interaction between these two factors (data not shown). It is also known that SREBPs are weak activators of transcription and function efficiently only in concert with other transcription factors such as nuclear factor Y (NF-Y) or Sp1 (34). Therefore, C/EBP␣ may in some manner regulate these factors. Indeed, it was reported that C/EBP␤ cooperates with Sp1 (35) or NF-Y (36) in regulating gene expression. Furthermore, it was recently demonstrated that C/EBP␣ activates gene expression by directly binding with DNA bound NF-Y (37). NF-Y interacts with the epoxide hydrolase 5/-1 bp CCAAT box in vitro and in vivo, but only NF-Y was unable to stimulate EPHX1 promoter activity. The NF-Y consensus sequence CCAAT is included in most genes involved in lipogenesis including FPP and 7DCR (38, 39). Therefore, the decrease in FPP and 7DCR expression in HC-fed ob/ob-C/EBP␣/Cre⫹ mice may be because of weaker synergistic activation of SREBP1 by NF-Y because of loss of C/EBP␣. This possibility remains to be explored at the experimental level. Even though loss of expression of C/EBP␣ in ob/ob results in attenuated induction of typical lipogenic genes and FPP and 7DCR, SREBP1 mRNA levels are unchanged or lower in the ob/ob-C/EBP␣/Cre⫹ mice. In addition, the activated nuclear form of SREBP1 in ob/ob-C/EBP␣/Cre⫹ mice is expressed at a higher level in the nucleus as compared with Cre⫺ mice, a situation that should lead to the induction of SREBP1 target genes. Therefore, hepatic C/EBP␣ may regulate gene expression by potentiating an SREBP1-independent pathway. Recent studies suggest the existence of SREBP1-independent pathway for induction of lipogenic genes by HC feeding and fasting-refeeding cycles. A carbohydrate-response element binding protein (ChoREBP) was identified as a transcription factor required for carbohydrate-responsive activation of transcription of the L-type pyruvate kinase as well as the lipogenic enzyme genes FAS and ACC (40). The expression of ChoREBP in ob/ob-C/EBP␣/Cre⫹ mice under a normal diet showed a tendency toward decreased levels. However, under a HC diet, the expression of ChoREBP was almost the same level in each genotype and thus did not correlate with the expression of lipogenic genes. This suggests that the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

ChoREBP pathway cannot account for the C/EBP␣dependent difference in inducible lipogenic genes in the ob/ob mouse. It is known that insulin is essential for the expression of lipogenic genes (41). The expression in liver is strikingly affected by insulin levels in the blood (42). Therefore, the decrease of insulin levels in ob/ob-C/EBP␣/ Cre⫹ mice as compared with ob/ob-C/EBP␣/Cre⫺ mice may be involved in one of the mechanisms of altered lipogenesis. This may also explain the specificity of ob/ob mice for “impaired induction” because no significant difference in insulin levels was observed between OB/OB-C/EBP␣/Cre⫹ and Cre⫺ mice under normal dietary conditions. Some reports have demonstrated that insulin treatment increases the amount of mRNA for SREBP-1c in parallel with the mRNAs of its target genes (43, 44), thus suggesting that SREBP1 gene expression may be influenced by insulin. However, in our study, the expression of SREBP1 in HCfed ob/ob-C/EBP␣/Cre⫹, is correlated with insulin levels; decreased insulin levels do not result in lower SREBP1 mRNA. The SREBP1-nuclear form is elevated in contrast to decreased expression of lipogenic genes. We currently have no explanation for these results. Additional studies are needed to determine the molecular mechanism for induction impaired by loss of hepatic C/EBP␣ expression. Hepatic C/EBP␣ Indirectly Regulates Glucose Homeostasis Additional phenotypes were observed in ob/ob-C/ EBP␣/Cre⫹ mice by HC feeding. Under a normal diet (nonfasting), the serum glucose levels of ob/ob-C/ EBP␣/Cre⫹ mice were significantly higher than in the ob/ob-C/EBP␣/Cre⫺ mice. These mice show a large individual difference in serum glucose levels under the normal dietary conditions. However, serum glucose levels in all ob/ob-C/EBP␣/Cre⫹ mice were dramatically elevated on the first day from start of HC feeding and remained at high levels throughout the course of feeding (12 d). The ob/ob-C/EBP␣/Cre⫹ mice also exhibited unusually high glucose levels after the GTT. GTT testing before and after HC feeding revealed that the diet does not appear to cause high glucose levels indicative of the worsening of insulin resistance, suggesting that ob/ob-C/EBP␣/Cre⫹ mice potentially have a unusual glucose clearance. Whereas, the insulin levels in ob/ob-C/EBP␣/Cre⫹ mice are significantly lower than in ob/ob-C/EBP␣/Cre⫺ mice. In addition, the result of insulin tolerance test clearly demonstrated lower insulin sensitivity in ob/ob-C/EBP␣/Cre⫹ mice. These phenotypes appear to sufficiently explain unusually high levels of glucose during HC feeding and the GTT. The lower insulin levels in ob/ob-C/EBP␣/ Cre⫹ mice are notably still elevated compared with that of the wild-type OB/OB group. This may explain why the ob/ob-C/EBP␣/Cre⫹ mice under a normal diet did not show a striking elevation in glucose levels compared with ob/ob-C/EBP␣/Cre⫺ mice.

Mol Endocrinol, November 2004, 18(11):2751–2764 2761

It is still uncertain why the deficiency of hepatic C/EBP␣ in ob/ob mice leads to lower insulin levels and worsen insulin resistance. The lower insulin appears to be caused by an insufficient insulin secretion from a defective pancreas that results from a yet to be defined mechanism. The pathological results did not reveal a clear difference in islets between ob/ob-C/ EBP␣/Cre⫺ and ob/ob-C/EBP␣/Cre⫹ mice. The islets of ob/ob-C/EBP␣/Cre⫹ and Cre⫺ mice were hypertrophic compared with OB/OB mice. However, in the ob/ob-C/EBP␣/Cre⫹ mice, there was a tendency toward more islet size variation and mild degenerative changes compared with ob/ob-C/EBP␣/Cre⫺ mice (data not presented). In agreement with a role for systemic FFA in the development of type 2 diabetes, it was shown that elevated plasma FFA induces peripheral insulin resistance in humans and rodent models within a few hours (45, 46). FFAs can also have positive or negative effects on insulin secretion, depending on the experimental conditions used (47, 48). Thus, FFAs might have a direct impact for worsening insulin secretion and resistance, and the persistent hyperglycemia may contribute to further worsening of the diabetic phenotype. It was reported that liver-specific disruption of glucokinase also leads to impaired insulin secretion (49). These authors suggested that chronic hyperglycemia may lead to an impairment of glucoseinduced insulin secretion by a mechanism that is still not established. More studies are needed to elucidate the mechanism for the more severe diabetic symptoms observed in the ob/ob-C/EBP␣/Cre⫹ mice. In summary, liver-specific disruption of the C/EBP␣ in obese diabetic mice decreased hepatic TG and TC by impairing induction of lipogenic genes and leaded to worsen diabetic phenotype. A single transcription factor expressed in liver can act globally to regulate both glucose and lipid concentrations in the whole diabetic animal. Further, the effects of deficiency of hepatic C/EBP␣ were more sensitive in diabetic mice, suggesting the existence of a type 2 diabetes-specific pathway. Therefore, the elucidation of a molecular mechanism might lead to potential new therapeutic opportunities for anti-diabetes therapy.

MATERIALS AND METHODS Generation of Liver-Specific C/EBP␣ Conditional-Null Mice C/EBP␣(fl/fl) mice, produced as described (9), were bred with a mouse containing the albumin-Cre transgene (AlbCre) (50). This transgene was used in an earlier study to create a peroxisome proliferator-activated receptor ␥ and hepatocyte nuclear factor 4␣-liver null mice (51, 52). C/EBP␣(fl/fl)AlbCre⫹ (C/EBP␣/Cre⫹) or C/EBP␣(fl/fl)AlbCre⫺ (C/EBP␣/Cre⫺) were crossed with heterozygotes C57BL/6J-Lepob, obtained from The Jackson Laboratory (Bar Harbor, ME) to generate double heterozygotes (C/EBP␣ fl/⫹, OB/ob). The OB/ob-C/EBP␣(fl/ fl)AlbCre⫹ or OB/ob-C/EBP␣(fl/fl)AlbCre⫺ mice were then crossed to generate the ob/ob-C/EBP␣/Cre⫹ or ob/ob-C/ EBP␣/Cre⫺, OB/OB-C/EBP␣/Cre⫹ or OB/OB-C/EBP␣/Cre⫺

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2762 Mol Endocrinol, November 2004, 18(11):2751–2764

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

mice. Mice were maintained on a 12-h light, 12-h dark cycle and fed water and pellet chow diet (NIH07) ad libitum. The NCI Animal Care and Use Committee approved all animal studies that were carried out in accordance with NIH Guide for the Care and Use of Laboratory Animals and ALAC guidelines.

Glucose tolerance tests were performed as previously described (51). Insulin tolerance tests were performed under the nonfasting and normal diet feeding. Mice were injected ip insulin (Humulin, Eli Lilly Indianapolis, IN) at 2 U/kg body weight.

Feeding Treatments

Histology

A HC/fat-free diet was purchased from Harlan Teklad (TD 88232, Madison, WI). For the HC study, mice were fed the diet for 12 d and killed in a nonfasted state. Before the fasting and refeeding study, the mice were fed a regular chow diet (NIH07) ad libitum until the treatment commenced. Mice in the refeeding group were fasted 24 h (from 1800 to 1800 h) and then refed with a HC/fat-free diet for 24 h (1800 to 1800 h). Mice in the fasting group were fasted 24 h (from 1800 to 1800 h). Both groups were immediately killed within 30 min.

Livers from 12- to 13-wk-old representative mice were fixed in 10% neutral buffered formalin and embedded in paraffin, and sections cut at a thickness of 4–6 ␮m were stained with hematoxylin and eosin. Some liver sections frozen in OCT compound were stained with Oil Red O using standard procedures.

RNA and Protein Analysis The cDNA probes used for Northern blotting were described in an earlier report (51, 53) except for the following probes. cDNA probes for FPP, 7-dehydrocholesterol reductase (7DCR), 6-phosphogluconate dehydrogenase (PGD), carbohydrate-responsive element-binding protein (ChREBP) were amplified by PCR from a mouse liver cDNA library by using gene-specific primers, and cloned into pGEM-T Easy Vector (Promega, Madison, WI). The primers used for PCR were as follows: ChREBP, 5⬘ primer CTCAACTCCATACAACCCTCGG and 3⬘ primer TGCCTCTCTGCTCAGGAACTAAGG. PGD, 5⬘ primer GCAAACCTCATCCAGGCTCAAC and 3⬘ primer TTATTACAAGTGGGACGGGGCG. FPP, 5⬘ primer GCTCCAGGCTTTCTTCCTTGTG and 3⬘ primer TCTATGAGACTCTTGAGGCGGTTG. 7DCR, 5⬘ primer TTGTGTACTACTTCATCATGGCATG and 3⬘ primer GGGTTGAACTCAATTCCCATCAT. The identities of the probes were confirmed by DNA sequencing. Nuclear extracts from ob/ob mouse liver were isolated according to the protocols provided with the NE-PER Nuclear and Cytoplasmic Extraction Reagents kit (Pierce, Rockford, IL). Nuclear extract protein was assayed by use of the BCA protein assay (Pierce) and 20 ␮g was subjected to electrophoresis on a 10% sodium dodecyl sulfate-polyacrylamide gel, transferred to Immobilon-P membranes (Millipore, Bedford, MA), and probed according to the manufacturer’s recommendations with anti-C/EBP␣ (14AA; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), C/EBP␤ (H7; Santa Cruz Biotechnology, Inc.) and SREBP-1 (IgG-2A4, BD-Pharmingen, San Diego, CA) antibodies. An enhanced chemiluminescence detection system was used to visualize immunoreactive proteins (Amersham, Inc., Arlington Heights, IL). Serum Lipids and Lipoproteins Serum lipid levels and lipoprotein profiles were determined as previously described (54). Post-heparin lipase activities were measured with CONFLUOLIP (PROGEN, Heidelberg, Germany) as noted in an earlier report (55). Measurement of Hepatic Lipids, Glycogen, and Plasma Insulin Hepatic lipids and glycogen were measure as described in previous report (56). Plasma insulin was measured with a RIA kit (Linco Research, St. Charles, MO). Glucose Levels, Glucose, and Insulin Tolerance Tests Glucose levels were measured analyzed for glucose concentrations using a Glucometer Elite (Bayer Corp., Elkhart, IN).

Acknowledgments We thank Jorge Paiz for technical assistance and Shioko Kimura and Linda Byrd of the Laboratory of Metabolism for their helpful suggestions.

Received May 25, 2004. Accepted July 27, 2004. Address all correspondence and requests for reprints to: Frank J. Gonzalez, Building 37, Room 3106, National Institutes of Health, Bethesda, Maryland 20892. E-mail: [email protected]. K.M. was supported by a postdoctoral fellowship from the Japanese Society for the Promotion of Science.

REFERENCES 1. Ramji DP, Foka P 2002 CCAAT/enhancer-binding proteins: structure, function and regulation. Biochem J 365:561–75 2. Williams SC, Cantwell CA, Johnson PF 1991 A family of C/EBP-related proteins capable of forming covalently linked leucine zipper dimers in vitro. Genes Dev 5:1553–1567 3. Landschulz WH, Johnson PF, Adashi EY, Graves BJ, McKnight SL 1988 Isolation of a recombinant copy of the gene encoding C/EBP. Genes Dev 2:786–800 4. Antonson P, Xanthopoulos KG 1995 Molecular cloning, sequence, and expression patterns of the human gene encoding CCAAT/enhancer binding protein ␣ (C/EBP ␣). Biochem Biophys Res Commun 215:106–113 5. Cao Z, Umek RM, McKnight SL 1991 Regulated expression of three C/EBP isoforms during adipose conversion of 3T3–L1 cells. Genes Dev 5:1538–1552 6. Lekstrom-Himes J, Xanthopoulos KG 1998 Biological role of the CCAAT/enhancer-binding protein family of transcription factors. J Biol Chem 273:28545–28548 7. Wang ND, Finegold MJ, Bradley A, Ou CN, Abdelsayed SV, Wilde MD, Taylor LR, Wilson DR, Darlington GJ 1995 Impaired energy homeostasis in C/EBP ␣ knockout mice. Science 269:1108–1112 8. Flodby P, Barlow C, Kylefjord H, Ahrlund-Richter L, Xanthopoulos KG 1996 Increased hepatic cell proliferation and lung abnormalities in mice deficient in CCAAT/enhancer binding protein ␣. J Biol Chem 271:24753–24760 9. Lee YH, Sauer B, Johnson PF, Gonzalez FJ 1997 Disruption of the c/ebp ␣ gene in adult mouse liver. Mol Cell Biol 17:6014–6022 10. Liu S, Croniger C, Arizmendi C, Harada-Shiba M, Ren J, Poli V, Hanson RW, Friedman JE 1999 Hypoglycemia and impaired hepatic glucose production in mice with a deletion of the C/EBP␤ gene. J Clin Invest 103:207–213

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

11. Arizmendi C, Liu S, Croniger C, Poli V, Friedman JE 1999 The transcription factor CCAAT/enhancer-binding protein ␤ regulates gluconeogenesis and phosphoenolpyruvate carboxykinase (GTP) gene transcription during diabetes. J Biol Chem 274:13033–13040 12. Ossipow V, Descombes P, Schibler U 1993 CCAAT/ enhancer-binding protein mRNA is translated into multiple proteins with different transcription activation potentials. Proc Natl Acad Sci USA 90:8219–8223 13. Shimomura I, Bashmakov Y, Horton JD 1999 Increased levels of nuclear SREBP-1c associated with fatty livers in two mouse models of diabetes mellitus. J Biol Chem 274:30028–30032 14. Horton JD, Goldstein JL, Brown MS 2002 SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J Clin Invest 109:1125–1131 15. Shimano H, Yahagi N, Amemiya-Kudo M, Hasy AH, Osuga J, Tamura Y, Shionoiri F, Iizuka Y, Ohashi K, Gotoda T, Harada K, Ishibashi S, Yamada N 1999 Sterol regulatory element-binding protein-1 as a key transcription factor for nutritional induction of lipogenic enzyme genes. J Biol Chem 274:35832–35839 16. Janowski BA 2002 The hypocholesterolemic agent LY295427 up-regulates INSIG-1, identifying the INSIG-1 protein as a mediator of cholesterol homeostasis through SREBP. Proc Natl Acad Sci USA 99:12675–12680 17. Sellem CH, Frain M, Erdos T, Sala-Trepat JM 1984 Differential expression of albumin and ␣-fetoprotein genes in fetal tissues of mouse and rat. Dev Biol 102:51–60 18. Tilghman SM, Belayew A 1982 Transcriptional control of the murine albumin/␣-fetoprotein locus during development. Proc Natl Acad Sci USA 79:5254–5257 19. Chen SS, Chen JF, Johnson PF, Muppala V, Lee YH 2000 C/EBP␤, when expressed from the C/ebp␣ gene locus, can functionally replace C/EBP␣ in liver but not in adipose tissue. Mol Cell Biol 20:7292–7299 20. Croniger C, Trus M, Lysek-Stupp K, Cohen H, Liu Y, Darlington GJ, Poli V, Handon RW, Reshef L 1997 Role of the isoforms of CCAAT/enhancer-binding protein in the initiation of phosphoenolpyruvate carboxykinase (GTP) gene transcription at birth. J Biol Chem 272: 26306–26312 21. Kaplan ML, Leveille GA 1981 Development of lipogenesis and insulin sensitivity in tissues of the ob/ob mouse. Am J Physiol 240:E101–E107 22. Goldstein JL, Brown MS 1990 Regulation of the mevalonate pathway. Nature 343:425–430 23. Jira PE, Waterham HR, Wanders RJ, Smeitink JA, Sengers RC, Wevers RA 2003 Smith-Lemli-Opitz syndrome and the DHCR7 gene. Ann Hum Genet 67:269–280 24. Shefer S, Salen G, Batta AK, Honda A, Tint GS, Irons M, Elias ER, Chen TC, Holick MF 1995 Markedly inhibited 7-dehydrocholesterol-␦ 7-reductase activity in liver microsomes from Smith-Lemli-Opitz homozygotes. J Clin Invest 96:1779–1785 25. Wassif CA, Maslen C, Kachilele-Linjewile S, Lin D, Linck LM, Connor WE, Steiner RD, Porter FD 1998 Mutations in the human sterol ␦7-reductase gene at 11q12–13 cause Smith-Lemli-Opitz syndrome. Am J Hum Genet 63:55–62 26. Fitzky BU, Moebius FF, Asaoka H, Waage-Baudet H, Xu L, Xu G, Maeda N, Kluckman K, Hiller S, Yu H, Batta AK, Shefer S, Chen T, Salen G, Sulik K, Simoni RD, Ness GC, Glossmann H, Patel SB, Tint GS 2001 7-Dehydrocholesterol-dependent proteolysis of HMG-CoA reductase suppresses sterol biosynthesis in a mouse model of Smith-Lemli-Opitz/RSH syndrome. J Clin Invest 108: 905–915 27. Dietschy JM, Siperstein MD 1967 Effect of cholesterol feeding and fasting on sterol synthesis in seventeen tissues of the rat. J Lipid Res 8:97–104 28. Bruenger E, Rilling HC 1988 Determination of isopentenyl diphosphate and farnesyl diphosphate in tissue samples

Mol Endocrinol, November 2004, 18(11):2751–2764 2763

29.

30.

31.

32. 33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

with a comment on secondary regulation of polyisoprenoid biosynthesis. Anal Biochem 173:321–327 Horton JD, Shah NA, Warrington JA, Andersson NN, Park SW, Brown MS, Goldstein JL 2003 Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes. Proc Natl Acad Sci USA 100:12027–12032 Amemiya-Kudo M, Shimano H, Hasty AH, Yahagi N, Yoshikawa T, Matsuzaka T, Okazaki H, Tamura Y, Iizuka Y, Ohashi K, Osuga J, Harada K, Gotoda T, Sato R, Kimura S, Ishibashi S, Yamada N 2002 Transcriptional activities of nuclear SREBP-1a, -1c, and -2 to different target promoters of lipogenic and cholesterogenic genes. J Lipid Res 43:1220–1235 Zhang F, Lin M, Abidi P, Thiel G, Liu J 2003 Specific interaction of Egr1 and c/EBP␤ leads to the transcriptional activation of the human low density lipoprotein receptor gene. J Biol Chem 278:44246–44254 McKnight SL 2001 McBindall—a better name for CCAAT/enhancer binding proteins? Cell 107:259–261 Misawa K, Horiba T, Arimura N, Hirano Y, Inoue J, Emoto N, Shimano H, Shimizu M, Sato R 2003 Sterol regulatory element-binding protein-2 interacts with hepatocyte nuclear factor-4 to enhance sterol isomerase gene expression in hepatocytes. J Biol Chem 278:36176–36182 Xiong S, Chirala SS, Wakil SJ 2000 Sterol regulation of human fatty acid synthase promoter I requires nuclear factor-Y- and Sp-1-binding sites. Proc Natl Acad Sci USA 97:3948–3953 Lee YH, Williams SC, Baer M, Sterneck E, Gonzalez FJ, Johnson PF 1997 The ability of C/EBP ␤ but not C/EBP ␣ to synergize with an Sp1 protein is specified by the leucine zipper and activation domain. Mol Cell Biol 17: 2038–2047 Yu L, Wu Q, Yang CP, Horwitz SB 1995 Coordination of transcription factors, NF-Y and C/EBP ␤, in the regulation of the mdr1b promoter. Cell Growth Differ 6:1505–1512 Zhu QS, Qian B, Levy D 2004 C/EBP␣ activates transcription of the human microsomal epoxide hydrolase gene (EPHX1) through the interaction with DNA Bound NF-Y. J Biol Chem 279:29902–29910 Kim JH, Lee JN, Paik YK 2001 Cholesterol biosynthesis from lanosterol. A concerted role for Sp1 and NF-Ybinding sites for sterol-mediated regulation of rat 7dehydrocholesterol reductase gene expression. J Biol Chem 276:18153–18160 Spear DH, Kutsunai SY, Correll CC, Edwards PA 1992 Molecular cloning and promoter analysis of the rat liver farnesyl diphosphate synthase gene. J Biol Chem 267: 14462–14469 Uyeda K, Yamashita H, Kawaguchi T 2002 Carbohydrate responsive element-binding protein (ChREBP): a key regulator of glucose metabolism and fat storage. Biochem Pharmacol 63:2075–2080 Foufelle F, Ferre P 2002 New perspectives in the regulation of hepatic glycolytic and lipogenic genes by insulin and glucose: a role for the transcription factor sterol regulatory element binding protein-1c. Biochem J 366: 377–391 Becard D, Hainault I, Azzout-Marniche D, Bertry-Coussot L, Ferre P, Foufelle F 2001 Adenovirus-mediated overexpression of sterol regulatory element binding protein-1c mimics insulin effects on hepatic gene expression and glucose homeostasis in diabetic mice. Diabetes 50: 2425–2430 Foretz M, Guichard C, Ferre P, Foufelle F 1999 Sterol regulatory element binding protein-1c is a major mediator of insulin action on the hepatic expression of glucokinase and lipogenesis-related genes. Proc Natl Acad Sci USA 96:12737–12742 Shimomura I, Bashmakov Y, Ikemoto S, Horton JD, Brown MS, Goldstein JL 1999 Insulin selectively in-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.

2764 Mol Endocrinol, November 2004, 18(11):2751–2764

45.

46.

47. 48.

49.

50.

creases SREBP-1c mRNA in the livers of rats with streptozotocin-induced diabetes. Proc Natl Acad Sci USA 96:13656–13661 Dresner A, Laurent D, Marcucci M, Griffin ME, Dufour S, Cline GW, Slezak LA, Andesen DK, Hundal RS, Rothman DL, Petersen KF, Shulman GI 1999 Effects of free fatty acids on glucose transport and IRS-1-associated phosphatidylinositol 3-kinase activity. J Clin Invest 103: 253–259 Roden M, Price TB, Perseghin G, Petersen KF, Rothman DL, Cline GW, Shulman GI 1996 Mechanism of free fatty acid-induced insulin resistance in humans. J Clin Invest 97:2859–2865 McGarry JD, Dobbins RL 1999 Fatty acids, lipotoxicity and insulin secretion. Diabetologia 42:128–138 Sako Y, Grill VE 1990 A 48-hour lipid infusion in the rat time-dependently inhibits glucose-induced insulin secretion and B cell oxidation through a process likely coupled to fatty acid oxidation. Endocrinology 127:1580–1589 Postic C, Shiota M, Niswender KD, Jetton TL, Chen Y, Moates JM, Shelton KD, Lindner J, Cherrington AD, Magnuson MA 1999 Dual roles for glucokinase in glucose homeostasis as determined by liver and pancreatic ␤ cell-specific gene knock-outs using Cre recombinase. J Biol Chem 274:305–315 Yakar S, Liu JL, Stannard B, Butler A, Accili D, Sauer B, LeRoith D 1999 Normal growth and development in the absence of hepatic insulin-like growth factor I. Proc Natl Acad Sci USA 96:7324–7329

Matsusue et al. • Role of C/EBP␣ in the Diabetic Mouse

51. Matsusue K, Haluzik M, Lambert G, Yim SH, Gavrilova O, Ward JM, Brewer Jr B, Reitman ML, Gonzalez FJ 2003 Liver-specific disruption of PPAR␥ in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes. J Clin Invest 111:737–747 52. Hayhurst GP, Lee YH, Lambert G, Ward JM, Gonzalez FJ 2001 Hepatocyte nuclear factor 4␣ (nuclear receptor 2A1) is essential for maintenance of hepatic gene expression and lipid homeostasis. Mol Cell Biol 21:1393–1403 53. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G, Gonzalez FJ 2000 Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102:731–744 54. Guo GL, Lambert G, Negishi M, Ward JM, Brewer Jr HB, Kliewer SA, Gonzalez FJ, Sinal CJ 2003 Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 278:45062–45071 55. Duque M, Graupner M, Stutz H, Wicher I, Zechner R, Paltauf F, Hermetter A 1996 New fluorogenic triacylglycerol analogs as substrates for the determination and chiral discrimination of lipase activities. J Lipid Res 37: 868–876 56. Akiyama TE, Lambert G, Nicol CJ, Matsusue K, Peters JM, Brewer Jr HB, Gonzalez FJ 2004 Peroxisome proliferator-activated receptor ␤/␦ regulates very low density lipoprotein production and catabolism in mice on a western diet. J Biol Chem 279:20874–20881

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 January 2017. at 21:19 For personal use only. No other uses without permission. . All rights reserved.