CCAAT/Enhancer-binding Protein α Mediates Induction of Hepatic Phosphoenolpyruvate Carboxykinase by p38 Mitogen-activated Protein Kinase*

Excessive hepatic gluconeogenesis and glucose production are important contributors to hyperglycemia in both type 1 and type 2 diabetes. In diabetic humans and animal models, elevated levels of p38 mitogen-activated protein kinase (p38) are observed in several tissues. Our study shows that activity of p38 is significantly elevated in livers of db/db or streptozocin-induced type 1 diabetic mice. Using cultured hepatoma cells, we find that activation of p38 enhances expression of hepatic gluconeogenic gene phosphoenolpyruvate carboxykinase (PEPCK). Furthermore, our studies demonstrate that activation of p38 stimulates phosphorylation of CCAAT/enhancer-binding protein α (C/EBPα) at serine 21 and increases its transactivation activity in the context of PEPCK gene transcription. Our results indicate that C/EBPα mediates p38-stimulated PEPCK transcription in liver cells.

The liver is an important organ for maintaining blood glucose homeostasis. After a meal, part of the absorbed glucose is transported into the liver and converted to glycogen and other molecules for energy storage. During fasting, in addition to glycogenolysis, the hepatocytes synthesize glucose via gluconeogenesis using non-carbohydrate substrates. The precise and dynamic switch between hepatic glucose uptake and output is essential for maintaining blood glucose concentration in a normal range. In both type 1 and type 2 diabetes, excessive hepatic gluconeogenesis contributes to hyperglycemia (1)(2)(3). Although new transcription factors and signaling pathways have recently been identified that regulate expression of hepatic gluconeogenic genes, the underlying molecular mechanisms that provoke inappropriate hepatic gluconeogenesis in diabetes are still poorly understood.
Hepatic gluconeogenesis is tightly controlled by hormones that regulate expression of gluconeogenic enzymes such as PEPCK 2 and glucose-6-phosphatase (G6Pase). PEPCK is gen-erally considered to be the first committed step in gluconeogenesis. The regulation of PEPCK is executed primarily at the transcriptional level because allosteric modification of PEPCK has not been described (4,5). Transcriptional regulation of PEPCK has been extensively studied, and a cluster of transcription factors or coactivators is involved in this regulatory process (5). One of these transcription factors is CCAAT/enhancer-binding protein-␣ (C/EBP␣), which is the founding member of the C/EBP family and which was first identified in crude nuclear extracts from rat liver (6). C/EBP␣ is preferentially expressed in the liver, adipose tissue, certain cells of the lung, and the mammary glands (7). There is overwhelming evidence that C/EBP␣ is an important regulator of integrative processes that control glucose homeostasis (5, 8 -11). Targeted deletion of the C/EBP␣ gene in mice results in a profound derangement of liver structure and function (8 -10). C/EBP␣ Ϫ/Ϫ mice die soon after birth because of hypoglycemia caused by impaired hepatic glycogen storage and low expression of gluconeogenic enzymes, including PEPCK and G6Pase (8). Our recent study revealed that despite the comparable levels of total C/EBP␣ protein in livers of wild type and db/db diabetic mice, knocking down C/EBP␣ decreases PEPCK and hepatic glucose production specifically in db/db mice (11). These results suggest that the transactivation activity of C/EBP␣ may be altered in the livers of db/db diabetic mice.
Inflammation has been proposed as a pathological mechanism for type 2 diabetes. Growing evidence supports this hypothesis and demonstrates that a variety of stresses or proinflammatory cytokines, especially cytokines from adipocytes, not only contribute to insulin resistance and diabetes but also play a critical role in the development of diabetic complications (12,13). The p38 family of proline-directed serine/threonine kinases has four members, each encoded by individual genes (14). p38␣ is the predominant isoform in liver (15). p38 can be strongly activated by environmental stresses and inflammatory cytokines but is inconsistently activated by insulin and growth factors (14). Activation of p38 has been observed in several tissues of diabetic animals and human subjects (16 -20). Early studies reported that p38 stimulates activating transcription factor 2 (ATF2), which increases PEPCK promoter activation in liver and kidney cells (21,22). While we were preparing this manuscript, Cao et al. (23) reported that p38 mediates cAMPstimulated PEPCK and G6Pase gene expression and gluconeogenesis. These data suggest that p38 not only is a component of the cAMP signaling cascade, but it also contributes to the excessive hepatic gluconeogenesis in diabetes.
Here our study shows that p38 induces phosphorylation of C/EBP␣ on serine 21 in liver cells. Furthermore, serine phosphorylation enhances C/EBP␣ transactivation activity and increases PEPCK gene expression, suggesting that C/EBP␣ mediates p38-induced gluconeogenic enzyme expression.
Cell Culture-FAO cells, derived from the H4IIE hepatoma cell line, possess a complete gluconeogenic enzyme system. Human hepatoma HepG2 cells were also used for transient transfection study. The cells were maintained at 37°C, 5% CO 2 in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Gemini Bio-Products, Woodland, CA).
Glucose Production Assay-Glucose production from FAO cells was measured as previous described (24) using a colorimetric glucose oxidase assay (Sigma-Aldrich). Briefly, 24 h after transfection the cells were washed three times with phosphatebuffered saline. The cells were then incubated for 6 h at 37°C, 5% CO 2 in glucose production buffer (glucose-free Dulbecco's modified Eagle's medium, pH 7.4, containing 20 mM sodium lactate, 1 mM sodium pyruvate, and 15 mM HEPES, without phenol red). To estimate rates of gluconeogenesis after activation of p38, glucose concentrations in FAO cell culture medium were measured. The glucose assays were conducted in duplicate, and the intra-assay coefficient of variation was less than 5%.
Nuclear Protein Isolation and Western Blot Analysis-Total cell lysate and nuclear protein extracts were isolated as described previously (24). For Western blot, 76 g of total cell lysates or 30 g of nuclear protein extract was separated by 10% acrylamide SDS-PAGE gel, and the proteins were transferred to polyvinylidene difluoride membrane (Bio-Rad) with a Mini Trans Blot cell from Bio-Rad. After blocking with 5% fat-free milk or bovine serum albumin, the membrane was blotted with 1:1000 dilution of antibody. After washing, the membrane was blotted with a 1:10,000 dilution of horseradish peroxidase-conjugated anti-rabbit or anti-mouse IgG antibody (Santa Cruz Biotechnology). Peroxidase activity was visualized with ECL-Plus from Amersham Biosciences. The specific band density was determined using Kodak Image Station 2000R and 1D software (Kodak, New Haven, CT).
RNA Extraction and Quantitative Reverse Transcription-PCR Analysis-Total RNA was prepared from plated FAO cells with TRIzol reagent following the manufacturer's protocol (Invitrogen). cDNA was synthesized using SuperScript III reverse transcriptase and oligo(dT) 12-18 primer (Invitrogen). Real time PCR was performed using the Mx3000P real time PCR system (Stratagene) and SYBR green dye (Molecular Probes, Eugene, OR). Sequences of primers are: PEPCK, 5Ј-ACGCCATTAAGACCATCCAG-3Ј and 5Ј-TCGATGC-CTTCCCAGTAAAC-3Ј; and 18S, 5Ј-CGAAAGCATTTGC-CAAGAAT-3Ј and 5Ј-AGTCGGCATCGTTTATGGTC-3Ј. Amount of PCR product was calculated from standard curves established from each primer pair. Expression data were normalized to the amount of 18S PCR product.
Luciferase Assay-Reporter constructs and expression plasmids were transfected into cells by FuGENE 6 transfection reagent (Roche Applied Science). A pCMV-␤-galactosidase plasmid was cotransfected to control for transfection efficiency (Clontech Laboratories, Inc., Palo Alto, CA). Twenty-four hours after transfection, the cells were lysed, and luciferase activity was measured (26). The luciferase data were normalized relative to ␤-galactosidase luminescence and are expressed as relative luciferase activity.
Experimental Animals-Male C57BL/6J and C57BL/6Jϩ/ϩLepr db/db (db/db) mice (9 -10 weeks old) were purchased from Jackson Laboratory (Bar Harbor, ME). The mice were housed in a pathogen-free animal facility under a standard 12 h of light/12 h of dark cycle with free access to food and water. The experiments using mouse models were carried out under the Association for Assessment and Accreditation of Laboratory Animal Care guidelines with approval of the University of Kentucky Animal Care and Use Committee. Type 1 diabetic C57BL/6J mice were created with a single intraperitoneal injection of steptozotocin (180 mg/kg of body weight). Six days after the onset of diabetes, liver tissues were collected. The mice were fasted overnight before collection of the samples.
Data Analysis-The data are expressed as the means Ϯ S.E. Statistical analysis was performed using Student's t test or analysis of variance, followed by a contrast test with Tukey or Dunnett error protection. The differences were considered significant at p Ͻ 0.05.

Hyperactivation of p38 and Increased Phosphorylation of C/EBP␣ at Serine 21 in the Livers of db/db and STZ-induced
Diabetic Mice-To study the roles of MAP kinase in the excessive hepatic glucose production associated with diabetes, activation of the three main MAP kinase pathways was studied in  AUGUST 25, 2006 • VOLUME 281 • NUMBER 34

JOURNAL OF BIOLOGICAL CHEMISTRY 24391
the livers of both type 1 and type 2 diabetic mouse models. Obese male db/db mice at 10 weeks of age were used as a model for type 2 diabetes, whereas administration of STZ, which destroys ␤-cells in the pancreas, was used to create a model for type 1 diabetes. Fasting blood glucose concentrations above 400 mg/dl were observed for both diabetic models (Table 1).
To assess whether diabetes is associated with activation of MAP kinases, liver samples were collected from db/db mice and from mice 1 week after the onset of STZ-induced diabetes. Activation of p38 and ERK1/2 was measured by Western blot using anti-phospho-p38 (Thr 180 /Tyr 182 ) or phospho-ERK1/2 (Thr 202 /Tyr 204 ) specific antibody (Cell Signaling). JNK activities were measured using a JNK assay kit (Cell Signaling), and c-Jun fusion protein was used as substrate. The results showed that phosphorylation of p38 was significantly increased in livers of db/db and STZ-induced diabetic mice (Fig. 1A). The phosphorylation of ERK1/2 was significantly higher only in livers of STZ-induced type 1 diabetic mice (Fig. 1A). Differences in ERK1/2 phosphorylation between db/db mice and their controls were not observed ( p Ͼ 0.05; Fig. 1A). Consistent with a prior study (27), JNK activity was significantly increased in the livers of db/db mice ( p Ͻ 0.05; Fig. 1B). In addition, there was a trend toward increased JNK activity in the livers of STZ-induced diabetic mice ( p ϭ 0.115; Fig. 1B). Taken together, these data indicate that p38 is the only MAP kinase that is activated in the livers of both type 1 and type 2 diabetic mice. C/EBP␣ is a key transcriptional regulator of PEPCK and G6Pase and thus hepatic gluconeogenesis (8 -11). We therefore conducted the following studies to verify whether C/EBP␣ is involved in the excessive hepatic gluconeogenesis in diabetes. The protein levels of C/EBP␣ in these diabetic mouse models were measured by Western blot. The results showed that the total C/EBP␣ protein levels in the livers were comparable between STZ-induced type 1 diabetic mice or db/db diabetic mice and their controls, respectively ( p Ͼ 0.05; Fig. 1C, top panel), suggesting that diabetes does not influence transcription or translation of C/EBP␣. Similar to most transcription factors, functional regulation of C/EBP␣ is another mechanism for controlling the transactivation activities. Several threonine and serine residues in C/EBP␣ are phosphorylated by protein kinase C or glycogen synthase kinase 3 (25,28,29). Furthermore, the phosphorylation of C/EBP␣ appears to increase or decrease its transactivation activity (25,28,29). A recent study demonstrated that serine 21 residue of C/EBP␣ can be directly phosphorylated by ERK1/2 in adipocytes (25). The above results indicate that some MAP kinase pathways are activated in the livers of diabetic mice. Thus, we measured the phosphorylation status of C/EBP␣ by Western blot using anti-phospho-C/EBP␣ (Thr 222/226 ) or anti-phospho-C/EBP␣ (Ser 21 ) antibody (Cell Signaling) and the nuclear extract protein samples. As shown in Fig. 1C, the phosphorylation of C/EBP␣ at serine 21 was significantly increased in both db/db and STZ-induced type 1 diabetic mice compared with their control, respectively ( p Ͻ 0.05; Fig. 1C). In contrast, significant changes were not observed in phosphorylation of C/EBP␣ on threonine 222 and 226 in livers from these diabetic mice (data not shown). In following sections, phosphorylation of C/EBP␣ refers to serine 21 phosphorylation.
Activation of p38 MAP Kinase Increases PEPCK Gene Expression-Studies have suggested that activation of p38 contributes to hyperglycemia by increasing hepatic gluconeogenesis (21)(22)(23). Our study revealed that there is a positive correlation between p38 phosphorylation and PEPCK mRNA levels in livers of db/db and STZ-induced diabetic mice (data not shown). Next we conducted the following in vitro studies to FIGURE 1. Activation of p38 MAP kinase in livers of db/db and STZ-induced diabetic mice. Liver tissues were collected after overnight fasting. A, activation p38 MAP kinase or ERK1/2 was measured by Western blot using anti-phospho-p38 MAP kinase (Thr 180 /Tyr 182 ) or anti-phospho-ERK1/2 (Thr 202 /Tyr 204 ) antibody. After normalization to total p38 MAP kinase or ERK1/2, the data are presented as the means Ϯ S.D. B, JNK activation was measured using recombinant c-Jun as substrate and following the protocol provided by the kit manufacturer. C, nuclear protein was extracted from liver tissues. Total and phosphorylated C/EBP␣ were detected by Western blot using anti-C/EBP␣ or anti-phospho-C/EBP␣ (Ser 21 ) antibody. The data are presented as the means Ϯ S.D. (n ϭ 6/group). WB, Western blot. investigate whether activation of p38 increases PEPCK gene expression in FAO rat hepatoma cells, which are an established model for studying gluconeogenesis (24,30,31). Activation of p38 was induced with anisomycin or the p38 upstream kinase MKK6. PEPCK mRNA levels were measured by quantitative PCR. Anisomycin is a reagent that activates both p38 and JNK pathways. Therefore, a p38-or JNK-specific inhibitor was used to establish specificity. The results showed that a 2-h treatment with anisomycin (1 g/ml) increased PEPCK mRNA nearly 2-fold ( p Ͻ 0.05; Fig. 2A). Although addition of the p38 inhibitor (SB203580) almost completely abolished the increase in anisomycin-induced PEPCK mRNA, the JNK inhibitor (SP600126) had no effect ( Fig. 2A), indicating that induction of PEPCK was mediated specifically by p38.
To confirm these results, we used a constitutively active MKK6, which selectively activates p38. As shown in Fig. 2B, ectopic expression of MKK6(Glu) and p38␣ increased PEPCK mRNA levels significantly compared with cells transfected with the empty vector pcDNA alone ( p Ͻ 0.05; Fig. 2B). Although overexpression of p38␣ tended to increase expression of PEPCK mRNA, the differences were not statistically significant (data not shown). However, ectopic expression of MKK6(Glu) alone increased PEPCK mRNA (Fig. 2B, second column), indicating that activation of p38 rather than expression of p38 is important for regulating transcription of PEPCK. Consistent with the important role of PEPCK in gluconeogenesis, we also observed that FAO cells that are treated with anisomycin or that express MKK6(Glu) and p38␣ have elevated production of glucose (data not shown). Taken together, these studies further support the hypothesis that activation of p38 increases PEPCK gene expression and hepatic glucose production.
p38 MAP Kinase Phosphorylates C/EBP␣ at Serine 21 in Liver Cells-To determine whether C/EBP␣ plays a role in the induction of PEPCK gene expression by p38, the following studies were conducted. Initially we studied the effects of p38 activation on C/EBP␣ gene expression and found that activation of p38 with anisomycin or with coexpression of MKK6(Glu) and p38␣ does not alter the C/EBP␣ gene expression in FAO cells (Fig. 3, A-C). We next studied the effects of p38 activation on C/EBP␣ phosphorylation in FAO cells. As showed in Fig. 3A, anisomycin robustly increased C/EBP␣ phosphorylation at the serine 21 site. Because anisomycin can activate both p38 and JNK, a specific inhibitor for either p38 or JNK was employed to distinguish which pathway is responsible for the C/EBP␣ phosphorylation. Although the addition of a p38-specific inhibitor (SB203580) completely abolished anisomycin-stimulated C/EBP␣ phosphorylation, the JNK inhibitor (SP600126) did not influence C/EBP␣ phosphorylation (Fig. 3A, last two lanes). These data suggest that activation of p38, but not JNK, increases C/EBP␣ phosphorylation. We also conducted a time course study to investigate the correlation between p38 activation and C/EBP␣ phosphorylation. As expected, anisomycin quickly activated p38, which was measured by Western blot using anti-phospho-p38 antibody (Fig. 3B). Importantly, phosphorylation of C/EBP␣ closely paralleled activation of p38 (Fig.  3B). Phosphorylation of p38 and C/EBP␣ was robustly induced after 10 min of anisomycin treatment and reached a peak at 30 min. Although overexpression of p38␣ alone did not increase C/EBP␣ phosphorylation (data not shown), in our transient transfection studies, overexpression of MKK6(Glu) and WT p38␣ robustly increased the phosphorylation of C/EBP␣ (Fig.  3C). Taken together, these data provide strong support for the hypothesis that activation of p38 induces C/EBP␣ phosphorylation in hepatoma cells.
It is known that insulin is a strong activator of ERK and other MAP kinases in liver cells (14). Thus, we treated FAO cells with insulin for the indicated times (Fig. 3D) and observed that although insulin treatment increased ERK1/2 and JNK phosphorylation, there was no effect on activation of p38. Phosphorylation of C/EBP␣ was also not altered by insulin treatment (Fig. 3D). Similar results were also observed in HepG2 and H4IIE cells (data not shown). These studies suggest that neither ERK1/2 nor JNK MAP kinase phosphorylates C/EBP␣ at serine 21 in these hepatocytes. Together, these results indicate that p38 MAP kinase stimulates C/EBP␣ phosphorylation at the serine 21 residue in FAO cells.
Phosphorylation of C/EBP␣ Increases Transcriptional Activity and Increases PEPCK Gene Transcription-The preceding results demonstrate that activation of p38 increases C/EBP␣ phosphorylation and PEPCK gene expression in FAO cells. We next tested whether phosphorylation of C/EBP␣ enhances its transactivation activity and mediates p38-induced PEPCK gene expression. In the first study, plasmids encoding wild type C/EBP␣ (C/EBP␣WT), C/EBP␣WT in which serine 21 is mutated to aspartate (C/EBP␣S21D; mimics phosphorylation), or alanine (C/EBP␣S21A; mimics dephosphorylation) were transfected into FAO cells. Transfection and expression efficiency is illustrated in Fig. 4A (top panel). The 42-kDa isoform was predominantly expressed in cells transfected with these plasmids. As expected, overexpression of wild type C/EBP␣ significantly increased PEPCK mRNA levels ( p Ͻ 0.05; Fig. 4A). Ectopic expression of C/EBP␣S21A also increased PEPCK mRNA ( p Ͻ 0.05; Fig. 4A), suggesting that the absence of serine 21 phosphorylation of expressed C/EBP␣ does not reduce its transactivation activity. This result is consistent with a previous observation in other types of cells (25). Interestingly, expression of C/EBP␣S21D increased the PEPCK mRNA level further (Fig.  4A). Because C/EBP␣ is a transcription factor mediating cAMP-induced PEPCK gene transcription (32) and p38 is a component of the cAMP signaling cascade (23), we treated transfected cells with dibutyryl cAMP to determine whether phosphorylation of C/EBP␣ increases cAMP-stimulated PEPCK transcription. The cAMP-induced PEPCK mRNA levels were significantly higher in cells expressing C/EBP␣S21D than in cells transfected with plasmid encoding wild type C/EBP␣ (Fig. 4B). These data suggest that phosphorylation of C/EBP␣ at serine 21 increases basal and cAMP-induced PEPCK gene expression.
Using a PEPCK promoter reporter gene assay, we studied the effects of p38 activation and C/EBP␣ phosphorylation on PEPCK promoter activity. We employed HepG2 cells, which are human hepatoma cells and are a well characterized model for studying regulation of PEPCK promoter activity (33,34). Although expression of p38␣ alone did not alter PEPCK promoter activity, activation of p38 by cotransfection with MKK6(Glu) increased PEPCK promoter activity slightly (Fig. 5A). Interestingly, coexpression of p38␣ and MKK6(Glu) increased the effect of C/EBP␣ by over 2-fold (Fig. 5A), suggesting that C/EBP␣ is involved in p38-induced PEPCK gene transcription. Furthermore, this synergistic effect of p38␣, MKK6(Glu), and C/EBP␣ on PEPCK promoter-directed luciferase expression suggests that p38 MAP kinase may increase PEPCK promoter activity by inducing C/EBP␣ phosphorylation.
To study the effects of C/EBP␣ phosphorylation on its transactivation activity, we cotransfected the PEPCK promoter-reporter gene with plasmids encoding WT or mutant C/EBP␣ into HepG2 cells. Although overexpression of wild type C/EBP␣ or C/EBP␣S21A increased luciferase activity ϳ6-fold ( p Ͻ 0.001; Fig. 5B), luciferase activities were induced almost 12-fold in cells transfected expressing C/EBP␣S21D, which mimics phosphorylation at serine 21 (p Ͻ 0.001; Fig. 5B). These studies strongly suggest that phosphorylation of C/EBP␣ at serine 21 increases C/EBP␣ transactivation activity and increases PEPCK gene transcription.

DISCUSSION
Excessive hepatic glucose production is an important contributor to hyperglycemia in both type 1 and type 2 diabetes. Although hepatic insulin resistance is causal, the mechanisms underlying excessive hepatic glucose production remain poorly understood. In this study, we report that p38 is activated in the livers of both type 1 and type 2 diabetic mice. We demonstrate that activation of p38 induces C/EBP␣ phosphorylation at serine 21. Furthermore, phosphorylation of C/ EBP␣ enhances its transactivation activity in the context of PEPCK gene transcription. Therefore, we propose that C/EBP␣ mediates p38-stimulated PEPCK transcription in liver cells.
Development of diabetes and diabetic complications has been associated with stress and inflammation. For example, increased expression of pro-inflammatory cytokines, particularly from adipose tissue of obese mice or humans, has been linked with the deterioration of insulin sensitivity (12,35). In addition, chronic hyperglycemia is a typical metabolic stress that promotes the development of diabetic complications (12,35). A study from Arkan et al. (13) provides evidence supporting the notion that inflammation underlies the metabolic disorders of insulin resistance and type 2 diabetes. The study demonstrates that IB kinase ␤ acts locally in liver and systemically in myeloid cells, where NF-B activation induces inflammatory mediators that cause insulin resistance. A recent study reported that endoplasmic reticulum stress activates JNK in the livers of ob/ob diabetic and high fat diet-induced obese mice (27). The study also indicated that hyperactivated JNK increases insulin receptor substrate 1 serine phosphorylation and protein degradation, which leads to insulin resistance (27). Furthermore, studies from Pagliassotti and co-workers (37,38) indicate that endoplasmic reticulum stress induced by metabolic turbulence impairs insulin signaling via JNK activation in liver and increases expression and activity of G6Pase and the capacity for glucose release in primary rat hepatocytes and H4IIE liver cells.
p38 is the second stress-activated MAP kinase (14). The activation of p38 by hyperglycemia or elevated pro-inflammatory cytokines has been reported in a variety of tissues from diabetic human subjects and animal models (16 -20). Our study shows that both p38 and JNK are activated in the livers of db/db diabetic mice. Studies have indicated that p38 activation enhances expression of PEPCK in both hepatocytes and kidney cells (21,22). Knocking down endogenous p38 in the liver reduced PEPCK mRNA level and hepatic glucose production in mice (23). Our current study provides further evidence demonstrating that activated p38 enhances PEPCK expression and hepatic gluconeogenesis. Together, these results suggest that hyperactivated p38 in the livers of diabetic mice contributes to hyperglycemia by increasing hepatic glucose production.
Available information suggests that several transcription factors or coactivators mediate p38-induced PEPCK gene expression. ATF2 and PGC-1␣ have been identified as mediators of p38 in liver or kidney cells (21,23,39). p38 induces phosphorylation of PGC-1␣ and ATF2 and increases their transactivation of PEPCK transcription (21,23). Our study provides an additional mechanism by which p38 induces PEPCK gene transcription in liver cells, adding C/EBP␣ as a new mediator. Our Twenty-four hours after transfection, the cells were lysed, and luciferase activity was measured. Luciferase activities were corrected for transfection efficiency by normalization to ␤-galactosidase activity. The data are presented as relative luciferase activities (A) or fold activation relative to luciferase activities of constructs cotransfected with pcDNA vector (B). *, p Ͻ 0.05 versus pcDNA transfected cells.
study demonstrated that C/EBP␣ mediates p38-induced PEPCK gene transcription in a similar fashion as PGC-1␣ and ATF2. Therefore, hyperactivated p38 in the livers of diabetes enhances C/EBP␣, PGC-1␣, and ATF2 transactivation activity, which may communally enhance PEPCK gene expression and increase hepatic glucose production. The activation of these transcription factors or coactivator as well as PEPCK transcription also imply that p38 increases hepatic glucose production directly through regulation of gene transcription. Therefore, metabolic and inflammatory stress induces excessive hepatic glucose production not only by JNK-induced insulin resistance but also through a p38-enhanced gluconeogenic enzyme transcription.
C/EBP␣ plays a pivotal role in regulating hepatic gluconeogenic gene expression and hepatic glucose production (5, 8 -10). However, it is not clear whether and how C/EBP␣ is involved in the development of excessive hepatic glucose production in diabetes. The C/EBP␣ protein levels are comparable in the liver between diabetic mice and their controls, suggesting that a functional alteration may be involved (11). C/EBP␣ transactivation activity can be regulated by covalent modifications such as phosphorylation or acetylation (25,28,29,36). Our current study demonstrated that p38 can induce C/EBP␣ phosphorylation at serine 21 and that the phosphorylation increases its transactivation activity in the context of PEPCK gene transcription. Therefore, we propose that C/EBP␣ contributes to the development of excessive hepatic glucose production by increasing transactivation activity. Interestingly, phosphorylation of serine 21 of C/EBP␣ does not alter the DNA binding affinity, it does change conformation of the C/EBP␣ protein (25). Thus, further effort will be required to elucidate mechanisms by which p38-induced C/EBP␣ phosphorylation increases its transactivation activity and enhances PEPCK gene transcription.