Abstract
Adipocyte differentiation is regulated by insulin and IGF-I, which transmit signals by activating their receptor tyrosine kinase. SH2-B is an adaptor protein containing pleckstrin homology and Src homology 2 (SH2) domains that have been implicated in insulin and IGF-I receptor signaling. In this study, we found a strong link between SH2-B levels and adipogenesis. The fat mass and expression of adipogenic genes including peroxisome proliferator-activated receptor γ (PPARγ) were reduced in white adipose tissue of SH2-B−/− mice. Reduced adipocyte differentiation of SH2-B-deficient mouse embryonic fibroblasts (MEFs) was observed in response to insulin and dexamethasone, whereas retroviral SH2-B overexpression enhanced differentiation of 3T3-L1 preadipocytes to adipocytes. SH2-B overexpression enhanced mRNA level of PPARγ in 3T3-L1 cells, whereas PPARγ levels were reduced in SH2-B-deficient MEFs in response to insulin. SH2-B-mediated up-regulation of PPARγ mRNA was blocked by a phosphatidylinositol 3-kinase inhibitor, but not by a MAPK kinase inhibitor. Insulin-induced Akt activation and the phosphorylation of forkhead transcription factor (FKHR/Foxo1), a negative regulator of PPARγ transcription, were up-regulated by SH2-B overexpression, but reduced in SH2-B-deficeint MEFs. These data indicate that SH2-B is a key regulator of adipogenesis both in vivo and in vitro by regulating the insulin/IGF-I receptor-Akt-Foxo1-PPARγ pathway.
ADIPOSE TISSUE PLAYS major roles in energy homeostasis, lipid metabolism, and insulin actions. It also acts as an endocrine organ that secretes a wide range of factors, such as leptin, adiponectin, plasminogen activator inhibitor-1 (PAI-1), and TNF-α, some of which are key regulators of energy homeostasis (1). Adipogenesis is tightly regulated by insulin and IGF-I signaling (2). Insulin is a major hormone controlling critical energy functions, such as glucose and lipid metabolism. IGF-I has been suggested to be a major regulator of adipose tissue growth and differentiation of preadipocytes into adipocytes. Recent studies have revealed that both insulin and IGF-I mediate adipocyte differentiation through the IGF-I receptor (IGF-IR) (3). The binding of insulin or IGF-I to their receptors induces their intrinsic tyrosine kinase activity, resulting in the recruitment and phosphorylation of multiple substrates, such as insulin receptor substrates (IRSs). These allow for the formation of macromolecular complexes close to the receptor. The two main transduction pathways are the phosphatidylinositol 3-kinase (PI3K) pathway and the MAPK pathway (4, 5). The MAPK pathway is considered to be involved in proliferation and differentiation, whereas the PI3K pathway plays a major role in metabolic functions, mainly via the activation of the Akt cascade. Activation of Akt stimulates glycogen synthesis, protein synthesis, cell survival, inhibition of lipolysis, and glucose uptake. This pathway is also considered to be important for adipogenesis (6–8).
It was previously believed that the activated insulin receptor (IR) and IGF-IR directly phosphorylate IRS. However, an adaptor protein, SH2-B, has recently been identified as a positive regulator of IR, IGF-IR tyrosine kinase activity and IRS phosphorylation (9, 10). SH2-B is a ubiquitously expressed cytoplasmic protein that contains a pleckstrin homology, a Src-homology-2 (SH2) domain and multiple phosphorylation sites (11). SH2-B binds via its SH2 domain to not only IR and IGF-IR but also multiple tyrosine kinases, including the platelet-derived growth factor receptor, the fibroblast growth factor receptor, the GH receptor and Janus kinase 2 (JAK2) (12–17). However, the precise molecular mechanism of SH2-B’s potentiation of tyrosine kinase activity remains to be clarified. Rui and his colleagues (18) also demonstrated that SH2-B directly binds, via its pleckstrin homology and SH2 domains, to both IRS1 and IRS2 and mediates the formation of a JAK2/SH2-B/IRS1 or IRS2 tertiary complex. Consequently, SH2-B dramatically enhances the leptin-stimulated tyrosine phosphorylation of IRS1 and IRS2, resulting in the promotion of PI3K and Akt activation.
To investigate the physiological role of SH2-B, two research groups, including ours, have independently generated SH2-B−/− mice by homologous recombination (19, 20). SH2-B−/− mice are infertile; thus, SH2-B is essential for reproduction (19). Impaired signal transduction of the IGF-I receptor in SH2-B-deficient mice results in poor gonad development. SH2-B−/− mice with a 129 and C57BL/6 mixed background also demonstrate neonatal growth retardation within 2–6 wk of birth (19). These phenotypes may be related to a reduced response to GH or IGF-I. [Duan et al. (20) also reported that SH2-B−/− mice develop insulin resistance and glucose intolerance.] This group further observed that SH2-B−/− males gain body weight rapidly and SH2-B−/− males were approximately two times heavier than their wild-type (WT) littermates at 21 wk of age. These phenotypes may be explained by the reduced leptin sensitivity of SH2-B−/− mice.
However, in the current study, we noticed a strong reduction in the mass of white adipose tissue (WAT) in SH2-B−/− mice under C57BL/6 background. In vitro adipocyte differentiation experiments indicated that, whereas SH2-B overexpression in 3T3-L1 fibroblasts enhanced adipocyte differentiation, SH2-B-deficient MEF possessed a reduced adipocyte differentiation potential. We also found that SH2-B enhanced peroxisome proliferator-activated receptor γ (PPARγ) mRNA expression, at least in part, by regulating the PI3-kinase-Akt-Foxo1 pathway. We conclude that SH2-B is a key mediator that positively regulates adipogenesis both in vivo and in vitro.
RESULTS
SH2-B Expression in Adipose Tissue and Induction during Differentiation in Vitro
To investigate the role of SH2-B in adipogenesis, we first examined the expression of SH2-B in various tissues. SH2-B protein was abundant in skeletal muscle, WAT, and the pancreas, as measured by immunoblotting with anti-SH2-B antibodies (Fig. 1A). We noticed that the SH2-B was expressed in WAT of normal mice (Fig. 1A: Ad). By RT-PCR analysis, we defined the SH2-Bβ and SH2-Bγ isoforms as major transcripts in WAT (Fig. 1B). Then we examined the expression of SH2-B in WAT in obese animal models; db/db mice, which lack the functional leptin receptor, and normal mice fed a high-fat diet for more than 20 wk. These mice possess a large mass of fat tissue and hyperlipidemia due to their abnormal lipid metabolism. SH2-B expression was strongly up-regulated in the WAT of these obese animal models (Fig. 1, C and D), suggesting that SH2-B is involved in adipocyte differentiation in these conditions.
SH2-B Expression in Adipose Tissues and during Differentiation in Vitro A, SH2-B expression detected by Western blotting in the tissues of normal mice. Br, Brain; Lu, lung; He, heart; Li, liver; Ki, kidney; Pa, pancreas; LI, large intestine; SI, small intestine; Mu, skeletal muscle; Ad, WAT; Te, testis. B, Detection of SH2-B isoforms by RT-PCR of RNA from Li, Mu, and Ad. C and D, Expression of SH2-B in obese WAT. SH2-B mRNA was measured by real-time RT-PCR in the WAT from 11- to 13-wk-old db/db mice or C57BL/6J mice fed a high-fat diet for 25 wk (n = 6–8). E, The expression of SH2-B during adipocyte differentiation of 3T3-L1 cells determined by real-time RT-PCR. Values are the mean ± sem (*, P < 0.05 compared with d 0).
To determine whether SH2-B is induced during adipocyte differentiation, we examined 3T3-L1 preadipocytes. 3T3-L1 cells were differentiated into mature adipocytes upon exposure to a mixture of hormonal stimuli including insulin, glucocorticoids, and a cAMP enhancer (21). As shown in Fig. 1E, during the differentiation process, quantitative real-time PCR analysis revealed that the expression of the SH2-B mRNA was increased 3 d after the induction of differentiation and the expression was maintained during differentiation. These data suggest that SH2-B expression is associated with adipocyte differentiation and/or maturation.
WAT Development Was Defective in SH2-B-Deficient Mice
To further define the physiological involvement of SH2-B in adipogenesis, we examined SH2-B-deficient mice. After breeding them on the C57BL/6 background, unlike the C57BL/6X129 mixed background, we found that the percentage of SH2-B−/− pups was much lower than the Mendelian expectation (data not shown). We found that the surviving SH2-B-null mice showed postnatal growth retardation and proportionate dwarfism (Fig. 2B). As shown in Fig. 2, A, C, and D, the total mass and weight of WAT were extremely reduced in SH2-B−/− mice compared with those of the WT littermates. These results indicate that SH2-B plays an essential role in adipose tissue development.
WAT Development in SH2-B-Deficient Mice A, Anatomical view of WT and SH2-B−/− mice after 24 wk on a limited normal diet. B, Body weight change in WT and SH2-B−/− mice. C, Total weight of adipose tissues of 24-wk-old WT and SH2-B−/− mice fed a normal diet. Total fat pads (inguinal, gonadal, retroperitoneal, and mesenteric fat pads) were collected and weighed. D, Relative weights of different fat depots (% of body weight) in WT and SH2-B−/− mice (n = 4 each). E, Histological examination of the EWAT of WT and SH2-B−/− mice stained with hematoxylin and eosin. F, Distribution of the number and size of adipocytes in WAT. Number of adipocytes and their size in fixed area (5,000,000 μm2) were quantitated by an image analysis system. The representative result from EWAT. G, Adipogenic gene expression in adipose tissue from SH2-B−/− mice. mRNA levels of indicated genes in inguinal fat pads from SH2-B−/− and WT mice were detected by RT-PCR.
We then examined SH2-B−/− mice histologically. In the epididymal WAT (EWAT) of SH2-B−/− mice, lipid accumulation was much less than that in the WT littermates. Hematoxylin and eosin staining and quantification of the images confirmed an extremely reduced content of lipids in SH2-B−/− mice. In WAT, the size of each adipocyte of SH2-B−/− mice was smaller than that of WT mice (Fig. 2E). Adipose tissues stained with hematoxylin and eosin were analyzed as described (22). Morphometric analysis of adipocyte distribution along with their size (sectional area) indicated that the size of each adipocyte of SH2-B−/− mice was smaller than that of WT mice (Fig. 2F).
Low Expression of Adipogenesis Genes in WAT of SH2-B−/− Mice
We investigated the expression of adipogenic genes in WAT of SH2-B−/− mice by RT-PCR (Fig. 2G). The expression of PPARγ and CCAAT/enhancer-binding protein (C/EBP), two master transcriptional regulators for controlling adipogenic gene transcription, were low in WAT of SH2-B−/− mice compared with that of controls (23–33). In addition, WAT of SH2-B−/− mice contained much lower levels of functional adipogenic genes such as fatty acid synthase, sterol regulatory element binding protein 1c, and adipocyte fatty acid binding protein (aP2) (Fig. 2G). The expression of adipocyte-derived hormones such as adiponectin and leptin was reduced in WAT of SH2-B−/− mice (Fig. 2G). These data indicate that WAT differentiation might be retarded in SH2-B−/− mice.
Effects of SH2-B on in Vitro Adipocyte Differentiation
To determine the effects of SH2-B on preadipocyte differentiation, we overexpressed each isoform of SH2-B in 3T3-L1 cells. SH2-Bα, SH2-Bβ, and SH2-Bγ were introduced into 3T3-L1 cells with a retroviral vector carrying an internal ribosomal entry site (IRES)-enhanced green fluorescent protein (EGFP), and stable transformants were isolated by fluorescence-activated cell sorting. Comparable levels of SH2-Bα, SH2-Bβ, and SH2-Bγ were detected by Western blotting (Fig. 3A). The accumulation of lipids was visualized with Oil Red O staining and quantified by dye extraction. As shown in Fig. 3, B and C, overexpression of SH2-Bβ resulted in strong promotion of accumulation of lipid droplets in a larger fraction of cells. SH2-Bβ was the most effective, and SH2-Bα also significantly promoted lipid accumulation, whereas SH2-Bγ showed little effect (Fig. 3B). RT-PCR analyses showed that SH2-Bβ overexpression increased the expression of the adipogenic transcriptional regulators, such as PPARγ and C/EBPα (Fig. 3D). Western blotting also indicated that the induction of PPARγ occurred 1 d earlier in SH2-Bβ transformants than in control 3T3-L1 cells. Consistent with these results, the expression of downstream adipogenic genes aP2 was increased by SH2-B overexpression (Fig. 3, D and E). These data suggest that SH2-Bβ overexpression led to early induction of PPARγ and aP2, resulting in stronger induction of lipogenesis in 3T3-L1 cells.
Effects of SH2-B Overexpression on in Vitro Adipocyte Differentiation A, Expression of exogenous SH2-Bα, SH2-Bβ, and SH2-Bγ in 3T3-L1 cells detected by Western blotting. B, Oil Red O staining of stable SH2-B transformants. Microscopic pictures were taken 5 d after induction of differentiation. C, Quantitative analysis of panel B. Differentiation into adipocytes was induced on d 0 by treating confluent cells for 2 d with insulin, Dex, and IBMX. The medium was replaced with DMEM containing 10% FBS and insulin on d 2. Oil Red O was extracted from stained 3T3-L1 cells and quantified by absorbance at 520 nm. Values are the mean ± sem (*, P < 0.01). D, Adipogenic gene expression detected by RT-PCR in 3T3-L1 transformants after induction of differentiation. E, PPARγ levels were detected by Western blotting after induction of differentiation.
To further analyze the physiological importance of SH2-B in adipocyte differentiation, primary mouse embryonic fibroblasts (MEFs) were prepared from WT and SH2-B−/− embryonic d 12.5 embryos and then used for hormone-induced adipocyte differentiation. SH2-B expression in WT MEFs was confirmed by Western blotting (Fig. 4A). On d 8 after induction, cells were stained with Oil Red O staining. As shown in Fig. 4, B and C, lipid droplets accumulated in WT MEFs, whereas SH2-B−/− MEFs contained a much smaller amount of lipid droplets than WT MEFs. RT-PCR analysis revealed that the expression of aP2 was markedly reduced in SH2-B−/− MEFs (Fig. 4D). Furthermore, forced expression of SH2-Bβ by retrovirus resulted in strong enhancement of lipid accumulation and aP2 expression in both WT and SH2-B−/− MEFs (Fig. 4, E–G). Taken together, these data indicate that SH2-B positively regulates adipogenesis.
Effects of SH2-B on in Vitro Adipocyte Differentiation A, Detection of SH2-B in MEFs from WT and SH2-B−/− mice by Western blot analysis. B and C, Oil Red O staining of MEFs 8 d after the induction of differentiation. C, Quantitative analysis of B. D, Adipogenic gene expression detected by RT-PCR in MEFs on day 8. Values are the mean ± sem. *, P < 0.01. E, Oil Red O staining of MEFs transfected with or without SH2-Bβ. Microscopic pictures were taken 8 d after induction of differentiation. F, Quantitative analysis of E (N.S., Not significant. *, P < 0.01). G, aP2 gene expression detected by RT-PCR in MEFs transfected with SH2-Bβ on d 8.
SH2-B Enhances PPARγ Expression through the PI3-Kinase Pathway
PPARγ (PPARγ2) has been shown to be a master regulator of adipocyte differentiation. We noticed that PPARγ mRNA and protein levels were higher in SH2-Bβ transformants than control transformants before insulin stimulation (Fig. 3E), which may account for the enhanced adipogenesis with SH2-B overexpression. Insulin treatment further up-regulated PPARγ levels within a few hours, and it was much more prominent in SH2-B overexpressing cells. In contrast, there was less PPARγ induction in SH2-B−/− MEFs (Fig. 5, A and B).
Effect of SH2-B on PPARγ Expression Levels and Activity A and B, PPARγ mRNA expression levels in SH2-Bβ transformants and SH2-B−/−MEFs stimulated with insulin for indicated periods were detected by real-time RT-PCR and normalized to that of β-actin (n = 3). C and D, 3T3-L1 cells were transiently transfected with a 3xPPRE-luciferase reporter plasmid vector and β-galactosidase vector. Cells were stimulated with insulin (300 ng/ml) in the presence or absence of PD-98059 and LY-294002 (50 μm). PPRE luciferase activity and PPARγ mRNA levels in empty vector transformants and SH2-Bβ transformants were measured as described in Materials and Methods. E, PPRE luciferase activity in SH2-Bβ transformants transfected with DN-Akt or CA-Akt was measured after stimulation with insulin (300 ng/ml). Values are the mean ± sem (N.S., not significant; *, P < 0.05).
It has been suggested that the PI3-kinase-Akt pathway enhances PPARγ expression, whereas the ERK pathway suppresses expression (34, 35). To examine the effect of SH2-B and downstream signaling pathways on PPARγ mRNA levels, SH2-B-overexpressing 3T3-L1 cells and control cells were stimulated with insulin in the presence or absence of PΙ3Κ inhibitor (LY294002) or MAPK kinase (MEK)-ERK inhibitor (PD98059). As shown in Fig. 5, C and D, LY294002 strongly inhibited PPARγ mRNA induction as well as PPARγ transcriptional activity in response to insulin. In addition, dominant-negative Akt (DN-Akt) inhibited PPARγ transcriptional activity induced by insulin, whereas constitutively active Akt (CA-Akt) enhanced it (Fig. 5E). In contrast, PD98059 enhanced PPARγ transcriptional activity. These inhibitory and stimulatory effects by ERK and Akt, respectively, were observed in both control and SH2-Bβ-overexpressing cells. The up-regulation of PPARγ mRNA in response to insulin was much more dramatic in SH2-B transformants than in parental 3T3-L1 cells. The effect of inhibitors was very clearly observed in SH2-B-overexpressing cells.
SH2-B Enhances the IRS-Akt-Foxo1 Pathway
To define the molecular mechanism for the effect of SH2-B on PPARγ induction, we examined insulin signaling in 3T3-L1 preadipocytes. As shown in Fig. 6A, overexpression of SH2-Bα and SH2-Bβ isoforms, but not SH2-Bγ, in 3T3-L1 cells resulted in stronger Akt activation in response to insulin than in control 3T3-L1 cells. Higher Akt activation by SH2-Bα and SH2-Bβ, but not SH2-Bγ, overexpression was well correlated with enhanced adipocyte differentiation (see Fig. 3B). We then examined insulin signal transduction using SH2-Bβ transformants (Fig. 6B) and SH2-B-deficient MEFs (Fig. 7, A and B). The insulin-induced phosphorylation of IRS-1 and Akt was enhanced in SH2-Bβ overexpressing cells (Fig. 6B), whereas IRS-1 and Akt phosphorylation was severely impaired in SH2-B−/− MEFs (Fig. 7, A and B). And the Akt phosphorylation levels were ameliorated by transfection of SH2-Bβ (Fig. 7C). ERK activation in SH2-B transformants and MEFs were not very evident in our experimental conditions (Figs. 6B and 7B). These data demonstrated that SH2-B is an important molecular link between the IR/IGF-IR and the IRS1-Akt pathways. See quantitative analysis for Figs. 6B and 7, A and B, in Supplemental Figure 1 published on The Endocrine Society’s Journals Online web site as supplemental data on http://mend.endojournals.org.
Effect of SH2-B on Insulin Signaling in 3T3-L1 Transformants Expressing SH2-B A and B, 3T3-L1 transformants expressing SH2-B isoforms (A) and SH2-Bβ transformants (B) were stimulated with insulin (300 ng/ml) for indicated periods. After lysis, cell extracts were analyzed by Western blotting using indicated antibodies. The band intensity was quantified by a densitometer.
Effect of SH2-B on Insulin and IGF-I Signaling in SH2-B−/− MEFs A and B, MEFs from WT and SH2-B−/− mice were stimulated with insulin (300 ng/ml) for the indicated periods. The band intensity of panels A and B was quantified. C, MEFs transfected with or without SH2-Bβ were stimulated with insulin (300 ng/ml) for the indicated periods. The band intensity was quantified. D, MEFs from WT and SH2-B−/− mice were stimulated with IGF-I (100 ng/ml). After lysis, cell extracts were analyzed by Western blotting using the indicated antibodies. The band intensity was quantified by a densitometer. P, Phosphorylated.
Recently, it has been shown that insulin at supraphysiological concentrations induces preadipocyte differentiation through the IGF-IR (3). Thus, we examined Akt activation in response to IGF-I. As shown in Fig. 7D, Akt activation levels were reduced in SH2-B−/− MEFs compared with WT MEFs. These data confirmed that SH2-B regulates not only insulin receptor but also IGF-IR signaling.
It has been shown that SH2-B is a positive regulator of nerve growth factor-mediated activation of Akt/Forkhead pathway in PC12 cells (36). One of the downstream targets of Akt for adipogenesis is the transcription factor FKHR/Foxo1. The inactivation of Foxo1 by phosphorylation promotes adipocyte differentiation (37) and PPARγ expression (34). We confirmed the hyperphosphorylation of Foxo1 in SH2-Bβ overexpressing 3T3-L1 cells in response to insulin but the reduced phosphorylation of Foxo1 in SH2-B-deficient MEFs (Figs. 6B and 7A). These data explain the mechanism of regulation of PPARγ expression and adipogenesis by SH2-B.
DISCUSSION
In this study, we demonstrated the pivotal role of SH2-B in adipogenesis both in vivo and in vitro. In the adipocyte differentiation process, the transcriptional regulation of adipocyte-specific genes during differentiation is relatively well characterized. PPARγ and C/EBPα are two master regulators for controlling adipogenic genes (33, 38–40). However, the signal transduction pathways or networks by which the differentiation inducers lead to the activation of the adipogenic transcription program are not as well understood. PI3K inhibitors have been shown to strongly suppress 3T3-L1 adipocyte differentiation. The Akt signal cascade appears to induce or activate PPARγ and C/EBP during the induction of 3T3-L1 adipocyte differentiation (33, 38–41). Recently, PPARγ (PPARγ2) expression has been shown to be negatively regulated by Foxo-1 (34). Foxo-1 directly interacts with the PPARγ promoter and inhibits its transcription. On the other hand, insulin induces phosphorylation and nuclear export of Foxo-1 through Akt; therefore, the IR/IGF-IR-IRS-PI3-kinase-Akt pathway promotes PPARγ expression. This pathway was clearly up-regulated in SH2-B overexpressing cells, and down-regulated in SH2-B-deficient cells. Thus, we propose that SH2-B is an important regulator of Akt and adipogenesis both in vitro and in vivo.
However, the precise molecular mechanism by which SH2-B enhances the IR/IGF-IR-IRS1 cascade remains to be investigated. Previously, SH2-B was shown to interact with both IR/IGF-IR and IRSs (9, 10). Thus, SH2-B may facilitate the recruitment of IRSs to the tyrosine kinase. This could explain why the loss of SH2-B in MEFs is more profound in Akt than in ERK. In addition, SH2-B has been suggested to enhance tyrosine kinase activity by binding to the phosphotyrosine residue of the kinase activation loop of several tyrosine kinases, including IR/IGF-IR and JAK2. SH2-B may form a dimer (42), therefore promoting the dimerization and cross-phosphorylation of tyrosine kinases. Enhanced IRS phosphorylation may be due to enhanced kinase activity.
On the other hand, the downstream pathway of Akt in adipogenesis is now clearer. Inactivation of Foxo-1 by phosphorylation promotes adipocyte differentiation, and its constitutively active mutant prevents adipogenesis (37). The direct phosphorylation of Foxo-1 by Akt is shown to inhibit its transcriptional activity (37, 43). Foxo-1 has been shown to interact with PPARγ, and it has been shown that they reciprocally antagonize each other’s activity (43). However, we found that the IR/IGF-IR-Akt pathway rather induces the transcription of PPARγ (Fig. 5C). Recently, the direct suppressive effect of Foxo-1 on the PPARγ transcription was shown (34) and our data strongly support this notion. We found that changes in the PPARγ transcriptional activity were parallel to its mRNA levels (Fig. 5D), suggesting that transcriptional control is a major regulator of PPARγ activity. Our data clearly demonstrate the importance of SH2-B for the IR/IGF-IR-IRS-PI3-kinase-Akt-Foxo-1-PPARγ pathway as well as adipocyte differentiation. Enhanced PPARγ transcription by insulin in SH2-B−/− MEFs was completely blocked by a PI3-kinase inhibitor and DN-Akt, which further supports our hypothesis. In contrast, a MEK-ERK inhibitor enhanced PPARγ mRNA levels, suggesting that ERK negatively regulates PPARγ transcription. This phenomenon has been described before (44), but its molecular basis has not been clarified. Because the effects of the ERK inhibitor on PPARγ mRNA levels were drastic in SH2-B overexpressing cells, these cells must be useful for searching the mechanism of ERK mediated suppression of PPARγ transcription.
Rui and colleagues (20) reported that SH2-B-homozygous null mice developed metabolic syndromes, such as hyperlipidemia, hyperglycemia, hyperleptinemia, and hyperinsulinemia, and hepatic steatosis, which are commonly associated with obesity. They suspected that one of the causes of the observed obesity is that SH2-B is an endogenous mediator of leptin signaling and is also required for maintaining normal energy metabolism and body weight. This is in contrast to our observations that SH2-B−/− mice had growth retardation and reduced fat mass. This may be due to the genetic background of mice or conditions under which mice were kept. Rui and colleagues (20) mentioned that when SH2-B−/− males and their WT littermates were housed individually and pair-fed, SH2-B−/− mice gained significantly less body weight than their WT littermates. They thought that the reduction of weight gain was most likely caused by the elevated energy expenditure in SH2-B−/− mice. These data raise a possibility that energy intake and expenditure may be controlled by two distinct pathways that may be differentially regulated by SH2-B; leptin and insulin signaling. Both groups observed impaired insulin receptor activation and signaling in the liver, skeletal muscle and fat, including impaired tyrosine phosphorylation of IRS1 and -2 and activation of PI3-kinase, and Akt pathways. Therefore, in our SH2-B−/− mice, the effect of impaired insulin action in the fat, muscle and liver may be more apparent than in mice of Rui and colleagues (18) in which the effect of impaired leptin signaling may be greater. In conclusion, SH2-B is a key regulator of adipogenesis both in vivo and in vitro by regulating the insulin/IGF-I receptor-Akt-Foxo1-PPARγ pathway.
MATERIALS AND METHODS
Animal Experiments
The generation of SH2-B−/− mice was reported previously (19). SH2-B+/− males were backcrossed with WT C57BL/6 females (The Jackson Laboratory, Bar Harbor, ME) more than seven times to obtain SH2-B−/− mice in a C57BL/6 background. Mice were housed on a 12-h light, 12-h dark cycle in specific pathogen-free conditions in the unit at Kyushu University with free access to water and standard mouse chow. All experiments using these mice were approved by and performed according to the guidelines of the Animal Ethics Committee of Kyushu University (Fukuoka, Japan).
Cell Culture and Adipocyte Differentiation
3T3-L1 preadipocytes were maintained as described previously (21). Their differentiation into adipocytes was induced first by treatment of confluent cells for 2 d with 0.5 mm isobutylmethylxanthine (IBMX), 1 μm dexamethasone (Dex), and insulin (10 μg/ml) in DMEM supplemented with 10% fetal bovine serum (FBS) and then by further culturing in DMEM containing 10% FBS and insulin (10 μg/ml) for the subsequent 4–6 d, and the medium was replenished every other day.
Primary MEFs were obtained from 11.5- to 12.5-d SH2-B+/− females crossed with SH2-B+/− males. For differentiation, confluent monolayers of MEFs were cultured in a differentiation medium (DMEM containing 10% FBS supplemented with 0.5 mm IBMX, 1 μm Dex, and 10 μg/ml insulin) for 8 d, and the medium was replenished every 2 d.
Retroviral Gene Transduction
The retrovirus vector carrying SH2-B (α, β, and γ) cDNA with IRES-EGFP (PMY-Flag-IRES-EGFP) was kindly provided by I. Nobuhisa (Kumamoto University, Kumamoto, Japan). PMY-IRES-EGFP (empty) was used as a control vector. Retrovirus packaging was performed as previously described (45). 3T3-L1 cells and MEFs were incubated in a virus stock medium containing polybrene (Roche Applied Science, Indianapolis, IN) at 4 μg/ml for 4 h. EGFP-positive cells were sorted with fluorescence-activated cell sorting.
Oil Red O Staining
The Oil-Red-O staining was performed as described previously (46). After staining, the cultures were rinsed with 60% isopropanol. The cultures were then thoroughly washed with PBS and visualized (47). Quantification of Oil Red O was done by extracting the dye with 100% isopropanol, and absorbance was measured by spectrophotometry at 520 nm.
RT-PCR Analysis
Total RNAs from adipose tissue and cells were prepared using the TRIzol reagent (Invitrogen Corp., Carlsbad, CA), reverse transcribed, and analyzed by semiquantitative PCR or quantitative real-time PCR. Quantitative real-time RT-PCR was performed using SYBR Green (Applied Biosystems, Foster City, CA) and the ABI 7000 sequence detector system (Applied Biosystems). The SH2-B and PPARγ quantity was normalized by the levels of β-actin. PCR was done by using the following forward and reverse primers: aP2 (5′-TGA TGC CTT TGT GGG AAC CT-3′, 5′-GCT TGT CAC CAT CTC GTT TTC TCT-3′), C/EBPα (5′-GGA ACA GCT GAG CCG TTGA AC-3′, 5′-GCG ACC CGA AAC CAT CCT-3′), PPARγ (5′-GCC AGT TTC GAT CCG TAG AAG-3′, 5′-AGT CCT TGT AGA TCT CCT GG-3′), murine sterol regulatory element binding protein 1c (5′-ATC GGC GCG GAA GCT GTC GGG GTA GCG TC-3′, 5′-ACT GTC TTG GTT GTT GAT GAG CTG GAG CAT-3′), fatty acid synthase (5′-GGC TTT GGC CTG GAA CTG GCC CGG TGG CT-3′, 5′-TCG AAG GCT ACA CAA GCT CCA AAA GAA TA-3′), m-Leptin (5′-ATT TCA CAC ACG CAG TCG GTA T-3′, 5′-AAG CCC AGG AAT GAA GTC CA-3′), m-Adiponectin (5′-GCC AGT CAT GCC GAA GA-3′, 5′-TCT CCA GCC CCA CAC TGA AC-3′) and SH2-B (5′-TTC GAT ATG CTT GAG CAC TTC CGG-3′, 5′-GCC TCT TCT GCC CCA GGA TGT-3′). The mouse β-actin (5′-ACT GGG ACG ACA TGG AGA AG-3′, 5′-GGG GTG TTG AAG GTC TCA AA-3′) gene was amplified as a control.
Immunoblot Analysis
Immunoblotting was performed as described (48). Anti-phospho-ERK1/2 (catalog no. 9106), anti-phospho-Akt (catalog no. 9271), and anti-Akt (catalog no. 9272), anti-PPARγ (catalog no. 07-466), anti-phospho-FKHR (Foxo1) (catalog no. 9461s) antibodies were purchased from Cell Signaling Technology, and anti-ERK2 antibody (catalog no. SC-154) and anti-IRS-1 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-phospho-IRS-1 was purchased from Calbiochem (La Jolla, CA). The antibody against mSH2-B (catalog no. 611527) was obtained from BD Biosciences (Franklin Lakes, NJ).
Measurement of PPARγ Transcriptional Activity
3Τ3-L1 cells were seeded 1 d before transfection. On the next day, cells usually reached 50% confluency and were transfected with 3xPPRE-luciferase reporter plasmid vector (kindly provided by H. Morinaga, Kyushu University) and β-galactosidase vector by FuGENE6 reagent following the manufacturer’s instructions (Roche Applied Science). Twenty-four hours after transfection, cells were stimulated with insulin (300 ng/ml) for 16 h in the presence or absence of LY294002 (50 μm) or PD98059 (50 μm). The luciferase activity was determined using a luminometer as described (48). Constitutively active-Akt (CA-Akt) and DN-Akt plasmids (kindly provided by Y. Gotoh, Tokyo University) were transfected in 3Τ3-L1 cells with 3xPPRE-luciferase reporter plasmid vector and β-galactosidase vector by FuGENE6 reagent. Twenty-four hours after transfection, cells were stimulated with insulin (300 ng/ml) for 16 h and the luciferase activity was determined.
Statistical Analysis
For statistical analysis, we used Student’s t test, and a 95% confidence limit was taken to be significant, defined as P < 0.05.
Acknowledgments
We thank T. Yoshioka, Ms. M. Othsu, and Ms. Y. Yamada for technical assistance and Y. Nishi for manuscript preparation.
This work was supported by special grants-in-aid from the Ministry of Education, Science, Technology, Sports, and Culture of Japan, the Haraguchi Memorial Foundation, the Yamanouchi Foundation for Research on Metabolic Disorders, the Takeda Science Foundation, the Mochida Memorial Foundation, the Kato Memorial Foundation, and the Uehara Memorial Foundation.
Disclosure Statement: All authors have nothing to disclose.
Abbreviations
- aP2
Adipocyte fatty acid binding protein;
- CA-Akt
constitutively active Akt;
- C/EBP
CCAAT/enhancer-binding protein;
- Dex
dexamethasone;
- DN-Akt
dominant-negative Akt;
- EGFP
enhanced GFP;
- EWAT
epididymal WAT;
- FBS
fetal bovine serum;
- IBMX
isobutylmethylxanthine;
- IGF-IR
IGF-I receptor;
- IR
insulin receptor;
- IRES
internal ribosomal entry site;
- IRSs
insulin receptor substrates;
- JAK2
Janus kinase 2;
- MEFs
mouse embryonic fibroblasts;
- MEK
MAPK kinase;
- PAI-1
plasminogen activator inhibitor-1;
- PI3K
phosphatidylinositol 3-kinase;
- PPARγ
peroxisome proliferator-activated receptor γ;
- WAT
white adipose tissue;
- WT
wild type.
