| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Helix Research Institute (R.N.-W., Y. Ma., H.W.), 1532-3 Yana Kisarazu-shi Chiba, 292-0812 Japan; The Second Department of Anatomy (Y. Mo., E.S.), Wakayama Medical College, Kimiidera, Wakayama 641-0012, Japan; Department of Hematology and Oncology (I.M.), Faculty of Medicine, University of Osaka, Yamadaoka Suita-shi Osaka 565-0871, Japan; and Hubid Genome Systems Co. (M.-A.M.), Hirakawa-cho, Chiyoda-ku, Tokyo 102-0093, Japan
Address all correspondence and requests for reprints to: Hiroshi Wakao, Ph.D., Laboratory of Immune Regulation, Riken Research Center for Allergy and Immunology, c/o Department of Molecular Immunology, School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku Chiba, 260-8670 Japan. E-mail: hwakao{at}med.m.chiba-u.ac.jp.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, which plays crucial roles in adipogenesis. We have also generated transgenic mice in which dn-Stat5A is expressed in an adipose tissue-specific fashion and found attenuation of peroxisome proliferator-activated receptor- and of many adipocyte-related genes. These results highlight a novel role of Stat5 in adipocyte differentiation. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, resistin, and adiponectin (2, 3, 4). Among these, leptin mediates the satiety signal in regulation of energy metabolism, whereas TNF and resistin are involved in insulin resistance in obesity (2, 5, 6). Adipocytes also take up and release glycerol and free fatty acids, which contribute to regulation of hepatic and peripheral glucose metabolism. With regard to glucose homeostasis, adipose tissue along with heart and skeletal muscle are the only tissues known to express and regulate the insulin-dependent glucose transporter 4, which facilitates the entry of glucose into these cells and out of the circulation postprandially (7). Therefore, the adipocyte differentiation program comprises regulation of genes involved in all aspects of molecular events such as cell morphology, storage and release of triglycerides, glucose metabolism, and energy homeostasis. Obesity, which is marked by an excessive accumulation of adipose tissues, affects many Western countries, is a major risk factor of insulin resistance, and contributes to pathological states such as diabetes, hypertension, and cardiovascular disease. Hence, a better understanding of the molecular basis for adipocyte differentiation opens a door to the development of therapeutic interventions.
Adipogenesis is a complex process controlled by the interplay of signals emanating from both environmental and intracellular factors (8, 9, 10). Preadipocyte cell lines such as 3T3-L1 and 3T3-F422A have served as in vitro models to delineate complex cascades of transcriptional events during the differentiation process. Three classes of transcription factors have been identified that directly influence fat cell development. These include CCAAT-enhancer binding proteins (C/EBPs), the nuclear hormone receptor peroxisome proliferator-activated receptor- (PPAR), and adipocyte determination differentiation factor 1/sterol response element binding protein-1c (ADD1/SREBP-1c) (8, 9, 10, 11). Among these, C/EBPß and -
are induced at an early stage of adipogenesis and required for subsequent fulfillment of the adipogenic program (12, 13, 14). In vitro study has demonstrated that ectopic expression of C/EBPß and, to a lesser extent, C/EBP leads to the conversion of preadipocytes or NIH-3T3 cells into adipocytes (13, 14). Gene knockout experiments have indicated that ablation of either C/EBPß or - isoform results in a slight reduction in the adipogenic potential of embryonic fibroblasts, whereas depletion of both leads to severe impedance (15). C/EBP is induced relatively late in differentiation, after the induction of PPAR but before the synthesis of many enzymes and proteins characteristic of fully differentiated cells (9). As in the case of C/EBPß, ectopic expression of C/EBP in fibroblasts is sufficient to convert them into adipocytes (16, 17). Furthermore, C/EBP knockout mice display dramatically reduced fat accumulation in white adipose tissue (WAT) and brown adipose tissue (BAT) (18). These findings demonstrate that C/EBPs play a vital role in fat cell development. At present, it is thought that induction of C/EBPß and - is prerequisite for the subsequent expression of PPAR (19). Gain-of-function experiments have shown that overexpression of PPAR in the presence of its agonists is sufficient to induce adipogenesis in a variety of cell types including fibroblasts and myoblasts (20, 21). In addition, gene ablation experiments disclosed an essential role of PPAR in the formation of fat (22, 23, 24). ADD1/SREBP-1c is a third transcription factor associated with adipogenesis (25). It is also reported that ectopic expression of ADD1 in fibroblasts leads to their conversion into adipocytes in the presence of the PPAR agonists (26). Recent experiments have proposed that ADD1/SREBP-1c activates PPAR expression directly or through the production of an endogenous ligand, thus enabling cells to differentiate into adipocytes (27, 28).
It is clear that these factors function as proadipogenic factors in adipogenesis. One has to keep in mind, however, that all these factors are undetectable or expressed at very low levels before differentiation, and their expression increases only after the induction of adipogenesis. It is highly conceivable, therefore, that the expression of these factors is controlled and coordinated by protein(s) present before adipogenic stimulation. Although C/EBPß and - are assumed to be required for PPAR induction, in vivo data from the C/EBPß and - double knockout mice indicate that there must be C/EBPß- and -independent pathway(s) leading to PPAR and C/EBP expression (15). 3T3-L1 cells undergo differentiation upon challenge with an adipogenic hormone cocktail consisting of methylisobutylxanthine (MIX), dexamethasone (DEX), insulin, and fetal bovine serum (FBS). Because MIX and DEX are direct inducers for C/EBPß and -, respectively, we hypothesized that factors posttranslationally activated by FBS may play a role in adipogenesis together with C/EBPß and - (12). As adipocytes contain several cytokine receptors such as the GH and the prolactin receptors (29, 30), we assumed that downstream molecule(s) underpinning the function of these receptors might be a candidate molecule(s). We have hypothesized that signal transducer and activator of transcription 5 (Stat5), a prolactin/GH-signaling mediator, may play a role in adipogenesis. This hypothesis is based on our previous report that prolactin stimulates NIH3T3 adipogenesis (31). Stat5 is composed of two highly homologous genes, Stat5A and Stat5B, and is used as a signal transducer by a variety of cytokines and growth factors (32, 33). It has been shown also that Stat5 as well as Stat1 proteins are up-regulated during the adipogenic conversion of 3T3-L1 or NIH-3T3ß/ cells, whereas Stat3 and Stat6 expression remains constant (34, 35). These observations suggested that Stat5 is involved in adipose cell development. In this report, we have constructed conditional expression systems in which the expression of either a wild-type (wt) or a dominant negative (dn) form of Stat5A was regulated under the doxycycline (Dox)-controlled promoter in 3T3-L1 cells (36). These systems allowed us to evaluate some roles of Stat5 in adipocyte differentiation. Our current experiments demonstrated that Stat5A and B were strongly activated at an early stage of differentiation, and ectopic expression of Stat5A influenced expression of PPAR, which ultimately resulted in altered triglyceride accumulation at a later stage of differentiation of 3T3-L1 cells. To further support the in vitro data, we have generated transgenic mice in which dn-Stat5A is expressed in an adipose tissue-specific manner. We found that overexpression of dn-Stat5A in adipose tissues also compromised expression of PPAR. These data demonstrate that Stat5 regulates, at least in part, adipogenesis through controlling the expression of PPAR.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Stat5A was strongly tyrosine phosphorylated at 1 h after hormone treatment, and this high level of activation was maintained for up to 3 h (Fig. 1, lanes 2 and 3, upper panel). By 24 h, however, the degree of phosphorylation was slightly attenuated (Fig. 1, lane 4, upper panel). Thereafter, a small increment of tyrosine phosphorylation was observed up to 120 h (lanes 5–7, upper panel). Even at 192 h, Stat5A was still tyrosine phosphorylated, although the degree was much less than that at 3 h (Fig. 1, lane 8, upper panel). In the lower panel, Stat5s ran as a doublet, the 95 kDa and 92 kDa, corresponding to Stat5A and Stat5B, respectively. Both Stat5A and Stat5B were strongly tyrosine phosphorylated at 1 h post hormone treatment, and this high level of activation was maintained for up to 3 h (Fig. 1, lanes 2 and 3, lower panel). Subsequently the tyrosine-phosphorylated 92-kDa band (Stat5B) gradually extinguished, whereas the 95-kDa band (Stat5A) was kept tyrosine phosphorylated, albeit weakly. Intriguingly, Western blot analysis revealed that Stat5A and Stat5B expression was augmented from 48 h post adipogenic challenge toward the end of differentiation, and these findings were in line with a previous report (Fig. 1, lanes 5–8, upper and lower panels) (34). We have also confirmed that no Stat5A and B remained in the cell lysates after immunoprecipitation (data not shown). It is noteworthy that the ratio between tyrosine-phosphorylated Stat5s and unphosphorylated Stat5s decreased as cells converted into adipocytes (Fig. 1, lanes 2–8, upper and lower panels). Other Stats, such as Stat1 and Stat3, were not activated during adipogenic conversion of 3T3-L1 cells (data not shown). These results showed that Stat5A and Stat5B were activated upon adipogenic hormone stimulation and that activation of Stat5A was sustained during adipogenic conversion of 3T3-L1 cells.
|
). Dox treatment resulted in a 5-fold increase for wt- and dn-Stat5A in 48 h (Fig. 2A, lanes 1–3). We found that maximal induction of both wt- and dn-Stat5A proteins lasted for 48–72 h. Thereafter, expression of wt- and dn-Stat5A gradually diminished even in the continuous presence of Dox (Fig. 2A, lanes 2–6). Dox had no effect on Stat5 protein expression in the control cells (termed "parent") stably transfected with reverse tetracycline-controlled transactivator (rtTA) alone (data not shown).
|
, Parent, Dox [-], PRL, and FBS). Similarly, challenge with prolactin or FBS stimulated the luciferase activity by 5.2- and 4.4-fold in L1-wt-Stat5 cells, respectively (Fig. 2B, L1-wt-Stat5, Dox [-], PRL, and FBS). Likewise, prolactin and FBS stimulated the luciferase activity by 7.1- and 5.6- fold in L1-dn-Stat5 cells, respectively (Fig. 2B, L1-dn-Stat5, Dox [-], PRL, and FBS). Addition of Dox did not affect the prolactin- and the FBS-induced transcriptional activity in parent cells (Fig. 2B, Parent, Dox [+], PRL, and FBS). When wt-Stat5A expression was induced, both prolactin- and FBS-dependent luciferase activities resulted in a further 2.0-fold increase (Fig. 2B, L1-wt-Stat5, Dox [+], PRL, and FBS). We also found a 3-fold increment in basal luciferase activity by Dox in L1-wt-Stat5 cells (Fig. 2B, L1-wt-Stat5, Dox [-] and [+], None). On the contrary, when dn-Stat5A was induced, both prolactin- and FBS-stimulated luciferase activities were repressed to 44% and 67%, respectively, of their control values (Fig. 2B, L1-dn-Stat5; compare Dox [-] and [+], PRL, and FBS). These results showed that exogenous Stat5A indeed influenced endogenous Stat5 signaling pathways, thus advocating the validity of this system.
Stat5 has been shown to be responsible for cell growth dependent on several cytokines; thus a possible mechanism by which Stat5 affects adipogenesis would be by promoting or inhibiting cell proliferation during clonal expansion (41, 42). To assess this possibility, we compared cell numbers before and after differentiation stimulation (d 0–4), in the presence or absence of Dox in L1-wt-Stat5 and L1-dn-Stat5 cells. As shown in Fig. 2C, there was about a 3-fold increase in cell number after adipogenic challenge, which is normally observed during mitotic clonal expansion of preadipocyte cells (Fig. 2C, compare d 0 and d 3) (43). Proliferation of L1-wt-Stat5 and L1-dn-Stat5 cells was not significantly affected regardless of Dox presence (Fig. 2C). We also found that the doubling time for L1-dn-Stat5 cells was slightly slower than that for L1-wt-Stat5 cells (Fig. 2C); however, no apparent effect of Dox was observed. When cell number was compared in the growing phase, the same results were obtained (data not shown). Because Stat5 controls the expression of several genes required for cell cycle progression in hematopoietic cells (42, 44), we evaluated whether wt- or dn-Stat5A affects the expression of these genes. Northern Blot analysis indicated that the expression levels of c-fos, cyclin A, B, D1, E, p21, and p27 were not influenced by either wt-Stat5A or dn-Stat5A (data not shown). These data suggested that Stat5 activation was not related to mitogenesis of 3T3-L1 cells.
Stat5 Exerts Its Stimulatory Role in Adipogenesis at Initial Stages of Differentiation
We then hypothesized that the expression of wild-type Stat5A would stimulate adipogenesis, whereas dn-Stat5A would inhibit differentiation. This hypothesis was first tested in experiments in which triglyceride storage was monitored as an index of adipose differentiation by Oil Red O staining (Fig. 3A). In the absence of Dox, L1-wt-Stat5 and L1-dn–Stat5 cells exhibited triglyceride accumulation and large, rounded morphology 9 d after exposure to a differentiation cocktail (Fig. 3A, microscopic view, Dox-). Macroscopic analysis also revealed the similar degree of differentiation of both cell lines as judged from Oil Red O staining (Fig. 3A, macroscopic view, Dox-). Thus, each cell line possessed normal differentiation potential. In L1-wt-Stat5 cells, the number of cells accumulating triglycerides increased when Dox was added before exposure to the differentiation cocktail (Fig. 3A, microscopic and macroscopic view, Dox+). On the contrary, ectopic expression of dn-Stat5A decreased the cell population undergoing adipogenesis (Fig. 3A, microscopic and macroscopic view, Dox+). Parent cells showed no difference in the differentiation potential regardless of Dox presence (data not shown). These observations suggested that Stat5 modulated adipogenesis. Because Stat5 was strongly activated at an early stage of adipogenesis (Fig. 1), we have tried to define the time window during which Stat5 exerted its effect on adipogenesis. wt-Stat5A expression was induced at different time points during the adipogenic regimen, and resulting triglyceride contents were compared (Fig. 3B). It must be noted that it takes 48 h to maximally induce exogenous Stat5A with Dox. When Dox was added 48 h before adipogenic hormone challenge, the relative amount of triglycerides was augmented to 1.8-fold over the control in L1-wt-Stat5 cells (Fig. 3B, left panel, compare - and -2 d). Addition of Dox on the same day as adipogenic hormone treatment resulted in a 30% increment of triglyceride contents relative to the control (Fig. 3B, left panel, compare - and 0 d). In contrast, when adipogenic induction preceded Dox treatment by 2 d, little enhancement of triglyceride accumulation was observed (Fig. 3B, left panel, compare - and +2 d). We also assessed the effects of dn-Stat5A on triglyceride loading in the cells by the same protocol (Fig. 3B, middle panel). When Dox was added 48 h before adipogenic hormone challenge, the relative amount of triglycerides was reduced to 67% over the control (Fig. 3B, middle panel, compare - and -2 d). Addition of Dox on the same day as adipogenic hormone treatment resulted in a 21% decrement of triglyceride contents relative to the control (Fig. 3B, middle panel, compare - and 0 d). In contrast, when adipogenic induction preceded Dox treatment by 2 d, little reduction of triglyceride accumulation was observed (Fig. 3B, middle panel, compare - and +2 d). Addition of Dox to parental cells did not alter the triglyceride content, implying that Dox alone had no impact on differentiation (Fig. 3B, right panel). These data demonstrated that Stat5 exerted its proadipogenic effect during early phases of differentiation, which correlated well with its strong activation periods (Figs. 1 and 3B).
|
are the initial factors transcriptionally induced upon adipogenic hormone challenge and are thought to be required for PPAR expression (12, 19). Administration of adipogenic hormones led to the immediate transcriptional activation of C/EBPß and -, as reported (Fig. 4A, lanes 1 and 2 and lanes 6 and 7, C/EBPß and C/EBP) (12). The increased level of C/EBPß and were maintained up to 2 d and then declined by 4 d in both L1-wt-Stat5 and L1-dn-Stat5 cells (Fig. 4A, lanes 3–5, C/EBPß and C/EBP). Addition of Dox caused no change in the expression levels of these factors in both cells (Fig. 4A, compare lanes 1–5 and lanes 6–10, C/EBPß and C/EBP). These data clearly demonstrated that Stat5 had little influence on C/EBPß and - expression. PPAR mRNA was induced very early (1–2 d postconfluence) during the differentiation of both L1-wt-Stat5 and L1-dn-Stat5 cells (Fig. 4A, lanes 1–5, PPAR). Addition of Dox to L1-wt-Stat5 cells robustly stimulated the level of PPAR from d 1 to d 9 (Fig. 4A, compare lanes 2–5 and lanes 7–10). In contrast, addition of Dox to L1-dn-Stat5 cells decreased PPAR from d 1 to d 9 (Fig. 4A, compare lanes 2–5 and lanes 7–10). These data unequivocally defined the role of Stat5 in regulating the expression of PPAR mRNA from the early phase of differentiation. C/EBP mRNA increased as cells differentiate in both L1-wt-Stat5 and L1-dn-Stat5 cells (Fig. 4A, lane 5, C/EBP). Overexpression of wt-Stat5A resulted in earlier induction of C/EBP in L1-wt-Stat5 (Fig. 4A, compare lanes 3–5 and lanes 8–10), whereas induction of dn-Stat5A decreased C/EBP expression in L1-dn-Stat5 cells by d 9 (Fig. 4A, compare lanes 5 and 10). SREBP1c was induced very early in both L1-wt-Stat5 and L1-dn-Stat5 cells (Fig. 4A, lanes 1–5, SREBP1c). Overexpression of wt-Stat5A or dn-Stat5A resulted in a slight increment or decrement of the SREBP1c at d 9, respectively (compare Fig. 4A, lanes 5 and 10, SREBP1c). Addition of Dox to parental cells did not alter the level of C/EBPß, -, PPAR, C/EBP, or SREBP1c, implying that Dox alone had no impact on the expression of these transcription factors (Fig. 4A).
|
expression was further confirmed at the protein level. In parent cells the amount of PPAR increased as the cell differentiated in the absence of Dox (Fig. 4B, parent, lanes 1–5). Addition of Dox to parent cells did not alter the level of PPAR (Fig. 4B, parent, lanes 6–10). However, Dox treatment resulted in enhanced expression of PPAR from d 1 to d 7 in L1-wt-Stat5 cells, whereas a profile of PPAR expression similar to that of parent cells was observed in the absence of Dox (Fig. 4B, WT and parent, compare lanes 1–5 and lanes 6–10). On the other hand, expression of dn-Stat5A repressed the expression of PPAR from d 1 to d 7, whereas L1-dn-Stat5 cells showed identical protein expression profiles as Parent cells in the absence of Dox (Fig. 4B, DN and parent, compare lanes 1–5 and lanes 6–10). These data are in line with the hypothesis that Stat5 regulates the expression of PPAR from the early phase of differentiation and suggest PPAR to be a direct target of Stat5. Intriguingly, a putative Stat5-consensus motif was present in the highly conserved domains of the human and mouse PPAR 2 promoter. Our preliminary data indicated that the mouse PPAR 2 promoter could be activated by Stat5 (Nanbu-Wakao, R., and H. Wakao, unpublished results).
Dn-Stat5A Expression in Adipose Tissue Results in Repression of PPAR
We have generated transgenic mice that express another form of dn-Stat5A in fat tissues with the aP2 enhancer/promoter to unequivocally demonstrate a role of Stat5 in adipogenesis in vivo (45). We used Stat5A with a COOH-terminal transactivation domain deletion as dn-Stat5A (46). The amount of WAT is reduced in the transgenic animals (data not shown). Northern blot analysis revealed that the transgenic animal indeed expressed the truncated form of Stat5A in WAT (Fig. 5). As expected from in vitro data, the expression of PPAR, C/EBP, and SREBP1c was reduced in the transgenic animal. It is noteworthy that dn-Stat5A partially blocked the expression of C/EBPß, whereas that of C/EBP was not influenced. As PPAR and C/EBP regulate expression of the adipocyte-specific gene, aP2 (47, 48), we examined whether dn-Stat5A affected its expression. As can be seen in Fig. 5, aP2 was reduced in the transgenic animal. In addition to aP2, the expression of adiponectin, a fat-derived hormone, was also repressed in the transgenic animal. In toto, these data demonstrated that Stat5 controlled expression of the proadipogenic factors such as PPAR, C/EBP, and SREBP1c as well as of adipocyte-specific genes during adipogenesis.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Particularly, strong activation of Stat5 at very early stages of adipogenesis implied that Stat5 plays some roles during the clonal expansion phase. Although Stat5A and B increased along with differentiation, the physiological relevance of this is yet to be explored (34) (Fig. 1). Because tyrosine phosphorylation of Stat5A was attenuated as differentiation proceeded but not extinguished, it is possible that prolonged activation may be required for maintaining differentiation (Fig. 1). Gene disruption experiments have disclosed that Stat5A and B are mainly responsible for GH and prolactin signaling (50). It is apparent from our results that both prolactin and FBS lead to Stat5 activation (Fig. 2B). Because prolactin, GH, and probably IGF-I, which all activate Stat5 or enhance activation of Stat5, are components of FBS, these hormones might be primary inducers for Stat5 activation (3, 51, 52). Intriguingly, these hormones can stimulate adipogenesis (8, 31). In agreement with these notions, reduction in body weight and in fat mass was observed in prolactin receptor knockout mice and in Stat5A, B double knockout mice (50, 53), although a leaner mouse is not direct evidence of a defect in adipogenesis. Involvement of the Stat5 signaling in adipogenesis is also suggested in 3T3-F442A cells, the differentiation of which is dependent on GH (54).
In this report we have generated reciprocal systems in which either the wt or dn form of Stat5A could be induced to gain insights into the mechanisms of Stat5 action in adipogenesis. Effects of exogenous Stat5s on endogenous signaling pathways have been addressed by transfecting a Stat5-dependent reporter gene into L1-dn-Stat5 and L1-wt-Stat5 cells (Fig. 2B). Results from these experiments further underpin the notion that exogenous Stat5s indeed impinge upon endogenous signaling pathways, resulting in either positive or negative output. Although we observed that Dox caused a 3-fold increase in basal luciferase activity in L1-wt-Stat5 cells (Fig. 2B), this might be due to the presence of insulin in the starvation medium. In fact, insulin is reported to activate Stat5 (55).
The fact that wt- or dn-Stat5A overexpression did not affect cell proliferation implies that Stat5 activation is unrelated to cell cycle progression in 3T3-L1 cells (Fig. 2C). A similar observation is seen in a leukemia cell line in which erythropoietin-dependent hemoglobin synthesis is solely dependent on Stat5 activation but not on cell growth (56). This is in clear contrast to the close relationship of Stat5 activation to the cell cycle progression in several hematopoietic and lymphoid cells (39, 41, 42, 57). Because the effects of dn-Stat5A on the endogenous Stat5 signaling pathways are partial (Fig. 2B), we cannot exclude the possibility that the activity of dn-Stat5A is not sufficient to impinge upon cell proliferation. Our Dox-inducible system also allowed delimiting the time window during which Stat5 effectively influences the adipogenic program. Stat5 exerted its maximal effect on triglyceride accumulation when ectopically expressed simultaneously to the adipogenic hormone treatment. This indicates that precocious activation of this factor is important for induction of PPAR and subsequent cell fate (Figs. 1 and 3B).
An earlier study has suggested that expression of PPAR is mediated by C/EBPß and/or - (19). However, C/EBPß and - double-knockout mice experiments indicate the presence of other pathways leading to PPAR induction (15). In accordance with this, our present study demonstrated that Stat5 regulates, at least in part, the expression of PPAR and C/EBP, without affecting C/EBPß and (Fig. 4A). Indeed, activation of Stat5s precedes expression of these factors (Fig. 1). Obviously, there are caveats to these conclusions based solely on studies with in vitro data. However, data from aP2-driven dn-Stat5A transgenic mice further reinforce in vitro results (Figs. 4A and 5). This allows us to conclude that Stat5 plays a role in adipogenesis both in vitro and in vivo. Although the expression of Stat5A affected PPAR in a C/EBPß- and C/EBP -independent manner in 3T3-L1 cells, C/EBPß was slightly reduced in transgenic mice (Fig. 5). This may be due to a different potential of two types of the dn forms employed in the study. It is possible that the C-terminal truncated form of Stat5A exerts more potent dn effect than that of Y694F mutant. Alternatively, there may be another signaling pathway interfering with the expression of C/EBPß in vivo. PPAR gene ablation resulted in decrease of the many adipocyte-related genes including aP2 and C/EBP. (22, 23). Given that Stat5 regulates the expression of PPAR, it is not surprising that expression of the PPAR-controlled genes are repressed by dn-Stat5A both in vitro and in vivo (Figs. 4A and 5). Analysis of other adipocyte-specific gene expression will reveal whether their expression is dependent on Stat5-PPAR or not.
Taking all into account, we would like to present a working model by which Stat5 stimulates adipogenesis (Fig. 6). Stat5 is activated following adipogenic hormone challenge and stimulates, either directly or through transcription of as-yet-unidentified genes, the expression of C/EBP and PPAR. These proadipogenic factors, in turn, induce a series of adipocyte-related genes such as aP2 and adiponectin. Enhanced expression of these genes and of lipogenic enzymes results in increased triglyceride accumulation within the cells. Therefore, Stat5 appears to be an initial transcription factor involved in adipogenesis and constitutes a direct link between external signals to the intracellular transcriptional cascade leading to execution of the differentiation program.
|
, DN). Inducible expression of wt-Stat5A alone in the absence of MIX and DEX was not sufficient to initiate adipogenesis (Nanbu-Wakao, R., and H. Wakao, unpublished results). Optimal differentiation of 3T3-L1 cells is dependent on the presence of FBS, MIX, DEX, and insulin. Deprivation of one or more components results in attenuation of the differentiation potential of the cells (14). This indicates that there must be several factors simultaneously activated by these hormones. As mentioned earlier, C/EBPß and - are transcriptionally induced by MIX and DEX, respectively (12). In contrast, Stat5 is activated posttranslationally, as demonstrated in this study. A recent report has also shown that activation of cAMP response element binding protein (CREB) is necessary and sufficient to induce adipogenesis (58). A similar activation profile between CREB and Stat5 during adipocyte differentiation implies that these pathways function in parallel to promote adipogenesis. While the contribution of CREB explains, at least in part, the requirement of cAMP in adipogenesis, Stat5 may play as a mediator of FBS that optimizes differentiation. Another pathway to be considered is Wnt signaling, which is shown to inhibit adipocyte differentiation (59). Because Stat5 and Wnt exhibit an opposite effect on the expression of PPAR and C/EBP as well as on adipogenesis, it will be interesting to examine whether there is any cross-talk between Wnt and Stat5 signaling. A previous report has indicated that glucocorticoid induces PPAR expression in a dose-dependent fashion (19). Given that Stat5 is responsible for the induction of PPAR, a direct interaction between Stat5 and the glucocorticoid receptor can be envisaged for this induction. In fact, a synergistic output elicited from such an interaction has been reported (60). In this regard, involvement of glucocorticoid receptor and Stat5 may explain, at least in part, the requirement of DEX for PPAR and C/EBP expression as well as for maximal differentiation of 3T3-L1 cells (14). Finally, it is noteworthy that not only is Stat5 involved in adipogenesis, but it also plays a mandatory role in mammopoiesis, particularly in alveoli formation in vivo (61, 62). Mammary epithelial cell differentiation in vitro is reminiscent of adipocyte differentiation in that both systems require the cells to be confluent before treatment with hormones comprising a similar but not identical cocktail (63, 64). In both cases, Stat5 is immediately activated upon hormone challenge (present study and Ref. 65). These observations indicate that Stat5 triggers cell differentiation in concert with other intracellular signaling molecules. Intriguingly, it is shown that activation of the Jak-Stat pathway is required for the self-renewal of stem cells in Drosophila (66, 67). Identification of Jak-Stat targeting genes will shed light on the molecular mechanisms underlying these phenomena. In the absence of a Stat5-specific inhibitor, our Dox-controlled expression system will allow identification of Stat5-regulated genes in adipogenesis. Molecular characterization of these genes will further our knowledge of adipocyte differentiation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Plasmids and cDNA
The cDNAs for C/EBP, C/EBPß, C/EBP, PPAR2, and aP2 were isolated from a 3T3-L1 preadipocyte library in pZL1 constructed at 9 d after the induction of differentiation by using the SuperScriptII 
system (Invitrogen, San Diego, CA) as described previously (31). Other cDNAs were prepared by RT-PCR as follows. First-strand cDNA was prepared from 3T3-L1 total RNA by using an RNA PCR (AMV) Ver2.1 kit (Takara, Ohtsu, Japan). The cDNAs were then used as templates in PCRs with the indicated primer sets. The resulting PCR products were subcloned into the pGEM-7Zf(+) vector (Promega Corp., Madison, WI). The primer sequences used in this study were as follows. ADD1/SREBP-1c, 5'-GACCCTGGTGAGTGGNGGAAC-3' (sense), 5'-GAGGCCAANGGGTTGCAGGNC-3' (antisense). Adiponectin, 5'-ATGCTACTGTTGCAAGCTCTCCT-3' (sense), 5'-TCAGTTGGTATCATGGTAGAGAAG-3' (antisense). All cDNA sequences were verified by DNA sequencing (ABI 377 DNA sequencer; Applied Biosystems, Foster City, CA).
Cell Culture and Induction Of Differentiation
3T3-L1 preadipocytes (NIHS cell bank, Japan, catalogue no. JCRB9014) were maintained in growth medium consisting of DMEM (Nisseiken, Kyoto, Japan) containing 10% normal calf serum. To differentiate 3T3-L1 cells, cells were grown to confluence and left for 2 d. Cells were then exposed to fresh differentiation medium consisting of DMEM, 10% FBS, 1 µM DEX, 0.5 mM MIX, and 10 µg/ml of insulin for 48 h. Thereafter cells were refed every other day with DMEM containing 10% FBS and 2.5 µg/ml of insulin. Differentiation was confirmed by Oil Red O staining of lipid vesicles. Triglycerides from differentiated cells were extracted with isopropanol and quantitated with a Triglycerides Test kit (Wako Chemical Co.). Cells were counted with a hemocytoanalyzer CDA-500 (Sysmex, Kobe, Japan).
Construction of the Dox-Inducible Expression Systems for Stat5
To establish inducible Stat5A expression cell lines, we used the rtTA system (36), in which transcription of a target gene is initiated by Dox treatment. 3T3-L1 cells were first transfected with pUD172–2neo (rtTA), and clones were selected in the presence of 0.4 mg/ml of G418. G418-resistant clones were assessed for their rtTA-dependent reporter gene expression by supertransfecting pUHC13–3. A clone that showed the greatest Dox-dependent luciferase activity was isolated, termed parent, and used for further construction. The expression vectors for wt-Stat5A and for dn-Stat5A (Y694F) were constructed by ligating an EcoRI-NotI fragment containing ovine Stat5 or Stat5 (Y694F), which corresponds to mouse Stat5A, into the modified multicloning site of pUHD10–3. These vectors were transfected together with pPUR (CLONTECH Laboratories, Inc., Palo Alto, CA) into parent cells. Clones were selected in the presence of 3 µg/ml of puromycin and then monitored for their Dox-dependent induction of Stat5 by Western blot analysis.
RNA Analysis
Total RNA, isolated according to the method of Chomczynski and Sacchi (68), was fractionated on a 1% agarose gel with a Northern Max-Gly kit (Ambion, Inc., Austin, TX) and transferred to a nylon membrane. rRNA was stained on the filters with methylene blue to assess RNA loading and transfer efficiency. Digoxigenin-labeled antisense RNA probes were synthesized with SP6/T7 RNA polymerase from linearized plasmids. Hybridization and washing were performed according to the protocol provided by the supplier (Roche Molecular Biology, Basel, Switzerland).
Immunoprecipitation and Immunoblotting
Whole-cell extracts from 3T3-L1 cells were prepared at the indicated time points with lysis buffer (50 mM HEPES-NaOH, pH 7.8; 150 mM NaCl; 1% Triton X-100; 50 mM NaF; 1 mM Na3VO4; 30 mM sodium pyrophosphate; 1 mM Pefablock; 0.1 mg/ml leupeptin). Cells were lysed on ice for 60 min, and insoluble materials were pelleted by centrifugation. Protein concentration was determined with a BCA protein assay kit (Pierce Chemical Co., Rockford, IL). Clarified lysates were incubated with antiserum against Stat5A and B proteins. After an overnight incubation with protein A-coupled beads, the immunoprecipitates were electrophoresed on a 10% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane. Immunoprecipitated proteins were blotted with 4G10 and then reblotted with the respective antibody. Anti-Stat5A (L-20), Stat5B (C-17), and anti-PPAR (H-100) were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Dox-Dependent Transcriptional Activation Assay
Forty-eight hours before transfection, L1-wt-Stat5 or L1-dn-Stat5 cells were treated with Dox or left untreated. At 50–60% confluence, cells were transfected with a Stat5-reporter plasmid together with the renilla luciferase expression vector (pRL-cytomegalovirus) to control the transfection efficiency (Promega Corp.) (40). After DNA removal, cells were serum starved for 16 h in OPTIMEM (Life Technologies, Inc., Gaithersburg, MD) and then left untreated or challenged with prolactin or FBS. Cells were lysed, and both firefly and renilla luciferase activities were measured.
Construction of aP2-Driven dn-Stat5 Transgenic Mice
The 5.4-kb aP2 promoter/enhancer was cloned by long and accurate-PCR (Takara Co., Ohtsu, Japan) (45). The C-terminal deletion mutant of Stat5A was digested with EcoRI and XbaI from pXM vector (a gift from Dr. B. Groner). After blunting, the truncated Stat5A was cloned into a vector containing the rabbit ß-globin poly (A). Then the truncated-Stat5A-rabbit ß-globin poly (A) was cloned into XbaI-ClaI sites of pBlue Script IIKS(+) (Stratagene, La Jolla, CA). After digestion with XbaI and blunting, a phosphorylated 5.4-kb aP2 promoter/enhancer was inserted. The resulting 9.1-kb aP2 truncated-Stat5A rabbit ß-globin poly (A) construct was cleaved with SalI and NotI and purified on agarose gel. The integrity of the plasmid was confirmed by DNA sequencing of the ligation junctions. The purified DNA was microinjected into more than 200 fertilized eggs of C57BL/6CrSlc (SLC Co., Hamamatsu, Japan). Among the 18 offspring, 5 had integrated the transgene as determined by PCR on tail DNA with transgene-specific primers (5'-ACATACAGGGTCTGGTCATG-3' and 5'-CGCCTTCTTCTGCAGCTCGT-3', giving 422 bp). Mice with high levels of transgene expression were bred to C57BL/6J mice, and several independent lines of transgenic mice were established. Mice were housed in colony cages and maintained on a 12-h light, 12-h dark cycle. Data from representative mice were shown.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: C/EBP, CCAAT-enhancer binding protein; CREB, cAMP response element binding protein; DEX, dexamethasone; dn, dominant negative; Dox, doxycycline; FBS, fetal bovine serum; MIX, methylisobutylxanthine; PPAR, peroxisome proliferator-activated receptor; rtTA, reverse tetracycline-controlled transactivator; Stat, signal transducer and activator of transcription; WAT, white adipose tissue; wt, wild type.
Received for publication May 14, 2001. Accepted for publication March 6, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and TNF receptors in obesity and insulin resistance. J Intern Med 245:621–625[CrossRef][Medline]
and stimulates adipogenesis. Genes Dev 9:2350–2363 gene. EMBO J 16:7432–7443[CrossRef][Medline]
is sufficient to initiate the 3T3-L1 adipocyte differentiation program. Proc Natl Acad Sci USA 91:8757–8761 promotes the adipogenic program in a variety of mouse fibroblastic cells. Genes Dev 8:1654–1663 knockout mice. Science 269:1108–1112, and glucocorticoids. Mol Cell Biol 16:4128–4136[Abstract]
2, a lipid-activated transcription factor. Cell 79:1147–1156[CrossRef][Medline]
and C/EBP. Proc Natl Acad Sci USA 92:9856–9860 is required for the differentiation of adipose tissue in vivo and in vitro. Mol Cell 4:611–617[CrossRef][Medline]
mediates high-fat diet-induced adipocyte hypertrophy and insulin resistance. Mol Cell 4:597–609[CrossRef][Medline]
is required for placental, cardiac, and adipose tissue development. Mol Cell 4:585–595[CrossRef][Medline]
expression by adipocyte differentiation and determination factor 1/sterol regulatory element binding protein 1: implications for adipocyte differentiation and metabolism. Mol Cell Biol 19:5495–5503 through the production of endogenous ligand. Proc Natl Acad Sci USA 95:4333–4337 (PPAR ) messenger RNA expression and stimulates adipogenic conversion of NIH-3T3 cells. Mol Endocrinol 14:307–316 ligand-dependent induction of STAT1, STAT5A, and STAT5B during adipogenesis. Biochem Biophys Res Commun 262:216–222[CrossRef][Medline]
disrupts mitotic clonal expansion and regulation of retinoblastoma proteins p130 and p107 during 3T3-L1 adipocyte differentiation. Biochem Biophys Res Commun 247:373–378[CrossRef][Medline]
2: tissue-specific regulator of an adipocyte enhancer. Genes Dev 8:1224–12342 regulates adipose expression of the phosphoenolpyruvate carboxykinase gene. Mol Cell Biol 15:351–357[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  U. A. White, A. A. Coulter, T. K. Miles, and J. M. Stephens The STAT5A-Mediated Induction of Pyruvate Dehydrogenase Kinase 4 Expression by Prolactin or Growth Hormone in Adipocytes Diabetes, June 1, 2007; 56(6): 1623 - 1629. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kang, C. N. Bennett, I. Gerin, L. A. Rapp, K. D. Hankenson, and O. A. MacDougald Wnt Signaling Stimulates Osteoblastogenesis of Mesenchymal Precursors by Suppressing CCAAT/Enhancer-binding Protein {alpha} and Peroxisome Proliferator-activated Receptor {gamma} J. Biol. Chem., May 11, 2007; 282(19): 14515 - 14524. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. E. Floyd, B. M. Segura, F. He, and J. M. Stephens Degradation of STAT5 proteins in 3T3-L1 adipocytes is induced by TNF-{alpha} and cycloheximide in a manner independent of STAT5A activation Am J Physiol Endocrinol Metab, February 1, 2007; 292(2): E461 - E468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kawai, N. Namba, S. Mushiake, Y. Etani, R. Nishimura, M. Makishima, and K. Ozono Growth hormone stimulates adipogenesis of 3T3-L1 cells through activation of the Stat5A/5B-PPAR{gamma} pathway J. Mol. Endocrinol., January 1, 2007; 38(1): 19 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Miyaoka, M. Tanaka, T. Naiki, and A. Miyajima Oncostatin M Inhibits Adipogenesis through the RAS/ERK and STAT5 Signaling Pathways J. Biol. Chem., December 8, 2006; 281(49): 37913 - 37920. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Viengchareun, H Bouzinba-Segard, J-P Laigneau, M-C Zennaro, P A Kelly, A Bado, M Lombes, and N Binart Prolactin potentiates insulin-stimulated leptin expression and release from differentiated brown adipocytes J. Mol. Endocrinol., December 1, 2004; 33(3): 679 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. A. Noon, A. J. L. Clark, and P. J. King A Peroxisome Proliferator-response Element in the Murine mc2-r Promoter Regulates Its Transcriptional Activation during Differentiation of 3T3-L1 Adipocytes J. Biol. Chem., May 28, 2004; 279(22): 22803 - 22808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Shang and M. J. Waters Constitutively Active Signal Transducer and Activator of Transcription 5 Can Replace the Requirement for Growth Hormone in Adipogenesis of 3T3-F442A Preadipocytes Mol. Endocrinol., December 1, 2003; 17(12): 2494 - 2508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Shipley and D. J. Waxman Down-Regulation of STAT5b Transcriptional Activity by Ligand-Activated Peroxisome Proliferator-Activated Receptor (PPAR) {alpha} and PPAR{gamma} Mol. Pharmacol., August 1, 2003; 64(2): 355 - 364. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Hurst and D. J. Waxman Activation of PPAR{alpha} and PPAR{gamma} by Environmental Phthalate Monoesters Toxicol. Sci., August 1, 2003; 74(2): 297 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. E. Floyd and J. M. Stephens STAT5A Promotes Adipogenesis in Nonprecursor Cells and Associates With the Glucocorticoid Receptor During Adipocyte Differentiation Diabetes, February 1, 2003; 52(2): 308 - 314. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
