Role of MAPK phosphatase-1 (MKP-1) in adipocyte differentiation.

Both time-dependent modulation of intracellular signaling molecules and sequential induction of transcriptional regulators are essential for the differentiation of preadipocytes into adipocytes. We have now shown that the activity, but not the abundance, of p42/p44 mitogen-activated protein kinase (MAPK) is down-regulated during adipocyte differentiation. This decrease in p42/p44 MAPK activity does not appear to be a direct effect of hormonal inducers of differentiation but rather represents a characteristic event of adipocyte differentiation that is achieved through a persistent change in intracellular signaling. Although the phosphorylation or abundance of MEK, an upstream kinase for p42/p44 MAPK, was not altered during differentiation, the abundance of MAPK phosphatase-1 (MKP-1), a negative regulator of p42/p44 MAPK, was increased with a time course similar to that of the down-regulation of p42/p44 MAPK activity. Ectopic expression of MKP-1 in preadipocytes reduced and depletion of endogenous MKP-1 in mature adipocytes increased the activity of p42/p44 MAPK. Prevention of the up-regulation of MKP-1 abundance in preadipocytes by expression of Mkp-1 antisense RNA resulted in persistence of p42/p44 MAPK activation and blocked differentiation, effects that were reversed by the MEK inhibitor PD98059. These results suggest that MKP-1 plays an essential role in adipocyte differentiation through down-regulation of p42/p44 MAPK activity.

The amount of adipose tissue, which is altered in various physiological and pathological conditions, is an important determinant of energy homeostasis in living animals (1). Given that mature adipocytes do not undergo cell division, the number of adipocytes in vivo is thought to increase as a result of the proliferation of preadipocytes and their subsequent differentiation into mature adipocytes. Studies with cultured cells, such as mouse 3T3-L1 and 3T3-F442A cell lines, have shed light on the mechanisms of growth and differentiation of preadipocytes (2,3). Differentiation of these cells occurs in two discrete steps. The cells first undergo several rounds of mitosis, known as clonal expansion, after which they become quiescent again, express adipocyte-specific proteins, and acquire biochemical and morphological characteristics of mature adipocytes (2,3). Both the proliferation and differentiation of preadipocytes are characterized by marked changes in the pattern of gene expression that are achieved by the sequential induction of transcription factors. Exposure of preadipocytes to appropriate hormonal inducers of differentiation thus results in early and transient expression of the ␤ and ␦ isoforms of the CCAAT/ enhancer-binding protein (C/EBP), 1 which in turn appear to contribute to the proliferation of these cells as well as to the subsequent increase in the expression of C/EBP␣ and peroxisome proliferator-activated receptor ␥ (PPAR␥) (4,5). The latter two proteins then mediate the activation of a variety of adipocyte-specific genes (2)(3)(4).
Both extracellular and intracellular signaling molecules also play key roles in the differentiation of preadipocytes. Downregulation of Pref-1 (6) and Wnt-10 (7), secreted proteins that are abundant in preadipocytes, is thus important for differentiation of these cells. Fibroblast growth factor-10, which is expressed transiently during adipocyte differentiation, is also required to maintain the expression of C/EBP␤ until a specific stage of differentiation (8). With regard to intracellular signaling molecules, phosphoinositide 3-kinase, which is specifically activated at a late phase of differentiation, contributes to the expression of proteins that characterize mature adipocytes (9). A preferential inhibitor of p38 mitogen-activated protein kinase (MAPK) prevents the differentiation of preadipocytes (10), and overexpression of MAPK kinase 6, an activator of p38 MAPK, triggers the spontaneous differentiation of these cells (11). Introduction of an active form of the protein kinase Akt also induces spontaneous differentiation of preadipocytes (12). Although these various observations indicate the importance of specific intracellular signaling in the differentiation of preadipocytes, the mechanisms by which such signaling is regulated remain largely unknown.
We now show that the activity of the p42 and p44 isoforms of MAPK is down-regulated during differentiation of preadipocytes. We provide evidence that the abundance of MAPK phosphatase (MKP)-1 increases during the differentiation of these cells and that this protein plays an essential role in the differentiation program by down-regulating the activity of p42/p44 MAPK.
Cell Culture and Staining-3T3-L1 and 3T3-F442A preadipocytes were maintained as described previously (9). The differentiation of 3T3-L1 preadipocytes into adipocytes was induced by incubation of confluent cells cultured in the presence of 10% fetal bovine serum (FBS) first for 2 days with insulin (5 g/ml), 0.25 M dexamethasone, and 0.5 mM isobutylmethylxanthine (IBMX) and then with insulin (5 g/ml) alone for an additional 2 days. Alternatively, the cells were cultured in the presence of 10% FBS and treated with IBMX (0.5 mM) and dexamethasone (0.25 M) for only 2 days. Differentiation of 3T3-F442A preadipocytes was induced by exposing confluent cells to troglitazone (10 M) in the presence of 10% FBS for 2 days. After incubation with these reagents, the 3T3-L1 and 3T3-F442A cells were returned to normal culture medium, which was replenished every other day. Oil red O staining (8) and staining for ␤-galactosidase activity (15) were performed as described.
Fractionation of Adipose Tissue-For preparation of stromal-vascular and mature adipocyte fractions of epididymal adipose tissue removed from 10-week-old C57BL/6 mice, the tissue was washed with and minced in Dulbecco's modified Eagle's medium (DMEM) and then incubated on a shaking platform for 20 min at 37°C with the same medium containing collagenase (0.5 mg/ml). The mixture was then passed through a nylon filter (pore size, 250 m) to remove undigested material, and the filtrate was centrifuged for 10 min at 200 ϫ g and at 4°C. Floating cells (the mature adipocyte fraction) and the pellet (the stromal-vascular fraction) were recovered, washed three times with DMEM, and then lysed and subjected to immunoblot analysis.
Expression Plasmids and Adenoviral Vectors-A mammalian expression vector encoding MKP-1 (pcDNA3.1/MKP-1) was constructed by subcloning a cDNA for the human MKP-1 ortholog (CL-100) (16), kindly provided by S. Keyse (Molecular Pharmacology Unit, Ninewells Hospital, Dundee, UK), into pcDNA3.1 (Invitrogen). To construct an adenoviral vector that encodes Mkp-1 antisense RNA, we subcloned a mouse Mkp-1 cDNA containing both the entire coding region and 157 bp of the 5Ј-untranslated region (17), kindly provided by N. Tonks (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), in the antisense orientation into pAxCAwt; the resulting construct was introduced into 293 cells together with DNA-TPC by transfection with the use of an adenovirus expression kit (Takara, Tokyo, Japan) as described previously (9). An adenoviral vector encoding ␤-galactosidase was kindly provided by I. Saito (Tokyo University, Japan). For adenovirus-mediated gene transfer into 3T3-L1 preadipocytes, cells cultured to subconfluency were infected with viruses at a multiplicity of infection of 10 plaque-forming units/cell. Two days after infection, differentiation was induced as described above. Alternatively, 3T3-L1 adipocytes at 6 days after the induction of differentiation were infected with viruses at a multiplicity of infection of 30 plaque-forming units/cell; the cells were subjected to immunoblot analysis 2 days after infection.
Electroporation-For transfection of 3T3-L1 preadipocytes with pcDNA3.1/MKP-1 or with an expression vector for ␤-galactosidase (SR␣/LacZ), the cells were detached from culture dishes by incubation with phosphate-buffered saline containing 0.25% trypsin and collagenase (0.5 mg/ml). Approximately 5 ϫ 10 6 cells were mixed with 5 g of plasmid in the solution provided with a cell line Nucleofector kit V (Amaxa, Cologne, Germany), and the plasmid was then introduced into the cells by electroporation with the use of a Nucleofector (Amaxa) instrument according to the T-20 program.

RESULTS
Down-regulation of p42/p44 MAPK Activity during Adipocyte Differentiation-3T3-L1 preadipocytes, a well characterized model for studying adipocyte differentiation, were induced to differentiate by incubation with insulin, dexamethasone, and IBMX for 2 days and then with insulin alone for an additional 2 days. At various times before, during, and after such treatment, the cells were harvested, and the activity of p42/p44 MAPK was assayed in immunoprecipitates prepared with antibodies that recognize both p42 and p44 isoforms of this kinase. The substantial p42/p44 MAPK activity that was apparent in preadipocytes before exposure to the inducers of differentiation had decreased by 4 days after the onset of differentiation induction (Fig. 1a). At 8 days postinduction, when Ͼ90% of the cells had acquired morphological and biochemical characteristics of mature adipocytes (data not shown), the activity of p42/p44 MAPK was Ͻ20% of that observed in the untreated cells. Immunoblot analysis of cell lysates with antibodies specific for phosphorylated p42/p44 MAPK revealed that the extent of phosphorylation of the isozymes, which reflects their kinase activity, decreased in parallel with the kinase activity (Fig. 1a); the abundance of the kinases was not affected by the induction of differentiation, consistent with previous observations (10). In contrast, phosphorylation of p38 MAPK was not affected by the inducers of differentiation but was increased by treatment of the cells with anisomycin, an activator of this isozyme (Fig. 1b).
Insulin activates p42/p44 MAPK in various cell types. However, the activity of p42/p44 MAPK in 3T3-L1 preadipocytes not exposed to insulin or the other inducers of differentiation was higher than that in cells at relatively late stages of differentiation induced by these agents (Fig. 1a). Induction of the differentiation of 3T3-L1 preadipocytes by exposure to dexamethasone and IBMX alone (without insulin) for 2 days resulted in the accumulation of lipid droplets in ϳ60% of the cells within 8 days after the onset of treatment (data not shown). Downregulation of the phosphorylation of p42/p44 MAPK was also observed during the differentiation process (Fig. 1c). Moreover, Total lysates prepared from cells harvested before (time 0) or at the indicated times after exposure to the differentiation inducers were subjected to immunoprecipitation with antibodies to p42/p44 MAPK, and the resulting precipitates were assayed for kinase activity with myelin basic protein (MBP) as substrate (a, upper panel). Alternatively, the cell lysates were subjected directly to immunoblot analysis with antibodies to phosphorylated p42/p44 MAPK, total p42/p44 MAPK, phosphorylated p38 MAPK, or total p38 MAPK, as indicated. b, lysates prepared from 3T3-L1 preadipocytes that had been incubated for 10 min with or without 10 M anisomycin were also subjected to immunoblot analysis. All data are representative of at least three independent experiments. when 3T3-F442A cells, a preadipocyte cell line closely related to 3T3-L1 cells (2), were induced to differentiate by incubation for 2 days with troglitazone, a synthetic agonist for PPAR␥ (18), phosphorylation of p42/p44 MAPK was decreased with a time course similar to that observed during the differentiation of 3T3-L1 cells (Fig. 1d). These results suggested that the downregulation of p42/p44 MAPK activity during the differentiation of preadipocytes is dependent neither on the specific hormonal inducers nor on the specific cell line but rather reflects an intrinsic modification of intracellular signaling characteristic of the differentiation program.
To verify this hypothesis, we deprived either 3T3-L1 preadipocytes or fully differentiated adipocytes (8 days after induction) of serum for 16 h, stimulated the cells by reexposure to serum for 10 min, and then assayed cell lysates for the phosphorylation of p42/p44 MAPK and of Akt, both of which are known to be activated by serum. Serum elicited a marked increase in the extent of phosphorylation of both p42/p44 MAPK and Akt in the nondifferentiated cells (Fig. 2a). However, the effect of serum on the phosphorylation of p42/p44 MAPK was greatly reduced in the differentiated adipocytes compared with that in the nondifferentiated cells. In contrast, the amount of phosphorylated Akt in the differentiated adipocytes was greater than that in the nondifferentiated cells in the absence or presence of serum, probably reflecting the increase in the abundance of Akt that occurred during differentiation. These results were thus consistent with the notion that the down-regulation of p42/p44 MAPK activity during differentiation of preadipocytes is intrinsic to the differentiation program.
We also investigated whether this biochemical change observed in cultured adipocytes also occurs in adipose tissue of living animals. Mouse epididymal adipose tissue was fractionated into the mature adipocyte fraction and the stromal-vascular fraction, the major cell population of which is thought to be preadipocytes, and extracts of both fractions were then sub-jected to immunoblot analysis. As expected, adipocyte marker proteins such as C/EBP␣ and aP2 were detected in the mature adipocyte fraction but not in the stromal-vascular fraction, whereas the abundance of C/EBP␤, which is induced early during the differentiation of cultured preadipocytes, was higher in the stromal-vascular fraction (Fig. 2b). Although the amount of p42/p44 MAPK did not differ substantially between the two fractions, the extent of p42/p44 MAPK phosphorylation was much greater in the stromal-vascular fraction, suggesting that the decrease in p42/p44 MAPK activity apparent during adipocyte differentiation in culture also occurs in adipose tissue of living animals.
Up-regulation of MKP-1 during Adipocyte Differentiation-The activity of p42/p44 MAPK is regulated both by the upstream kinase MEK, which phosphorylates both threonine and tyrosine residues of p42/p44 MAPK, and by a dual specificity phosphatase, which dephosphorylates the residues of p42/p44 MAPK phosphorylated by MEK (19). Although phosphorylated MEK was readily detected in 3T3-L1 preadipocytes that had been stimulated with serum for 10 min, it was detected in cells treated with inducers of differentiation only at low levels revealed by prolonged exposure of immunoblots (Fig. 3a). However, neither the amount of phosphorylated MEK nor that of total MEK changed during differentiation. Moreover, the extent of serum-induced phosphorylation of MEK was similar in nondifferentiated preadipocytes and in differentiated adipo-  proteins (a, c, and d). a, nondifferentiated 3T3-L1 preadipocytes were also deprived of serum for 16 h and then incubated in the absence or presence of 10% FBS for 10 min. Total RNA was also prepared from 3T3-L1 cells during differentiation and was subjected to Northern blot analysis with 32 P-labeled cDNA probes specific for mouse Mkp-1 or mouse Ppar␥ mRNAs (b); the positions of 28 S and 18 S rRNA on the ethidium bromide-stained gel are also shown. The amounts of Ppar␥ mRNA (b) and protein (c) were monitored as markers of adipocyte differentiation. All data are representative of at least three independent experiments. cytes (Fig. 2a). These results indicate that the down-regulation of p42/p44 MAPK activity during adipocyte differentiation is not attributable to a change in MEK signaling. In contrast, the amount of mRNA for Mkp-1, a member of the dual specificity phosphatase family (19), was increased during differentiation of 3T3-L1 cells induced by insulin, dexamethasone, and IBMX with a time course similar to that of the decrease in p42/p44 MAPK activity (Fig. 3b). Consistent with this result, the abundance of MKP-1 protein also increased during the differentiation of 3T3-L1 preadipocytes, whereas that of the dual specificity phosphatases MKP-2 and MKP-3 (19) remained largely unchanged (Fig. 3c). Moreover, a similar increase in the abundance of MKP-1 was observed in 3T3-L1 cells induced to differentiate with IBMX plus dexamethasone as well as in 3T3-F442A cells induced to differentiate with troglitazone (Fig. 3d). The amount of MKP-1 protein was also greater in the mature adipocyte fraction than in the stromal-vascular fraction of mouse epididymal adipose tissue (Fig. 2b). These results indicate that the increase in the abundance of MKP-1 is also a characteristic feature of adipocyte differentiation.
MKP-1 Expression Determines p42/p44 MAPK Activity during Adipocyte Differentiation-We next investigated whether the increase in the abundance of MKP-1 is related to the down-regulation of p42/p44 MAPK activity during differentiation of preadipocytes. For these experiments, we used electroporation to introduce exogenous genes into 3T3-L1 preadipocytes. Transfection with an expression vector for ␤-galactosidase revealed that Ͼ70% of the cells expressed the exogenous gene (Fig. 4a), indicative of a high efficiency of transfection with this method. The amount of MKP-1 in 3T3-L1 preadipocytes transfected with a vector for this protein was similar to that in nontransfected differentiated adipocytes (Fig.  4b). Transfection with the MKP-1 vector did not affect the abundance of other proteins such as CREB, a transcription factor involved in adipocyte differentiation (2). The seruminduced phosphorylation of p42/p44 MAPK, however, was greatly diminished in preadipocytes transfected with the MKP-1 vector compared with that apparent in cells transfected with the corresponding empty vector (Fig. 4c), consistent with the notion that the change in the abundance of MKP-1 is a major determinant of that in the activity of p42/p44 MAPK during differentiation of 3T3-L1 cells.
To establish more directly a causal link between MKP-1 expression and p42/p44 MAPK activity during adipocyte differentiation, we next examined the effect of depletion of endogenous MKP-1 on p42/p44 MAPK activity. Given that transfection of fibroblasts with a vector encoding Mkp-1 antisense RNA was shown to prevent the expression of Mkp-1 (20), we constructed an adenoviral vector (AxCAMkp-1-1AS) that contains Mkp-1 cDNA in the antisense orientation. As expected, infection of 3T3-L1 differentiated adipocytes with AxCAMkp-1-1AS, but not with a control adenovirus encoding ␤-galactosidase (AxCAlacZ), resulted in a marked reduction in the amount of Mkp-1 without an effect on that of CREB (Fig. 4d). Conversely, the extent of seruminduced phosphorylation of p42/p44 MAPK was greater in the differentiated cells infected with AxCAMkp-1AS than in those infected with the control adenovirus (Fig. 4e). These results thus support the conclusion that the increase in the abundance of MKP-1 is a major contributor to the decrease in p42/p44 MAPK activity during adipocyte differentiation.
Up-regulation of MKP-1 Expression Is Required for Adipocyte Differentiation-We next investigated the physiological importance of the up-regulation of MKP-1 expression during adipocyte differentiation. 3T3-L1 preadipocytes were infected with AxCAMkp-1AS or with the control virus AxCAlacZ and were subsequently exposed to hormonal inducers of differentiation (insulin, dexamethasone, and IBMX). Eight days after exposure to the inducers of differentiation, the abundance of MKP-1 in cells infected with AxCAMkp-1AS was ϳ10% that in cells infected with the control virus (Fig. 5a). Consistent with this observation, the differentiation-associated down-regulation of p42/p44 MAPK activity apparent in the cells infected with the control virus did not occur in the cells infected with AxCAMkp-1AS (Fig. 5b). Furthermore, although the cells infected with 3T3-L1 preadipocytes that had not been exposed to inducers of differentiation (time 0) were transfected (or not) with an expression vector for MKP-1 (pcDNA3.1/MKP-1) or with the corresponding empty vector. The cells were lysed 48 h after transfection and subjected to immunoblot analysis with antibodies to MKP-1 or to CREB. Lysates prepared from nontransfected 3T3-L1 mature adipocytes 8 days after exposure to insulin, dexamethasone, and IBMX were also subjected to immunoblot analysis. c, effect of MKP-1 overexpression on the phosphorylation of p42/p44 MAPK. 3T3-L1 preadipocytes transfected as in b were deprived of serum for 16 h and then incubated in the absence or presence of 10% FBS for 10 min, lysed, and subjected to immunoblot analysis with antibodies to phosphorylated p42/p44 MAPK or to total p42/p44 MAPK. d, depletion of MKP-1 by expression of Mkp-1 antisense RNA. 3T3-L1 mature adipocytes 6 days after exposure to insulin, dexamethasone, and IBMX were infected (or not) with an adenoviral vector encoding either ␤-galactosidase (AxCA-lacZ) or Mkp-1 antisense RNA (AxCAMkp-1AS). The cells were lysed 48 h after infection and subjected to immunoblot analysis with antibodies to MKP-1 or to CREB. Lysates of nontransfected 3T3-L1 preadipocytes (time 0) were also subjected to immunoblot analysis. e, effect of Mkp-1 antisense RNA expression on the phosphorylation of p42/p44 MAPK. 3T3-L1 mature adipocytes were infected with adenoviral vectors as in d. Forty eight hours after infection, the cells were deprived of serum for 16 h, incubated for 10 min in the absence or presence of 10% FBS, lysed, and subjected to immunoblot analysis with antibodies to phosphorylated p42/p44 MAPK or to total p42/p44 MAPK. All data are representative of at least three independent experiments.
AxCAlacZ exhibited substantial accumulation of lipid droplets 8 days after exposure to the inducers of differentiation, lipid accumulation was markedly inhibited in the cells infected with AxCAMkp-1AS (Fig. 5c). The induction of PPAR␥, C/EBP␣, and aP2 was also greatly reduced in cells infected with AxCAMkp-1AS compared with that apparent in cells infected with the control virus (Fig. 5d). The abundance of CREB was not affected by expression of Mkp-1 antisense RNA (Fig. 5, a and d), and the up-regulation of C/EBP␤ expression, which is maximal 2 days after exposure to the hormonal inducers of differentiation, was similar in cells infected with AxCAMkp-1AS and in those infected with AxCAlacZ (Fig. 5e). These results indicate that the induction of MKP-1 expression is essential for adipocyte differentiation and that MKP-1 contributes to this process at a step after the induction of C/EBP␤. 3T3-L1 cells that had been infected with AxCAMkp-1AS and exposed to inducers of differentiation in the presence of PD98059, a pharmacological inhibitor of MEK, manifested a reduced level of p42/p44 MAPK phosphorylation similar to that apparent in cells that had been infected with the control virus before induction of differentiation (Fig. 5b). Such treatment with PD98059 also prevented the inhibitory effect of MKP-1 depletion on adipocyte differentiation, as monitored by the accumulation of lipid droplets (Fig. 5c) and by the up-regulation of adipocyte marker proteins including PPAR␥, C/EBP␣, and aP2 (Fig. 5d). These results indicate that the inhibitory effect of AxCAMkp-1AS on adipocyte differentiation is attributable to the sustained activation of p42/p44 MAPK. DISCUSSION We have shown that the activity of p42/p44 MAPK decreases during the differentiation of cultured preadipocytes, consistent with previous observations (10). Furthermore, our results suggest that this down-regulation of p42/p44 MAPK activity is intrinsic to the differentiation program, reflecting a change in the intracellular signaling machinery, and is independent of the specific inducers of differentiation. The extent of the seruminduced phosphorylation of p42/p44 MAPK was thus greater in nondifferentiated preadipocytes than in mature adipocytes, even though the abundance of p42/p44 MAPK did not differ between the two states. An insulin-induced increase in the activity of p42/p44 MAPK apparent several hours after exposure of 3T3-L1 cells to insulin, dexamethasone, and IBMX has been shown to enhance the induction of PPAR␥ and C/EBP␣ expression during differentiation (21). This finding is consistent with the fact that insulin increases the proportion of differentiated cells but is not absolutely required for differentiation. Signaling by MEK and p42/p44 MAPK has also been shown to contribute to the mitotic clonal expansion that occurs within several days after the induction of differentiation of 3T3-L1 cells (22,23). In contrast, incubation of preadipocytes with growth factors that induce a prolonged activation of p42/ p44 MAPK prevents the cells from differentiating into adipocytes (24 -26), and such inhibitory effects are attenuated by the MEK inhibitor PD98059 (24). Together, these various observations suggest that, although p42/p44 MAPK signaling is impor- tant for early steps of the differentiation program, sustained activation of this signaling inhibits adipocyte differentiation. The down-regulation of p42/p44 MAPK activity that occurs during the differentiation of preadipocytes thus appears to be essential for completion of the differentiation process.
The Mkp-1 gene is an immediate early gene whose expression is induced rapidly by a variety of extracellular stimuli. Agents used to induce the differentiation of 3T3-L1 cells, including insulin (27), glucocorticoids (28), and cyclic AMP analogs (29), have thus been shown to increase expression of the Mkp-1 gene. However, the induction of MKP-1 gene expression by such stimuli is usually transient (19,(27)(28)(29). Indeed, we found that the abundance of Mkp-1 was increased within several hours after exposure of 3T3-L1 preadipocytes to insulin, dexamethasone, and IBMX but had returned to basal levels by 16 h. 2 The pronounced increase in the amounts of both MKP-1 mRNA and protein apparent 4 days after exposure of 3T3-L1 cells to these inducers of differentiation as well as the persistence of the increased level of Mkp-1 gene expression in differentiated adipocytes are thus likely attributable to an alteration of the transcriptional machinery programmed during differentiation and not to a direct effect of the differentiation inducers per se. Given that the time courses of the expression of C/ebp␣ 2 and Ppar␥ were similar to that of Mkp-1 expression, one or both of these transcription factors may play a role in transcriptional activation of the Mkp-1 gene. In this regard, the promoter of the mouse Mkp-1 gene contains a sequence (TGGTGGAAT, nucleotides Ϫ1155 to Ϫ1147) to which both C/EBP␣ and C/EBP␤ are able to bind. Furthermore, the induction of Mkp-1 expression normally observed during liver regeneration does not occur in mice that lack the C/ebp␤ gene (30). The mechanism by which expression of the Mkp-1 gene is up-regulated during the differentiation of preadipocytes remains to be established, however.
The down-regulation of the activity and phosphorylation of p42/p44 MAPK appeared related to the up-regulation of Mkp-1 expression during the differentiation both of 3T3-L1 preadipocytes in culture as well as of cells in adipose tissue derived from animals. Furthermore, ectopic expression of MKP-1 in preadipocytes at a level similar to that of the endogenous protein in mature adipocytes resulted in a marked decrease in the extent of p42/p44 MAPK phosphorylation, and conversely, depletion of endogenous MKP-1 in mature adipocytes triggered an increase in the level of p42/ p44 MAPK phosphorylation, suggesting that MKP-1 is a major regulator of p42/p44 MAPK activity during the differentiation of adipocytes. Prevention of the up-regulation of MKP-1 expression by transfection of preadipocytes with a vector for Mkp-1 antisense RNA also greatly inhibited adipogenesis in cells exposed to insulin, dexamethasone, and IBMX. The phosphorylation of p42/p44 MAPK was not downregulated even 8 days after exposure of 3T3-L1 preadipocytes expressing Mkp-1 antisense RNA to the inducers of differentiation. Moreover, inhibition of p42/p44 MAPK activity by treatment with PD98059 restored the ability of cells expressing Mkp-1 antisense RNA to undergo differentiation into adipocytes. The sustained activation of p42/p44 MAPK in these cells thus appeared to be the cause and not a result of the prevention of differentiation by MKP-1 depletion. Although MKP-1 dephosphorylates and inactivates kinases other than p42/p44 MAPK (19), the restorative effect of PD98059 indicates the importance of p42/p44 MAPK as a target of MKP-1 during adipocyte differentiation.
Whereas the induction of PPAR␥ and C/EBP␣ expression during adipocyte differentiation was markedly attenuated by depletion of MKP-1, that of C/EBP␤ expression was unaffected. These results are consistent with our observation that the activity of p42/p44 MAPK began to decrease after the onset of the increase in C/EBP␤ expression during differentiation. The up-regulation of MKP-1 expression and the down-regulation of p42/p44 MAPK activity therefore likely contribute to adipocyte differentiation at a step after the induction of C/EBP␤ expression. The activity of many members of the nuclear receptor family, including that of PPAR␥, is affected by their phosphorylation status. Phosphorylation of PPAR␥2 on Ser 112 (equivalent to Ser 82 of PPAR␥1) is catalyzed by p42/p44 MAPK and inhibits its transactivation activity (31,32). Depletion of MKP-1 in 3T3-L1 preadipocytes thus likely promotes the phosphorylation of PPAR␥ by p42/p44 MAPK and thereby prevents the cells from acquiring the characteristics of mature adipocytes. PPAR␥ appears capable of inducing the expression both of itself and of C/EBP␣ (33,34). The prevention of the increase in the abundance of PPAR␥ and of C/EBP␣ by expression of Mkp-1 antisense RNA might thus be attributable to inhibition of the transactivation activity of PPAR␥. We have found that growth factor-induced inhibition of transactivation activity of PPAR␥2 was prevented either by MKP-1 overexpression or by PD98059, 2 which is consistent with our hypothesis that MKP-1 can regulate the transactivation activity of PPAR␥. In summary, we have shown that the down-regulation of p42/p44 MAPK activity is a characteristic feature of and is crucial for adipocyte differentiation and that this biochemical change is attributable to an increase in the abundance of MKP-1. The abundance of MKP-1 is regulated by a variety of stimuli at both the transcriptional and post-transcriptional levels (19). Given that the putative MAPK phosphorylation sites in PPAR␥ influence both the size of fat cells and whole body insulin sensitivity in living animals (35), MKP-1 signaling in adipose tissue may provide a target for the development of new treatments for obesity and diabetes mellitus.