Because leptin has recently been shown to induce proliferation and/or differentiation of different cell types through different pathways, the aim of the present study was to investigate, in vitro, the influence of leptin on adipogenesis in rat preadipocytes. A prerequisite to this study was to identify leptin receptors (Ob-Ra and Ob-Rb) in preadipocytes from femoral subcutaneous fat. We observed that expressions of Ob-Ra and Ob-Rb increase during adipogenesis. Furthermore, leptin induces an increase of p42/p44 mitogen-activated protein kinase phosphorylated isoforms in both confluent and differentiated preadipocytes and of STAT3 phosphorylation only in confluent preadipocytes. Moreover, exposure to leptin promoted activator protein-1 complex DNA binding activity in confluent preadipocytes. Finally, exposure of primary cultured preadipocytes from the subcutaneous area to leptin (10 nM) resulted in an increased proliferation ([3H]thymidine incorporation and cell counting) and differentiation (glycerol-3-phosphate dehydrogenase activity and mRNA levels of lipoprotein lipase, peroxisome proliferator-activated receptor-γ2, and c-fos). Altogether, these results indicate that, in vitro at least, leptin through its functional receptors exerts a proadipogenic action in subcutaneous preadipocytes.
- leptin receptors
- mitogen-activated protein kinase
leptin, the recently identified product of the ob gene, is a 16-kDa protein produced and secreted primarily by adipocytes (61). Leptin acts through functional receptors (Ob-R) on specific regions of the brain to regulate food intake and energy expenditure. These receptors are related to class I cytokine receptors and their different isoforms (Ob-Ra, b, c, d, e, and f) are transcribed from a single gene via alternative splicing (54). These receptors have the same extracellular domain. Each isoform is expressed in a wide variety of tissues and in a tissue-specific manner. The Ob-Ra, which has a short intracellular domain, is expressed predominantly in kidney, lung, intestine, heart, testes, choroid plexus, brain microvessels, and adipose tissue and less in liver, skeletal muscle, adrenal, spleen, and pancreatic β-cells (29, 54). The Ob-Rb, which has the longest cytoplasmic domain, is highly expressed in the hypothalamus, cerebellum, and pancreatic β-cells but weakly expressed in spleen, heart, choroid plexus, and kidney (29, 54).
The leptin-bound Ob-Rb stimulates gene transcription via an activation of different pathways. Indeed, Ob-Rb has the capacity to activate the Janus kinase/signal transducers and activators of transcription (Jak/STAT) and the mitogen-activated protein kinase (MAPK) pathways (54) and to increase transcription of suppressor of cytokine signaling 3 (SOCS-3) (6, 19). In the hypothalamus, the Ob-Rb transduces the leptin signal via an activation of STAT3, which results in a negative regulation of food intake and a positive influence on energy expenditure (54). The leptin-bound Ob-Ra activates Jak2 but is unable to induce STAT activation and is thought to play a role in the transport of leptin into the brain (7, 54).
Besides these important metabolic effects in the hypothalamus, leptin elicits various peripheral actions, as recently shown. Indeed leptin intervenes as a modulator of hematopoiesis, thermogenesis, angiogenesis, lipid metabolism, reproduction, immunological responses, and of pancreatic, intestinal, and kidney functions (24, 50, 52,54). Moreover, the observation that leptin stimulates lipolysis both in vivo and in vitro supports the concept that leptin acts as an autocrine regulatory signal in adipose tissue (21, 22). Cytokines and growth factors other than leptin are also produced and secreted by adipocytes [i.e., tumor necrosis factor-α (TNF-α) and insulin-like growth factor I]. These factors play an important role in the regulation of adipogenesis (15, 49) by modulating the STAT and the MAPK pathways (8, 26, 31, 62). Because leptin signaling also involves the STAT and MAPK cascades (7,54), a role for leptin as a putative autocrine/paracrine regulatory signal controlling body fat mass development at the level of the adipoconversion process is questionable. The purpose of the present study was to test this hypothesis.
We have at first attempted to identify the presence of Ob-Ra and Ob-Rb in primary cultured confluent and differentiated preadipocytes from the stromal vascular fraction of rat subcutaneous adipose tissue. Second, we have studied the influence of leptin on the activation of MAPK and Jak/STAT pathways during the adipoconversion process. After these receptors were shown to be expressed and functional in preadipocytes, we then investigated the direct influence of leptin on the adipoconversion of these cells in vitro. The results presented herein clearly show that leptin stimulates in vitro the proliferation and differentiation capacities of rat subcutaneous preadipocytes.
MATERIALS AND METHODS
DMEM, DMEM/Ham's F-12 (50:50 mix), penicillin, streptomycin, HEPES, transferrin, triiodothyronine (T3), leupeptin, aprotinin, phenylmethylsulfonyl fluoride (PMSF) and BSA were purchased from Sigma Chemical (St. Louis, MO). Collagenase was from Roche Molecular Biochemicals (Mannheim, Germany). Recombinant murine leptin was from Pepro Tech (London, UK). Superscript II RNase H-RT was from GIBCO BRL (Grand Island, NY). Taq polymerase and RNA guard were provided by Pharmacia Biotechnology (Uppsala, Sweden). The MEK inhibitor, U-0126, and T4 polynucleotide kinase were from Promega (Madison, WI).
Male Sprague-Dawley rats were kept under controlled lighting conditions (light, 6:00 AM; dark, 8:00 PM) and constant temperature (21°C) with free access to water and food. Animals (250–275 g) were killed by decapitation, and femoral subcutaneous adipose tissue was immediately removed under sterile conditions.
The stromal vascular fraction was obtained after digestion of subcutaneous adipose tissues by collagenase as previously described (46). The cells were plated in DMEM supplemented with HEPES (20 mM), streptomycin (0.1 mg/ml), penicillin (100 U/ml), and 8% FCS and maintained at 37°C under 5% CO2 atmosphere. After they were plated, the cells were extensively washed and maintained in primary culture as follows.
First, for cell growth experiments, the cells were maintained for 24 h in DMEM supplemented with 8% FCS and then for another 24 h in DMEM with 2% FCS in the absence or presence of recombinant leptin (10 nM). In preliminary experiments, different concentrations of leptin (1, 5, 10, 50, and 100 nM) and different exposure times to leptin (24 and 48 h) were tested. The optimal conditions found were 24 h of exposure to 10 nM leptin.
Second, for differentiation studies, the cells were maintained in DMEM with 8% FCS until 70–80% confluence was reached (1 day after plating), and then in DMEM-F12 (1:1) supplemented with insulin (5 μg/ml), transferrin (10 μg/ml), T3 (2 nM), and antibiotics (0.1 mg/ml streptomycin and 100 U/ml penicillin) (ITT medium) as described in Ref. 14. Two days later, the ITT medium was removed and replaced by fresh ITT medium with or without leptin (10 nM) and the cell culture was extended by 24 or 48 h. Preliminary experiments again revealed that the best leptin effectiveness occurs with 10 nM leptin.
To detect Ob-Ra and Ob-Rb mRNA, total RNA was isolated using Trizol reagent (12) from cultures at total confluence (2 days after plating) and at the differentiated state (6 days after plating, i.e., after 4 days in ITT medium). For the study of lipoprotein lipase (LPL) and peroxisome proliferator-activated receptor-γ2 (PPARγ2) mRNA expressions, total RNA was extracted on the third and fourth day of culture in ITT medium from differentiated preadipocytes exposed or not to leptin for 24 and 48 h. Finally, the effect of leptin on c-fos mRNA expression was shown on the third day after 2 days in ITT medium and an overnight in a DMEM/Ham's F-12 medium. Total RNA was then extracted after 15, 30, and 60 min of exposure to 10 nM leptin. RNA recovery and quality were checked by measuring the 260/280 optical density ratio and by electrophoresis under denaturing conditions on 2% agarose gel.
As described previously (37), 2.5 μg total RNA (Ob-Ra and Ob-Rb expressions) or 0.5 μg total RNA (LPL, PPARγ2, c-fos, cyclophilin, and β-actin expressions) were submitted to RT-PCR (47). PCR was performed with a thermocycler Gene Amp PCR 2400 (Perkin Elmer) followed by quantification realized with Bio-gene software. To ensure that amplifications of the genes tested were within the exponential range, different numbers of PCR were run. The number of PCR amplification cycles chosen for each studied gene is indicated in Table1.
At day 1 postplating, culture medium was replaced by fresh medium containing 2% FCS, and cells were then exposed or not to recombinant leptin (10 nM). Twenty-four hours later, cells were washed three times with Hanks' buffer (in mM: 136.7 NaCl, 5.36 KCl, 0.42 Na2HPO4, 0.44 KH2PO4, and 4.16 NaHCO3) followed by addition of 0.2% trypsin in Hanks' buffer to dishes for 2–3 min at 37°C while shaking. Cells were then collected in Hanks' supplemented buffer with 10% FCS (vol/vol). Twenty microliters of this suspension were diluted with five microliters of crystal violet, and the number of cells was established using a hemocytometer.
In addition to direct cell counts, [3H]thymidine incorporation was used as a measure of DNA synthesis. At day 1 postplating, the culture medium was changed into DMEM containing 2% FCS, [3H]thymidine (1 mCi/ml) and supplemented or not with recombinant leptin (10 nM). Twenty-four hours later, the dishes were washed three times with saline, and cells were treated for 5 min with 1% SDS and then with 10% TCA for 45 min at 4°C. After filtration on GF/C glass fiber filters (Whatman, Clifton), radioactivity was counted.
Glycerol-3-Phosphate Dehydrogenase Activity Assay
After 24 and 48 h of culture in the absence or presence of leptin (10 nM) in ITT medium, differentiation media were discarded and cells were scraped and sonicated [3 blasts for 15 s; VibraCell (Bioblock, Strasbourg, France)] in ice-cold buffer containing (in mM) 50 Tris · HCl (pH 7.4), 0.25 sucrose, 1 EDTA, and 1 β-mercaptoethanol. The homogenates were centrifuged at 20,000g for 10 min at 4°C, and the resulting supernatants were used for glycerol-3-phosphate dehydrogenase (GPDH) assays according to Wise and Green (58). Activities are expressed in milliunits per milligram of protein (1 mU being equal to the oxidation of 1 nmol of NADH2/min).
Ob-R Protein Expression
Rat preadipocyte membranes were prepared at the confluent and differentiated states. Cells were scraped on ice with buffer A containing 10 mM Tris, 0.25 M sucrose, 5 mM EDTA, and protease inhibitors (0.57 mM PMSF, 5 μg/ml leupeptin, and 5 μg/ml aprotinin). Next, cells were sonicated for 15 s and centrifuged (21,000 g at 4°C) for 20 min, and the resulting supernatant was collected and stored in Laemmli's buffer (35) (vol/vol) at −20°C.
In parallel experiments, membranes were obtained from rat hypothalamus and brain, following the procedure previously described in Ref.36, and used as positive controls for Ob-Rb and Ob-Ra. Equal amounts of membrane proteins (100 μg) were subjected to SDS-PAGE (10% acrylamide). Proteins were transferred to polyvinylidene difluoride (PVDF) membrane for 90 min at room temperature. After blocking in buffer A (20 mM Tris · HCl, 137 mM NaCl, and 0.1% Tween 20) containing 2.5% gelatin for 2 h at room temperature, filters were incubated overnight at room temperature with goat polyclonal anti-Ob-R antibody (K-20) (1: 300 dilution; Santa Cruz Biotechnology, Santa Cruz, CA). Blots were then extensively washed with buffer A and incubated with the secondary antiserum (horseradish peroxidase-labeled anti-goat IgG 1:10,000 dilution) for 1 h at room temperature and washed. Finally, an enhanced chemiluminescence kit from Pierce (Interchim) was used for signal detection. Control experiments with various protein amounts (50–200 μg) were performed to ensure that the densitometric signal intensity was proportional to the loaded amount of protein.
Specificity of the immunoreactive Ob-R proteins was verified by loss of the immunoreactivity of samples when incubated with the antiserum neutralized by the corresponding specific peptide.
STAT3 and MAPK Activation
The time course of the STAT3 phosphorylated isoform was investigated as follows: confluent or differentiated preadipocytes were maintained overnight in a serum-free culture medium at day 1postplating or after 2 days in ITT medium, respectively. After exposure to recombinant leptin (10 nM) for 5, 15, and 30 min, cells were scraped and sonicated on ice in buffer containing 50 mM Tris, 120 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.5 mM desoxycholate, 0.1% SDS, 1 mM sodium vanadate, 0.57 mM PMSF, 30 mM β-glycerophosphate, 5 μg/ml aprotinin, and 12.5 μg/ml leupeptin. After centrifugation at 100,000g for 10 min at 4°C, supernatants were diluted in Laemmli's buffer (vol/vol).
Equal amounts (50 μg) of cellular extracts were subjected to SDS-PAGE (7.5%). Proteins were transferred to PVDF membrane and blocked inbuffer A with 5% milk for 1 h. Membranes were then incubated overnight at 4°C with rabbit polyclonal phospho-STAT3 (Tyr-705) antibody (1:1,000 dilution; New England Biolabs) inbuffer A with 5% milk or with rabbit polyclonal STAT3 antibody (C-20) (1:200 dilution; Santa Cruz Biotechnology) inbuffer A with 2.5% gelatin. Blots were then extensively washed with buffer A and incubated with the secondary antiserum (horseradish peroxidase-labeled anti-rabbit IgG 1:5,000 dilution) for 1 h at room temperature and washed. Finally, signal detection was performed as described above.
To study the MAPK pathway, confluent or differentiated preadipocytes were also maintained overnight in serum-free culture medium before in vitro leptin stimulation. The time course of activation of the p42/p44 MAPK phosphorylated form was investigated as follows: cells were exposed for 2, 5, 15, and 30 min to recombinant leptin (10 nM); for 5 min to U-0126 (10 μM) in the presence or absence of leptin (10 nM); or for 5 min to 10% FCS. Cells were then scraped and sonicated on ice in buffer containing 10 mM Tris, 150 mM NaCl, 2 mM EGTA, 0.5 mM dithiothreitol (DTT), 1 mM sodium vanadate, 30 mM β-glycerophosphate, 0.57 mM PMSF, 5 μg/ml leupeptin, and 5 μg/ml aprotinin. After cells were centrifuged at 100,000 g for 15 min at 4°C, cytosolic extracts were diluted in Laemmli's buffer (vol/vol).
Equal amounts (10 μg) of cytosolic extracts were subjected to SDS-PAGE (12.5%). Proteins were transferred to a PVDF membrane and blocked in buffer A with 2.5% gelatin for 2 h. Membranes were then incubated overnight at room temperature with rabbit polyclonal p42/p44 MAPK phosphorylated antibody (pTEpY, V8031; 1: 7,000 dilution; Promega) or with mouse monoclonal anti-total MAPK antibody (1:500 dilution; Transduction Laboratories, Lexington, KY) inbuffer A with 2.5% gelatin. The resulting blots were extensively washed with buffer A and incubated with the secondary antiserum (horseradish peroxidase-labeled anti-rabbit IgG or anti-mouse IgG 1:10,000 dilution) for 1 h at room temperature and washed. Finally, signal detection was performed as described above.
Activator Protein-1 Complex DNA Binding
The influence of leptin on activator protein-1 complex (AP-1) DNA binding activity was tested as follows: confluent or differentiated preadipocytes were maintained overnight in a serum-free culture medium before in vitro stimulation. Activation of AP-1 DNA binding was measured after 15, 30, and 60 min exposure to recombinant leptin (10 nM) or 30 min exposure to 10% FCS. Then, as described in Ref.30, cells were scraped on ice in buffer A containing 10 mM Tris, 0.15 M NaCl, 1 mM EDTA, 0.6% Nonidet P-40, 1 mM sodium vanadate, 0.57 mM PMSF, 20 mM β-glycerophosphate, 10 mM NaF, 0.5 mM DTT, 5 μg/ml aprotinin, and 5 μg/ml leupeptin, pH 7.9. The homogenates were centrifuged at 2,400 g for 5 min at 4°C, and the nuclear extracts were prepared as follows: supernatants were removed, and the resulting pellets were resuspended in cold buffer B (10 mM HEPES, 420 mM NaCl, 0.1 mM EDTA, 0.1 mM EGTA, 1.5 mM MgCl2, 25% glycerol, 0.6% Nonidet P-40, 0.57 mM PMSF, 1 mM sodium vanadate, 20 mM β-glycerophosphate, 10 mM NaF, 0.5 mM DTT, 5 μg/ml aprotinin, and 5 μg/ml leupeptin, pH 7.9). Suspensions were vigorously shaken at 4°C for 20 min followed by centrifugation at 20,000 g for 20 min at 4°C. The resulting supernatants containing the nuclear extracts were then used for electrophoretic mobility shift assay.
Protein-DNA complexes were formed by incubating 2 μg of nuclear extracts in 10 μl binding cocktail (10 mM HEPES, 50 mM KCl, 5 mM MgCl2, 0.1 mM EDTA, 50 mM NaCl, 4 mM spermidin, 2 mM DTT, 100 μg/ml albumin, and 35% glycerol, pH 8) in the presence of 2.5 μg of poly(dI-dC) for 15 min at 4°C. Then 100,000 cpm of32P-labeled double-stranded oligonucleotide containing a binding site for AP-1 (5′-CGC TTG ATG AGT CAG CCG GAA-3′) was added, and the incubations were further extended for 15 min at room temperature. The resulting DNA-protein complexes were separated from the unbound probes by electrophoresis on a native 6% polyacrylamide gel in 0.5× Tris-borate-EDTA buffer. Gels were then dried and subjected to autoradiography. In competition experiments, 1, 10, and 100-fold molar excesses of unlabeled AP-1 double-stranded oligonucleotides were included in the binding reaction mixture. The double-stranded oligonucleotides were labeled with [γ-32P]ATP (3,000 Ci/mmol) using T4polynucleotide kinase.
Protein concentration was measured according to Bradford (10) with BSA as standard.
All values were expressed as means ± SE of three to four different experiments, and statistical analysis was performed using unpaired Student's t-test.
We chose to investigate the effects of leptin on rat adipogenesis in subcutaneous preadipocytes, a superficial fat deposit that presented lower replication and differentiation capacities than other deep fat localizations (16, 34) and thus is more appropriate to observe a possible modulation of adipogenesis by an effector.
Expression of Leptin Receptors in Rat Subcutaneous Preadipocytes
Using specific primers for Ob-Ra (short isoform) and Ob-Rb (long isoform) (see Table 1), we measured the expression of both Ob-R isoforms in rat subcutaneous preadipocytes at confluence and differentiation states. By this semiquantitative RT-PCR method, we observed that Ob-Ra and Ob-Rb mRNA expressions increased in subcutaneous preadipocytes during the adipoconversion process (Fig.1 A). The same picture was provided by Western blot analysis (Fig. 1 B) using an antibody directed against the amino terminus of the rat leptin receptor, which recognized ∼120- and ∼210-kDa bands, in preadipocytes. These sizes are consistent with the short and long forms of rat leptin receptor (50). As also shown in Fig.1 B, these bands were identical to the positive controls (brain and hypothalamus) (38) and disappeared in peptide-saturating experiments.
These results indicate that, at both confluent and differentiated states, rat subcutaneous preadipocytes express the two Ob-Ra and Ob-Rb isoforms of the leptin receptor.
Activation of STAT3 by Leptin
Because the phosphorylation-dependent activation of STAT3 is a major transduction pathway for leptin signaling (33, 50,54), expression of phosphorylated STAT3 was investigated in cellular extracts from preadipocytes during adipogenesis in response to recombinant leptin. As shown in Fig. 2, exposure of confluent subcutaneous preadipocytes to leptin (10 nM) for 15 min resulted in an increase in the STAT3 phosphorylated form (×1.8 ± 0.35), and the magnitude of this effect was decreased after 30 min. In differentiated subcutaneous preadipocytes, however, no effect of leptin on STAT3 activation could be observed.
Activation of p42/p44 MAPK by Leptin
Because it was previously shown that Ob-Rb and Ob-Ra can also lead to activation of the MAPK pathway (18, 53, 59), the activated form of p42/p44 MAPK was investigated in cytosolic extracts of confluent and differentiated subcutaneous preadipocytes in response to recombinant leptin. As shown in Fig.3, addition of leptin (10 nM) induced a clear increase in the p42/p44 MAPK phosphorylated isoforms in both confluent (×2.2 ± 0.25) and differentiated (×3.05 ± 0.57) preadipocytes. The level of activation was maximal after 5 min. The same effect was observed after preincubation with 10% FCS (positive control). Preincubation of preadipocytes for 5 min with the two specific MAPK inhibitors U-0126 (20) or PD-98059 (17) (data not shown) blunted the leptin effect, which confirmed that p42/p44 MAPK-activated isoforms were targets of leptin (Fig. 3). Moreover, we observed that total MAPK protein expression was unaltered by the leptin treatment.
Because activation of the MAPK signaling pathway by growth factors results in AP-1 activation (for review see Ref. 57), we next investigated whether the AP-1 DNA binding activity in vitro could be a possible target for leptin in preadipocytes.
As shown in Fig. 4, after 15 min exposure to leptin, an increase (×1.5 ± 0.12) in AP-1 DNA binding activity occurred in confluent but not in differentiated preadipocytes (data not shown). Moreover, this band completely disappeared in the presence of 100-fold unlabeled AP-1 probe, and, after 15 min of exposure to 10% FCS, a twofold increase in AP-1 binding activity was observed, thus confirming the validity of our assay.
In vitro Effects of Leptin on Adipogenesis
In several reports, leptin was shown to control cell proliferation (24, 50, 56). Conversely, other cytokines like TNF and interleukin-6 have been reported to inhibit cell differentiation (25, 43, 49). In contrast, leptin has been shown to enhance differentiation of human marrow stromal cells into osteoblasts and to inhibit their adipoconversion (55). In another studies, leptin was described as an inductor of differentiation for hematopoietic (24) and germ cells (18). These different observations led us to consider the possibilities that leptin may modulate the proliferation and differentiation capacities of rat subcutaneous preadipocytes.
As shown in Fig. 5, a 24 h exposure to leptin (10 nM) resulted in a significant increase in subcutaneous preadipocyte growth (×1.44 ± 0.07), as measured by [3H]thymidine incorporation into DNA. It should be noted that the same results were obtained whether 2% FCS was present or not in the culture media and also that insulin (1 μM), used as a positive control, enhanced preadipocyte cell growth (data not shown). Moreover, the addition of U-0126 (10 μM), which per se had no effect on cell proliferation, abolished the positive effect of leptin on subcutaneous preadipocyte growth.
Finally, the leptin-stimulated proliferation of subcutaneous preadipocytes was also confirmed by direct cell counting (×1.4 ± 9, n = 3). These results show that leptin increases subcutaneous preadipocyte replication in vitro.
As shown in Fig. 6 A, exposure to leptin (10 nM) for 48 h results in a clear increase in lipid droplet accumulation in subcutaneous preadipocytes. These morphological observations were confirmed by measurements of GPDH activity, a late marker of adipocyte differentiation that faithfully reflects cellular triacylglycerol content (23). As a matter of fact, as shown in Fig. 6 B, exposure to leptin (10 nM) for 48 h resulted in an increase in the level of GPDH activity (×1.5) in subcutaneous preadipocytes. This effect was suppressed in the presence of the specific MEK inhibitor, U-0126 (10 μM), which per se had little effect on GPDH activity (data not shown). No effect of leptin was seen, however, after 24 h of exposure to leptin (data not shown) probably because GPDH expression occurs later.
Because the LPL gene is expressed early during the adipogenic process (1), we investigated the effect of leptin on LPL mRNA levels in primary cultured subcutaneous preadipocytes. After 2 days of culture in the differentiation medium, the preadipocytes were exposed to leptin (10 nM) for either 24 or 48 h, and then RT-PCR analysis of LPL mRNA was performed. As shown in Fig.7 A, LPL mRNA expression was increased in subcutaneous preadipocytes (×3.07 ± 0.27) after 24 h exposure to leptin. After 48 h, however, leptin failed to elicit any effect on LPL mRNA, which is not surprising considering the early expression of the LPL gene during the adipoconversion process and the observation that LPL mRNA levels increase (×3.5) between 24 and 48 h of culture in control cells (Fig. 7 A).
PPARγ is a major transactivating factor involved in the adipoconversion process (39). To investigate whether leptin might affect PPARγ2 mRNA expression, cells were maintained in the differentiating medium for 2 days before exposure to leptin for either 24 or 48 h (Fig. 7 B).
As observed for LPL mRNA expression, the results of RT-PCR analysis show that PPARγ2 mRNA levels in subcutaneous preadipocytes were increased after 24 h exposure to leptin (×1.5 ± 0.1), but this effect could no longer be observed after 48 h leptin exposure, probably because in control cells PPARγ2 mRNA increased 1.5-fold between 24 and 48 h of culture (Fig. 7 B).
These experiments, indicating that leptin modulates in vitro the adipogenic process in subcutaneous preadipocytes, led us to investigate the underlying mechanism(s). AP-1 consists of Fos and Jun protein homo- or heterodimers and binds to regulatory sequences in the promoter of various target genes involved in cell growth, differentiation, and metabolism (45). Indeed, various observations have led to the assignment of an important role to the AP-1 complex in the regulation of the adipocyte differentiation process (27,60). This complex induces transcription of both the adipocyte intracellular lipid-binding protein P2 (aP2) gene (51) and the LPL gene in Ob1771 preadipose cells (3). These observations led us to investigate the effects of leptin on c-fos mRNA levels in preadipocytes during differentiation (after 2 days in ITT medium). Using RT-PCR analysis, we found that exposure of differentiating preadipocytes to 10 nM leptin resulted in a significant increase in c-fos transcript levels after 15 min (×3.3 ± 1), an effect that was further amplified (×6.5 ± 0.6) after 30 min (Fig.8).
Leptin, the ob gene product, is synthesized in adipose tissue and plays an important role in body fat mass homeostasis (28, 42). Leptin acts by binding to specific receptors, the Ob-R receptors, which belong to the class I cytokine receptor family. Recently, the presence of Ob-R in adipose tissue has been reported (29), suggesting that the leptin effects on this tissue such as lipolysis stimulation (21, 22) are mediated through these receptors.
In the present study, Ob-R mRNA and protein were characterized by RT-PCR and immunoblotting in both rat confluent and differentiated subcutaneous preadipocytes. In agreement with a recent report in human differentiated preadipocytes (9), we show here that rat preadipocytes express the two Ob-R isoforms, Ob-Ra and Ob-Rb. Moreover, our experiments indicate that these expressions increase during the adipoconversion process. The latter results strongly suggest that leptin intervenes as an autocrine/paracrine signal controlling preadipocyte replication and/or differentiation, as do other cytokines such as TNF-α (49) and the leukemia inhibitory factor (LIF) (2).
The functionality of these Ob-R receptors in preadipocytes at both confluent and differentiated states was evaluated by measuring the activation of the STAT3 and MAPK signaling pathways. Because the concentration of leptin in our in vitro experiments (10 nM) was 40 times higher than the normal rat blood leptin level, the physiological relevance of the present data could be questionable. However, the leptin concentration used in our study is in the lower part of the range of those tested in other in vitro studies (10–120 nM) (40, 11), and, due to the lack of soluble leptin receptor in our culture medium, the stability of the recombinant leptin is certainly lower than in the plasma in vivo. Moreover, a recent in vivo study reported much higher leptin levels in fat interstitial fluid than in blood (41), suggesting that the leptin concentration tested in the present study is physiological.
In the present in vitro study, it was demonstrated that leptin rapidly and transiently activated the MAPK pathway through phosphorylation of the p42 and p44 isoforms in both confluent and differentiated preadipocytes. Activation of MAPK generally results in a phosphorylation-dependent modulation of the transactivation activity of various transcriptional factors including the AP-1. As shown here, exposure to leptin increases AP-1 DNA-binding activity in confluent but not in differentiated preadipocytes, suggesting that, in the former cells, leptin acts like growth factors (for review see Ref.57) through MAPK activation and AP-1 DNA binding activity stimulation. Moreover, we observed, for the first time, that leptin in vitro increases preadipocyte proliferation and that this positive effect was blocked by the specific MEK inhibitor, U-0126. These results suggest that leptin stimulates the proliferation of rat subcutaneous preadipocytes through activation of both the MAPK cascade and AP-1 DNA binding activity as leptin does in other cell types (48,53).
In parallel to MAPK pathway activation, we also found that leptin activates the STAT pathway through increased STAT3 phosphorylation in rat confluent preadipocytes. In a recent report, Deng et al. (13) have shown that STAT3 phosphorylation correlates with postconfluent preadipocyte mitotic clonal expansion, suggesting a role for STAT3 in the proliferative phase of adipogenesis. This observation allows postulation that the positive effect of leptin on rat preadipocyte growth may be linked to STAT3 phosphorylation in addition to MAPK pathway activation. Moreover, our in vitro present finding in confluent preadipocytes is consistent with in vivo experiments showing the same leptin effects in white adipose tissue (4, 33,53).
However, in differentiated preadipocytes leptin was ineffective on STAT3 phosphorylation. Possible explanations to this negative finding are 1) the occurrence of a sustained STAT3 activation state during the course of the adipoconversion process or 2) an increased expression of the suppressor of cytokine signaling, SOCS-3, whose expression is STAT3-dependent and whose role is to prevent leptin-induced STAT3 activation through inhibition of Jak2 tyrosine phosphorylation (6).
Another signaling cascade that could also be involved in the leptin effects on adipose tissue is the phosphatidylinositol 3-kinase (PI3-kinase)-mediated pathway. It is known that in C2C12 myotubes (5) leptin was shown to display various effects of insulin such as stimulation of glucose transport and glycogen synthesis through stimulation of PI3-kinase (32). However, a recent study reported that leptin rapidly activates STAT3 and MAPK in adipose tissue explants ex vivo and in 3T3-L1 adipocytes but has no significant effect on PI3-kinase activity and Akt phosphorylation and activity (33). Although not investigated in the current study, the effects of leptin on the PI3-kinase pathway during preadipocyte-adipocyte conversion remain to be established.
This study describes a proadipogenic effect of leptin on subcutaneous preadipocytes. We observed that this effect was correlated with increased expressions of the early and late markers of differentiation: LPL and GPDH, respectively, in these cells. In the same cells, leptin also stimulated expression of two key adipogenic transcriptional factors c-fos and PPARγ2, an effect that was accompanied by an increased fat storage in these cells. Moreover, in the differentiated cells, leptin induced a clear activation of p42/p44 MAPK, which seems to play a pivotal role in the proadipogenic effect of leptin, since, in the presence of the specific MEK inhibitor U-0126, the positive effect of leptin on GPDH activity was abolished. These results are to be compared with those observed with another cytokine, LIF, which also increase the differentiation of the preadipose cells Ob 1771 and 3T3-F442A via the MAPK cascade (2).
The present observation of increased adipogenesis in response to leptin in subcutaneous preadipocytes leads to speculation of a possible physiological role for leptin in vivo in the setting of subcutaneous adipose tissue in rats. Such a role, if any, would be consistent with a recent study reporting that transgenic mice overexpressing leptin showed at first a severe loss of body weight followed by a rebound effect characterized by an increase in body fat mass and number of lipid-filled adipocytes (44).
In conclusion, the present study demonstrates that leptin increases both the proliferation and differentiation of subcutaneous preadipocytes in vitro and suggests that leptin in vivo might be an autocrine/paracrine activator of adipogenesis in rat subcutaneous adipose tissue.
This work was supported by the Université Paris V and the Comité des Yvelines de la Ligue Contre le Cancer.
Address for reprint requests and other correspondence: Y. Giudicelli, Service de Biochimie, Centre Hospitalier, 78303 Poissy Cedex, France (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society