Protein Kinase B/AKT 1 Plays a Pivotal Role in Insulin-like Growth Factor-1 Receptor Signaling Induced 3T3-L1 Adipocyte Differentiation*

During 3T3-L1 preadipocyte differentiation induction, the insulin-stimulated insulin-like growth factor-1 (IGF-1) receptor signal is responsible for the induction of adipocyte differentiation. Treatment with inhibitors of phosphatidylinositol 3-kinase, LY294002 or wortmannin, leads to the complete blockade of adipocyte differentiation in 3T3-L1 preadipocytes. Of the three factors (1-methyl-3-isobutylxanthine, dexamethasone, and insulin) inducing 3T3-L1 preadipocyte differentiation, only insulin was able to activate the phosphatidylinositol 3-kinase-protein kinase B/Akt signal cascade. In 3T3-L1 preadipocytes, protein kinase B/Akt 1 RNA interference not only suppressed the expression of protein kinase B/Akt 1 but also blocked hormone-induced adipocyte differentiation. In these protein kinase B/Akt 1 RNA interference cells, the signal molecules upstream of protein kinase B/Akt 1, such as IGF-1 receptor and insulin receptor substrate-1, were normally activated by insulin stimulation, whereas insulin-stimulated phosphorylation of forkhead transcription factor (FKHR), which is a downstream molecule of PKB/Akt 1, was blocked. Thus, protein kinase B/Akt 1 is an important signal mediator in IGF-1 receptor signal cascade for inducing adipocyte differentiation.

Most of our current understandings on the cellular mechanisms of adipocyte differentiation regulation have come from the studies on established in vitro preadipocyte cell lines (1)(2)(3). These in vitro preadipocyte cell lines, e.g. 3T3-L1, 3T3-F442A, and TA1, can be induced to differentiate into adipocytes by insulin or in combination with other hormones. Subsequent studies on 3T3-L1 adipocyte differentiation induction indicate that the authentic hormone to induce the adipocyte differentiation is insulin-like growth factor-1 (IGF-1), 1

and insulin acts
through the IGF-1 receptor on the cell membrane to induce the differentiation (4). In post-confluent 3T3-L1 preadipocytes, IGF-1 receptor signal initiates the adipocyte differentiation process. Immediately after the hormonal stimulation, postconfluent G 0 3T3-L1 preadipocytes reenter the cell cycle and start the adipocyte differentiation program. It is clear that the IGF-1 receptor signal plays an irreplaceable role in inducing the adipocyte differentiation process in 3T3-L1 cells.
The signaling pathways or networks involved in mediating IGF-1 receptor signal for adipocyte differentiation are not fully understood. Two kinase systems, MAP kinases and phosphatidylinositol 3-kinase-protein kinase B/Akt, in 3T3-L1 cells can be activated by IGF-1 receptor signaling. The function of the MAP kinase system in IGF-1 receptor signal-induced 3T3-L1 cell adipogenesis is not clear. It is further complicated by the different effects exhibited by p38 MAP kinase and ERK1/2 on 3T3-L1 cell adipogenesis (5)(6)(7)(8)(9)(10)(11)(12). More molecular studies are required in order to understand the relationship between MAP kinase system and 3T3-L1 cell adipogenesis.
PI 3-kinase-PKB/Akt as an important intracellular signal cascade has been involved in the regulation of many cellular activities, including cell proliferation and apoptosis. It is also involved in IGF-1 receptor signal-regulated cell growth and anti-apoptosis in many types of cell (13)(14)(15)(16). Several lines of evidence have indicated the important function of PI 3-kinase-PKB/Akt signal cascade in adipogenesis (17)(18)(19)(20)(21)(22). Inhibition of PI 3-kinase with wortmannin blocks the adipocyte differentiation in 3T3-L1 cells (17). The ectopic expression of the constitutively active form of PKB/Akt will cause spontaneous adipocyte differentiation in 3T3-L1 cells (18,19). The generation of 3-phosphorylated phosphatidylinositols in response to the hormonal stimulation in 3T3-L1 cell is responsible for the activation of PKB/Akt and the induction of adipocyte differentiation (20,21). Recently, the results from a gene knockout animal study have strongly supported the results of in vitro cell line studies. In mice with PKB/Akt 1 and 2 gene knockouts, the development of adipose tissue is impaired along with other abnormalities, including muscle, bone, and skin (22). Thus, the PI 3-kinase-PKB/Akt signal system is important not only in the regulation of adipose tissue development but also in the development of other tissues originating from mesodermal cells.
Because the PI 3-kinase-PKB/Akt signal system promotes the adipocyte differentiation (18 -21) and mediates the IGF-1 receptor signal (13)(14)(15)(16), which is also the key inducer for 3T3-L1 adipocyte differentiation, it is clear that PI 3-kinase-PKB/Akt will be the important signal system in IGF-1 receptor signal-induced adipocyte differentiation. In the present study, by using RNA interference we have demonstrated that PKB/ Akt 1 is in a pivotal position for mediating the IGF-1 receptor signal during 3T3-L1 adipocyte differentiation. Without PKB/ Akt 1, neither the adipocyte differentiation process nor the downstream transcription factor can be activated by the receptor signal in 3T3-L1 cells.
For the effect of PI 3-kinase inhibitors (LY294002 and wortmannin) on 3T3-L1 adipocyte differentiation induction, post-confluent cells were pretreated with individual inhibitor at the indicated concentration for 1 h and then induced to differentiate following the standard differentiation induction protocol. The inhibitor was supplemented in the differentiation induction medium during the first 4 days (i.e. 2 days of induction with MIX, DEX, and insulin and an additional 2 days with insulin) and removed after day 4. Cell numbers were counted during the differentiation process, and the cells were stained with Oil Red-O on day 8.
Small Double Strand RNA-mediated Interference (RNAi)-The RNA interference experiment was carried out by using plasmid expressing small interfering RNA fragment (27). Sequences for RNAi were selected from the PKB/Akt 1 cDNA coding region started with GGG and were analyzed by BLAST research to ensure that they did not have significant sequence homology with other genes. The sequence for PKB/Akt 1 RNAi was 5Ј-GGGCTGAAGAGATGGAGGTGT-3Ј, corresponding to nucleotides 389 -409 in the coding region (translation initiation site ATG as nucleotide 1). The following two pairs of oligonucleotides were synthesized to construct the RNAi plasmid following the protocol described by Sui et al. (27): Akt-i-1, 5Ј-GGCTGAAGAGATGGAGGTGTA-3Ј (forward) and 5Ј-AGCTTACACCTCCATCTCTTCAGCC-3Ј (complement); Akt-i-2, 5Ј-AGCTTACACCTCCATCTCTTCAGCCCTTTTTG-3Ј (forward) and 5Ј-AATTCAAAAAGGGCTGAAGAGATGGAGGTGTA-3Ј (complement). The annealed Akt-i-1 double strand oligonucleotide was inserted into the U6 promoter Bluescript plasmid (27) between the ApaI (nuclease-blunted end) and the HindIII site to make the intermediate plasmid. The annealed Akt-i-2 oligonucleotide was then inserted into the HindIII and EcoRI sites of the intermediate plasmid to make the PKB/Akt 1 RNAi plasmid. Thus, the constructed PKB/Akt 1 RNAi plasmid can express an RNA fragment, which will form a 21-bp double strand RNA hairpin with a 6-bp loop. The sequence of the 21-bp double strand RNA hairpin is corresponding to the PKB/Akt 1 RNAi sequence. As control, the corresponding shuffle sequence with the same nucleotide composition but different sequence was designed. The shuffle sequence for the PKB/Akt 1 RNAi sequence was 5Ј-GGGATCGGAGTGTAGAG-GTGA-3Ј, and the two pairs of oligonucleotide for the construction of plasmid were Akt-Sh-1, 5Ј-GGATCGGAGTGTAGAGGTGAA-3Ј (forward) and 5Ј-AGCTTTCACCTCTACACTCCGATCC-3Ј (complement), and Akt-Sh-2, 5Ј-AGCTTTCACCTCTACACTCCGATCCCTTTTTG-3Ј (forward) and 5Ј-AATTCAAAAAGGGATCGGAGTGTAGAGGTGAA-3Ј (complement). The plasmids were constructed in the same way as the RNAi plasmids.
To make stable cell lines expressing the small interfering RNA fragment, the cassette of the U6 promoter with RNAi sequence insertion was digested from the constructed plasmids and inserted into pcDNA3 plasmid with the cytomegalovirus promoter deleted. The 3T3-L1 preadipocyte was transfected with the RNAi plasmids, and the stable cell lines were selected by the resistance to G418. Four plasmids (pAkt-i and pAkt-sh) were transfected into 3T3-L1 preadipocytes, respectively, and independent foci were selected and propagated.
Cell Transfection-1 ϫ 10 5 exponentially growing low passage 3T3-L1 preadipocytes were plated into a 60-mm culture dish and cultured in DMEM with 10% calf serum at 37°C in a CO 2 incubator for 24 h. The cells were then transfected with the desired plasmid by LipofectAMINE TM following the protocol provided by Invitrogen. 400 g/ml G418 was added to the medium 24 h after the transfection. G418-resistant foci were generally formed in 3 weeks. Individual focus was selected and propagated for further analysis.
Western Immunoblot-35-mm 3T3-L1 cell monolayers were lysed directly in 1ϫ boiling Laemmli SDS sample buffer with 20 mM dithiothreitol. Samples were subjected to SDS-PAGE and transferred to Immobilon-P membrane (Millipore). After blocking with 5% nonfat dried milk in 1ϫ Tween/Tris-buffered saline (TTBS) (0.05% Tween, 25 mM Tris-HCl, pH 7.5, and 150 mM NaCl) for 2 h at room temperature, membranes were incubated with primary antibody for 2 h at room temperature, followed by horseradish peroxidase-conjugated secondary antibody for 45 min. The targeted protein was revealed by enhanced chemiluminescence.
Immunoprecipitation-10-cm 3T3-L1 cell monolayers were washed with ice-cold phosphate-buffered saline (PBS) and lysed in 1% Triton X-100 buffer (1% Triton X-100, 50 mM Hepes, pH 7.4, 2.5 mM EDTA, 150 mM NaCl, 30 mM ␤-glycerophosphate, 1 mM sodium vanadate, 1 mM PMSF, and 2 l/ml protease inhibitor mixture 1 and 2 (28)). The cell lysate was homogenized and extracted at 4°C for 1 h. After centrifugation at 12,000 ϫ g for 15 min, the supernatant (1000 g of protein) was mixed with 1 g of primary antibody for 2 h at 4°C and then with protein A-agarose overnight at 4°C. The protein A-agarose was collected by centrifugation at 1000 ϫ g for 5 min, washed with lysis buffer four times, and resuspended in 1ϫ Laemmli SDS sample buffer for Western blot analysis.
Analysis of PKB/Akt Activation by IGF-1 Receptor Signaling in 3T3-L1 Cell during Differentiation Induction-Two-day post-confluent 3T3-L1 preadipocytes were fed with DMEM containing 10% fetal bovine serum (FBS) for 24 h to minimize the serum effect. The cells were then induced with hormone mixture (MIX, DEX, and insulin) or each individual hormone. The hormonal concentrations are the same as used for the standard differentiation induction. Cells were washed with ice-cold PBS and harvested in 1ϫ boiling Laemmli SDS sample buffer with 20 mM dithiothreitol. The samples were subjected to Western blot analysis.
For kinetic analysis of PKB/Akt activation by insulin, cells were treated with 1 M insulin and harvested at 1 and 2 min and up to 16 h after the insulin stimulation. For insulin concentration-dependent analysis, cells were treated with insulin at the indicated concentrations ( Fig. 2) for 20 min and then harvested for Western blot analysis.
Analysis of Inhibitor Effects on IGF-1 Receptor Autophosphorylation, IRS-1 Phosphorylation, and PKB/Akt Activation-After being fed with DMEM containing 10% FBS for 24 h, post-confluent cells were pretreated or not with LY294002, PD98059, or U0126 for 1 h and then stimulated with 1 M insulin. After the indicated times, the cells were harvested with SDS sample buffer for Western blot analysis of PKB/Akt activation.
After the inhibitor and insulin treatment, cells were washed with ice-cold PBS and harvested with immunoprecipitation buffer for analysis of IGF-1 receptor autophosphorylation and IRS-1 phosphorylation. After immunoprecipitation with the anti-IGF-1 receptor ␤-subunit antibody or anti-IRS-1 antibody, the samples were subjected to Western blot analysis with anti-phosphotyrosine antibody.
Analysis of PI 3-Kinase Activity-Post-confluent cells were fed with DMEM containing 1% FBS overnight to reduce the serum effect and then were pretreated or not with 20 M LY294002 for 1 h followed by 1 M insulin stimulation for 10 min. After washing twice with ice-cold PBS, cells were scraped and lysed in lysis buffer containing 50 mM Hepes, pH 7.4, 150 mM NaCl, 0.5 mM EGTA, 10% glycerol, 1% Triton X-100, 1 mM sodium vanadate, 2 mM PMSF, and 2 l/ml protease inhibitors mixture 1 and 2 (28). The cell lysate was centrifuged, and the supernatant was collected. 1 g of primary antibody (anti-phosphotyrosine or anti-p110␣ PI 3-kinase) was added to the clarified cell lysate (500 g of protein), incubated at 4°C overnight, and precipitated with 10 l of protein A-agarose beads. After washing twice with lysis buffer, once with 20 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, and 1 mM sodium vanadate buffer, and once with 20 mM Tris-HCl, pH 7.5, 100 mM NaCl, and 0.5 mM EGTA buffer, the beads were resuspended in 50 l of reaction mixture (20 mM Tris-HCl, pH 7.5, 100 mM NaCl, 10 mM MgCl 2 , 0.5 mM EGTA, 0.1 mg/ml phosphatidylinositol, 0.1 mg/ml phosphatidylserine, 0.2 mM adenosine, 20 M ATP, 10 Ci of [␥-32 P]ATP). The reaction was carried out at 30°C for 30 min and terminated by adding 100 l of 1 M HCl. The supernatant of the reaction mixture was extracted with 160 l of chloroform/methanol (1:1), and the chloroform/methanol extract was washed once with 80 l of methanol, 100 mM HCl and 2 mM EDTA buffer (1:1). The washed extract was spotted on a TLC plate (Whatman, LK6) that was precoated in 1% potassium oxalate, 40% methanol, and 1.2 mM EDTA buffer for 30 min at room temperature and baked at 100°C for 1 h. The chromatography was carried out with chloroform/methanol/NH 4 OH/water (45:35:1.5:8.5) buffer. Phosphatidylinositol 3-phosphate was revealed by Phosphor-Imager analysis.
Analysis of PKB/Akt 1 RNAi Cells-The G418-resistant cells transfected with RNAi vector were analyzed for their protein expression of PKB/Akt 1 and 2. Stable RNAi cell lines and control cells (wild type 3T3-L1 cell and shuffle-sequence vector-transfected cell) were lysed in 1ϫ Laemmli SDS sample buffer. Samples were then subjected to SDS-PAGE and Western blot with anti-PKB/Akt 1 and 2 antibodies, respectively.
For analysis of the phosphorylation of PKB/Akt 1 and 2, the RNAi cells and control cells were treated with or without insulin and LY294002 as described in Fig. 4. PKB/Akt 1 and PKB/Akt 2 were immunoprecipitated with anti-PKB/Akt 1 antibody and anti-PKB/Akt 2 antibody, respectively. The immunoprecipitated samples were then analyzed by immunoblot with anti-phospho-PKB/Akt antibody.
The analysis of PKB/Akt 1 activity in RNAi cells used the Akt kinase assay kit from Cell Signaling Technology, Inc., and followed the protocol provided by the manufacturer. Briefly, post-confluent RNAi cells and control cells were treated with 1 M insulin. At the indicated times, the cells were washed with ice-cold PBS and lysed in 1 ml of ice-cold lysis buffer containing 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM sodium vanadate, 1 g/ml leupeptin, and 1 mM PMSF. After centrifugation at 12,000 ϫ g for 10 min at 4°C to remove the insoluble material, the supernatant was collected, and the protein concentration was determined. An equal amount of cell lysate (500 g of protein) from each cell was mixed with 10 l of immobilized PKB/Akt 1 antibody incubated at 4°C for 3 h on a rotating wheel. The antibody beads were collected, washed once with lysis buffer and twice with kinase buffer containing 25 mM Tris-HCl, pH 7.5, 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM sodium vanadate, and 10 mM MgCl 2 , and resuspended in 40 l of kinase buffer supplemented with 200 M ATP and 1 g of glycogen synthase kinase-3 (GSK-3) fusion protein. After incubation at 30°C for 30 min, the reaction was terminated with 20 l of 3ϫ SDS sample buffer. Samples were then subjected to SDS-PAGE and blotted with anti-phospho-GSK-3 antibody.
The analysis of adipocyte differentiation, activation of PKB/Akt 1 and ERK, and phosphorylation of the IGF-1 receptor and IRS-1 followed the procedures described above.
For analysis of FKHR phosphorylation, PKB/Akt 1 RNAi cells were treated and harvested in the same way as for analysis of PKB/Akt 1 phosphorylation. The Western blot was carried out with FKHR and phospho-FKHR antibodies.

Inhibition of 3T3-L1 Adipogenesis by the Treatment of PI 3-Kinase Inhibitors, LY294002
and Wortmannin-In post-confluent 3T3-L1 preadipocytes, the adipocyte differentiation process was induced by the combination of MIX, DEX, and insulin. Previous studies indicate (4) that the IGF-1 receptor signal activated by insulin plays a key role in inducing the adipocyte differentiation process. PI 3-kinase-PKB/Akt is an important signal cascade for mediating IGF-1 receptor signal. By inhibiting PI 3-kinase with LY294002 or wortmannin, cellular events induced by PI 3-kinase-PKB/Akt-mediated IGF-1 receptor signal will be blocked. As shown in Fig. 1, in 3T3-L1 cells treated with LY294002 or wortmannin, the combination of inducers could no longer induce the adipocyte differentiation process. No adipocyte differentiation was observed in the PI 3-kinase inhibitor-treated 3T3-L1 cells. In addition, the mitotic clonal expansion induced during the differentiation process was also inhibited by PI 3-kinase inhibitors (Fig. 1, B and C).
The Activation of PI 3-Kinase-PKB/Akt by Insulin but Not by MIX and DEX-During differentiation induction with MIX, DEX, and insulin, insulin activated the IGF-1 receptor on the cell membrane and initiated the IGF-1 receptor signaling cascade (4). Although three hormones (MIX, DEX, and insulin) are used to initiate the 3T3-L1 adipocyte differentiation process, only insulin was capable of inducing mitotic clonal expansion (8). Thus, the ability of the PI 3-kinase inhibitor, LY294002 or wortmannin, to block the mitotic clonal expansion as well as adipocyte differentiation indicated the blocked IGF-1 receptor signaling cascade but not MIX and DEX signaling pathways by these inhibitors.
To ascertain the effect of PI 3-kinase inhibitor on the IGF-1 receptor signaling, signal molecules upstream of PI 3-kinase (IGF-1 receptor and IRS-1) and downstream of PI 3-kinase (PKB/Akt) were analyzed. The results indicated that the PI 3-kinase inhibitor, LY294002, had no inhibitory effect on the receptor itself. Neither ligand-stimulated receptor autophosphorylation nor the phosphorylation of IRS-1 (the direct downstream substrate of IGF-1 receptor kinase) by receptor tyrosine kinase was blocked by the treatment of LY294002 (Fig. 2, A and  B). PD98059 that blocks the MEK-ERK signal but not the PI 3-kinase signal was used as an additional control. It appeared that these inhibitors had no adversarial effect on the receptor itself. In the presence of the PI 3-kinase inhibitor, insulin could still activate the IGF-1 receptor and initiate the IGF-1 receptor signaling cascade.
As expected, LY294002 did not affect the signal molecules upstream of PI 3-kinase. However, the signal molecules downstream of PI 3-kinase and the kinase itself should be inhibited by the treatment with LY294002. As shown in Fig. 5, C and D, in 3T3-L1 preadipocyte insulin stimulation significantly activated the PI 3-kinase activity, as more phosphatidylinositol 3-phosphate was generated from phosphatidylinositol by PI 3-kinase immunoprecipitated from insulin-stimulated cells than control cells. In 3T3-L1 cells treated with LY294002, the activation of PI 3-kinase by insulin was completely inhibited. PI 3-kinase isolated from cells treated with both insulin and LY294002 exhibited little kinase activity by in vitro kinase activity analysis (Fig. 5, C and D). In addition, the PI 3-kinase activity could also be inhibited by LY294002 when the inhibitor was added directly to the PI 3-kinase isolated from insulinstimulated 3T3-L1 preadipocytes.
As the major downstream effector of PI 3-kinase, the activation of PKB/Akt is regulated by PI 3-kinase. Phosphatidylinositol 3-phosphate, the product of PI 3-kinase, is important for the activation of PKB/Akt. During 3T3-L1 preadipocyte differentiation induction, PKB/Akt was significantly activated (Fig. 3A). Antibody against the critical serine-phosphorylated peptide of PKB/Akt 1 (also cross-reactive with phosphorylated forms of PKB/Akt 2 and 3) detected a dramatic increase of the phosphorylated form of PKB/Akt (Fig. 3A). After stimulation, PKB/Akt 1 antibody detected a new PKB/Akt 1 protein with slow mobility on SDS-PAGE (Fig. 3A). Because the antibody against phospho-PKB/Akt was mouse monoclonal and antibody against PKB/Akt 1 was rabbit polyclonal, sequential Western blots on the same membrane with these two antibodies were carried out. The results indicated that the slow mobility PKB/Akt 1 protein appeared after the hormonal stimulation exactly corresponded to the band detected by the phospho-PKB/Akt antibody (results not shown). In addition, PKB/Akt 2 was presented in much less amounts in 3T3-L1 preadipocytes than PKB/Akt 1 (Fig. 4, C and D). Thus, the phosphorylated PKB/Akt band was most likely phospho-PKB/Akt 1.
To initiate the 3T3-L1 adipocyte differentiation process, three hormones (MIX, DEX, and insulin) were used. Of these three hormones, only insulin was capable of activating PKB/ Akt and induced significant phosphorylation of PKB/Akt 1 (Fig.  3A). MIX and DEX had little effect on the activation of PKB/ Akt. The activation of PKB/Akt 1 by insulin stimulation was also inhibited in the cells treated with LY294002 (Fig. 3C). With 20 M LY294002 treatment, the activation of PKB/Akt was completely inhibited. However, at 10 M, LY294002 only partially inhibited the PKB/Akt activation (results not shown). This was consistent with the effect of LY294002 on adipocyte differentiation (Fig. 1). In addition, the activation of PKB/Akt by insulin stimulation was not affected by PD98059 or U0126, two inhibitors of MEK (Fig. 3D), neither was insulin-stimulated activation of ERK affected by LY294002 (results not shown). The result that LY294002 did not block the insulinstimulated activation of ERK further supported the inhibition of PI 3-kinase but not the receptor tyrosine kinase by PI 3-kinase inhibitors. Thus, in 3T3-L1 cells, the PI 3-kinase-PKB/Akt signaling system was only activated by the insulin-stimulated IGF-1 receptor. This IGF-1 receptor-PI 3-kinase-PKB/Akt signaling cascade was independent from the IGF-1 receptor signal-activated MEK-ERK signaling cascade.

Involvement of PI 3-Kinase-PKB/Akt Signal Cascade in 3T3-L1 Adipocyte Differentiation Induction-
The insulin-stimulated activation of PI 3-kinase-PKB/Akt was blocked by the treatment of the PI 3-kinase inhibitor as well as the hormoneinduced 3T3-L1 adipocyte differentiation and mitotic clonal expansion. These studies with PI 3-kinase inhibitor indicated that the PI 3-kinase-PKB/Akt signaling system played a pivotal role in mediating the IGF-1 receptor signal for inducing 3T3-L1 adipocyte differentiation. However, these results represented the inhibition of both PKB/Akt 1 and 2, because the inhibitor, LY294002 or wortmannin, blocked the activity of PI 3-kinase that is the upstream signal molecule of PKB/Akt. Neither PKB/Akt 1 nor 2 could be activated in cells treated with these inhibitors. Western blot analysis indicated that PKB/Akt 1 was the dominant form of PKB/Akt in 3T3-L1 preadipocytes (results not shown). Thus, small double-strand RNA-mediated interference targeting PKB/Akt 1 was used to block specifically the protein expression of PKB/Akt 1. As shown in Fig. 4A, four independent 3T3-L1 cell lines stably transfected with PKB/Akt 1 RNAi vector were selected, and they had significantly decreased expression of PKB/Akt 1 protein. More important, in these PKB/Akt 1 RNAi cells the amount of PKB/Akt 2 protein was not changed compared with that in control cells (Fig. 4B). These RNAi cell lines provided the cell model to analyze specifically the function of PKB/Akt 1 in PI 3-kinase-PKB/Akt signaling system during 3T3-L1 cell differentiation induction.
In PKB/Akt 1 RNAi cells, the activation of PI 3-kinase-PKB/ Akt 2 should not be affected. In order to differentiate PKB/Akt 1 and 2, PKB/Akt 1 and 2 antibodies were used to immunoprecipitate PKB/Akt 1 and 2, respectively. The immunoprecipitated samples were then analyzed for their phosphorylation. As indicated in Fig. 4C, there was no difference between the control cells and PKB/Akt 1 RNAi cells in the phosphorylation of PKB/Akt 2, which was stimulated by insulin and blocked with the treatment of LY294002. However, the phospho-PKB/Akt 1 was significantly less in PKB/Akt 1 RNAi cells than in the control cells. Only a residual amount of phospho-PKB/Akt 1 was detected in these RNAi cells activated by insulin stimulation.
Although the amount of PKB/Akt 1 protein was dramatically decreased in these RNAi cells, it is possible that the residual PKB/Akt 1 still possessed enough enzymatic activity to suffice the signal mediation. Therefore, the PKB/Akt 1 kinase activity in RNAi cells was determined by its ability to phosphorylate GSK-3 (glycogen synthase kinase-3) fusion protein substrate (Fig. 4D). Without insulin stimulation, there was no detectable PKB/Akt 1 kinase activity in either control cells or RNAi cells. No GSK-3 fusion protein was phosphorylated. However, after insulin stimulation, PKB/Akt 1 kinase activity was dramatically induced in wild type 3T3-L1 preadipocytes and shuffle-sequence vector transfected cells. The GSK-3 fusion protein substrate was significantly phosphorylated by PKB/Akt 1 immunoprecipitated from these insulin-stimulated control cells (Fig. 4D). In contrast, little PKB/Akt 1 kinase activity was isolated from PKB/Akt 1 RNA interference cells after the insulin stimulation. It is clear that in PKB/Akt 1 RNAi cells the residual PKB/Akt 1 did not exhibit With the suppression of PKB/Akt 1 protein and its kinase activity by PKB/Akt 1 RNA interference, further analysis was carried out to investigate the effect of PKB/Akt 1 RNAi interference on the hormone-induced adipocyte differentiation process. As shown in Fig. 4, E and F, following the standard differentiation induction protocol, adipocyte differentiation and mitotic clonal expansion could no longer be induced in these PKB/Akt 1 RNAi cells. These results of suppressing PKB/Akt 1 by RNA interference paralleled the results of PI 3-kinase inhibitors (LY294002 and wortmannin). Taken together, our results demonstrated that PKB/Akt 1 is an important signal molecule in regulation of adipocyte differentiation process induced by IGF-1 receptor signal.
The effect of PI 3-kinase activity is reversed by PTEN, a phosphatase-dephosphorylating 3-position phosphate from phosphatidylinositol phosphates (29 -31). Thus, PI 3-kinase and PTEN regulated the intracellular phosphatidylinositol 3-phosphates that regulated the PKB/Akt 1 activity. The activation of PI 3-kinase and PKB/Akt mediated the signal, whereas the activity of PTEN attenuated the signal. With inhibition of PI 3-kinase by LY294002 or suppression of PKB/ Akt 1 by RNAi, the IGF-1 receptor signal-induced adipocyte differentiation process was blocked in 3T3-L1 preadipocytes ( Figs. 1 and 4). By using PTEN RNA interference, we suppressed the protein level of PTEN in 3T3-L1 preadipocytes (results not shown). These PTEN RNAi cells exhibited enhanced activation of PKB/Akt than control cells and differentiated into adipocytes normally when induced with differentiation induction hormones. Thus, these results suggested that it was the activation of the PI 3-kinase-PKB/Akt system by ligand stimulation, but not the attenuation of PTEN activity, that was important for inducing 3T3-L1 adipocyte differentiation.  indicates minutes)). PKB/Akt 1 was isolated from these cells by immunoprecipitation. Glycogen synthase kinase-3 was used as substrate for PKB/Akt 1. The phosphorylated glycogen synthase kinase-3 was revealed by Western blot with anti-phosphoglycogen synthase kinase-3 antibody. C, S, Ri1, and Ri2 are the same as in A. The arrow and pGSK-3 indicate the phosphorylated glycogen synthase kinase-3 on Western blot. E, adipocyte differentiation in PKB/Akt 1 RNAi cells. Cells were induced following the standard differentiation induction protocol, and triglyceride droplets were stained with Oil Red-O. C, S, Ri1, Ri2, Ri3, and Ri4 are the same as in A. F, mitotic clonal expansion in PKB/Akt 1 RNAi cells. During differentiation induction, cell numbers at different stages of the differentiation induction process were determined and normalized against that on day 0. The cell numbers of three culture plates were averaged. The fold increases of cell numbers during differentiation induction over that on day 0 are shown, and the six bars in each group represent the six cell lines: wild type 3T3-L1 cell (C), Shuffle sequence vector transfected cell (S), and four PKB/Akt 1 RNAi cells (Ri1, Ri2, Ri3, and Ri4). receptor autophosphorylation, IRS-1 tyrosine phosphorylation, and PI 3-kinase activation were analyzed. As shown in Fig. 5, A and B, insulin-stimulated autophosphorylation of IGF-1 receptor or the phosphorylation of the receptor kinase substrate, IRS-1, was not affected by the PKB/Akt 1 RNA interference.
The association with certain phosphotyrosine proteins via the Src homology 2 domain of the regulatory subunit of PI 3-kinase is important for the activation of PI 3-kinase (32). In addition, the activated growth factor receptor tyrosine kinases also phosphorylate both catalytic and regulatory subunits of PI 3-kinase on tyrosine. This tyrosine phosphorylation increased the PI 3-kinase activity (33)(34)(35)(36). Thus, the activated PI 3-kinase could be immunoprecipitated with antiphosphotyrosine antibody. All forms of activated PI 3-kinase should be present in the anti-phosphotyrosine antibody immunoprecipitated sample. As shown in Fig. 5D, from nonstimulated cells, the immunoprecipitated samples by antiphosphotyrosine antibody exhibited little PI 3-kinase activity. After insulin stimulation, the significantly increased PI 3-kinase activity was found in anti-phosphotyrosine antibody immunoprecipitated samples. This PI 3-kinase activity in anti-phosphotyrosine antibody immunoprecipitated samples from insulin-stimulated cells could be inhibited by the addition of LY294002 during in vitro PI 3-kinase activity assays. In addition, the activation of PI 3-kinase by insulin stimulation was inhibited in LY294002-treated cells. The anti-phosphotyrosine antibody immunoprecipitated samples from cells treated with both LY294002 and insulin had very low PI 3-kinase activity. It is important to point out that PKB/Akt 1 RNAi cells exhibited the same regulation of PI 3-kinase activity by insulin and PI 3-kinase inhibitor.
Because the immunoprecipitation by anti-phosphotyrosine antibody precipitated only the forms of PI 3-kinase associated with phosphotyrosine protein or tyrosine phosphorylated by receptor tyrosine kinase (presumably activated forms), the PI 3-kinase not associated with phosphotyrosine proteins or not tyrosine-phosphorylated (presumably nonactivated forms) was not immunoprecipitated by anti-phosphotyrosine antibody. Thus, antibody against PI 3-kinase was used to immunoprecipitate the PI 3-kinase protein itself. The PI 3-kinase activity was analyzed for the immunoprecipitated samples from cells treated with insulin or inhibitor. With antibody against p110␣, the catalytic subunit of ␣ type PI 3-kinase, all the forms of ␣ type PI 3-kinase (both activated and nonactivated) were immunoprecipitated. The samples immunoprecipitated by anti-p110␣ antibody from cells with various treatments contained the same amount PI 3-kinase protein (results not shown). However, the analysis of PI 3-kinase activity in these anti-p110␣ antibody immunoprecipitated samples indicated the increased kinase activity by insulin stimulation and inhibited kinase activity by LY294002 treatment (Fig. 5C). This result clearly showed the activation of ␣-type PI 3-kinase by insulin stimulation. Combined with the results from anti-phosphotyrosine immunoprecipitated samples, it was clear that insulin stimu-lation significantly activated the PI 3-kinase activity in both control and PKB/Akt 1 RNAi cells.
In PKB/Akt 1 RNAi cells, the signal pathway from the IGF-1 receptor down to PI 3-kinase was not affected (Fig. 5, A-D). However, the activation of PKB/Akt 1 by insulin stimulation was dramatically decreased because of the decreased amount of PKB/Akt 1 protein (Fig. 4 and Fig. 5E). Thus, PKB/Akt 1 RNAi should block the downstream signal pathway from PKB/Akt. In contrast to the normal signal transduction upstream of PKB/ Akt 1 (from IGF-1 receptor to PI 3-kinase), the insulin-stimulated phosphorylation of FKHR, the forkhead transcription factor, was dramatically decreased in PKB/Akt 1 RNAi cells but not in control cells (Fig. 5G). As a downstream molecule, FKHR was regulated by PKB/Akt-dependent phosphorylation (37)(38)(39)(40). In wild type 3T3-L1 preadipocytes, the phosphorylation of FKHR was stimulated by insulin and inhibited by the treatment of LY294002 (Fig. 5G). In PKB/Akt 1 RNAi cells, insulin-stimulated FKHR phosphorylation was decreased even without any treatment of PI 3-kinase inhibitor (Fig. 5G). Thus, PKB/Akt 1 RNA interference blocked the activation of the downstream signal molecules of PKB/Akt 1 but not the activation of the upstream signal molecules of PKB/Akt 1.
In order to assess the effect of PKB/Akt 1 RNA interference on other IGF-1 receptor signal pathways, insulin-induced activation of ERK was investigated. As shown in Fig. 5F, the activation of ERK by IGF-1 receptor signal was not affected by the suppression of PKB/Akt 1 protein expression. Thus, PKB/ Akt 1 RNA interference only affected the IGF-1 receptor signalactivated PI 3-kinase-PKB/Akt signal cascade but not the MEK-ERK signal cascade. DISCUSSION Intracellular signaling cascades are the link between the extracellular hormonal inducers and the transcription factors that receive the signals and cause the cellular responses. The inhibition of multiple cellular responses to IGF-1 receptor signaling in 3T3-L1 preadipocytes by PKB/Akt RNA interference or PI 3-kinase inhibitors indicates the essential position of the PI 3-kinase-PKB/Akt signal cascade in IGF-1 receptor signal transduction. In combination with the work of other researchers (18,19) in which the constitutively activated PKB/Akt is found to cause the spontaneous adipocyte differentiation in 3T3-L1 preadipocytes, the activation of PI 3-kinase-PKB/Akt appears to be essential and sufficient for the induction of 3T3-L1 adipocyte differentiation.
IGF-1 receptor exists in most type of cells and is essential for regulation of cell proliferation and apoptosis. IGF-1 receptor signal could stimulate cell growth and has anti-apoptotic effects (13,(41)(42)(43)(44). Beside these general effects, IGF-1 receptor signal is also involved in inducing some specific cellular responses in certain type of cells, for example adipogenesis in 3T3-L1 cells (4). To induce the specific cellular response in preadipocytes, the IGF-1 receptor can activate a specific signal pathway existing only in preadipocytes that lead to the adipogenesis or activate some general signal pathways that lead to the activation of specific effectors or a specific combination of effectors. The results from our current study supported signaling through the general signal pathways. In 3T3-L1 preadipocytes, the IGF-1 receptor signal leads to adipocyte differentiation (the specific cellular response) through the PI 3-kinase-PKB/Akt signal cascade, which is a general signal pathway and is activated by the IGF-1 receptor signal in most types of cells. Thus, it is the downstream effectors but not the signal pathways that confer the specificity of the cellular responses to IGF-1 receptor signaling in preadipocytes.
In the 3T3-L1 adipocyte differentiation process, the transcriptional regulation of adipocyte-specific genes during differentiation is relatively well characterized. Peroxisome proliferator-activated receptor ␥ (PPAR␥) and CCAAT/enhancerbinding protein ␣ (C/EBP␣) are two master regulators for controlling the adipogenic gene transcription program (2,3,45). However, the signal transduction pathways or network by which the differentiation inducers lead to the activation of adipogenic transcription program is relatively less understood.
The results in the current study implied an important association between the PI 3-kinase-PKB/Akt signal cascade and the transcription factors, PPAR␥ and C/EBP␣, in 3T3-L1 adipocyte differentiation induction.
The inactivation of transcription factor FKHR (Foxo 1) promotes the adipocyte differentiation, and its constitutively active mutant prevents adipogenesis (46). The direct phosphorylation of FKHR by PKB/Akt inhibits its transcriptional activity (47,48). Based on these studies, the inactivation of FKHR by PKB/Akt 1-catalyzed phosphorylation is one of the prerequisites for adipocyte differentiation. In our current study, PKB/ Akt 1 RNAi diminished the phosphorylation of FKHR (Fig. 5G) and prevented the adipocyte differentiation (Fig. 4E). Recently, it was reported (49) that FKHR interacts with PPAR␥ and reciprocally antagonizes each other's activity. Thus, it could be expected that the activation of PKB/Akt by the IGF-1 receptor signal leads to the phosphorylation and inactivation of transcription factor FKHR and removes its ''braking'' effect on the adipogenic transcription factors.