Cross-talk between the Platelet-derived Growth Factor and the Insulin Signaling Pathways in 3T3-L1 Adipocytes*

Phosphatidylinositol (PI) 3-kinase is activated by various growth factors such as PDGF (platelet-derived growth factor) and insulin. The aim of the present study was to determine whether PDGF could modulate insulin activation of PI 3-kinase in 3T3-L1 adipocytes. When cells were preincubated for 5–15 min with PDGF, PI 3-kinase activity associated to insulin receptor substrate 1 (IRS 1) in response to insulin was decreased, due to reduced association of the PI 3-kinase p85 subunit with IRS 1. In addition, following this PDGF pretreatment, the tyrosine phosphorylation of IRS 1 in response to insulin and its electrophoretic mobility were diminished. The change in the mobility of IRS 1 could be attributed to PDGF-induced serine/threonine phosphorylation of the protein which was partly inhibited by PI 3-kinase inhibitors. By contrast, epidermal growth factor, which does not stimulate PI 3-kinase, had no effect on the association of PI 3-kinase with IRS 1 in response to insulin. This series of results indicates that the PDGF-induced serine/threonine phosphorylation of IRS 1 could be due to activation of PI 3-kinase pathway. Furthermore, this phosphorylation of IRS 1 is associated with a decrease in its tyrosine phosphorylation by insulin and in its association with the p85 subunit of PI 3-kinase. This study suggests that a cross-talk exists between the different pathways stimulated by PDGF and insulin in intact cells.

Phosphatidylinositol (PI) 3-kinase 1 is a common element of the signaling pathway of a large number of tyrosine kinase receptors. PI 3-kinase is a heterodimer consisting of an 85-kDa regulatory subunit (p85) containing one Src homology 3 (SH3) domain and two Src homology 2 (SH2) domains (1)(2)(3) and an 110-kDa catalytic subunit (4). The catalytic subunit phosphorylates inositol lipids at the D-3 position of the inositol ring and has been shown to possess a serine kinase activity (5). By contrast, the p85 regulatory subunit functions as an adaptor which, via its SH2 domains, links PI 3-kinase to tyrosine-phosphorylated proteins such as autophosphorylated tyrosine kinase receptors (6). This association leads to the stimulation of the kinase activities of the p110 subunit (6,7). Since PI 3-kinase is activated by a large range of peptide growth factors, this enzyme activity appears to be implicated in various cellular responses including promotion of cell growth, regulation of cell differentiation, and metabolism (for review see Ref. 6). Despite this, each growth factor triggers distinct and specific biological responses in each particular cell type. In the present study, we looked at the effects of a prior stimulation by platelet-derived growth factor (PDGF) on the further ability of insulin to activate PI 3-kinase. We took advantage of the 3T3-L1 adipocytes where PI 3-kinase can be activated by both insulin and PDGF but not by EGF (8,9). When PDGF stimulates tyrosine phosphorylation of its receptor, the p85 regulatory subunit of PI 3-kinase associates to phosphorylated Tyr-Xaa-Xaa-Met motifs of the PDGF receptor (10,11). In contrast, insulin activates the tyrosine kinase activity of its receptor that subsequently phosphorylates insulin receptor substrate 1 (IRS 1) on YMXM motifs allowing binding of p85 to IRS 1 and PI 3-kinase activation (12)(13)(14). Our data show that pretreatment with PDGF decreases the ability of insulin to phosphorylate IRS 1 on tyrosine residues and consequently decreases both the amount of the p85 subunit and the PI 3-kinase activity associated with IRS 1. These results demonstrate that, in 3T3-L1 adipocytes, a crosstalk exists between the pathways stimulated by PDGF and insulin. This phenomenon could play a role in the regulation of the cell responses to growth factors and may explain the modulation of the cellular responses to different stimuli.

EXPERIMENTAL PROCEDURES
Antibodies-Antibodies to IRS 1 were obtained from a rabbit injected with a peptide corresponding to the sequence comprising amino acids 489 -507 of the protein and used at 1/100 dilution for immunoblotting. Antibodies used for immunoblotting of the p85 subunit of the PI 3-kinase and of phosphotyrosine-containing proteins were from UBI (Lake Placid, NY). Antibodies to phosphotyrosine used in immunoprecipitation assays were obtained after injection of a rabbit with phosphotyrosine coupled to bovine immunoglobulins.
Cell Culture-3T3-L1 fibroblasts were cultured in DMEM, 10% fetal calf serum and induced to differentiate into adipocytes as described (8). 3T3-L1 adipocytes were used between 8 and 12 days after initiation of differentiation, when more than 95% of the cells exhibited an adipocytelike phenotype. Sixteen hours before each experiment, the cells were changed to serum-free DMEM supplemented with 0.5% (w/v) bovine serum albumin.
Measurement of PI 3-Kinase Activity-3T3-L1 adipocytes were pretreated at 37°C without or with PDGF (50 ng/ml) or EGF (100 nM) for different periods. Then the cells were stimulated or not for 5 min with insulin (100 nM). The cells were solubilized for 40 min at 4°C in 700 l of buffer A (20 mM Tris, pH 7.4, 137 mM NaCl, 100 mM NaF, 10 mM * This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale (France), the University of Nice, and the Association pour la Recherche contre le Cancer ARC No. 2111. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 1 The abbreviations used are: PI 3-kinase, phosphatidylinositol 3-kinase; PDGF, platelet-derived growth factor; EGF, epidermal growth factor; DMEM, Dulbecco's modified Eagle's medium; PVDF, polyvinylidene difluoride; IRS 1, insulin receptor substrate 1; PAGE, polyacrylamide gel electrophoresis. EDTA, 2 mM Na 3 VO 4 , 10 mM pyrophosphate, 1 mM PMSF, 100 units/ml aprotinin) containing 1% Nonidet P-40 (v/v). Lysates were centrifuged for 10 min at 13,000 ϫ g. Supernatants were incubated for 2 h at 4°C with antibodies to IRS 1 or to phosphotyrosine coupled to protein A-Sepharose beads. Immune pellets were washed twice with each of the three following buffers: (a) phosphate-buffered saline containing 1% Nonidet P-40 and 200 M Na 3 VO 4 ; (b) 100 mM Tris, pH 7.4, 0.5 M LiCl, 200 M Na 3 VO 4 , and (c) 10 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 200 M Na 3 VO 4 . Bead-associated PI 3-kinase activity was assayed from the phosphorylation of PI in the presence of [␥-32 P]ATP as described previously (15). The reaction products were separated by thin layer chromatography. Quantification was performed after autoradiography by Cerenkov counting of the spot corresponding to phosphatidylinositol 3-phosphate.
Immunodetection of Phosphotyrosine-containing Proteins, PI 3-Kinase p85 Subunit, and IRS 1-Cells were pretreated with PDGF (50 ng/ml) for different periods and then stimulated with insulin (100 nM) for 5 min. When indicated, cells were treated with 100 nM wortmannin or with 50 M LY294002 for 20 min prior to PDGF stimulation. Then the cells were solubilized in buffer A containing 1% Nonidet P40. Cell lysates were incubated for 2 h with antibodies to IRS 1 preadsorbed on protein A-Sepharose. Pellets were washed as described above and were treated with Laemmli buffer, boiled for 10 min, and proteins analyzed by SDS-PAGE with a 7.5% acrylamide gel. Proteins were transferred to a polyvinylidene difluoride (PVDF) sheet. The sheet was incubated with blocking buffer (phosphate-buffered saline, 5% bovine serum albumin, w/v) for 2 h at room temperature and then overnight at 4°C with antibodies to IRS 1, to the p85 subunit of PI 3-kinase, or to phosphotyrosine. Membranes were washed three times (10 min each) in phosphate-buffered saline containing 1% Nonidet P-40. When antibodies to phosphotyrosine were used, a further incubation was performed for 1 h at room temperature with rabbit anti-mouse immunoglobulins. Finally, sheets were incubated for 1 h at room temperature with 125 I-protein A (5 ϫ 10 5 cpm/ml blocking buffer) and washed as above. Blots were submitted to autoradiography. 32 P-Orthophosphate Labeling of 3T3-L1 Adipocytes-3T3-L1 adipocytes (one 100-mm dish/condition) in phosphate-free DMEM supplemented with 0.5% bovine serum albumin (w/v) were labeled for 2.5 h at 37°C with [ 32 P]orthophosphate (1 mCi/5 ml). When indicated, cells were treated with 100 nM wortmannin for 20 min prior to a 15-or 60-min stimulation with 50 ng/ml PDGF. Then the cells were stimulated or not for 5 min with 100 nM insulin. Cells were washed with ice-cold buffer A and solubilized for 40 min at 4°C in 800 l of buffer A supplemented with 1% Nonidet P-40 (v/v). Samples were centrifuged at 13,000 ϫ g for 10 min. Supernatants were immunoprecipitated for 2 h at 4°C with antibodies to IRS 1 preadsorbed on protein A-Sepharose. The immune pellets were washed as described previously, treated with Laemmli buffer, boiled for 10 min, and separated on a 7.5% acrylamide SDS-PAGE. The gel was dried and autoradiographed.

PDGF Stimulates PI 3-Kinase Activity Associated with Phos-
photyrosine-containing Proteins-We first determined the time course of PI 3-kinase activation by PDGF in 3T3-L1 adipocytes. Cells were incubated for various periods with 50 ng/ml PDGF. PI 3-kinase activity was then measured in immunoprecipitates obtained with antibodies to phosphotyrosine (Fig. 1). PDGF maximally stimulated the PI 3-kinase activity between 5 and 15 min of stimulation. Then the activity progressively decreased to 60 min. This time course of PI 3-kinase activation paralleled the tyrosine phosphorylation of the PDGF receptor ( Fig. 1) suggesting that PI 3-kinase activity measured in antiphosphotyrosine immunoprecipitates was due to the association of the enzyme with the tyrosine-phosphorylated PDGF receptors. By contrast, EGF did not induce change in the PI 3-kinase activity in 3T3-L1 adipocytes but was able to activate mitogen-activated protein kinase (data not shown).
Effect of PDGF Pretreatment on Insulin-induced PI 3-Kinase Activity Associated with IRS 1-Since PI 3-kinase associates to activated PDGF receptors, we investigated whether a pretreatment of the cells by PDGF could interfere with a subsequent activation of PI 3-kinase by insulin. To test this hypothesis, 3T3-L1 adipocytes were stimulated with PDGF for different periods, before a subsequent 5-min insulin stimulation. PI 3-ki-nase activity was then measured in immunoprecipitates obtained with antibodies to IRS 1 (Fig. 2). Insulin alone markedly increased PI 3-kinase activity associated to IRS 1 (15-Ϯ 3-fold increase in four different experiments). When the cells were pretreated with PDGF for 5-15 min before insulin stimulation, PI 3-kinase activity associated to IRS 1 decreased. Maximal inhibition (32%) was observed after 15 min of PDGF pretreatment. No inhibition was observed at 30 min of PDGF pretreatment and PI 3-kinase activity decreased again by 30% after 60 min pretreatment. Different from the results observed with PDGF, EGF pretreatment of the cells did not modify the level of PI 3-kinase activity associated to IRS 1 in response to a subsequent insulin stimulation.

Effect of PDGF Pretreatment on Insulin-induced Tyrosine Phosphorylation of IRS 1 and Its Association with PI 3-Kinase-
The previous experiments indicate that the PI 3-kinase activity associated with IRS 1 following insulin stimulation is diminished by a PDGF pretreatment. We then wanted to determine whether it reflected a diminution of the amount of the enzyme associated with IRS 1 or an inhibition of the PI 3-kinase activity itself. 3T3-L1 adipocytes were pretreated with PDGF for different periods before insulin stimulation (5 min). Proteins were solubilized and immunoprecipitated with antibodies to IRS 1. The immunoprecipitated proteins were analyzed by SDS-PAGE and transferred to PVDF membranes. The PI 3-kinase associated with IRS 1 was detected by antibodies to FIG. 1. Effect of PDGF on PI 3-kinase activity associated to phosphotyrosine-containing proteins. 3T3-L1 adipocytes were incubated with or without 50 ng/ml PDGF for the indicated periods before homogenization and immunoprecipitation with antibodies to phosphotyrosine. PI 3-kinase assay was performed on the immune pellets as described under "Experimental Procedures." Quantification of the PI 3-kinase product was performed by scintillation counting. Results are expressed relative to the maximal activity measured after 5 min of PDGF stimulation (bottom panel). Results are the means Ϯ S.E. of three independent experiments. Top panel, cells were stimulated with 50 ng/ml PDGF for the times indicated. Cells were solubilized and proteins were separated on SDS-PAGE. Following transfer, phosphotyrosinecontaining proteins were immunodetected as described under "Experimental Procedures." The autoradiogram shows the results obtained in one typical experiment in which duplicate samples have been analyzed. its p85 subunit (Fig. 3A). In the absence of PDGF pretreatment, insulin induced the association of the PI 3-kinase p85 subunit with IRS 1. The pretreatment with PDGF for 5-60 min markedly reduced the amount of the p85 associated to IRS 1. In parallel, the tyrosine phosphorylation of proteins was visualized with an antibody to phosphotyrosine. The insulin-induced autophosphorylation of the ␤-subunit of its receptor was not affected by PDGF (data not shown). By contrast, a PDGF pretreatment induced a decrease in the insulin-induced tyrosine phosphorylation of IRS 1 (Fig. 3B). It should be noted that PDGF did not induce any tyrosine phosphorylation of IRS 1 (data not shown and Ref. 8). The decrease in insulin-induced IRS 1 tyrosine phosphorylation was not due to a modification in the amount of IRS 1 present in the immune pellets since it was similar in all conditions, indicating that PDGF pretreatment modified neither the total cellular amount of IRS 1 nor the ability of the antibody to immunoprecipitate IRS 1 (Fig. 3C). PDGF induced an increase in the apparent molecular weight of IRS 1, which was similar to that induced by insulin (Fig. 3C).
Effect of PDGF Treatment on IRS 1 Phosphorylation-To test whether the slower IRS 1 electrophoretic migration in PDGFtreated cells was due to a change in its phosphorylation state, 3T3-L1 adipocytes were labeled with [ 32 P]orthophosphate. Cells were then stimulated for 15 or 60 min without or with PDGF (50 ng/ml), before insulin treatment (100 nM, 5 min). At the end of the incubation, the proteins were solubilized and subjected to immunoprecipitation with antibodies to IRS 1. Immunoprecipitated proteins were analyzed by SDS-PAGE followed by autoradiography (Fig. 4). A low level of IRS 1 phosphorylation was observed in basal conditions. Both insulin and PDGF induced the incorporation of [ 32 P]orthophosphate into IRS 1 (2.5-and 2-fold increase, respectively). Since insulin (Fig.  3), but not PDGF (data not shown and Ref. 8), induced the tyrosine phosphorylation of IRS 1, these results suggest that the phosphorylation occurred mostly on tyrosine residues in the presence of insulin and on serine/threonine residues following PDGF stimulation. Furthermore, the phosphorylations in-duced by insulin and PDGF were not additive since an acute insulin stimulation after a PDGF pretreatment did not markedly increase 32 P incorporation into IRS 1. These results indicate that PDGF treatment could induce the phosphorylation of IRS 1 in intact cells on serine/threonine residues and decreased its ability to be tyrosine-phosphorylated by insulin and thus to bind the p85 subunit of PI 3-kinase.
Effect of PI 3-Kinase Inhibitors on PDGF-induced IRS 1 Phosphorylation-We then looked at the possibility that PI 3-kinase was involved in the phosphorylation of IRS 1 in response to PDGF. Indeed, PI 3-kinase is activated in response to PDGF (Refs. 8 and 9 and Fig. 1), and PI 3-kinase exerts a serine

FIG. 2. Effect of PDGF or EGF treatment on insulin-induced IRS 1-associated PI 3-kinase activity.
3T3-L1 adipocytes were incubated without or with PDGF (50 ng/ml) or EGF (100 nM) for different periods. The cells were then stimulated without insulin (100 nM) for 5 min before homogenization and immunoprecipitation with antibodies to IRS 1. The immune pellets were used to measure PI 3-kinase activity as described under "Experimental Procedures." Results are the means Ϯ S.E. from four to six independent experiments. The results are expressed as a percentage of the activity measured after 5 min of insulin stimulation without any pretreatment with PDGF or EGF.

FIG. 3. Effect of PDGF on the association of insulin-induced p85 with IRS 1 and on tyrosine phosphorylation of IRS 1.
Cells were incubated without (Ϫ) or with 50 ng/ml PDGF for the times indicated. Cells were then treated without (basal) or with 100 nM insulin for 5 min. Cellular homogenates were immunoadsorbed to antibodies to IRS 1. Proteins were separated by SDS-PAGE, transferred to PVDF sheets, and immunoblotted with antibodies to the p85 subunit of the PI 3-kinase (A), phosphotyrosine (B), or to IRS 1 (C) as described in methodology. Typical autoradiograms (out of four different experiments) are presented.

FIG. 4. PDGF induces phosphorylation of IRS 1.
3T3-L1 adipocytes were labeled for 2.5 h at 37°C with [ 32 P]orthophosphate. Then the cells were incubated for 15 or 60 min without or with 50 ng/ml PDGF prior to being stimulated with or without 100 nM insulin for 5 min as indicated. Cells were homogenized and immunoprecipitated with antibodies to IRS 1 as described under "Experimental Procedures." Proteins of the immune pellets were separated on SDS-PAGE, and labeled proteins were revealed by autoradiography. kinase activity toward IRS 1 in vitro (16 -18). Cells were incubated with two pharmacological inhibitors of the enzyme, i.e. with 100 nM wortmannin or 50 M LY294002, for 20 min prior to any incubation with PDGF or insulin (Fig. 5). IRS 1 tyrosine phosphorylation was then followed by immunoblotting with antibodies to phosphotyrosine as described previously. Pretreatment of the cells with wortmannin enhanced the tyrosine phosphorylation of IRS 1 induced by insulin alone, as also reported (19), whereas LY294002 was without effect. Pretreatment with LY294002 and wortmannin blocked the inhibitory effect of PDGF on the tyrosine phosphorylation of IRS 1 induced by insulin, and concomitantly, they prevented the electrophoretic mobility shift induced by PDGF (Fig. 5). When cells were labeled with [ 32 P]orthophosphate as described above, wortmannin markedly inhibited, by 80%, the serine/threonine phosphorylation induced by PDGF (Fig. 6).

DISCUSSION
The present study was designed to look at a possible modulation of insulin action by growth factors, focusing at the level of PI 3-kinase. PI 3-kinase is involved in multiple signaling pathways (20), such as those activated by insulin and PDGF. Activation of this enzyme results from the association of the p85 subunit of the PI 3-kinase to tyrosine-phosphorylated IRS 1 in response to insulin (12)(13)(14)21) or directly to tyrosinephosphorylated PDGF receptor in response to PDGF (10, 11). Here we have taken advantage of 3T3-L1 adipocytes, which express insulin, PDGF, and EGF receptors, to show that a pretreatment of the cells with PDGF, but not with EGF, decreased the PI 3-kinase activity and the amount of p85 subunit associated with IRS 1 in response to a subsequent insulin stimulation. This effect was rapid, since it occurred after a very short PDGF treatment (5-15 min). Moreover, in PDGF-treated cells, the insulin-induced tyrosine phosphorylation of IRS 1 was reduced, and this could explain the lower amount of p85 subunit associated with IRS 1 in response to insulin. The decrease in IRS 1 tyrosine phosphorylation was not due to an inhibition of insulin receptor tyrosine kinase activity since PDGF treatment did not modify the insulin-induced autophosphorylation of the receptor. By contrast, PDGF treatment lowered the electrophoretic mobility of IRS 1, the likely consequence of a phosphorylation of IRS 1 following PDGF treatment. Since PDGF did not induce any tyrosine phosphorylation of IRS 1 (data not shown and Ref. 8), the shift can only result from an increase in the serine/threonine phosphorylation of IRS 1 following PDGF treatment. These results are reminiscent of previous studies using okadaic acid (22) or tumor ne-crosis factor-␣ (23, 24) agents which simultaneously increase IRS 1 serine/threonine phosphorylation and decrease insulinstimulated IRS 1 tyrosine phosphorylation. It is thus tempting to think that the stimulation of IRS 1 serine/threonine phosphorylation induced by PDGF treatment was responsible for the inhibition of insulin-induced IRS 1 tyrosine phosphorylation. The mechanism by which the serine/threonine phosphorylation of IRS 1 could modulate its tyrosine phosphorylation is not well defined. In the case of tumor necrosis factor-␣ treatment, it was proposed that the phosphorylated form of IRS 1 could act as an inhibitor of the insulin receptor tyrosine kinase activity (24). This explanation seems unlikely in the present study since insulin receptor autophosphorylation was not altered by a PDGF pretreatment (data not shown). When cells were treated with okadaic acid, the tyrosine kinase activity of the insulin receptor was not changed, but the serine/threoninephosphorylated form of IRS 1 had a decreased ability to interact with the insulin receptor (22). Recently, it has been proposed that the phosphorylation of the serine residues of IRS 1, in the proximity of the tyrosine residues involved in the binding of PI 3-kinase, could decrease the association between PI 3-kinase and IRS 1 (25). Such a mechanism could be involved in the effect we observed in PDGF-treated cells.
Several studies have focused on the identification of kinases involved in the serine/threonine phosphorylation of IRS 1. Casein kinase II (26) and the serine kinase activity of PI 3-kinase (16 -18) have been shown to phosphorylate IRS 1 in vitro. A series of data argue for a role of PI 3-kinase in the phosphorylation of IRS 1 following PDGF stimulation. First, EGF, which is unable to activate PI 3-kinase in 3T3-L1 adipocytes but which activates multiple other signaling pathways such as the mitogen-activated protein kinase kinase cascade, does not modify the insulin-induced PI 3-kinase activity associated with IRS 1. Second, the serine/threonine phosphorylation of IRS 1 induced by PDGF is partly inhibited by two pharmacological inhibitors of PI 3-kinase as indicated by the reversal of the mobility shift of IRS 1 or by the cell orthophosphate labeling experiments. These data thus suggest that PI 3-kinase can induce a serine/ threonine phosphorylation of IRS 1 not only in vitro in immunoprecipitates obtained from insulin-stimulated cells (16 -18) but also in cells stimulated by PDGF. It is also possible that the phosphorylation of IRS 1 observed after a PDGF treatment results from the activation of another serine kinase(s) activated by the PI 3-kinase pathway. Indeed, novel serine/threonine kinases, targets of PI 3-kinase, such as the protein kinase B (27)(28)(29) and atypical protein kinase C and - (30,31) are activated by PDGF. Serine kinase activities, independent of the PI 3-kinase pathway, could also participate in the phosphorylation of IRS 1 since the PDGF-induced phosphorylation of IRS 1 was not completely blocked by wortmannin. Further studies are necessary to identify the enzyme implicated in the serine/ threonine phosphorylation of IRS 1.
The decrease in insulin-induced PI 3-kinase activity associated with IRS 1 was the likely consequence of the reduced IRS 1 tyrosine phosphorylation at early time points of PDGF pretreatment. However, other mechanisms could be involved in such an effect. Indeed, it has been shown that the p85 subunit becomes tyrosine-phosphorylated after a PDGF stimulation (32,33) and that this tyrosine phosphorylation decreased its ability to associate with tyrosine-containing proteins such as the activated PDGF receptor. Such an hypothesis is unlikely in our case since we never observed such a tyrosine phosphorylation of p85 in our cell line. It is also possible that the activated PDGF receptor binds a large amount of p85 subunit following PDGF stimulation thus decreasing, by a competitive process, the amount of p85 available to associate with IRS 1 after insulin stimulation. Indeed, a displacement of PI 3-kinase from the PDGF receptors toward IRS 1 following insulin stimulation occurred in a recent study (34) performed with cells transfected by high levels of insulin receptors while only the endogenous PDGF receptors were present. The present series of experiments has been performed in 3T3-L1 adipocytes, expressing only the endogenous insulin and PDGF receptors. Since it has recently been shown in NIH 3T3 cells that only 5% of the total pool of p85␣ associated with the PDGF receptor (33), this hypothesis is not very plausible to explain our results. It is still unknown why PI 3-kinase activity associated with IRS 1 returned to maximally insulin-stimulated values after 30 min of PDGF pretreatment. This occurred although serine/threonine phosphorylation of IRS 1 by PDGF was maintained for at least 60 min, concomitant with a constant decrease in the tyrosine phosphorylation of IRS 1 and of its association with the p85 subunit. The reason for this observation is still unclear.
Our results clearly show that intercommunications exist between the different growth factors and hormones signaling pathways. Enzymes common to multiple pathways, such as the PI 3-kinase, create the possibility of many different types of interactions among the various pathways. The diversity of these interactions could result in additive, synergistic, or an-tagonistic effects. The studies of such cross-talks between the different signaling pathways should be new approaches to understand the physiological significance of such events.