Evidence for two distinct phosphatidylinositol kinases in fibroblasts. Implications for cellular regulation

The SH3 domain of p56lck is involved in binding to phosphatidylinositol 3'-kinase from T lymphocytes

Many of the Src-like tyrosine kinases are thought to participate in multiprotein complexes that modulate transmembrane signalling through tyrosine phosphorylation. We have used in vitro binding studies employing bacterially expressed glutathione S-transferase-p56lck fusion proteins and cell extracts to map regions on p56lck that are involved in binding to phosphatidylinositol 3'-kinase (PI3K). Deletions within the SH3 domain of p56lck abolished binding of PI3K activity from T-cell lysates, whereas deletion of the SH2 domain caused only a slight reduction in the level of PI3K activity bound to p56lck sequences. The binding of PI3K from T-cell extracts to p56lck was not blocked by antiphosphotyrosine antibodies, but p56lck-bound PI3K activity was sensitive to phosphatase treatment. The SH3 domain of p56lck also bound the majority of PI3K activity from uninfected chicken embryo fibroblasts. However, a drastically different binding specificity was observed with use of extracts of Rous sarcoma virus v-src-transformed cells, in which the majority of PI3K activity bound to the SH2 domain of p56lck in a phosphotyrosine-dependent manner. These results suggest that are two modes of PI3K binding to p56lck, and presumably to other Src-like tyrosine kinases. In one mode, PI3K from T cells or uninfected chicken embryo fibroblasts binds predominantly to the SH3 domain of p56lck. In the other mode, involving PI3K from Rous sarcoma virus-transformed cells, binding is largely phosphotyrosine dependent and requires the SH2 domain of p56lck.

Cloning of a novel, ubiquitously expressed human phosphatidylinositol 3-kinase and identification of its binding site on p85

Phosphatidylinositol 3-kinase (PI 3-kinase) has been implicated as a participant in signaling pathways regulating cell growth by virtue of its activation in response to various mitogenic stimuli. Here we describe the cloning of a novel and ubiquitously expressed human PI 3-kinase. The 4.8-kb cDNA encodes a putative translation product of 1,070 amino acids which is 42% identical to bovine PI 3-kinase and 28% identical to Vps34, a Saccharomyces cerevisiae PI 3-kinase involved in vacuolar protein sorting. Human PI 3-kinase is also similar to Tor2, a yeast protein required for cell cycle progression. Northern (RNA) analysis demonstrated expression of human PI 3-kinase in all tissues and cell lines tested. Protein synthesized from an epitope-tagged cDNA had intrinsic PI 3-kinase activity and associated with the adaptor 85-kDa subunit of PI 3-kinase (p85) in intact cells, as did endogenous human PI 3-kinase. Coprecipitation assays showed that a 187-amino-acid domain between the two src homology 2 domains of p85 mediates interaction with PI 3-kinase in vitro and in intact cells. These results demonstrate the existence of different PI 3-kinase isoforms and define a family of genes encoding distinct PI 3-kinase catalytic subunits that can associate with p85.

C-FMS dependent HL-60 cell differentiation and regulation of RB gene expression

The dependence of induced myelomonocytic cell differentiation, and regulation of the RB tumor suppressor gene during this process, on the c-fms gene product, the CSF-1 lymphokine receptor, was determined in HL-60 promyelocytic leukemia cells. Adding a monoclonal antibody with specificity for the c-fms gene product to cells treated with various inducers of myelomonocytic or macrophage differentiation, including retinoic acid and 1,25-dihydroxy vitamin D3, inhibited the rate of differentiation. During the period of inducer treatment usually preceding onset of differentiation, longer periods of antibody exposure caused greater inhibition of differentiation. In a stable HL-60 transfectant overexpressing the CSF-1 receptor at the cell surface due to a constitutively driven c-fms trans gene, the rate of differentiation was enhanced compared to the wild type cell, consistent with a positive regulatory role for the CSF-1 receptor. The anti-fms antibody caused much less inhibition of differentiation in the transfectants than in wild type cells, consistent with a larger number of receptors causing reduced sensitivity. During the induced metabolic cascade leading to differentiation, the typical early down regulation of RB gene expression was inhibited by the antibody. The antibody itself caused an increase in RB expression per cell, which offset the decrease normally caused by differentiation inducers (1,25-dihydroxy vitamin D3 and retinoic acid). The changes in RB expression preceded changes in the RB protein to the hypophosphorylated state. Most of the RB protein in proliferating cells was phosphorylated and no significant accumulation of hypophosphorylated RB protein occurred until after onset of G0 arrest. Thus the metabolic cascade leading to myelomonocytic differentiation of HL-60 cells appears to be driven by a function of the c-fms protein. Inhibiting that process by attacking this receptor impedes differentiation and also compromises the early down regulation of RB tumor suppressor gene expression which normally precedes differentiation. These findings provide additional support for a potential role for down regulating RB expression in promoting cell differentiation, and suggest the possibility that RB may be either a target or intermediate mediator of CSF-1 actions.

cAMP antagonizes p21ras-directed activation of extracellular signal-regulated kinase 2 and phosphorylation of mSos nucleotide exchange factor

In fibroblasts, stimulation of receptor tyrosine kinases results in the activation of the extracellular signal-regulated kinase 2 (ERK2). The major signalling pathway employed by these receptors involves the activation of p21ras and raf-1 kinase. Here we show that in NIH3T3 and rat-1 fibroblasts, elevation of the intracellular cAMP level results in the inhibition of ERK2 activation induced by PDGF, EGF and insulin treatment. Analysis of various signalling intermediates shows that cAMP interferes at a site downstream of p21ras, but upstream of raf-1 kinase. Inhibition by cAMP depends on both the cAMP concentration and the absolute amount of p21ras molecules bound to GTP, suggesting a mechanism of competitive inhibition. Also TPA-induced, p21ras-independent, activation of raf-1 kinase and ERK2 is inhibited by cAMP. We have used the inhibitory effect of cAMP to investigate whether phosphorylation of mSos, a p21ras nucleotide exchange factor, is dependent on the activity of the raf-1 kinase/ERK2 pathway. We found that phosphorylation of mSos, as monitored by a mobility shift, is delayed with respect to p21ras and ERK2 activation and is inhibited by cAMP in a similar cell type- and concentration-dependent manner as the inactivation of ERK2. These results provide evidence for a model of p21ras-directed signalling towards ERK2 that feeds back on mSos by regulating its phosphorylation status and that can be negatively modulated by protein kinase A and positively modulated by protein kinase C action.

Erythropoietin induces the association of phosphatidylinositol 3'-kinase with a tyrosine-phosphorylated protein complex containing the erythropoietin receptor

Stimulation of sensitive cells with erythropoietin results in rapid induction of protein tyrosine phosphorylation. Other than tyrosine phosphorylation of one chain of the erythropoietin receptor, the identities of the remaining tyrosine-phosphorylated proteins are undefined. In this report, we demonstrate that the stimulation of the erythropoietin-sensitive human UT7 cells by erythropoietin rapidly resulted in the appearance of phosphatidylinositol 3-kinase activity in anti-phosphotyrosine immunoprecipitates. Erythropoietin action was rapid, detectable after as early as 1 min stimulation, transient, returning to control level after 30 min stimulation and was observed using the erythropoietin concentrations able to stimulate the cell proliferation. Anti-(phosphatidylinositol 3-kinase) antibodies specifically immunoprecipitated 125I-erythropoietin bound to its receptor, strongly suggesting that phosphatidylinositol 3-kinase associated with a protein complex containing the activated erythropoietin receptor. To confirm this result, phosphatidylinositol 3-kinase was immunoprecipitated from erythropoietin-stimulated cells using mild conditions followed by Western analysis using anti-phosphotyrosine antibodies. Five tyrosine phosphorylated proteins were revealed: the cloned chain of the erythropoietin receptor, the regulatory subunit of phosphatidylinositol 3-kinase and three unidentified proteins of 111, 97 and 64 kDa. None of these tyrosine phosphorylated proteins was detected in anti-(phosphatidylinositol 3-kinase) immunoprecipitates from unstimulated cells. Thus, our results show that phosphatidylinositol 3-kinase associates with a tyrosine-phosphorylated protein complex containing the activated erythropoietin receptor.

Activation of phosphatidylinositol-3 kinase by ligation of the interleukin-7 receptor on human thymocytes

Interleukin-7 (IL-7) is a glycoprotein that regulates lymphocyte precursor growth and differentiation. However, the exact mechanism whereby the IL-7 receptor (IL-7R) mediates these cell growth signals remains unknown. One of the earliest metabolic events linked to mitogenic responses in other growth factor receptor systems is the activation of phosphatidylinositol-3 kinase (PI-3 kinase). We demonstrate here that ligation of the IL-7R results in dose- and time-dependent increases in PI-3 kinase activity. These results suggest that PI-3 kinase is involved in signal transduction via the IL-7R in human thymocytes.

Activation of phosphatidylinositol 3-kinase and phosphatidylinositol 4-kinase during rat parotid acinar cell proliferation

We have recently shown that beta-adrenergic agonist, isoproterenol-induced parotid acinar cell proliferation is in part mediated by elevated levels of surface galactosyltransferase which undergoes interaction with the EGF-R. The receptor subsequently undergoes autophosphorylation on the tyrosine residues in a manner similar to its 'receptor-ligand' interaction (Purushotham et al. (1992) Biochem. J. 284, 767-776). In this study, we provide evidence for phosphatidylinositol 3-kinase and 4-kinase as cytoplasmic signalling proteins involved in both the isoproterenol and EGF-stimulated signal transduction upon in vitro and in-vivo stimulation of parotid acinar cells. Total cell lysate activity for the PtdIns 4-kinase was 2- and 3-fold higher than unstimulated control cells, while the PtdIns 3-kinase was 1.4- and 2.8-fold higher following stimulation by isoproterenol or EGF, respectively. Increases of 6- and 2-fold in phosphatidylinositol 3-kinase were observed in anti-phosphotyrosine-antibody-immunoprecipitated cell lysates upon in-vitro growth stimulation with isoproterenol or EGF, respectively. There was an increase in tyrosine phosphorylation of the holoenzyme and association of the p85 subunit of phosphatidylinositol 3-kinase with EGF-R in response to both isoproterenol and EGF treatments. This corresponded with the mobilization of p85 from the cytoplasm to the plasma membrane upon growth stimulation. These results further implicate the phosphoinositide metabolites in the second messenger signalling pathways of isoproterenol-induced rat parotid cell proliferation. The parallel utilization of EGF indicate that the post-transductional mechanisms of isoproterenol-induced acinar cell proliferation are similar to the growth-factor-mediated activation of intracellular signalling pathways for cell growth.

Receptor regulation of phosphoinositide 3-hydroxykinase in the NG115-401L-C3 neuronal cell line: stimulation by insulin-like growth factor-I

The activation of phosphoinositide 3-hydroxykinase (P13K) is currently believed to represent the critical regulatory event which leads to the production of a novel intracellular signal. We have examined the control of this pathway by a number of cell-surface receptors in NG115-401L-C3 neuronal cells. Insulin-like growth factor-I stimulated the accumulation of 3-phosphorylated inositol lipids in intact cells and the appearance of P13K in antiphosphotyrosine-antibody-directed immunoprecipitates prepared from lysed cells, suggesting that P13K had been activated by a mechanism involving a protein tyrosine kinase. In contrast, P13K in these cells was not regulated by a variety of G-protein-coupled receptors, nerve growth factor acting via a low affinity receptor, or receptors for transforming growth factor-beta and interleukin-1. The receptor-specificity of P13K activation in these cells places significant constraints on the possible physiological function(s) of this pathway.

Characterization of the bovine brain cytosolic phosphatidylinositol 3-kinase complex

Receptor-linked phosphatidylinositol (PtdIns) 3-kinase may generate a second-messenger signal. Here a large-scale purification of the bovine brain enzyme, based on methods developed by Morgan, Smith and Parker [(1990) Eur. J. Biochem. 191, 761-767] and Fry, Panayotou, Dhand, Ruiz-Larrea, Gout, Nguyen, Courtneidge and Waterfield [(1992) Biochem. J. 288, 383-393] is described. The purified enzyme is shown to be a heterodimer of 85 kDa and 110 kDa protein subunits (p85 and p110). Labelling with 5'-p-fluorosulphonylbenzoyladenosine shows that p110 contains an ATP-binding site and confers catalytic activity to the complex. The purified complex is known to be highly phosphorylated on both p85 alpha and p110 subunits, and dephosphorylation generates a deactivated complex, indicating that phosphorylation is an important covalent modification of the complex and may modulate PtdIns 3-kinase activity.

Different properties of monomer and heterodimer forms of phosphatidylinositol 3-kinases

Phosphatidylinositol (PI) 3-kinase plays an important role in the signalling of cell growth. We previously purified two types of PI 3-kinase from bovine thymus, a monomer from (PI 3-kinase I) and a heterodimer form (PI 3-kinase II) [Shibasaki, Homma and Takenawa (1991) J. Biol. Chem. 266, 8108-8114]. Here we examine the properties of these purified PI 3-kinases. Both PI 3-kinases were inhibited strongly by quercetin and isoquercetin. The inhibition of PI 3-kinase I and PI 3-kinase II by quercetin appears to be non-competitive, with apparent Ki values of 4 microM and 2.5 microM respectively. PI 3-kinase II, but not PI 3-kinase I, co-immunoprecipitates with pp60v-src and polyoma middle T (mT)/pp60c-src, even under conditions where the PI 3-kinases are not phosphorylated, suggesting that non-phosphorylated PI 3-kinase recognizes autophosphorylated pp60v-src. PI 3-kinase II is phosphorylated by pp60v-src and binds to it. Anti-p85 (85 kDa subunit of PI 3-kinase II) antibody precipitates not only PI 3-kinase II but also co-immunoprecipitates pp60v-src in src-transformed cells, suggesting that PI 3-kinase II binds to pp60v-src in vivo. These data suggest that the two PI 3-kinases may be regulated independently.

Modulating effects of the extracellular sequence of the human insulinlike growth factor I receptor on its transforming and tumorigenic potential

We reported previously that an N-terminally truncated insulinlike growth factor I receptor (IGFR) fused to avian sarcoma virus UR2 gag p19 had a greater transforming potential than did the native IGFR, but it failed to cause tumors in vivo. To investigate whether the 36 amino acids (aa) of the IGFR extracellular (EC) sequence in the gag-IGFR fusion protein encoded by the retrovirus UIGFR have a modulatory effect on the biological and biochemical properties of the protein, four mutants, NM1, NM2, NM3, and NM4 of the EC sequence were constructed. NM1 lacks the entire 36 aa residues; NM2 lacks the N-terminal 16 aa residues (aa 870 to 885), including two potential N-linked glycosylation sites of the EC sequence; NM3 contains a deletion of the C-terminal 20 aa residues (aa 886 to 905) of the EC sequence; and NM4 contains N-to-Q substitutions at both N-linked glycosylation sites. NM1 was the strongest of the four mutants in promoting anchorage-independent growth of transfected chicken embryo fibroblasts, while NM2 and NM4 had weaker transforming potential than did the original UIGFR virus. Only NM1 and NM3 were able to induce sarcomas in chickens. The four NM mutant-transformed cells expressed the expected proteins with comparable steady-state levels. The in vitro tyrosine kinase activity of P53NM1 was about fourfold higher than that of the parental P57-75UIGFR, whereas NM2 and NM4 proteins exhibited four- to fivefold-lower kinase activities. Despite lacking the IGFR EC sequence, P53NM1 formed covalent dimers similar to those formed by the parental P57-75UIGFR. Increased phosphatidylinositol (PI) 3-kinase activity was found to be associated with the mutant IGFR proteins. Among NM4 proteins. Elevated tyrosine phosphorylation of cellular proteins of 35, 120, 140, 160, and 170 kDa was detected in all mutant IGFR-transformed cells. We conclude that the EC 36-aa sequence of IGFR in the gag-IGFR fusion protein exerts intricate modulatory effects on the protein's transforming and tumorigenic potential. The 20 aa residues immediately upstream of the transmembrane domain have an inhibitory effect on the tumorigenic potential of gag-IGFR, whereas N-linked glycosylation within the EC sequence appears to have a positive effect on the transforming potential of UIGFR. Increased in vitro kinase activity and, to a lesser extent, in vivo tyrosine phosphorylation as well as the elevated association of PI 3-kinase activity with IGFR proteins seem to be correlated with the transforming potential of IGFR mutant proteins.

The insulinomimetic agents H2O2 and vanadate stimulate tyrosine phosphorylation of potential target proteins for the insulin receptor kinase in intact cells

H2O2 and vanadate are known insulinomimetic agents. Together they induce insulin's bioeffects with a potency which exceeds that seen with insulin, vanadate or H2O2 alone. We have previously shown that a combination of H2O2 and vanadate, when added to intact cells, rapidly stimulates protein tyrosine phosphorylation, owing to the inhibitory effects of these agents on intracellular protein tyrosine phosphatases (PTPases). Employing Western blotting with anti-phosphotyrosine antibodies, we have now identified in Chinese-hamster ovary (CHO) cells transfected with a wild-type insulin-receptor gene (CHO.T cells) several proteins (e.g. pp180, 125, 100, 60 and 52) whose phosphotyrosine content is rapidly increased upon treatment of the cells with a combination of insulin and 3 mM-H2O2. Tyrosine phosphorylation of these and additional proteins was further potentiated when 100 microM-sodium orthovanadate was added together with H2O2. The effects of insulin, insulin/H2O2, and H2O2/vanadate on tyrosine phosphorylation were markedly decreased in CHO cells transfected with an insulin-receptor gene where the twin tyrosines 1162 and 1163 were replaced with phenylalanine (CHO.YF-3 cells). Similarly, most of these proteins failed to undergo enhanced tyrosine phosphorylation in parental CHO cells incubated in the presence of insulin or the insulinomimetic agents. Our findings suggest that inhibition of PTPase activity by H2O2/vanadate augments the autophosphorylation of tyrosines 1162 and 1163 of the insulin receptor kinase, leading to its activation in an insulin-independent manner. As a result, tyrosine phosphorylation of potential targets for this enzyme takes place. Failure of H2O2/vanadate to induce phosphorylation of these proteins in receptor mutants lacking these twin tyrosine residues supports this hypothesis.

Expression and characterization of the p85 subunit of the phosphatidylinositol 3-kinase complex and a related p85 beta protein by using the baculovirus expression system

PtdIns 3-kinase associates with certain activated protein-tyrosine kinase receptors and with the pp60c-src/polyoma middle-T complex, suggesting that the enzyme is involved in growth regulation. The purified PtdIns 3-kinase appears to have two subunits, of 85 kDa and 110 kDa. Structural analysis at protein and cDNA levels revealed two forms of the 85 kDa subunit, one which associates with PtdIns 3-kinase activity termed p85 alpha, and a protein of unknown function, p85 beta. Both 85 kDa proteins contain src-homology regions 2 and 3 (SH2 and SH3), but lack enzymic activity, suggesting that they may be regulatory subunits of PtdIns 3-kinase. To probe their structure and function further, p85 alpha and p85 beta have been expressed and purified in large amounts from insect cells by using baculovirus vectors. Specific antisera detect p85 alpha, but not p85 beta, associated with PtdIns 3-kinase activity in various cell types. Co-expression studies in insect cells have shown that p85 alpha and p85 beta are substrates for the protein-tyrosine kinases of epidermal growth factor, colony-stimulating factor 1 and c-erbB2 receptors and the src family kinase p59c-fyn. Both p85 alpha and p85 beta form tight complexes with these protein-tyrosine kinases as measured by immunoprecipitation and kinase assays in vitro. The specificity of binding of free p85 is less restricted than that of p85 in the active PtdIns 3-kinase complex with the 110 kDa protein. The relevance of these results to growth-factor-induced PtdIns 3-kinase activation is discussed.

Purification and characterization of a phosphatidylinositol 3-kinase complex from bovine brain by using phosphopeptide affinity columns

Specific phosphorylated tyrosine residues in the kinase insert region of the human platelet-derived-growth-factor beta-receptor mediate the formation of multienzyme complexes with this receptor. When phosphorylated, tyrosine residue 751 within the kinase insert region mediates binding of PtdIns 3-kinase to this receptor. A 17-amino-acid peptide containing this tyrosine residue was synthesized, phosphorylated by using epidermal-growth-factor receptor and then coupled to an Actigel matrix. The tyrosine-751 phosphopeptide column is used here as a final affinity step in the purification of the PtdIns 3-kinase from bovine brain to apparent homogeneity. The active resin-bound PtdIns 3-kinase is composed of two polypeptides, p110 and p85, which are elutable with SDS-containing buffers and detectable by silver staining of polyacrylamide gels. The 85 kDa protein is shown to be identical with the recently cloned p85 alpha. Phosphotyrosine is demonstrated to be an essential part of the structure required for binding of both of these proteins and PtdIns 3-kinase activity to this peptide. The active PtdIns 3-kinase complex from bovine brain, but not recombinant p85 subunits, shows specificity for binding to phosphopeptides containing a YXXM consensus sequence. Neither PtdIns 3-kinase activity, nor the complex of p85 and 110 kDa proteins, binds to several other phosphopeptide affinity columns lacking this sequence motif. The selectivity of binding of baculovirus-expressed free p85 alpha subunit of bovine brain PtdIns 3-kinase, the closely related protein p85 beta and purified bovine brain PtdIns 3-kinase to these and other phosphopeptide columns is examined.

Functional asymmetry of the regions juxtaposed to the membrane-binding sequence of polyomavirus middle T antigen

The functional importance of the two clusters of positively charged amino acids which flank the hydrophobic membrane-anchoring sequence of polyomavirus middle T (mT) protein has been investigated by using site-directed mutagenesis. A clear asymmetry was apparent. No effect on transformation was seen following multiple alterations or complete removal of the cluster at the carboxyl end of the protein. In contrast, a single substitution replacing the first arginine amino terminal to the hydrophobic stretch with glutamic acid, but not with lysine, histidine, or methionine, produced a partially transformation-defective mutant with a novel phenotype. This mutant failed to confer anchorage-independent growth on F111 established rat embryo fibroblasts but induced foci with altered morphology compared with wild-type mT. Biochemical studies on this mutant revealed that F111 clones expressing levels of mutant mT equivalent to those of wild-type controls showed a 65% reduction in pp60c-src activation and an 87% reduction in mT-associated phosphatidylinositol 3-kinase activity. However, F111 clones expressing seven times more mutant mT than did wild-type controls showed equal or greater levels of kinase activities yet remained incompletely transformed. Possible mechanisms involving this transformation-sensitive region of mT are discussed.

Functional asymmetry of the regions juxtaposed to the membrane-binding sequence of polyomavirus middle T antigen

The functional importance of the two clusters of positively charged amino acids which flank the hydrophobic membrane-anchoring sequence of polyomavirus middle T (mT) protein has been investigated by using site-directed mutagenesis. A clear asymmetry was apparent. No effect on transformation was seen following multiple alterations or complete removal of the cluster at the carboxyl end of the protein. In contrast, a single substitution replacing the first arginine amino terminal to the hydrophobic stretch with glutamic acid, but not with lysine, histidine, or methionine, produced a partially transformation-defective mutant with a novel phenotype. This mutant failed to confer anchorage-independent growth on F111 established rat embryo fibroblasts but induced foci with altered morphology compared with wild-type mT. Biochemical studies on this mutant revealed that F111 clones expressing levels of mutant mT equivalent to those of wild-type controls showed a 65% reduction in pp60c-src activation and an 87% reduction in mT-associated phosphatidylinositol 3-kinase activity. However, F111 clones expressing seven times more mutant mT than did wild-type controls showed equal or greater levels of kinase activities yet remained incompletely transformed. Possible mechanisms involving this transformation-sensitive region of mT are discussed.

Polyomavirus middle T-antigen NPTY mutants

A polyomavirus middle T-antigen (MTAg) mutant containing a substitution of Leu for Pro at amino acid 248 has previously been described as completely transformation defective (B. J. Druker, L. Ling, B. Cohen, T. M. Roberts, and B. S. Schaffhausen, J. Virol. 64:4454-4461, 1990). This mutant had no alterations in associated proteins or associated kinase activities compared with wild-type MTAg. Pro-248 lies in a tetrameric sequence, NPTY, which is reminiscent of the so-called NPXY sequence in the low-density-lipoprotein receptor. In the low-density-lipoprotein receptor, mutations in the NPXY motif but not in the surrounding amino acids abolish receptor function, apparently by decreasing receptor internalization (W. Chen, J. L. Goldstein, and M. S. Brown, J. Biol. Chem. 265:3116-3123, 1990). To determine whether this sequence represents a functional motif in MTAg as well, a series of single amino acid substitutions was constructed in this region of MTAg. All of the mutations of N, P, T, or Y, including the relatively conservative substitution of Ser for Thr at amino acid 249, resulted in a transformation-defective MTAg, whereas mutations outside of this sequence allowed mutants to retain near-wild-type transformation capabilities. Transformation-defective mutants with mutations in the NPTY region behaved similarly to the mutant with the original Pro-248-to-Leu-248 mutation when assayed for associated proteins and activities in vitro; that is, they retained a full complement of wild-type activities and associated proteins. Further, insertion of the tetrameric sequence NPTY downstream of the mutated motif restored transforming abilities to these mutants. Thus, the tetrameric sequence NPTY in MTAg appears to represent a well-defined functional motif of MTAg.

Activation of phosphatidylinositol-3 kinase by nerve growth factor involves indirect coupling of the trk proto-oncogene with src homology 2 domains

Growth factor receptor tyrosine kinases can form stable associations with intracellular proteins that contain src homology (SH) 2 domains, including the p85 regulatory subunit of phosphatidylinositol (PI)-3 kinase. The activation of this enzyme by growth factors is evaluated in PC12 pheochromocytoma cells and NIH 3T3 fibroblasts expressing the pp140c-trk nerve growth factor (NGF) receptor (3T3-c-trk). NGF causes the rapid stimulation of PI-3 kinase activity detected in anti-phosphotyrosine, but not in anti-trk, immunoprecipitates. This effect coincides with the tyrosine phosphorylation of two proteins, with molecular masses of of 100 kd and 110 kd, that coimmunoprecipitate with p85. Similar phosphorylation patterns are induced when an immobilized fusion protein containing the amino-terminal SH2 domain of p85 is used to precipitate tyrosine-phosphorylated proteins. Thus, although NGF produces the rapid activation of PI-3 kinase through a mechanism that involves tyrosine phosphorylation, there is no evidence for tyrosine phosphorylation of p85, or for its ligand-dependent association with the NGF receptor. Perhaps another phosphoprotein may link the NGF receptor to this enzyme.

Signal transduction in neutrophil activation. Phosphatidylinositol 3-kinase is stimulated without tyrosine phosphorylation

Treatment of human neutrophils with the peptide f-Met-Leu-Phe (FMLP) results in neutrophil activation concomitant with stimulation of phosphatidylinositol (PtdIns) 3-kinase activity as measured by production of PtdIns-3,4,5-P3 in [32P]orthophosphate labeled cells. Antiphosphotyrosine immunoprecipitates were assayed for PtdIns 3-kinase activity; essentially no activity was present in lysates from either stimulated or unstimulated cells. The 85 kDa regulatory subunit of PtdIns 3-kinase, which normally serves as a substrate for tyrosine kinases, was not detected by SDS-PAGE or Western blot analysis in antiphosphotyrosine immunoprecipitates. In addition, no radioactive band corresponding to PtdIns 3-kinase was observed by SDS-PAGE following antiPtdIns 3-kinase immunoprecipitations. However, immunoprecipitates using polyclonal antibodies against PtdIns 3-kinase showed high PtdIns 3-kinase activity in neutrophil lysates and the 85 kDa subunit of PtdIns 3-kinase was detected in Western blots; no differences in activity were observed in FMLP-stimulated and unstimulated cells. These results suggest that, in contrast to polypeptide growth factor signal transduction systems, the activation of PtdIns 3-kinase by FMLP does not require tyrosine phosphorylation.

Phosphatidylinositol 3'-kinase is activated by association with IRS-1 during insulin stimulation

IRS-1 undergoes rapid tyrosine phosphorylation during insulin stimulation and forms a stable complex containing the 85 kDa subunit (p85) of the phosphatidylinositol (PtdIns) 3'-kinase, but p85 is not tyrosyl phosphorylated. IRS-1 contains nine tyrosine phosphorylation sites in YXXM (Tyr-Xxx-Xxx-Met) motifs. Formation of the IRS-1-PtdIns 3'-kinase complex in vitro is inhibited by synthetic peptides containing phosphorylated YXXM motifs, suggesting that the binding of PtdIns 3'-kinase to IRS-1 is mediated through the SH2 (src homology-2) domains of p85. Furthermore, overexpression of IRS-1 potentiates the activation of PtdIns 3-kinase in insulin-stimulated cells, and tyrosyl phosphorylated IRS-1 or peptides containing phosphorylated YXXM motifs activate PtdIns 3'-kinase in vitro. We conclude that the binding of tyrosyl phosphorylated IRS-1 to the SH2 domains of p85 is the critical step that activates PtdIns 3'-kinase during insulin stimulation.

Phosphatidylinositol 3-kinase: structure and expression of the 110 kd catalytic subunit

Purified bovine brain phosphatidylinositol 3-kinase (Pl3-kinase) is composed of 85 kd and 110 kd subunits. The 85 kd subunit (p85 alpha) lacks Pl3-kinase activity and acts as an adaptor, coupling the 110 kd subunit (p110) to activated protein tyrosine kinases. Here the characterization of the p110 subunit is presented. cDNA cloning reveals p110 to be a 1068 aa protein related to Vps34p, a S. cerevisiae protein involved in the sorting of proteins to the vacuole. p110 expressed in insect cells possesses Pl3-kinase activity and associates with p85 alpha into an active p85 alpha-p110 complex that binds the activated colony-stimulating factor 1 receptor. p110 expressed in COS-1 cells is catalytically active only when complexed with p85 alpha.

An Epstein-Barr virus transformation-associated membrane protein interacts with src family tyrosine kinases

In latently infected growth-transformed human lymphocytes, Epstein-Barr virus (EBV) encodes two integral plasma membrane proteins: LMP1, which constitutively induces B-lymphocyte activation and intercellular adhesion, and LMP2A, which associates with LMP1 and is a tyrosine kinase substrate. We now demonstrate that LMP2A associates with src family protein tyrosine kinases, particularly lyn kinase, in nonionic detergent extracts of transfected B lymphoma cells or in extracts of EBV-transformed B lymphocytes. The LMP2A and tyrosine kinase association is stable in nonionic detergents and includes a 70-kDa cell protein which is also an in vitro or in vivo kinase substrate. This LMP2A association with B-lymphocyte src family tyrosine kinases is likely to be an important pathway in EBV's effects on cell growth.

The inhibition of phosphatidylinositol 3-kinase by quercetin and analogs

Phosphatidylinositol (PtdIns) 3-kinase is an enzyme involved in cellular responses to growth factors. Quercetin (2-(3,4-dihydroxyphenyl)-3,5,7-trihydroxy-4H-1-benzopyrano-4-one), a naturally occuring bioflavinoid, was found to inhibit PtdIns 3-kinase with an IC50 of 1.3 micrograms/ml (3.8 microM); inhibition appears to be directed towards the ATP binding site of the kinase. Analogs of quercetin were also investigated as PtdIns 3-kinase inhibitors, with the most potent compounds exhibiting IC50's in the range of 1.7-8.4 micrograms/ml (5-19 microM). In contrast, genistein, a potent tyrosine kinase inhibitor of the isoflavone class, did not inhibit PtdIns 3-kinase significantly (IC50 greater than 30 micrograms/ml). These findings suggest that flavinoids may serve as potent inhibitors of PtdIns 3-kinase. Furthermore, the enzyme is much more sensitive to substituents at the 3-position of the flavinoid ring than are other protein and PtdIns kinases, suggesting that specific inhibitors of PtdIns 3-kinase can be developed to explore the biological role of the enzyme in cellular proliferation and growth factor response.

Insulin stimulates phosphatidylinositol-3-kinase activity in rat adipocytes

Phosphatidylinositol (PtdIns) 3-kinase is thought to participate in the signal transduction pathways initiated by the activation of receptor tyrosine kinases including the insulin receptor. To approach the physiological relevance of this enzyme in insulin signaling, we studied the activation of PtdIns-3-kinase in adipocytes, a major insulin target tissue for glucose transport and utilisation. To analyze possible interactions of the enzyme with cellular proteins, immunoprecipitations with the following antibodies were performed: (a) anti-phosphotyrosine antibodies, (b) two antibodies to the 85-kDa subunit of PtdIns-3-kinase (p85) and (c) an antibody to the 185-kDa major insulin receptor substrate (p185). We show that in cell extracts from adipocytes exposed to insulin, and after immunoprecipitation with an anti-phosphotyrosine antibody and an antibody to p85, we are able to detect a PtdIns-3-kinase activity stimulated by the hormone. Similarly, after immunoprecipitation with an antibody to p185, an increase in the PtdIns-3-kinase activity could be demonstrated. Taken together these results suggest that, upon insulin stimulation of fat cells, PtdIns-3-kinase itself is tyrosine phosphorylated and/or associated with an insulin receptor substrate, such as p185, which could function as a link between the insulin receptor and PtdIns-3-kinase. The PtdIns-3-kinase was activated within 1 min of exposure to insulin, and the half-maximal effect was reached at the same concentration, i.e. 3 nM, as for stimulation of the insulin receptor kinase. Subcellular fractionation showed that PtdIns-3-kinase activity was found both in the membranes and in the cytosol. Further, immunoprecipitation with an antibody to p85, which possesses the capacity to activate PtdIns-3-kinase, suggests that the presence of the enzyme in the membrane may be due to an insulin-induced recruitment of the PtdIns-3-kinase from the cytosol to the membrane. Finally, we used isoproterenol, which exerts antagonistic effects on insulin action. This drug was found to inhibit both the PtdIns-3-kinase and the insulin receptor activation by insulin, suggesting that the activation of the PtdIns-3-kinase was closely regulated by the insulin receptor tyrosine kinase. The occurrence of an insulin-stimulated PtdIns-3-kinase in adipocytes leads us to propose that this enzyme might be implicated in the generation of metabolic responses induced by insulin.

Platelet-derived growth factor stimulates synthesis of PtdIns(3,4,5)P3 by activating a PtdIns(4,5)P2 3-OH kinase

Although the hormone-stimulated synthesis of 3-phosphorylated inositol lipids is known to form an intracellular signalling system, there is no consensus on the crucial receptor-regulated event in this pathway and it is still not clear which of the intermediates represent potential output signals. We show here that the key step in the synthesis of 3-phosphorylated inositol lipids in 3T3 cells stimulated by platelet-derived growth factor is the activation of a phosphatidylinositol(4,5)-bisphosphate (3)-hydroxy (PtdIns(4,5)P2 3-OH) kinase. A similar conclusion has been applied to explain the actions of formyl-Met-Leu-Phe on neutrophils, and it may be that receptors that couple through intrinsic tyrosine kinases or through G proteins stimulate the same step in 3-phosphorylated inositol lipid metabolism. The close parallel between these two mechanisms for the activation of PtdIns(4,5)P2 3-OH kinase and those described for the activation of another key signalling enzyme, phospholipase C (ref. 7), focuses attention on the product of the PtdIns(4,5)P2 3-OH kinase, PtdIns(3,4,5)P3, as a possible new second messenger.

GTPase-activating protein and phosphatidylinositol 3-kinase bind to distinct regions of the platelet-derived growth factor receptor beta subunit

In response to binding of platelet-derived growth factor (PDGF), the PDGF receptor (PDGFR) beta subunit is phosphorylated on tyrosine residues and associates with numerous signal transduction enzymes, including the GTPase-activating protein of ras (GAP) and phosphatidylinositol 3-kinase (PI3K). Previous studies have shown that association of PI3K requires phosphorylation of tyrosine 751 (Y751) in the kinase insert and that this region of receptor forms at least a portion of the binding site for PI3K. In this study, the in vitro binding of GAP to the PDGFR was investigated. Like PI3K, GAP associates only with receptors that have been permitted to autophosphorylate, and GAP itself does not require tyrosine phosphate in order to stably associate with the phosphorylated PDGFR. To define which tyrosine residues are required for GAP binding, a panel of PDGFR phosphorylation site mutants was tested. Mutation of Y771 reduced the amount of GAP that associates to an undetectable level. In contrast, the F771 (phenylalanine at 771) mutant bound wild-type levels of PI3K, whereas the F740 and F751 mutants bound 3 and 23%, respectively, of the wild-type levels of PI3K but wild-type levels of GAP. The F740/F751 double mutant associated with wild-type levels of GAP, but no detectable PI3K activity, while the F740/F751/F771 triple mutant could not bind either GAP or PI3K. The in vitro and in vivo associations of GAP and PI3K activity to these PDGFR mutants were indistinguishable. The distinct tyrosine residue requirements suggest that GAP and PI3K bind different regions of the PDGFR. This possibility was also supported by the observation that the antibody to the PDGFR kinase insert Y751 region that blocks association of PI3K had only a minor effect on the in vitro binding of GAP. In addition, highly purified PI3K and GAP associated in the absence of other cellular proteins and neither cooperated nor competed with each other's binding to the PDGFR. Taken together, these studies indicate that GAP and PI3K bind directly to the PDGFR and have discrete binding sites that include portions of the kinase insert domain.

Cytosolic calcium oscillations induced by hepatocyte growth factor (HGF) in single fura-2-loaded cultured hepatocytes: effects of extracellular calcium and protein kinase C

Hepatocyte growth factor (HGF) induced the periodic fluctuations of cytosolic calcium concentration ([Ca2+]i) in primary cultured rat hepatocytes, which were dependent on extracellular calcium. The HGF-induced [Ca2+]i oscillations were suppressed by the pretreatment with phorbol 12-myristate 13-acetate (PMA). Administration of PMA during oscillations also caused their blockade, but the subsequent addition of protein kinase C (PKC) inhibitor H-7 reversed the inhibitory effects of PMA, thereby resulting in the resumption of the oscillatory responses. Moreover, the prior exposure to H-7 caused apparent increases in [Ca2+]i spike peaks elicited by HGF. These results suggest a negative modulation via PKC in HGF-induced repetitive [Ca2+]i transients. The absence of HGF-induced oscillations after the thapsigargin treatment indicates that the agonist-sensitive intracellular Ca2+ pool plays a crucial role in the [Ca2+] oscillations.

GTPase-activating protein and phosphatidylinositol 3-kinase bind to distinct regions of the platelet-derived growth factor receptor beta subunit

In response to binding of platelet-derived growth factor (PDGF), the PDGF receptor (PDGFR) beta subunit is phosphorylated on tyrosine residues and associates with numerous signal transduction enzymes, including the GTPase-activating protein of ras (GAP) and phosphatidylinositol 3-kinase (PI3K). Previous studies have shown that association of PI3K requires phosphorylation of tyrosine 751 (Y751) in the kinase insert and that this region of receptor forms at least a portion of the binding site for PI3K. In this study, the in vitro binding of GAP to the PDGFR was investigated. Like PI3K, GAP associates only with receptors that have been permitted to autophosphorylate, and GAP itself does not require tyrosine phosphate in order to stably associate with the phosphorylated PDGFR. To define which tyrosine residues are required for GAP binding, a panel of PDGFR phosphorylation site mutants was tested. Mutation of Y771 reduced the amount of GAP that associates to an undetectable level. In contrast, the F771 (phenylalanine at 771) mutant bound wild-type levels of PI3K, whereas the F740 and F751 mutants bound 3 and 23%, respectively, of the wild-type levels of PI3K but wild-type levels of GAP. The F740/F751 double mutant associated with wild-type levels of GAP, but no detectable PI3K activity, while the F740/F751/F771 triple mutant could not bind either GAP or PI3K. The in vitro and in vivo associations of GAP and PI3K activity to these PDGFR mutants were indistinguishable. The distinct tyrosine residue requirements suggest that GAP and PI3K bind different regions of the PDGFR. This possibility was also supported by the observation that the antibody to the PDGFR kinase insert Y751 region that blocks association of PI3K had only a minor effect on the in vitro binding of GAP. In addition, highly purified PI3K and GAP associated in the absence of other cellular proteins and neither cooperated nor competed with each other's binding to the PDGFR. Taken together, these studies indicate that GAP and PI3K bind directly to the PDGFR and have discrete binding sites that include portions of the kinase insert domain.

Purification and characterization of human erythrocyte phosphatidylinositol 4-kinase. Phosphatidylinositol 4-kinase and phosphatidylinositol 3-monophosphate 4-kinase are distinct enzymes

PtdIns 4-kinase has been purified 83,000-fold from human erythrocyte membranes. The major protein detected by SDS/PAGE is of molecular mass 56 kDa, and enzymic activity can be renatured from this band of the gel. The characteristics of this enzyme are similar to other type II PtdIns kinases previously described: PtdIns presented in Triton X-100 micelles is preferred as a substrate over PtdIns vesicles, the enzyme possesses a relatively low Km for ATP (20 microM), and adenosine is an effective inhibitor. A monoclonal antibody raised against bovine brain type II PtdIns 4-kinase is an effective inhibitor of the purified enzyme. PtdIns(4,5)P2 inhibits by approx. 50% when added in equimolar amounts with PtdIns; PtdIns4P has little effect on activity. A PtdIns3P 4-kinase activity has also been detected in erythrocyte lysates. Approximately two-thirds of this activity is in the cytosolic fraction and one-third in the membrane fraction. No PtdIns3P 4-kinase activity could be detected in the purified type II PtdIns 4-kinase preparation, nor could this activity be detected in a bovine brain type III PtdIns 4-kinase preparation. The monoclonal antibody that inhibits the type II PtdIns 4-kinase does not affect the PtdIns3P 4-kinase activity in the membrane fraction. The cytosolic PtdIns3P 4-kinase can be efficiently recovered from a 60%-satd.-(NH4)2SO4 precipitate that is virtually free of PtdIns 4-kinase activity. We conclude that PtdIns3P 4-kinase is a new enzyme distinct from previously characterized PtdIns 4-kinases, and that this enzyme prefers PtdIns3P over PtdIns as a substrate.

Interaction of phosphatidylinositol 3-kinase-associated p85 with epidermal growth factor and platelet-derived growth factor receptors

One of the immediate cellular responses to stimulation by various growth factors is the activation of a phosphatidylinositol (PI) 3-kinase. We recently cloned the 85-kDa subunit of PI 3-kinase (p85) from a lambda gt11 expression library, using the tyrosine-phosphorylated carboxy terminus of the epidermal growth factor (EGF) receptor as a probe (E. Y. Skolnik, B. Margolis, M. Mohammadi, E. Lowenstein, R. Fischer, A. Drepps, A. Ullrich, and J. Schlessinger, Cell 65:83-90, 1991). In this study, we have examined the association of p85 with EGF and platelet-derived growth factor (PDGF) receptors and the tyrosine phosphorylation of p85 in 3T3 (HER14) cells in response to EGF and PDGF treatment. Treatment of cells with EGF or PDGF markedly increased the amount of p85 associated with EGF and PDGF receptors. Binding assays with glutathione S-transferase (GST) fusion proteins demonstrated that either Src homology region 2 (SH2) domain of p85 is sufficient for binding to EGF and PDGF receptors and that receptor tyrosine autophosphorylation is required for binding. Binding of a GST fusion protein expressing the N-terminal SH2 domain of p85 (GST-N-SH2) to EGF and PDGF receptors was half-maximally inhibited by 2 and 24 mM phosphotyrosine (P-Tyr), respectively, suggesting that the N-SH2 domain interacts more stably with PDGF receptors than with EGF receptors. The amount of receptor-p85 complex detected in HER14 cells treated with EGF or PDGF. Growth factor treatment also increased the amount of p85 found in anti-PDGF-treated HER14 cells, suggesting that the vast majority of p85 in the anti-P-Tyr fraction is receptor associated but not phosphorylated on tyrosine residues. Only upon transient overexpression of p85 and PDGF receptor did p85 become tyrosine phosphorylated. These are consistent with the hypothesis that p85 functions as an adaptor molecule that targets PI 3-kinase to activated growth factor receptors.

Interaction of phosphatidylinositol 3-kinase-associated p85 with epidermal growth factor and platelet-derived growth factor receptors

One of the immediate cellular responses to stimulation by various growth factors is the activation of a phosphatidylinositol (PI) 3-kinase. We recently cloned the 85-kDa subunit of PI 3-kinase (p85) from a lambda gt11 expression library, using the tyrosine-phosphorylated carboxy terminus of the epidermal growth factor (EGF) receptor as a probe (E. Y. Skolnik, B. Margolis, M. Mohammadi, E. Lowenstein, R. Fischer, A. Drepps, A. Ullrich, and J. Schlessinger, Cell 65:83-90, 1991). In this study, we have examined the association of p85 with EGF and platelet-derived growth factor (PDGF) receptors and the tyrosine phosphorylation of p85 in 3T3 (HER14) cells in response to EGF and PDGF treatment. Treatment of cells with EGF or PDGF markedly increased the amount of p85 associated with EGF and PDGF receptors. Binding assays with glutathione S-transferase (GST) fusion proteins demonstrated that either Src homology region 2 (SH2) domain of p85 is sufficient for binding to EGF and PDGF receptors and that receptor tyrosine autophosphorylation is required for binding. Binding of a GST fusion protein expressing the N-terminal SH2 domain of p85 (GST-N-SH2) to EGF and PDGF receptors was half-maximally inhibited by 2 and 24 mM phosphotyrosine (P-Tyr), respectively, suggesting that the N-SH2 domain interacts more stably with PDGF receptors than with EGF receptors. The amount of receptor-p85 complex detected in HER14 cells treated with EGF or PDGF. Growth factor treatment also increased the amount of p85 found in anti-PDGF-treated HER14 cells, suggesting that the vast majority of p85 in the anti-P-Tyr fraction is receptor associated but not phosphorylated on tyrosine residues. Only upon transient overexpression of p85 and PDGF receptor did p85 become tyrosine phosphorylated. These are consistent with the hypothesis that p85 functions as an adaptor molecule that targets PI 3-kinase to activated growth factor receptors.

Transformation-defective mutants of polyomavirus middle T antigen associate with phosphatidylinositol 3-kinase (PI 3-kinase) but are unable to maintain wild-type levels of PI 3-kinase products in intact cells

Middle T antigen (MT) of polyomavirus causes transformation by associating with a number of cellular proteins. The association with and activation of two such proteins, phosphatidylinositol 3-kinase (PI 3-kinase) and pp60c-src, appears to be necessary for transformation by MT. The tyrosine kinase activity of MT-associated pp60c-src is significantly increased when assayed in vitro, and levels of phosphotyrosine-containing proteins are elevated in vivo. Similarly, levels of the PI 3-kinase products phosphatidylinositol-3,4-bisphosphate [PI(3,4)P2] and phosphatiylinositol-3,4,5-trisphosphate [PI(3,4,5)P3] are constitutively elevated in MT-transformed cells. However, the formation of a complete MT/cellular protein complex and the activation of tyrosine kinase are not sufficient to cause transformation, since the transformation-defective mutants 248m and dl1015 associate with all wild-type MT-associated proteins, including PI 3-kinase and pp60c-src, and neither mutant appears to be defective in MT-associated tyrosine kinase activity. Studies presented here compared (i) the amount of PI 3-kinase activity associated with the MT complex and (ii) levels of [3H]inositol incorporation into PI 3-kinase products in cells expressing mutant or wild-type MT. The results show that dl1015 is defective in both assays, whereas 248m is defective only for incorporation of [3H]inositol into PI(3,4,5)P2 and PI(3,4)P3. These findings identify a biochemical defect in the 248m mutant and corroborate previous results correlating transformation and elevated levels of PI 3-kinase products in vivo. In addition, they indicate that PI 3-kinase product levels are affected by factors other than simply the amount of PI 3-kinase activity associated with the MT complex.

Platelet membrane phosphatidylinositol kinase activity. Triton X-100 effects provide evidence for intramicellar reaction

Phosphatidylinositol (PI) kinase activity of platelet membranes was solubilized and partially purified by anion-exchange chromatography to measure the initial enzymatic rates. Kinetic studies were performed in the presence of Triton X-100 to obtain mixed micelles. The partially purified enzyme exhibited a Michaelian behaviour towards ATP, with a Km of 58 microM. The enzymatic rates were dependent upon Triton concentrations. Upon increasing its concentration, this detergent exhibited an activating effect followed by an inhibitory one. The optimal micellar Triton concentration was proportionnal to the PI concentration used in the assay. Conversely, the behaviour of the enzyme towards PI was dependent upon the Triton concentration. However, when PI concentration was expressed as its surfacic concentration within the micelles, the activity became independent of the detergent concentration, and a Km value of 0.09 mol/mol was estimated. Therefore, in vitro phosphorylation of phosphatidylinositol by PI kinase is rate-limited by an intramicellar reaction, and provides a study model for the in vivo reaction.

Phosphoinositides and cell growth

The findings described above illustrate how the src kinase can influence several new pathways of inositol phosphate metabolism, both at the membrane level with the production of novel D-3 phosphoinositides and the activation of PI-3 kinase, and at the cytosolic level by altering the expression of certain inositol polyphosphates, in particular Ins(1,4,5,6)P4. At present, it is difficult to speculate on the role these phenomena play in cellular transformation by src, since the functions of D-3 phosphoinositides and most inositol polyphosphates are unclear. There is evidence, however, that these new pathways of phosphoinositide metabolism occur in response to other types of cellular stimulations besides src transformation. Novel D-3 phosphoinositides are expressed in a variety of nonneoplastic cells, including human platelets treated with thrombin, smooth muscle cells and stimulated neutrophils. In addition, unusual InsP4 isomers such as D/L-Ins(1,4,5,6)P4 are found in chicken erythrocytes, murine macrophages, AR4-2J rat pancreatoma cells and adrenal glomerulosa cells, to name only a few. Recently, associations have been reported between PI-3 kinases and cytoskeletal elements in thrombin- stimulated platelets, and between activated ras proteins in rat liver epithelial cells. The latter discovery is particularly intriguing since GTP-binding proteins such as ras are known to influence cell shape and serve as downstream effector proteins in the signal transduction pathways of numerous growth factor receptors. Thus, one function of novel phosphoinositides and their metabolites may lie at the level of cytoskeletal and cell shape regulation. Clearly, additional roles for phosphoinositides exist in cells besides their traditional use as precursors for the generation of Ins(1,4,5)P3 and diacylglycerol.(ABSTRACT TRUNCATED AT 250 WORDS)