Regulation of Phosphoinositide Kinases in T Cells EVIDENCE THAT PHOSPHATIDYLINOSITOL 3-KINASE IS NOT A SUBSTRATE FOR T CELL ANTIGEN RECEPTOR-REGULATED TYROSINE KINASES*

A phosphoinositide kinase that can phosphorylate phosphatidylinositol (PtdIns) is present in 4G10 monoclonal antibody (mAb) phosphotyrosine immunoprecip- itates isolated from T cells activated via the T cell antigen receptor (TCR) . CD3 complex. This PtdIns ki- nase is not the PtdIns 3-kinase that associates with activated protein tyrosine kinases in fibroblasts, since Western blotting and immunoprecipitation experiments with antibodies specific for the p85a subunit of the PtdIns 3-kinase indicate that this polypeptide is not immunoprecipitated by the 4G10 mAb from TCR. CD3-activated Jurkat cells. Moreover, immunoprecipitated PtdIns 3-kinase isolated from T cells with p85 antibodies is inhibited when PtdIns is presented in Nonidet P-40, whereas the PtdIns kinase activity present in 4G10 mAb phosphotyrosine immunoprecipitates is enhanced in the presence of Nonidet P-40. I n vitro kinase assays of PtdIns 3-kinase immunoprecipitated with p85 antibodies from T cells indicate that it associates with a serine kinase that can phosphorylate a p85 polypeptide. However, no protein tyrosine kinase activity capable of tyrosine phosphorylating p85 in vitro associates with p85a immunoprecipitates in quiescent or TCR.CD3-activated T cells. These data suggest that the TCR.CD3 complex does not regulate PtdIns 3-kinase activity by a mechanism that involves protein tyrosine kinases.

Tyrosine phosphorylation and its regulation are essential for TCR. CD3-mediated T cell activation (14,17). Furthermore, PTK activation precedes phosphoinositide breakdown in T cells (12) and is an obligate event for TCR. CD3-mediated PLC activation. The mechanism that the TCR-CD3 uses to couple to PLC probably involves TCR. CD3-directed tyrosine phosphorylation of the PLCyl isozyme, resulting in increased enzyme activity and hence elevated phosphoinositide metabolism (18)(19)(20). TCR. CD3-mediated PtdIns (4,5)P2 hydrolysis is a sustained response in T cells which requires a substantial increase in the net synthesis of PtdIns (4,5)P2 (21,22). Analysis of cellular PtdIns (4,5)P2 levels during T cell activation support that such an increase occurs (21,22) and indicate that it is regulated by the actions of ill-defined phosphoinositide lipid kinases, namely phosphatidylinositol (PtdIns) 4-kinase and phosphatidylinositol-(4)-monophosphate (PtdIns(4)P) 5-kinase (10). Little is known about the mechanisms that regulate the activity of the PtdIns 4-kinase, although there are reports that in fibroblasts this kinase is a substrate for PTKs and may therefore be regulated by tyrosine phosphorylation (23).
Recently, it has been shown that T cells express the lipid products of a novel PtdIns 3-kinase (24): PtdIns(3)P, PtdIns (3,4)P2, and PtdIns (3,4,5)P3 (21). The function of these D-3 (po1y)phosphoinositides has not been determined, but they are not substrates for the known cellular PLCs and as such are not thought to be hydrolyzed during cellular activation (25,26). During T cell activation via the TCR.CD3 complex, levels of PtdIns(3)P do not change, but there is an increase in cellular levels of PtdIns (3,4)P2 and PtdIns (3,4,5)P3 (21). It has also been described that stimulation of the IL-2R, which results in T cell growth, also induces increases in levels of D-3 (po1y)phosphoinositides (27). Hence, these lipid prod-ucts may be involved in the signaling mechanisms that regulate both T cell activation and growth.
The PtdIns 3-kinase comprises two major subunits of 85 and 110 kDa (24). The p85 polypeptide may function as the mediator for the interactions between the catalytic p l l 0 subunit of PtdIns 3-kinase and activated receptor and nonreceptor PTKs (28)(29)(30). The p85 subunit contains Src homology (SH) regions (one SH3 and two SH2 regions) (28,29) and associates with and is a substrate for many tyrosine kinases including p60""" (30-32), the platelet-derived growth factor (PDGF) receptor (28-32), the colony-stimulating factor-1 (CSF-1) receptor (33,34), the epidermal growth factor (EGF) receptor (35), and the insulin receptor (36). In T lymphocytes, recent studies have suggested that the PtdIns 3-kinase is a substrate for the S r c family kinase ~5 9 '~" (37) and may also be tyrosine-phosphorylated by as yet unidentified IL- 2Rcontrolled PTKs (37,38). However, although these data implicate PTKs in the regulation of the PtdIns 3-kinase that occurs during IL-2R-directed T cell growth, it is unknown whether it is a substrate for the TCR. CD3-controlled PTKs that regulate T cell activation.
It has been described that tyrosine phosphorylation of PtdIns 3-kinase correlates with its increased activity (30,32,36). Therefore, the object of the present study was to use the 4G10 monoclonal antibody (mAb) for phosphotyrosine and antibodies to the p85 subunit of the PtdIns 3-kinase to determine whether the subunits of the PtdIns 3-kinase are tyrosine-phosphorylated or associated with tyrosine kinases subsequent to triggering of the TCR-CD3. The results of this study demonstrate that immunoprecipitates of tyrosine-phosphorylated proteins isolated from TCR. CD3-stimulated T cells do contain a PtdIns kinase activity, but this is distinct from PtdIns 3-kinase. Moreover, PtdIns 3-kinase is not detectably phosphorylated on tyrosine in TCR. CD3-activated T cells nor is there any evidence that it associates with tyrosine kinases during T cell activation.

EXPERIMENTAL PROCEDURES
Reagents and Antibodies-The CD3 antibody UCHTl and (Fab'), fragments of UCHTl were generated as described (13). Polyclonal rabbit antisera to the p85 subunit of the PtdIns 3-kinase was generated as described (28). Full purification details and characterization of the p85 recombinant protein are described elsewhere.' The 4G10 phosphotyrosine antibody was kindly supplied by Dr. Brian Druker and Dr. Tom Roberts (Harvard Medical School). The following were obtained from the sources indicated. Potassium oxalate and phosphoinositides (soybean PtdIns, bovine PtdIns(4)P, and PtdIns(4,5)P2) were obtained from Sigma (Poole, Dorset, United Kingdom); [y-:"P]ATP (5000 Ci/mmol) and '"I-protein A (30 mCi/ mg) were purchased from Amersham International (United Kingdom). Silica gel 60 thin layer chromatography plates were from Whatman. Organic solvents were from BDH Chemicals (Poole, United Kingdom).
Ptdlns Kinase Assay-PtdIns kinase activity was determined by the modified method of Whitman et al. (39). Briefly, the washed immunoprecipitates were resuspended in 30 pl of kinase buffer. A lipid mixture (50 pl of 0.1 mg/ml PtdIns and 0.1 mg/ml phosphatidylserine, dispersed by sonication in 25 mM HEPES buffer, pH 7.4, and 1 mM EDTA) was added to the immunoprecipitates. The reaction was initiated by the addition of 20 pCi of [y-32P]ATP and 100 p M ATP and terminated after 15 min by the addition of 80 p1 HCl (1 M) and 200 p1 of ch1oroform:methanol (1:l). After vigorous mixing and centrifugation to separate the phases, the organic layer was removed, dried under NP, and resuspended in chloroform. The extracted phospholipids were then separated by TLC in 1-propano1:acetic acid (2 N) (65:35 (v/v)) developing solvents (40) and visualized by exposure to iodine vapor and autoradiography. Phospholipids were identified by comparison with nonlabeled standards.
immunoblotting with 4GIO mAb and p85 Polyclonal Antisera-Samples for immunoblotting were electrophoresed on 7-17% SDSpolyacrylamide gel and transferred by electroblotting onto PVDF membranes (Immobilon P, Millipore, Bedford, MA) as described previously (13). The transfer buffer used was 192 mM glycine, 25 mM Tris, and blotting was carried out for 20-24 h a t 0.2 A and 50 V. The blots were probed with either iodinated 4G10 for 2 h (13) or p85 polyclonal antisera overnight (28). Blots probed with the p85 antibodies were washed three times in blocking buffer, followed by incubation for 1 h a t room temperature with '"I-protein A (28). Blots were dried with Whatman 3MM paper and bands visualized by autoradiography of the filter a t -70 "C on XAR-5 film (Kodak).
I n Vitro Protein Kinase Assays-Immunoprecipitates were prepared as above, and the immunoprecipitates were washed three times in lysis buffer and twice in protein kinase assay buffer (100 mM NaC1, 25 mM HEPES, pH 7.4, 10 mM MgCl,, 5 mM MnC12, 100 pM sodium orthovanadate) as described (31). In vitro kinase assays were initiated with 20 p1 of kinase buffer containing 10 p~ ATP and 10 pCi of [y-"'PIATP. After 10 min at 37 "C, the reaction was stopped by the addition of 1 ml of lysis buffer containing 20 mM EDTA. The immunoprecipitates were then washed three times in this buffer, and the immunoprecipitated proteins were solubilized in SDS sample buffer prior to separation by SDS-PAGE. "P-Labeled proteins were visualized by autoradiography a t -70 "C. For immunoprecipitation of proteins labeled in the in vitro kinase assay, reactions were stopped with 0.1 ml of assay buffer containing 1% SDS and boiled. Samples were then diluted 10-fold in assay buffer followed by the addition of 4G10 mAb or p85a polyclonal antisera. After overnight rotation a t 4 "C, beads were washed as above, followed by boiling in 200 pl of 1% reducing SDS sample buffer and separated by SDS-PAGE.
Phosphoamino Acid Analysis-After SDS-PAGE, labeled protein was electroblotted onto a PVDF filter as described (13), and the blot was then autoradiographed to determine the position of the labeled bands. Portions of the filter which contained labeled protein were excised, soaked in methanol, and then washed in distilled water. Bound protein was hydrolyzed in 6 M HC1 (150-250 pl) for 1 h at 110 "C. The hydrolysate was dried in a SpeediVac (Savant, Farmingdale, NY) and redissolved in 5 p1 of 0.1 M acetic acid containing 3 pg each of unlabeled phosphothreonine, phosphoserine, and phosphotyrosine as standards. The phosphoamino acids were separated a t 20 mA for 80 min on a cellulose TLC (Polygram Cell 300, Macherey-Nagel, Duren, Federal Republic of Germany) in acetic acid:pyridine:water (10:1:189, v:v:v). The labeled phosphoamino acids were visualized by autoradiography a t -70 "C.

RESULTS
The 4GlO Phosphotyrosine mAb Immunoprecipitates a PtdIns Kinase from Activated T Cells-Activation of the TCR. CD3 complex with the CD3 mAb UCHTl (41) (-) and unstimulated (lanes I , 3,5,7,9). 4G10 mAb immunoprecipitates were prepared from Nonidet P-40 lysates, washed, and divided equally into two aliquots. a, one aliquot of immunoprecipitated phosphotyrosyl polypeptides was resolved by SDS-PAGE under reducing conditions, transferred to a PVDF membrane, detected by probing with iodinated 4G10 mAb, and visualized by autoradiography. The migration of molecular mass calibration standards is indicated to the left of the autoradiograph in kilodaltons. b, the second aliquot of immunoprecipitated phosphotyrosyl polypeptides was assayed for associated PtdIns kinase activity using PtdIns as a substrate. Extraction and separation of the PtdIns kinase products was performed as described under "Experimental Procedures." The standard PtdIns(4)P and PtdIns(4,5)P2 were co-chromatographed with the samples and visualized by exposure to iodine vapor and autoradiography. Data are from a single experiment, representative of five others. Cell equivalents per lane are 2 X lo7 cells. stimulated and control cells and confirm previous reports (13) that the induction of protein tyrosine phosphorylation after TCR CD3 stimulation was maximal within 30 s of activating the cells and had returned to control levels by 30 min. To determine if PtdIns kinase(s) were tyrosine phosphorylated or associated with phosphotyrosyl polypeptides, a parallel set of 4G10 immunoprecipitates were assayed for PtdIns kinase activity. The data (Fig. lb) show that a PtdIns kinase activity was present in 4G10 immunoprecipitates prepared from Jurkat cells after 1-10-min exposure to the TCR-CD3 agonist UCHTl (10 pg/ml). The association of PtdIns kinase with 4G10 mAb was transient, since no PtdIns kinase was detectable in 4G10 mAb immunoprecipitates at 30 min post-TCR-CD3 ligation. 4G10 mAb immunoprecipitation of tyrosine phosphorylated polypeptides (Fig. 2a) was prevented by the addition of 50 mM phenyl phosphate (Fig. 24 lanes 3 and 4 ) and 10 mM phosphotyrosine (Fig. 2u, lanes 5 and 6) during immunoprecipitation (Fig. 2a). Similarly, the association of PtdIns kinase activity with 4G10 mAb immunoprecipitates prepared from activated Jurkat cells was prevented by 50 mM immunoprecipitates were washed and divided equally into two aliquots. a, one aliquot of immunoprecipitated phosphotyrosyl polypeptides was resolved by SDS-PAGE under reducing conditions, transferred to a PVDF membrane, detected by probing with iodinated 4G10 mAb, and visualized by autoradiography. The migration of molecular mass calibration standards is indicated to the left of the autoradiograph in kilodaltons. b, the second aliquot of immunoprecipitated phosphotyrosyl polypeptides was assayed for associated PtdIns kinase activity using PtdIns as a substrate. Extraction and separation of the PtdIns kinase products was performed as described under "Experimental Procedures." The standard PtdInsP and Ptd-InsP2 were co-chromatographed with the samples and visualized by exposure to iodine vapor and autoradiography. Data are from a single experiment, representative of five others. Cell equivalents per lane are 2 X lo7 cells. P-Tyr, phosphotyrosine. phenyl phosphate (Fig. 2b, lanes 3 and 4 ) and 10 mM phosphotyrosine (Fig. 2b, lanes 5 and 6). Therefore, the association of PtdIns kinase activity with the 4G10 mAb appeared to be specific and dependent on the recognition of phosphotyrosine.

Tyrosine-phosphorylated or Associated with PTKs in T Cells-
In initial efforts to characterize the PtdIns kinase activity associated with the 4G10 mAb, we examined whether 4G10 immunoprecipitates isolated from TCR-CD3-activated Jurkat cells contained the p85 subunit of the PtdIns 3-kinase. At least two forms of p85 have been isolated, designated p85a and p85j3 (28). Initial Western blotting and immunoprecipitation studies with specific antisera to p85a and p85p established that Jurkat cells expressed p85a but not p85B (data not shown). In Western blotting studies with the polyclonal antisera to the p85a subunit, a single p85 polypeptide could be detected in IO6 cell equivalents of total Nonidet P-40 cell lysates prepared from both TCR. CD3-activated and nonactivated Jurkat cells (Fig. 3u, lanes 2 and 3). These cell lysates also contained phosphotyrosyl polypeptides (Fig. 3b, lanes 2  and 3 ) . In contrast, 4G10 immunoprecipitates from TCR-CD3-activated Jurkat cell lysates, which contained phospho-  (lanes 1 and 3 ) . Control cells were treated with vehicle (lane 2). Cells were lysed as described under "Experimental Procedures." Phosphotyrosyl polypeptides in the total cell lysate (lanes 2 and 3 ) or in the 4G10 mAb immunoprecipitates from UCHT1-stimulated cells ( l a n e 1 ) were divided in two and resolved by SDS-PAGE under reducing conditions, transferred to a PVDF membrane, and ultimately detected by probing with p85a polyclonal antisera followed by "'1-protein A ( a ) or iodinated 4G10 mAb ( b ) and visualized by autoradiography. The migration of molecular mass calibration standards in indicated to the left of each autoradiograph kilodaltons. Each autoradiograph is from a single experiment, representative of 3 others. Cell equivalents per lane are 2 X 10' (Lane 1) and 10' (Lanes 2 and 3).
tyrosyl proteins (Fig. 3b, lane I ) , failed to Western blot with the p85a antisera, even though 2 X lo7 cell equivalents of a 4G10 mAb immunoprecipitate were loaded onto these gels (Fig. 3a, lane 1 ). The p85a antisera can immunoprecipitate PtdIns 3-kinase activity from both activated and nonactivated Jurkat cells (Fig. 4a). However, p85a immunoprecipitates, when resolved by SDS-PAGE and transferred onto a PVDF filter, contained no detectable polypeptides reactive with the iodinated 4G10 mAb as determined by Western blotting analysis (Fig. 4b, lanes 3 and 4 ) , whereas phosphotyrosyl polypeptides were present in parallel 4G10 mAb immunoprecipitates analyzed under identical Western blotting conditions (Fig. 4b, lanes 1 and 2).
4G10 immunoprecipitates derived from TCR. CD3-activated (10 pg/ml UCHTl or 10 pg/ml UCHTl (Fab')* fragments) Jurkat cells have high levels of associated in vitro PTK activity (Fig. 5a) such that a number of the co-immunoprecipitated proteins were phosphorylated on tyrosine in vitro (Fig. 5b). The in vitro labeled substrates from the 4G10 mAb immunoprecipitates prepared from TCR -CD3-activated and nonactivated Jurkat cells were reprecipitated with the 4G10 mAb, confirming that they were phosphorylated on tyrosine residues (Fig. 6c, lanes 4-6). However, none of these proteins could be reprecipitated with the p85a polyclonal antiserum (Fig. 6b, lanes 4-6). As controls in this series of assays, parallel in vitro kinase protocols were carried out on p85a immunoprecipitates isolated from Jurkat cells and subjected to reprecipitation with either p85a or 4G10 antibodies. The data (Fig. 6 4 lanes 1-3) show that p85a immunoprecipitates did contain associated in vitro protein kinase activity, which resulted in the phosphorylation of a p85 polypeptide. The in vitro kinase activity associated with p85a immunoprecipitates was present in TCRsCD3-activated and nonactivated Jurkat cells and could reflect the association of p85a with a PTK. However, the in vitro 32P-labeled p85 polypeptide was reprecipitated with the p85a antisera (Fig. 6b, lanes 1-3) - Jurkat cells were stimulated with 10 pg/ml UCHTl for 2 min. Control cells were treated with vehicle and unstimulated. Jurkat cells were lysed with Brij 96 as described under "Experimental Procedures," and the Brij 96 lysates were immunoprecipitated with protein A-Affi-Gelcoupled polyclonal antisera to the p85a subunit of PtdIns 3 kinase. These immunoprecipitates were then assayed for associated PtdIns kinase activity using PtdIns as a substrate. Extraction and separation of the products was performed as described under "Experimental Procedures." The standard PtdInsP and PtdInsP2 were co-chromatographed with the samples and visualized by exposure to iodine vapor and autoradiography. The TLC is from a single experiment, representative of four others. b, 4G10 mAb Western blot analysis of p85a immunoprecipitates. Jurkat cells were stimulated with 10 pg/ml UCHTl (lanes 2 and 4 ) for 2 min. Control cells (lanes I and 3) were treated with vehicle and unstimulated. Jurkat cells were lysed as described under "Experimental Procedures," and the Brij 96 lysates were immunoprecipitated as described under "Experimental Procedures" with polyclonal antisera to p85a (lanes 3 and 4 ) or the 4G10 mAb (lanes 1 and 2). The immunoprecipitates were resolved by SDS-PAGE under reducing conditions, transferred to a PVDF membrane, detected by probing with iodinated 4G10 mAb, and visualized by autoradiography. The migration of molecular mass calibration standards is indicated to the left of each autoradiograph in kilodaltons. Each autoradiograph is from a single experiment, representative of three others. Cell equivalents per lane are 2 X 10' cells.
but not the 4G10 mAb (Fig. 6c, lanes 1-3). Moreover, phosphoamino acid analysis of in vitro 32P-labeled p85 that was immunoprecpitated by p85a antisera from TCR. CD3-activated and nonactivated Jurkat cells established that it was phosphorylated exclusively on serine residues (data not shown). In addition, p85a immunoprecipitates derived from either TCR. CD3-activated or nonactivated Jurkat cells did not contain a kinase activity that could phosphorylate the tyrosine kinase substrate Raytide" (42), although a serine kinase activity that resulted in the phosphorylation of a serine peptide substrate (Arg-Phe-Ala-Arg-Lys-Gly-Ser-Leu-Arg-Gln-Lys-Asn-Val) (43, 44) was observed (data not shown). These reciprocal Western blotting (Fig. 4b) and reprecipitation experiments (Fig. 6) with p85a and 4G10 antibodies indicate that p85a associated with a serine kinase but was not a substrate for PTKs and was not detectably associated with tyrosine kinases in Jurkat cells. The experiments in Figs. 4-6 were carried out following extraction in Brij 96 detergent. Brij 96 is a mild detergent which preserves loosely associated oligomeric peptides?
Characterization of the PtdIns Kinase Activity Associated with the 4GlO rmlb-Although 4G10 mAb immunoprecipitates did not apparently contain the p85 subunit of the PtdIns 3-kinase, they did contain PtdIns kinase(s) activity. This could reflect the presence of the p l l 0 PtdIns 3-kinase catalytic subunit. The most direct way to address this issue is to S. Ley and N. Osman, unpublished observations. Jurkat cells were stimulated with 10 pg/ml UCHTl mAb for 2 min. Control cells were treated with vehicle and unstimulated. Jurkat cells were lysed with Brij 96, and the lysates were immunoprecipitated with the 4G10 mAb. In vitro kinase reactions were performed on these immunoprecipitates as described under "Experimental Procedures." The products were then separated by SDS-PAGE (a). Polypeptides were visualized by autoradiography. The migration of molecular mass calibration standards is indicated to the left of each autoradiograph in kilodaltons. Each autoradiograph is from a single experiment, representative of two others. b, phosphoamino acid analysis of in vitro :'2P-labeled polypeptides associated with the 4G10 mAb were separated by SDS-PAGE and then transferred to a PVDF filter. After autoradiography, the indicated phosphorylated bands were excised and bound protein hydrolyzed in HCI. Phosphoamino acids were resolved by TLC and compared with unlabeled standards. P-Tyr, phosphotyrosine; P-Ser, phosphoserine; P-Thr, phosphothreonine.
analyze by anion exchange on HPLC the lipid products of the PtdIns kinase activity present in 4G10 mAb immunoprecipitates. However, it proved impossible to label these products sufficiently in uitro for such studies. Lipid products from PtdIns kinase assays of p85a immunoprecipitates could readily be analyzed by HPLC techniques, and such studies confirmed that the p85a antisera immunoprecipitated PtdIns 3kinase activity (data not shown). However, the 4G10 mAb immunoprecipitate had less than 0.1% of the PtdIns kinase activity of the p85a immunoprecipitate (Fig. 7), which made it impossible to obtain sufficient labeled lipid products for quantitative HPLC analysis.
In an alternative approach, we examined the detergent and adenosine sensitivity of the PtdIns kinase activity present in the 4G10 mAb immunoprecipitates. Previous work has shown that PtdsIns 3-kinase is inhibited when PtdIns is presented in detergent but is resistant to adenosine (23,24). In contrast, PtdIns 4-kinase can utilize PtdIns presented in detergent and is inhibited by adenosine (23, 45). Fig. 8a, lane 1, shows that the PtdIns activity associated with 4G10 mAb immunoprecipitates from TCR.CD3-activated Jurkat cells was enhanced by 0.5% Nonidet P-40 (Fig. 8a, lane 3) and was inhibited (approximately 30%) by 300 pM adenosine (Fig. 8a, lane 2).
In marked contrast, the PtdIns 3-kinase activity immunoprecipitated by the p85a antibody from TCR.CD3-activated Jurkat cells (Fig. 8b, lane I ) was completely inhibited by Nonidet P-40 (Fig. 8b, lane 3) and unaffected by 300 p~ adenosine (Fig. 8b, lane 2 ) . Previous studies have established that PtdIns 3-kinase can be immunoprecipitated with phosphotyrosine antibodies from PDGF-activated fibroblasts (28)(29)(30)(31)(32). The data in Fig. &, lane 2, confirm increased levels of PtdIns kinase activity in 4G10 mAb immunoprecipitates isolated from PDGF-stimulated Swiss 3T3 cells compared with unstimulated quiescent cells (Fig. &, lane 1 ). The PtdIns kinase activity in 4G10 mAb immunoprecipitates derived from PDGF-stimulated fibroblast (Fig. &, lane 2) was inhibited by the detergent Nonidet P-40 (Fig. &, lane 4 ) . DISCUSSION The present study demonstrates that activation of the Jurkat leukemic T cell line via the TCR -CD3 induces tyrosine phosphorylation of a large number of cellular substrates and results in a rapid increase in the level of PtdIns kinase activity associated with phosphotyrosine antibodies. The increased levels of PtdIns kinase activity in 4G10 immunoprecipitates isolated from TCR.CD3 activated cells reflect either an increase in tyrosine phosphorylation of the enzyme or an associated protein. The present data also indicate that Jurkat cells express PtdIns 3-kinase which has been identified in previous studies to be associated with activated receptor and non-receptor PTKs in fibroblasts and hemopoietic cells (28)(29)(30)(31)(32)(33)(34)(35)(36).
Recent studies have shown that in T lymphocytes, triggering of the IL-2R results in an increase in the association of PtdIns 3-kinase with phosphotyrosine antibodies (27). However, several lines of evidence in this study indicate that the PtdIns kinase activity immunoprecipitated by the 4G10 mAb antibody from TCR. CD3-activated Jurkat cells is not PtdIns 3-kinase. First, in fibroblasts, it is the p85 subunit of the PtdIns 3-kinase that associates with PTKs (33, 35, 46), but specific immunoprecipitation and Western blotting studies failed to identify p85 in the 4G10 mAb immunoprecipitates derived from TCR. CD3-activated Jurkat cells. Second, a comparison of the detergent and adenosine sensitivity of the 4G10 phosphotyrosine mAb-associated PtdIns kinase activity and authentic PtdIns 3-kinase immunoprecipitated with p85a antibodies, isolated from TCR. CD3-activated Jurkat cells, revealed different biochemical properties of these lipid kinases. The PtdIns kinase activity immunoprecipitated with p85a antibodies from Jurkat cells or with 4G10 antibodies from PDGF-activated fibroblasts was sensitive to inhibition by the detergent Nonidet-P-40. Moreover, the p85a antibodyassociated PtdIns kinase activity derived from Jurkat cells was resistant to adenosine inhibition. In contrast, the PtdIns kinase activity associated with 4G10 mAb immunoprecipitates from TCR.CD3-activated Jurkat cells was enhanced by Nonidet P-40 and sensitive to inhibition by adenosine. These data indicate a PtdIns kinase activity that is tyrosine-phosphorylated or associated with phosphotyrosyl proteins in TCR. CD3-activated Jurkat cells and which is not identifiable as PtdIns 3-kinase. The 4G10 mAb-associated PtdIns kinase activity may be attributable to PtdIns 4-kinase or a PtdIns 5kinase (23).
A PtdIns 4-kinase has been shown to associate with and be a substrate for ligand activated EGF receptors (23), although the functional significance of this phenomenon is unclear. Nevertheless, in T cells it has been demonstrated that T cell activation is associated with an increase in the activity of . This mechanism functions to replenish the PLC substrate PtdIns (4,5)Pz, during a sustained period of TCR.CD3-induced PLC activation (21,22). It is well established that tyrosine phosphorylation is essential for TCR. CD3 induction of PtdIns (4,5)P2 hydrolysis (14-17). This requirement is attributed to the role of PTKs in coupling the TCR.CD3 to PLC, since tyrosine phosphorylation of PLCrl is known to be induced by the TCR.CD3 and is thought to be essential for TCR.CD3 stimulation of this enzyme (18)(19)(20). The present observations of the tyrosine phosphorylation of a PtdIns kinase or an associated protein raises the possibility that PTKs may also be involved in TCR CD3 regulation of PtdIns metabolism at the level of the replenishment of PtdIns (4,5)Pz during prolonged periods of TCR. CD3-induced PLC activation.
In fibroblasts, an increase in the activity of cellular PtdIns  p85a polyclonal antisera (b). The washed immunoprecipitates were analyzed for PtdIns kinase activity using PtdIns as a substrate. Extraction and separation of the products was performed as described under "Experimental Procedures." The standard PtdInsP and Ptd-InsP2 were co-chromatographed with the samples and visualized by exposure to iodine and autoradiography. The TLC is from a single experiment, representative of four others.
3-kinase and an increase in cellular levels of D-3 phosphoinositides (47) correlates with tyrosine phosphorylation of the p85 subunit of the [30][31]. Similarly, IL-2R regulation of PtdIns 3-kinase activity correlates with tyrosine phosphorylation of the enzyme (27), although the requirement of tyrosine phosphorylated p85 subunit for PtdIns 3-kinase activity remains to demonstrated. Ligation of the TCR.CD3 complex results in an increase in certain D-3 phosphoinositide lipids in intact cells (21), but data in this study indicate that p85a is not a detectable substrate for TCR.CD3-regulated PTKs. It would seem therefore that TCR-CD3 regulation of PtdIns 3-kinase is by an alternative mechanism that does not involve tyrosine phosphorylation of the p85a subunit. No increase in in vitro PtdIns 3-kinase activity immunoprecipitated by p85 antisera could be detected upon TCR CD3 activation of Jurkat cells. Similarly, PDGF treatment of fibroblasts has been reported not to increase the in vitro PtdIns 3-kinase activity of p85 immunoprecipitates (35). Only 5% of the total PtdIns 3-kinase activity associates with the ligand-activated PDGF receptor in NIH 3T3 cells: suggesting that only a small percentage of the total PtdIns 3kinase population is receptor-activated. Ligation of the TCR. CD3 complex may similarly activate a small percentage of the total cytosolic PtdIns 3-kinase, such that the remaining nonreceptor-activated PtdIns 3-kinase, which is also immunoprecipitated by p85a antisera, may mask a rise in a receptor-  . Extraction and separation of the products was performed as described "Experimental Procedures." The standard PtdIns was co-chromatographed with the samples and visualized by exposure to iodine and autoradiography. The TLC is from a single experiment, representative of two others. in vitro to label the substrates and allow their detection by reprecipitation with the 4G10 mAb. Under these conditions, i t was still not possible to detect tyrosine-phosphorylated polypeptides in the p85a immunoprecipitates by reprecipitation with the 4G10 mAb. The p85 subunit of PtdIns 3-kinase is not only a substrate for PTKs in fibroblasts (28,29) but is also known to associate with activated PTKs such as c-src (31-32) and v-src (48), as well as ligand activated PDGF, EGF, and CSF-1 receptors (28-33). These associations between p85 and PTKs are possibly mediated via SH2 domains (28,29,33,49,50,51). The associations between p85 and PTKs may be important to the regulatory function of p85 by targeting the catalytic subunit of PtdIns to activated tyrosine kinases (35). Hence, the p85 subunit may associate with, but not act as a substrate for, TCR. CD3-regulated PTKs. Indeed, PtdIns 3-kinase is known to associate with an IL-2R-associated PTK (27) CD4-~56'~ complex (52) and with p59fy" in IL-2-dependent T cell clones (37). In addition to evidence that suggests that the p85 subunit of PtdIns 3-kinase does not act as a substrate for PTKs, there is also evidence that p85 does not associate with PTKs in TCR.CD3-activated Jurkat cells. First, immune complex kinase assays performed on p85a immunoprecipitates from activated T cells revealed that a serine kinase, but no PTK activity, co-precipitated with p85a under low stringency cell lysis conditions used previously to maintain PTK-TCR. CD3 associations (53). Second, the 4G10 antibody is known to immunoprecipitate PTKs in T cells: including p59fy" and p56Ick, but it did not ' N. Osman and S. Ley, unpublished observations. co-precipitate PtdIns 3-kinase in our studies. This could result from the disruption of the binding of PtdIns 3-kinase to PTKs by the 4G10 mAb. However, this phosphotyrosine antibody has been used effectively to co-precipitate PTKs and PtdIns 3-kinase in IL-PR-activated T cells (27). Similarly, immune complex kinase assays of p85a immunoprecipitates in Swiss 3T3 cells can readily demonstrate the existence of a p85aassociated PTKs,G which indicates that the p85a antibody used in these studies does not interfere with p85-PTK associations. Moreover, these antisera did allow p85 to co-precipitate with a serine kinase in Jurkat cells. The functional relevance or identity of this associated protein kinase is unknown, but a similar association of PtdIns 3-kinase with a serine/threonine kinase has been described in fibroblasts (54).
In conclusion, the present data suggest that the p85a subunit of PtdIns 3-kinase is not a detectable substrate for TCR. CD3-regulated PTKs and suggests that any regulation of this enzyme activity by the TCR -CD3 must involve an alternative mechanism. The data also indicate that the PtdIns 3-kinase does not detectably associate with PTKs during T cell activation by the TCR.CD3, in contrast to its reported association with PTKs during IL-2R-directed T cell growth (27) and CD4 activation (52). It is interesting to note that both the IL-2R (55) and CD4 (56) are associated with ~5 6 "~, whereas the TCR.CD3 associates with p 5 P (57). Hence, the regulation of PtdIns 3-kinase may be entirely different, depending on the association of the relevent activated cell surface molecule with either ~5 6 "~ or p59fy". Thus, tyrosine phosphorylation of PtdIns 3-kinase may be important to signal transduction mechanisms associated with the IL-2R and CD4 ligation but not those associated with the TCR. CD3 complex. Finally, the TCR. CD3 complex does induce the phosphotyrosyl polypeptide association of a PtdIns kinase distinct from the PtdIns 3-kinase, and the identity of this lipid kinase and its functional relevance to phosphoinositide metabolism in T cells remain to be resolved.