The Journal of Immunology

Role of Two Adaptor Molecules SLP-76 and LAT in the PI3K Signaling Pathway in Activated T Cells

Eun Kyung Shim,*,1 Seung Hee Jung,*,1 and Jong Ran Lee*,†,‡

Previously, we identified p85, a subunit of PI3K, as one of the molecules that interacts with the N-terminal region of Src homology 2 domain-containing leukocyte protein of 76 kDa (SLP-76). We also demonstrated that tyrosine phosphorylation either at the 113 and/ or 128 position is sufficient for the association of SLP-76 with the Src homology 2 domain near the N terminus of p85. The present study further examines the role of the association of these two molecules on the activation of PI3K signaling cascade. Experiments were done to determine the role of SLP-76, either wild-type, tyrosine mutants, or membrane-targeted forms of various SLP-76 constructs, on the membrane localization and phosphorylation of Akt, which is an event downstream of PI3K activation. Recon- stitution studies with these various SLP-76 constructs in a Jurkat variant cell line that lacks SLP-76 or linker for activation of T cells (LAT) show that the activation of PI3K pathway following TCR ligation requires both SLP-76 and LATadaptor proteins. The results suggest that SLP-76 associates with p85 after T cell activation and that LAT recruits this complex to the membrane, leading to Akt activation. The Journal of Immunology, 2011, 186: 2926–2935.

rc homology (SH) 2 domain-containing leukocyte phos- recruits multiple signaling intermediates through their SH2 phoprotein of 76 kDa (SLP-76) is an adaptor protein that is domains, including Grb2 (and associated Sos), phospholipase Cg1, S comprised of three motifs allowing for protein–protein and p85, a subunit of PI3K (12–14). The recruitment of SLP-76 interactions: an N-terminal acidic region containing tyrosine to GEMs through its indirect association with LAT by Grb2- phosphorylation sites, a middle proline-rich motif that binds to related adaptor downstream of Shc (Gads) binding is a critical the SH3 domain of Grb2 family members, and a C-terminal SH2 event in the early TCR signaling (10, 11). domain (1). The importance of SLP-76 in T cell development and Previously, we identified p85 as one of the molecules that function has been well characterized in the experiments in the interacts with an N-terminal acidic region of SLP-76 (15). We also Jurkat human T cell leukemia line and in mice made deficient for demonstrated that tyrosine phosphorylation either at the 113 or SLP-76 expression by homologous recombination (2–9). During 128 position is sufficient for the association of SLP-76 with the T cell activation, SLP-76 is recruited to the glycolipid-enriched SH2 domain near the N terminus of p85 (15). A regulatory subunit membrane microdomains (GEMs), also called lipid rafts, upon of PI3K, p85, contains two SH2 domains that are separated by which signaling complexes are formed (10, 11). After the ligation an interacting domain, which binds with a catalytic subunit (16). of TCR, a transmembrane adaptor protein in GEMs, linker for Together with other protein-interaction domains in the catalytic activation of T cells (LAT), is tyrosine-phosphorylated and and regulatory subunits, the SH2 domains recruit PI3K to mem- brane-associated signaling complexes after the activation of protein tyrosine kinases followed by TCR ligation (16). PI3K *Division of Life and Pharmaceutical Sciences, Ewha Womans University, Seoul constitutes an active form of the enzyme by membrane targeting 120-750, Korea; †Department of Life Science, College of Natural Sciences, Ewha Womans University, Seoul 120-750, Korea; and ‡Center for Cell Signaling and Drug of the p110 catalytic subunit and generates phospholipid second Discovery Research, Ewha Womans University, Seoul 120-750, Korea messengers, such as phosphatidylinositol 4,5-biphosphate (PIP2) 1E.K.S. and S.H.J. contributed equally to this study. and phosphatidylinositol 3,4,5-triphosphate (PIP3) (16). Concur- Received for publication May 28, 2010. Accepted for publication December 28, rently, the enzymes phosphoinositide-dependent protein kinase 1 2010. (PDK1) and protein kinase B/Akt are translocated to the plasma This work was supported by grants from the Basic Research Program of the Korea membrane by PIP3 binding, and PDK1 directs phosphorylation Science and Engineering Foundation (R01-2006-000-10429-0 to J.R.L.), the Korea and activation of Akt ensues (16). Once phosphorylated on Thr308 Research Foundation (2009-0074129 to J.R.L.), and the National Core Research 473 Center Program of the Ministry of Science and Technology and the Korea Science and Ser , fully activated Akt then dissociates from the plasma and Engineering Foundation (R15-2006-020) through the Center for Cell Signaling membrane and phosphorylates a variety of substrates, including and Drug Discovery Research at Ewha Womans University. E.K.S. and S.H.J. were supported in part by the Brain Korea 21 Project of the Korea Ministry of Education. glycogen synthase kinase 3b (Gsk3b) and FKHR (FOXO1) (17). Activation of these and other downstream molecules by Akt col- Address correspondence and reprint requests to Dr. Jong Ran Lee, Department of Life Science, College of Natural Sciences, Ewha Womans University, Science Build- lectively results in signals for cell survival, cytokine gene in- ing C-407, 11-1 Daehyun-dong, Seodaemun-gu, Seoul 120-750, Korea. E-mail ad- duction, and cell cycle progression in T cells (17). dress: [email protected] Although an essential role for PI3K signaling in lymphocyte Abbreviations used in this article: Gads, Grb2-related adaptor downstream of Shc; activation is indicated, it remains unknown how TCR activates GEM, glycosphingolipid-enriched membrane domain; Gsk3b, glycogen synthase kinase 3b; IB, immunoblotting; IP, immunoprecipitation; IS, immunological synapse; PI3K. A number of studies have shown possible mechanisms: LAT, linker for activation of T cells; PDK1, phosphoinositide-dependent protein binding to phosphotyrosines in TRIM, LAT, SLP-76, or ZAP-70 kinase 1; PH, pleckstrin homology; PIP2, phosphatidylinositol 3,4-biphosphate; PIP3, phosphatidylinositol 3,4,5-triphosphate; PTEN, phosphatase and tensin homo- through the p85 SH2 domain (15, 18–20); association with Src log; SEE, staphylococcal enterotoxin E; SH, Src homology; siRNA, small interfering kinases by the p85 SH3 domain (21, 22); or direct activation of the RNA; SLP-76, Src homology 2 domain-containing leukocyte protein of 76 kDa; TM, p110 subunit by Ras (23–25). In this study, we describe the role transmembrane region; Wt, wild-type. of two adaptor molecules, SLP-76 and LAT, in the activation of Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 PI3K signaling following TCR stimulation. We show that SLP-76 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1001785 The Journal of Immunology 2927 associates with p85 in Jurkat T cells after the stimulation of TCR. above. All Jurkat-derived T cell lines were cultured in RPMI 1640 medium We also investigated whether the association of these two mole- containing 10% heat-inactivated FCS, 100 U/ml penicillin, 100 mg/ml cules plays a role in the activation of PI3K by examining mem- streptomycin, and 20 mM glutamine. Cell lines were maintained at 37˚C with 5% CO2/95% O2. For activation, Jurkat T cells were washed and brane localization and phosphorylation of Akt. Furthermore, we resuspended in RPMI 1640 medium at 1 3 107 cells/ml and incubated with directly demonstrate the role of SLP-76 on the activation of PI3K/ either isotype control Ab, anti-CD3 (2 mg/ml), and/or anti-28 (10 mg/ml) Akt by performing reconstitution experiments in SLP-76–deficient Abs for 5 min at 37˚C. (J14) or LAT-deficient (JCam2) Jurkat T cells with wild type (Wt), Expression of plasmid DNA tyrosine mutants, and membrane-targeted constructs of SLP-76. Jurkat, JCam2, and J14 cells were electroporated using a Gene Pulser (Bio- Results from the study not only show that TCR-induced PI3K/Akt Rad) at a setting of 250 V and 950 microfarad or a BTX-T820 Electro- activation is dependent on membrane translocation and tyrosine SquarePorator (Genetronics, San Diego, CA) at a setting of 220 V and 65 phosphorylation of SLP-76, but that LAT is also required for the ms in cuvettes containing 2 3 107 cells in 0.4 ml Cytomix (31) in- localization of SLP-76 to the membrane. Collectively, these re- tracellular buffer (pH 7.6; 120 mM KCl, 0.15 mM CaCl2,10mMK2HPO4/ sults suggest distinct roles of two adaptor molecules in the acti- KH2PO4, 25 mM HEPES, 2 mM EGTA, and 5 mM MgCl2) and 20–40 mg indicated plasmid DNA. For transient expression of plasmid DNA, cells vation of PI3K signaling pathway following TCR ligation: SLP- were cultured in growth media for 24 h before performing experiments. To 76 for membrane translocation of p85 and LAT for membrane generate stable cell lines, after 7 d in culture, cells were sorted for ex- translocation of SLP-76. pression of GFP using a FACSAria (BD Biosciences, Mountain View, CA). After an additional 2 wk in culture, cells were sorted again to obtain stable Materials and Methods transfectants showing GFP fluorescence. Similar levels of surface TCR expression were confirmed. The cell lines were monitored regularly for Abs and reagents expression of transfected proteins. Anti-human CD3 mAb (UCHT1) and anti-human CD28 mAb (CD28.2) Human T cell purification and nucleofection from BD Pharmingen (San Diego, CA) was used for cell stimulation. For immunoprecipitation (IP) and immunoblotting (IB), we used anti-PI3K p85 Samples of peripheral blood were collected from normal healthy donors. and anti-LAT polyclonal Abs and anti-myc (9E10) and anti-phos- Written informed consent was obtained from all donors, and all studies were photyrosine (4G10) mAbs from Upstate Biotechnology (Lake Placid, NY), approved by the Institutional Human Ethics Review Board of Ewha anti-SLP-76 and anti-Akt polyclonal Abs from Santa Cruz Biotechnology Womans University Medial Center. Mononuclear cells were isolated by (Santa Cruz, CA), and anti-flag mAb (M2) from Sigma-Aldrich (St Ficoll-Paque Plus density centrifugation, washed, and suspended in PBS Louis, MO). Abs against phospho-Akt (pAkt, Ser473 and Thr308), phos- containing 0.5% BSA. The T cell fraction was isolated by depletion of pho-Gsk3b (pGSK3b, Ser9), and phospho-FKHR (pFKHR, Ser256) non-T cells with superparamagnetic microbead selection using a mixture were from Cell Signaling Technology (Beverly, MA). Secondary Abs, of mAbs and MiniMACS columns (Miltenyi Biotec). The efficiency of anti-mouse IgG and anti-rabbit IgG conjugated with HRP, were pur- purification was verified by flow cytometry, counterstaining with R-PE– chased from Bio-Rad (Hercules, CA). CellTracker Orange CMTMR antihuman CD3 mAb, and usually reached the range of 94–99%. and superantigen staphylococcal enterotoxin E (SEE) were purchased Nucleofection was performed using an Amaxa human T cell Nucleofector from Molecular Probes (Eugene, OR) and Toxin Technology (Sarasota, kit (Lonza, Cologne, Germany) following the manufacturer’s optimized FL), respectively. Ficoll-Paque Plus and ECL reagents were purchased protocol for unstimulated human T cells. Briefly, purified human T cells from Amersham Pharmacia Biotech (Arlington Heights, IL). A Pan (4 3 106 cells/sample) were washed and the cell pellet was carefully T cell isolation kit II from Miltenyi Biotec (Bergisch Gladbach, resuspended in 100 ml Nucleofector solution. After combining 100 ml cell Germany) was used for depletion of non-T cells (negative selection). suspension simultaneously with 5 mg indicated plasmid DNA and 300 nM An Amaxa human T cell Nucleofector kit was obtained from Lonza indicated siRNA, the suspension was transferred into a certified cuvette (Cologne, Germany). ON-TARGET plus SMARTpool small interfering and the selected program (U-014) of the Amaxa Nucleofector kit was RNAs (siRNAs) targeting human SLP-76 and LAT, and ON-TARGET applied. The cells were cultured in growth media for 48 h before per- plus nontargeting pool were purchased from Thermo Scientific Dhar- forming experiments. For activation, the cells were washed and resus- macon (Chicago, IL). pended in RPMI 1640 medium at 1 3 107 cells/ml and incubated with either isotype control Ab or anti-CD3 Ab (5 mg/ml) for 5 min at 37˚C. Plasmids IP and IB The plasmids carrying Wt SLP-76 and its various tyrosine mutants cloned into a modified pEF-Bos vector with an N-terminal sequence encoding the T cell stimulation was terminated by adding an equal volume of ice-cold flag epitope were provided by Dr. G.A. Koretzky (University of Penn- medium, and cell lysates were prepared in 1% Nonidet P-40 lysis buffer sylvania, Philadelphia, PA) (2–4). SLP-76 Wt or Gads binding domain containing protease inhibitors (50 mg/ml aprotinin, 10 mg/ml leupeptin, 40 mutant (SLP-G2) fused with the first 35 aa (transmembrane region [TM]) mg/ml pepstatin A, and 1 mM PMSF) and phosphatase inhibitors (400 mM of LAT in MigR1 and SLP-G2 in pEGFP were also provided by Dr. G.A. sodium vanadate, 10 mM sodium fluoride, and 10 mM sodium pyro- Koretzky (26). The plasmid expressing the LAT-TM fusion of three tyro- phosphate), as described previously (32). For IB, lysates from 1 3 106 sine mutants (TYF) of SLP-76 was generated by ligating PCR fragments of cells were mixed with 23 Laemmli sample buffer, boiled, and subjected to the LAT-TM flanked by BamHI/BglII sites and of the SLP-TYF flanked by 10% SDS-PAGE. After transferring proteins to a nitrocellulose membrane, BamHI/EcoR1 into MigR1 vector (a gift from W. Pear, University of IB was performed by blocking the membrane with 5% nonfat dried milk Pennsylvania, through G.A. Koretzky) (27) at the BglII/EcoR1 sites. LAT and incubating with various primary Abs, followed by HRP-conjugated Wt c-DNA was obtained from Jurkat T RNA by RT-PCR and subcloned secondary Abs. Detection was conducted with ECL reagents. For IP, into MigR1 vector at the BglII/EcoR1 sites. All constructs were confirmed lysates from 10 3 106 cells were tumbled with GammaBind G-Sepharose by sequencing. GFP-Akt-pleckstrin homology (PH) was a gift from Dr. T. beads conjugated with individual Abs as indicated. The immune com- Meyer (Stanford University, Stanford, CA) (28). Hemagglutinin-tagged plexes were then subjected to SDS-PAGE, followed by IB. phosphatase and tensin homolog (PTEN [Wt]) in pCGN vector and myc-tagged Akt (Wt) in pUSEamp vector (Upstate Biotechnology) were Isolation of GEM fractions provided by Drs. S.R. Lee (Chonnam National University, Kwangju, GEM fractions were isolated as described previously with slight mod- Korea) and S.S. Kang (Chungbuk National University, Cheongju, Korea), ifications (33). Briefly, TCR-activated Jurkat T cells (1 3 108) were respectively (29, 30). washed twice in PBS, resuspended in 900 ml buffer A (25 mM HEPES, Cell lines, cell culture, and cell activation 150 mM NaCl, and 1 mM EGTA, with protease and phosphatase inhibitors described for the Nonidet P-40 lysis buffer), and preincubated for 4 min at Human Jurkat T leukemia cells, JCam2, and J14 were gifts from Dr. A. Weiss 37˚C. The detergent Brij-98 in buffer A (100 ml) was added to achieve (University of California, San Francisco, San Francisco, CA). Additional a final concentration of 1%, and the cell suspensions were further in- stable J14 derivatives created using the MigR1 vector and expressing Wt cubated for 5 min at 37˚C. Cell lysates were then mixed with 1 ml 80% SLP-76 or SLP-G2 were gifts from Dr. G.A. Koretzky. A stable J14 de- sucrose in prewarmed buffer A and immediately chilled on ice for 55 min. rivative expressing membrane-bound SLP-TYF (TM-TYF) and stable After transfer to an ultracentrifuge tube, the samples were overlaid with 6.5 JCam2 derivatives expressing Wt LAT, membrane-bound SLP-Wt (TM- ml 30% sucrose in buffer A, followed by 1 ml 5% sucrose in buffer A. All Wt), and TM-TYF were created using the MigR1 vector described procedures were performed on ice. The Brij-insoluble fractions were 2928 REGULATION OF PI3K SIGNALING BY SLP-76 IN T CELLS separated from the cell lysates by ultracentrifugation for 18 h at 40,000 the activation of PI3K signaling in Jurkat or Jurkat variant cells, rpm in a Beckman SW 40.1 Ti rotor at 4˚C. Fractions (1 ml) were collected phosphorylation of Akt was measured in these cells following sequentially from the top of the gradient and analyzed by IB. TCR stimulation. To enhance the sensitivity of detection of Akt Immunological synapse formation and immunocytochemistry phosphorylation, especially in cells transfected with various DNA For conjugation of Jurkat T cells, Raji B cells were used as APCs (34). Prior plasmids throughout the study, myc-tagged Akt was expressed and to conjugation, Raji B cells were stained with CMTMR (4 mM) for 5 min phosphorylation was analyzed by anti-myc IP. As shown in Fig. at 37˚C, washed, and incubated in RPMI 1640 medium with or without 1B, both endogenous and transfected myc-tagged Akt was phos- SEE (5 mg/ml) for 1 h at 37˚C. CMTMR-stained Raji B cells and Jurkat horylated in Jurkat T cells, but not in SLP-76 or LAT-deficient T cells that were transfected with expression plasmids encoding GFP-Akt- mutants, J14 or JCam2 cells, after TCR activation. These results PH or an empty vector were mixed at a 1:1 ratio, immediately transferred to poly-L-lysine–coated slides, and incubated for 15 min at 37˚C. Slides show a correlation between tyrosine-phosphorylated SLP-76 as- were fixed in 3.7% formaldehyde for 20 min at room temperature and sociation with p85 and PI3K activation in TCR-stimulated T cells. analyzed using a Zeiss Axiovert 200 fluorescence microscope that was Because Jurkat T cells are deficient in PTEN (35), we next ex- 3 equipped with a 40 objective lens (Carl Zeiss, Oberkochen, Germany). amined the phosphorylation of Akt in Jurkat T cells supplemented Background fluorescence levels in T cells that express vector-associated GFP were minimal. The settings of the microscopy that corrected for any with a Wt PTEN plasmid. As shown in Fig. 2A, the phosphoryla- 473 308 background fluorescence in T cells were used for fluorescence measure- tion of Akt (both Ser and Thr ) was similar in Jurkat and ments. All images were acquired and analyzed with Zeiss software. Images PTEN-transfected Jurkat T cells after TCR activation. Furthermore, were further processed and quantified with Wright Cell Imaging Facility because the binding site for PI3K lies within the YMNM motif of ImageJ. the CD28 cytoplasmic domain (36), we analyzed the phosphory- Results lation of Akt in Jurkat T cells after CD28 costimulation. Similar levels of Akt phosphorylation were detected after TCR stimulation Association of p85 with tyrosine-phosphorylated SLP-76 alone or together with CD28 costimulation, but a slight phos- results in the phosphorylation of Akt following TCR stimulation phorylation of Akt was seen after CD28 stimulation alone (Fig. 2B). We previously reported that SLP-76 associates with p85, a subunit of PI3K, in Jurkat T cells after the activation of TCR (15). To Tyrosine phosphorylation of SLP-76 at position 113 and/or 128 examine the role of this association in the activation of PI3K is required for association with p85 and consequent signaling, we assayed phosphorylation of Akt, an event that occurs phosphorylation of Akt after TCR activation downstream of PI3K. We first measured the association of p85 Previously, we also demonstrated that tyrosine phosphorylation at with tyrosine-phosphorylated SLP-76 in TCR-stimulated Jurkat residue 113 and/or 128 is sufficient for the association of SLP-76 cells or a Jurkat variant cell line that lacks SLP-76 or LAT (Fig. with the SH2 domain near the N terminus of p85 (15). In this study, 1A). Reciprocal IP detected the association of SLP-76 with p85 we examined the role of these SLP-76 tyrosines in the activation only in Jurkat T cells in which tyrosine phosphorylation of SLP- of PI3K signaling. J14 cells were reconstituted with Wt or various 76 was induced following TCR stimulation (Fig. 1A). To detect tyrosine mutants of SLP-76 along with myc-tagged Akt, and TCR-

FIGURE 1. Association of tyrosine-phosphorylated SLP-76 with p85 leads to Akt phosphorylation following TCR activation. A, Jurkat, SLP-762, and LAT2 cells were either unstimulated (2) or TCR stimulated (+) for 5 min at 37˚C. Lysates were immunoprecipitated with Ab against either SLP-76 or p85. Immunoprecipitates were divided equally and both sets were subjected to SDS-10% PAGE. For IP with anti–SLP-76, tyrosine phosphorylation of SLP-76 following TCR ligation was detected by IB with 4G10, and the same blot was reprobed with anti–SLP-76 for loading controls for IP reactions. To measure phophorylation-dependent association of SLP-76 with p85, another blot was detected by IB with anti-p85 Ab. For IP with anti-p85, tyrosine phosphory- lation of SLP-76 following TCR ligation was detected by IB with 4G10, and the same blot was reprobed with anti–SLP-76 to measure phophorylation- dependent association of SLP-76 with p85. For loading controls for IP reactions, another blot was detected by IB with anti-p85 Ab. Data are representatives of three independent experiments. B, Jurkat, SLP-762, and LAT2 cells were transfected with a plasmid carrying Akt (myc). The cells were either left unstimulated (2) or TCR stimulated (+) for 5 min at 37˚C, and lysates were prepared and subjected to SDS-10% PAGE. Protein expression was detected by IB with anti-myc Ab. Activation induced by TCR stimulation was determined by IB with anti-pAkt Ab. Immunoprecipitates with anti-myc Ab, as well as control anti-mouse Ig (data not shown), were subjected to SDS-10% PAGE. The amounts of IP from unstimulated or TCR stimulated lysates were de- termined by IB with anti-myc Ab and phoshorylation of Akt, both endogenous (endo) and transfected (myc), following TCR stimulation was confirmed by IB with anti-pAkt Ab. Data shown are representative of three separate experiments. The Journal of Immunology 2929

FIGURE 2. Akt phosphorylation in Jurkat T cells after reconstitution with PTEN followed by TCR activation, or after costimulation with TCR and CD28. A, Jurkat T cells were transfected with empty vector (2) or PTEN Wt plasmid [HA (+)]. The cells were either unstimulated (2) or TCR stimulated (+) for 5 min at 37˚C, and lysates were prepared and subjected to SDS-10% PAGE. PTEN expression was detected by IB with anti-HA Ab. Activation induced by TCR stimulation was determined by IB with anti-pAkt Ab (both Ser473 and Thr308). For loading controls in each lane, the same blot probed with anti-pAkt Ab (Thr308) was reprobed with anti-Akt Ab. Data are representative of three independent experiments. B, Jurkat cells were either unstimulated (2), stimulated with TCR+ or CD28+ alone, or costimulated with TCR+ and CD28+ for 5 min at 37˚C. Lysates were prepared and subjected to SDS-10% PAGE. Activation induced by TCR stimulation was determined by IB with anti-pAkt Ab (both Ser473 and Thr308). For loading controls in each lane, the same blot probed with anti-pAkt Ab (Thr308) was reprobed with anti-Akt Ab. Data are representative of three independent experiments. induced phosphorylation of Akt was detected by blotting with that the association of SLP-76 with p85 through phosphorylated anti-myc Ab (Fig. 3A). Consistent with results of the previous tyrosines after TCR stimulation correlates with the activation of interaction studies of SLP-76 with p85, Akt phosphorylation fol- PI3K signaling. lowing TCR stimulation was diminished in J14 cells reconstituted with the SLP-76-Y113F mutant, but not with SLP-76-Y128F or Tyrosine phosphorylation of SLP-76 in the absence of LAT is SLP-76-Y145F (Fig. 3A). Akt phosphorylation following TCR not sufficient for Akt phosphorylation following TCR activation stimulation was further diminished in J14 cells reconstituted with To determine whether tyrosine phosphorylation of SLP-76 is the double tyrosine mutant (Y113, 128F) or triple tyrosine mutant sufficient for Akt phosphorylation, we compared the expression of (TYF) (Fig. 3A). The normalization of pAkt to the level of anti- Wt SLP-76 with the phosphorylation of Akt following TCR myc IP, followed by calculation of induced phosphorylation of Akt stimulation in Jurkat, J14, and JCam2 cells. As shown in Fig. 4A, after TCR activation, is shown in Fig. 3B. These results suggest the levels of TCR-stimulated phosphorylation of Akt were similar

FIGURE 3. Phosphorylation of SLP-76 at tyrosines 113 and/or 128 is required for Akt phosphorylation following TCR activation. A, J14 cells were reconstituted with plasmids carrying SLP-76 Wt (flag) or various tyrosine substitution mutants (flag) along with Akt (myc). The cells were either left unstimulated (2) or TCR stimulated (+) for 5 min at 37˚C, and lysates were prepared and subjected to SDS-10% PAGE. Protein expression was detected by IB with anti-flag or anti-myc Ab. Immunoprecipitates with anti-myc Ab, as well as control anti-mouse Ig (data not shown), were subjected to SDS-10% PAGE. The amounts of IP from lysates were determined by IB with anti-myc Ab, and phosphorylation of Akt following TCR stimulation was confirmed by IB with anti-pAkt Ab. Representative data from three independent experiments are shown. B, The intensities of the bands in pAkt and myc blots were quantified and the normalization of pAkt to the level of anti-myc IP was followed. Values indicate the intensity after TCR stimulation relative to the unstimulated control. 2930 REGULATION OF PI3K SIGNALING BY SLP-76 IN T CELLS in Jurkat T cells overexpressing Wt SLP-76, J14 cells recon- presence of LAT and GM1 (Fig. 5A). The IS formed between stituted with Wt SLP-76, and Jurkat cells transfected with a vector T cells and APCs in the presence and absence of SEE, as well as control. In contrast, overexpression of Wt SLP-76 did not induce the subsequent recruitment of Akt in T cells, is shown in Fig. 5B. phosphorylation of Akt after TCR stimulation in JCam2 cells (Fig. Maximum fluorescence intensity by Akt recruitment to the IS in 4A). These results clearly demonstrate that overexpression of SLP- each T cell–APC conjugate in the presence of SEE was quantified 76 in the absence of LAT is not sufficient to cause the phos- and the intensity values of 35 conjugates were compared between phorylation of Akt following TCR stimulation. Moreover, over- Jurkat, J14, and JCam2 cells, as shown in Fig. 5C. The GFP- expression of tyrosine mutant SLP-76 (TYF) in Jurkat cells did tagged PH domain of Akt (GFP-Akt-PH) was fully recruited to not affect TCR-induced phosphorylation of Akt in Jurkat cells, as IS, where phospholipid second messengers are generated by ac- shown in Fig. 4B. These results suggest that overexpression of the tivated PI3K in Jurkat cells. In contrast, partial or no recruitment SLP-76 triple tyrosine mutant did not interfere with the interaction of Akt-PH to IS was observed in J14 or JCam2 cells (Fig. 5B,5C). between the endogenous tyrosine-phosphorylated SLP-76 and p85, and that the interaction is mediated only by the phosphory- Membrane-targeted SLP-76 is sufficient for the activation of lation of SLP-76. Fig. 4B also shows the results of overexpressing PI3K signaling following TCR ligation SLP-76 (TYF) in J14 and JCam2 cells. The fact that reconstitution The association of p85 with tyrosine-phosphorylated SLP-76 or of J14 cells with SLP-76 (TYF) did not induce phosphorylation of LAT following TCR activation has been demonstrated previously Akt following TCR stimulation further demonstrates that the as- (15, 18, 37). However, p85 association with either SLP-76 or LAT sociation of SLP-76 with p85 through its phosphorylated tyrosines alone did not lead to the activation of PI3K, as shown in Figs. 4 is required for TCR-induced activation of the PI3K cascade. and 5. To investigate the roles of the two adaptor proteins, SLP- 76 and LAT, in PI3K activation, we performed reconstitution GEM recruitment of p85 and immunological synapse experiments with various constructs of SLP-76 and LAT in SLP- localization of Akt are fully induced by TCR activation in the 76–deficient J14 or LAT-deficient JCam2 cells. presence of both SLP-76 and LAT As shown in Fig. 6A, following stimulation of TCR alone or We next examined the membrane translocation of both p85 and together with CD28, the PI3K signaling pathway was activated in SLP-76 and the consequent Akt recruitment to immunological Jurkat cells, but not in J14 and JCam2 cells, as demonstrated by synapse (IS) after TCR activation. As expected, translocation of the phosphorylation of downstream signaling molecules Akt, both SLP-76 and p85 to GEMs following TCR stimulation occurred Gsk3b, and FKHR. To examine whether SLP-76 association with only in Jurkat T cells, and not in J14 and JCam2 cells, as de- LAT or membrane targeting of SLP-76 is required for the PI3K termined by IB analysis of GEM fractions identified by the signaling after TCR ligation, SLP-76–deficient J14 cells were reconstituted with SLP-G2 (Gads binding domain mutant of SLP- 76; consequently not associated with LAT) or TM-SLP-G2 (SLP- G2 fused to LAT TM; consequently membrane-targeted). Ex- pression of SLP-76 Wt and TM-SLP-G2 constructs, but not the SLP-G2 construct alone, reconstituted the TCR- and TCR/CD28- induced PI3K signaling pathways in J14 cells (Fig. 6B). Moreover, transformation of J14 cells with membrane-targeted SLP-76 (TM- SLP-Wt), but not with membrane-targeted tyrosine mutants of SLP-76 (TM-SLP-TYF), induced PI3K signaling following stim- ulation with TCR alone or together with CD28 (Fig. 6C). Ex- pression of LAT Wt and TM-SLP-Wt constructs, but not the TM- SLP-TYF construct, also reconstituted the TCR- and TCR/CD28- induced PI3K signaling pathways in JCam2 cells (Fig. 6D). These results demonstrate that both events, that is, tyrosine phosphory- lation of SLP-76 and membrane targeting of SLP-76 by LAT, are required for the activation of PI3K through T cell activation. The translocation of p85 to GEMs following TCR stimulation was also measured by IB analysis of GEM fractions in J14 cells reconstituted with the TM-SLP-G2 construct and in JCam2 cells reconstituted with TM-SLP-Wt (Fig. 7A). In contrast, the GEM recruitment of p85 was not induced by TCR activation in J14 cells reconstituted with the SLP-G2 construct or in JCam2 cells reconstituted with the TM-SLP-TYF construct (Fig. 7A). In this study, membrane-targeted SLP-76 proteins (TM-SLP-G2, TM- SLP-Wt, and TM-SLP-TYF) were also shown in GEM fractions, FIGURE 4. Phosphorylation of SLP-76 tyrosines is required but not but SLP-G2 was not translocated to GEMs following TCR stim- sufficient for Akt phosphorylation following TCR activation. A, Jurkat, 2 2 ulation due to lack of association with LAT (Fig. 7A). Similarly, SLP-76 , and LAT cells were transfected with plasmids carrying SLP-76 resident Wt SLP-76 in LAT-deficient JCam2 cells was not iden- Wt (flag; or empty vector, pEF) along with Akt (myc). B, Jukat, J14, and tified in GEMs when probed with anti-STP-76, because the re- JCam2 cells were also transfected with plasmids carrying SLP-76 TYF cruitment of SLP-76 to GEMs occurs through its indirect (flag) along with Akt (myc). Stimulation of the cells and IP of lysates with anti-myc Ab were followed, as described in Fig. 3. Protein expression was association with LAT by Gads binding (Fig. 7A). detected by IB with anti-flag or anti-myc Ab. The amounts of IP from The subsequent recruitment of Akt to the IS formed with APCs lysates were determined by IB with anti-myc Ab, and phosphorylation of in the presence and absence of SEE was detected by fluorescence Akt following TCR stimulation was determined by IB with anti-pAkt Ab. detection of the transfected GFP-Akt-PH (Fig. 7B). As shown in Data shown are representatives of three independent experiments. Fig. 5C, the fluorescence intensities representing Akt recruitment The Journal of Immunology 2931

FIGURE 5. Recruitment of p85 and Akt to GEMs and IS following TCR stimulation in Jurkat cells, but not in SLP-762 and LAT2 cells. A,Jurkat,SLP-762, and LAT2 cells were stimulated with either control or anti-CD3 Ab for 5 min at 37˚C. Cells were lysed and GEM fractions were isolated by density gradient ultracentrifugation, as described in Materials and Methods. Fractions 1–9 (50 ml) were resolved by SDS-10% PAGE, and p85 and SLP-76 were detected by IB. GEM-resident proteins such as LAT and GM1 were found primarily in fractions 2 and 3 from all cell lines. IB data in fractions 2 and 3, representative of three separate experiments, are shown. B, Jurkat, SLP-762,andLAT2 cells that were transfected with the GFP-Akt-PH plasmid were stimulated for 15 min with CMTMR-labeled Raji B cells, either pulsed with SEE (+) or without SEE (2), as described in Materials and Methods. Localization of GFP-Akt-PH to the IS (arrowhead) was analyzed, and representative fields of the merged fluorescence pattern are shown for each group. Experiments were repeated at least three times. C, Maximum fluorescence intensity by Akt recruitment at the IS in the presence of SEE was quantified, and the intensity values of 35 conjugates were compared between Jurkat, SLP-762, and LAT2 cells. *p , 0.05, one-way ANOVA (Tukey multiple comparison test). to the IS in T cell–APC conjugates in the presence of SEE are in which each endogenous gene was suppressed, were also elec- compared in Fig. 7C. The IS localization of Akt was induced by troporated with a TM-SLP-Wt or TM-SLP-TYF plasmid. As shown TCR activation in J14 cells reconstituted with SLP-Wt (Fig. 7B, in Fig. 8A and 8C, endogenous SLP-76 or LAT was suppressed 7C). However, recruitment of Akt to the IS was not detected in J14 by specific siRNA, but not by scrambled siRNA. Exogenously cells reconstituted with the SLP-G2 construct (Fig. 7B,7C). These transfected plasmids, TM-SLP-Wt and TM-SLP-TYF, were also results, together with those of p85 recruitment to the GEM, sug- expressed in human T cells in which each gene was suppressed gest that the induction of TCR-activated PI3K signaling requires (Fig. 8A,8C). the recruitment of SLP-76 to the GEM through an indirect asso- Following stimulation with TCR, the PI3K signaling pathway ciation with LAT by Gads binding and the subsequent recruitment was activated in cells transfected with a scrambled siRNA, but not of Akt to the IS formed with APCs. with siRNA specific for either SLP-76 or LAT, as demonstrated by Similarly, the IS localization of Akt was induced by TCR ac- the phosphorylation of downstream signaling molecules Akt, tivation in JCam2 cells reconstituted with TM-SLP-Wt (Fig. 7B, Gsk3b, and FKHR (Fig. 8B,8D). Expression of TM-SLP-Wt, but 7C). However, recruitment of Akt to the IS following TCR stim- not the TM-SLP-TYF construct, reconstituted TCR-induced PI3K ulation was not observed in JCam2 cells reconstituted with the signaling pathways in cells that had either the SLP-76 or LAT TM-SLP-TYF construct (Fig. 7B,7C). These results further sug- gene suppressed (Fig. 8B,8D). We also obtained the same results gest that membrane-targeted tyrosine-phosphorylated SLP-76 is from the same experiments performed in primary T cells isolated sufficient for the induction of TCR-activated PI3K signaling. from mouse spleen (data not shown). The results from experi- ments in primary T cells further demonstrate that tyrosine phos- Tyrosine phosphorylation of SLP-76 and its membrane phorylation of SLP-76 and membrane targeting of SLP-76 by LAT translocation by LAT are also required for TCR-induced PI3K are both required for the activation of PI3K following TCR activation in primary T cells isolated from human peripheral stimulation. blood From experiments performed using various Jurkat T cell systems, Discussion we conclude that SLP-76 and LAT are both required for trans- In this report, we describe the role of two adaptor molecules, SLP- location of p85 to the membrane and subsequent phosphorylation 76 and LAT, in the activation of PI3K signaling following TCR of Akt, and that the two adaptor proteins have different functions in stimulation. Both SLP-76 and LAT are substrates of ZAP-70/Syk the process. SLP-76 associates with p85 after T cell activation, and protein tyrosine kinases and are known to associate with several then LAT recruits this complex to the membrane. To verify these SH2 domain-containing molecules, including the p85 subunit of findings in primary T cell systems, we carried out key experiments PI3K, through phosphotyrosine residues upon T cell activation (15, in primary T cells isolated from human peripheral blood. 18, 37). Human T cells were electroporated with a scrambled siRNA or Upon T cell activation, LAT associates directly with Grb2, a specific siRNA for either SLP-76 or the LAT gene. Human T cells, phospholipase Cg1, and the p85 subunit of PI3K. Two tyrosine 2932 REGULATION OF PI3K SIGNALING BY SLP-76 IN T CELLS

FIGURE 6. Both tyrosine phosphorylation and membrane recruitment of SLP-76 are required for the activation of PI3K signaling pathway after TCR stimulation. A, Jurkat, LAT2, and SLP-762 cells were stimulated either with control or anti-CD3 Ab (+) or costimulated with anti-CD3 Ab (+) and anti- CD28 Ab (+). Cell lysates were prepared and subjected to SDS-10% PAGE. Phosphorylation of proteins following stimulation was detected by IB with indicated phospho-protein–specific Abs. The same blot probed with anti-pAkt was reprobed with anti-Akt Ab to demonstrate equal sample loading. B, The empty vector (MigR1), SLP-76 Wt (SLP-Wt), Gads binding domain mutant (SLP-G2), and the plasmid expressing the TM of LAT fused to SLP-G2 (TM- SLP-G2) were stably expressed in SLP-762 cells. C, The empty vector (MigR1), SLP-76 Wt (SLP-Wt), and the plasmids expressing the TM of LAT fused to SLP-Wt (TM-SLP-Wt), or three tyrosine mutants of SLP-76 (TM-SLP-TYF), were stably expressed in SLP-762 cells. D, The empty vector (MigR1), LAT Wt, TM-SLP-Wt, and TM-SLP-TYF were stably expressed in LAT2 cells. Stimulation of all cells and analyses of activation were followed, as described in A. Representative data from three independent experiments are shown. residues, Y171 and Y191 of LAT, appear to be required for as- canonical phosphotyrosines in LAT, SLP-76, or ZAP-70 through sociation with these signaling molecules (38). LAT binds Cbl, Sos, the p85 SH2 domain (15, 18–20); association with Src kinases by Vav, and SLP-76, probably indirectly via Grb2 (38). SLP-76 the p85 SH3 domain (21, 22); or direct activation of p110 subunit associates with Vav1, noncatalytic kinase, IL-2–induced tyrosine by Ras (23–25). kinase, and the p85 subunit of PI3K through phosphorylation of The clearest link between T lymphocyte signaling and PI3K tyrosine residues Y113, Y128, and Y145 upon T cell activation activation is the costimulatory molecule CD28. The p85 regulatory (15, 39). Upon TCR activation, the tyrosine phosphorylation and subunit of PI3K, being associated with a phosphorylated YXXM consequent association of these proteins occur at a special junc- motif in the cytoplasmic tail of CD28, recruits the p110 catalytic tion, the IS, formed at the T cell–APC interface (40, 41). subunit, which converts PIP2 to PIP3 at the membrane (36, 52, 53). The family of PI3Ks includes the p85/p110 isoform, which Locally generated PIP3 serves as a docking site for the PH domain comprises two associated proteins, the p85 regulatory subunit and of PDK1and its target Akt, resulting in the sequential activation of the catalytic subunit p110 (42–45). p85 contains two SH2 domains these kinases. Activated Akt phosphorylates multiple proteins, and one SH3 domain, which allow interactions with other sig- including Gsk3b and FKHR, enabling them to affect numerous naling proteins, resulting in membrane translocation and activa- cellular responses (36, 52, 53). tion of the p110 catalytic subunit. p85 is recruited to the plasma In this report, we studied the role of two adaptor molecules, SLP- membrane through its SH2 domain by tyrosine-phosphorylated 76 and LAT, in PI3K activation following TCR ligation. It has been adaptor molecules, and then 39-phosphoinositides are produced shown that both SLP-76 and LAT associate with the SH2 domain of by the activated PI3K catalytic subunit. The production of PIP2 p85 through their phosphotyrosine residues upon TCR activation and PIP3, which are undetectable in unstimulated cells, is a key (15, 18, 37). In this study, we demonstrated the TCR-induced event during signaling, as it allows the membrane recruitment of interaction of p85 and SLP-76 occurs in Jurkat T cells, but not many PH domain-containing proteins involved in cell growth, in J14 or JCam2 cells, and results in Akt activation upon T cell survival, or cytoskeletal rearrangement (43, 44). activation (Fig. 1). Reconstitution of J14 cells with various SLP- In T cells, Ag recognition is followed by rapid and sustained 76 mutants demonstrated that tyrosine phosphorylation at 113 or accumulation of the PI3K product, PIP3 at the plasma membrane, 128 is sufficient for the activation of Akt (Fig. 3). Additionally, we particularly concentrated at the IS (36, 45, 46). Genetic manipu- show that membrane-targeted forms of Wt SLP-76, but not of its lations that enhance PI3K pathway activity cause lymphoprolif- tyrosine mutants, reconstituted Akt phosphorylation in activated eration in mice (47–49). Conversely, pharmacologic inhibitors of J14 or JCam2 cells (Fig. 6). PI3K, such as wortmannin and LY294002, potently block T cell Basal levels of Akt activation in Jurkat T cells have been shown proliferation (50, 51). These observations are indicative of an to be high as a consequence of PTEN loss in these cells (35, 54). essential role for PI3K signaling in lymphocyte activation. How Because we obtained our results using various Jurkat derivatives, TCR activates PI3K remains unknown, although a number of we carefully examined the endogenous phosphorylation of Akt, as possible mechanisms have been reported: binding to YXXM well as downstream targets Gsk3b and FKHR. However, the level motifs in the transmembrane adaptor protein TRIM or to non- of endogenous Akt activation was not as high as we expected and The Journal of Immunology 2933

FIGURE 7. Recruitment of p85 and Akt to GEMs and IS following TCR stimulation in SLP-762 or LAT2 cells reconstituted with TM-SLP-G2 or TM- SLP-Wt, but not with SLP-G2 or TM-SLP-TYF plasmid. A, SLP-762 cells stably reconstituted with TM-SLP-G2 or SLP-G2 plasmid (SLP-76(2)/TM- SLP-G2 or SLP-G2) and LAT2 cells stably reconstituted with TM-SLP-Wt or TM-SLP-TYF plasmid (LAT(2)/TM-SLP-Wt or TM-SLP-TYF) were stimulated with either control or anti-CD3 Abs for 5 min at 37˚C. GEM fractions were isolated from these cells by density gradient ultracentrifugation, and IB was performed with anti-p85 and anti-SLP-76 Abs, as described for Fig. 5A. Data are representative of three separate experiments. B and C, SLP-76(2)/ SLP-Wt or SLP-G2, and LAT(2)/TM-SLP-Wt or TM-SLP-TYF cells were transfected with the GFP-Akt-PH plasmid. Cell stimulation and IS localization analysis of GFP-Akt-PH (arrowhead) were followed, as described for Fig. 5B. Representative fields of the merged fluorescence pattern for each group are shown from three separate experiments. *p , 0.05, one-way ANOVA (Tukey multiple comparison test). it varied greatly with experimental conditions, such as resting ments. Above all, we did not see much of a difference in the level condition of the cells. A similar level of endogenous phosphory- of Akt activation after PTEN was added back to Jurkat T cells lation of Akt was also observed in primary human T cell experi- (Fig. 2A).

FIGURE 8. Expression of membrane-targeted and tyrosine-phosphorylated SLP-76 activates PI3K signaling pathway after TCR stimulation in primary human T cells in which the SLP-76 or LAT gene was suppressed. A and C, Primary T cells isolated from human peripheral blood were transfected with scrambled or SLP-76 (or LAT) gene-specific siRNA. Either TM-SLP-Wt or TM-SLP-TYF plasmid was expressed in SLP-76 (or LAT) gene-suppressed primary human T cells. Cell lysates were prepared and subjected to SDS-10% PAGE. Gene suppression, as well as expression of both endogenous (endo) and exogenously transfected (exo) SLP-76, was detected by IB with anti-SLP-76 and anti-LAT Abs. B and D, The cells prepared in A and C were either unstimulated (2) or TCR stimulated (+) for 5 min at 37˚C, and lysates were prepared and subjected to SDS-10% PAGE. Phosphorylation of proteins following stimulation was detected by IB with indicated phospho-protein–specific Abs. The same blot probed with anti-pAkt was reprobed with anti-Akt Ab to demonstrate equal sample loading. All experiments were repeated twice and similar results were obtained. 2934 REGULATION OF PI3K SIGNALING BY SLP-76 IN T CELLS

CD28 costimulation also does not seem to enhance TCR-induced SLP-76 to the membrane in Jurkat cells (Figs. 4, 6, 7) and primary PI3K activation in our Jurkat system. As shown in Fig. 2B, human T cells (Fig. 8C,8D). These findings suggest a new par- stimulation of CD28 alone induced a small amount of Akt acti- adigm for PI3K/Akt activation mechanisms that are known to be vation, whereas costimulation of TCR/CD28 did not significantly required for T cell activation. At the same time, this new mech- enhance the TCR-induced Akt activation detected by phosphory- anism of collaboration between SLP-76 and LAT adaptor mole- lation of either Ser473 or Thr308. These results are somewhat cules may provide a reason to re-evaluate many controversial contradictory to the results obtained in primary mouse T cells, results and issues regarding PI3K/Akt activation signaling in where CD28 stimulation and TCR/CD28 costimulation strongly T lymphocytes, such as the Vav1 independence of the activation of induced Akt activation compared with TCR stimulation alone Akt (55). Collectively, the results of this study illustrate distinct (55). However, in the same report, the level of Akt activation by roles for two adaptor molecules, SLP-76 and LAT, in the activa- TCR or CD28 stimulation or TCR/CD28 costimulation of Jurkat tion of PI3K following TCR ligation. T cells was similar to what we measured (55). This variability could be related to differences in the degree of surface receptor Acknowledgments cross-linking by different Abs or a difference between the origin We thank Drs. G.A. Koretzky, W. Pear, T. Meyer, S.S. Kang, S.R. Lee, and and maintenance of T cells used. A. Weiss for providing valuable reagents and cell lines. Three tyrosine residues in the adaptor protein SLP-76 reside within YE(S/P)P motifs: both Y113 and Y128 within YESP, and Disclosures Y145 within YEPP (3). These residues are phosphorylated upon The authors have no financial conflicts of interest. receptor ligation and associate with many signaling molecules that contain SH2 domains (39, 56). Association of Y113 and Y128 with Vav and noncatalytic kinase is important for TCR-induced References phosphorylation of proximal signaling substrates (39, 56). Y145 is 1. Jackman, J. K., D. G. Motto, Q. Sun, M. Tanemoto, C. W. Turck, G. A. Peltz, G. A. Koretzky, and P. R. Findell. 1995. Molecular cloning of SLP-76, a 76-kDa important for the optimal association of SLP-76 with IL-2–in- tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270: duced tyrosine kinase, and it is the most critical tyrosine for T cell 7029–7032. development and functions (56, 57). We have previously shown 2. Motto, D. G., S. E. Ross, J. Wu, L. R. Hendricks-Taylor, and G. A. Koretzky. 1996. Implication of the GRB2-associated phosphoprotein SLP-76 in T cell that SLP-76 associates with the p85 subunit of PI3K and that receptor-mediated interleukin 2 production. J. Exp. Med. 183: 1937–1943. phosphorylation of Y113 or Y128 is sufficient for binding to p85 3. Fang, N., D. G. Motto, S. E. Ross, and G. A. Koretzky. 1996. Tyrosines 113, 128, (15). TCR-induced Akt phosphorylation was inhibited in SLP-76– and 145 of SLP-76 are required for optimal augmentation of NFAT promoter activity. J. Immunol. 157: 3769–3773. deficient J14 cells reconstituted with mutant SLP-76 with Y113F, 4. Musci, M. A., D. G. Motto, S. E. Ross, N. Fang, and G. A. Koretzky. 1997. Three Y113 and 128F, or TYF (Y113, 128, and 145F), and this result domains of SLP-76 are required for its optimal function in a T cell line. J. Immunol. 159: 1639–1647. further demonstrates SLP-76 involvement in the PI3K signaling 5. Yablonski, D., M. R. Kuhne, T. Kadlecek, and A. Weiss. 1998. Uncoupling of pathway (Fig. 3). However, it is possible that other molecules that nonreceptor tyrosine kinases from PLC-g1 in an SLP-76-deficient T cell. Science associate with the same residues within SLP-76 are involved in 281: 413–416. 6. Clements, J. L., B. Yang, S. E. Ross-Barta, S. L. Eliason, R. F. Hrstka, PI3K activation. Vav, which associates with the Y113 residue of R. A. Williamson, and G. A. Koretzky. 1998. Requirement for the leukocyte- SLP-76 after TCR ligation, is one candidate because Vav also has specific adapter protein SLP-76 for normal T cell development. Science 281: a role in PI3K activation (58, 59). 416–419. 7. Pivniouk, V., E. Tsitsikov, P. Swinton, G. Rathbun, F. W. Alt, and R. S. Geha. LAT undergoes a post-translational modification in the form of 1998. Impaired viability and profound block in thymocyte development in mice palmitoylation, which leads to its localization to GEM fractions, lacking the adaptor protein SLP-76. Cell 94: 229–238. 8. Myung, P. S., G. S. Derimanov, M. S. Jordan, J. A. Punt, Q. H. Liu, B. A. Judd, and this localization is essential for LAT signaling (12, 18, 38). E. E. Meyers, C. D. Sigmund, B. D. Freedman, and G. A. Koretzky. 2001. Many signaling molecules, including p85, Vav, and SLP-76, are Differential requirement for SLP-76 domains in T cell development and func- associated either directly or indirectly with LAT after TCR acti- tion. Immunity 15: 1011–1026. 9. Kumar, L., V. Pivniouk, M. A. de la Fuente, D. Laouini, and R. S. Geha. 2002. vation, and they also localize to GEM (18, 38). In this study, we Differential role of SLP-76 domains in T cell development and function. Proc. show that TCR-induced localization of p85 and SLP-76 did not Natl. Acad. Sci. USA 99: 884–889. occur at the GEM of LAT-deficient JCam2 cells (Fig. 5A). Con- 10. Boerth, N. J., J. J. Sadler, D. E. Bauer, J. L. Clements, S. M. Gheith, and G. A. Koretzky. 2000. Recruitment of SLP-76 to the membrane and glycolipid- sequently, Akt localization to IS of T cell–APC conjugates was enriched membrane microdomains replaces the requirement for linker for acti- completely lost in JCam2 cells, as shown in Fig. 5B and 5C. When vation of T cells in T cell receptor signaling. J. Exp. Med. 192: 1047–1058. 11. Ishiai, M., M. Kurosaki, K. Inabe, A. C. Chan, K. Sugamura, and T. Kurosaki. Akt activation was measured by phosphorylation of Akt, we found 2000. Involvement of LAT, Gads, and Grb2 in compartmentation of SLP-76 to 473 308 that Akt phosphorylation at Ser , as well as Thr , was un- the plasma membrane. J. Exp. Med. 192: 847–856. detectable after stimulation with TCR or TCR/CD28 in the ab- 12. Zhang, W., R. P. Trible, and L. E. Samelson. 1998. LAT palmitoylation: its essential role in membrane microdomain targeting and tyrosine phosphorylation sence of LAT (data not shown). Other studies have shown that during T cell activation. Immunity 9: 239–246. PI3K is stably activated in the IS of T cells/APCs to stimulate Akt 13. Lin, J., A. Weiss, and T. S. Finco. 1999. Localization of LAT in glycolipid- signaling (46), and that GEM-associated kinase, mammalian tar- enriched microdomains is required for T cell activation. J. Biol. Chem. 274: 473 28861–28864. get of rapamycin complex 2, phosphorylates Akt Ser to induce 14. Wange, R. L. 2000. LAT, the linker for activation of T cells: a bridge between full activation of Akt, which requires phosphorylation on residues T cell-specific and general signaling pathways. Sci. STKE 2000: re1. 308 473 15. Shim, E. K., C. S. Moon, G. Y. Lee, Y. J. Ha, S.-K. Chae, and J. R. Lee. 2004. Thr and Ser (60, 61). Several studies also showed re- Association of the Src homology 2 domain-containing leukocyte phosphoprotein cruitment of Akt to the GEM in activated T cells (62–64). of 76 kD (SLP-76) with the p85 subunit of phosphoinositide 3-kinase. FEBS Based on the results of this study, we propose a new role for SLP- Lett. 575: 35–40. 16. Patel, R. K., and C. Mohan. 2005. PI3K/AKT signaling and systemic autoim- 76 adaptor protein in the signaling cascade leading to Akt acti- munity. Immunol. Res. 31: 47–55. vation, mediated by direct association with the p85 subunit of 17. Manning, B. D., and L. C. Cantley. 2007. AKT/PKB signaling: navigating PI3K (Fig. 1). We present data showing that TCR-induced PI3K/ downstream. Cell 129: 1261–1274. 18. Zhang, W., J. Sloan-Lancaster, J. Kitchen, R. P. Trible, and L. E. Samelson. Akt activation requires membrane translocation and tyrosine 1998. LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to phosphorylation of the SLP-76 adaptor protein in either the Jurkat cellular activation. Cell 92: 83–92. 19. Bruyns, E., A. Marie-Cardine, H. Kirchgessner, K. Sagolla, A. Shevchenko, T cell system (Figs. 4, 6, 7) or primary human T cells (Fig. 8A, M. Mann, F. Autschbach, A. Bensussan, S. Meuer, and B. Schraven. 1998. T cell 8B). We also show that LAT is required for the localization of receptor (TCR) interacting molecule (TRIM), a novel disulfide-linked dimer The Journal of Immunology 2935

associated with the TCR-CD3-z complex, recruits intracellular signaling proteins 41. Huppa, J. B., and M. M. Davis. 2003. T-cell-antigen recognition and the im- to the plasma membrane. J. Exp. Med. 188: 561–575. munological synapse. Nat. Rev. Immunol. 3: 973–983. 20. Moon, K. D., C. B. Post, D. L. Durden, Q. Zhou, P. De, M. L. Harrison, and 42. Okkenhaug, K., K. Ali, and B. Vanhaesebroeck. 2007. Antigen receptor sig- R. L. Geahlen. 2005. Molecular basis for a direct interaction between the Syk nalling: a distinctive role for the p110d isoform of PI3K. Trends Immunol. 28: protein-tyrosine kinase and phosphoinositide 3-kinase. J. Biol. Chem. 280: 80–87. 1543–1551. 43. Lemmon, M. A. 2007. Pleckstrin homology (PH) domains and phosphoinosi- 21. Prasad, K. V., O. Janssen, R. Kapeller, M. Raab, L. C. Cantley, and C. E. Rudd. tides. Biochem. Soc. Symp. 74: 81–93. 1993. Src-homology 3 domain of protein kinase p59fyn mediates binding to 44. Deane, J. A., and D. A. Fruman. 2004. Phosphoinositide 3-kinase: diverse roles phosphatidylinositol 3-kinase in T cells. Proc. Natl. Acad. Sci. USA 90: 7366–7370. in immune cell activation. Annu. Rev. Immunol. 22: 563–598. 22. Pleiman, C. M., W. M. Hertz, and J. C. Cambier. 1994. Activation of phos- 45. Costello, P. S., M. Gallagher, and D. A. Cantrell. 2002. Sustained and dynamic phatidylinositol-39 kinase by Src-family kinase SH3 binding to the p85 subunit. inositol lipid metabolism inside and outside the immunological synapse. Nat. Science 263: 1609–1612. Immunol. 3: 1082–1089. 23. Rodriguez-Viciana, P., P. H. Warne, R. Dhand, B. Vanhaesebroeck, I. Gout, 46. Fabre, S., V. Lang, J. Harriague, A. Jobart, T. G. Unterman, A. Trautmann, and M. J. Fry, M. D. Waterfield, and J. Downward. 1994. Phosphatidylinositol-3-OH G. Bismuth. 2005. Stable activation of phosphatidylinositol 3-kinase in the T cell kinase as a direct target of Ras. Nature 370: 527–532. immunological synapse stimulates Akt signaling to FoxO1 nuclear exclusion and 24. Suire, S., P. Hawkins, and L. Stephens. 2002. Activation of phosphoinositide 3- cell growth control. J. Immunol. 174: 4161–4171. kinase g by Ras. Curr. Biol. 12: 1068–1075. 47. Okkenhaug, K., A. Bilancio, G. Farjot, H. Priddle, S. Sancho, E. Peskett, 25. Wabnitz, G. H., G. Nebl, M. Klemke, A. J. Schro¨der, and Y. Samstag. 2006. W. Pearce, S. E. Meek, A. Salpekar, M. D. Waterfield, et al. 2002. Impaired B Phosphatidylinositol 3-kinase functions as a Ras effector in the signaling cascade and TCR signaling in p110d PI 3-kinase mutant mice. Science 297: 1031–1034. that regulates dephosphorylation of the actin-remodeling protein cofilin after co- 48. Okkenhaug, K., D. T. Patton, A. Bilancio, F. Garc¸on, W. C. Rowan, and stimulation of untransformed human T lymphocytes. J. Immunol. 176: 1668–1674. B. Vanhaesebroeck. 2006. The p110d isoform of phosphoinositide 3-kinase 26. Singer, A. L., S. C. Bunnell, A. E. Obstfeld, M. S. Jordan, J. N. Wu, P. S. Myung, controls clonal expansion and differentiation of Th cells. J. Immunol. 177: 5122– L. E. Samelson, and G. A. Koretzky. 2004. Roles of the proline-rich domain in 5128. SLP-76 subcellular localization and T cell function. J. Biol. Chem. 279: 15481– 49. Deane, J. A., M. J. Trifilo, C. M. Yballe, S. Choi, T. E. Lane, and D. A. Fruman. 15490. 2004. Enhanced T cell proliferation in mice lacking the p85b subunit of phos- 27. Pear, W. S., J. P. Miller, L. Xu, J. C. Pui, B. Soffer, R. C. Quackenbush, phoinositide 3-kinase. J. Immunol. 172: 6615–6625. A. M. Pendergast, R. Bronson, J. C. Aster, M. L. Scott, and D. Baltimore. 1998. 50. Ward, S. G., C. H. June, and D. Olive. 1996. PI 3-kinase: a pivotal pathway in Efficient and rapid induction of a chronic myelogenous leukemia-like myelo- T-cell activation? Immunol. Today 17: 187–197. proliferative disease in mice receiving P210 bcr/abl-transduced bone marrow. 51. Ward, S. G., A. Wilson, L. Turner, J. Westwick, and D. M. Sansom. 1995. In- Blood 92: 3780–3792. hibition of CD28-mediated T cell costimulation by the phosphoinositide 3-ki- 28. Harriague, J., and G. Bismuth. 2002. Imaging antigen-induced PI3K activation in nase inhibitor wortmannin. Eur. J. Immunol. 25: 526–532. T cells. Nat. Immunol. 3: 1090–1096. 52. Page`s, F., M. Ragueneau, R. Rottapel, A. Truneh, J. Nunes, J. Imbert, and 29. Lee, S. R., K. S. Yang, J. Kwon, C. Lee, W. Jeong, and S. G. Rhee. 2002. Re- D. Olive. 1994. Binding of phosphatidylinositol-3-OH kinase to CD28 is re- versible inactivation of the tumor suppressor PTEN by H2O2. J. Biol. Chem. 277: quired for T-cell signalling. Nature 369: 327–329. 20336–20342. 53. Acuto, O., and F. Michel. 2003. CD28-mediated co-stimulation: a quantitative 30. Yusoff, P., D.-H. Lao, S. H. Ong, E. S. Wong, J. Lim, T. L. Lo, H. F. Leong, support for TCR signalling. Nat. Rev. Immunol. 3: 939–951. C. W. Fong, and G. R. Guy. 2002. Sprouty2 inhibits the Ras/MAP kinase 54. Ward, S. G., and D. A. Cantrell. 2001. Phosphoinositide 3-kinases in pathway by inhibiting the activation of Raf. J. Biol. Chem. 277: 3195–3201. T lymphocyte activation. Curr. Opin. Immunol. 13: 332–338. 31. van den Hoff, M. J., A. F. Moorman, and W. H. Lamers. 1992. Electroporation in 55. Wood, J. E., H. Schneider, and C. E. Rudd. 2006. TcR and TcR-CD28 en- “intracellular” buffer increases cell survival. Nucleic Acids Res. 20: 2902. gagement of protein kinase B (PKB/AKT) and glycogen synthase kinase-3 32. Lee, J. R., and G. A. Koretzky. 1998. Extracellular signal-regulated kinase-2, but (GSK-3) operates independently of guanine nucleotide exchange factor VAV-1. not c-Jun NH2-terminal kinase, activation correlates with surface IgM-mediated J. Biol. Chem. 281: 32385–32394. apoptosis in the WEHI 231 B cell line. J. Immunol. 161: 1637–1644. 56. Jordan, M. S., J. Sadler, J. E. Austin, L. D. Finkelstein, A. L. Singer, 33. Drevot, P., C. Langlet, X. J. Guo, A. M. Bernard, O. Colard, J. P. Chauvin, P. L. Schwartzberg, and G. A. Koretzky. 2006. Functional hierarchy of the N- R. Lasserre, and H. T. He. 2002. TCR signal initiation machinery is pre- terminal tyrosines of SLP-76. J. Immunol. 176: 2430–2438. assembled and activated in a subset of membrane rafts. EMBO J. 21: 1899–1908. 57. Bunnell, S. C., M. Diehn, M. B. Yaffe, P. R. Findell, L. C. Cantley, and 34. Charvet, C., A. J. Canonigo, D. D. Billadeau, and A. Altman. 2005. Membrane L. J. Berg. 2000. Biochemical interactions integrating Itk with the T cell localization and function of Vav3 in T cells depend on its association with the receptor-initiated signaling cascade. J. Biol. Chem. 275: 2219–2230. adapter SLP-76. J. Biol. Chem. 280: 15289–15299. 58. Reynolds, L. F., L. A. Smyth, T. Norton, N. Freshney, J. Downward, D. Kioussis, 35. Shan, X., M. J. Czar, S. C. Bunnell, P. Liu, Y. Liu, P. L. Schwartzberg, and and V. L. Tybulewicz. 2002. Vav1 transduces T cell receptor signals to the ac- R. L. Wange. 2000. Deficiency of PTEN in Jurkat T cells causes constitutive tivation of phospholipase C-g1 via phosphoinositide 3-kinase-dependent and localization of Itk to the plasma membrane and hyperresponsiveness to CD3 -independent pathways. J. Exp. Med. 195: 1103–1114. stimulation. Mol. Cell. Biol. 20: 6945–6957. 59. Inabe, K., M. Ishiai, A. M. Scharenberg, N. Freshney, J. Downward, and 36. Harada, Y., E. Tanabe, R. Watanabe, B. D. Weiss, A. Matsumoto, H. Ariga, T. Kurosaki. 2002. Vav3 modulates B cell receptor responses by regulating O. Koiwai, Y. Fukui, M. Kubo, C. H. June, and R. Abe. 2001. Novel role of phosphoinositide 3-kinase activation. J. Exp. Med. 195: 189–200. phosphatidylinositol 3-kinase in CD28-mediated costimulation. J. Biol. Chem. 60. Hill, M. M., J. Feng, and B. A. Hemmings. 2002. Identification of a plasma 276: 9003–9008. membrane Raft-associated PKB Ser473 kinase activity that is distinct from ILK 37. Paz, P. E., S. Wang, H. Clarke, X. Lu, D. Stokoe, and A. Abo. 2001. Mapping the and PDK1. Curr. Biol. 12: 1251–1255. Zap-70 phosphorylation sites on LAT (linker for activation of T cells) required 61. Bozulic, L., and B. A. Hemmings. 2009. PIKKing on PKB: regulation of PKB for recruitment and activation of signalling proteins in T cells. Biochem. J. 356: activity by phosphorylation. Curr. Opin. Cell Biol. 21: 256–261. 461–471. 62. Kane, L. P., M. N. Mollenauer, and A. Weiss. 2004. A proline-rich motif in the C 38. Sommers, C. L., L. E. Samelson, and P. E. Love. 2004. LAT: a T lymphocyte terminus of Akt contributes to its localization in the immunological synapse. J. adapter protein that couples the antigen receptor to downstream signaling Immunol. 172: 5441–5449. pathways. Bioessays 26: 61–67. 63. Bauer, B., M. Jenny, F. Fresser, F. Uberall, and G. Baier. 2003. AKT1/PKBa is 39. Koretzky, G. A., F. Abtahian, and M. A. Silverman. 2006. SLP76 and SLP65: recruited to lipid rafts and activated downstream of PKC isotypes in CD3- complex regulation of signalling in lymphocytes and beyond. Nat. Rev. Immunol. induced T cell signaling. FEBS Lett. 541: 155–162. 6: 67–78. 64. Na, S. Y., A. Patra, Y. Scheuring, A. Marx, M. Tolaini, D. Kioussis, 40. Grakoui, A., S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen, B. A. Hemmings, T. Hu¨nig, and U. Bommhardt. 2003. Constitutively active and M. L. Dustin. 1999. The immunological synapse: a molecular machine protein kinase B enhances Lck and Erk activities and influences thymocyte se- controlling T cell activation. Science 285: 221–227. lection and activation. J. Immunol. 171: 1285–1296.