INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The plasma membrane lipid bilayer contains membrane domains with distinct composition (32). These membrane lipid rafts, alternatively known as detergent-insoluble membranes, detergent-resistant membranes, detergent-insoluble glycolipid-rich membranes, triton-insoluble floating fraction, or glycosphingolipid-enriched membranes, were initially defined by their insolubility in cold, nonionic detergents, a characteristic that, together with their comparatively high buoyancy, facilitates their isolation on sucrose gradients (2). Lipid rafts are enriched in sphingolipids stabilized by intercalating cholesterol (2, 11) and form liquid-ordered assemblies separate from the phospholipid bilayer. These special physicochemical characteristics are associated with distinct functions, including a role in signal transduction (32).

Numerous observations support a function for lipid rafts in T-cell receptor (TCR) signaling. Several molecules implicated in TCR signal transduction associate constitutively with lipid rafts. They include two Src kinase family members, Lck and Fyn, implicated in TCR signal initiation (31). The T-cell-specific adapter Lat is also constitutively associated with lipid rafts, where it functions as an activation scaffold after tyrosine phosphorylation (8, 19, 46, 47). Furthermore, TCR signaling requires proper compartmentalization of Lck to the lipid rafts (12) and a Lat mutant which failed to associate with lipid rafts was not phosphorylated and resulted in defective TCR signaling (48).

In addition to molecules constitutively present in the lipid rafts, TCR engagement induces raft recruitment of proximal and distal elements of the TCR activation sequence, including the TCR-associated  chains, and several adapter and effector molecules (22, 42, 48). Moreover, disruption of raft organization by cholesterol sequestration impairs T-cell activation (42).

A central issue in the role played by membrane rafts in signal transduction is whether recompartmentalization of effector molecules to the lipid rafts controls their activation status and induces downstream signaling events. In TCR signal transduction, a fraction of phospholipase C-1 (PLC1), a cytosolic protein, is inducibly recompartmentalized to the lipid rafts (42, 48). PLC1-mediated hydrolysis of phosphatidylinositol (4,5)-bisphosphate to inositol (1,4,5)-trisphosphate and diacylglycerol controls Ca²⁺ mobilization and protein kinase C activation, respectively (25), critical steps that regulate interleukin 2 (IL-2) transcription (4). We therefore questioned whether PLC1 activation is controlled by recompartmentalization to the lipid rafts.

The TCR-induced signaling cascade ensues with Lck and Fyn activation, leading to the phosphorylation of conserved immunereceptor tyrosine-based activation motifs (ITAMs) on the TCR-associated CD3 and  chains (13). Phosphorylated ITAMs recruit the T-cell-specific kinase, Zap-70, which is subsequently activated via Lck phosphorylation (13). Activated Zap-70 phosphorylates Lat (40, 47) and a second T-cell-specific adapter, Slp-76 (43). PLC1 then engages tyrosine-phosphorylated Lat, thereby localizing PLC1 to the lipid rafts (40, 47). PLC1 activation is regulated by tyrosine phosphorylation (39) via a mechanism that, in addition to Zap-70 and Lat (40, 47), requires Slp-76 (43) and members of the Tec kinase family, Itk and Rlk (27). Interestingly, tyrosine-phosphorylated PLC1 is enriched in lipid rafts (42, 48).

To investigate if the activation status of PLC1 is controlled by recompartmentalization to the lipid rafts, we engineered a form of the enzyme that was constitutively present in the lipid rafts. Expression of a raft-targeted form of the enzyme in Jurkat T leukemia cells resulted in the constitutive tyrosine phosphorylation and activation of PLC1 with ensuing Ca²⁺-dependent transcriptional activation. Phosphorylation of raft-targeted PLC1 required Lck but dispensed with Zap-70, Lat, and Slp-76. Thus, relocalization to membrane rafts is sufficient for PLC1 phosphorylation and bypasses TCR engagement and the activation of early steps in the TCR-signaling sequence.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Plasmids, cell lines, and reagents. The HA-tagged bovine PLC1 in the expression vector pCIneo (Promega, Madison, Wis.) has been previously described (33). The amino-terminal palmitoylation signal sequence, (M)GCVQCKDKEA, and the control sequence, (M)ASVQCKDKEA, were introduced by PCR using appropriate oligonucleotides. The amplified fragments were shuttled via XbaI and BamHI sites into a pBluescript II SK() vector (Stratagene, La Jolla, Calif.) and excised via XbaI and Eco72I and cloned into PLC1-HA in pCIneo. pSX-LckY505F, encoding an activated form of Lck, was a gift of R. Perlmutter (Merck Co. Inc, Rahway, N.J.). pSX-Zap-70Y492F, encoding a partially active form of Zap-70, was described previously (15, 37). pR1k-GFP, encoding an R1k fusion protein with green fluorescent protein in the pCI vector, was a gift from P. Schwartzberg (National Institute for Human Genome Research, Bethesda, Md.). The pSRa-Flag-PTEN (WT-PTEN) and pSRa-Flag-PTEN C124S (PTEN C/S) constructs, encoding the wild-type protein and a catalytically inactive mutant, respectively, were described previously (29). All Jurkat lines were maintained in RPMI 1640 (Life Technologies, Gaithersburg, Md.) containing 10% fetal bovine serum. The Zap-70-negative cell line P116 and a stable Zap-70-reconstituted line were from R. T. Abraham (Duke University, Durham, N.C.). The Slp-76-deficient Jurkat clone J14, the stable Slp-76-reconstituted clone J14-76-11, and the anti-clonotypic TCR antibody C305, were gifts from A. Weiss (University of California, San Francisco, Calif.). Human embryonic kidney HEK 293T cells were maintained in Dulbecco's modified Eagle's medium (Life Technologies) containing 10% fetal bovine serum. The anti-influenza virus hemagglutinin (HA) antibody 12CA5 was a gift from A. Weissman (National Cancer Institute, Bethesda, Md.). The anti-HA antibody 3F10 was from Boehringer Mannheim (Indianapolis, Ind.). The antiphosphotyrosine antibody 4G10 and the anti-LAT rabbit polyclonal antiserum were from Upstate Biotechnology (Lake Placid, N.Y.). The rabbit antiphospholipase C1 [pY⁷⁸³] phospho-specific antibody was from Biosource (Camarillo, Calif.). The antibody to phospho-Akt (Ser 473) was from Cell Signaling Technology (Beverly, Mass.), and the anti-Flag antibody (M2) was from Sigma (St. Louis, Mo.).

Transfection assays. Jurkat E6.1 cells, Jurkat TAg cells, and all Jurkat derivatives were grown to log phase and transfected by using a square-wave electroporator. HEK 293T cells were grown to subconfluence, serum starved overnight, and transfected with the indicated constructs (0.3 µg of each constructs when not otherwise indicated) by using LipofectAMINE (Life Technologies).

Cell activation, immunoprecipitation, and immunoblotting. Transfected Jurkat cells were cultured at 0.5 × 10⁶ /ml for 24 h and treated with 0.1 mg of DNase per ml, and nonviable cells were removed by Ficoll gradient centrifugation immediately prior to experimental use. For stimulation, 10⁷ Jurkat cells were treated with medium or C305 antibody for 2 min at 37°C. The cells were lysed in a buffer composed of 60 mM Tris-HCl (pH 7.8), 150 mM NaCl, 5 mM EDTA, 10% glycerol, and 60 mM n-octyl--D-glucopyranoside (Sigma) as the detergent, together with protease and phosphatase inhibitors. HEK 293T cells were harvested 48 h after transfection in lysis buffer, and the protein concentration was determined. A 3-mg portion of protein per sample was immunoprecipitated. Cleared lysates were precipitated with specific antibodies prebound to protein A or protein A/G beads. Proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblot analysis with ¹²⁵I-protein A or chemiluminescence for detection. Radioimmunoblots were scanned on a PhosphorImager (Molecular Dynamics, Sunnyvale, Calif.) and analyzed with no manipulation except for adjustment of the exposure range.

Metabolic labeling of Jurkat cells. Transiently transfected Jurkat cells (8.5 × 10⁶) were transferred to 1 ml of RPMI containing 5 mM sodium pyruvate, 5% saline-dialyzed fetal calf serum, and 0.25 mCi of [³H]myristate or [³H]palmitate (Amersham Pharmacia Biotech, Inc., Piscataway, N.J.). After 3 h at 37°C, the cells were placed in lysis buffer. Anti-HA precipitates were resolved by SDS-PAGE (10% polyacrylamide) (InVitrogen, Carlsbad, Calif) and then treated with En³Hance (Dupont NEN, Boston, Mass.) for 1 h followed by 5% polyethylene glycol 8000 for 30 min. Dried gels were exposed to film for 4 weeks.

Subcellular fractionation and isolation of lipid rafts. Cell fractionation was performed as previously described (5). To isolate lipid rafts, transiently transfected cells (22 × 10⁶) were suspended in lysis buffer without detergent or glycerol and sonicated five times with 5-s pulses. Samples were adjusted to a total volume of 1 ml in lysis buffer containing 1% Brij 58 (Pierce, Rockford, Ill.) and maintained on ice for 1 h. The lysates were then mixed with 1 ml of 85% sucrose in 60 mM Tris HCl (pH 7.8) containing 150 mM NaCl and 5 mM EDTA and transferred to polyallomer centrifuge tubes (Beckman, Palo Alto, Calif.). A 2-ml volume of 30% sucrose was layered on top, followed by 1 ml of 5% sucrose. Samples were centrifuged in a Beckman SW55Ti rotor at 200,000 × g for 16 to 18 hr at 4°C, and 400-µl fractions were recovered from the top and immunoprecipitated and/or subjected to SDS-PAGE and immunoblot analysis.

Immunofluorescence confocal microscopy. Cells were stained with tetramethylchodamine-5-isothiocyanate (TRITC) (rod)-conjugated cholera toxin B subunit (CT-B; List, Campbell, Calif.), aggregated with an anti-CT-B antibody fixed in 2.0% formaldehyde in phosphate-buffered saline (PBS), and permeabilized by immersion in PBS-0.1% Triton X-100. The cells were stained with anti-HA in PBS containing 4 mg of normal goat serum per ml and 4 mg of human immunoglobulins per ml, washed, reacted with fluorescein isothiocyanate (FITC)-conjugated secondary antibody (Molecular Probes), washed, and mounted in glycerol. Internal controls, performed by omitting the primary antibody, show no detectable FITC staining (data not shown). Confocal analysis was carried out with a TCS 4D confocal setup (Leica) mounted on a Leitz DMRB microscope, equipped with a 100×/1.3 NA oil immersion objective. High-resolution fluorescence images were obtained by exciting FITC and rhodamine at 488 and 552 nm, respectively, with an argon ion laser. Images were acquired with an averaging function line-by-line, top down, with a scanning mode format of 512 × 512 pixels. Single optical sections of FITC signal, merged with the corresponding rhodamine images, were elaborated by a two-dimensional image-processing system.

NF-AT luciferase assay. Jurkat cells were transfected with 3 or 5 µg of WT-PLC1-HA or Palm-PLC1-HA along with 10 µg of an NF-AT-Luc plasmid containing a luciferase reporter under the control of the IL-2 minimal promoter and three optimized NF-AT sites (a gift of G. Crabtree, Stanford University, Stanford, Calif.) and 10 µg of pAD (Clontech, Palo Alto, Calif.), a control vector for transfection efficiency in which the expression of -galactosidase is regulated by the adenovirus major late promoter. At 3 h after transfection, the cells were incubated with medium alone, 10 nM phorbol myristate acetate (PMA), or 10 nM PMA plus 1 µM ionomycin for additional 6 h. The cells were disrupted in lysis buffer (Promega) and assayed using luciferin (Promega) and for -galactosidase activity using Galacton Plus and Emerald Enhance (Tropix, Bedford, Mass.).

IL-2 enzyme-linked immunosorbent assay. Jurkat cells were transfected with 20 µg of WT-PLC1-HA or Palm-PLC1-HA along with 20 µg of WT-PTEN or PTEN C/S. At 3 h after transfection, duplicate samples were incubated at 10⁶ cells/ml with medium alone, 10 nM PMA, or 10 nM PMA plus 1 µM ionomycin for an additional 24 h. Supernatants were harvested and assayed for IL-2 by using a QuantiGlo IL-2 chemiluminescence immunoassay kit (R&D System, Minneapolis, Minn.) as specified by the manufacturer.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Acylation constitutively targets PLC1 to lipid rafts. Numerous proteins are posttranslationally modified by the covalent addition of lipid moieties. Acylation by the covalent addition of myristate to an N-terminal glycine via an amide linkage and/or that of palmitic acid to cysteine residues confers specific targeting properties to the cytoplasmic leaflet of membrane microdomains (32, 36). The acylation sequence of the Src family member Fyn, a dually acylated protein, is sufficient for targeting Fyn to lipid rafts (41). To force the compartmentalization of PLC1 to the lipid rafts, we engineered a dually acylated form of an influenza virus HA-tagged PLC1 with an N-terminal myristoylation and palmitoylation motif from the human Fyn sequence (Palm-PLC1-HA) (Fig. 1A). In addition, an unmodified PLC1-HA (wild type, WT-PLC1-HA) and a control protein in which the glycine and cysteine acyl acceptors of the acylation motif were mutated (GCAS) were constructed as controls.

View larger version (32K): [in this window] [in a new window]   FIG. 1.   Characterization and subcellular fractionation of wild-type and acylated PLC1 in Jurkat cells. (A) Schematics of the constructs encoding WT-PLC1-HA and Palm-PLC1-HA. (B) Metabolic labeling of WT-PLC1-HA and Palm-PLC1-HA. Jurkat TAg cells were transiently transfected with the indicated constructs, labeled with [3H]myristate (top panel) or [3H]palmitate (middle panel), and lysed. Anti-HA precipitates were resolved by SDS-PAGE. A parallel gel loaded with whole-cell lysates (WCL) was immunoblotted (WB) with anti-HA antibody to control for PLC1-HA expression (bottom panel). (C) Subcellular fractionation of WT-PLC1-HA and Palm-PLC1-HA. Jurkat TAg cells were transiently transfected with the indicted constructs and fractionated into cytosol (C) and membrane (M) fractions. A 25-µg amount of protein was resolved by SDS-PAGE and immunoblotted with anti-HA.

View larger version (48K): [in this window] [in a new window]   FIG. 2.   Lipid raft compartmentalization of wild-type and acylated PLC1 in Jurkat cells. Lysates from transiently transfected Jurkat TAg cells were separated on discontinuous sucrose gradients. Fractions were analyzed by SDS-PAGE and immunoblotted (WB) with anti-HA (top and middle panels) or for endogenous Lck (Santa Cruz) (bottom panel).

View larger version (21K): [in this window] [in a new window]   FIG. 3.   Confocal analysis of the subcellular distribution of wild-type and acylated PLC1 in Jurkat cells. Transiently transfected Jurkat TAg cells were activated with an anti-TCR (C305) or left unstimulated as indicated. Cells were stained with CT-B and aggregated with anti-CT-B antibody to visualize lipid rafts (left panels, red). Transfected PLC1 was stained with anti-HA (middle panels, green). Areas of colocalization are shown in yellow in overlay images (right panels).

Raft-targeted PLC1 is constitutively tyrosine phosphorylated and induces Ca²⁺-dependent transcriptional activation. The ability to constitutively localize Palm-PLC1-HA to the rafts allowed us to test whether raft recruitment is sufficient for PLC1 activation and the transmission of signals downstream of PLC1. Since phosphorylation of tyrosine residues controls PLC1 enzymatic activation and tyrosine-phosphorylated PLC1 is abundant in lipid rafts (42, 48), we first questioned whether forcing PLC1 to the rafts affected its phosphorylation status. To test this possibility, WT-PLC1-HA and Palm-PLC-HA were transfected in Jurkat T cells and immunoprecipitated from lysates with anti-HA and probed by immunoblotting with antiphosphotyrosine antibody. While WT-PLC1-HA was tyrosine phosphorylated only on TCR triggering, Palm-PLC1-HA was constitutively phosphorylated to an extent similar to that observed for TCR-stimulated WT-PLC1-HA (Fig. 4A). This constitutive phosphorylation could be augmented by TCR engagement only marginally (Fig. 4B). Tyr⁷⁸³, a residue critical for PLC1 activation (14), was among the tyrosine residues constitutively phosphorylated in Palm-PLC1-HA (Fig. 4C). The relationship between subcellular localization and the phosphorylation status of wild-type and acylated PLC1 in resting cells is shown in Fig. 4D. Fractions from sucrose gradient floatation corresponding to detergent-soluble and detergent-resistant portions were pooled and probed by immunoblotting with anti-HA and antiphosphotyrosine. WT-PLC1-HA was exclusively present in the soluble fraction and was not phosphorylated, while Palm-PLC1-HA was mostly compartmentalized to the lipid rafts and constitutively phosphorylated. The small amount of Palm-PLC1-HA detected in the soluble fraction is most probably due to contamination during sucrose gradient flotation of the raft fractions.

View larger version (34K): [in this window] [in a new window]   FIG. 4.   Raft-targeted PLC1 is constitutively tyrosine phosphorylated and induces NF-AT activation in Jurkat cells. (A) Transfected Jurkat E6.1 cells were activated with an anti-TCR antibody (C305) as indicated. Lysates were immunoprecipitated (IP) with anti-HA, resolved by SDS-PAGE, immunoblotted (WB) with an antiphosphotyrosine 4G10 (anti-pY, upper panel), stripped, and reprobed with anti-HA (lower panel). (B) Jurkat E6.1 cells were transiently transfected with the control GCAS PLC1-HA or Palm-PLC1-HA and stimulated with increasing amounts of anti-TCR antibody (C305). Lysates were immunoprecipitated with anti-HA, resolved by SDS-PAGE, immunoblotted with 4G10 (anti-pY, upper panel), stripped, and reprobed with anti-HA (lower panel). (C) Lysates from transfected E6.1 Jurkat cells were immunoprecipitated with anti-HA, resolved by SDS-PAGE, and immunoblotted with 4G10 (anti-pY, upper panel) or a phospho-specific anti-PLC1[pY783] (lower panel). (D) The detergent-soluble (S) and raft (R) fractions were separated by discontinuous sucrose gradient, pooled, immunoprecipitated with anti-HA, resolved by SDS-PAGE, immunoblotted with 4G10 (anti-pY, upper panel), stripped, and reprobed with anti-HA (lower panel). (E) Jurkat cells were transiently transfected with the indicated constructs along with an NF-AT luciferase reporter and an adenovirus major late promoter -galactosidase-encoding vector to control for transfection efficiency. Transfected cells were treated with medium alone, a phorbol ester (PMA), or PMA plus ionomycin. Data were normalized for -galactosidase activity and expressed as the percentage of maximum activation with PMA plus ionomycin. The data shown are the means and standard errors of the mean of four separate experiments.

Tyrosine-phosphorylation of raft-targeted PLC1 requires Lck. Since Lck is necessary for Ca²⁺ mobilization (34), we investigated the role of Src kinases in the phosphorylation of raft-targeted PLC1. Treatment with the Src kinase inhibitor 4 - amino - 5 - (4 - chlorophenyl) - 7 - (t - butyl)pyrazolo - [3,4 - d]pyrimidine (PP2) (9) abolished Palm-PLC1-HA tyrosine phosphorylation in both unstimulated and TCR-stimulated Jurkat cells (Fig. 5A). Furthermore, transient expression of Palm-PLC1-HA in JCaM1.6, a Jurkat line deficient in functional Lck (34), did not result in constitutive tyrosine phosphorylation, and phosphorylation was not detectable following TCR engagement (Fig. 5B). Coexpression of Lck in JCaM1.6 cells led to the reconstitution of Palm-PLC1-HA phosphorylation. Therefore, Lck is required for the phosphorylation of raft-targeted PLC1.

View larger version (41K): [in this window] [in a new window]   FIG. 5.   Tyrosine phosphorylation of raft-targeted PLC1 in Jurkat cells requires Lck. (A) Transiently transfected Jurkat E6.1 cells were cultured in medium containing vehicle alone (0.2% dimethyl sulfoxide [DMSO]) or 20 µM PP2. The cells were activated with anti-TCR antibody (C305) as indicated. Lysates were immunoprecipitated (IP) with anti-HA, resolved by SDS-PAGE, immunoblotted (WB) with 4G10 (anti-pY, upper panel), stripped and reprobed with anti-HA (lower panel). (B) The Lck-defective Jurkat cell line JCam1.6 was transfected with Palm-PLC1-HA and cotransfected with an empty vector or pSX-LckY505F. Lysates were immunoprecipitated with anti-HA, resolved by SDS-PAGE, immunoblotted with 4G10 (anti-pY, upper panel), stripped, and reprobed with anti-HA (lower panel).

Tyrosine-phosphorylation of raft-targeted PLC1 does not require Zap-70 or interaction with the adapters Lat and Slp-76. Since Lck phosphorylates and activates Zap-70 (3, 13), whose expression is required for PLC1 activation (40), we questioned whether Zap-70 played a role in the phosphorylation of raft-targeted PLC1. WT-PLC1-HA or Palm-PLC1-HA was expressed in the Zap-70-deficient Jurkat cell somatic mutant P116 (40). P116 cells display no detectable tyrosine phosphorylation of WT-PLC1-HA in either unstimulated or stimulated cells, while the phosphorylation defect was fully reversed in P116 cells stably transfected with Zap-70 (Fig. 6A). Palm-PLC1-HA, however, was constitutively tyrosine phosphorylated in the Zap-70-deficient P116 cells, demonstrating that phosphorylation of raft-targeted PLC1 is independent of Zap-70.

View larger version (32K): [in this window] [in a new window]   FIG. 6.   Tyrosine phosphorylation of raft-targeted PLC1 in Jurkat cells is independent of Zap-70, Lat, and Slp-76. (A) The Zap-70-negative Jurkat line P116 and a stable Zap-70-reconstituted line were transiently transfected and activated as indicated. Lysates were immunoprecipitated (IP) with anti-HA, resolved by SDS-PAGE, immunoblotted (WB) with 4G10 (anti-pY, upper panel), stripped, and reprobed with anti-HA (lower panel). (B) Jurkat E6.1 cells were transfected and activated as indicated. Lysates were immunoprecipitated with anti-HA, resolved by SDS-PAGE, and separately immunoblotted with 4G10 (anti-pY) (top panel), with anti-Lat (middle panel), or with anti-HA (bottom panel). Calf intestinal phosphatase was used to remove phosphate groups from the anti-HA immunoprecipitates probed with anti-Lat to enhance blotting. (C) The Slp-76-deficient Jurkat clone J14 and a stable Slp-76-reconstituted clone (J14-76-11) were transiently transfected and activated as indicated. Lysates were immunoprecipitated with anti-HA, resolved by SDS-PAGE, immunoblotted with 4G10 (anti-pY) (upper panel), stripped, and reprobed with anti-HA (lower panel).

View larger version (34K): [in this window] [in a new window]   FIG. 7.   NF-AT activation by raft-targeted PLC1 does not require Zap-70 or Slp-76. The Zap-70-negative Jurkat line P116, a Zap-70-reconstituted P116 cell-derived line, the Slp-76-deficient Jurkat clone J14, and a Slp-76-reconstituted J14 cell clone (J14-76-11) were transiently transfected with the indicated PLC1 constructs along with a NF-AT luciferase reporter and an adenovirus major late promoter -galactosidase-encoding vector to control for transfection efficiency. Transfected cells were treated with medium alone, PMA, or PMA plus ionomycin. Data were normalized for -galactosidase activity and expressed as percentage of maximum activation with PMA plus ionomycin. The data shown are the means and standard errors of the mean of three separate experiments.

Reconstitution of PTEN in Jurkat T cells does not affect the constitutive tyrosine phosphorylation of Palm-PLC1-HA or decrease NF-AT induction or IL-2 production in cells expressing raft-targeted PLC1. Membrane localization of certain members of the Tec kinase family is mediated by the high-affinity interaction of the pleckstrin homology domain of these kinases with specific D3 phosphoinositides, principally phosphatidylinositol (3, 4, 5)-trisphosphate (PIP₃) (17, 27, 28). Localization of Tec kinases at the plasma membrane promotes their activation (16). PIP₃ levels are regulated by the D3 phosphoinositide phosphatase PTEN, whose expression is deficient in Jurkat T cells (29). The resulting high basal levels of D3 phosphoinositides in these cells lead to the constitutive membrane localization of the Tec kinase Itk and contribute to the hyperresponsiveness of Jurkat cells to exogenous stimuli (29, 30). To investigate whether the lack of PTEN plays a role in the phosphorylation of raft-targeted PLC1 in Jurkat cells, we have transiently reconstituted these cells with WT-PTEN or an inactive form of PTEN (PTEN C/S) along with WT-PLC1-HA or Palm-PLC1-HA (Fig. 8). Nearly equivalent levels of PTEN expression were observed in cells transfected with the PTEN constructs. Furthermore, reconstitution of PTEN activity by the wild-type protein was confirmed by the reduced levels of phosphorylation of the ubiquitous serine/threonine kinase Akt (protein kinase B) on Ser⁴⁷³, a residue whose phosphorylation is critically dependent on the presence of D3 phosphoinositides. The levels of tyrosine phosphorylation of WT-PLC1-HA were greatly reduced on TCR stimulation, confirming the critical role played by D3-phosphoinositides in the activation of cytoplasmic PLC1, possibly by regulating its interaction with the plasma membrane. Interestingly, PTEN reconstitution did not affect the tyrosine phosphorylation of Palm-PLC1-HA. Therefore, PTEN expression levels do not contribute to the constitutive phosphorylation of raft-targeted PLC1.

View larger version (56K): [in this window] [in a new window]   FIG. 8.   PTEN expression does not affect tyrosine phosphorylation of raft-targeted PLC1 in Jurkat T cells. Jurkat TAg cells were transiently transfected with 20 µg of empty vector, Flag-PTEN C/S, or Flag-WT-PTEN along with 20 µg of WT-PLC1-HA or Palm-PLC1-HA. The cells were treated as indicated. Cell lysates were immunoprecipitated (IP) with anti-HA, resolved by SDS-PAGE, and immunoblotted (WB) with 4G10 (anti-pY) (top panel). Whole-cell lysates (WCL) were immunoblotted with anti-HA (second panel from the top), an anti-phospho-Akt (pAKT, Ser473) (second panel from the bottom), or anti-Flag (bottom panel).

View larger version (37K): [in this window] [in a new window]   FIG. 9.   PTEN reconstitution does not affect NF-AT activation or IL-2 secretion in Jurkat T cells expressing raft-targeted PLC1. (A). Jurkat TAg cells were transiently transfected with 20 µg of Flag-PTEN C/S or Flag-WT-PTEN along with 5 µg of WT-PLC1-HA or Palm-PLC1-HA, a NF-AT luciferase reporter, and an adenovirus major late promoter -galactosidase-encoding vector as described in Materials and Methods. Transfected cells were treated with medium alone, PMA, or PMA plus ionomycin. Parallel samples were processed for PLC1-HA expression by immunoblot analysis. Data were normalized for PLC1-HA expression levels and presented as percentage of maximum activation with PMA plus ionomycin. The data shown are the means and standard error of the mean of four separate experiments. (B) Jurkat TAg cells were transiently transfected with Flag-PTEN C/S or Flag-WT-PTEN along with WT-PLC1-HA or Palm-PLC1-HA. Transfected cells were treated with medium alone, PMA, or PMA plus ionomycin. Parallel samples were processed for PLC1-HA expression by immunoblot analysis. Data were normalized for PLC1-HA expression and are presented as a percentage of maximum activation with PMA plus ionomycin. IL-2 secretion by cells treated with PMA plus ionomycin ranged from 2,400 to 3,900 U per 5 × 104 cells. ELISA, enzyme-linked immunosorbent assay.

Raft-targeted PLC1 is preferentially phosphorylated by Rlk or Lck in HEK 293T cells. In addition to Lck and Zap-70, the Tec family kinases Itk and Rlk are involved in PLC1 activation. T lymphocytes from Itk^//Rlk^/ mice display a decrease in early PLC1 phosphorylation together with impaired Ca²⁺ mobilization, indicating that Tec kinases regulate TCR-induced PLC1 activation (26). To investigate the ability of these kinases to phosphorylate wild-type or raft-targeted PLC1, we transfected HEK 293T cells with WT-PLC1-HA or Palm-PLC1-HA along with Itk, Rlk, or active forms of Lck or Zap-70. HEK 293T cells do not express any of these proteins, allowing reconstitution of their activity in the absence of interference by endogenous kinases.

View larger version (35K): [in this window] [in a new window]   FIG. 10.   Preferential phosphorylation of raft-targeted PLC1 by Rlk and Lck in HEK 293T cells. HEK 293T cells were transiently transfected with 1 µg of PLC1-HA and 0.3 µg of the indicated kinase plasmids. Lysates were immunoprecipitated (IP) with anti-HA, resolved by SDS-PAGE, immunoblotted (WB) with 4G10 (anti-pY) (upper panel), stripped, and reprobed with anti-HA (lower panel).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Our data show that compartmentalization of PLC1 to the lipid rafts is sufficient to trigger constitutive tyrosine phosphorylation and activation of this molecule. Phosphorylation of raft-targeted PLC1 required Lck but was independent of Zap-70, Slp-76, or the interaction with Lat.

Effectors and regulatory molecules appear to be segregated in resting cells based in part on their "solubility" in lipid rafts. Receptor-induced recompartmentalization to these membrane microdomains may promote their interaction by increasing their local concentration or via the formation of localized scaffolds with molecules constitutively present in this compartment. In the experiments reported here, we focused on TCR-induced translocation of PLC1 to the lipid rafts as a regulatory mechanism controlling its phosphorylation and activation.

Under physiologic conditions of TCR signaling, tyrosine-phosphorylated Lat serves as a raft-associated scaffold that recruits PLC1 via direct binding to its amino-terminal SH2 domain (33, 40, 47). We therefore anticipated that a raft-targeted PLC1 could bypass the need for Lat interaction. Accordingly, neither Lat nor Zap-70, the kinase responsible for Lat phosphorylation (47), was required to phosphorylate raft-targeted PLC1. Phosphorylated Lat recruits Slp-76 to the lipid rafts via the intercession of another adapter, Gads (21, 49). It has been proposed that Slp-76 nucleates a molecular complex with PLC1 and Itk (35). While an association between Slp-76 and Itk has been observed in vivo (35), the interaction between PLC1 and phosphorylated Slp-76 has been inferred from SH2 fusion protein studies (10) and appears to be redundant, given the direct interaction between Lat and PLC1. Recently, however, a raft-targeted Slp-76 has been shown to substitute for Lat in PLC1 activation (1), suggesting that a direct Slp-76/PLC1 interaction can take place. Cells deficient in Slp-76 recruit PLC1 to Lat, but no PLC1 tyrosine phosphorylation and activation is observed in these cells. This contrasts with our observation of no requirement for Slp-76 for tyrosine phosphorylation and activation of raft-targeted PLC1. Lat interaction, however, may not directly correlate with raft compartmentalization, since not all Lat is compartmentalized to the lipid rafts. Indeed, even phosphorylated Lat can be found outside of the raft compartment (48). A plausible explanation in light of our data is that Lat functions by initially recruiting PLC1 with Slp-76 stabilizes the compartmentalization of PLC1 to the lipid rafts. This is consistent with other data showing that inactivation of the Gads binding sites in Lat (and thereby the Slp-76 recruitment sites) blocks stable association of PLC1 with Lat and PLC1 activation (49). Furthermore, Gads-deficient thymocytes exhibit significantly reduced PLC1 phosphorylation and Gads-deficient T cells fail to flux Ca²⁺ on CD3 activation (45). Experiments are in progress to determine if Slp-76 and Lat cooperate in TCR-induced PLC1 compartmentalization to the lipid rafts.

Zhang et al. (49) have also observed TCR-induced transient Ca²⁺ mobilization in the absence of PLC1 binding to Lat, possibly by a PLC1-independent mechanism. PLC1 binding to Tyr¹³² of Lat, however, was still required for PLC1 phosphorylation, prolonged Ca²⁺ flux, and NF-AT activation (49). By stably targeting PLC1 to the rafts, we are more likely to recapitulate those events that lead to the long-term activation of PLC1. Whether prolonged PLC1 activation is due to phosphorylation of specific sites (possibly by specific kinases) or correlates with its residence time in the raft compartment remains to be established. In either case, our data suggest that a stable association of PLC1 with the lipid rafts results in its persistent phosphorylation and activation, with ensuing Ca²⁺ flux, NF-AT activation, and IL-2 secretion, by promoting approximation of PLC1 with critical kinase(s) or by affecting its dwelling time.

Phosphorylation of PLC1 in TCR signal transduction involves multiple kinases. While several pieces of evidence indicate that Lck, Zap-70, and the Tec kinases are all required, the role of individual players is less well defined. In particular, whether each kinase directly phosphorylates PLC1 has not been established. We observed no PLC1 phosphorylation when Palm-PLC1-HA was coexpressed with Zap-70 Y492F in HEK 293T cells. Together with the lack of a Zap-70 requirement in the phosphorylation of raft-targeted PLC1 in Jurkat cells, these data bring into question a direct intervention of Zap-70 in PLC1 phosphorylation. These data are consistent with the lack of TCR-induced PLC1 phosphorylation in cells defective for Slp-76, despite normal levels of Zap-70 activation, as indicated by the phosphorylation of endogenous Lat (43). A role for Slp-76 in facilitating Zap-70 interaction with PLC1 cannot be ruled out, albeit no stable interaction of this adapter with Zap-70 has been reported.

Lck and Rlk play a role in the phosphorylation of raft-associated PLC1, as shown by the Lck requirement for the phosphorylation of Palm-PLC1-HA in Jurkat cells and the preferential phosphorylation of raft-targeted PLC1 by either kinase in reconstituted HEK 293T cells. Both kinases are acylated and associate with lipid rafts or membranous vesicles (6, 27). We speculate that raft compartmentalization promotes the proximity of PLC1 to these kinases. Lck could function by both trans-activation of Rlk (27) and direct phosphorylation of raft-associated PLC1. This latter possibility is consistent with the phosphorylation of PLC1 in reconstituted HEK 293T cells and the observation that viral Src is capable of phosphorylating PLC1 (23). These data support the notion that Lck plays important signaling roles beyond the initial ITAM phosphorylation and the activation of Zap-70 (7, 44).

In contrast to Rlk, Itk failed to phosphorylate either wild-type or raft target PLC1 in reconstituted HEK 293T cells. While Rlk is constitutively acylated, Itk is recruited to the lipid rafts via PIP₃-dependent anchoring (27). The absence of significant phosphorylation of Palm-PLC1-HA in HEK 293T cells coexpressing Itk may therefore be due to the absence of PIP₃. In Jurkat cells, however, a PTEN defect maintains high basal levels of PIP₃, favoring the constitutive enrichment of Itk, which is normally present in the cytosol, in the lipid rafts (29). Reintroduction of PTEN phosphatase into Jurkat cells restores the normal Itk distribution pattern but does not affect the constitutive tyrosine phosphorylation of raft-targeted PLC1. Furthermore, PTEN reconstitution does not affect PMA-enhanced NF-AT induction or IL-2 production in cells expressing raft-targeted PLC1. These data suggest that Itk, even when constitutively associated with the plasma membrane, plays only a minor role in the phosphorylation and activation of raft-targeted PLC1 in Jurkat cells, although they do not exclude a role for Itk in the phosphorylation of PLC1 in response to TCR activation. Furthermore, these data, taken together with the observation that the decrease in PLC1 phosphorylation seen in Itk-deficient lymphocytes is only partial (18, 20, 26), support the notion that kinases other than Itk contributes to PLC1 phosphorylation in T lymphocytes. Our study supports a model whereby translocation to the lipid rafts is rate limiting for PLC1 activation by promoting its interaction with regulatory kinases present in these microdomains. Lck and Rlk are two potential candidates that may preferentially phosphorylate raft-compartmentalized PLC1, either directly by playing a hierarchical role or indirectly by activating other downstream kinases. Thus, lipid rafts play a central role in TCR signal transduction by segregating signaling molecules in resting cells and by providing a compartment for their association and activation upon receptor engagement.