Abstract
Stimulation of the CD3/TCR results within minutes in an increase in T cell adhesion mediated by β1 integrins. The biochemical pathways that control CD3-mediated increases in β1 integrin-mediated adhesion remain poorly characterized. In this study, the role of the tyrosine kinase ZAP-70 in the regulation of β1 integrin activity by the CD3/TCR was investigated. CD3 stimulation did not increase β1 integrin-mediated adhesion of the ZAP-70-deficient Jurkat T cell line, P116, to the β1 integrin ligand fibronectin. Reintroduction of wild-type ZAP-70, but not a kinase-inactive variant, K369R, corrected the adhesive defect observed in P116 T cells. In addition, the kinase-inactive ZAP-70 mutant inhibited CD3-induced adhesion of primary human T cell blasts. Interestingly, a ZAP-70 mutant with a tyrosine to phenylalanine substitution at position 319 (Y319F) restored the adhesive defect in P116 T cells, even though Y319F ZAP-70 failed to fully reconstitute CD3-initiated NF-AT-dependent transcription and tyrosine phosphorylation of the LAT adapter protein. Finally, expression of mutants of LAT and the SLP-76 adapter protein that modulate CD3-mediated activation of an NF-AT reporter gene failed to block CD3-induced increases in β1 integrin-mediated adhesion. These observations support a model in which the tyrosine kinase activity of ZAP-70 kinase is critical for regulation of β1 integrin activity by CD3/TCR. However, the signaling events downstream of ZAP-70 that regulate CD3/TCR-mediated activation of β1 integrin function exhibit key differences when compared with the signaling pathways that regulate transcriptional events initiated by CD3/TCR stimulation.
The integrin family of cell surface adhesion receptors is profoundly important to the dynamic functions of T lymphocytes (1, 2). These cells use specific integrin receptors to facilitate interactions with other cell types and with components of the extracellular matrix that are critical for Ag recognition and travel into, within, and between different anatomical compartments (3). Members of the β1 subfamily of integrin receptors, including α4β1, α5β1, and α6β1, mediate T cell interactions with extracellular matrix components such as fibronectin (FN)3 and laminin, as well as cell surface counterreceptors such as VCAM-1 (3). The functional activity of β1 (and other) integrins is dynamically regulated by the activation state of the T cell. Signaling through the Ag-specific CD3-TCR complex results within minutes in a rapid, transient increase in β1 and β2 integrin-mediated T cell adhesion that does not involve an increase in levels of integrin expression on the T cell surface (4, 5, 6). Similar changes in β1 integrin activity can be initiated by treatment of T cells with the phorbol ester PMA. This dynamic regulation of integrin-mediated adhesion by the CD3/TCR is one of the earliest manifestations of T cell activation that ultimately leads to T cell cytokine production, proliferation, and differentiation.
The biochemical events that occur subsequent to CD3/TCR ligation have been extensively analyzed with regard to transcriptional regulation of the IL-2 gene (7, 8). Activation of src family kinases results in tyrosine phosphorylation of immunoreceptor tyrosine-based activation motifs on CD3 subunits of the TCR complex, as well as tyrosine phosphorylation of the Syk family tyrosine kinase ZAP-70 (9, 10, 11, 12). Recruitment of ZAP-70 to the TCR complex via association with tyrosine-phosphorylated CD3 immunoreceptor tyrosine-based activation motifs (13, 14, 15) is a critical proximal event that links the enzymatically inert CD3-TCR complex to downstream biochemical signaling events. Activated ZAP-70 plays a critical role in the subsequent tyrosine phosphorylation of two adapter proteins, LAT (16) and SLP-76 (17). LAT is a lymphocyte-specific, transmembrane adapter protein that becomes heavily tyrosine phosphorylated in T cells after CD3 cross-linking (16) and is also targeted to glycolipid-enriched membrane regions (18, 19). The critical role of LAT in T cell signaling is illustrated by inhibitory effects of a phosphorylation-deficient LAT mutant on CD3-dependent transcription in Jurkat cells (16) and a severe SCID phenotype in mice lacking LAT (20). SLP-76 is also tyrosine phosphorylated in TCR-stimulated T cells, and has been implicated in transcriptional regulation of the IL-2 gene (17, 21). SLP-76-deficient mice have a phenotype similar to that of LAT-deficient mice, suggesting that these two adapter proteins play a critical role in coordinating downstream TCR signals essential for T cell development (22, 23).
Although the biochemical pathways that regulate CD3/TCR-mediated transcriptional activation of the IL-2 gene have been extensively mapped, much less is known regarding the biochemical events that regulate CD3/TCR-mediated activation of β1 integrins. Previous studies have implicated protein kinase C and phosphatidylinositol 3-kinase (PI 3-K) in activation-dependent regulation of β1 integrin-mediated adhesion (5, 24, 25, 26, 27). CD3/TCR stimulation leads to the association of tyrosine-phosphorylated LAT with the p85 subunit of PI 3-K (16), and SLP-76 has been implicated in CD3/TCR-dependent regulation of actin polymerization (28). However, no studies have directly assessed a role for ZAP-70 in TCR-mediated regulation of β1 integrin-mediated adhesion, a functional response of T cells that does not require the initiation of transcriptional events in the nucleus.
In this report, we demonstrate a critical role for the kinase activity of ZAP-70 in CD3/TCR-mediated increases in β1 integrin-mediated adhesion of Jurkat T cells and human peripheral T cells to FN. However, a ZAP-70 mutant that impairs CD3-mediated tyrosine phosphorylation of LAT and transcriptional activation had minimal effects on CD3-induced increases in β1 integrin function. Mutations in LAT and SLP-76 that impair CD3-mediated transcriptional activation also failed to block CD3-induced activation of β1 integrins, suggesting key differences in the biochemical signaling pathways downstream of ZAP-70 that regulate transcriptional events in the nucleus and β1 integrin function.
Materials and Methods
Cells
The Jurkat E6-1 cell line was obtained from the American Type Culture Collection (Manassas, VA). The ZAP-70-deficient Jurkat mutant, P116, was provided by R. Abraham and B. Irvin (Mayo Clinic, Rochester, MN) (29). Both Jurkat lines were maintained in RPMI 1640 supplemented with 10% FCS (Atlanta Biological, Norcross, GA), l-glutamine, and penicillin/streptomycin (complete medium).
T cell blasts were prepared by culturing PBMC for 3 days in RPMI 1640 complete medium with 20% FCS and a 1:100 dilution of PHA (Life Technologies, Grand Island, NY) as described previously (30).
Abs and other reagents
The CD3-specific mAb OKT3 and the β1 integrin-specific activating mAb TS2/16 were obtained from American Type Culture Collection. The inhibitory β1 integrin-specific mAb AIIB2 was obtained from the Developmental Studies Hybridoma Bank (Iowa City, IA). Goat anti-mouse IgG was purchased from ICN Biomedicals (Costa Mesa, CA). Goat anti-mouse IgG-PE conjugates were purchased from Southern Biotechnology Associates (Birmingham, AL). The phorbol ester PMA was purchased from LC Laboratories (Woburn, MA). Human FN was purchased from Life Technologies. The phosphotyrosine-specific mAb 4G10 and the anti-LAT antiserum were purchased from Upstate Biotechnology (Lake Placid, NY). The anti-hemagglutinin (HA) mAb 3F10 was purchased from Boehringer Mannheim (Indianapolis, IN). The anti-ZAP-70 antiserum was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The anti-FLAG M2 mAb was purchased from Sigma (St. Louis, MO). The anti-SLP-76 antiserum was kindly provided by G. Koretsky (University of Pennsylvania, Philadelphia, PA). The goat anti-mouse IgG HRP conjugate was purchased from Caltag Laboratories (Burlingame, CA). The donkey anti-rabbit IgG HRP conjugate was purchased from Amersham Life Science (Piscataway, NJ). The rabbit anti-sheep IgG HRP conjugate was purchased from Bio-Rad (Hercules, CA).
DNA constructs
The bicistronic enhanced green fluorescent protein (EGFP) expression vector pIRES2-EGFP was purchased from Clontech (Palo Alto, CA). The wild-type human ZAP-70 cDNA construct was kindly provided by A. Weiss (University of California, San Francisco, CA). The ZAP-70 cDNA sequence was PCR amplified and a hemagglutinin (HA) tag was added to the amino terminus using custom primers from Life Technologies. These primers also added a 5′ HindIII site and a 3′ BglII site. The HindIII site was blunt-ended using the Klenow fragment (Promega, Madison, WI). ZAP-70 was ligated into the Eco47III and BglII sites of the pIRES2-EGFP multiple cloning region. ZAP-70 mutations were generated with the QuickChange Mutagenesis kit (Stratagene, La Jolla, CA) using both sense and antisense oligonucleotides (Life Technologies) encoding for the substitution of tyrosine at position 319 with phenylalanine (Y319F) and the substitution of lysine at position 369 with arginine (K369R).
A wild-type human LAT cDNA was amplified from total Jurkat mRNA by RT-PCR. The 734-bp reading frame was simultaneously amplified and HA tagged using custom primers purchased from Life Technologies. A HindIII site was inserted at the 5′ end of the fragment, and a BglII site was inserted at the 3′ end. RT-PCR amplification was performed on a Perkin-Elmer DNA Thermal Cycler (Perkin-Elmer, Norwalk, CT) by 35 cycles of the following parameters: 94°C for 30 s, 55°C for 30 s, and 68°C for 4 min. The HindIII end was blunt-ended to ligate the LAT cDNA into pIRES2-EGFP as described for ZAP-70 subcloning above. The Y171F (TAC to TTC) and Y191F (TAT to TTT) mutations were performed by PCR site-directed mutagenesis as described above. All mutations were confirmed by sequencing at the University of Minnesota Microchemical Facilities (Minneapolis, MN).
The pEF-FLAG-SLP-76 and pEF-FLAG-Y3F SLP-76 constructs were provided by G. Koretzky (University of Pennsylvania) (31).
Transient transfections
Jurkat, P116, and PHA T cells were transfected as described previously (30, 32). All constructs in the pIRES2-EGFP vector were electroporated with 30 μg DNA/10 × 106 cells. SLP-76 constructs were coelectroporated with empty pIRES2-EGFP in a 2:1 ratio; 30 μg pEF-FLAG-SLP-76 and 15 μg pIRES2-EGFP were used per 10 × 106 cells. Electroporations were performed in Opti-MEM medium (Life Technologies) using a BTX square wave electroporator. Jurkat and P116 cells were electroporated at 240 V for 25 ms. T cell blasts were electroporated at 240 V for 40 ms. Transfected cells were then cultured for 12–18 h in complete medium. Viable cells were recovered via gradient density centrifugation with Ficoll-Hypaque (Life Technologies).
Intracellular staining
Jurkat cells were transfected as above and fixed in 4% paraformaldehyde. Cells were permeabilized with 0.5% saponin and stained with the anti-HA or anti-FLAG Ab followed by PE-conjugated secondary Ab in a 0.5% saponin solution as described previously (33). Cells were analyzed by flow cytometry on a Becton Dickinson FACScan (Mountain View, CA).
NF-AT-luciferase assays
Jurkat cells were cotransfected with the indicated ZAP-70, LAT, or SLP-76 expression vectors as well as a reporter construct containing the NF-AT-binding sequence upstream of the luciferase promoter (provided by D. Mueller, University of Minnesota, Minneapolis, MN). An aliquot of cells was analyzed for EGFP expression by flow cytometry to assess transfection efficiency. Cells were left untreated or stimulated with immobilized OKT3 or PMA (50 ng/ml) plus ionomycin (1.5 μM) for 5 h at 37°C. Cells were harvested and prepared for luciferase activity analysis per the manufacturer’s instructions (Promega). Luciferase activity was determined in a Berthold luminometer LB9501 in relative light units. Luciferase activity was standardized by transfection efficiency in each sample, and results from triplicate samples for each activation condition are shown as a percentage of the maximal luciferase activity achieved through stimulation of cells with PMA plus ionomycin.
Adhesion assays
Adhesion assays with untransfected Jurkat and P116 T cells were performed as previously described (32) using calcein-acetoxymethyl ester labeling of T cells. Adhesion assays were performed in flat-bottom 96-well plates (Costar, Cambridge, MA) precoated with 0.3 μg/well FN and blocked with PBS supplemented with 2.5% BSA. Cells were resuspended in PBS supplemented with 0.5% human serum albumin (PBS/HSA) at 1.6 × 106 cells/ml after Ficoll separation. For CD3 stimulation, cells were preincubated on ice for 30 min with the anti-CD3 mAb OKT3 and then washed twice with PBS/HSA. After washing, 80,000 cells/well were added to the appropriate wells. The following stimulating agents were added to the wells before the addition of cells: 10 ng/ml PMA, a 1:10 dilution of the activating β1 integrin-specific mAb TS2/16, or 1 μg/ml of goat anti-mouse IgG (for CD3 stimulation). Cells were incubated at 4°C for 1 h, plates were floated in a 37°C water bath for the indicated times, and nonadherent cells were removed by washing as described previously (32).
For transiently transfected cells, adhesion assays were performed as described previously (30). Briefly, cells were transfected and adhesion assays were performed as described above, except that the cells were not labeled with calcein-acetoxymethyl ester. An aliquot of each cell sample representing the same volume used in each well for the adhesion assay was prepared for flow cytometric analysis to verify the number of cells added to each well. Following washing to remove nonadherent cells, adherent cells were collected using PBS/0.1% EDTA and cells from six replicate wells were pooled into a tube containing 1 ml PBS/10% bovine calf serum. Cells were pelleted and resuspended in exactly 200 μl HBSS supplemented with 1% bovine calf serum and 0.2% sodium azide plus 50 μl PKH26 microbeads (Sigma) and 25 μl propidium iodide (Sigma) for a total sample volume of 275 μl. Each sample was analyzed on the flow cytometer, acquiring a total of 30,000 events.
Analysis of adherent transfected cells was performed as described previously (30). For each sample, forward scatter/side scatter profiles were used to identify the transfected cells, and the reference microbeads. FL2 events in the microbead gate were used to enumerate the total number of beads acquired in each sample. FL1 fluorescence reflects EGFP expression and PI-negative, FL1-positive events were gated into four subpopulations: EGFP-negative cells and three subpopulations with increasing levels of EGFP expression. The total volume of sample acquired was determined by the number of microbeads acquired divided by the bead density in each sample. The total number of T cells in each sample was then determined by the equation: [(T cells acquired)/(ml of sample acquired)](0.275 ml). Results are expressed as the mean percentage of adhesion of cells expressing no EGFP (labeled 0 on the x-axis) or increasing levels of EGFP (labeled 1, 2, and 3 on the x-axis) to FN from duplicate samples for each activation condition tested.
Immunoprecipitation and Western blotting
CD3 stimulation was performed as described above, except in L-15 medium (Life Technologies). After stimulation for 2 min at 37°C, cells were lysed by adding an equivalent volume of 2× lysis buffer (1% Triton X-100, 1% sodium deoxycholate, 158 mM NaCl, 10 mM Tris-HCl, 5 mM EDTA, 2 mM sodium vanadate, 20 μg/ml leupeptin, 20 μg/ml aprotinin, and 2 mM PMSF). Lysates were clarified by centrifugation at 12,000 rpm for 20 min at 4°C. Immunoprecipitations were performed as previously described (34) using goat anti-rat IgG-Sepharose beads precoated with 4 μg of the anti-HA mAb 3F10, goat anti-mouse IgG-Sepharose beads precoated with 4 μg of anti-FLAG M2, or protein A-Sepharose beads precoated with anti-LAT. All beads were purchased from Zymed (San Francisco, CA). Cell lysates (2–10 × 106 cells) were incubated with Ab-bead preparations at 4°C overnight. After washing with lysis buffer, proteins were removed from beads by boiling. Samples were separated on 10% SDS-polyacrylamide gels, and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA) for Western blotting.
Western blots were performed as previously described (34) using 1:1000 dilutions of primary Ab and appropriate dilutions of HRP-conjugated secondary Abs (1:50,000 for donkey anti-rabbit IgG, 1:20,000 for goat anti-mouse IgG, and 1:2000 for rabbit anti-sheep IgG). Membranes were developed by enhanced chemiluminescence (Pierce, Rockford, IL) according to the manufacturer’s instructions. Reprobing a membrane was performed as described previously (34).
Results
CD3/TCR signaling in the absence of ZAP-70 does not enhance β1 integrin-mediated T cell adhesion
We analyzed β1 integrin expression and function on P116 T cells, a ZAP-70-deficient variant of Jurkat T cells (29). P116 cells express CD3, and the α4β1 and α5β1 integrins at levels comparable to wild-type Jurkat T cells (Fig. 1,A). However, although CD3 stimulation of Jurkat T cells resulted in increased adhesion to FN, CD3 stimulation of P116 T cells did not enhance the adhesion of P116 T cells to FN when compared with unstimulated cells (Fig. 1,B). This was not due to global defects in β1 integrin function, since stimulation of P116 T cells with the phorbol ester PMA or the activating β1 integrin-specific mAb TS2/16 resulted in increased adhesion to FN, similar to what was observed after PMA stimulation of Jurkat T cells (Fig. 1,B and data not shown). In addition, both basal and PMA-induced adhesion of Jurkat cells and P116 cells to FN was mediated by β1 integrins, since the inhibitory β1 integrin-specific mAb AIIB2, but not the control HLA class I-specific mAb W6/32, potently blocked adhesion (Fig. 1,B). Similar to what was observed with Jurkat T cells (24), PMA-induced adhesion of P116 cells peaked at 10 min of stimulation (Fig. 1,C). CD3 stimulation did not enhance the adhesion of P116 cells at any time point tested (Fig. 1 C and data not shown).
CD3/TCR-mediated activation of β1 integrins in P116 T cells is restored by kinase-active ZAP-70
To verify that the inability of CD3 signaling to enhance β1 integrin-mediated adhesion of P116 T cells was in fact due to loss of ZAP-70 expression, we analyzed the adhesion of P116 T cells transiently expressing human ZAP-70 cDNA constructs tagged with an HA epitope tag. We used an EGFP bicistronic vector, which allowed us to assess the expression level of ZAP-70 in P116 T cells with EGFP fluorescence without the potential complications of tagging the ZAP-70 itself with EGFP. Two-color flow cytometry and intracytoplasmic staining with an anti-HA Ab and PE-conjugated secondary Ab was used to demonstrate a linear relationship between EGFP expression and expression of wild-type and kinase-inactive (K369R) ZAP-70 in transiently transfected P116 cells (Fig. 2,A). These results also illustrate that ZAP-70 can be expressed at various levels following transient transfection in P116 cells. Consistent with previous results (29, 35, 36), re-expression of wild-type, but not kinase-inactive, ZAP-70 restored CD3-induced transcriptional activation of an NF-AT-luciferase reporter gene in P116 T cells (Fig. 2,B). To assess the effects of ZAP-70 expression on tyrosine phosphorylation of the LAT adapter protein, LAT was immunoprecipitated from unstimulated and CD3-stimulated P116 cells and Western blots were probed with an anti-phosphotyrosine mAb. The results in Fig. 3 show that LAT was not tyrosine phosphorylated in unstimulated P116 cells and CD3 stimulation did not increase LAT phosphorylation. Re-expression of wild-type, but not kinase-inactive, ZAP-70 restored CD3-induced tyrosine phosphorylation of LAT in P116 cells (Fig. 3).
To determine whether ZAP-70 expression also restored CD3-induced adhesion of P116 T cells to FN, we analyzed the adhesion of transiently transfected P116 cells to FN in the absence of stimulation or following CD3 or PMA stimulation (Fig. 4). This assay quantitates the adhesion of P116 T cells to FN by collecting the FN-adherent cells and analyzing each cell population by flow cytometry as described previously (30). Approximately 20–50% of transiently transfected P116 cells expressed EGFP (data not shown). With each sample of cells, postacquisition gating was used to assess the adhesion of P116 cells expressing no EGFP (labeled 0 on the x-axis of the graphs in Fig. 4) or increasing levels of EGFP (labeled 1, 2, and 3 in Fig. 4). Thus, adhesion of cells expressing varying levels of the ZAP-70 construct can be assessed in each set of transfectants. The results show that EGFP expression by itself did not alter the inability of CD3 stimulation to enhance the adhesion of P116 T cells to FN. In contrast, CD3 stimulation enhanced the adhesion of P116 T cells expressing wild-type ZAP-70. This effect of wild-type ZAP-70 was dose dependent, since increasing expression of wild-type ZAP-70 resulted in increased adhesion following CD3 stimulation (Fig. 4). Furthermore, the ability of ZAP-70 to restore CD3-induced regulation of β1 integrin function in P116 T cells required kinase-active ZAP-70, since CD3 stimulation failed to enhance adhesion of P116 transfectants expressing kinase-inactive ZAP-70, even at high levels (Fig. 4). The effects of ZAP-70 re-expression on β1 integrin regulation in P116 T cells were specific to CD3/TCR signaling, since PMA stimulation was unaffected by the expression of either EGFP alone, wild-type ZAP-70, or kinase-inactive ZAP-70 (Fig. 4). These results demonstrate a role for the kinase activity of ZAP-70 in CD3-induced regulation of β1 integrin activity.
Kinase-inactive ZAP-70 inhibits CD3-induced increases in adhesion of human peripheral T cell blasts to FN
To analyze a role for ZAP-70 in CD3/TCR regulation of β1 integrin activity in normal T cells, we also examined the effects of the ZAP-70 constructs on the adhesion of PHA-stimulated T cell blasts to FN. Transient transfection of T cell blasts resulted in EGFP expression in 5–7% of the cells (top panels, Fig. 5). Using the adhesion assay employed with P116 T cells, analysis of the EGFP-negative cells in each set of transient transfectants demonstrated that stimulation of T cell blasts via CD3 cross-linking for 10 min resulted in increases in adhesion to FN comparable to that observed with PMA stimulation (Fig. 5). Expression of wild-type ZAP-70 expression did not affect either the basal adhesion of T cell blasts to FN or the increased adhesion induced by CD3 stimulation or PMA stimulation. In contrast, expression of the kinase-inactive K369R ZAP-70 mutant specifically inhibited CD3-induced increases in adhesion of T cell blasts to FN in a dose-dependent manner (Fig. 5). At high levels of expression of kinase-inactive ZAP-70, the adhesion of CD3-stimulated T cell blasts was similar to the adhesion of unstimulated cells. These results provide further evidence that ZAP-70 plays a critical role in CD3/TCR-mediated regulation of β1 integrin activity.
The Y319F ZAP-70 mutant restores CD3/TCR-mediated increases in P116 T cell adhesion to FN
We also explored the effects of mutating the tyrosine at position 319 in ZAP-70 to phenylalanine in CD3/TCR regulation of β1 integrin-mediated adhesion. Although the Y319F ZAP-70 mutant was expressed in P116 cells at levels comparable to wild-type ZAP-70 (Figs. 2,A and 3), the Y319F ZAP-70 mutant was not able to restore CD3-mediated activation of NF-AT luciferase (Fig. 2,B). Although expression of Y319F ZAP-70 in P116 cells resulted in some restoration of CD3-induced tyrosine phosphorylation of LAT, the level of LAT phosphorylation was significantly lower than that observed in CD3-stimulated P116 cells expressing wild-type ZAP-70 (Fig. 3). However, the Y319F ZAP-70 mutant was able to restore CD3-induced adhesion of P116 T cells to FN to levels similar to that observed with wild-type ZAP-70 (Fig. 4). Furthermore, the Y319F mutant had minimal inhibitory effects on CD3-mediated adhesion of human T cell blasts to FN, with only a slight inhibitory effect at the highest expression level of EGFP analyzed (Fig. 5). These results suggest that tyrosine 319 in ZAP-70 does not play a critical role in CD3/TCR regulation of β1 integrin activity.
Mutations in LAT that impair CD3-induced transcriptional activation of NF-AT luciferase do not impair CD3 regulation of β1 integrin-mediated adhesion
Our results with the Y319F ZAP-70 mutant suggest that CD3-mediated regulation of β1 integrin activity may be dependent on ZAP-70 but independent of LAT. To explore this issue further, we determined the effects of HA-tagged wild-type LAT and an HA-tagged LAT mutant, Y171/191F, on CD3-induced adhesion of human T cell blasts to FN. The Y171/191F LAT mutant has tyrosine to phenylalanine substitutions at two critical tyrosine residues in the LAT cytoplasmic domain (16). Consistent with previous studies (16), expression of Y171/191F LAT inhibited CD3-mediated transcriptional activation of NF-AT luciferase in Jurkat T cells (Fig. 6,A). In addition, CD3 stimulation resulted in tyrosine phosphorylation of wild-type LAT but not Y171/191F LAT (data not shown). However, CD3-induced increases in adhesion of Jurkat T cells to FN were unaffected by expression of either wild-type LAT or Y171/191F LAT, even at high levels (Fig. 6,B). Both wild-type LAT and Y171/191F LAT were also expressed at comparable levels in human T cell blasts (Fig. 7,A). Similar to what was observed with Jurkat T cells, expression of either wild-type LAT or Y171/191F LAT failed to alter CD3-induced adhesion of human T cells blasts to FN (Fig. 7 B). These results suggest differences in the role of LAT in CD3-dependent signaling pathways that regulate NF-AT-dependent transcription and β1 integrin activity.
Mutations in SLP-76 that impair enhancement of CD3-induced transcriptional activation of NF-AT luciferase by SLP-76 do not impair CD3 regulation of β1 integrin-mediated adhesion
Since ZAP-70 also plays a critical role in TCR-mediated tyrosine phosphorylation of the adapter protein SLP-76 (17), we analyzed the effect of SLP-76 overexpression on CD3-dependent regulation of adhesion of human T cell blasts to FN. We used FLAG epitope-tagged wild-type SLP-76 and a mutant SLP-76 with tyrosine to phenylalanine substitutions at positions 113, 128, and 145 (Y3F SLP-76). Although the Y3F SLP-76 construct was expressed in both Jurkat T cells and human T cell blasts, FLAG-tagged wild-type SLP-76 was difficult to detect in human T cell blasts derived from several different donors (Fig. 8,A). As previously reported (31), expression of a FLAG-tagged form of wild-type SLP-76 enhanced CD3-induced activation of NF-AT luciferase in Jurkat T cells. In contrast, expression of a mutant SLP-76 with tyrosine to phenylalanine substitutions at positions 113, 128, and 145 (Y3F SLP-76) failed to enhance NF-AT-dependent transcription (Fig. 8,B). To assess the effect of the Y3F SLP-76 mutant on T cell adhesion, human T cell blasts were cotransfected with an EGFP expression vector and FLAG-tagged Y3F SLP-76. Expression of Y3F SLP-76 did not affect CD3-induced increases in adhesion of human T cell blasts to FN (Fig. 8 C). Wild-type SLP-76 or Y3F SLP-76 also had no effect on CD3-induced increases in Jurkat T cell adhesion to FN (data not shown).
Discussion
One early consequence of stimulation through the Ag-specific CD3-TCR complex is a rapid and transient increase in β1 integrin-mediated adhesion to FN (5). The intracellular signals critical for CD3/TCR-mediated regulation of β1 integrins remain poorly understood. In this report, we have determined whether signals similar to those used for other manifestations of stimulation in T cells, such as NF-AT activation, are also involved in CD3/TCR-mediated increases in β1 integrin activity. Our studies suggest a critical role for the ZAP-70 tyrosine kinase in the regulation of β1 integrin activity by the CD3/TCR. However, CD3-induced activation of β1 integrins was unaffected by mutant forms of LAT and SLP-76 that inhibited CD3-mediated activation of NF-AT reporter genes. This suggests that distinct effector functions generated by CD3/TCR ligation use both common and distinct signaling intermediates.
The generation and analysis of P116 Jurkat cells, which lack endogenous ZAP-70 expression, made it possible to determine the role of ZAP-70 in CD3-mediated regulation of β1 integrin activity in a well-established model system. These studies showed that, unlike wild-type Jurkat T cells, ligation of the CD3/TCR on P116 T cells did not result in enhanced β1 integrin-mediated adhesion to FN. This defect in TCR signaling was not due to differences between Jurkat and P116 T cells in the expression of CD3/TCR or β1 integrins or in the kinetics of adhesion. Furthermore, the ability of both PMA stimulation and an activating β1 integrin-specific mAb to enhance β1 integrin-mediated adhesion of P116 T cells in a manner similar to Jurkat T cells suggests that β1 integrins expressed on P116 T cells are not functionally defective.
Several lines of evidence suggest a critical role for ZAP-70 in CD3/TCR regulation of β1 integrins. First, re-expression of wild-type ZAP-70 restored CD3-induced increases in β1 integrin activity in P116 T cells. Second, a kinase-inactive form of ZAP-70 (K369R ZAP-70) failed to restore CD3-induced β1 integrin activation in P116 T cells. This suggests a critical role for ZAP-70 kinase activity in this signaling pathway rather than a role for ZAP-70 as an adapter protein (37). Finally, the kinase-inactive form of ZAP-70 also inhibited CD3-mediated increases in β1 integrin activity in human peripheral blood T cell blasts. This result also argues for a central role for ZAP-70 in the regulation of β1 integrin function by the CD3/TCR in normal as well as transformed T cells.
Mutation of the tyrosine at position 319 in the interdomain B region of ZAP-70 to phenylalanine has previously been shown to affect CD3/TCR-initiated activation of a NF-AT reporter gene, calcium flux, Ras activation, and tyrosine phosphorylation of LAT, a downstream adapter protein critical to TCR signaling (35, 38, 39). However, we demonstrate here that Y319F ZAP-70 was as efficient as wild-type ZAP-70 in restoring CD3-mediated activation of β1 integrins. In addition, Y319F ZAP-70 did not inhibit CD3-induced β1 integrin activation in T cell blasts, except for some slight inhibition at the highest levels of ZAP-70 expression obtained in these transient transfection experiments. Similar to previous studies, we also observed that the Y319F ZAP-70 mutant did not restore CD3-mediated activation of a NF-AT luciferase reporter gene in P116 T cells. In addition, expression of Y319F ZAP-70 in P116 T cells resulted in minimal CD3-induced tyrosine phosphorylation of LAT when compared with P116 T cells expressing comparable levels of wild-type ZAP-70. Thus, these results suggest that 1) tyrosine 319 in ZAP-70 is not critical for ZAP-70-dependent regulation of β1 integrin-mediated adhesion and 2) CD3-induced increases in β1 integrin function may not involve ZAP-70-dependent activation of calcium mobilization and Ras activity.
Although the Y319F ZAP-70 mutant was previously reported to lack the ability to restore CD3-induced tyrosine phosphorylation of LAT in P116 T cells (35), we observed some detectable LAT tyrosine phosphorylation upon expression of Y319F ZAP-70. This discrepancy may relate to high levels of ZAP-70 expression that can be achieved in transient transfection experiments when compared with the stable transfectants reported in previous studies (35). Nevertheless, it remains possible that CD3-induced increases in β1 integrin activity in T cells may be dependent on a low threshold level of tyrosine phosphorylation of LAT that can be mediated by expression of Y319F ZAP-70. However, our studies with the Y171/191F LAT mutant argue against this possibility, since expression of this LAT mutant blocked TCR-initiated activation of NF-AT luciferase but did not inhibit CD3-induced activation of β1 integrins.
We also tested a potential role for the SLP-76 adapter protein in CD3-dependent activation of β1 integrins, since tyrosine phosphorylation of SLP-76 is also ZAP-70 dependent (17, 35). Overexpression of wild-type SLP-76 enhanced CD3-mediated activation of NF-AT luciferase, but had no effect on CD3-mediated activation of β1 integrins. Mutation of three critical tyrosine residues in SLP-76 (Y3F SLP-76) blocked the enhancement of NF-AT-dependent transcription by wild-type SLP-76 (31) and inhibited CD3-induced actin polymerization in human T cells (28). However, similar to the results obtained with Y171/191F LAT, expression of Y3F SLP-76 did not affect CD3-mediated regulation of β1 integrin functional activity. Although Y3F SLP-76 does not act in a dominant negative manner for NF-AT activation, Y3F SLP-76 has been reported to block CD3-induced actin polymerization in Jurkat T cells (28). Other studies clearly suggest that LAT and SLP-76 are linked biochemically via the Gads adaptor (40) and functionally, as suggested by the very similar phenotypes of LAT- and SLP-76-deficient mice (20, 22, 23). Thus, although LAT and SLP-76 clearly play a critical, and possibly cooperative, role in many T cell-signaling pathways, our studies currently do not provide any evidence for a role for either LAT or SLP-76 in the regulation of β1 integrin activity initiated by CD3/TCR stimulation.
Previous studies have demonstrated a critical role for the lipid kinase PI 3-K in the regulation of β1 integrin activity by several cell surface receptors expressed on human T cells, including the CD3/TCR (25, 30, 41, 42, 43). The biochemical signaling pathways that regulate activation of PI 3-K by the CD3/TCR remain unclear. CD3-mediated tyrosine phosphorylation of LAT results in the association of PI 3-K with LAT in a manner that is dependent on the tyrosines at positions 171 and 191 (16). However, our studies with the Y171/191F LAT mutant suggest that LAT-mediated recruitment of PI 3-K is not required for CD3-mediated activation of β1 integrins. The role of ZAP-70 in regulating PI 3-K activity is also unclear, although studies of the Syk tyrosine kinase in B cells suggest that B cell receptor-mediated activation of PI 3-K is dependent on Syk (44, 45). Thus, it is possible that ZAP-70 plays a similar role in T cells in regulating CD3/TCR-mediated activation of PI 3-K. ZAP-70 may also play a role in PI 3-K regulation via association of ZAP-70 with the adapter protein Cbl (36, 46, 47, 48), since Cbl associates with PI 3-K upon CD3/TCR stimulation and PI 3-K activity is elevated in Cbl-deficient mice (49). Although Cbl has been implicated in the regulation of β1 integrin activity by the CD28 receptor (50), the role of Cbl in CD3/TCR-mediated activation of β1 integrins remains unclear. Since the kinase activity of ZAP-70 is critical to CD3-dependent regulation of β1 integrin activity, ZAP-70-mediated tyrosine phosphorylation of substrates that play a role in facilitating cytoskeletal changes, such as α tubulin (51, 52), must also be considered.
In summary, these studies demonstrate a critical role for the tyrosine kinase activity of ZAP-70 in the regulation of β1 integrin activity by the CD3/TCR. However, a role for either LAT or SLP-76, two downstream substrates of ZAP-70, in CD3-mediated regulation of β1 integrins is not evident. These results demonstrate key differences between regulation of β1 integrin-mediated adhesion and other T cell effector responses with respect to these TCR proximal signaling events.
Acknowledgements
We thank Drs. R. Abraham, B. Irvin, G. Koretzky, A. Weiss, and D. Mueller for providing valuable cells and reagents, N. H. Lazarus for technical assistance, and Dr. M. L. Woods for helpful discussions.
Footnotes
This work was supported by National Institutes of Health Grants AI31126 and AI38474 (to Y.S.), a Leukemia Research Foundation Postdoctoral Fellowship Grant (to H.C.), and National Institutes of Health Training Grant T32 CA09138 (to R.L.). Y.S. is the Harry Kay Chair of Cancer Research at the University of Minnesota.
Abbreviations used in this paper: FN, fibronectin; EGFP, enhanced green fluorescent protein; HA, hemagglutinin; LAT, linker of activation of T cells; PI 3-K, phosphatidylinositol 3-kinase; SLP-76, Src homology 2 domain-containing leukocyte phosphoprotein of 76 kDa.