Antigen Receptor–Induced Activation and Cytoskeletal Rearrangement Are Impaired in Wiskott-Aldrich Syndrome Protein–Deficient Lymphocytes

Antibodies used for these studies included the following: FITC-conjugated Ly9.1, anti–Thy 1.2, anti-CD8, anti-Fas, anti-IgM, anti-CD4, anti-CD5, and anti-IgD; PE-conjugated anti-CD4, anti-CD3∈, anti-CD44, anti-CD43, anti-CD23, and anti–IL-2 and anti-B220; biotin-conjugated anti–TCR-α/β, anti-IgD, anti-CD19, and anti-CD25; allophycocyanin-conjugated anti–TCR-α/β, anti-CD3, anti-CD4, anti-CD8, anti-B220, anti-CD40, and anti-IgM antibodies and Cy5-conjugated streptavidin, all from PharMingen; and anti–extracellular signal regulatory kinase (ERK)1 and anti-ERK2 antibodies from Santa Cruz, anti–TCR-ζ and anti-ZAP70 antibodies supplied by Drs. Larry Samelson (National Institutes of Health, Bethesda, MD) and André Veillette (McGill Cancer Centre, Montreal, Quebec, Canada), respectively; and antiphospho-SAPK/JNK and anti-SAPK/JNK antibodies (reactive against all SAPK/JNK isoforms) from New England Biolabs. Texas red–conjugated streptavidin was purchased from GIBCO BRL. Monoclonal hamster anti–mouse CD3∈ was produced by the 145-2C11 hybridoma (provided by Dr. R. Miller, Ontario Cancer Institute) and purified from the supernatant by protein G chromatography; polyclonal rabbit anti-WASp antibody was derived by immunization with a polylysine-conjugated peptide corresponding to a putative nuclear localization signal within the WAS cDNA; anti–murine β-actin mAb, PMA, ionomycin, and murine recombinant IL-2 were purchased from Sigma Chemical Co.; rabbit anti–hamster and anti–mouse IgG antibodies and FITC-conjugated streptavidin were purchased from Jackson ImmunoResearch Laboratories. Murine recombinant IL-4 was obtained from PharMingen, [³H]thymidine from Dupont/NEN, 7-amino-actinomycin D (7-AAD) from Calbiochem, Indo-1 from Molecular Probes, myelin basic protein (MBP) from Upstate Biotechnology, Inc., and an IL-2 ELISA kit from Genzyme. Brefeldin A was a gift from Dr. R. Miller.

A WAS gene targeting vector was derived by subcloning a 3.3-kb segment of the WAS gene encompassing ∼2 kb of the 5′ flanking sequence upstream of the initiation codon through exon 4 of the gene into the Ssc8387I site of the pPNT expression cassette 46 and a 3.4-kb fragment encompassing the XbaI site in exon 11 through to a BamHI site within intron 11 into the pPNT XbaI-BamHI sites. Embryonic stem (ES) cells from the male-derived R1 ES cell line (129/Sv) were electroporated with this targeting vector and selected with neomycin and gancyclovir 47. Surviving clones were screened for homologous recombination at the WAS locus by PCR using the following primer pair: 5′-GTGAAGGATAACCCTCAGAAGTCC-3′ (forward primer, S1, derived from sequences within exon 2 of the WAS gene) and 5′-CGGAGCAGAATCTAGATGGCAGAGT-3′ (reverse primer, S2, representing sequences in the 3′ region downstream from exon 12 of the WAS gene) (see Fig. 1 A). Targeted clones were verified by Southern blotting with a probe derived from a 450-bp segment external to the 5′ region of homology (see Fig. 1 B). Two independently derived WAS^−/− clones were then either aggregated with CD1 eight cell stage embryos or microinjected into 3.5-d recombinase activating gene 2–deficient (RAG-2^−/−) blastocysts 48, and the aggregates or blastocysts were then implanted into pseudopregnant foster mothers.

WAS genotypes of the chimeric progeny were confirmed using a PCR assay including the forward primer (G1; 5′-ACTGAAGGCTCTTTACTATTGCT-3′), derived from a sequence within the neomycin resistance gene, and a reverse primer (G2; 5′-ACTGAAGCCTCTTTACTATTGCT-3′), corresponding to a sequence within exon 11 of the WAS gene. Chimeric male mice carrying the mutation in the germline were bred to the C57BL/6 background by backcrossing over six generations.

Cells were resuspended in immunofluorescence staining buffer (PBS containing 1% BSA and 0.05% NaN₃) and incubated with the appropriate fluorochrome-conjugated antibodies for 30 min at 4°C. For three- and four-color staining, Texas red– or Cy5-conjugated streptavidin was used after staining with biotin-conjugated antibodies. Stained cells were analyzed using a FACScan™ with CELLQuest™ software (Becton Dickinson). For detection of expression of the early activation marker CD69, thymocytes (2 × 10⁶ cells/ml) were cultured with medium alone, with plate-bound anti-CD3 antibody (10 μg/ml), or with plate-bound anti-CD3 plus anti-CD28 antibodies (10 and 0.2 μg/ml, respectively) for 24 h. Cells were stained with FITC-conjugated anti-CD8, PE-conjugated anti-CD4, and biotinylated anti-CD69 antibodies followed by Cy5-conjugated streptavidin. Expression of CD69 was measured on gated CD4⁺ CD8⁺ thymocytes.

For analysis of cell proliferation, single cell suspensions were prepared from thymi, lymph nodes, and spleen of age-matched WAS^−/− and WAS^+/+ (C57BL/6) mice bred at the Samuel Lunenfeld Research Institute, and the cells were then subjected to erythrocyte lysis in ammonium chloride buffer. Thymocytes and lymph node cells were cultured in 96-well flat-bottomed microtiter plates (2 × 10⁶ cells/ml) for 48 h in culture medium alone or in the presence of either plate-bound (0–25 μg/ml) or soluble (2 μg/ml) anti-CD3∈ antibody with or without anti-CD28 antibody (0.2 μg/ml) or with soluble anti-CD3 antibody plus PMA (1 μg/ml and 5 ng/ml, respectively), Con A (1 μg/ml) or PMA plus ionomycin (5 μg/ml and 250 ng/ml, respectively), or soluble anti-CD3 antibody plus IL-2 (1 μg/ml and 50 U/ml, respectively). Splenocytes (2 × 10⁶ cells/ml) were cultured for 48 h with medium alone, LPS (5 μg/ml), IL-4 (2 ng/ml), or IL-4 plus anti-IgM antibody F(ab′)₂ fragment (1.25 and 5 μg/ml), or anti-CD40 antibody (2 μg/ml). Cultured cells were pulsed with [³H]thymidine (1 μCi/well) for 16 h before terminating the incubation. Incorporated radioactivity was measured using an automated β scintillation counter. Alternatively, for biochemical assays, 4 × 10⁷ thymocytes or lymph node T cells were resuspended in 100 μl 2% FCS–containing RPMI and incubated for 30 min at 4°C in the presence or absence of biotin-conjugated anti-CD3 antibody (25 μg/ml). After removal of unbound antibody, the cells were resuspended at a concentration of 2 × 10⁸ cells/ml, warmed to 37°C for 1 min, and incubated for varying periods with 10 μg/ml streptavidin.

To evaluate IL-2 secretion, supernatants were collected from unstimulated and TCR-stimulated cells and the amount of IL-2 was quantified by ELISA (Genzyme). Numbers of IL-2–producing cells were assayed by incubating stimulated and unstimulated lymph node T cells (2 × 10⁶) in 24-well plates precoated with anti-CD3 antibody (5 μg/ml) with or without addition of soluble anti-CD28 antibody (4 μg/ml) and in the presence or absence of Brefeldin A (5 μM) at 37°C for 8 h. Cells were harvested, washed with immunofluorescence buffer, and labeled with an FITC-conjugated anti-CD4 antibody. Cells were then washed again and incubated for 15 min in 150 μl of 4% paraformaldehyde at room temperature. Cells were then washed, permeabilized, and labeled by 2-h incubation with a PE-conjugated anti–IL-2 antibody in 100 μl of 0.1% saponin–containing immunofluorescence buffer. Cells were washed twice with permeabilization buffer, resuspended in immunofluorescence buffer, and analyzed by flow cytometry.

Splenocytes and thymocytes (2 × 10⁶ cells) from WAS^+/+ and WAS^−/− mice were lysed in buffer containing 50 mM Tris, pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM PMSF, and 1 μg/ml each of leupeptin, pepstatin, and aprotinin as protease inhibitors. After 15 min incubation in cold lysis buffer, unlysed cells were removed by centrifugation and the lysates were then boiled in the presence of 6× reducing sample buffer, electrophoresed through 10% SDS-polyacrylamide, and transferred to nitrocellulose (Schleicher & Schuell). The filters were incubated at 4°C for at least 1 h in TBST solution (150 mM NaCl, 10 mM Tris-HCl, pH 7.4, 0.05% Tween 20) plus 3% gelatin. Filters were then incubated for 2 h at room temperature with anti-WASp antibody (1 μg/ml) in TBST, followed by the addition of goat anti–mouse antiserum labeled with peroxidase (Amersham Pharmacia Biotech) and horseradish peroxidase conjugate (Bio-Rad Laboratories). The blot was stripped and reprobed with an anti–β-actin antibody in order to assess loading. For immunoprecipitation, lysates prepared from unstimulated or TCR-stimulated thymocytes or lymph node T cells (4 × 10⁷) were precleared by incubating an equal amount (100–250 μg) lysate protein with protein A–Sepharose (Amersham Pharmacia Biotech) for 1 h at 4°C and for an additional 1 h with 40 μl rabbit preimmune sera. Lysates were then incubated for 3 h at 4°C with specific antibody or rabbit preimmune serum and 25 μl packed protein A–Sepharose beads, and the immune complexes were then collected by centrifugation, washed four times in lysis buffer, and resuspended in SDS sample buffer for immunoblotting analyses. Alternatively, ERK1/ERK2 immune complexes immunoprecipitated using a mixture of anti-ERK1 and anti-ERK2 antibodies were washed in MAPK buffer (5 mM Hepes, pH 7.4, 10 mM MgCl₂, 100 mM NaVO₄) and resuspended in 50 μl reaction buffer (30 mM Tris-HCl, pH 8.0, 20 mM MgCl₂, 2 mM MnCl₂ containing 5 μg MBP [Upstate Biotechnology], 1 mM cold ATP, and 10 μCi [γ-³²P]ATP [Dupont/NEN]). After 15 min incubation at 30°C, reactions were terminated by addition of 12 μl 6× SDS-PAGE loading buffer, and the samples were boiled, electrophoresed through 12% polyacrylamide gels, and transferred to nitrocellulose. The phosphorylated MBP bands were visualized by autoradiography, and ERK1 and ERK2 expression levels were determined by anti-ERK2 immunoblotting analysis.

Thymocytes and lymph node T cells (5 × 10⁶ cells/ml) were labeled with Indo-1 (5 μM) and incubated at 37°C in the dark for 30 min. Cells were washed, resuspended in RPMI containing 2% FBS and 10 mM Hepes, pH 7.4, and incubated with biotinylated anti-TCR antibody (5 μg/ml) for 15 min at 4°C. After washing, the cells were resuspended in RPMI buffer at a concentration of 10⁷ cells/ml and stimulated with 5 μg/ml of streptavidin. Calcium levels were detected by flow cytometric analysis of Indo-1 violet-blue fluorescence ratio 49.

Freshly isolated thymocytes were plated at a density of 2 × 10⁶ cells/ml in 96-well tissue culture plates precoated with either anti-CD3 antibody (20 μg/well), anti-CD3 plus anti-CD28 (each 20 μg/well), anti-Fas antibody (0.1, 1, or 5 μg/ml), or PMA plus ionomycin (10 and 500 ng/ml, respectively). After 24 h, the cells were harvested, washed with PBS, and stained with FITC-conjugated anti-CD4, PE-conjugated anti-CD8, and 7-AAD (10 μg/ml) for 30 min on ice 50. After three washes, the cells were analyzed using the Becton Dickinson FACScan™ and CELLQuest™ software.

10⁶ cells were incubated with 1 μg/ml of biotinylated anti-TCR (thymus) or 1 μg/ml of biotinylated anti–mouse IgD (spleen), for 30 min on ice. After washing, cells were labeled for 30 min on ice with 1 μg/ml streptavidin-FITC. Cells were washed and resuspended in 100 μl of RPMI and incubated at 37°C for 5 min. The cells were pelleted in a microcentrifuge for 15 s and then fixed for 15 min in 2% paraformaldehyde. After washing twice in 0.5% BSA–containing PBS, the cells were cytospun onto slides and mounted with SlowFade Light (Molecular Probes). The cells were visualized with a confocal fluorescence microscope (MRC-600; Bio-Rad Laboratories). The average percentage of capped cells was obtained from 10 microscope fields per sample (100–200 cells/field).

Internalization of the TCR was assayed by incubating 10⁶ lymph node T cells with anti-CD3 antibody (1 μg/ml) for 30 min on ice. After washing, cells were incubated with biotinylated goat anti–hamster antibody (2 μg/ml) for 30 min at 4°C and then warmed at 37°C. Aliquots were removed at varying time points thereafter, transferred to ice, and the reaction was stopped by addition of 0.1% NaN₃. Cells were then stained with FITC-conjugated streptavidin, fixed for 15 min in 4% paraformaldehyde, and washed before analysis by flow cytometry.

Thymocytes were incubated with anti-CD3∈ (1 μg/ml) for 30 min on ice. After washing, the primary antibodies were cross-linked using goat anti–hamster Ig (1 μg/ml) for 5 min. The control cells were incubated only with the cross-linking antibody. Activation was terminated by addition of 4% paraformaldehyde. After fixation, cells were incubated with FITC-conjugated phalloidin (Sigma Chemical Co.) for 30 min, washed three times in PBS, and analyzed using a FACSCalibur™.

Long bones (humerus, femur, tibia) were removed from selected mice, and the ends were then clipped and flushed using a 27-gauge needle and ice-cold HBSS. Clumps of marrow were then broken up by repeated pipetting, and the cells were spun at 500 g at 4°C for 5 min, resuspended in HBSS in a 15-ml polypropylene tube, and underlaid with a discontinuous Percoll gradient (52, 65, and 75% Percoll solution). Cells were then centrifuged at 1,500 g for 30 min at 4°C, and the neutrophil-enriched fraction (at the interface of the 65 and 75% gradients) was harvested, diluted with an equal volume of HBSS, and spun down in a microfuge for 10 s. These cells were resuspended in 1 ml of RPMI and counted using a Coulter counter. To evaluate phagocytosis, cells were mixed with opsonized zymosan (10 particles/cell) and lucifer yellow at 0.5 mg/ml, pelleted in a microfuge tube, and then incubated for 5 min at 37°C. Cells were then washed three times in ice-cold PBS, and the number of cells with lucifer yellow–containing phagosomes was counted using a fluorescence microscope.

To derive ES cells carrying a disruption of the WAS gene, a targeting construct was developed in which a 3.5-kb segment encompassing intron 4 to exon 11 of the WAS gene was substituted with the pPNT expression vector. This vector was then introduced into R1 ES cells, and two clones carrying a WAS mutant allele were then aggregated with CD1 eight cell embryos. Once germline transmission of the targeted allele was confirmed by PCR and Southern analysis (Fig. 1a and Fig. b), the WAS mutation was bred into the C57BL/6 background. The two WAS^−/− ES cell clones were also injected into RAG^−/− blastocysts so as to derive WAS^−/−RAG^−/− chimeric mice which were used in addition to the WAS^−/− mice for some studies of lymphocyte function. As illustrated by immunoblotting analysis, male mice carrying the WAS mutant allele showed no WASp expression in either thymic or splenic cells (Fig. 1 C) despite the presence of surface (s)Ig⁺ B cells and mature CD4⁺ CD8⁻ and CD4⁻CD8⁺ T cells (Fig. 1 D).

Although lymphoid cells were present in the WAS^−/− mice, an analysis of cell numbers in these animals revealed lymphocyte cellularity to be reduced in all lymphoid organs examined. A reduction in lymphoid cell number was most evident in the thymus, where the number of thymocytes was about one quarter of that observed in wild-type mice (Table). Reduced thymic cellularity appeared to reflect a decrease in the total and relative numbers of CD4⁺CD8⁺ double-positive thymocytes, an abnormality which, in turn, was associated with relative increases in the numbers of double-negative CD4⁻CD8⁻ and single-positive CD4⁺ CD8⁻ and CD4⁻CD8⁺ cells (Fig. 1 D). Although this defect was not detected in a previous analysis of WASp-deficient mice 45, the current findings are consistent with several lines of evidence revealing lymphocyte maturation to be altered in WAS patients 17,18,22. To further examine this maturation defect, the CD4⁻CD8⁻ double-negative precursors, identified by virtue of negative staining for CD4, CD8, B220, TCR-α/β, and TCR-γ/δ, were evaluated for expression of the maturation markers CD25 and CD44. These latter markers have been previously shown to identify four double-negative thymocyte subsets, CD44⁺CD25⁻, CD44⁺CD25⁺, CD44⁻CD25⁺, and CD44⁻ CD25⁻, which demarcate progressive stages of thymocyte maturation 51. As shown in Fig. 1 D (lower panel), among these four double-negative subsets, the CD44⁻CD25⁺ population was relatively increased and the CD44⁻CD25⁻ population relatively decreased in the WAS^−/− compared with wild-type mice. Thus the reduction in total thymocyte numbers observed in the absence of WASp appears to reflect impaired progression of the CD44⁻CD25⁺ to the CD44⁻CD25⁻ stage and by extension, reduced transit from the double-negative to the double-positive stage. Interestingly, this same perturbance in thymocyte maturation has also been detected in animals deficient for the CD45 52, Lck 53, and Vav 54,55 signaling effectors, an observation which strongly suggests that this defect reflects diminished signaling, most likely via the pre-TCR 56.

WAS^−/− thymocytes were also examined for expression of TCR-α/β, CD3, and the maturation marker CD5, all of which were found to be expressed normally on the mutant thymocytes (data not shown). Similarly, analysis of splenic and bone marrow B cell populations revealed expression of CD43, CD19, CD23, CD86, CD25, B220, sIgM, and sIgD to be comparable in WAS^−/− and wild-type mice (data not shown). However, for thymocytes, total numbers of peripheral CD4⁺CD8⁻ and CD4⁻CD8⁺ T cells and peripheral B220⁺sIgM⁺ B cells were reduced by ∼50% in the WAS^−/− chimeric compared with wild-type animals (Table). Together, these data indicate lymphopoiesis to be impaired, albeit not arrested, in the context of WASp deficiency. WASp effects on lymphoid ontogeny appear to be predominantly realized in the thymic compartment at the double-negative stage of thymocyte differentiation, and it may be for this reason, as well as sensitivity of this defect to strain background 54, that impaired thymic ontogeny was not detected in an independently derived WAS^−/− mouse 45. Importantly, however, in the latter mice, as well as in some WAS patients 6, significant reductions of peripheral T lymphocyte number have been observed. Thus, the available evidence strongly suggests a role for WASp in driving the maturation and/or expansion of T and possibly mature B cell populations. Whether this latter role is subserved by WASp modulation of the proliferation/expansion and/or survival of lymphoid precursors remains to be determined. Alternatively, the reduction in lymphocyte number observed in the WAS^−/− mice might reflect perturbation in the transit of immature precursors from the bone marrow to other lymphoid organs.

In addition to lymphocytes, platelets were also found to be reduced in the periphery of WAS^−/− mice. As shown in Table and consistent with the detection of thrombocytopenia in virtually all WAS patients, platelet number was reduced by ∼40% in the WAS^−/− relative to wild-type mice. At present, the basis for this finding remains unclear, with preliminary data revealing that platelet survival is normal in these animals (data not shown). The physiologic significance of this defect is also unclear, as the WAS^−/− mice, in contrast to classical WAS patients, show no overt signs of a hemorrhagic diathesis.

In view of previous data suggesting that antigen receptor–evoked lymphocyte activation is impaired in WAS patients 9,10,12, thymocytes and peripheral B and T lymphocytes from the WAS^−/− chimeras were next studied with respect to their responses to antigen receptor ligation. The results of these analyses revealed that anti-CD3 antibody–evoked proliferation was markedly reduced in WAS^−/− relative to WAS^+/+ thymocytes (Fig. 2 A) at all anti-CD3 antibody concentrations tested (data not shown). Similarly, Con A stimulation elicited significantly less proliferation in WAS^−/− thymocytes than wild-type cells. By contrast, the mutant cells proliferated normally in response to PMA and ionomycin, a maneuver which triggers proliferation in the absence of antigen receptor engagement. In addition, anti-CD3 as well as anti-CD3 plus anti-CD28 antibody–induced upregulation of the CD69 early activation marker 57,58 were markedly diminished in the mutant cells (Fig. 2 D).

Similar to WAS^−/− thymocytes, WASp-deficient peripheral T cells were also observed to have reduced proliferative responses to anti-CD3 antibody, but a normal response to PMA and ionomycin (Fig. 2 B). However, WAS^−/− peripheral T cells also responded, albeit less than wild-type cells, to anti-CD3 antibody stimulation combined with PMA treatment, whereas addition of CD28 could partially restore proliferation of these cells (Fig. 2 B, left panel). Anti-CD3–induced proliferative responses of the WASp-deficient cells were also restored to almost normal levels by the addition of IL-2 (Fig. 2 B, right panel). To investigate the basis for this latter finding, anti-CD3–induced production of IL-2 was also evaluated in the WAS^−/− T cells. As shown in Fig. 2 E, wild-type T cells secreted only a small amount of IL-2 after anti-CD3 antibody treatment, but dramatically increased their IL-2 production in conjunction with anti-CD28 antibody–mediated costimulation. However, in the WAS^−/− cells, IL-2 production was not only negligible after anti-CD3 antibody stimulation, but also increased by much less (50%) in these cells than in wild-type cells after CD28 costimulation (Fig. 2 E). This defect in the induction of IL-2 does not reflect impairment of IL-2 secretion, as an evaluation of IL-2 production based on intracellular staining of T cells stimulated with anti-CD3/anti-CD28 antibodies in the presence of the secretion blocker Brefeldin A 59,60 revealed the numbers of IL-2–staining cells to be ∼50% less in the WAS^−/− relative to wild-type cell cultures (Fig. 2 E). By contrast, in the absence of Brefeldin A, IL-2 accumulation was not detected in the mutant or wild-type cells, indicating the IL-2 secretion pathway to be intact in the context of WASp deficiency. Thus, these observations indicate that WASp effects on TCR signaling are modulated by costimulatory signals and also suggest that the T cell proliferative defect conferred by WASp deficiency is due, at least in part, to a defect in IL-2 production. Taken together, these data confirm the participation of WASp in the transduction of activation signals via the TCR and suggest that the effects of WASp on TCR-mediated activation are most profound in immature T cells and may be tempered by concomitant activation signals delivered via CD28, IL-2, and possibly other stimulatory receptors. The data also identify potential functional differences between the T cells of the WASp-deficient mice studied here and those of WAS patients in whom anti-CD3–induced proliferative responses appear to be only partially restored by exogenous IL-2 10. While further studies are required to determine whether this discrepancy reflects the use of transformed WAS patient cell lines, the current data provide definitive evidence of a role for WASp in promoting the coupling of TCR stimulation to T cell activation and to the IL-2 production integral to normal T cell mitogenesis.

In contrast to thymocytes and peripheral T cells, WASp-deficient splenic B cells showed only a marginal reduction in response to antigen receptor ligation, a finding that was not altered by varying the concentrations of anti-Ig antibody (Fig. 2 C). As observed in WAS^−/− T cells, this mild reduction in responsiveness of the WASp-deficient B cells to mitogenic stimulus appears to be antigen receptor specific, as these cells proliferated normally in response to LPS, IL-4 (Fig. 2 C), and the combination of IL-4 with anti-CD40 antibody (data not shown). Thus, WASp deficiency appears to be associated with impairment in T cell and to a much lesser extent, B cell antigen receptor (BCR) signaling. While a defect in BCR signaling was not reported in a previous study of WASp-deficient mice 45, the current findings are consistent with data indicating that antigen receptor–evoked B cell proliferative responses are impaired in patients with classical WAS 9. However, it is currently unclear whether the divergent effects of WASp deficiency on TCR versus BCR signaling reflect differences in the biochemical requirements required for signal delivery by the respective receptors or the capacity of another effector, such as N-WASp, to adequately subserve WASp functions in B, but not T cells.

In view of these data identifying a role for WASp in coupling antigen receptor engagement to lymphocyte proliferation, the possibility that this protein also modulates antigen receptor capacity to activate cell death cascades was also investigated. Specifically, WAS^−/− and wild-type double-positive thymocytes were cultured for 24 h in the presence of anti-CD3 antibody alone or combined with anti-CD28 antibody, and the viability and rate of cellular apoptosis were then evaluated in CD4⁺CD8⁺ cells by immunofluorescence analysis of 7-AAD staining. As shown in Fig. 2 F, results of this analysis revealed that viability of the anti-CD3–stimulated WAS^−/− CD4⁺CD8⁺ cells was ∼35% greater than that of wild-type cells. This difference was even more marked in the context of anti-CD3/anti-CD28 stimulation. By contrast, cell viability after anti-Fas antibody or PMA/ionomycin treatment was similar in the mutant and wild-type cells. Therefore, these data indicate that WASp specifically modulates the apoptotic as well as mitogenic signaling cascades evoked by TCR engagement. At present, the physiologic significance of these findings remains unclear, although the involvement of WASp in TCR-mediated apoptosis implies a potential role for this protein in facilitating negative selection events in the thymus, a possibility which might be etiologically relevant to the autoimmune phenomena frequently expressed by WAS patients 6.

To elucidate the biochemical basis for the T cell functional defects conferred by WASp deficiency, WAS^−/− T cells were evaluated with respect to the signaling events elicited by TCR engagement. As indicated by the antiphosphotyrosine immunoblots shown in Fig. 3a and Fig. b, the levels of TCR-ζ, ZAP70, and total cellular protein tyrosine phosphorylation detected in resting and TCR-stimulated WAS^−/− thymocytes and peripheral T cells were essentially identical to those observed in similarly treated wild-type cells. Along similar lines, TCR-evoked increases in MAPK activation and in phosphorylation, and presumably activation of SAPK/JNK were comparable (Fig. 3C and Fig. D). By contrast, the increase in intracellular calcium levels induced by TCR ligation was less sustained in the WAS^−/− compared with wild-type cells, with intracellular calcium concentrations 20 and 25% reduced in the WAS^−/− relative to wild-type cells at the 600- and 700-s time points, respectively (Fig. 3 E). At present, it is unclear whether this quantitative difference in intracellular calcium mobilization observed in wild-type compared with WAS^−/− cells translates to a sufficiently significant perturbance of other downstream signaling events, such as NF-AT translocation to the nucleus, so as to account for the very significant impairment in TCR-induced IL-2 expression by the mutant cells.

As WASp has also been implicated by many lines of evidence in the regulation of cytoskeletal architecture, the relevance of WASp to lymphocyte cytoskeletal rearrangements induced by antigen receptor engagement was also investigated. To this end, WAS^−/− lymphocytes were studied with respect to the induction of antigen receptor capping, a phenomenon that requires de novo actin polymerization and microfilament rearrangement. As indicated in Fig. 4 A and illustrated by the representative micrograph (Fig. 4 B), a defect in ligand-induced capping was detected in WAS^−/− thymocytes, the number of these cells showing T cell antigen receptor clustering to be ∼40% less than that of wild-type thymocytes. As is consistent with a defect, albeit mild in BCR-induced proliferation, anti-Ig–mediated capping of the BCR was also impaired in the context of WASp deficiency. In addition, immunofluorescence analysis of anti-CD3–treated phalloidin-labeled WAS^−/− thymocytes, an assay that selectively detects expression of polymerized F-actin, revealed that antigen receptor–mediated actin polymerization was markedly reduced in WAS^−/− thymocytes (Fig. 4 C) and lymph node T cells (data not shown). Together, these data demonstrate a requirement for WASp in the regulation of cytoskeletal rearrangement in response to antigen receptor engagement.

In addition to patch and cap formation, internalization of antigen receptors after their engagement appears to require cytoskeletal rearrangement, as endocytosis of these receptors is abrogated by cytoskeleton-disrupting agents such as dihydrocytochalasin B 61. In view of this observation, together with data from our lab (Siminovitch, K., unpublished observations) and another 62 revealing the capacity of WASp and the Las17/Beel yeast WASp homologue, respectively, to interact with proteins implicated in endocytosis, the possibility that WASp plays a role in ligand-triggered TCR endocytosis was investigated by assaying the levels of biotinylated TCR present on WAS^−/− T cells at varying times after cell stimulation. As illustrated in Fig. 4 C, the results of this analysis revealed the rate of TCR endocytosis to be markedly reduced in the context of WASp deficiency, with 50% of the receptors internalized in wild-type cells, but only 10% internalized in WAS^−/− cells at 1 h after stimulation. Therefore, these results suggest the involvement of WASp in the process of receptor endocytosis. At present, the level at which the endocytic pathway is disrupted by WASp deficiency remains unclear. However, this defect may also impact on the properties of nonlymphoid cells, as opsonized zymosan particle internalization was also found to be significantly reduced in WAS^−/− compared with wild-type neutrophils (Fig. 4 E). Together, these data raise the possibility that the altered endocytosis of the TCR and/or other receptors and consequent changes in levels of surface receptor expression impact on a broad range of immune cellular functions, including not only cell activation, but also maintenance of cell polarity, antigen presentation, and/or the uptake of extracellular nutrients and microorganisms.

In summary, the current data reveal that transduction of TCR-evoked proliferation, apoptosis, and actin polymerization–inducing signals are impaired in WAS^−/− T cells. WASp deficiency also engenders a defect in thymocyte maturation, by impairment of the progression of CD4⁻CD8⁻ progenitors from the CD44⁻CD25⁺ to a CD44⁻CD25⁻ stage. This developmental defect may reflect diminished signaling capacity of the pre-TCR in the absence of WASp, as has been previously proposed in relation to the similar, albeit more severe, maturational defect detected in Lck-deficient thymocytes 53. Similarly, the reduction in peripheral T cell numbers conferred by WASp deficiency may reflect not only impaired thymopoiesis, but also diminished proliferative/expansion potential consequent to defective transduction of TCR mitogenic signals. As is consistent with the immunologic abnormalities described in WAS patients, the early stages of B cell ontogeny appear intact and BCR-induced proliferation is only modestly diminished in the context of WASp deficiency. However, the reduction in splenic B cell numbers and in capping capacity of stimulated antigen receptors on WAS^−/− B cells implies that the modest diminution in antigen receptor–evoked B cell proliferation is physiologically relevant and may engender impaired expansion of B cells within the peripheral compartment.

Despite the plethora of functional defects detected in WAS^−/− T cells and thymocytes, many of the biochemical events involved in signal relay through the TCR (such as tyrosine phosphorylation and MAPK and SAPK/JNK activation) remain intact in these cells. In this context, the WAS^−/− cells are strikingly similar to the lymphoid cells of mice genetically rendered deficient for Vav 63,64,65, a guanine nucleotide exchange factor for the Rho family of GTPases 66,67. Although the effect of Vav deficiency on B cell ontogeny and function is more prominent than that of WASp deficiency, Vav^−/− mice manifest defects in T cell maturation, TCR-evoked proliferation, cytokine production, and actin polymerization that closely resemble those observed in the WAS^−/− mice. In addition, as with WAS^−/− T cells, Vav^−/− T cells show little or no abnormalities with respect to the early activation events induced by TCR ligation. At present, the biochemical basis whereby both the structural (receptor clustering and actin polymerization) and physiologic (proliferation and cytokine production) sequelae of TCR ligation are disrupted by WASp or Vav deficiency remains unclear. However, a relationship between these facets of T cell activation has become increasingly well recognized and is strongly supported by recent data indicating a pivotal role for cytoskeletal rearrangement in reorienting the T cell–APC interface so as to create a framework for the incorporation of signaling molecules into supramolecular activation clusters (SMACs) that amplify signal delivery 68,69. Taken together with the T cell defects observed in WAS^−/− mice, these observations suggest that while TCR activation signals can be initiated in the absence of cytoskeletal-driven events such as receptor clustering, signal delivery in this context is insufficient to achieve optimal activation and cytokine production. Thus, by virtue of its capacity to interact with both cdc42 and the Arp2/3 complex, WASp may act to promote the spatial rearrangements required for SMAC formation and thereby facilitate the juxtaposition of activated antigen receptors and signaling molecules that is required for efficient transduction of activation signals. Accordingly, the defective actin polymerization and TCR clustering engendered by WASp deficiency would predictably translate to signaling of insufficient amplitude and/or duration for optimal activation and cytokine production. This hypothesis is supported by preliminary data revealing that T cell microtubule organizing center (MTOC) polarization toward the T cell–APC contact site is reduced in stimulated WAS^−/− T cells (Burkhardt, J., and K. Siminovitch, data not shown) and, if correct, might explain the disparate effects of WASp deficiency on B cell versus T cell activation, the latter of which appears more highly dependent on proper actin filament assembly 70,71. In view of data indicating CD28–B7 interaction to be both necessary and sufficient for inducing movement of the cortical actin cytoskeleton toward the TCR–ligand interface 68, the capacity of CD28 costimulation to partially normalize TCR-evoked proliferative responses of WAS^−/− T cells suggests that WASp effects on cytoskeletal rearrangement may be particularly important for the propagation of TCR-initiated signals evoked in the absence of costimulatory signal. The capacity of WASp to interact with Nck may also be of some relevance to WASp effects on TCR signaling, as the Nck adaptor protein has been shown to promote TCR-mediated NF-AT activation by virtue of an interaction with the Pak1 serine/threonine kinase 40. As Pak1, like WASp, appears to act downstream of cdc42 activation in a JNK-independent fashion, these observations suggest that WASp effects on TCR-evoked IL-2 production may reflect not only WASp modulation of upstream cytoskeletal-mediated TCR rearrangements, but also WASp interactions with both activated cdc42 and Nck. While these hypotheses require further investigation, the current data indicate that mice with WASp-deficient lymphoid cells provide a valuable system for dissecting out the complex structural and biochemical properties of WASp and delineating the molecular events that couple antigen receptor stimulation to cytoskeletal rearrangement and lymphocyte activation.

The authors thank Dr. Gordon Mills for helpful discussions, Denis Bouchard for assistance with immunofluorescence analysis, Marina Gertsenstein for assistance with the ES cell work and aggregations, and Dr. Tak Mak for his generosity in facilitating the genesis of the WAS^−/− mice.

This work was supported by grants to K.A. Siminovitch and J.M. Penninger from the Medical Research Council of Canada. K.A. Siminovitch is a Research Scientist of the Arthritis Society of Canada; L.A.G. da Cruz is a recipient of a fellowship from the Fundação para a Ciência e a Tecnologia of Portugal; and A.K. Somani is supported by a Steve Fonyo studentship from the National Cancer Institute of Canada.