Syk and ZAP-70 subserve nonredundant functions in B and T lymphopoiesis. In the absence of Syk, B cell development is blocked, while T cell development is arrested in the absence of ZAP-70. The receptors and the signaling molecules required for differentiation of NK cells are poorly characterized. Here we investigate the role of the Syk protein tyrosine kinase in NK cell differentiation. Hemopoietic chimeras were generated by reconstituting alymphoid (B, T, NK) recombinase-activating gene-2 × common cytokine receptor γ-chain double-mutant mice with Syk−/− fetal liver cells. The phenotypically mature Syk−/− NK cells that developed in this context were fully competent in natural cytotoxicity and in calibrating functional inhibitory receptors for MHC molecules. Syk-deficient NK cells demonstrated reduced levels of Ab-dependent cellular cytotoxicity. Nevertheless, Syk/− NK cells could signal through NK1.1 and 2B4 activating receptors and expressed ZAP-70 protein. We conclude that the Syk protein tyrosine kinase is not essential for murine NK cell development, and that compensatory signaling pathways (including those mediated through ZAP-70) may sustain most NK cell functions in the absence of Syk.

The Syk family of protein tyrosine kinases (PTKs)3 contains two members, p72syk (Syk) and ZAP-70 (1), which share structural features, including a consensus tyrosine kinase domain and two Src homology type 2 domains. The Src homology type 2 domains are required for the association of Syk and ZAP-70 to the immunoreceptor tyrosine-based activation motif (ITAM) contained within the cytoplasmic domain of activating cell surface receptors on hemopoietic cells (2). Syk is expressed in many hemopoietic cell types, although its levels are greatest in B cells (3). Expression of ZAP-70, in contrast, is restricted to T and NK cells (3). The coexpression of Syk and ZAP-70 in the same lymphoid cell subsets (T and NK cells) suggests that these two related PTKs could subserve redundant functions. Nevertheless, the intracellular targets of the Syk and ZAP-70 kinases are not entirely elucidated, and it remains possible that Syk and ZAP-70 phosphorylate a unique set of protein substrates.

Syk is primarily activated following engagement of ITAM-containing receptors present in the B cell receptor (BCR) complex and in the Fc receptors (FcR) for IgG (FcγRI, FcγRII, FcγRIII) and IgE (FcεRI) (1, 4). Engagement of platelet integrins has been shown to activate Syk (5), and Syk appears complexed to the IL-2R β-chain (6), a functional component of the IL-2 and IL-15 receptors (reviewed in Ref. 7). Thus, Syk activation potentially intervenes in signaling from Ag receptors, FcRs, integrins and cytokine receptors.

Gene-targeting experiments have unraveled some of the essential roles of the Syk PTK in vivo. Mice deficient in Syk (Syk−/−) die in utero or during the prenatal period from excessive hemorrhage (8, 9). This phenotype could be explained by the failure of collagen-activated Syk−/− platelets to phosphorylate PLCγ2, which, in turn, results in compromised platelet secretion and aggregation (10). However, collagen signaling in platelets is also defective in Fcγ-deficient mice, which do not show a bleeding diathesis, and bleeding times in chimeric Syk−/− mice are normal (10). Alternatively, a defect in the function of other hemopoietic cell types (perhaps macrophages) in the absence of Syk could lead to microvascular destruction (10).

Due to the lethal nature of Syk deficiency, any role for the Syk PTK in lymphoid development has been addressed in somatic or hemopoietic chimeras using recombinase-activating gene-2 (RAG2) mutant mice as hosts (8, 9). In these studies B cell development failed to progress beyond the pro-B cell stage due to an inability of Syk-deficient early B cells to couple pre-BCR ligation to downstream intracellular processes (8, 9, 11). In contrast, αβ T cell development appeared normal in Syk−/− mice, although overall T cell numbers were reduced (9), and the dendritic epidermal γδ T cell subpopulation was not detected (8, 12). Reciprocally, in ZAP-70-deficient mice (13, 14) and humans (15) the development of αβ T cells, γδ T cells, and NK1.1+ αβ T cells is abnormal. In contrast, NK and B lymphopoiesis in ZAP-70 mutants are not affected. The T cell defect in ZAP-70 deficiency is caused by abnormal signaling following TCR triggering, thereby highlighting the analogous functions of Syk and ZAP-70 in BCR and TCR pathways, respectively.

Syk could intervene in the differentiation of NK cells via 1) signal transduction through the IL-2R β-chain in response to IL-15, a vital growth and differentiation factor for NK cell progenitors and mature NK cells (16); 2) activating NK cell surface receptors, including those involved in natural cytotoxicity; or 3) FcγRIII activation for Ab-dependent cellular cytotoxicity (ADCC) (reviewed in Ref. 17). Recently, Brumbaugh et al. (18) have suggested that Syk plays a major role in mediating the natural cytotoxicity of human NK cells. However, a role for Syk in NK cell differentiation could not easily be assessed in the previously reported Syk/− chimeras, because the host RAG2 mice can generate their own functional NK cells (19).

We have developed a novel alymphoid mouse strain harboring both the RAG2 and common cytokine receptor γ-chain (γc) mutations. RAG2/γc mutant mice have certain advantages for the study of lymphoid development in vivo, including the complete absence of all mature T, B, and NK cells and a reduced number of lymphoid precursors (20). The lymphoid system in RAG2/γc mice can be stably and efficiently reconstituted using normal or mutant hemopoietic stem cells (HSC) from adult bone marrow (BM) or fetal liver (FL), and thus RAG2/γc mutant mice provide a new tool to study the role of any gene in NK cell differentiation in vivo. In this report we have analyzed the role of the Syk PTK in murine NK cell development and function.

Mice with a null mutation in the γc (21) were from the fourth generation backcross to the C57BL/6 background. RAG2 mice (19) from the 10th generation backcross to C57BL/6 were provided by Dr. B. Rocha (Paris, France). Mice doubly deficient in RAG2 and γc (RAG2/γc; H-2b) were obtained by intercrossing as previously described (20). C57BL/6 and C57BL/6.β2m−/− mice were obtained from CDTA/CNRS (Orleans, France).

Mice heterozygous for the Syktm1Tyb mutation (8) that had been backcrossed five generations onto the B10.D2 background (H-2d) were intercrossed to generate day 15.5–16.5 Syk−/− and control (Syk+/+ or Syk+/−) embryos. The observation of a vaginal plug was designated day 0.5. FL cell suspensions were obtained by passage of the tissue through a 23-gauge needle. RAG2/γc mice (>6 wk of age) were irradiated with 0.3 Gy from a cobalt source and 4 h later were injected i.v. with 5 × 106 FL cells. No obvious differences were seen between Syk+/+ or Syk+/− FL chimeras, which will be referred to as Syk+ chimeras. All mice received tetracycline and bactrim in the drinking water for the period following FL cell transfer. The Syk genotypes of the embryos were determined by Southern blotting as previously described (8).

Single-cell suspensions were prepared from spleen, BM, thymus, and liver. Erythrocytes were lysed in ammonium chloride, and cells were resuspended in PBS with 3% FCS and 0.01% sodium azide. mAbs directly conjugated to FITC, PE, Tricolor, or biotin were used for immunofluorescence analysis as previously described (20), including Abs specific for CD2, CD3, CD4, CD8, TCRαβ, TCRγδ, CD11b (Mac-1), CD16 (FcRγII/III), CD19, CD24, B220, IgM, CD90 (Thy-1), CD117 (c-Kit), CD122 (IL2Rβ), NK1.1, DX5, 2B4, Ly49A, Ly49G2, CD132 (γc), H-2Kd, H-2Kb, and Gr-1 (all from PharMingen, San Diego, CA).

Splenocytes were passed through nylon wool columns to remove most B cells and macrophages. Nylon wool-nonadherent cells were cultured in flat-bottom 24-well plates at 5 × 106 cells/ml in complete medium (RPMI 1640 with 10% FCS, 10−5 M β-ME, 100 μg/ml streptomycin, and 100 U/ml penicillin) supplemented with 20 ng/ml of human IL-15 (R&D Systems, Minneapolis, MN) or 1000 U/ml of human IL-2 (PeproTech, Rocky Hill, NJ). After 3–4 days the nonadherent cells were removed, and the adherent lymphokine-activated killer (A-LAK) cells were refed and cultured until days 8–10. A-LAK cultures produced in this manner routinely contained >95% NK1.1+/CD3 cells.

A 51Cr release assay was used to measure NK activity in vitro as previously described (22). Target cells (YAC-1, EL-4, P815, mouse fibroblasts (L cells) or their Qa-Ib-transfected derivatives (F12) (23), or Con A-activated B6 and β2m−/− blasts) were labeled with 100 μCi of 51Cr (ICN Pharmaceutical, Costa Mesa, CA), and 2.5–5 × 103 targets were incubated with graded numbers of effector cells in 200 μl of medium for 4 h. For natural cytotoxicity, effector cells were NK-enriched, B cell-depleted splenocytes from poly(I:C)-primed mice (40 h after injection with 0.2 mg of poly(I:C)/mouse). A-LAK cells were used as effectors for 1) Ab-dependent cell cytotoxicity (ADCC) using EL-4 cells with or without Thy 1.1 mAb (20); 2) reverse ADCC (rADCC) using FcR+ P815 cells in the presence or the absence of NK1.1, 2B4, or CD16 mAbs; or 3) MHC inhibition and recognition of “missing self” using Con A-activated B6 and β2m−/− blasts or L and F12 cells. Radioactivity released into the cell-free supernatant was measured, and the percent specific lysis was calculated as follows: 100 × [(experimental release − spontaneous release)/(maximum release − spontaneous release)].

Splenic NK cells were FACS sorted using the pan-NK cell mAb DX5, which is coexpressed on 95% of splenic NK cells and shows no NK cell-activating or -blocking activities (PharMingen). Purified DX5+ NK cells (105 cells/200 μl) were cultured in flat-bottom microtiter plates in human IL-15 (20 ng/ml) or human IL-2 (1000 U/ml). Stimuli included mIL-12 (2 ng/ml; PeproTech), or mAbs NK1.1 (clone PK136; 20 μg/ml), 2B4 (5 μg/ml), CD16 (75 μg/ml), or control Gr-1 (10 μg/ml). Wells were precoated with mAbs (50 μl/well) for 4 h before adding NK cells. After 24 h, the cell-free supernatants were collected, and the amount of IFN-γ was quantitated by ELISA (Genzyme, Cambridge, MA).

Analysis of ZAP-70 protein expression in FACS-purified NK1.1+/CD3 NK cells was performed by immunoblotting as previously described (24).

Reconstitution of RAG2/γc mice with Syk−/− FL cells generated the same developmental arrest at the BM pre-B cell stage that was originally described using RAG2−/− blastocyst complementation (8, 9). Few IgM+B220+ B cells could be detected in the BM of Syk−/ chimeras, but these cells were almost completely absent in the spleen (Fig. 1,A, Table I, and data not shown). In contrast, normal numbers of BM and splenic IgM+B220+ B cells were present in Syk+ chimeras. Consistent with previous reports (8, 9), Syk−/− chimeras showed no obvious defect in thymocyte development as defined by CD4 and CD8 expression (Fig. 1 B), although thymocyte numbers were reduced 2-fold in Syk−/− thymi (data not shown). Mature CD4 and CD8 T cells were found in the peripheral lymphoid organs of these mice, which expressed normal levels of TCR-αβ (data not shown). Moreover, γδ T cells and NK T cells could be readily detected in the thymus, spleen, and liver and lining the gut intestinal tract of Syk−/− FL→RAG2/γc chimeras (F.C., D.G.-G., A. Wilson, M.T., E.S., V.L.J.T., and J.P.D., manuscript in preparation).

FIGURE 1.

Lymphoid development in the absence of Syk. BM cells (A) were stained with FITC-anti-IgM and PE-anti-B220, while thymocytes (B) were stained with FITC-anti-CD4 and PE-anti-CD8. T cell subsets are normally represented in Syk−/− chimeras, while B cells (B220+ IgM+) are barely detectable in BM and are absent in spleen, as previously described (9 ). C, Splenocytes were stained with FITC-anti-IL-2Rβ, PE-anti-NK1.1, biotinylated anti-CD19, and biotinylated anti-TCRαβ. B cells and αβ T cells were electronically excluded from the analysis, and expression of IL-2Rβ vs NK1.1 is shown. IL-2Rβ+ NK1.1+ NK cells represented between 30–60% of non-B, non-T cells in the spleens of Syk−/− chimeras. D, The donor origin (H-2d) of NK cells was verified in TCRαβ CD19 NK1.1+ splenocytes. The NK1.1 TCRαβ CD19 H-2d- cells are of host origin and stain positively for H-2b (data not shown).

FIGURE 1.

Lymphoid development in the absence of Syk. BM cells (A) were stained with FITC-anti-IgM and PE-anti-B220, while thymocytes (B) were stained with FITC-anti-CD4 and PE-anti-CD8. T cell subsets are normally represented in Syk−/− chimeras, while B cells (B220+ IgM+) are barely detectable in BM and are absent in spleen, as previously described (9 ). C, Splenocytes were stained with FITC-anti-IL-2Rβ, PE-anti-NK1.1, biotinylated anti-CD19, and biotinylated anti-TCRαβ. B cells and αβ T cells were electronically excluded from the analysis, and expression of IL-2Rβ vs NK1.1 is shown. IL-2Rβ+ NK1.1+ NK cells represented between 30–60% of non-B, non-T cells in the spleens of Syk−/− chimeras. D, The donor origin (H-2d) of NK cells was verified in TCRαβ CD19 NK1.1+ splenocytes. The NK1.1 TCRαβ CD19 H-2d- cells are of host origin and stain positively for H-2b (data not shown).

Close modal
Table I.

Splenic cellularity in Syk+ and Syk−/− chimerasa

ChimerasTotalNK1.1+TCRαβ+B220high
Syk+ (n = 6) 110.6 ± 24.0 1.45 ± 0.53 27.3 ± 3.61 69.7 ± 20.6 
Syk−/− (n = 7) 12.5 ± 2.53 0.98 ± 0.29 9.18 ± 1.64 0.19 ± 0.10 
ChimerasTotalNK1.1+TCRαβ+B220high
Syk+ (n = 6) 110.6 ± 24.0 1.45 ± 0.53 27.3 ± 3.61 69.7 ± 20.6 
Syk−/− (n = 7) 12.5 ± 2.53 0.98 ± 0.29 9.18 ± 1.64 0.19 ± 0.10 
a

Spleen cell suspensions were stained with a mixture of anti-B220, anti-NK1.1, and anti-TCRαβ mAbs. Absolute numbers (×106) were calculated based on the percentages of single positive cells out of the total number of lymphoid cells. For B lymphocytes, only the B220high cells were calculated, because most B220low cells coexpress NK1.1.

These results confirm previous observations using Syk-deficient HSC transfers (8, 9) and rule out any effects of competition from host early B cell precursors on the Syk−/− B cell developmental block. In contrast to previous reports (10), we found no evidence of early deaths in Syk−/− chimeras due to hemorrhagic ascites or anemia. Syk−/− chimeras remained healthy up to 8 mo postgraft, and histological examination of the intestinal walls of these mice were completely normal (data not shown). The observed pathology following transfer of Syk-deficient fetal liver cells into lethally irradiated recipients (10) probably relates to the degree of chimerism in nonlymphoid hemopoietic lineages. In our study the mild irradiation of the host RAG2/γc mice (0.3 Gy) might result in low levels of chimerism in nonlymphoid compartments.

Syk-deficient FL cells reconstituted NK cell development in RAG2/γc chimeras. Flow cytometric analysis revealed that NK1.1+ IL-2Rβ+ NK cells were readily detectable in the CD19 TCRαβ fraction of splenic, hepatic, and BM cells from Syk+ and Syk−/− chimeras (Fig. 1,C and data not shown). Absolute numbers of NK cells were similar between the two groups of mice (<2-fold reduction was found in Syk−/− chimeras; Table I). The phenotype of Syk−/− NK cells was analyzed by flow cytometry using a panel of NK cell-associated markers. Both Syk+ and Syk−/− NK1.1+ cells demonstrated a normal percentage and expression level of CD2, CD11b, CD16, B220, DX5, 2B4, CD90, and CD117 (data not shown). Finally, the donor origin of NK cells was confirmed by PCR amplification of the RAG2 gene in sorted NK cells (data not shown) and by the cell surface expression of H-2Kd molecules by NK1.1+ splenocytes in the H-2b RAG2/γc hosts (Fig. 1 D). Reciprocal staining with an anti-H-2Kb mAb confirmed that all lymphoid cells were donor derived (data not shown). These results demonstrate that the Syk PTK is not essential for murine NK cell development in vivo.

The signal transduction pathways required for the generation of NK cells are poorly characterized. Cytokine receptor signaling via the IL-15/IL-15R complex is required for NK cell development, presumably at an early stage during the commitment of HSC to the NK lineage (16). A potential role for Syk in IL-15 responses has been implied as well, because an association of Syk with the IL-2R β-chain has been documented (6), and this chain is required for signaling via IL-2 or IL-15. Previous studies, however, have demonstrated that Syk-deficient T cells can respond to IL-2 in vitro (8), and in this study we did not observe any manifestations of IL-2 deficiency (such as deregulated T cell homeostasis, lymphoproliferation, or colitis) in Syk−/− chimeras (data not shown). Moreover, IL-15 responses of Syk-deficient NK cells appear normal; IL-15-stimulated A-LAK cultures from Syk+ and Syk−/− chimeras were generated similarly, and IL-15 induced a comparable level of proliferation from Syk+ and Syk−/− cells (data not shown). These results suggest that Syk association with the IL-2R β-chain is not essential for many of the functions of IL-2 or IL-15. Finally, our data extend the list of nonessential PTKs for murine NK cell development, which includes Lck, ZAP-70, Fyn, and now Syk (13, 25, 26). It remains possible that the receptors controlling NK cell development do not require any PTK signals.

To analyze the effector functions of Syk−/− NK cells, the lytic capacity of freshly isolated splenic NK cells was tested in vitro. Syk-deficient NK-enriched splenocytes demonstrated normal levels of natural cytotoxicity against YAC-1 targets (Fig. 2 A). This result contrasts with a recent report suggesting that Syk is essential for the natural cytotoxicity of human NK cells (18). Although species-specific differences in NK cell function may account for these differences, one must consider the possibility that the dominant negative (DN) Syk construct used in the studies of Brumbaugh et al. (18) might have inhibited not only Syk, but other signaling pathways (including, perhaps, ZAP-70) as well. Along these lines, studies using a DN LAT construct revealed an essential role for LAT in human NK cell functions (27), while LAT-deficient mice appear to have normal NK activities (28).

FIGURE 2.

Natural cytotoxicity and ZAP-70 expression in Syk−/− NK cells. A, A classical 4-h 51Cr release assay was performed to evaluate NK cell lytic capacity. Specific lysis was determined at the indicated E:T cell ratios using NK-enriched (B cell-depleted) spleen cells from poly(I:C)-treated mice as effectors and 51Cr-labeled YAC-1 cells as targets. One representative experiment of three is shown. B, Sorted NK cells from Syk+ or Syk−/− RAG2/γc chimeras or control Jurkat T cells were lysed, and total cellular proteins were analyzed by anti-ZAP-70 immunoblotting. The arrow indicates the migration of ZAP-70, and m.w. standards are indicated on the right.

FIGURE 2.

Natural cytotoxicity and ZAP-70 expression in Syk−/− NK cells. A, A classical 4-h 51Cr release assay was performed to evaluate NK cell lytic capacity. Specific lysis was determined at the indicated E:T cell ratios using NK-enriched (B cell-depleted) spleen cells from poly(I:C)-treated mice as effectors and 51Cr-labeled YAC-1 cells as targets. One representative experiment of three is shown. B, Sorted NK cells from Syk+ or Syk−/− RAG2/γc chimeras or control Jurkat T cells were lysed, and total cellular proteins were analyzed by anti-ZAP-70 immunoblotting. The arrow indicates the migration of ZAP-70, and m.w. standards are indicated on the right.

Close modal

The lack of any effects of Syk deficiency on YAC-1 killing could be explained by compensation via other related PTKs, such as ZAP-70. Consistent with this hypothesis, we found that purified NK1.1+ CD3 NK cells from both Syk+ and Syk−/− chimeras expressed ZAP-70 protein (Fig. 2 B). Moreover, the levels of ZAP-70 were similar in both types of NK cells, suggesting that Syk and ZAP-70 expression are independently regulated within NK cells. The demonstration of the Syk-related ZAP-70 PTK in Syk-deficient NK cells raises the possibility of ZAP-70-mediated compensation for certain effector functions in the absence of Syk.

The ability of Syk and ZAP-70 to partially compensate for each other in T cell development has been demonstrated in ZAP-70 mutant mice, in which intrathymic differentiation is arrested at the double-positive stage (13, 14). Enforced expression of Syk in ZAP-70-deficient mice can rescue T cell development (29). In B cells, the absence of Syk arrests development with no compensation by ZAP-70, due to its lack of expression in early B cell progenitors (3). However, ZAP-70 can restore BCR signaling in a Syk-deficient B cell line (30), demonstrating that compensation is possible. Redundancy between Syk- and ZAP-70-mediated signaling may also operate during NK cell development and for NK cell effector functions. In this respect, NK cells are the only mouse lymphocyte subset in which Syk family PTKs are nonessential for differentiation. In contrast, αβ T cells, γδ T cells, and B cells are strictly dependent either on ZAP-70 (αβ T cells and some γδ T cells (13, 14) or on Syk (some γδ T cells and B cells (8, 9, 12).

Previous studies have suggested a role for PTKs in the function of NK cell receptors, including members of the Ly49 family. In particular, the role of signaling molecules in the shaping of the NK cell repertoire (31) or in calibrating Ly 49 receptor expression (32) has not been defined. Ly49A binds H-2Dd and H-2Dk, while Ly49G2 binds to H-2Dd and H-2Ld (reviewed in Ref. 31). Because the donor FL cells were H-2d, the resultant RAG2/γc chimeras harbored cells bearing either H-2b (host) or H-2d (donor). This allowed us to determine whether Syk played a role in the regulation of expression of Ly49 receptors by H-2Dd (33).

We examined the expression levels of Ly49A and Ly49G2 receptors and the frequencies of these Ly49+ NK subsets in Syk and Syk−/− RAG2/γc chimeras. As shown in Fig. 3,A, the percentages of Ly49A+ or Ly49G2+ NK cells were not altered in the absence of Syk. Moreover, both Ly49A and Ly49G2 levels were similarly decreased in Syk+ and Syk−/− NK cells compared with that observed in NK cells derived from C57BL/6 mice (Fig. 3 A). These results demonstrate that the Syk PTK signaling is not required for Ly49 receptor calibration and suggest that expression of H-2d molecules by hemopoietic cells is sufficient to shape the NK cell repertoire.

FIGURE 3.

Inhibitory receptors are normally regulated and function efficiently in the absence of Syk. A, Spleen cells were stained with anti-Ly49A or Ly49G2, anti-NK1.1, and a combination of anti-CD19 and anti-TCRαβ Abs. CD19+ B cells and TCRαβ+ T cells were electronically eliminated from the analysis, and the percentage and mean fluorescence intensity of the Ly49 molecules were determined on the gated NK1.1+ cells. Mean fluorescence intensities and SDs for three B6 mice and four to six Syk+ or Syk−/− chimeras are shown. The percentages of NK1.1+ cells expressing Ly49A were 16 ± 4.1 in B6 mice (n = 3), 3.1 ± 1.3 in Syk+ chimeras (n = 6), and 3.6 ± 0.6 in Syk−/− chimeras (n = 6). The percentage of NK1.1+ cells expressing Ly49G2 were 46.3 ± 1.5 in B6 mice (n = 3), 44.1 ± 7.7 in Syk+ chimeras (n = 4), and 31.3 ± 4.5 in Syk−/− chimeras (n = 6). B,Syk+ and Syk−/− A-LAK cells were tested for their ability to distinguish between targets expressing classical and nonclassical MHC molecules. Con A-activated (5 μg/ml for 48 h) B6 (filled circles) and β2m−/− (filled triangles) splenocytes were used as targets. Qa-1b-transfected (F12 cells, empty circles) or control (L cells, empty triangles) fibroblasts were used to test inhibitory receptors of the CD94/NKG2 family.

FIGURE 3.

Inhibitory receptors are normally regulated and function efficiently in the absence of Syk. A, Spleen cells were stained with anti-Ly49A or Ly49G2, anti-NK1.1, and a combination of anti-CD19 and anti-TCRαβ Abs. CD19+ B cells and TCRαβ+ T cells were electronically eliminated from the analysis, and the percentage and mean fluorescence intensity of the Ly49 molecules were determined on the gated NK1.1+ cells. Mean fluorescence intensities and SDs for three B6 mice and four to six Syk+ or Syk−/− chimeras are shown. The percentages of NK1.1+ cells expressing Ly49A were 16 ± 4.1 in B6 mice (n = 3), 3.1 ± 1.3 in Syk+ chimeras (n = 6), and 3.6 ± 0.6 in Syk−/− chimeras (n = 6). The percentage of NK1.1+ cells expressing Ly49G2 were 46.3 ± 1.5 in B6 mice (n = 3), 44.1 ± 7.7 in Syk+ chimeras (n = 4), and 31.3 ± 4.5 in Syk−/− chimeras (n = 6). B,Syk+ and Syk−/− A-LAK cells were tested for their ability to distinguish between targets expressing classical and nonclassical MHC molecules. Con A-activated (5 μg/ml for 48 h) B6 (filled circles) and β2m−/− (filled triangles) splenocytes were used as targets. Qa-1b-transfected (F12 cells, empty circles) or control (L cells, empty triangles) fibroblasts were used to test inhibitory receptors of the CD94/NKG2 family.

Close modal

We further tested the ability of Syk−/− NK cells to distinguish between MHC class I-positive or -negative target cells. We found that both Syk−/− and Syk+ A-LAK cells were able to lyse β2m-deficient blasts, while the presence of class I MHC was able to inhibit lysis by both Syk−/− and Syk+ A-LAK cells (Fig. 3,B and Table II). We also found that Syk−/− A-LAK cells express functional inhibitory receptors for the nonclassical MHC molecule Qa-Ib (Fig. 3 B) and that similar percentages (∼50%) of Syk−/− and Syk+ NK cells stained positive using Qa-Ib tetramers (data not shown). These results demonstrate that inhibitory receptors of the CD94/NKG2 family can signal independently of Syk.

Table II.

A-LAK cytotoxicity in the absence of Syk

Target% Specific 51Cr Releasea
SykSyk−/−
YAC-1 (n = 6b78 ± 8 76 ± 12 
EL4 (n = 4) 10 ± 1 12 ± 5 
EL4+ anti-Thy-1 (n = 4) 69 ± 3 41 ± 12  
P815 (n = 4) 11 ± 1 12 ± 2 
P815+ anti-NKR-P1 (n = 3) 31 ± 5 32 ± 4 
P815+ anti-2B4 (n = 3) 27 ± 2 30 ± 6 
P815+ anti-CD16 (n = 3) 27 ± 2 16 ± 2  
L cells (n = 3) 25 ± 3 26 ± 1 
F12 (n = 3) 9 ± 1 9 ± 1 
Con A-B26 blasts (n = 3) 14 ± 3 15 ± 2 
Con A-β2m−/− blasts (n = 3) 41 ± 11 41 ± 4 
Target% Specific 51Cr Releasea
SykSyk−/−
YAC-1 (n = 6b78 ± 8 76 ± 12 
EL4 (n = 4) 10 ± 1 12 ± 5 
EL4+ anti-Thy-1 (n = 4) 69 ± 3 41 ± 12  
P815 (n = 4) 11 ± 1 12 ± 2 
P815+ anti-NKR-P1 (n = 3) 31 ± 5 32 ± 4 
P815+ anti-2B4 (n = 3) 27 ± 2 30 ± 6 
P815+ anti-CD16 (n = 3) 27 ± 2 16 ± 2  
L cells (n = 3) 25 ± 3 26 ± 1 
F12 (n = 3) 9 ± 1 9 ± 1 
Con A-B26 blasts (n = 3) 14 ± 3 15 ± 2 
Con A-β2m−/− blasts (n = 3) 41 ± 11 41 ± 4 
a

A-LAK cells were incubated for 4 h with 51Cr-labeled targets (in the presence or absence of the indicated mAbs) at an E:T ratio of 9:1. CD16 responses are reduced in Syk−/− NK cells (bold).

b

Number of experiments.

A specific NK cell effector function denoted ADCC is elicited when Ig-coated target cells are recognized by NK cells via their sole FcR, FcγRIII, consisting of CD16 associated with the Fcγ chain. The essential role of Fcγ in ADCC is demonstrated by Fcγ-deficient mice, which fail to express FcγRIII and as such have no demonstrable ADCC activity (34). Syk interacts with the ITAM of Fcγ and is activated during ADCC (35). To address the role of Syk in ADCC, lymphokine-activated NK cells from Syk+ or Syk−/− chimeras were tested for their ability to lyse Ab-coated target cells in vitro. Syk+ NK cells lysed Ab-coated EL-4 thymoma cells with high efficiency (Fig. 4,A and Table II). In contrast, Syk−/− NK cells showed abnormal ADCC responses, corresponding to about 50% of normal (Fig. 4,A and Table II). The defective ADCC was not attributable to an overall decrease in Syk−/− NK cell effector function, as these same lymphokine-activated NK cells efficiently lysed several different targets, including YAC-1, P815, class I-deficient blasts, and L cells (Figs. 3,B and 4, A and B, and Table II). Based on these results, it appears that Syk plays an important role in FcγRIII triggering in NK cells, but that ADCC can proceed in vitro in the absence of Syk.

FIGURE 4.

Signaling through FcRγIII, NKR-P1, and 2B4 in Syk−/− NK cells. A, A-LAK cells where tested for the ability to lyse Ab (anti-Thy-1)-coated EL-4 thymoma targets (▴) or control EL-4 cells (•). The specific lysis was determined at the indicated E:T cell ratios after a standard 4-h 51Cr release assay. ▵, YAC-1 killing. The mean and SD of four independent experiments are shown. B, FcR+ P815 cells were used as targets for rADCC. A-LAK cells were preincubated for 15 min at room temperature with 5 μg/ml of 2B4, 20 μg/ml of NK1.1, 75 μg/ml of CD16 mAbs, or medium alone. A representative of three independent experiments is shown.

FIGURE 4.

Signaling through FcRγIII, NKR-P1, and 2B4 in Syk−/− NK cells. A, A-LAK cells where tested for the ability to lyse Ab (anti-Thy-1)-coated EL-4 thymoma targets (▴) or control EL-4 cells (•). The specific lysis was determined at the indicated E:T cell ratios after a standard 4-h 51Cr release assay. ▵, YAC-1 killing. The mean and SD of four independent experiments are shown. B, FcR+ P815 cells were used as targets for rADCC. A-LAK cells were preincubated for 15 min at room temperature with 5 μg/ml of 2B4, 20 μg/ml of NK1.1, 75 μg/ml of CD16 mAbs, or medium alone. A representative of three independent experiments is shown.

Close modal

Signaling via FcRs can generate either cellular activation or inhibition, where the type of response is determined by the cell type-specific FcR engaged and the intracellular proteins that are assembled following FcR binding (reviewed in Ref. 4). The only FcR expressed by NK cells is FcγRIII, which is required for NK cell-mediated ADCC (34). ADCC has been shown to activate both Syk and ZAP-70 (reviewed in Ref. 17). Although the effects of ZAP-70 deficiency on ADCC activity have not been reported (13, 15), we found that Syk−/− NK cells show abnormal, but demonstrable, ADCC activity, which may reflect the reduced ability of ZAP-70 to subserve this function in the absence of Syk. The partial redundancy of Syk in NK cell ADCC contrasts with the absolute requirement for Syk in FcR-mediated signaling in macrophages and mast cells (36, 37). It will be interesting to further characterize these Syk-independent pathways for FcR activation in NK cells.

Cross-linking of the NKR-P1C molecule (defined by the mAb PK136; anti-NK1.1) on murine NK cells activates NK cell lytic functions and induces IFN-γ production (38). Previous studies have demonstrated that the Fc γ-chain is required for NK1.1-dependent signaling (39). Cross-linking of the 2B4 receptor enhances cytotoxicity, granule exocytosis, and IFN-γ production (40), while CD16 ligation on NK cells can induce IFN-γ production (41). To assess whether Syk is essential in any of these signaling pathways, Syk−/− NK cells were tested for their capacity to mediate rADCC or to produce IFN-γ upon cross-linking of NK1.1, 2B4, or CD16 receptors. As shown in Fig. 4,B, Syk−/− LAK cells mediated rADCC after preincubation with either NK1.1 or 2B4 mAbs. In contrast, while cross-linking of CD16 in Syk−/− NK cells resulted in some rADCC activity, the levels were clearly reduced relative to those in Syk+ LAK cells. Similarly, cross-linking of NK1.1 or 2B4 molecules on purified Syk−/− NK cells resulted in IFN-γ production comparable with that by Syk+ NK cells, while IFN-γ production from Syk−/− NK cells following CD16 ligation was defective, reaching only half the control levels (Fig. 5).

FIGURE 5.

IFN-γ production by purified NK cells. Splenic DX5+ NK cells were FACS sorted and expanded in IL-2 or IL-15. After 6–8 days, 2 × 104 cells were plated in 200 μl of complete medium supplemented with human IL-2 (1000 U/ml), or on plates precoated with 2B4, NK1.1, CD16, or Gr-1 (control) mAbs. Stimulation with IL-12 served as a positive control. After 24 h, the supernatants were collected, and the amount of IFN-γ was quantitated by ELISA. Results are expressed as the mean and SD of three independent experiments.

FIGURE 5.

IFN-γ production by purified NK cells. Splenic DX5+ NK cells were FACS sorted and expanded in IL-2 or IL-15. After 6–8 days, 2 × 104 cells were plated in 200 μl of complete medium supplemented with human IL-2 (1000 U/ml), or on plates precoated with 2B4, NK1.1, CD16, or Gr-1 (control) mAbs. Stimulation with IL-12 served as a positive control. After 24 h, the supernatants were collected, and the amount of IFN-γ was quantitated by ELISA. Results are expressed as the mean and SD of three independent experiments.

Close modal

Based on these results, we conclude that signaling through NK1.1 and 2B4 is intact in the absence of Syk, while signaling through CD16 is less efficient, although present in Syk−/− NK cells. Moreover, the association between NK1.1 and FcγRIII and the subsequent activation of a signaling pathway via this complex does not have an absolute requirement for Syk. The Syk-independent mechanisms of NK1.1 signaling in murine NK cells may also involve ZAP-70 in an analogous fashion to other Fcγ-containing receptors.

The analysis of Syk-deficient NK cells in our novel alymphoid mouse strain demonstrates that this PTK plays a nonessential role in murine NK cell development and is dispensable for crucial NK cell effector functions, including natural cytotoxicity, generation of LAK activity, and cytokine production. Stimulation of a variety of NK surface receptors (including the IL-2/15R, NK1.1, 2B4, Ly49, and CD94/NKG2 complex) resulted in normal responses by Syk−/− NK cells. Although the signaling pathways that permit NK cells to function in the absence of Syk remain unknown, a number of observations suggest that this may be achieved by the Syk-related PTK ZAP-70. We demonstrate that Syk−/− NK cells express ZAP-70 protein. Moreover, previous studies have demonstrated that activation of NK cells can result in engagement of both Syk and ZAP-70 signaling pathways (reviewed in Ref. 17). These observations are consistent with interchangeable roles of Syk and ZAP-70 in NK cell differentiation. The analysis of NK cell development and NK cell effector functions in mice deficient in both Syk and ZAP-70 will allow these hypotheses to be addressed directly.

We thank Dr. P. J. Dyson (London, U.K.) for kindly providing the F12 cells, Drs. O. Acuto (Pasteur Institute, Paris, France) and P. Hoglund (Strasbourg, France) for helpful discussions, and M. Malassis for technical assistance.

1

This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale, the Association pour le Recherche sur le Cancer, the Ligue Nationale Contre le Cancer, the Fondation pour la Recherche Médicale, and the Medicinska Forskningsrådet (to F.C. and J.P.D.) and by the Medical Research Council, U.K. (to V.L.J.T.).

3

Abbreviations used in this paper: PTK, protein tyrosine kinase; ITAM, immunoreceptor tyrosine-based activation motif; BCR, B cell Ag receptor; ADCC, Ab-dependent cell cytotoxicity; rADCC, reverse ADCC; HSC, hemopoietic stem cells; BM, bone marrow; FL, fetal liver; LAK, lymphokine-activated killer cells; A-LAK, adherent LAK; FcR, fragment c receptor; γc, common γ-chain; RAG, recombinase-activating gene; DN, dominant negative.

1
Bolen, J. B., J. S. Brugge.
1997
. Leukocyte protein tyrosine kinases: potential targets for drug discovery.
Annu. Rev. Immunol.
15
:
371
2
Reth, M..
1989
. Antigen receptor tail clue.
Nature
338
:
383
3
Chan, A. C., N. S. van Oers, A. Tran, L. Turka, C. L. Law, J. C. Ryan, E. A. Clark, A. Weiss.
1994
. Differential expression of ZAP-70 and Syk protein tyrosine kinases, and the role of this family of protein tyrosine kinases in TCR signaling.
J. Immunol.
152
:
4758
4
Daeron, M..
1997
. Fc receptor biology.
Annu. Rev. Immunol.
15
:
203
5
Gao, J., K. E. Zoller, M. H. Ginsberg, J. S. Brugge, S. J. Shattil.
1997
. Regulation of the pp72syk protein tyrosine kinase by platelet integrin αIibβ3.
EMBO J.
16
:
6414
6
Minami, Y., Y. Nakagawa, A. Kawahara, T. Miyazaki, K. Sada, H. Yamamura, T. Taniguchi.
1995
. Protein tyrosine kinase Syk is associated with and activated by the IL-2 receptor: possible link with the c-myc induction pathway.
Immunity
2
:
89
7
Di Santo, J. P., H.-R. Rodewald.
1998
. In vivo roles of receptor tyrosine kinases and cytokine receptors in early thymocyte development.
Curr. Opin. Immunol.
10
:
196
8
Turner, M., P. J. Mee, P. S. Costello, O. Williams, A. A. Price, L. P. Duddy, M. T. Furlong, R. L. Geahlen, V. L. Tybulewicz.
1995
. Perinatal lethality and blocked B-cell development in mice lacking the tyrosine kinase Syk.
Nature
378
:
298
9
Cheng, A. M., B. Rowley, W. Pao, A. Hayday, J. B. Bolen, T. Pawson.
1995
. Syk tyrosine kinase required for mouse viability and B-cell development.
Nature
378
:
303
10
Poole, A., J. M. Gibbins, M. Turner, M. J. van Vugt, J. G. van de Winkel, T. Saito, V. L. Tybulewicz, S. P. Watson.
1997
. The Fc receptor γ-chain and the tyrosine kinase Syk are essential for activation of mouse platelets by collagen.
EMBO J.
16
:
2333
11
Turner, M., A. Gulbranson-Judge, M. E. Quinn, A. E. Walters, I. C. M. MacLennan, V. Tybulewicz.
1997
. Syk tyrosine kinase is required for the positive selection of immature B cells into the recirculating B cell pool.
J. Exp. Med.
186
:
2013
12
Mallick-Wood, C. A., W. Pao, A. C. Cheng, J. M. Lewis, S. Kulkarni, J. B. Bolen, B. Rowley, R. T. Tigelaar, T. Pawson, A. C. Hayday.
1996
. Disruption of epithelial gd T cell repertoires by mutation of the Syk tyrosine kinase.
Proc. Natl. Acad. Sci. USA
93
:
9704
13
Negishi, I., N. Motoyama, K. Nakayama, K. Nakayama, S. Senju, S. Hatakeyama, Q. Zhang, A. C. Chan, D. Y. Loh.
1995
. Essential role for ZAP-70 in both positive and negative selection of thymocytes.
Nature
376
:
435
14
Kadlecek, T. A., N. S. C. van Oers, L. Lefrancois, S. Olson, D. Finlay, D. H. Chu, K. Connolly, N. Kileen, A. Weiss.
1998
. Differential requirements for ZAP-70 in TCR signaling and T cell development.
J. Immunol.
161
:
4688
15
Elder, M. E., D. Lin, J. Clever, A. C. Chan, T. J. Hope, A. Weiss, T. G. Parslow.
1994
. Human severe combined immunodeficiency due to a defect in ZAP-70, a T cell tyrosine kinase.
Science
264
:
1596
16
Puzanov, I. J., N. S. Williams, J. Schatzle, P. V. Sivakumar, M. Bennett, V. Kumar.
1997
. Ontogeny of NK cells and the bone marrow microenvironment: where does IL15 fit in?.
Res. Immunol.
148
:
195
17
Leibson, P. J..
1997
. Signal transduction during natural killer activation: inside the mind of a killer.
Immunity
6
:
655
18
Brumbaugh, K. M., B. A. Binstadt, D. D. Billadeau, R. A. Schoon, C. J. Dick, R. M. Ten, P. J. Leibson.
1997
. Functional role for Syk tyrosine kinase in natural killer cell-mediated natural cytotoxicity.
J. Exp. Med.
186
:
1965
19
Shinkai, Y., G. Rathbun, K. P. Lam, E. M. Oltz, V. Stewart, M. Mendelsohn, J. Charron, M. Datta, F. Young, A. M. Stall, et al
1992
. RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement.
Cell
68
:
855
20
Colucci, F., C. Soudais, E. Rosmaraki, L. Vanes, V. L. J. Tybulewicz, J. P. Di Santo.
1999
. Dissecting NK cell development using a novel alymphoid mouse model: investigating the role of c-abl proto-oncogene in murine NK differentiation.
J. Immunol.
162
:
2761
21
Di Santo, J. P., W. Muller, D. Guy-Grand, A. Fischer, K. Rajewsky.
1995
. Lymphoid development in mice with a targeted deletion of the interleukin 2 receptor γ chain.
Proc. Natl. Acad. Sci. USA
92
:
377
22
Guy-Grand, D., J. B. Cuenod, S. M. Malassis, F. Selz, P. Vassalli.
1996
. Complexity of the mouse gut T cell immune system: identification of two distinct natural killer T cell intraepithelial lineages.
Eur. J. Immunol.
26
:
2248
23
Cotteril, L. A., H. J. Stauss, M. M. Millrain, D. J. Pappin, D. Rahaman, B. Canas, P. Chandler, A. Stackpoole, P. J. Robinson, J. P. Dyson.
1997
. Qa-1 interaction and T cell recognition of the Qa-1 determinant modifier peptide.
Eur. J. Immunol.
27
:
2123
24
Mege, D., V. Di Bartolo, V. Germain, L. Tuosto, F. Michel, O. Acuto.
1996
. Mutation of tyrosines 492/493 in the kinase domain of ZAP-70 affects multiple T-cell receptor signaling pathways.
J. Biol. Chem.
271
:
32644
25
Wen, T., L. Zhang, S. K. Kung, T. J. Molina, R. G. Miller, T. Mak.
1995
. Alloskin graft rejection, tumor rejection and natural killer activity in mice lacking p56lck.
Eur. J. Immunol.
25
:
3155
26
van Oers, N. S., B. Lowin-Kropf, D. Finlay, K. Connolly, A. Weiss.
1996
. αβ T cell development is abolished in mice lacking both Lck and Fyn protein tyrosine kinases.
Immunity
5
:
429
27
Jevremovic, D., D. D. Billadeau, R. A. Schoon, C. J. Dick, B. J. Irvin, W. Zhang, L. E. Samelson, R. T. Abraham, P. J. Leibson.
1999
. A role for the adaptor protein LAT in human NK cell-mediated cytotoxicity.
J. Immunol.
162
:
2453
28
Zhang, W., C. L. Sommers, D. N. Burshtyn, C. C. Stebbins, J. B. DeJarnette, R. P. Trible, A. Grinberg, H. C. Tsay, H. M. Jacobs, C. M. Kessler, et al
1999
. Essential role of LAT in T cell development.
Immunity
10
:
323
29
Gong, Q., L. White, R. Johnson, M. White, I. Negishi, M. Thomas, A. C. Chan.
1997
. Restoration of thymocyte development and function in zap-70−/− mice by the Syk protein tyrosine kinase.
Immunity
7
:
369
30
Kong, G. H., J. Y. Bu, T. Kurosaki, A. S. Shaw, A. C. Chan.
1995
. Reconstitution of Syk function by the ZAP-70 protein tyrosine kinase.
Immunity
2
:
485
31
Raulet, D. H., W. Held, I. Correa, J. R. Dorfman, M. F. Wu, L. Corral.
1997
. Specificity, tolerance and developmental regulation of natural killer cells defined by expression of class I-specific Ly49 receptors.
Immunol. Rev.
155
:
41
32
Sentman, C. L. M. Y. Olsson, K. Karre.
1995
. Missing self recognition by natural killer cells in MHC class I transgenic mice: a ‘receptor calibration’ model for how effector cells adapt to self.
Semin. Immunol.
7
:
109
33
Olsson, M. Y. K. Karre, C. L. Sentman.
1995
. Altered phenotype and function of natural killer cells expressing the major histocompatibility complex receptor Ly-49 in mice transgenic for its ligand.
Proc. Natl. Acad. Sci. USA
92
:
1649
34
Takai, T., M. Li, D. Sylvestre, R. Clynes, J. V. Ravetch.
1994
. FcR γ chain deletion results in pleiotropic effector cell defects.
Cell
76
:
519
35
Stahls, A., G. E. Liwszyc, C. Couture, T. Mustelin, L. C. Andersson.
1994
. Triggering of human natural killer cells through CD16 induced tyrosine phosphorylation of p72syk kinase.
Eur. J. Immunol.
24
:
2491
36
Crowley, M. T., P. S. Costello, C. J. Fitzer-Attas, M. Turner, F. Meng, C. Lowell, V. L. Tybulewicz, A. L. De Franco.
1997
. A critical role for Syk in signal transduction and phagocytosis mediated by Fcγ receptors on macrophages.
J. Exp. Med.
186
:
1027
37
Costello, P. S. T., M. Walters, A. E. Cunningham, C. N. Bauer, P. H. Donward, V. L. J. Tybulewicz.
1996
. Critical role for the tyrosine kinase Syk in signaling through the high affinity IgE receptor mast cells.
Oncogene
13
:
2595
38
Arase, H., N. Arase, T. Saito.
1996
. Interferon-γ production by natural killer cells and NK1.1+ T cells upon NKR-P1 cross-linking.
J. Exp. Med.
183
:
2391
39
Arase, N., H. Arase, S. Y. Park, H. Ohno, C. Ra, T. Saito.
1997
. Association with FcRγ is essential for activation signal through NKR-P1 (CD161) in natural killer (NK) cells and NK1.1+ T cells.
J. Exp. Med.
186
:
1957
40
Garni-Wagner, B., A. Purohit, P. A. Mathew, M. Bennet, V. Kumar.
1993
. A novel function-associated molecule related to non-MHC restricted cytotoxicity mediated by activated natural killer cells and T cells.
J. Immunol.
151
:
60
41
Cassatella, M. A., I. Anegon, M. C. Cuturi, P. Griskey, G. Trinchieri, B. Perussia.
1989
. FcγR (CD16) interaction with ligand induces Ca2+ mobilization and phosphoinositide turnover in human natural killer cells. Role of Ca2+ in FcγR (CD16)-induced transcription and expression of lymphokine genes.
J. Exp. Med.
169
:
549