The chemokine, stromal cell-derived factor-1 (SDF-1 /

Transcription

1 Distinct Role of ZAP-70 and Src Homology 2 Domain-Containing Leukocyte Protein of 76 kda in the Prolonged Activation of Extracellular Signal-Regulated Protein Kinase by the Stromal Cell-Derived Factor-1 /CXCL12 Chemokine 1 Kimberly N. Kremer, Troy D. Humphreys, Ashok Kumar, Nan-Xin Qian, and Karen E. Hedin 2 Stimulation of T lymphocytes with the ligand for the CXCR4 chemokine receptor stromal cell-derived factor-1 (SDF-1 / CXCL12), results in prolonged activation of the extracellular signal-regulated kinases (ERK) ERK1 and ERK2. Because SDF-1 is unique among several chemokines in its ability to stimulate prolonged ERK activation, this pathway is thought to mediate special functions of SDF-1 that are not shared with other chemokines. However, the molecular mechanisms of this response are poorly understood. In this study we show that SDF-1 stimulation of prolonged ERK activation in Jurkat T cells requires both the ZAP-70 tyrosine kinase and the Src homology 2 domain-containing leukocyte protein of 76 kda (SLP-76) scaffold protein. This pathway involves ZAP-70-dependent tyrosine phosphorylation of SLP-76 at one or more of its tyrosines, 113, 128, and 145. Because TCR activates ERK via SLP-76-mediated activation of the linker of activated T cells (LAT) scaffold protein, we examined the role of LAT in SDF-1 -mediated ERK activation. However, neither the SLP-76 proline-rich domain that links to GADS and LAT, nor LAT, itself are required for SDF-1 to stimulate SLP-76 tyrosine phosphorylation or to activate ERK. Together, our results describe the distinct mechanism by which SDF-1 stimulates prolonged ERK activation in T cells and indicate that this pathway is specific for cells expressing both ZAP-70 and SLP-76. The Journal of Immunology, 2003, 171: The chemokine, stromal cell-derived factor-1 (SDF-1 / CXCL12) 3 and its receptor, CXCR4, are critical for the regulation of T lymphocyte functions under normal conditions and disease (1). SDF-1 is a member of a superfamily of 50 chemokines that are classified as C, CC, CXC, or CX3C depending on the spacing and number of their conserved cysteine residues (2). Interestingly, CXCR4, the sole known receptor for SDF-1, is not unique to T lymphocytes, but rather is ubiquitously expressed in mammalian cells. In the immune system, SDF-1 modulates T cell development in the thymus (3, 4), T cell migration, and the expression of genes controlling T cell signaling, migration, and survival (5). In addition, SDF-1 costimulates the immune activation of T lymphocytes stimulated by the TCR (6). CXCR4 has also elicited significant attention because it is exploited for HIV-1 infection, a function that is antagonized by SDF-1 (1). Thus, whereas little is known regarding the function Departments of Surgery and Immunology, Mayo Graduate and Medical Schools, Mayo Clinic, Rochester, MN Received for publication June 24, Accepted for publication April 21, The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported in part by the philanthropy of Barbara Lipps to the Mayo Clinic, and by National Institutes of Health Training Grant AI (to T.D.H.) and RO1 Grant GM59763 (to K.E.H). 2 Address correspondence and reprint requests to Dr. Karen E. Hedin, Department of Immunology, Mayo Clinic, Medical Sciences Building 2nd Floor, 200 First Street Southwest, Rochester, MN address: hedin.karen@mayo.edu 3 Abbreviations used in this paper: SDF-1, stromal cell-derived factor-1 ; [Ca 2 ] i, intracellular calcium ion concentration; ERK, extracellular signal-regulated kinase; LAT, linker of activated T cells; SH2, Src homology 2; MAP, mitogen-activated protein; SLP-76, SH2 domain-containing leukocyte protein of 76 kda; MEK, MAP/ ERK kinase. of SDF-1 in most mammalian cells, the role of this chemokine and its receptor in the regulation of T cell homeostasis and viral infection is well established. Unfortunately, the molecular mechanisms that mediate the effects of SDF-1 on T cells are not well understood. In particular, the signaling pathways that permit SDF-1 to regulate T lymphocyte gene expression are incompletely characterized. However, emerging data indicate that chemokine receptors use extracellular signal-regulated kinase (ERK) mitogen-activated protein (MAP) kinase for gene expression. SDF-1 -mediated activation of the ERK MAP kinase pathway is required for protection against serum-deprived apoptosis and for the expression of many genes, including those that inhibit apoptosis (5). In addition, because the ERK pathway also contributes to T lymphocyte immune activation (7) and critically regulates thymocyte development (8), ERK activation is likely involved in the regulation of these processes by SDF-1. In contrast to most chemokines, SDF-1 stimulates ERK activation in T cells for a prolonged period of time (9). Because prolonged rather than transient ERK activation provokes distinct cellular responses downstream of other cell-surface receptors (8, 10 12), it is thought that prolonged SDF-1 -dependent ERK activation may subserve a special biological role. Therefore, in this study we have focused on delineating the molecular mechanisms of SDF-1 -mediated ERK activation in T cells. CXCR4 and other chemokine receptors are members of the G protein-coupled receptor superfamily. Although CXCR4-mediated ERK activation has been extensively studied (9, 13 16), the molecular machinery that links CXCR4 signaling to ERK activation in T cells has not been fully characterized. SDF-1 signaling via CXCR4 uses pertussis toxin-sensitive G i -type G proteins for both migration and ERK activation. SDF-1 -mediated ERK activation Copyright 2003 by The American Association of Immunologists, Inc /03/$02.00

2 has also been shown to require phosphatidylinositol 3-kinase activity (15), Ras (16), and the ERK kinase, MAP/ERK kinase (MEK)-1. In other cell types, tyrosine kinases can participate in G i protein-mediated activation of phosphatidylinositol 3-kinase and the Ras-Raf-MEK-ERK cascade (17). Moreover, both tyrosine kinase- and G i protein-mediated signal transduction often result in increased intracellular calcium ion concentrations ([Ca 2 ] i ), and [Ca 2 ] i is elevated in T cells in response to SDF-1 (2, 9, 18). Although SDF-1 has been shown to use the lymphocyte-specific tyrosine kinase, ZAP-70, for adhesion and migration (19, 20), the role of ZAP-70 in SDF-1 -mediated ERK activation and [Ca 2 ] i signaling has not been addressed. Similarly, beside MEK-1 and ERK themselves, key tyrosine kinase substrates that link CXCR4 signaling with ERK activation remain to be identified. In this study we show that ZAP-70 as well as the tyrosine kinase substrate and scaffold protein, Src homology 2 (SH2) domain-containing leukocyte protein of 76 kda (SLP-76), are required for the distinctive prolonged ERK activation that occurs in response to treatment of Jurkat T cells with SDF-1, but not other chemokines. We further demonstrate that this pathway requires ZAP-70-dependent tyrosine phosphorylation of SLP-76, thereby implicating SH2 domain-containing signaling proteins that bind to this region of SLP-76 in prolonged ERK signaling. However, SDF-1 -dependent ERK activation does not require either the SLP-76 prolinerich region or linker of activated T cells (LAT), domains that were previously shown to mediate ERK activation by SLP-76 downstream of the TCR. In addition to showing the unique roles that ZAP-70 and SLP-76 play in ERK activation, we also found that these two molecules are important for SDF-1 -induced signaling that results in increased [Ca 2 ] i. These findings significantly advance our understanding of the molecular mechanisms underlying SDF-1 effects on T cells and begin to address the mechanisms that distinguish SDF-1 signaling from that of other cell-surface receptors that also activate ERK. Materials and Methods Cells Jurkat T cells and mutant sublines were maintained in medium A (RPMI 1640 supplemented with 5% FCS, 5% calf serum, 10 mm HEPES (ph 7.4), 2mML-glutamine, and 2 M 2-ME) at 10 6 cells/ml. Robert T. Abraham (Duke University, Durham, NC) generously provided Jurkat sublines that are LAT-deficient (called ANJ-3 (21)), ZAP-70-deficient (called P116 (18)), or stably reconstituted with ZAP-70 (called P116-C40 (18)). SLP- 76-deficient Jurkat T cells (J14) were a generous gift from A. Weiss (University of California, San Francisco, CA) (22). Assay for active phosphorylated ERK1 and ERK2 Cells were stimulated for the indicated times at 37 C, pelleted by rapid centrifugation, lysed in SDS-PAGE sample buffer, fractionated on SDS- PAGE, and transferred to Immobilon-P membrane (Millipore, Bedford, MA). Unless otherwise indicated, cells were stimulated with M SDF-1 (R&D Systems, Minneapolis, MN) or M CCL19, doses determined to give maximum ERK responses. Both SDF-1 and CCL19 (macrophage inflammatory protein-3 ) were obtained from R&D Systems. The presence of active ERK1 and ERK2 phosphorylated on threonine 202 and tyrosine 204 was detected by immunoblotting with phospho-specific p44/p42 ERK1 and ERK2 (Thr 202 /Tyr 204 ) antiserum, whereas total ERK2 protein was detected using p44 ERK kinase antiserum (both antisera from New England Biolabs, Beverly, MA). ERK MAP kinase assay Jurkat cells ( /point) were stimulated as above, then lysed and analyzed for ERK1 plus ERK2 MAP kinase activity toward the epidermal growth factor receptor peptide, as described (23, 24). ERK activity measured by this assay is completely inhibited by pretreatment with the MEK-1 inhibitor, PD (data not shown). Transient transfections Cells (10 7 ) were mixed with 3 20 g of the indicated expression plasmid plus irrelevant plasmid DNA (pcdna3.1; Invitrogen, San Diego, CA) to make a total of 30 g, electroporated using a BTX electro-square-porator model T820 (BTX, San Diego, CA), cultured in medium A for h, then harvested and assayed as indicated. Under these conditions transient transfection with a green fluorescent protein expression plasmid routinely yielded high expression by 60 80% of Jurkat T cells. Expression plasmids encoding FLAG-tagged wild-type or mutant SLP-76 proteins were generous gifts from G. A. Koretzky (University of Pennsylvania, Philadelphia, PA (25)). An expression plasmid encoding myc-tagged ZAP-70 was a gift from R. T. Abraham (Duke University, Durham, NC (18)). Expression of transfected proteins was detected by immunoblotting with SLP-76 or ZAP-70 antisera (gifts from P. J. Leibson, Mayo Clinic, Rochester, MN) or anti-myc mab 9E10 (Berkeley Antibody, Richmond, CA). Detection of tyrosine phosphorylated SLP-76 Cells were stimulated with SDF-1 as indicated and lysed (26). Then, either endogenous SLP-76 was immunoprecipitated with SLP-76 antiserum and protein A-Sepharose (Sigma-Aldrich, St. Louis, MO), or FLAGtagged transiently transfected SLP-76 was immunoprecipitated with anti- FLAG agarose (Sigma-Aldrich). Phosphotyrosine incorporation into SLP-76 was detected by immunoblotting with anti-phosphotyrosine mab 4G10 (Upstate Biotechnology, Lake Placid, NY). [Ca 2 ] i analysis 361 Cells were loaded with indo-1 (Molecular Probes, Eugene, OR), then [Ca 2 ] i was assayed by using a FACStar Plus flow cytometer (BD Biosciences, San Jose, CA) to measure the ratio of 405:495 nm fluorescence, as described (27). Stimuli were either M SDF-1 or 500 ng/ml anti-cd3 mab (OKT3) cross-linked with 0.1 mg/ml goat-anti-mouse IgG (ICN Pharmaceuticals, Costa Mesa, CA). Data were analyzed and plotted by using FlowJo software (Tree Star, Palo Alto, CA). Results SDF-1 stimulates prolonged ERK activation in Jurkat T lymphocytes via a mechanism requiring SLP-76 Treatment of Jurkat T cells with SDF-1 results in the accumulation of active phosphorylated ERK1 and ERK2 with high levels of ERK activation extending to at least 10 min (Fig. 1A). To explore the roles of different signaling proteins in SDF-1 -mediated ERK activation in T cells, we used well-characterized somatic mutants of the Jurkat T cell line that are deficient in distinct signaling proteins. Interestingly, SDF-1 evokes only transient ERK activation in J14 cells, which are deficient in the SLP-76 scaffold protein (22) (Fig. 1, A and B). ERK activation at 8 or 10 min was significantly lower in J14 cells compared with the parental Jurkat T cell line (Fig. 1B). In contrast, SDF-1 -dependent ERK activation at 2 min did not significantly differ between J14 and wild-type Jurkat T cells (n 4, p 0.6). Thus, only the later, more prolonged phase of SDF-1 -mediated ERK response is defective in J14 cells. The control experiments demonstrate that the signaling defect in J14 cells results from a deficiency in SLP-76 and not some other mutation, because transient expression of SLP-76, but not vector (pcdna3.1) alone, significantly increased SDF-1 -dependent ERK activation at both 8 and 10 min (Fig. 1, C and D). The deficiency in SDF-1 -dependent ERK activation is not caused by decreased levels of CXCR4 on the surface of J14 cells, because flow cytometric analyses detected no deficiencies in CXCR4 cellsurface expression by J14 cells compared with the parental Jurkat T cell line (data not shown). Therefore, these results demonstrate that SDF-1 stimulates prolonged ERK activation in Jurkat T cells via a mechanism requiring SLP-76. Phosphorylation of SLP-76 is required for prolonged ERK activation in response to SDF-1 We next analyzed the molecular mechanisms that permit SLP-76 to mediate prolonged ERK activation by SDF-1. SLP-76 is a

3 362 SDF-1 USES ZAP-70 AND SLP-76 TO STIMULATE PROLONGED ERK FIGURE 1. SDF-1 stimulates prolonged ERK activation in Jurkat T cells via a mechanism requiring SLP-76. A, Cells were stimulated for the indicated times with M SDF-1 and assayed for active, phosphorylated ERK1 and ERK2 (P-ERK1 and P-ERK2) by immunoblotting (upper gel). The same blot was stripped and immunoblotted with antisera to total ERK2 as a control (lower gel). B, Summary of three independent experiments as in A, where the responses of SLP-76-deficient cells are represented as a percentage of the responses of the parental Jurkat cell line in the same experiment SEM., Significantly different (p 0.05) from Jurkat control (100%). C, SLP-76-deficient J14 cells were transiently transfected with either vector alone (pcdna3.1) or a plasmid encoding wild-type SLP-76 (SLP-76-WT), then assayed for SDF-1 -dependent ERK activation as in A. D, Summary of experiments as in C. The responses of SLP-76-transfected cells are represented as a percentage of the responses of cells transfected with pcdna3 in the same experiment SEM., Significantly different (p 0.05) from pcdna3.1 alone (100%). scaffold protein composed of several domains. TCR signal transduction results in SLP-76 tyrosine phosphorylation at the aminoterminal tyrosines 113, 128, and 145 via a mechanism requiring the ZAP-70 tyrosine kinase. Phosphorylated SLP-76 subsequently binds SH2 domain-containing signaling proteins that contribute to TCR signal transduction (7, 25, 28). Therefore, we asked if SDF-1 signals to ERK via a mechanism involving SLP-76 tyrosine phosphorylation. SDF-1 treatment induced the tyrosine phosphorylation of endogenous SLP-76 in Jurkat T cells (Fig. 2A) as well as SLP-76 introduced into J14 Jurkat T cells by transient transfection (Fig. 2B). Site-directed disruption of the previously described (25, 29) tyrosine phosphorylation sites 113, 128, and 145 of SLP-76 abolished the tyrosine phosphorylation of SLP-76 in response to SDF-1 (SLP-76-3YF; Fig. 2B). These results suggest a role for SLP-76 tyrosine phosphorylation in mediating the ERK response to SDF-1. To test this hypothesis, we asked if SLP-76-3YF could rescue ERK activation in response to SDF-1 treatment of J14 cells. However, SLP-76-3YF failed to rescue ERK activation, despite being expressed at equivalent or higher levels than wild-type SLP-76 that did rescue prolonged ERK activation in the J14 cells (Fig. 3A). In multiple experiments, the SDF-1 -dependent prolonged ERK activation of J14 cells transfected with SLP-76-3YF was significantly different from that of J14 cells transfected with SLP-76-WT (n 4, p 0.05) (Fig. 3B) and was not significantly different from J14 cells transfected with pcdna3 alone (n 4, p 0.4). Together, the results in this section indicate that one or more of the previously described amino-terminal tyrosine phosphorylation sites of SLP-76 are required for prolonged ERK activation in response to SDF-1. SDF-1 stimulates tyrosine phosphorylation of SLP-76 via a mechanism requiring ZAP-70 Tyrosines 113, 128, and 145 of SLP-76 are targets of the ZAP-70 tyrosine kinase in the TCR signaling pathway (25, 29). We therefore used a ZAP-70-deficient somatic mutant of the Jurkat T cell line (P116 (18)) to determine whether ZAP-70 is required for SLP-76 tyrosine phosphorylation in response to SDF-1. SDF-1 treatment of P116 cells did not result in SLP-76 tyrosine phosphorylation (Fig. 4A). SDF-1 -dependent tyrosine phosphorylation of SLP-76 was rescued by transient expression of ZAP-70 (Fig. 4A). Furthermore, stable expression of ZAP-70 (creating the P116-C40 cell line (18)) also rescued SDF-1 -dependent tyrosine phosphorylation of SLP-76 (Fig. 4B). The deficiency in SDF-1 dependent SLP-76 tyrosine phosphorylation is not caused by decreased levels of CXCR4 on the surface of P116 cells, because P116 cells express normal cell-surface levels of CXCR4 (data not

4 363 FIGURE 2. One or more amino-terminal tyrosines of SLP-76 are required for SDF-1 to stimulate SLP-76 tyrosine phosphorylation. A, Jurkat T cells were stimulated with SDF-1 for 10 min, and endogenous SLP-76 was immunoprecipitated and immunoblotted with anti-phosphotyrosine mab or SLP-76 antiserum. B, SLP-76-deficient (J14) cells were transiently transfected with plasmids encoding either wild-type (SLP-76-WT) or mutant (SLP-76-3YF) SLP-76 and stimulated with SDF-1 for 8 min. SLP-76 proteins were immunoprecipitated and analyzed by immunoblotting with anti-phosphotyrosine mab or SLP-76 antiserum. Results in A and B are each representative of three separate experiments. shown). These results strongly support a role for ZAP-70 in the SDF-1 -dependent tyrosine phosphorylation of SLP-76. Similar to SLP-76, ZAP-70 is required for SDF-1 to stimulate prolonged ERK activation Because phosphorylation of SLP-76 is required for prolonged ERK activation in response to SDF-1, and because SLP-76 tyrosine phosphorylation correlates with prolonged ERK activation in response to this chemokine, we examined the role of ZAP-70 in SDF-1 -dependent ERK regulation. Interestingly, the ZAP-70-deficient P116 cell line exhibits a defect in SDF-1 -stimulated ERK activation that closely resembles the ERK activation defect of SLP-76-deficient J14 cells (Fig. 5, A and B, compared with Fig. 1, A and B). ERK activation at 8 or 10 min was significantly lower in P116 cells compared with the parental Jurkat T cell line. Only the later phase of ERK activation requires ZAP-70, as ERK activation at 2 min did not significantly differ between P116 and Jurkat cells (n 6, p 0.1)(Fig. 5B). Thus, as for the SLP-76-deficient J14 cells, only the prolonged phase of SDF-1 -dependent ERK activation is deficient in P116 cells. Genetic reconstitution experiments show that the signaling defect in P116 cells results from a deficiency in ZAP-70, as stable re-expression of ZAP-70 in P116 cells (P116-C40 cells) restored prolonged ERK activation in response to SDF-1 (Fig. 5, A and B). Stable ZAP-70 expression in P116 cells restored SDF-1 -dependent ERK activation at 8 min to levels indistinguishable from those of the parental Jurkat T cell line (n 3, p 0.8), a significant increase from ERK activation at 8 min of the ZAP-70-deficient P116 cells (n 3, p 0.05). Together, the results in Figs. 4 and 5 indicate that ZAP-70 is required for SDF-1 to stimulate both SLP-76 tyrosine phosphorylation and prolonged ERK activation. FIGURE 3. One or more amino-terminal tyrosines of SLP-76 are required for prolonged ERK activation in response to SDF-1. A, SLP- 76-deficient J14 cells were transiently transfected with either wild-type (SLP-76-WT) or mutant (SLP-76-3YF) SLP-76, then assayed for SDF-1 dependent ERK activation as in Fig. 1A. B, A summary of four independent experiments as in A, where the responses of SLP-76-3YF-transfected cells are represented as a percentage of the responses of cells transfected with SLP-76-WT in the same experiment SEM., Significantly different from 100% (p 0.05). SLP-76 and ZAP-70 are required for [Ca 2 ] i mobilization in response to SDF-1 In addition to ERK, both SLP-76 and ZAP-70 have been shown to participate in signaling that results in increased [Ca 2 ] i (22, 21). Therefore, we asked if these proteins participate in [Ca 2 ] i mobilization in response to SDF-1. Fig. 6 shows the results of stimulating cells and assessing [Ca 2 ] i responses by using a flow cytometer to measure changes in intracellular indo-1 fluorescence. SDF-1 treatment induced a [Ca 2 ] i response in wild-type Jurkat T cells. In contrast, SDF-1 -dependent [Ca 2 ] i responses were defective in both the SLP-76-deficient J14 and the ZAP-70-deficient P116 cell lines. As previously described, TCR/CD3-induced [Ca 2 ] i responses were also deficient in J14 and P116 cells (18, 22). Thus, in addition to being important for SDF-1 -induced ERK activation, both SLP-76 and ZAP-70 are also required for SDF-1 -induced mobilization of [Ca 2 ] i. Prolonged ERK activation in response to SDF-1 does not require the SLP-76 GADS-LAT binding domain or LAT In the TCR signaling pathway, SLP-76 binds to GADS-LAT via its proline-rich domain (30). Because LAT could provide a link to Ras and ERK activation (21, 31), we asked if SLP-76 stimulates prolonged ERK activation via LAT. Therefore, we used a SLP-76 deletion mutant lacking the proline-rich domain (SLP ) that was previously shown to be unable to bind GADS-LAT (30). Transient expression of SLP in J14 cells rescued SDF-1 -mediated ERK activation in a manner similar to

5 364 SDF-1 USES ZAP-70 AND SLP-76 TO STIMULATE PROLONGED ERK FIGURE 4. SDF-1 stimulates tyrosine phosphorylation of SLP-76 via a mechanism requiring ZAP-70. A, ZAP-70-deficient (P116) cells were transiently transfected with plasmids encoding FLAG-SLP-76 myc- ZAP-70, then stimulated with SDF-1 for 8 min. FLAG-SLP-76 was immunoprecipitated and immunoblotted with anti-phosphotyrosine mab, SLP-76 antiserum, or anti-myc mab. B, Tyrosine phosphorylation of endogenous SLP-76 in Jurkat, ZAP-70-deficient cells (P116), and P116 cells stably reconstituted with ZAP-70 (P116-C40). Results are representative of three independent experiments. wild-type SLP-76 (Fig. 7, A and B)(p 0.6). Consistent with this result, SLP was also inducibly phosphorylated in response to SDF-1 (Fig. 7C, upper gel); moreover, phosphorylation was approximately proportionate to protein expression levels (Fig. 7C, lower gel). Thus, the domain of SLP-76 that binds to GADS- LAT is not required for SDF-1 to inducibly phosphorylate SLP-76 and stimulate prolonged ERK activation. To confirm that SDF-1 stimulates prolonged ERK activation via a mechanism independent of SLP-76 binding to GADS-LAT, we assayed ERK activation in a LAT-deficient Jurkat T cell line (ANJ-3). The ANJ-3 LAT-deficient Jurkat T cell line exhibited no defect in SDF-1 stimulation of prolonged ERK activation (Fig. 7D). ERK activation in ANJ-3 was not significantly different from ERK activation in the parental Jurkat cell line at 2, 8, or 10 min (n 3; p 0.4, 0.3, and 0.4 at 2, 8, and 10 min, respectively). Thus, LAT is not required for SDF-1 to stimulate prolonged ERK activation. Therefore, the pathway by which SLP-76 mediates prolonged ERK activation in response to SDF-1 differs from previously described SLP-76 signaling downstream of the TCR, which involves both the SLP-76 GADS-LAT binding domain and LAT. Discussion Chemokines are emerging as pleiotropic regulators of immune cell functions, acting as critical regulators of migration and gene expression important for the control of cell division, death, and differentiation, and for the synthesis and secretion of cytokines. For example, in addition to stimulating chemotaxis, CXCR4 signaling in T cells induces the expression of multiple genes, including those that inhibit apoptosis (5), and contributes to T cell proliferation and cytokine gene expression following TCR stimulation (6). As with FIGURE 5. ZAP-70 is required for SDF-1 to stimulate prolonged ERK activation. A, Jurkat, ZAP-70-deficient (P116) cells, or P116 cells stably reconstituted with ZAP-70 (P116-C40) were stimulated for the indicated times with SDF-1, lysed in SDS sample buffer, and assayed for active, phosphorylated ERK1 and ERK2 as in Fig. 1A. B, A summary of three independent experiments as in A, comparing the SDF-1 -dependent ERK activation of the indicated cell lines. The levels of ZAP-70 expressed by P116-C40 cells are approximately equivalent to those expressed by the parental Jurkat cell line (18). The responses of ZAP-70-deficient P116 cells and ZAP-70-reconstituted P116-C40 cells are represented as a percentage of the responses of the parental Jurkat cell line in the same experiment SEM., Significantly different from 100% (p 0.05). growth factor receptors, activation of the ERK MAP kinase pathway is important for gene regulation by CXCR4 (5). Other receptors use a large number of distinctive molecular mechanisms for FIGURE 6. SLP-76 and ZAP-70 are required for [Ca 2 ] i mobilization in response to SDF-1. The indicated cell lines were loaded with indo-1, then [Ca 2 ] i was assayed by using a flow cytometer to measure the ratio of 405:495 nm fluorescence (which correlates with [Ca 2 ] i ). Cells were stimulated at the times indicated by the arrows with either M SDF-1 (top) or0.5 g/ml anti-cd3 mab cross-linked with goat-anti-mouse IgG (bottom).

6 365 FIGURE 7. SDF-1 signaling via SLP-76 does not require the SLP-76 region that links to GADS-LAT. A, SLP-76-deficient J14 cells were transiently transfected with wild-type (SLP-76-WT) or mutant (SLP ) SLP-76 and then stimulated with SDF-1. Active ERK, total ERK2, and SLP-76 were assayed by immunoblotting as in Fig. 1A. B, A summary of 4 8 independent experiments as in A, where the responses of SLP-76- or SLP transfected cells are represented as a percentage of the responses of cells transfected with pcdna3 alone in the same experiment SEM., Significantly different from 100% (p 0.05). C, SLP-76-deficient J14 cells were transiently transfected with plasmids encoding either SLP-76-WT or SLP Following SDF-1 stimulation, SLP-76 proteins were immunoprecipitated and analyzed by immunoblotting with anti-phosphotyrosine mab or SLP-76 antiserum. Upper gel, anti-phosphotyrosine immunoblot showing inducible tyrosine phosphorylation of SLP in response to SDF-1. Lower gel, SLP-76 immunoblot showing protein expression levels. The asterisk indicates a cross-reacting band that is apparent just below SLP-76 when this particular batch of SLP-76 antiserum is used. This band is also seen in J14 cells that are either untransfected or transfected with a control plasmid, and is therefore not SLP-76. D, Cells were stimulated for the indicated times with M SDF-1 and assayed for active, phosphorylated ERK1 and ERK2 (P-ERK1 and P-ERK2) as in Fig. 1A. ERK activation, a property that permits selective regulation and cross-regulation of this important gene-regulatory pathway (17). However, the pathways leading to ERK activation downstream of chemokine receptors are incompletely characterized. In this study we analyze the molecular mechanisms of CXCR4-mediated ERK activation in T cells, focusing particularly on the mechanism by which the CXCR4 ligand, SDF-1, but not other chemokines (9), stimulates prolonged ERK activation in these cells. Using Jurkat T cell lines deficient in different signaling proteins, we show that normal prolonged ERK activation by the SDF-1 chemokine requires both the tyrosine kinase, ZAP-70, and the tyrosine kinase substrate and scaffold protein, SLP-76. We further show that these molecules function in a signaling pathway wherein SDF-1 treatment stimulates ZAP-70-dependent tyrosine phosphorylation of SLP-76, and subsequently, tyrosine phosphorylated-slp-76 mediates prolonged ERK activation in response to SDF-1. These results strongly suggest that SLP-76 mediates SDF-1 signaling by binding to SH2 domain-containing signaling molecules. When phosphorylated, SLP-76 tyrosines 113, 128, and/or 145 recruit multiple signaling proteins, including the Rac/ CDC42 guanine-nucleotide exchange factor, Vav-1 (25, 28), the adaptor, Nck (32), and the kinase, Itk (33, 34). Therefore, Vav-1, Nck, Itk, and/or other proteins binding to this region of SLP-76 may mediate ERK activation in response to CXCR4 signaling. Because we also found that both ZAP-70 and SLP-76 are required for normal [Ca 2 ] i mobilization following treatment with SDF-1, it is likely that SLP-76 binding proteins also function in other pathways stimulated by SDF-1. We found that both ZAP-70 and SLP-76 were required for CXCR4-mediated ERK activation. Therefore, we initially hypothesized that CXCR4 activates ERK in a manner similar to the TCR. The GADS-LAT binding region of SLP-76 plays a role in TCRmediated ERK activation (35), and TCR-mediated ERK activation requires LAT (21, 31). However, in contrast to TCR signaling, we found that neither the SLP-76 GADS-LAT binding region nor LAT itself are necessary for SDF-1 to stimulate SLP-76 tyrosine phosphorylation and to activate ERK. Therefore, our results indicate the existence of a new mechanism by which ZAP-70 and SLP-76 collaborate in mediating ERK activation downstream of a cell-surface receptor, in addition to characterizing the previously unknown roles of ZAP-70 and SLP-76 in CXCR4-mediated ERK activation. Our results provide a molecular explanation for the distinctive prolonged ERK activation that results from treating T cells with SDF-1 (9). SDF-1 stimulates only transient ERK activation in Jurkat T cells in the absence of either ZAP-70 or SLP-76, a result

7 366 SDF-1 USES ZAP-70 AND SLP-76 TO STIMULATE PROLONGED ERK indicating that both ZAP-70 and SLP-76 are required for prolonged ERK activation in response to SDF-1. Therefore, we asked if SLP-76 is also required for transient ERK activation by other chemokines. However, CCL19, which induces only transient ERK activation in Jurkat T cells (9), signals normally to ERK in both SLP-76-deficient and ZAP-70-deficient Jurkat T cells (data not shown). Thus, use of ZAP-70 and SLP-76 for prolonged ERK activation is a characteristic of SDF-1 but not CCL19. Moreover, our results indicate that prolonged ERK activation by SDF-1 results from CXCR4 using the ZAP-70 and SLP-76 signaling pathway described above, in contrast to CCL19, which does not use this pathway. In addition to CCL19, CCL2, CCL4, CCL20, and CXCL10 stimulate only transient ERK activation in T cells (9). Therefore, prolonged ERK activation in response to SDF-1 seems likely to distinguish effects of SDF-1 from those of other chemokines. In PC12 cells, nerve growth factor induces prolonged ERK activation, which in turn leads to the expression of specific genes that provoke neuronal differentiation (10 12). Moreover, a recent report suggests that positive selection during T cell development in the thymus requires prolonged rather than transient ERK activation (8). Therefore, SDF-1 may use prolonged ERK activation via ZAP-70 and SLP-76 as a way to modulate T cell development in the thymus (3, 4). CXCR4 is expressed by many different cell types in addition to T cells, and the developmental defects of the vasculature, heart, brain, and hematopoietic system in mice deficient in either SDF-1 or CXCR4 indicate a critical requirement for SDF-1 /CXCR4 signaling in multiple cell types (1). Our results identify one mechanism that is apparently used by SDF-1 to coordinate the actions of many different CXCR4-expressing cell types: the use of a tissue-specific signaling pathway. SLP-76 is selectively expressed in platelets (36) and T-lymphoid, monocyte, and granulocyte lineages (37, 38), whereas ZAP-70 expression is restricted to T cells and NK cells (39), and the related Syk kinase is restricted to hematopoietic cell types (40). Therefore, CXCR4 regulation of ERK and [Ca 2 ] i signaling via SLP-76 may underlie the distinctive effects of SDF-1 on certain hematopoietic cell types compared with other cell types that express CXCR4. Similarly, impaired CXCR4 signaling via SLP-76 may contribute to defects in developing T cells, mast cells, and platelets that are seen in SLP-76-deficient mice (14, 36, 41, 42), because all three cell types express CXCR4 and ZAP-70 and/or the related Syk kinase (3, 43, 44). In summary, this study demonstrates that stimulation of prolonged, as opposed to transient, ERK activation by SDF-1 / CXCR4 signaling requires phosphorylation of the ZAP-70 phosphorylation sites at the amino-terminus of SLP-76. Other previously described protein-binding regions of SLP-76, including the SLP domain that binds phospholipase C- 1 (35) and the SH2 domain that binds the SLAP-130/Fyb/ADAP scaffold protein (45, 46), may also contribute to SDF-1 -dependent ERK activation in T cells. However, in contrast to the TCR, which also uses SLP-76 for signaling, SDF-1 -mediated ERK activation is independent of the scaffold function of SLP-76 as it specifically relates to the GADS-LAT binding domain. Together, these results provide a molecular explanation for the prolonged ERK activation that is characteristic of SDF-1 /CXCR4 signaling in T cells and indicate that this pathway is restricted to certain cell types. Acknowledgments We are grateful to Drs. G. A. Koretzky, P. J. Leibson, and R. T. Abraham for providing critical reagents and cell lines. References 1. Murdoch, C CXCR4: chemokine receptor extraordinaire. Immunol. Rev. 177: Thelen, M Dancing to the tune of chemokines. Nat. Immunol. 2: Suzuki, G., Y. Nakata, Y. Dan, A. Uzawa, K. Nakagawa, T. Saito, K. Mita, and T. Shirasawa Loss of SDF-1 receptor expression during positive selection in the thymus. Int. Immunol. 10: Onai, N., Y. Zhange, H. Yoneyama, T. Kitamura, S. Ishikawa, and K. Matsushima Impairment of lymphopoiesis and myelopoiesis in mice reconstituted with bone marrow-hematopoietic progenitor cells expressing SDF-1-intrakine. Blood 96: Suzuki, Y., M. Rahman, and H. Mitsuya Diverse transcriptional response of CD4 T cells to stromal cell-derived factor (SDF)-1: cell survival promotion and priming effects of SDF-1 on CD4 T cells. J. Immunol. 167: Nanki, T., and P. E. Lipsky Cutting edge: stromal cell-derived factor-1 is a costimulator for CD4 T cell activation. J. Immunol. 164: Koretzky, G. A., and P. S. Myung Positive and negative regulation of T-cell activation by adaptor proteins. Nat. Rev. Immunol. 1: Welen, G., B. Hausmann, and E. Palmer A motif in the T-cell receptor controls positive selection by modulating ERK activity. Nature 406: Tilton, B., L. Ho, E. Oberlin, P. Loetscher, F. Baleux, I. Clark-Lewis, and M. Thelen Signal transduction by CXC chemokine receptor 4: stromal cell-derived factor 1 stimulates prolonged protein kinase B and extracellular signal-regulated kinase 2 activation in T lymphocytes. J. Exp. Med. 192: Harada, T., T. Morooka, S. Ogawa, and E. Nishida ERK induces p35, a neuron-specific activator of CDK5, through induction of Egr1. Nat. Cell Biol. 3: Woods, D., H. Cherwinski, E. Venetsanakos, A. Bhat, S. Gysin, M. Humbert, P. F. Bray, V. L. Saylor, and M. McMahon Induction of 3 -integrin gene expression by sustained activation of the Ras-regulated Raf-MEK-extracellular signal regulated kinase signaling pathway. Mol. Cell. Biol. 21: Muller, J., A. M. Cacace, W. E. Lyons, C. B. McGill, and D. K. Morrison Identification of B-KSR1, a novel brain-specific isoform of KSR1 that functions in neuronal signaling. Mol. Cell. Biol. 20: Ganju, R. K., S. A. Brubaker, J. Meyer, P. Dutt, Y. Yang, S. Zin, W. Newman, and J. E. Groopman The -chemokine, stromal cell-derived factor-1, binds to the transmembrane G protein coupled CXCR4 receptor and activates multiple signal transduction pathways. J. Biol. Chem. 273: Vicente-Manzanares, M., M. Rey, D. R. Jones, D. Sancho, M. Mellado, J. M. Rodriguez-Frade, M. A. del Pozo, M. Yanez-Mo, A. M. de Ana, C. Martinez-A. et al Involvement of phosphatidylinositol 3-kinase in stromal cell-cerived factor-1 -induced lymphocyte polarization and chemotaxis. J. Immunol. 163: Sotsios, Y., G. C. Whittaker, J. Westwick, and S. G. Ward The CXC chemokine stromal cell-derived factor activates a G i -coupled phosphoinositide 3-kinase in T lymphocytes. J. Immunol. 163: Weber, K. S. C., G. Ostermann, A. Zernecke, A. Schroder, L. B. Klickstein, and C. Weber Dual role of H-Ras in regulation of lymphocyte function antigen-1 activity by stromal cell-derived factor-1 : implications for leukocyte transmigration. Mol. Biol. Cell 12: Marinissen, M. J., and J. S. Gutkind G protein-coupled receptors and signaling networks: emerging paradigms. Trends Pharmacol. Sci. 22: Williams, B. L., B. J. Irvin, S. L. Sutor, C. C. S. Chini, E. Yacyshyn, J. B. Wardenburg, M. Dalton, A. C. Chan, and R. T. Abraham Phosphorylation of Tyr319 in ZAP-70 is required for T-cell antigen receptor-dependent phospholipase C- 1 and Ras activation. EMBO J. 18: Soede, R. D. M., Y. M. Wijnands, I. V. Kouteren-Cobzaru, and E. Roos ZAP-70 tyrosine kinase is required for LFA-1-dependent T cell migration. J. Cell Biol. 142: Ottoson, N. C., J. T. Pribila, A. S. H. Chan, and Y. Shimizu Cutting edge: T cell migration regulated by CXCR4 chemokine receptor signaling to ZAP-70 tyrosine kinase. J. Immunol. 167: Zhang, W., B. J. Irvin, R. P. Trible, R. T. Abraham, and L. E. Samelson Functional analysis of LAT in TCR-mediated signaling pathways using a LATdeficient Jurkat cell line. Int. Immunol. 11: Yablonski, D., M. R. Kuhne, T. Kadlecek, and A. Weiss Uncoupling of nonreceptor tyrosine kinases from PLC- 1 in an SLP-76-deficient T cell. Science 281: Mitchell, F. M., L. E. Heasley, N. X. Qian, J. Zamarripa, and G. L. Johnson Differential modulation of bombesin-stimulated phospholipase C and mitogen-activated protein kinase activity by [D-Arg 1,D-Phe 5,D-Trp 7,9,Leu 11 ] substance P. J. Biol. Chem. 270: Gardner, A. M., C. A. Lange-Carter, R. R. Vaillancourt, and G. L. Johnson Measuring activation of kinases in mitogen-activated protein kinase regulatory network. Methods Enzymol. 238: Fang, N., D. G. Motto, S. E. Ross, and G. A. Koretzky Tyrosines 113, 128, and 145 of SLP-76 are required for optimal augmentation of NFAT promoter activity. J. Immunol. 157: Kohn, A. D., K. S. Kovacina, and R. A. Roth Insulin stimulates the kinase activity of Rac-PK, a pleckstrin homology domain containing Ser/Thr kinase. EMBO J. 14: Leibson, P. J., D. E. Midthun, K. P. Windebank, and R. T. Abraham Transmembrane signaling during natural killer cell-mediated cytotoxicity: regulation by protein kinase C activation. J. Immunol. 145: Fang, N., and G. A. Koretzky SLP-76 and Vav function in separate, but overlapping pathways to augment interleukin-2 promoter activity. J. Biol. Chem. 274:16206.

8 Wardenburg, J. B., C. Fu, J. K. Jackman, H. Flotow, S. E. Wilkinson, D. H. Williams, R. Johnson, G. Kong, A. C. Chan, and P. R. Findell Phosphorylation of SLP-76 by the ZAP-70 protein-tyrosine kinase is required for T cell receptor function. J. Biol. Chem. 271: Liu, S. K., N. Fang, G. A. Koretzky, and C. J. McGlade The hematopoietic-specific adaptor protein Gads functions in T cell signaling via interactions with the SLP-76 and LAT adaptors. Curr. Biol. 9: Finco, T. S., T. Kadlecek, W. Zhang, L. E. Samelson, and A. Weiss LAT is required for TCR-mediated activation of PLC 1 and the Ras pathway. Immunity 9: Bubeck Wardenburg, J., R. Pappu, J. Y. Bu, B. Mayer, J. Chernoff, D. Straus, and A. C. Chan Regulation of PAK activation and the T cell cytoskeleton by the linker protein, SLP-76. Immunity 9: Bunnell, S. C., M. Diehn, M. B. Yaffe, P. R. Findell, L. C. Cantley, and L. J. Berg Biochemical interactions integrating Itk with the T cell receptor-initiated signaling cascade. J. Biol. Chem. 175: Su, Y. W., Y. Zhang, J. Schweikert, G. A. Koretzky, M. Reth, and J. Wienands Interaction of SLP adaptors with the SH2 domain of tec family kinases. Eur. J. Immunol. 29: Yablonski, D., T. Kadlecek, and A. Weiss Identification of a phospholipase C- 1 (PLC- 1) SH3 domain-binding site in SLP-76 required for T cell receptor-mediated activation of PLC- 1 and NFAT. Mol. Cell. Biol. 21: Judd, B. A., P. S. Myung, L. Leng, A. Obergfell, W. S. Pear, S. J. Shattil, and G. A. Koretzky Hematopoietic reconstitution of SLP-76 corrects hemostasis and platelet signaling through IIb 3 and collagen receptors. Proc. Natl. Acad. Sci. USA 97: Clements, J., S. E. Ross-Barta, L. T. Tygrett, T. J. Waldschmidt, and G. A. Koretzky SLP-76 expression is restricted to hemopoietic cells of monocyte, granulocyte, and T lymphocyte lineage and is regulated during T cell maturation and activation. J. Immunol. 161: Jackman, J. K., D. G. Motto, Q. Sun, M. Tanemoto, C. W. Turck, G. A. Peltz, G. A. Koretzky, and P. R. Findell Molecular cloning of SLP-76, a 76-kD tyrosine phosphoprotein associated with Grb2 in T cells. J. Biol. Chem. 270: Wange, R. L., A. N. T. Kong, and L. E. Samelson A tyrosine-phosphorylated 70-kDa protein binds a photoaffinity analogue of ATP and associates with both the -chain and CD3 components of the activated T cell antigen receptor. J. Biol. Chem. 267: Turner, M., E. Schweighoffer, F. Colucci, J. P. Di Santo, and V. L. Tybulewicz Tyrosine kinase Syk: essential functions for immunoreceptor signalling. Immunol. Today 21: Clements, J. L., B. Yang, S. E. Ross-Barta, S. L. Eliason, R. F. Hrstka, R. A. Williamson, and G. A. Koretzky Requirement for the leukocytespecific adapter protein SLP-76 for normal T cell development. Science 281: Pivniouk, V., E. Tsitsikov, P. Swinton, G. Rathbun, F. W. Alt, and R. S. Geha Impaired viability and profound block in thymocyte development in mice lacking the adaptor protein SLP-76. Cell 94: Lin, T. J., T. B. Issekutz, and J. S. Marshall Human mast cells transmigrate through human umbilical vein endothelial monolayers and selectively produce IL-8 in response to stromal cell-derived factor-1. J. Immunol. 165: Abi-Younes, S., A. Sauty, F. Mach, G. K. Sukhova, P. Libby, and A. D. Luster The stromal cell-derived factor-1 chemokine is a potent platelet agonist highly expressed in atherosclerotic plaques. Circ. Res. 86: Peterson, E. J., M. L. Woods, S. A. Dmowski, G. Derimanov, M. S. Jordan, J. N. Wu, P. S. Myung, Q. H. Liu, J. T. Pribila, B. D. Freedman, et al Coupling of the TCR to integrin activation by SLAP-13-/Fyb. Science 293: Griffiths, E. K., C. Krawczyk, Y. Y. Kong, M. Raab, S. J. Hyduk, D. Bouchard, V. S. Chan, I. Kozieradzki, A. J. Oliveira-dos-Santos, A. Wakeham, et al Positive regulation of T cell activation and integrin adhesion by the adapter Fyb/ SLAP. Science 293:2260.