INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Naïve, mature T lymphocytes undergo proliferation and acquisition of effector function when they encounter antigen on the surface of specialized antigen-presenting cells that also express required co-stimulatory surface proteins. Our understanding of the molecular signaling pathways that are initiated by these stimulatory events has advanced considerably over the past decade, largely as a result of studies in transformed T cell lines in which T cell antigen receptor (TCR)¹ engagement has been mimicked by agents such as lectins and anti-TCR antibodies, which cause aggregation of the TCR (1-3). Oxidants such as H₂O₂, pervanadate, and ultraviolet light have been found to mimic the intracellular signals initiated by TCR aggregation and have also been used to study this signaling pathway (4).

The earliest detectable event upon TCR engagement is the tyrosine phosphorylation of a number of substrates involved in downstream signaling events. These include TCR and the , , and  chains of CD3, ZAP-70, Itk, phospholipase C1, p95^Vav, SLP-76, c-Cbl, p36^Lat as well as others (1, 2). The accumulation of these tyrosine-phosphorylated proteins is the result of a shift in the net balance of competing protein-tyrosine kinases (PTK) and protein tyrosine phosphatase activities and results in part from the activation of several PTKs, including members of the Src, Syk/ZAP-70, and Btk/Itk families of PTKs (1, 3). Three PTKs in particular, Fyn, Lck, and ZAP-70, have been extensively studied, and their involvement in TCR signaling has been demonstrated by a number of different approaches (3).

The cascade of early TCR-proximal signaling events is initiated by the phosphorylation of conserved tyrosine residues within immunoreceptor tyrosine-based activation motifs (ITAMs) present within the CD3 and TCR chains. This event requires the activity of a src-family PTK (5, 6). Phosphorylated ITAMs act as high affinity binding sites for certain SH2 domain-containing proteins, including ZAP-70, which binds to ITAMs via its two SH2 domains (5, 7, 8). Recruitment of ZAP-70 to the TCR is required for the subsequent tyrosine phosphorylation and activation of ZAP-70 (9, 10), which requires the activity of a heterologous kinase, presumably Lck (11-13). ZAP-70 appears to function both as a kinase, phosphorylating downstream signaling proteins such as SLP-76 (14, 15), p36^LAT,² and tubulin (16) and as a scaffolding protein by binding particular SH2 domain-containing proteins by virtue of key sites of tyrosine phosphorylation within ZAP-70 (17). To date, biochemical analyses have identified a number of proteins that can bind to tyrosine-phosphorylated ZAP-70, including c-cbl, p95^Vav, p59^fyn, p56^lck, SHP-1, c-abl, and Ras-GAP (1-3).

These early TCR-proximal signaling events lead to the activation of several other signaling molecules, including phospholipase C1, protein kinase C, and the low molecular weight G protein Ras (2). Together, these signals combine to activate multiple transcription factors that contribute toward the production of the essential autocrine growth factor interleukin 2. Activated, GTP-bound Ras contributes to this process by activating the Raf-1/mitogen-activated protein kinase kinase/Erk signaling pathway, which is required for the production and activation of the critical transcription factor AP-1 (2). The importance of the signaling pathway between the TCR and Ras is underscored by the discovery that anergy is the result of a block in this pathway (18, 19).

How the activation of the TCR-proximal protein-tyrosine kinases causes activation of the Ras/Raf-1/mitogen-activated protein kinase kinase/Erk pathway remains an open question, even though their importance in this pathway was established by studies demonstrating that PTK inhibitors could block TCR-initiated activation of this pathway (20). A role for ZAP-70 has also been suggested on the basis of studies in which overexpression of dominant-negative ZAP-70 in Jurkat T cells could block Erk activation and activation of the interleukin 2 promoter in response to TCR cross-linking (10). SLP-76, which is an in vivo substrate of ZAP-70 (14, 15, 21, 22), can also cause Erk activation when overexpressed in Jurkat T cells (23).

In the present study we have reexamined the importance of ZAP-70 in regulating Erk activity in T cells in response to two different stimuli: 1) TCR cross-linking with the anti-CD3 monoclonal antibody, OKT3 and 2) an oxidative stimulus, hydrogen peroxide. Using the ZAP-70-negative Jurkat T cell line, P116, we find that ZAP-70 is required for the activation of Erk1 and 2 in response to H₂O₂ stimulation. Interestingly, we also find a ZAP-70-independent pathway for the activation of Erk 1 and 2 in P116 cells in response to TCR cross-linking, which is also independent of SLP-76 tyrosine phosphorylation. The implications of the presence of two pathways for the activation of Erk1 and 2 with differing requirements for ZAP-70 is discussed.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Cells and Antibodies-- E6 Jurkat T cells, the ZAP-70-negative Jurkat cell mutant, P116, and all stably transfected cell lines derived thereof were maintained as described (9, 24). The characteristics of the P116 cells have been described (24). In brief, they were derived from E6 Jurkat by mutagenesis and selection for inability to flux calcium in response to pervanadate. They lack ZAP-70 protein, and normal signaling capacity is restored by transfection with wild-type ZAP-70. P116 cells express normal levels of other signaling proteins, such as TCR/CD3 chains, Lck, Fyn-T, phospholipase C1, SLP-76, and Erk 1 and 2 (24). The OKT3 monoclonal antibody to human CD3 and the polyclonal rabbit antiserum specific for ZAP-70 have been described (9). The anti-phosphotyrosine monoclonal antibody, 4G10, and a sheep antiserum against human SLP-76 were from Upstate Biotechnology, Inc. (Lake Placid, NY). The anti-Erk2 polyclonal antiserum used for immunoprecipitation was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). A Transduction Laboratories (Lexington, KY) monoclonal antibody to Erk2 was used to blot for total Erk2 protein. A rabbit polyclonal antiserum specific only for Erk phosphorylated at both Thr-183 and the Tyr-185 and therefore specific for activated Erk was obtained from Promega (Madison, WI).

Plasmids and Transfection-- The cDNA encoding full-length, wild-type ZAP-70 fused with a C-terminal 9-amino acid myc epitopic tag was subcloned as a BamHI fragment from the previously described pSXSR-ZAP_myc plasmid (12) into the BamHI site of pBJneo to generate pBJ-ZAP_myc. The pBJ-ZAPKD_myc vector, which expresses a kinase-dead, dominant-negative form of ZAP-70, was similarly constructed from ZAP-70 cDNA carrying a K369R mutation, which was generated by site-directed mutagenesis using the Transformer kit (CLONTECH Laboratories, Inc. (Palo Alto, CA). P116 transfectants stably expressing wild-type or kinase-dead ZAP-70 were prepared as described previously (25).

Stimulation of Cells and Erk Kinase Assay-- The cells were serum-starved for 16 h before being harvested and washed in 4 °C RPMI using minimal manipulation of the cells, since it had been noted that extensive manipulation of the cells resulted in high basal Erk activity. The cells were resuspended to 5 × 10⁷ cells/ml in 4 °C RPMI and maintained on ice until stimulated. After equilibration to 37 °C for 5-10 min, the cells were stimulated by the addition of either OKT3 (1:100 of ascites), H₂O₂ (10 mM, unless otherwise indicated), or 50 ng/ml PMA. The duration of stimulation was 3 min unless otherwise indicated. Stimulation was terminated by the addition of 5 volumes of 4 °C lysis buffer (20 mM HEPES, pH 7.4, 1% Triton X-100, 50 mM -glycerophosphate, 2 mM EGTA, 10 mM sodium fluoride, 1 mM sodium orthovanadate, 10% glycerol, 10 µg/ml leupeptin, 10 µg/ml aprotinin, 100 µg/ml 4-(2-aminoethyl)-benzenesulfonyl fluoride). After a 30-min incubation on ice, post-nuclear lysates were prepared by a 10-min centrifugation at 21,000 × g at 4 °C. The lysates were then either directly analyzed by immunoblotting or subjected to immunoprecipitation followed by immunoblotting or kinase assay. Immunoprecipitates that were to be analyzed by immunoblotting were washed three times with lysis buffer supplemented with 150 mM NaCl. Those immunoprecipitates that were to be subjected to immune complex kinase assays were washed two times each sequentially with lysis buffer, LiCl wash buffer (100 mM Tris-HCl, pH 7.5, 0.5 M LiCl, 0.1% Triton X-100, 1 mM dithiothreitol), and kinase buffer (20 mM MOPS, pH 7.2, 2 mM EGTA, 10 mM MgCl₂, 1 mM dithiothreitol, 0.1% Triton X-100). Immunoprecipitated Erk was subjected to an immune complex kinase assay essentially as described previously (26).

Electrophoresis, Immunoblotting and Autoradiography-- Samples to be analyzed by electrophoresis were prepared in NuPAGE sample buffer by heating to 100 °C for 5 min. Samples were analyzed on NuPAGE 4-12% gradient gels either in MOPS or MES sample buffers and then transferred to nitrocellulose membranes according to the manufacturer's instructions (NOVEX, San Diego, CA). Immunoblots were developed by the ECL system of Amersham Pharmacia Biotech according to manufacturer's instructions. Autoradiograms were obtained by exposure of BMR film (Kodak, Rochester, NY).

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

ZAP-70 Is Required for TCR and H₂O₂-induced Accumulation of Tyrosine-phosphorylated Proteins-- H₂O₂ has been shown previously to cause accumulation of tyrosine-phosphorylated proteins in Jurkat T cells (27, 28). To test the role of ZAP-70 in this process, we examined tyrosine phosphorylation of proteins in cell lysates from ZAP-70-negative and ZAP-70-replete Jurkat T cells (Fig. 1). Hydrogen peroxide within the concentration range of 1-10 mM did not stimulate tyrosine phosphorylation of cellular substrates in the P116 cells. Normal Jurkat cells, on the other hand, showed a rapid accumulation of tyrosine-phosphorylated proteins in response to incubation with H₂O₂. The effect of H₂O₂ was concentration-dependent, and within the concentration range tested (1-10 mM), had a greater effect on tyrosine phosphorylation than that seen with anti-CD3 cross-linking, causing robust phosphorylation of proteins that were only minimally phosphorylated by anti-CD3. OKT3 stimulation of the P116 cells caused only weak tyrosine phosphorylation, with only proteins of 42, 85, and 150 kDa responding. Of note is the absence in both OKT3 and H₂O₂-stimulated P116 cells of the tyrosine-phosphorylated 36- and 76-kDa proteins that are predominant substrates in normal Jurkat cells and which are likely to correspond to the Grb2-associated p36 protein and SLP-76. These results are consistent with the severe early signaling defect that has been reported for P116 cells (24).

View larger version (74K): [in this window] [in a new window]   Fig. 1.   ZAP-70 is required for OKT3- and H2O2-stimulated accumulation of tyrosine-phosphorylated proteins. Whole-cell lysates were prepared from P116 and Jurkat cells incubated for 3 min at 37 °C with either no addition (Control), OKT3 ascites (CD3), or the indicated concentration of H2O2. 2 × 105 cell equivalents were separated on a 4-12% NuPAGE gel in MOPS buffer, transferred to nitrocellulose, and immunoblotted for phosphotyrosine. Results are representative of three independent experiments.

ZAP-70 Is Required for Erk Activation in Response to H₂O₂ but Not TCR Cross-linking-- In the same experiment, the requirement for ZAP-70 in mediating Erk activation in response to TCR cross-linking or H₂O₂ was also assessed, using both an immune complex kinase assay and immunoblotting for doubly phosphorylated, active Erk in whole-cell lysates. The presence of equal amounts or Erk2 in the immunoprecipitations and of Erk 1 and 2 in the Jurkat and P116 whole-cell lysates was confirmed by immunoblotting for these proteins (not shown). Erk2 was immunoprecipitated from the cell lysates and was used in an in vitro kinase assay using myelin basic protein as an exogenous substrate. Stimulation of normal Jurkat cells with either anti-CD3 or H₂O₂ (1-10 mM) gave an apparent maximal activation of Erk2 (Fig. 2A). H₂O₂ was effective in activating Erk2 down to a concentration of 10 µM (the lowest concentration tested) and demonstrated a graded concentration-response relationship between 10 µM and 1 mM (not shown). Activated Erk1 and Erk2 could also be detected in immunoblots of lysates from Jurkat T cells stimulated with H₂O₂ or OKT3, confirming the results of the kinase assay (Fig. 2B). However, in the P116 cells, the effect of H₂O₂ on Erk activation was undetectable by anti-active Erk immunoblotting of whole-cell lysates and only weakly detectable in the immune complex kinase assay. Unexpectedly, although the P116 cells were unable to activate Erk in response to H₂O₂, they remained capable of activating Erk in response to CD3 cross-linking. Please note that the apparent slower migration of Erk1 and 2 isolated from Jurkat cells as compared with P116 was not a consistent observation of these studies (see Fig. 4). The presence of ZAP-70 in Jurkat cells and its absence in P116 cells was demonstrated by immunoprecipitation of ZAP-70 followed by ZAP-70 immunoblotting (Fig. 2C). A replicate membrane was blotted for phosphotyrosine and showed that both anti-CD3 and H₂O₂ caused increased tyrosine phosphorylation of ZAP-70 (Fig. 2D).

View larger version (54K): [in this window] [in a new window]   Fig. 2.   ZAP-70 is required for Erk activation in response to H2O2 but not OKT3. Cell lysates from the same experiment shown in Fig. 1 were immunoprecipitated with an antibody to Erk2 and subjected to an immune complex kinase assay using myelin basic protein as a substrate (1 × 107 cell equivalents) (A), blotted with an antibody recognizing only the activated forms of Erk1 (upper band) and Erk2 (lower band) (2 × 105 cell equivalents) (B), or immunoprecipitated with an antibody to ZAP-70 (C and D), and either immunoblotted for ZAP-70 (C, 1 × 107 cell equivalents) or for tyrosine-phosphorylated ZAP-70 (D, 1 × 107 cell equivalents). Results are representative of three independent experiments. WCL, whole cell lysate.

SLP-76 Tyrosine Phosphorylation Occurs in Jurkat but Not in P116 Cells in Response to Erk-activating Stimuli-- Having found a dependence upon ZAP-70 for stimulating Erk activation in response to H₂O₂, we examined the tyrosine phosphorylation status of SLP-76, an in vivo substrate for ZAP-70 that has previously been implicated in Erk activation in response to TCR engagement. In Jurkat cells, anti-CD3 and H₂O₂ caused a striking increase in tyrosine phosphorylation of SLP-76 (Fig. 3). The effect of H₂O₂ was concentration-dependent and robust. In addition to increased tyrosine phosphorylation, there was also an increase in the association of tyrosine-phosphorylated proteins with SLP-76. However, in the ZAP-70-negative P116 cells, neither anti-CD3 nor H₂O₂ had any effect on SLP-76 tyrosine phosphorylation or its association with other tyrosine-phosphorylated proteins.

View larger version (47K): [in this window] [in a new window]   Fig. 3.   Erk activation is correlated with SLP-76 tyrosine phosphorylation in Jurkat cells but not P116 T cells. Cell lysates (1 × 107 cell equivalents) from the same experiment shown in Fig. 1 were immunoprecipitated with an antibody to SLP-76. Tyrosine-phosphorylated SLP-76 and co-precipitating tyrosine-phosphorylated proteins were separated on a 4-12% NuPAGE gel in MOPS and detected by immunoblotting with 4G10. Results are representative of three independent experiments.

Expression of Wild-type ZAP-70 but Not Kinase-dead ZAP-70 in P116 Cells Restores Ability to Activate Erk in Response to H₂O₂-- If the failure of P116 cells to activate Erk in response to H₂O₂ is actually due to the absence of ZAP-70, then reconstitution of these cells with wild-type ZAP-70 should correct this loss of signaling capacity. To test this, we compared Erk activation in P116 cells to P116 cells that had been stably transfected with either myc-tagged, wild-type ZAP-70 or myc-tagged, kinase-dead ZAP-70 carrying a K369R mutation. Three independent clones expressing wild-type ZAP-70 (P.WT2, P.WT5, and P.WT18) and two independent clones expressing kinase-dead ZAP-70 (P.DK2 and P.DK33) were analyzed. The clones expressed varying amounts of ZAP-70, as detected by an anti-ZAP-70 immunoblot, and only the results from the clones expressing the highest levels of ZAP-70 are shown (Fig. 4A). The P.WT18 clone had a similar level of ZAP-70 expression as that seen in Jurkat cells. Wild-type ZAP-70 restored the capacity of P116 cells to activate Erk1 and 2 in response to stimulation by H₂O₂ while having no apparent effect on OKT3-stimulated Erk activation in these cells. The relative amount of activated Erk detected in the different clones was directly correlated with the relative level of wild-type ZAP-70 expressed (not shown). It should be noted that the slower electrophoretic migration of ZAP-70 in the P.WT18 and P.DK2 clones was due to the myc tag (12). As shown previously, stimulation of normal Jurkat cells with anti-CD3 or H₂O₂ caused rapid tyrosine phosphorylation of ZAP-70 (Fig. 4B), which was correlated with the increased activation of Erk. Kinase-dead ZAP-70 did not support Erk activation in response to H₂O₂ in the stably transfected P116 cells, indicating that the kinase activity of ZAP-70 is required for Erk activation in response to H₂O₂. Surprisingly, the expression of kinase-dead ZAP-70 in P116 cells blocked the ZAP-70-independent activation of Erk1 and 2 that was observed in response to CD3 cross-linking. The ability to block OKT3-mediated Erk activation in K369R-ZAP-70-transfected P116 cells was correlated with the relative expression level of the mutant ZAP-70 (not shown). The ZAP-70 expressed in the P.DK2 cells became strongly tyrosine-phosphorylated upon CD3 cross-linking despite the absence of its own kinase activity and was only weakly phosphorylated in response to H₂O₂.

View larger version (50K): [in this window] [in a new window]   Fig. 4.   Expression of wild-type, but not K369R ZAP-70 in P116 T cells restores ability to activate Erk in response to H2O2. Jurkat, P116, P.WT18, and P.DK2 cells were incubated for 3 min at 37 °C as in Fig. 1. A, whole cell lysates (2 × 105 cell equivalents) were immunoblotted for ZAP-70 and activated Erk. B, ZAP-70 was immunoprecipitated from 1 × 107 cell equivalents of the same lysates used in (A) and immunoblotted for phosphotyrosine. C, Jurkat, P116, and P.WT18 cells were incubated for 5 min at 37 °C with the indicated stimuli, and then whole cell lysates (2 × 105 cell equivalents) were immunoblotted for activated Erk (upper band is Erk 1, and lower band is Erk 2).

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

The studies reported here were initiated to gain a better understanding of the mechanism whereby H₂O₂ can mimic TCR signaling. It had previously been demonstrated that oxidative stressors such as H₂O₂ and UV can cause the accumulation of tyrosine-phosphorylated proteins, raise the intracellular concentration of Ca²⁺, and activate NF-B and AP-1 in Jurkat T cells (27, 33, 34). ZAP-70 was found to be among the proteins that became tyrosine-phosphorylated in response to UV or H₂O₂ treatment of Jurkat T cells, resulting in activation of its autocatalytic activity (27). These former studies did not, however, examine the role of ZAP-70 in signaling the presence of these agents to downstream signaling events. In this study we demonstrate that ZAP-70 is a key mediator of the signals generated in T cells by oxidative stress. ZAP-70-negative Jurkat T cells showed no increase in tyrosine phosphorylation of cellular proteins nor Erk activation in response to H₂O₂ but could be made competent for H₂O₂-induced Erk activation by stable transfection with wild-type ZAP-70 cDNA. A cDNA encoding a kinase-dead mutant of ZAP-70 could not restore competence, indicating a requirement for ZAP-70 kinase activity in mediating Erk activation.

How ZAP-70 leads to Erk activation in response to H₂O₂ (or CD3 cross-linking) remains unclear. The principal pathway leading to Erk activation involves activation of Ras, which can be activated either by inhibiting the activity of Ras-specific GTPase activating proteins (GAPs) or by promoting the binding of GTP to Ras by the action of a guanine nucleotide exchange factor. Although there is evidence that TCR cross-linking results in decreased Ras-GAP activity, the mechanism of this response remains unknown (20). Other studies have suggested that TCR engagement can stimulate the recruitment of Ras-specific guanine nucleotide exchange factors such as SOS to the plasma membrane as a consequence of tyrosine phosphorylation of plasma membrane-associated proteins. As in other cell types, this process appears to involve the adapter protein Grb2, which uses its N-terminal SH3 domain to bind to a proline-rich region of SOS and its SH2 domain to bind tyrosine-phosphorylated membrane proteins. Grb2 can directly bind to tyrosine-phosphorylated TCR ITAMs in vitro (35, 36) and binds in vivo to a membrane-associated 36-kDa protein that is rapidly tyrosine-phosphorylated after TCR stimulation (reviewed in Ref. 37). This 36-kDa protein has recently been shown to be a specific substrate for ZAP-70 in vitro,² and it may be that the requirement for ZAP-70 in H₂O₂-stimulated Erk activation may be for tyrosine phosphorylation of this protein.

Alternatively, a second substrate for ZAP-70, SLP-76, has also been suggested to be involved in TCR-stimulated Erk activation (14, 15, 21-23), and ZAP-70 may be required to phosphorylate SLP-76 in order to get Erk activation in response to H₂O₂. That SLP-76 may have to become tyrosine-phosphorylated in order to activate Erk is suggested by the fact that overexpression of a truncation mutant of SLP-76, which lacks the prominent sites of tyrosine phosphorylation, fails to augment Erk activation in response to TCR stimulation. This is in contrast to full-length SLP-76, which activates Erk (23). We did observe SLP-76 tyrosine phosphorylation in Jurkat cells, but not in P116 cells, in response to either TCR cross-linking or incubation with H₂O₂ despite the presence of equivalent levels of SLP-76 in the two cell lines (not shown). This is consistent with SLP-76 being downstream of ZAP-70 in the pathway leading to Erk activation. However, SLP-76 tyrosine phosphorylation cannot be an absolute requirement for Erk activation in T cells, since Erk activation could occur in OKT3-stimulated P116 cells in the absence of detectable SLP-76 tyrosine phosphorylation.

How SLP-76 then affects Erk activation remains to be established; however, like p36, it too is able to bind to Grb2. Unlike p36 though, this association is mediated by the C-terminal SH3 domain of Grb2. Whether or not Grb2 can simultaneously engage SOS and SLP-76 remains controversial. In addition to the proline-rich SH3 domain acceptor site, the hematopoietic-specific SLP-76 has a C-terminal SH2 domain and a cluster of N-terminal tyrosines that become phosphorylated upon TCR engagement. SLP-76 thereby forms complexes with many signaling molecules, and activates Erk, NF-AT, and AP-1 when overexpressed in T cells (22, 38). It seems likely that some of these SLP-76-associated proteins may modulate the activity of proteins involved in Ras activation and that ZAP-70-dependent tyrosine phosphorylation may regulate some of these events.

The ability of exogenous oxidants such as H₂O₂ to usurp myriad signaling pathways in many different cell types indicates the existence of multiple regulatory molecules that can be controlled at least in part by oxidation-reduction mechanisms, suggesting that endogenous reactive oxygen species (ROS) may play a role in normal signaling pathways. In fact, it has recently been shown that a number of cytokine and growth factor receptors, including the receptors for interleukin 1, tumor necrosis factor , platelet-derived growth factor, and epidermal growth factor, generate reactive oxygen species as second messengers, which are required for the activation of key downstream signaling molecules such as NF-B and phospholipase C1 (39-44). One mechanism of action of ROS is through the inactivation of protein tyrosine phosphatases by oxidizing a critical conserved cysteine residue within the active site (reviewed in Ref. 45). It has been proposed that a transient inhibition of protein tyrosine phosphatases is required in addition to activation of PTKs in order to increase the basal tyrosine phosphorylation of key signaling proteins above a threshold capable of propagating a stimulatory signal. Interestingly, it has been suggested for many years that T cell activation may also require the production of ROS on the basis that treatment of T cells with antioxidants renders them unresponsive to mitogenic stimuli (46-48). In addition, acute exposure of T cells to H₂O₂ results in interleukin 2 production and increased proliferation (Ref. 49 and references therein).

Taken together, the above observations lead us to speculate that ZAP-70 may normally be regulated, in part, by an upstream enzymatic activity that produces an ROS. To speculate further, it may be that it is the presence or strength of such an ROS signal that determines whether a given antigen-MHC TCR ligand gives rise to an agonist, partial agonist, or antagonist biochemical response (50). Perhaps activation of the TCR-proximal PTKs in the absence of ROS production and consequent protein tyrosine phosphatase inhibition results in the partial phosphorylation pattern seen with partial agonists and antagonists. The ability of ZAP-70 to activate Erk in response to ROS may also be important at sites of inflammation, where activated neutrophils and monocytes generate large quantities of ROS (51). This may provide a mechanism for these cells to regulate the proliferative capacity and activity of T cells present at inflammatory sites. The testing of these hypotheses and the identification of the enzyme(s) involved in ROS production is the subject of continuing investigation.

The most unexpected finding of these studies was the discovery that TCR cross-linking in P116 cells can initiate a ZAP-70-independent activation of Erk1 and 2. Most current models for TCR-initiated Erk activation invoke ZAP-70 activation as part of this process, and indeed, overexpression of dominant-negative ZAP-70 in Jurkat T cells results in a block in their ability to activate Erk in response to TCR ligation (10). However, these earlier studies may need to be reinterpreted in light of our finding that stable expression of kinase-dead ZAP-70 blocked the ZAP-70-independent activation of Erk in P116 cells. The ability of kinase-dead ZAP-70 to block OKT3-stimulated Erk activation in P116 cells is not due to an inhibitory effect on Syk activation, as there is no Syk present in P116 cells. The P116 cells were derived from the E6.1 Jurkat subclone and are not only ZAP-70-negative but are also negative for Syk protein and mRNA (24, 52). We are currently working to understand how Erk is activated in the absence of ZAP-70 in these cells and how it is that dominant-negative ZAP-70 is able to block a signaling pathway that does not involve ZAP-70 or Syk.

In conclusion, we have found that two stimuli that are considered to mimic the intracellular signaling events initiated by TCR engagement give completely different responses in a ZAP-70-negative Jurkat T cell line. The oxidative stressor H₂O₂ demonstrates a requirement for ZAP-70 in mediating Erk activation, whereas OKT3 cross-linking of CD3 can activate Erk in the absence of ZAP-70. This is the first demonstration of a ZAP-70-independent pathway for Erk activation in T cells and adds to the complexity of the signaling pathways that can lead to Erk activation in response to TCR engagement. Further study will be required to determine whether or not this pathway is functionally relevant in normal TCR signaling.