INTRODUCTION

The transcription factor AP-1 plays a central role by integrating the signals elicited by a large number of extracellular stimuli including growth factors, phorbol esters, UV light, and cytokines (1). Homodimers of members of the Jun family of transcription factors and heterodimers formed with these members and those of the Fos family comprise the AP-1 transcription factor, which binds to the TRE¹ site to regulate gene transcription (1, 2).

The capacity of AP-1 to respond to a wide array of extracellular signals is largely mediated by the transcription factors that regulate the c-fos and c-jun promoters whose binding and activity are in turn regulated through different signaling pathways (2, 3, 4). Thus, the major regulatory element within the c-fos promoter, the serum response element (SRE), is recognized by a complex that contains the ternary complex factor (TCF)/Elk-1 and the serum response factor (3). TCF/Elk-1 is phosphorylated rapidly in response to a variety of stimuli by the mitogen-activated protein kinases ERK-1 and ERK-2, which are activated by the Ras/Raf/ERK kinase cascade (4, 5). The sequential activation and phosphorylation of ERK and TCF/Elk-1 facilitate the SRE ternary complex formation and the subsequent transcriptional induction of c-fos (6, 7, 8). Transcription of c-jun is activated by binding of preformed homodimers of c-Jun or heterodimers of c-Jun and ATF-2 to the TRE of the c-jun promoter. Both transcription factors c-jun and ATF-2 are phosphorylated by another group of mitogen-activated protein kinases denominated JNKs/SAPKs (c-Jun NH₂-terminal kinases/stress-activated protein kinases) (9, 10, 11), which are activated early by upstream signal transduction pathways involving kinase cascades different from those regulating the ERKs (12, 13).

Although several growth factors stimulate the ERK pathway and induce c-fos involving the phosphorylation of TCF/Elk-1, exposure of cells to stress induced by UV radiation (14) or by protein synthesis inhibitors (15), which occurred with c-fos gene expression, did not involve activation of the ERK pathway. Recently, studies on the mechanisms involved in TCF/Elk-1 phosphorylation indicated that this factor is a point of convergence of mitogen-activated protein kinases/ERKs and JNK/SAPK signaling pathways, since JNKs/SAPKs can also phosphorylate TCF/Elk-1 in response to a variety of stimuli (13, 16, 17, 18, 19). Therefore, JNK activation itself could account for AP-1 activation, since TCF/Elk-1, c-Jun, and ATF-2, factors critically involved in the transcription of c-fos and c-jun, are targets that can be phosphorylated by this kinase.

The activity of the AP-1 and NFB transcription factors is regulated differentially in T lymphocytes in response to changes in the cell redox state (20). The use of antioxidants with radical scavenger properties has evidenced a role for reactive oxygen intermediates, generated under prooxidant conditions, on the activation of NFB. Thus, the activation of NFB in response to oxidative stress is inhibited by several structurally unrelated antioxidants including dithiocarbamates (21, 22, 23, 24), the antioxidant enzyme thioredoxin (23, 24, 25), and the glutathione precursor N-acetylcysteine (21, 26, 27). AP-1 is also activated by a number of stimuli that generate oxidative stress (23, 28, 29, 30, 31). However, these antioxidants that block NFB activation efficiently paradoxically stimulate the DNA binding and transcriptional activities of AP-1 involving induction of the c-fos and c-jun mRNAs (23, 24, 32) and the transcriptional activation of the SRE (23). Therefore, signals triggered under both prooxidant and antioxidant conditions can lead to the activation of AP-1.

To understand the mechanisms by which antioxidants activate AP-1, we analyzed their effects on different signal transduction pathways involved in T cell activation. We found that different antioxidant agents stimulate JNK activity. Furthermore, we studied the antioxidant-induced signaling in T lymphocytes using pyrrolidine dithiocarbamate (PDTC), a potent antioxidant agent that activates AP-1. Our results are discussed in the context of recent works that implicate JNK as a mediator in the activation of both c-fos and c-jun transcriptional expression.

EXPERIMENTAL PROCEDURES

The human Jurkat T cell line was grown in RPMI 1640 with GLUTAMAX-I (Life Technologies, Inc.) supplemented with 10% fetal calf serum. Phorbol 12-myristate 13-acetate (PMA), PDTC, butylated hydroxyanisole (BHA), N-acetylcysteine, and the calcium ionophore A23187 were purchased from Sigma. Cyclosporin A was from Sandoz. Bisindolylmaleimide and herbimycin A were from Calbiochem. The purified activating anti-CD28.2 monoclonal antibody was provided by Dr. D. Olive, and the anti-CD3 SPV-T3 has been described previously (33).

Jurkat cells were incubated with PDTC, PMA plus ionophore, or pretreated with PDTC and then exposed to PMA plus ionophore. After various time points, cells were harvested, and total RNA was isolated using the Ultraspec system (Biotox Laboratories, Inc.). For dot-blot analysis, RNA from each sample (10 µg) was denatured and blotted onto nitrocellulose membranes in a Bio-Dot SF microfiltration apparatus (Bio-Rad). For Northern blot analysis denatured RNA (20 µg) was electrophoresed on a 1% agarose gel and blotted onto a nitrocellulose membrane.

After UV cross-linking, the filters were hybridized with the corresponding specific probes: a 0.8-kb BglII-NcoI fragment of c-fos cDNA, a 0.8-kb HindIII-PstI fragment of c-jun cDNA, a 1.8-kb EcoRI fragment of human PAC-1 cDNA (34), and a 0.6-kb HindIII-BamHI fragment of -actin cDNA.

The plasmids used for the expression of activated v-Ha-ras, dominant negative ras, and dominant negative raf were provided by Dr. D. Cantrell. These plasmids were generated by inserting a 0.7-kb BamHI fragment of v-Ha-ras (activated ras), a 4.4-kb XhoI-BamHI fragment of c-Ha-ras N17 (ras dominant negative), and a 0.8-kb polymerase chain reaction fragment containing a carboxyl-terminal truncated form of raf (dominant negative) into the pEFBOS expression vector (35). The IL-2Luc reporter construct containing the 326 to +45 human IL-2 promoter/enhancer region fused to the luciferase gene (36) was provided by Dr. G. Crabtree. The 73colLuc and 60colLuc plasmids including the AP-1-responsive (73/+63 base pairs) and the AP-1-unresponsive (60/+63 base pairs) regions of the human collagenase promoter fused to the luciferase gene have been described previously (37) and were provided by Dr. M. Karin.

Small scale nuclear extracts were obtained from 10⁷ Jurkat cells following a procedure described previously (38) but included in the extracting buffers 10 mM sodium orthomolybdate, 1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin, and 0.5 mM phenylmethylsulfonyl fluoride. EMSAs were performed as described (32) except for the binding reactions that were performed using 5 µg of nuclear proteins and 1 µg of poly(dI·dC) DNA carrier. The double-stranded oligonucleotide used as a probe in EMSAs, 5-GCCCCCTCTGACTCATGCTGACA-3, included the 68 to 46 base pairs of the CD11c promoter (32) containing an AP-1 consensus site.

Jurkat cells (10⁷) were transfected with Lipofectin (Life Technologies, Inc.) as recommended by the manufacturer. Briefly, exponentially growing cells (3-7 × 10⁵/ml) were washed twice with phosphate-buffered saline, resuspended in Opti-MEM (Life Technologies, Inc.), and incubated for 12 h at 37 °C with a mixture of 5 µg/ml plasmid DNA and 10 µg/ml Lipofectin in 1 ml of Opti-MEM. Then, cells were diluted with 14 ml of complete medium (RPMI and 10% fetal bovine serum) and incubated for 36 h. Cells were then treated with different stimuli for 12 additional h, collected by centrifugation, and luciferase activity was determined following instructions described in a luciferase assay kit (Promega, Madison, WI). In cotransfection experiments with expression vectors, the same protocol was followed, but the reporter and expression plasmids were added at concentrations of 5 and 10 µg/ml, respectively, and the Lipofectin to 10 µg/ml.

Western blot was performed as described previously (39) with the following modifications. Cells (10⁶) were washed and lysed in 200 µl of buffer A (20 mM Hepes, pH 7.8, 50 mM -glycerophosphate, 0.1 mM sodium orthovanadate, 0.5% Triton X-100, 2 mM MgCl₂, 1 mM EGTA, 1 mM dithiothreitol, 2 µg/ml aprotinin, 2 µg/ml leupeptin, 3 nM okadaic acid). After vortex, cells were kept on ice for 15 min and microcentrifuged for 15 min. Supernatants were collected, mixed with loading buffer, and separated by 8% SDS-polyacrylamide gel electrophoresis. Gels were transferred to Immobilon-P membranes and then incubated with blocking solution (5% w/v skim milk TBS-T (TBS, 0.05% Tween 20)) for 60 min at 45 °C, washed three times in TBS-T, and incubated with anti-ERK 1, 2 (Zymed, San Francisco) 0.1% v/v in TBS-T, 5% w/v skim milk. Then, membranes were washed six times for 5 min each in 0.5% w/v skim milk in TBS-T, and peroxidase-labeled goat anti-mouse IgG (ICN) (0.1% v/v in TBS-T) was incubated for 60 min at room temperature. After three washes with TBS-T and one with TBS, membrane-bound antibody was visualized by ECL detection reagent (Amersham Corp.).

In-gel kinase assays were performed as described previously (40) with some modifications. Lysates from 10⁶ cells were diluted in Laemmli sample buffer and separated by electrophoresis in a 10% SDS-polyacrylamide gel polymerized in the presence of GST-c-Jun_1-79 fusion protein. GST-c-Jun protein was isolated from 250 ml of bacterial cells expressing pGEX-c-Jun_1-79 plasmid (provided by Dr. R. Davis) using GSH-Shepharose 4B beads (Pharmacia Biotech Inc.). After electrophoresis, the gel was washed twice for 15 min in 100 ml of 20% v/v 2-propanol, 50 mM Hepes, pH 7.6, and then equilibrated with two 15-min changes of 100 ml of buffer A (50 mM Hepes, pH 7.6, 5 mM -mercaptoethanol). Proteins were denatured by incubating the gel twice for 15 min in 100 ml of 6 M urea in buffer A at room temperature. This was followed by four sequential renaturation steps (15 min each) by discarding half of the volume (50 ml) and replacing it with ice-cold renaturation buffer (0.05% v/v Tween 20 in buffer A) at 4 °C. After washing three or four times with 150 ml of renaturation buffer at 4 °C, the gel was immersed in 25 ml of kinase buffer (20 mM Hepes, pH 7.6, 20 mM MgCl₂, 5 mM -glycerophosphate, 0.1 mM sodium vanadate, 2 mM dithiothreitol) for 30 min at 4 °C and incubated further in 15 ml of kinase buffer containing 20 µM ATP and 100 µCi of [-³²P]ATP at 30 °C for 1-2 h. Finally, the gel was washed four times with 150-200 ml of 5% v/v trichloroacetic acid and 1% v/v sodium pyrophosphate at room temperature, dried, and exposed for 1-2 days.

RESULTS

Since different antioxidant agents, including PDTC, activate AP-1, we analyzed further the effect of PDTC on c-fos and c-jun mRNA steady state. Kinetic experiments using total RNA from Jurkat cells treated with PDTC and from cells treated with the combination of the phorbol ester PMA and the Ca²⁺ ionophore A23187, either pretreated or not with PDTC, were carried out (Fig. 1A). Incubation with PMA plus A23187 resulted in transient induction of both c-fos and c-jun mRNA levels, having maximal values at 30-60 min, whereas exposure to PDTC caused a prolonged expression of c-jun mRNA reaching maximal levels between 1 and 4 h and remained high for at least 10 h. PDTC induced a weaker expression of c-fos mRNA although displaying kinetics different from that exerted by the combination of PMA and ionophore. Strikingly, the incubation of Jurkat cells with PDTC prior to stimulation with PMA plus ionophore led to a prolonged induction not only of the c-jun mRNA levels, but also of the c-fos levels, which were, on the other hand, higher than those obtained with PDTC and PMA-ionophore separately. As shown previously for other cell types (23, 24, 32), EMSA using nuclear extracts from Jurkat T cells and a consensus AP-1 probe confirmed that PDTC (and PMA-ionophore) induced AP-1 DNA binding activity (Fig. 1B) and TRE-dependent transcriptional activation (Fig. 2). Taken together these data indicated that in Jurkat T cells, PDTC is able, by itself, to induce a transient expression of c-fos, a strong and prolonged expression of c-jun, and a durable activation of both c-fos and c-jun triggered by PMA plus ionophore.

-actin

The Ras/ERK pathway is a common route that serves multiple extracellular signals that activate the SRE of the c-fos promoter (4, 5). Since the SRE has been shown to act as a primary antioxidant responsive element that can be activated by PDTC (23), we analyzed whether the activation of AP-1 by this antioxidant was mediated through activation of this pathway in Jurkat T cells. As shown in Fig. 2A, the activation of an AP-1 reporter construct (73colLuc) in response to PDTC was not affected by the expression of a dominant negative ras, c-Ha-ras N17 (Fig. 2A). The efficient expression of the ras construct was confirmed in parallel cotransfections, where expression of c-Ha-ras N17 caused a severe transcriptional inhibition of both a luciferase plasmid directed by the human IL-2 promoter (Fig. 2C) and the AP-1 reporter plasmid in response to PMA plus ionophore (data not shown). Similar experiments were performed to analyze whether Raf-1 kinase was required for the activation of AP-1 induced by PDTC. AP-1 trans-activation by the antioxidant was not significantly affected by the expression of a trans-dominant negative mutant of Raf-1 kinase (Fig. 2B) which efficiently inhibited the IL-2Luc trans-activation, as well as the AP-1Luc trans-activation by PMA plus ionophore (Fig. 2D and data not shown).

To explore whether the signals elicited by PDTC affected ERK activity, we performed Western blot analysis using anti-Erk-1, -2 antibodies to detect any possible decrease in the electrophoretic mobility of ERKs which would be indicative of its activation (41). PDTC failed to modify the mobility of ERK-2, which appears to be the predominant ERK isoform expressed by Jurkat cells (Fig. 3A). As expected, PMA caused a very rapid transient shift in the mobility of ERK-2 which was detected as early as 30 s and declined after 1 h of treatment (Fig. 3A). Interestingly, when Jurkat cells were pretreated with PDTC and stimulated further with PMA, ERK-2 persisted in its activated form after 24 h of treatment (Fig. 3A).

The persistent phosphorylation of ERK-2 observed might be due to an inhibition of the PAC-1 T cell phosphatase, which has been shown to dephosphorylate ERK-2 in vivo (34, 42). However, Northern blot analysis indicated that mRNA steady-state levels of PAC-1, induced early by PMA, were not significantly affected by previous treatment with PDTC after 90 and 180 min (Fig. 3B) when a complete activation of ERK-2 was found in cotreated cells (Fig. 3A). Taken together, our results indicated that the Ras/Raf/ERK pathway did not appear to mediate the effect of the antioxidant PDTC. In addition, the combined treatment of PDTC and PMA, which has been shown to synergize on AP-1-dependent trans-activation (23, 24) caused a persistent phosphorylation of ERK-2 by mechanisms different from the inhibition of the gene expression of PAC-1.

Signals leading to T cell activation by costimulation with phorbol ester and Ca²⁺ ionophore or by agonistic antibodies to the T cell receptor-CD3 complex and to CD28 involve the activation of JNKs (43). Besides the ability to phosphorylate c-Jun (40), JNKs have recently been shown to phosphorylate ATF-2 (11, 44, 45) and TCF-Elk-1 (16, 17, 18, 19). Thus, JNKs phosphorylate factors involved in the activation of AP-1 not only by regulating c-jun but also by c-fos transcription. Therefore, we decided to determine whether PDTC affected the activity of JNKs, a mechanism that could account for the observed c-fos and c-jun activation in the apparent absence of ERK activation. In-gel kinase assays of GST-c-Jun_1-79 revealed activation of the 46- and 55-kDa forms of JNK in extracts of Jurkat cells treated with PDTC (Fig. 4A). Strikingly, whereas the treatment with either PMA plus ionophore or anti-CD3 plus anti-CD28 resulted in a transient activation of JNK (Fig. 4, A and B), PDTC exerted a prolonged activation of JNK which was maintained for at least 8 h (Fig. 4A). To determine whether the activation by PDTC was due to its antioxidant effect, we then analyzed the effects on JNK activity of two structurally unrelated antioxidants, N-acetylcysteine and BHA, which have been shown to increase AP-1 activity (23, 24). Exposure of Jurkat cells to BHA resulted in a weak although sustained activation of JNK which persisted for more than 4 h (Fig. 4D). Treatment with N-acetylcysteine led, on the other hand, to a transient activation of JNK, although in this case it was even stronger than that mediated by PMA plus ionophore (Fig. 4C). Thus, signals triggered by different antioxidant agents that increase or activate AP-1 involve the activation of JNK.

To search for mechanisms involved in the activation of JNK by antioxidant agents, a panel of inhibitors that interfere with upstream signals involved in T cell activation was used. Thus, inhibition of protein tyrosine kinases with herbimycin A resulted in significant inhibition of the JNK activation mediated by PMA plus ionophore (Fig. 5A) and by anti-CD3 plus anti-CD28 antibodies (data not shown), but it did not interfere with that mediated by PDTC. In addition, blockade of the Ca²⁺/calcineurin signaling pathway with cyclosporin A, which has been shown to attenuate JNK activation in response to either PMA plus ionophore or anti-CD3 plus anti-CD28 (43), had no effect on PDTC-mediated JNK activation at the doses analyzed (Fig. 5, A and B, and data not shown). Moreover, inhibition of PKC with bisindolylmaleimide (46), which completely abrogated the activation of JNK induced by PMA plus ionophore (Fig. 5A), had the opposite effect on the PDTC-mediated activation, resulting in increased JNK activity when bisindolylmaleimide was added (Fig. 5B). This effect, which was not observed either with extracts from cells treated with PMA plus ionophore, with anti-CD3 plus anti-CD28 cells, or with control cells pretreated with the PKC inhibitor (Fig. 5A and data not shown), requires further study. Nevertheless, the effect of bisindolylmaleimide clearly discriminates the signaling pathways induced by PDTC versus PMA plus ionophore. Thus, the signals that converge in JNK activation mediated by PDTC appear to be distinct from those elicited through stimulation of T cells with PMA plus ionophore or anti-CD3 plus CD28.

DISCUSSION

The physiological and pharmacological modifications of the cellular redox state provoke drastic changes in the activities of the transcription factors AP-1 and NFB which play a central role in the regulation of the immune response. NFB and AP-1 are affected differentially by signals generated by prooxidant and antioxidant agents (20); whereas NFB is induced by a large number of T cell stimuli that cause reactive oxygen intermediate generation (20, 47), its activation is inhibited by different antioxidant agents with radical scavenger properties. However, AP-1 is not only induced by a wide array of stimuli that generate oxidative stress (23, 28, 29, 30, 31), but also by a number of antioxidant agents (23, 24, 25, 32). In this study, we compared the signaling pathways through which antioxidants and T cell stimuli lead to the activation of AP-1 and identify JNK as a target activated by different antioxidant agents.

Although treatment of Jurkat T cells with either PMA plus ionophore or anti TCR/CD3 plus CD28 on one hand and with antioxidants on the other increases the activity of JNK, our data clearly indicate that the signal transduction mechanisms involved in both activations are different. (i) JNK activation by PDTC was refractory to inhibitors of protein tyrosine kinases, protein kinase C, or Ca²⁺ signaling, whereas the JNK activation mediated by either T cell receptor-CD3/CD28 antibodies or PMA plus ionophore was inhibited to a variable extent by such inhibitors. (ii) TRE and IL-2 promoter-mediated trans-activation by PMA and ionophore was sensitive to the expression of trans-dominant negative versions of ras and raf which failed to inhibit the AP-1 trans-activation induced by PDTC. (iii) Transient phosphorylation of ERK-2, detected at early times of activation with PMA, was not observed after treatment with PDTC.

Although the expression of c-Ha-ras N17 dominant negative did not affect the trans-activation of AP-1 by PDTC (Fig. 2A), this was augmented by the expression of a constitutively active v-Ha-ras (data not shown). Since activated Ha-ras has been shown to partially activate JNK (10), it is likely that the additional effect displayed by activated ras on the activation of AP-1 by PDTC is mediated through JNK. Hence, the Ras/Raf/ERK pathway can modulate the PDTC-induced signaling to AP-1, although it did not appear to mediate the effect of the antioxidant.

An interesting feature of ERK regulation arising during the analysis of its phosphorylation in extracts of Jurkat cells cotreated with PMA and PDTC was the persistent activated state of ERK-2. Whether this prolonged activation is related to the synergism observed on the activation of AP-1 by both agents (23, 24) remains to be investigated further. Since PAC-1 is a phosphatase rapidly expressed after T cell activation which has been shown to inhibit ERK-2 phosphorylation and activation (34, 42), we have analyzed its involvement in the activation induced by PMA plus PDTC. However, our studies on PAC-1 gene expression indicated that PAC-1 mRNA levels induced by PMA were not significantly affected by PDTC (Fig. 3B). Other explanations that could account for the constitutive activation of ERK include the possibility that PAC-1 phosphatase activity was post-transcriptionally inhibited by PDTC, or more likely, that other phosphatases could be inhibited by the antioxidant. In this regard, the expression of mitogen-activated protein kinase phosphatase-1 in T lymphocytes has recently been shown to block PMA-induced ERK activity and to inhibit IL-2 promoter-dependent transcription in response to PMA plus ionophore (48). Since mitogen-activated protein kinase phosphatase-1 has also been shown to dephosphorylate JNK (49), it will be of interest to study the activity and expression of this phosphatase in response to antioxidants.

In HeLa cells PDTC has been shown to induce c-fos and c-jun transcription (23). Although we have found induction of both c-fos and c-jun by PDTC in Jurkat cells, important kinetic differences are observed when both cell types are compared. Thus, in HeLa cells PDTC triggers c-fos gene expression faster than c-jun, with a transient induction in both cases (23). However, in Jurkat cells PDTC-mediated c-jun expression was prolonged and preceded that of c-fos. Therefore, PDTC can operate by different mechanisms depending on the cell type analyzed.

The SRE within the c-fos promoter has been shown to act as a primary antioxidant element that can be activated by PDTC in HeLa cells (23). However, as discussed above, the mechanisms by which the antioxidant acted in Jurkat cells does not appear to be mediated by the Ras/Raf/ERK pathway. Hence, PDTC seems to be one of the stimuli that can induce c-fos in the absence of ERK activation. Since JNK has been shown to phosphorylate TCF/Elk-1, the major transcription factor involved in SRE activation, it is possible that PDTC can activate c-fos transcription through the phosphorylation of TCF by JNK. Additional experiments, including in-gel kinase analysis of GST-Elk-1 in extracts of cells treated with antioxidants, will be required to elucidate this issue.

We have found that structurally unrelated antioxidant agents activate JNK in Jurkat T cells. However, the kinetics and strength of this activation as well as the abilities to stimulate AP-1 displayed by the different antioxidants are different. Thus, PDTC per se exerts a potent and prolonged JNK activation and activates both AP-1 DNA binding activity and the TRE-dependent trans-activation. BHA displays a weak but sustained activation of JNK; and N-acetylcysteine, which has been shown to induce AP-1 binding but not significant TRE trans-activation (23), exerts a transient activation of JNK. Moreover, N-acetylcysteine was the strongest antioxidant stimulating JNK. Therefore, sustained activation of JNK appears to be a mechanism through which some antioxidants can operate, but it is not a general mechanism that mediates the signaling of all of the antioxidants. Likewise, these results suggest that additional signaling, different from JNK activation, must be required to activate TRE-dependent transcription, since N-acetylcysteine, which is not able to activate it per se (23, and data not shown), activates JNK in a fashion similar to that of PMA plus ionophore.

It is important to note that both prooxidant and antioxidant conditions lead to the activation of AP-1. JNK appears to be tuned differentially under both conditions. Thus, it would be possible that those antioxidants that trigger sustained activation of JNK might elicit specific antioxidative responses through AP-1-regulated genes, subtly regulated by the duration and the strength of JNK stimulation. In this regard, in PC12 cells the cellular responses elicited by growth factors which lead cells to proliferate or to differentiate seem to be determined by the transient or sustained activation of ERK, respectively (5). On the other hand, we have shown recently that several dithiocarbamates, including PDTC, trigger myeloid differentiation involving the activation of AP-1 (32). Hence, it will be of the utmost interest to determine whether the duration of JNK activation is involved in selectively triggering these cellular responses.

Future experiments will be designed to identify the mechanisms by which antioxidants target JNK activity including their effects on the upstream signaling pathway components that activate JNK/SAPKs and on the phosphatases that regulate its phosphorylation. These studies, which could also contribute to trace the signal transduction mechanisms involved in T lymphocyte activation, are currently in progress.