Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Immunosuppressive drugs have radically altered the treatment of organ transplantation. The drug cyclosporin A (CsA), which has proved very effective as an immunosuppressive agent, was the first drug in a class of relatively inexpensive small-molecule inhibitors of T cell activation and function. Due to the dose-limiting side effects of CsA (Noble and Markham, 1995), other immunosuppressive drugs have emerged that have attempted to replace it. Many of these are biological inhibitors of T cell function, such as those that block activation of the IL-2 receptor  chain (Waldmann and O'Shea, 1998), and although effective in certain circumstances, their wide spread use has been limited by their cost. This has stimulated the search for new drugs that can block signal transduction pathways emanating from the T cell receptor.

Src kinase members, p56^Lck (Lck) and p59^Fyn (Fyn), are two proteins that play an important role in TCR signaling. The role of Lck in TCR signaling was elucidated using a mutant of the Jurkat cell line, termed JCAM-1 (Straus and Weiss, 1992). Upon TCR stimulation, these cells failed to give a calcium flux. Analysis of these cells revealed a defect in tyrosine phosphorylation, which was due to a defect in functionally active Lck (Straus and Weiss, 1992). Reconstitution of JCAM-1 cells with wild-type Lck restored functionality to the cell line. TCR signaling was also antagonized by the microinjection of Lck antibodies into T cells (Nakamura et al., 1994). Finally, Lck-deficient mice were shown to have suboptimal T cell-proliferative responses when T cells were stimulated through the TCR (Molina et al., 1992).

Fyn is also associated with the TCR directly, being coexpressed with the cytoplasmic domains of CD3 and  and TCR  and  chains. The role of Fyn in TCR signaling is not as clear as that of Lck. Overexpression of Fyn in T cells, however, augments TCR stimulation, as measured by tyrosine phosphorylation and IL-2 production (Cooke et al., 1991; Davidson et al., 1992). The analysis of the function of both Lck and Fyn has been somewhat hampered by the role they play in T cell development (van Oers et al., 1996). This has emphasized the need to assess the role that Lck and Fyn play in mature human T cells.

The Src kinase Lck, as indicated, plays a key role in mediating signals transduced through the T cell receptor. Although inhibitors of Lck currently exist (Faltynek et al., 1995; Hanke et al., 1996; Gimsa et al., 1999; Trevillyan et al., 1999), they are limited in both their potency and selectivity for Lck over other tyrosine kinase enzymes. In this study, we show that a novel series of Src kinase inhibitors (Celltech Src kinase inhibitors or CT-SKI) are more potent and selective for Src kinase enzymes than previous inhibitors, such as PP1 (Hanke et al., 1996). CT-SKI inhibit the Src kinases Lck, Fyn, and Lyn at concentrations lower than those required to inhibit Zap-70, PKC, EGFR, and critically Csk, a kinase known to regulate Src kinase activity (Chow et al., 1993). These inhibitors block T cell proliferation and cytokine production when stimulated through CD3 but not when T cells are activated with more complex stimuli such as in the case of a MLR. Finally, CT-SKI inhibition of CD3-activated proliferation could be overcome by additionally stimulating the IL-2 receptor pathway by adding exogenous IL-2.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

General Reagents and Antibodies. Unless stated otherwise, all chemical and biological reagents were obtained from Sigma-Aldrich (Poole, UK). All tissue culture plastic ware was obtained from Falcon, Becton Dickinson Labware (Franklin Lakes, NJ) with the exception of U-bottomed 96-well tissue culture plates, which were obtained from Costar (Cambridge, MA). Centrifugation was performed using bench-top centrifuges 8R and GP8R (IEC, Needham Heights, MA). All biological buffers, including tissue culture medium, were obtained from Invitrogen (Paisley, UK) unless stated otherwise. Fetal calf serum (FCS) was sourced from Helena Biosciences [no. NS3005, lot 7329-2-NS31 (USA herd); Sunderland, UK]. Anti-IL-4 and IFN- antibody pairs for sandwich ELISA and recombinant protein were obtained from BD PharMingen (San Diego, CA), and anti-IL-2 antibodies and recombinant protein were obtained from BioSource (Nivelles, Belgium). Anti-CD3 conjugated to FITC (fluorescein isothiocyanate) and anti-CD25 conjugated to phycoerythrin were obtained from BD Biosciences (San Jose, CA).

Cell Culture. Peripheral blood mononuclear cells were isolated from normal healthy volunteers. Whole blood was taken by venous puncture using heparinized Vacutainers (BD Biosciences), diluted 1:4 in RPMI 1640 (Invitrogen), and centrifuged at 400g for 35 min over a Ficoll-Paque gradient (Amersham Pharmacia Biotech UK, Ltd., Little Chalfont, Buckinghamshire, UK). Cells at the interface were removed and washed once followed by a low-speed spin to remove platelets. Unless stated otherwise, cells were then resuspended in RPMI 1640 containing 10% FCS and 100 units ml¹ penicillin, 50 µg ml¹ streptomycin, and 2 mM glutamine (Invitrogen). Anti-CD3 proliferation experiments were performed using PBMC resuspended at a density of 2 × 10⁵ cells/well in round-bottomed 96-well tissue culture-treated plates. Cells were stimulated with an optimal dose of the anti-CD3 antibody OKT3 (Celltech; at 0.04 µg ml¹) and incubated at 37°C in 5% CO₂/95% air. Cellular proliferation was measured by the incorporation of [³H]thymidine (0.5 µCi/well), and cells were pulsed with [³H]thymidine for 8 h before harvesting. Cells were harvested onto glass fiber filter mats using a Skatron 96-well harvester (Molecular Devices, Sunnyvale, CA), and [³H]thymidine incorporation was measured using a -plate counter (PerkinElmer Wallac, Cambridge, UK). IC₅₀ values were determined as the concentration sufficient to inhibit 50% of the specific proliferation induced by anti-CD3. Proliferation in a one-way MLR was performed using two human leukocyte antigen mismatched donors. The first donor's cells, referred to as the "responder", were resuspended at a density of 1 × 10⁵ cells/well in round-bottomed 96-well tissue culture treated plates. These cells were stimulated with the second donor's cells, referred to as the "stimulator", which had been irradiated for a period of 45 min with a total dose of 2500 rads. These cells had no proliferative capacity but were determined to be alive by staining with trypan blue and propidium iodide. The stimulator cells were added to the responder cells at a density of 1 × 10⁵ in an equal volume. This gave a ratio of 1:1 responders to stimulators. Cells were incubated for 5 days at 37°C in 5% CO₂/95% air. Cellular proliferation was measured by the incorporation of [³H]thymidine over the final day of the experiment. The production of cytokines by activated PBMC was measured by sandwich cytokine ELISA (see ELISA methods). Anti-CD3 (0.05 µg ml¹), PMA (at 1 µg ml¹), and ionomycin (Ca²⁺ ionophore derived from Streptomyces conglobatus at 10 ng ml¹) were used to activate human PBMC to produce cytokines. Cytokine production (IL-2, IFN-, and IL-4 for OKT3 and IL-2 for PMA/ionomycin) was measured in cell free supernatants. During experiments in which inhibitors were compared in terms of their inhibition of anti-CD3 and PMA/ionomycin-induced IL-2 production, cytokine levels were measured after 48 h. In experiments designed to test the reversibility of inhibition by Src kinase inhibitors, recombinant IL-2 was added to cultures of PBMC 4 h after stimulation with anti-CD3 antibody (0.05 µg ml¹), and cells were incubated for 48 h. Cellular proliferation was measured by the incorporation of [³H]thymidine over the final 6 h of the experiment.

Sandwich ELISA. Nunc Maxisorb plates (Nalge Nunc International, Naperville, IL) were coated with 2.5 µg ml¹ of capture antibody (anti-IL-4, anti-IFN-, and anti-IL-2) overnight at 4°C in coating buffer (4.3 g of NaHCO₃ and 5.3 g of Na₂CO₃ made up to 1 liter with distilled H₂O, pH 9.4). Wells were then aspirated and blocking buffer added (8.0 g of NaCl, 1.42 g of Na₂HPO₄ · 2H₂O, 0.2 g of KH₂PO₄, 0.2 g of KCl, and 5.0 g of bovine serum albumin (fraction V) made up to 1 liter with distilled H₂O, pH 7.4) while plates were rotated (250 rpm) on an orbital shaker (Stuart Scientific, Bibby Sterilin, Staffordshire, UK) at room temperature (RT) for 1.5 h. Plates were then washed four times with wash buffer (9.0 g of NaCl and 1 ml of Tween 20 made up to 1 liter with distilled H₂O, pH 7.4.), using a Denley Wellwash 4 plate washer (Thermo Labsystems, Vantaa, Finland). Standards were diluted in assay buffer and added along with samples to plates and incubated at RT for 2 h. The plates were then washed four times before adding biotinylated detection antibody at a concentration of 2.5 µg ml¹ in assay buffer, which was incubated at RT for 1.5 h. The plates were then washed four times and streptavidin conjugated to horseradish peroxidase (Amdex; Amersham Biosciences UK, Ltd.) added at a concentration of 1:500 in assay buffer. The plates were incubated at RT for 30 min before being washed four times, and tetramethylbenzidine (Intergen, Purchase, NY) was substrate added. Plates were allowed to develop for between 10 to 30 min, and the reaction was terminated using stop solution (1.8 M H₂SO₄). Plates were then read at 450 nm, with a reference reading taken at 630 nm using a Labsystems Multiskan Ex plate reader (Labsystems, Thermo Labsystems, UK). Standard curves were constructed and data analyzed using Genesis II software (Thermo Labsystems). Minimum detection limits of each assay were determined to be at least two standard deviations above background readings.

Flow Cytometry. Human PBMC (1 × 10⁵) in round-bottomed 96-well plates were stimulated with OKT3 (0.05 µg ml¹). Cells were assessed for the expression of CD25 (IL-2 receptor  chain) after 1, 2, 3, or 4 days poststimulus. OKT3-stimulated cells were stained with CD3-FITC and CD25-phosphatidylethanolamine. Briefly, 1 × 10⁵ cells were spun down in 5-ml polypropylene Falcon tubes (BD Biosciences, San Jose, CA) and washed once in PBS and once in 3 ml of staining buffer (PBS, 0.5% heat-inactivated FCS and 0.1% sodium azide, pH 7.4). Cells were then resuspended in 50 µl of staining buffer with 0.1 µg µl¹ of anti-CD3-FITC and anti-CD25-phycoerythrin in a reaction volume of 10 µl. Tubes were incubated at 4°C for 30 min in the dark. Before analysis, all cells were washed twice in cold PBS and resuspended in 500 µl of PBS without azide. Cell staining data were acquired using CellQuest software (BD Biosciences) on either a FACScan or FACSCalibur flow cytometer (BD Biosciences). Specific antibody staining was compared with staining with isotype antibody controls. Lymphocytes were gated according to their characteristic forward and side light scattering properties. Lymphocytes that were deemed to be positive for CD3 using quadrant markers were also assessed for CD25 expression. CD25 positivity was expressed in mean (geometrical) fluorescent units. Where human PBMC were tested for cell viability, the fluorescent dye propidium iodide (Pharmingen, San Diego) was added to cells at a concentration of 50 µg ml¹ in a volume of 50 µl of staining buffer. Propidium iodide staining was measured by excitation of the dye at 532 nm and emission at 617 nm.

Enzyme Assays (Reagents). Staurosporin, ATP (Tris salt), DTT, HEPES, pEY (polyglutamic acid tyrosine ratio of 4:1), and manganese chloride were obtained from Sigma-Aldrich. The protein kinase C assay kit, streptavidin-scintillation proximity assay beads, and ³³P--ATP were obtained from Amersham Pharmacia Biotech UK, Ltd. Brij-35 was obtained from Pierce Chemical (Indianapolis, IN), and magnesium chloride was obtained from BDE (Poole, UK). Microtitre plates for scintillation proximity assay were purchased from PerkinElmer Wallac. 6-Amino hexanoyl AEEIYGVLAKKK Lck substrate was synthesized by IBMS (Southhampton, UK).

Enzymes. Lck was cloned from a Jurkat cDNA library and expressed as a glutathione S-transferase-fusion protein (GST affinity tag to aid purification) in mammalian NS0 cells. GST-Lyn, GST-Fyn, GST-EGFR, and GST-ZAP-70 were all produced in-house as a GST catalytic domain fusion by a baculovirus-SF9 expression system. PKC was purchased from Roche Molecular Biochemicals (Mannheim, Germany). cdc-2 was purchased from Amersham Pharmacia Biotech UK, Ltd..

Enzyme Assays. GST-Lck, Lyn, and Fyn enzyme activity reactions were carried out in a total volume of 200 µl at room temperature in 96-well microtiter plates. The reaction mixture contained 20 mM HEPES, pH 7.4, 2 mM magnesium chloride, 2 mM manganese chloride, 0.05% Brij 35, 5 mM DTT, 1 µM 6-amino hexanoyl AEEIYGVLAKKK peptide substrate, 0.6 µM ATP (Tris salt,) and 5 µCi/ml ³³P--ATP. The compounds were added in DMSO so that the final DMSO concentration was 1%. The assay was run for 15 min before being stopped with 50 µl of stop solution, 3 mM ATP in 125 mM EDTA. The final mixture (200 µl) was then transferred to a Millipore MAPH filtration plate (Millipore Corporation, Bedford, MA) containing 100 µl of 75 mM phosphoric acid. The plate was then left for at least 60 min at room temperature. The plate was then washed (6 × 100 µl) with 75 mM phosphoric acid, and then 100 µl of scintillant (Packard Ultima Gold) was added before counting in a Wallac Microbeta plate counter.

1

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

CT-SKI Are Selective for Src Family Kinases over Other Regulatory T Cell Kinases. We have synthesized a novel series of chemical inhibitors of the Src kinase family of enzymes (CT-SKI) using a rational drug design approach. These compounds represent examples of phenylsulfarylpyrimidine and benzodihdroquinazolines chemical structures. Their selectivity against a panel of kinases was tested. Table 1 indicates the IC₅₀ values for six inhibitors, CT5102-00, CT5215-10, CT5263-00, CT5264-00, CT5269-10, and CT5276-00, the chemical structures of which are shown in Fig. 1 (Davis et al., 1997, 1998a,b; Batchelor et al., 1998), for the inhibition of in vitro enzyme activity of Lck, Lyn and Fyn (Src kinase enzymes), Zap-70, PKC, EGF-R kinase, Csk, and cdc-2. The known Src kinase inhibitor PP2 (Hanke et al., 1996) was included to act as a reference standard. All compounds tested were potent inhibitors of Src kinase activity, as exemplified by inhibition of Lck, with each inhibitor having IC₅₀ values against Lck of less than 5 nM. These inhibitors showed much reduced activity against Zap-70, PKC, and cdc-2 and improved selectivity over EGFR and Csk compared with PP2.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   The structure of a series of phenylsulfarylpyrimidine and benzodihdroquinazolines inhibitors of Src kinase enzymes.

Inhibition of TCR-Induced T Cell Proliferation by CT-SKI. Before commencing studies to assess the ability of CT-SKI to block T cell function, it was determined that the inhibitors were able to block both protein tyrosine phosphorylation of Lck and Fyn (autophosphorylation) and calcium flux in the T cell line E6.1 activated by cross-linking CD3 (data not shown). To determine whether CT-SKI blocked T cell function in primary human cells, anti-CD3 antibodies were administered to human PBMC, inducing a T cell-proliferative response. When CT-SKI were added at a fixed concentration to anti-CD3 activated human PBMC, they blocked T cell proliferation, measured by incorporation of [³H]thymidine, over the 72 h duration of the experiment (Fig. 2A). This was exemplified by the inhibitor CT5269. The inhibition of proliferation caused by CT5269 was dose-dependent when measured 48 h after human PBMC were stimulated with an anti-CD3 stimulus (Fig. 2B).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   The effect of Src kinase inhibitors on the proliferation of human PBMC induced by anti-CD3 antibodies. A, PBMC were pretreated with 400 nM of the Src kinase inhibitors for 30 min before stimulation with the anti-CD3 antibody OKT3 (40 ng ml1). Proliferation was measured at different times after stimulation of PBMC. Values are mean of triplicates ± S.D. and are representative of three separate experiments. B, PBMC were pretreated with the concentrations of CT5269 indicated for 30 min before stimulation with OKT3 (40 ng ml1). Proliferation was measured 48 h poststimulation with OKT3. Values are expressed as a percentage inhibition of OKT3-only stimulated cultures. Values are the mean of three separate experiments ± S.E.M.

CT-SKI Inhibit T Cell Cytokine Production and Cytokine Receptor Expression. Since CT-SKI were discovered to inhibit T cell proliferation in cells activated with anti-CD3 antibodies, we tested whether a potent inhibitor (CT5269) could block the production of cytokines and the expression of cytokine receptors involved in T cell activation and effector function. Human PBMC were activated with an anti-CD3 antibody (0.04 µg ml¹) over a period of 45 h in the presence of CT5269 (500 nM). Under these conditions, IL-2 production was almost completely blocked, with CT5269 inhibiting 92% of IL-2 produced as measured by area under the curve (Fig. 3A). CT5269 also blocked production of the Th2-priming cytokine IL-4, inhibiting 93% of IL-4 produced as measured by area under the curve (Fig. 3B). In addition, CT5269 also blocked production of the Th1-priming cytokine IFN-, inhibiting 72% of IFN- production as measured by area under the curve (Fig. 3C). The inhibition of IL-2, IL-4, and IFN- was demonstrated to be dose-dependent when measured 24 h after activation with anti-CD3 (Fig. 4). Although CT5269 appeared to inhibit IFN- production to a lesser extent than the other two cytokines tested, these differences were not statistically significant (p = 0.879).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   The effect of CT5269 on the synthesis of IL-2, IFN-, and IL-4 by PBMC stimulated with anti-CD3. PBMC were pretreated with 500 nM of CT5269 for 30 min before stimulation with OKT3 (40 ng ml1). IL-2 (A), IL-4 (B), and IFN- (C) were measured in cell-free culture supernatants at different times following stimulation with OKT3. Values are the mean of three separate experiments ± S.D.

View larger version (87K): [in this window] [in a new window]   Fig. 4.   Titration of the effect of CT5269 on the synthesis of IL-2, IFN-, and IL-4 by PBMC stimulated with anti-CD3. PBMC were pretreated with CT5269 at the concentrations indicated for 30 min before stimulation with OKT3 (40 ng ml1). Cytokine levels were measured 24 h after stimulation. Cytokine production was expressed as a percentage inhibition of total cytokine produced from OKT3-stimulated PBMC cultures only. Values are the mean of three separate experiments ± S.D.

View larger version (95K): [in this window] [in a new window]   Fig. 5.   The effect of Src kinase inhibitors on the expression of CD25 (IL-2 receptor  chain) on human PBMC stimulated with anti-CD3. PBMC were pretreated with Src kinase inhibitors for 30 min before stimulation with OKT3 (40 ng ml1). IL-2 receptor  chain (CD25) expression was measured 48 h after stimulation. CD25 expression was calculated as a percentage of CD25 mean expression levels on CD3HI/CD25 double-positive cells stimulated with OKT3 alone. Values are the mean of three separate experiments ± S.D.

Inhibition of Anti-CD3 but Not PMA- and Ionomycin-Induced IL-2 Production by CT-SKI. To determine where in the signaling cascade CT-SKI intervene to inhibit IL-2 production, human PBMC were activated with an anti-CD3 antibody (0.04 µg ml¹) or PMA(1 µg ml¹) and ionomycin (10 ng ml¹) to induce IL-2 production. A panel of 31 CT-SKI were chosen with a wide range of potency against isolated Src kinase (from 3 nM to 1736 nM). When the inhibition of Lck enzyme activity (taken as a representative Src kinase family member) and inhibition of anti-CD3-induced IL-2 production were analyzed (using a Pearson correlation equation) the correlation between the two parameters, for a series of close analogs, was good (r² = 0.913) and was significant (p < 0.0001) (Fig. 6A). When inhibition of Lck activity and PMA/ionomycin-induced IL-2 production were analyzed (using a Pearson correlation equation), however, the correlation between the two parameters was poor (r² = 0.110) and was not significant (p = 0.07) (Fig. 6B).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Correlation of inhibition of Src kinase enzyme with inhibition of IL-2 production from human PBMC stimulated with OKT3 or PMA/ionomycin. PBMC were treated with the Src kinase inhibitors for 30 min before stimulation with either OKT3 (A) (40 ng ml1) or PMA (1 µg ml1) and ionomycin (10 ng ml1) (B). The IC50 values of each inhibitor were obtained 48 h after stimulation of PBMC with either OKT3 or PMA and ionomycin. Values are the mean of three separate IC50 determinations and are plotted without error bars.

Inhibition of Anti-CD3 but Not a MLR-Induced T Cell-Proliferative Response by CT-SKI. Since these compounds blocked human PBMC proliferation induced by mitogenic stimuli directed through the TCR, they were tested for their ability to block more complex nonmitogenic stimuli, in this case a one-way MLR. A group of 10 structurally related CT-SKI were tested for their capacity to inhibit anti-CD3- and MLR-induced human T cell proliferation. The compounds selected had a wide range in potency against isolated Lck enzyme (from 1.7 nM to 2871 nM). The 10 compounds selected were more potent inhibitors of anti-CD3-induced human T cell proliferation than MLR-induced proliferation. This was reflected by the mean IC₅₀ values for the group of compounds, which were 400 nM for anti-CD3-induced proliferation but 2000 nM for MLR-induced proliferation, a difference that was statistically significant (p = <0.008) (Fig. 7A).

View larger version (19K): [in this window] [in a new window]   Fig. 7.   The effect of Src kinase inhibitors on the proliferation of human PBMC induced by anti-CD3 antibodies and in a MLR. A, PBMC were pretreated with Src kinase inhibitors for 30 min before stimulation. Proliferation was measured 48 h after stimulation with OKT3 and 5 days after stimulation in the MLR. IC50 values for each compound were obtained from dose response curves. Values are the mean of three separate experiments; errors are not shown. A straight line indicates the mean of all separate points. B and C, PBMC were pretreated with Src kinase inhibitors for 30 min before stimulation with either OKT3 (40 ng ml1) (B) or an equal number of mismatched irradiated PBMC in the case of the MLR (C). IC50 values for each inhibitor were obtained 48 h after stimulation of PBMC with OKT3 and 5 days after stimulation in the MLR. Values are the mean of three separate IC50 determinations and are plotted without error bars.

Inhibition of TCR-Induced T Cell Proliferation by CT-SKI Can Be Reversed by the Addition of Exogenous IL-2. Since CT-SKI inhibit anti-CD3-induced but not antigen-induced proliferation, we derived a model system whereby the initial activation of human PBMC by anti-CD3 antibodies was augmented by a source of exogenous IL-2. This system of IL-2 amplified signaling may not be evident under normal circumstances (of anti-CD3 activated human PBMC), as CT-SKI can inhibit both IL-2 protein production and IL-2 receptor expression.

View larger version (17K): [in this window] [in a new window]   Fig. 8.   The effect of CT5269 on the proliferation of human PBMC induced by anti-CD3 antibodies and the addition of exogenous human recombinant IL-2. PBMC were preincubated with 500 nM of CT5269 for 30 min before stimulation with OKT3 (40 ng ml1). Four hours after stimulation PBMC were treated by the addition of human recombinant IL-2 at the concentrations indicated. Proliferation was measured 72 h after OKT3 stimulation. The arrow refers to the percent inhibition of CT5629 without the presence of exogenous IL-2. Data are represented as the percentage inhibition of OKT3-stimulated cultures. Values are the mean of three separate experiment ± S.E.M.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we report the effect of a new class of Src kinase inhibitor on T cell function. These compounds were shown to be more potent than first generation inhibitors such as Genistein and WIN61651 (Akiyama et al., 1987; Faltynek et al., 1995). In addition, they also improved upon second generation inhibitors, such as pyrazolopyrimidines (PP1 and PP2) (Hanke et al., 1996), which until now were the most potent and specific inhibitors of the Src kinase family of enzymes.

This new class of Src inhibitor (CT-SKI) was shown to have little activity against other kinase enzymes involved in T cell receptor activation (e.g., PKC and ZAP-70) and showed selective inhibition over the regulatory kinase Csk. This was in contrast to PP2, which showed equivalent activity against Src and Csk (Table 1). This distinction between the two classes of compound may be important, as Csk is a negative regulator of T cell signaling and thus inhibition of Csk may counteract any inhibitory effect against Src kinases, Lck and Fyn, in the signaling cascade (Chow et al., 1993). It was confirmed that CT-SKI acted to block phosphorylation of Lck and Fyn (autophosphorylation) in the T cell line E6.1 stimulated by cross-linking CD3 using an anti-CD3 antibody (data not shown). The calcium flux induced in such cells was also blocked (Allen and Rapecki, 2000), and these effects correlated with inhibition of isolated Src kinase enzyme (data not shown).

To study their effect on cellular function, we tested the ability of CT-SKI to block activation of human T cell function. Since T cell clonal expansion (i.e., proliferation) is an important step in the immune response mediated by T cells, CT-SKI were tested for their ability to inhibit this process. When T cells were activated through the T cell receptor, using cross-linked anti-CD3 antibodies, CT-SKI completely blocked cellular proliferation with a greater potency than PP2 (Allen and Rapecki, 2000 and Hanke et al., 1996). To rule out that this inhibition was due to nonspecific effects, CT-SKI were tested for inhibition of proliferation of the JY cell line. Inhibition was only seen at concentrations at least 10-fold greater than that for anti-CD3 induced T cell proliferation indicating that the effects were specific (data not shown). In addition when compounds were tested for their effect on PBMC viability using propidium iodide, similar results were obtained (data not shown).

Since T cell proliferation is a multifactorial readout involving the activation and coordination of many genes and proteins, we measured the effect of CT-SKI on an earlier readout of cellular activation, namely the production of T cell specific cytokines. T cell specific cytokines, such as IL-2, are inducible and highly regulated and are also important in the proliferative response. CT-SKI potently inhibited IL-2, IL-4, and IFN- production, and their efficacy was unaltered throughout the time frame over which cytokine production was measured. The effect of CT-SKI on IL-2 production was more potent than PP2 (Allen and Rapecki, 2000). Interestingly, CT5269, which was chosen for its potent inhibition of Lck activity, at a concentration that inhibited approximately 50% of anti-CD3-induced proliferation, completely blocked the production of both IL-2 and IL-4; CT5269 also inhibited IFN- synthesis, although it showed a trend to weaker inhibition than the other two cytokines tested. This could be due to the production of IFN- by non-T cells (e.g., NK cells), which may be refractory to the effects of Src kinase inhibitors. These effects on cytokine synthesis were in contrast to those observed with the pyrazolopyrimidine compound, PP1, which acted differentially on Th cytokines, inhibiting IFN- production and augmenting IL-4 production (Gimsa et al., 1999). These differences may be due to the altered kinase selectivity profile of CT-SKI compared with PP1/2. Indeed using these two classes of inhibitor to block TCR-induced cytokine production may indicate that IFN- and IL-4 gene activation requires the activity of specific kinase enzymes.

Since CT-SKI inhibited anti-CD3-stimulated proliferation and IL-2 production, the effect of CT-SKI on IL-2 receptor expression was studied. IL-2 receptor expression (IL-2 receptor , , and common  chains) is required for full IL-2 signaling, so the effect of CT-SKI on the IL-2 receptor  chain was measured as an indicator of functional IL-2 receptor complex. We discovered that CT-SKI inhibited IL-2 receptor  expression in an equivalent manner to their effect on proliferation; this indicates that in cells activated through the TCR, IL-2 receptor ( chain) expression may be dependent on active Src enzyme.

We have shown that CT-SKI inhibit proliferation, IL-2 production, and IL-2 receptor expression in T cells under the same conditions. To rule out a nonspecific suppressive effect on IL-2 production, we tested CT-SKI using alternative stimuli to activate IL-2 production. When PBMC were stimulated with PMA and ionomycin, which are known to activate signaling at the level of PKC and calcium, CT-SKI were considerably weaker at inhibiting the production of IL-2. In addition, inhibition of PMA/ionomycin-induced IL-2 production by CT-SKI did not correlate with inhibition of Src, contrasting with inhibition of anti-CD3-induced IL-2 production, which showed a good correlation. This would suggest that the majority of the inhibitory action of CT-SKI in blocking IL-2 production is above the level of PKC/calcium, as would be expected if these inhibitors blocked Src kinase activity at the level of the TCR.

When CT-SKI were tested in other models of T cell activation, such as the MLR, their ability to block proliferation was considerably weaker. This is in contrast to the immunosuppressive drug, CsA, which inhibits with an equal inhibitory potency anti-CD3-, phytohemagglutinin-, and MLR-induced proliferation (Hess et al., 1982; Aiello et al., 1986; Abecassis et al., 1988). Inhibition of proliferation induced by anti-CD3, but not the MLR, correlated with inhibition of the activity of the Src enzyme p56Lck. This indicates that distinct signaling cascades may be activated by the different stimuli, anti-CD3 and MLR, whose requirement for Src kinases differs.

The final link in the signal transduction pathway from TCR through production of IL-2 and expression of IL-2 receptor is IL-2 binding and signaling through its receptor. We hypothesized that if CT-SKI failed to block this pathway, this may limit their efficacy in blocking IL-2 driven proliferation. There is evidence to implicate that protein tyrosine kinases (PTK) in IL-2 receptor signaling (Ihle, 1995; Taniguchi, 1995); these protein-tyrosine kinases (PTK) include the Src kinase Lck (Hatakeyama et al., 1991; Horak et al., 1991; Minami et al., 1993). To test CT-SKI in IL-2 receptor-dependent T cell activation, an IL-2-driven model of proliferation was established. In anti-CD3-stimulated PBMC, exogenous IL-2 was added 4 h after stimulation, and its effect on proliferation was measured. We discovered that the addition of IL-2 under such circumstances reversed, in a dose-dependent manner, the inhibitory effect of CT-SKI. This is in accordance with what is known about IL-2 signaling and subsequent signal transducer and activator of transcription activation, which is not dependent on Lck (Zhou et al., 2000). These findings may also explain why CT-SKI failed to adequately block proliferation in a MLR. If proliferation, which in a MLR is only evident after 3 days, is dependent upon IL-2 to amplify T cell proliferation, then CT-SKI may not be able to block such a response. Since CT-SKI also blocks TCR-induced IL-2 receptor expression, it may also be the case that IL-2 receptor expression in a MLR is independent of Src kinase activity.

To conclude, we have shown that CT-SKI appear to be potent inhibitors of mitogen-stimulated T cell activation. Nevertheless, when more physiologically relevant stimuli are used to activate T cells, as in the case of the MLR, they are less potent. This shift in potency may represent the activation of distinct signaling pathways, which may be Src-independent. In addition, it was discovered that CT-SKI potency shows considerable drift when going from the isolated enzyme to cell functional assays. This may be due to a number of factors, such as poor cell penetration, protein binding, or more likely high (micromolar) intracellular ATP concentrations (Traut, 1994). High ATP concentrations are likely to alter the potency of CT-SKI, as these compounds were discovered to be ATP-competitive but not substrate-competitive (data not shown). Despite this, their capacity to block T cell cytokine production and proliferation suggests an immunomodulatory role, and further investigation, particularly of their effects in vivo, will be needed to verify this function.