- Email: [email protected]
Dexamethasone and Prostaglandin E2Modulate T-Cell Receptor Signaling through a cAMP-Independent Mechanism
CELLULAR IMMUNOLOGY ARTICLE NO.
169, 117–124 (1996)
0099
Dexamethasone and Prostaglandin E2 Modulate T-Cell Receptor Signaling through a cAMP-Independent Mechanism LUCINDA H. ELLIOTT,1 AGATA K. LEVAY, BRAD SPARKS, MARIANNE MILLER,
AND
THOMAS L. ROSZMAN
Department of Microbiology and Immunology, College of Medicine, University of Kentucky, MS415 UKMC, Lexington, Kentucky 40536-0084 Received October 3, 1995; accepted December 7, 1995
One possible explanation for the link between stress and increased incidence of infection can be attributed to concomitant increases in levels of glucocorticoids (GS) and prostaglandin E2 (PGE2), both of which possess potent immunoregulatory activities. We have previously demonstrated that concentrations of PGE2 and the synthetic glucocorticoid, dexamethasone (DEX), which individually do not inhibit human T-cell responsiveness to anti-CD3 monoclonal antibody (mAb), act synergistically to inhibit IL-2 secretion and subsequent T-cell proliferation. In the present paper, we demonstrate that treatment of anti-CD3 mAb-stimulated T-cells with low (1008 and 1009 M) concentrations of DEX and PGE2 results in the inhibition of steadystate levels of IL-2 mRNA. Initial studies to elucidate the biochemical mechanisms involved indicate that the inhibitory effects of DEX and PGE2 cannot be correlated with increased levels of intracellular cAMP or the induction of apoptosis. However, the data indicate that DEX and PGE2 when added together interrupt anti-CD3 mAb-induced tyrosine phosphorylation of substrate proteins. Furthermore, the synergistic effect of DEX and PGE2 is mimicked by agonists for the cAMP-independent EP3 subtype of the PGE2 receptor. These data suggest that DEX and PGE2 elicit cAMPindependent signaling pathways which interact to inhibit the T-cell receptor-linked signal transduction cascade in anti-CD3 mAb-stimulated T-cells. q 1996 Academic Press, Inc.
INTRODUCTION There is compelling evidence to support bidirectional communication between the nervous and immune systems (1, 2). One example of this communication derives from studies which document stress-induced neural mediators inhibition of immune function (3, 4). Chief 1
To whom correspondence should be addressed.
among the neural mediators responsible for this immunosuppression are glucocorticoids (GS).2 They have been shown to have a profound effect on immunity including inhibition of lymphocyte proliferation (5), impairment of inflammatory cytokine production (5), and alterations in lymphocyte trafficking (5, 6). Collectively these studies suggest that one physiologic role of GS may be the homeostatic control of immune defense mechanisms and prevention of damage to normal tissue (5). Other physiologic modes of immunoregulation can be activated as a consequence of external stimuli. An example of this is a localized inflammatory response involving recruitment of phagocytic cells, including macrophages as well as immunocompetent T-cells to the site of antigen deposition. During this response another homeostatic mechanism becomes operant and involves the secretion of prostaglandin E2 (PGE2) by activated macrophages. As with GS, the immunomodulatory effects of PGE2 are evidenced by the inhibition of T-cell proliferation and cytokine production (7). The collective observations that GS and PGE2 possess potent immunoregulatory function and are concomitantly increased during stress and/or inflammation offer one hypothesis for the link between stress and increased susceptibility to infection. Hence normal and pathologic conditions may coexist where T-cells encounter immunomodulatory compounds which act synergistically to suppress their function. The immunoregulatory effects mediated by GS and PGE2 on human T-cells are correlated with negative regulation of IL-2 in stimulated T-cells (7, 8). Recent studies using the synthetic GS, dexamethasone (DEX), suggest that negative regulation of IL-2 gene transcription results from direct repression, via protein – protein interactions, of a key IL-2 DNA binding protein by ligand-activated glucocorticoid receptors 2 Abbreviations used: mAb, monoclonal antibody; IL-2, interleukin2; DEX, dexamethasone; PGE2 , prostaglandin E2 ; PGE2R, prostaglandin E2 receptor; GS, glucocorticoids; GR, glucocorticoid receptor; PTK, protein tyrosine kinase; TCR, T-cell receptor.
117
/ 6c0d$$8090
03-08-96 11:23:11
cias
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP: Cell Immuno
118
ELLIOTT ET AL.
(GR) (8). PGE2 mediates its effect by binding to a specific G-protein-linked seven transmembrane EP receptor, of which four subtypes (EP1 , EP2 , EP3 , and EP4) have been identified (reviewed in 9). Of these, only EP2 and EP3 are present on lymphoid cells. The inhibitory effect of PGE2 on IL-2 gene transcription has been attributed to increased levels of cAMP and the subsequent activation of protein kinase A (PKA) which interferes, via a phosphorylation event, with downstream signaling events coupled to the T-cell receptor (TCR/CD3) complex (10). However, these studies utilized high concentrations (1006 M ) of DEX or PGE2 , and hence may not reflect some in vivo situations. Because concomitant increased levels of GS and PGE2 can occur during stress and/or inflammation, we have previously determined the combined immunomodulatory effects of DEX, and PGE2 on Tcell activation. Thus, low concentrations (1009 to 1008 M ) of PGE2 and DEX, which individually do not inhibit T-cell responsiveness to anti-CD3 monoclonal antibody (mAb) act synergistically to inhibit T-cell proliferation (11). Furthermore, the synergistic antiproliferative effect of DEX and PGE2 is correlated with suppression of IL-2 synthesis by activated Tcells and can be reversed by the addition of recombinant IL-2 (12). In the present study, we have further investigated the mechanism(s) which contribute to the synergistic inhibitory effect of low concentrations of DEX and PGE2 on T-cell activation. The data indicate that addition of DEX and PGE2 to anti-CD3 mAb-stimulated T-cells are correlated with a synergistic decrease in steady-state levels of IL-2 mRNA. However, the synergistic antiproliferative effect of these compounds is not the result of apoptosis nor increased accumulation of cAMP. Rather, the data indicate that DEX and PGE2 when added together to anti-CD3 mAb-stimulated T-cells inhibit the induction of protein tyrosine phosphorylation events. Thus, these data indicate that low concentrations of DEX and PGE2 elicit biochemical events which synergistically interact to inhibit the TCR/CD3-stimulated signal transduction cascade. MATERIALS AND METHODS Isolation of purified resting T-cells. Peripheral blood leukocytes (PBL) were isolated from heparinized venous blood, obtained from healthy volunteers, by centrifugation over Ficoll–Hypaque gradients (13). Small resting T-cells were isolated on Percoll gradients (14) followed by rosetting with sheep erythrocytes (13). Percoll was diluted to a final density of 1.099 g/cm as follows. To prepare 100% Percoll solution, 716 ml of stock Percoll (Pharmacia, Piscataway, NJ) was mixed with 100 ml 1.5 M NaCl, 20 ml 1 M Na–Hepes, 164 ml tissue culture grade water, the pH adjusted to 7.45 and filter
/ 6c0d$$8090
03-08-96 11:23:11
cias
sterilized (personal communication from Dr. Carl June). The Percoll solution was further diluted in 11 RPMI 1640 to 80, 70, 60, and 50%. Percoll gradients were prepared by sequentially layering 3 ml of each Percoll dilution [80, 70, 60, 50%] into a 15-ml conical. The PBL (1 1 108) contained in 3 ml of RPMI were layered onto the top of the gradient and the gradients were centrifuged to equilibrium at 2000g for 30 min at 47C (14). The cells banding at the 60–70 and 80–90% interfaces were pooled, washed, and rosetted with sheep erythrocytes to obtain purified resting T-cells. The purity of these cells was greater than 95% as determined by immunofluorescence. Proliferation assays. Purified resting T-cells (1 1 105 cells/well) were stimulated in triplicate with immobilized anti-CD3 monoclonal antibody (mAb 500 ng/ well) in 96-well culture plates alone (control) and in the presence of low concentrations (1009 to 1008 M) of DEX (Sigma, St. Louis, MO) and/or PGE2 (Sigma) as previously described (11). In some of the wells the PGE2 receptor agonists, Butaprost (generously provided by Dr. Kluender, Miles Inc., West Haven, CT) and MB28.767 (generously provided by Rhone-Poulenc Rorer, Essex, England) at a final concentration of 1006 M were substituted for PGE2 . The plates were incubated for 72 hr in a moist 5% CO2 atmosphere, with 2.5 mCi of [3H]thymidine (specific activity 6.7 Ci/mM) added during the last 5 hr. The cells were harvested using a Skatron multiwell harvester (Skatron, Sterling, VA) and the cpm of [3H]thymidine was incorporated per well determined by scintillation counting. The percentage inhibition of the control proliferative response in each treatment group was determined as previously described (11). The synergistic index (SI) was calculated as previously described (11). Quantitation of steady-state levels of IL-2 mRNA. Purified T-cells (5 1 106 cells/well) were stimulated in 24-well culture plates with immobilized anti-CD3 mAb (7.5 mg/well) alone (control) and in the presence of low concentrations of DEX and/or PGE2 . The plates were incubated as previously described for 6 hr and the cells from five wells (2.5 1 107) were pooled from each treatment group for RNA extraction. Total cellular RNA was extracted by the method of Chomcyznski and Sacchi (15) and steady-state levels of IL-2 mRNA were quantitated by Northern analysis utilizing 40 – 60 mg of total RNA per treatment group. Briefly, the RNA was electrophoresed on a 1% agarose gel containing formaldehyde, transferred to nitrocellulose, and hybridized at 427C for 48 hr with the IL-2 cDNA clone pTCGF-11 (ATCC, Rockville, MD). The pTCGF-11 800-bp cDNA clone, which corresponds to the entire coding sequence of human IL-2 (16) was labeled with [a-32P]dCTP by random priming using a random-primed DNA labeling kit (Boeh-
AP: Cell Immuno
119
MODULATION OF T-CELL RECEPTOR SIGNALING
ringer Mannheim, Indianapolis, IN). The autoradiograms were analyzed using the Visage 2000 Bioimage Analyzer to determine the relative levels of IL-2 mRNA in each lane. The blots were stripped and hybridized with a 32P-labeled cDNA probe for ribosomal RNA (rRNA, generous gift from Dr. Jackie Featherstone, Univ. of Kentucky, Lexington, KY) to verify that similar amounts of total RNA were loaded per well. Qualitative analysis of DNA fragmentation. Purified T-cells (5 1 106 cells/well) were stimulated in 24well culture plates as described above. The plates were incubated for 24 hr and the cells from two wells pooled from each treatment group for DNA extraction. A modification of the method of Kyprianou and Isaacs (17) was used to assay for DNA fragmentation in each treatment group. Briefly, 1 1 107 cells from each treatment group were resuspended in 1 ml of lysis buffer (10 mM Tris– HCl, pH 8.0, 0.5% Triton X-100, 5 mM EDTA, 0.5% SDS, and 300 mg/ml proteinase K) and incubated for 18 hr in 377C water bath followed by 1 hr incubation at 377C with RNase A (40 mg/ml). The lysates were extracted sequentially, twice with equal volumes of buffer (10 mM Tris–HCl, pH 8.0)-saturated phenol, twice with phenol:chloroform:isoamyl alcohol (25:24:1), and twice with chloroform:isoamyl alcohol (24:1). The DNA was precipitated overnight at 0207C with 2 vol of absolute ethanol in the presence of 10 mM MgCl2 and 0.3 M sodium acetate, pH 5.2. Precipitated DNA was pelleted by centrifugation at 15,000g for 30 min at 47C, washed with 70% ethanol, and vacuum dried and the pellet resuspended in TE buffer (10 mM Tris–HCl, pH 8.0, 1 mM EDTA). The DNA concentration was determined spectrophotometrically and the DNA samples (10 mg/well) were analyzed by electrophoresis on a 1.6% agarose gel using TAE running buffer (40 mM Tris– acetate, 1 mM EDTA, pH 7.8). DNA fragmentation was visualized under a UV light after staining with ethidium bromide (0.5 mg/ml). Tyrosine phosphorylation and Western analysis. Resting T-cells (5 1 106/cells) in a final volume of 250 ml of RPMI were preincubated for 10 min on ice with soluble anti-CD3 mAb (10 mg) alone (control) and in the presence of low concentrations of DEX and/or PGE2 to allow ligand binding to come to equilibrium. The Eppendorf tubes were then transferred to a 377C water bath and the anti-CD3 mAb was cross-linked by the addition of 20 ml of goat anti-mouse IgG antibody (Organon Teknika Corp, Durham, NC). After the addition of cross-linking antibody, the cells were stimulated for 2 min and the reaction was stopped by the addition of 1/10 vol of 101 RIPA lysis buffer containing phosphatase and protease inhibitors (18) and incubation on ice for 15 min. After centrifugation at 12,000 rpm for 15 min, the postnuclear lysates (1 1 106 cell equivalents)
/ 6c0d$$8090
03-08-96 11:23:11
cias
were mixed with 51 Laemmli sample buffer (19) and boiled for 5 min and the proteins were separated by SDS–PAGE (10% acrylamide). The proteins were transferred to 0.2 nitrocellulose and the resulting blots were blocked for 2 hr with 20 mM Tris-buffered saline (TBS) containing 0.3% Tween 20 and 3% bovine serum albumin (BSA). The blots were incubated with gentle rocking at RT overnight with horseradish peroxidase (HRPO)-conjugated recombinant anti-phosphotyrosine mAb (RC-20, Transduction Laboratories, Lexington, KY) diluted according to manufacturer’s directions. Positive protein bands were detected by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL) according to the manufacturer’s directions. RESULTS Low concentrations of DEX and PGE2 synergize to inhibit steady-state levels of IL-2 mRNA induced by anti-CD3 mAb stimulation of T-cells. We have previously reported that the synergistic antiproliferative effect of low concentrations (1009 to 1008 M) of DEX and PGE2 on anti-CD3 mAb-stimulated human T-cells is correlated with the inhibition of IL-2 synthesis (12). Experiments were performed to determine the individual and combined effects of 1009 M of DEX and PGE2 on anti-CD3 mAb-induced steady-state levels of IL-2 mRNA in T-cells. Results indicated that peak levels of IL-2 mRNA are observed in human T-cells 6 hr after stimulation with immobilized anti-CD3 mAb and decline thereafter over the next 18 hr (data not shown). Thus, IL-2 mRNA was quantitated 6 hr after anti-CD3 mAb stimulation of T-cells in subsequent experiments. The data indicate that while neither DEX nor PGE2 alone significantly inhibits the induction of IL-2 mRNA in anti-CD3 mAb-stimulated T-cells, simultaneous addition of these substances does inhibit the induction of IL-2 mRNA, with a synergistic index of 1.43 (Fig. 1). The synergistic inhibitory effect of DEX and PGE2 on anti-CD3 mAb induced activation of T-cells is not the result of apoptosis. DEX or anti-CD3 mAb stimulation has been shown to induce apoptosis in thymocytes, Tcell leukemia and T-cell clones (20). Although freshly isolated peripheral blood T-cells are largely resistant to the induction of apoptosis by anti-CD3 mAb (21), the possibility remains that apoptosis could occur as the result of the synergistic action of anti-CD3 mAb, DEX, and PGE2 on these cells. Furthermore, the observed reversible effect of exogenously added rIL-2 on the inhibitory actions of these agents could be attributed to the ability of IL-2 to rescue T-cells from apoptosis (12). Experiments were performed to determine whether concurrent stimulation of T-cells with anti-CD3 mAb, DEX, and PGE2 induced apoptosis, as evidenced by
AP: Cell Immuno
120
ELLIOTT ET AL.
FIG. 1. DEX and PGE2 synergistically inhibit steady-state levels of IL-2 mRNA in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 6 hr with immobilized anti-CD3 mAb alone (column 1) or in the presence of 1009 M DEX / PGE2 (column 2), 1009 M DEX (column 3), or 1009 M PGE2 (column 4). Steady-state levels of IL-2 mRNA were quantitated by Northern analysis utilizing 40–60 mg of total RNA from each sample. To normalize for RNA loading, the results are expressed as the ratio of the OD for IL-2 mRNA to 24S ribosomal RNA. The results are from a representative of three experiments.
Stimulation of the EP3 subtype of the PGE2 receptor mimics the synergy observed with PGE2 and DEX. Recent evidence from our laboratory suggests that human T-cells express at least two subtypes of receptors, EP2 and EP3 , for PGE2 (25). Functional studies utilizing specific agonists demonstrate that stimulation of the EP2 subtype of the PGE2 receptor (PGE2R) is correlated with a rise in intracellular cAMP, while stimulation of the EP3 subtype is not (9). However, stimulation of either PGE2R subtype with 1004 M of the appropriate agonists results in the inhibition of T-cell activation comparable to that observed with 1006 M PGE2 (25). Experiments were performed to determine whether the synergy observed with DEX and PGE2 is linked to a specific subtype of PGE2R human T-cells. Concentrations of specific agonists for the EP2 (Butaprost) and EP3 (MB28.767) subtypes of the PGE2R, which individually do not affect T-cell responsiveness to anti-CD3 mAb, were used in these studies. The data presented in Fig. 4 indicate that stimulation of the EP3 subtype of the PGE2R with the 1006 M of the EP3-specific agonist MB28.767 concurrently with 1008 M DEX results in the synergistic inhibition (SI Å 1.9) of anti-CD3 mAbinduced proliferation of T-cells which is comparable to that observed with 1008 M PGE2 (SI Å 2.0). However,
DNA fragmentation. As shown in Fig. 2 neither the stimulation of T-cells with anti-CD3 mAb alone (lane 5) nor that in the presence of 1008 M PGE2 (lane 4), 1008 M DEX (lane 3), or 1008 M PGE2 / 1008 M DEX (lane 2) induced DNA fragmentation. Mouse thymocytes stimulated for 24 hr with 1007 M DEX were used as a positive control in these experiments (lane 1). Identical results were obtained using T-cells stimulated with anti-CD3 mAb for 48 hr in the presence of as much as 1006 M DEX (data not shown). Increased cAMP accumulation is not responsible for the synergistic antiproliferative effect of DEX and PGE2 on anti-CD3 mAb-stimulated T-cells. It is generally accepted that high concentrations (1006 M) of PGE2 inhibit T-cell activation by increasing intracellular levels of cAMP (22). Although DEX mediates its effects via a cytosolic steroid receptor not linked to cAMP induction, suprahigh concentrations (1004 M) have been shown to upregulate PGE2-induced cAMP levels in human lymphocytes (23). Moreover, stimulation of T-cells with anti-CD3 mAb can amplify the cAMP response to high concentrations of PGE2 (24). Thus, experiments were performed to determine whether the concurrent stimulation of T-cells with low concentrations of DEX, PGE2 , and anti-CD3 mAb induced a synergistic rise in cAMP. The results presented in Fig. 3 show that 1008 M DEX does not amplify the cAMP response to 1008 M PGE2 in anti-CD3 mAb-stimulated T-cells.
/ 6c0d$$8090
03-08-96 11:23:11
cias
FIG. 2. DEX and PGE2 do not induce DNA fragmentation in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 24 hr with immobilized anti-CD3 mAb alone (lane 5) or in the presence of 1008 M PGE2 (lane 4), 1008 M DEX (lane 3), or DEX / PGE2 (lane 2) and the extracted DNA was analyzed for fragmentation on a 1.6% agarose gel. DNA extracted from mouse thymocytes stimulated with 1007 M DEX (lane 1) was used as a positive control for DNA fragmentation. The results are from a representative of three experiments.
AP: Cell Immuno
MODULATION OF T-CELL RECEPTOR SIGNALING
121
signaling events elicited by the EP2-specific agonist, butaprost, do not synergize with DEX.
FIG. 3. DEX does not increase PGE2-induced cytosolic levels of cAMP in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 10 min with biotinylated anti-CD3 mAb cross-linked with streptavidin alone (column 2) or in the presence of 1008 M DEX (column 3), 1008 M PGE2 (column 4), or 1008 M DEX / 1008 M PGE2 (column 5). Cytosolic levels of cAMP for resting T-cells are in column 1. The results are from a representative of two experiments.
Synergistic concentrations of DEX and PGE2 inhibit protein tyrosine phosphorylation in anti-CD3 mAbstimulated T-cells. One possible explanation for the observed inhibition of anti-CD3 mAb stimulation of Tcells by synergistic concentrations of DEX and PGE2 involves the potential for positive cross-talk between the distinct signaling cascades linked to DEX, and PGE2 , with those emanating from the TCR/CD3. The earliest known signaling event linked to the TCR/CD3 is the activation of protein tyrosine kinases (PTK) which occurs within seconds after receptor perturbation and is essential for the induction of the subsequent biochemical signals (26). Thus, experiments were performed to determine the effect of DEX and PGE2 on anti-CD3 mAb-induced tyrosine phosphorylation events. The data shown in Fig. 5 indicate that neither DEX nor PGE2 when added individually, even at high concentrations (1006 M), significantly alter anti-CD3 mAb-induced protein tyrosine phosphorylation (compare lanes 2–6). However, concurrent addition of low concentrations (1008 M) of DEX and PGE2 (lane 7) reduced anti-CD3 mAb-induced protein tyrosine phosphorylation to that observed in resting T-cells (lane 1).
FIG. 4. Stimulation of the EP3 subtype of the PGE2 R with specific agonist synergizes with DEX to inhibit anti-CD3 mAb-induced proliferation of T-cells. Human T-cells (5 1 105 cells/ml) were stimulated as described under Materials and Methods for 72 hr with immobilized anti-CD3 mAb alone (control) and in the presence of the indicated concentrations of DEX and/or PGE2 , MB28.767 (EP3 agonist), or butaprost (EP2 agonist). The percentage inhibition of the control response and SI were calculated as previously described (11). The results are from a representative of three experiments where the mean cpm { SEM of the control response was 59,433 { 11,088.
/ 6c0d$$8090
03-08-96 11:23:11
cias
AP: Cell Immuno
122
ELLIOTT ET AL.
FIG. 5. DEX and PGE2 inhibit protein tyrosine phosphorylation events induced in anti-CD3 mAb-stimulated T-cells. Postnuclear lysates were obtained from unstimulated human T-cells (lane 1) and human T-cells, stimulated as described under Materials and Methods for 2 min with anti-CD3 mAb alone (lane 2) or in the presence of 1008 M DEX (lane 3), 1006 M DEX (lane 4), 1008 M PGE2 (lane 5), 1006 M PGE2 (lane 6), or 1008 M DEX / 1008 M PGE2 (lane 7), were separated by SDS–PAGE (10% acrylamide). After transfer to nitrocellulose, tyrosine phosphorylated proteins were analyzed by ECL Western detection utilizing RC-20 mAb to phosphotyrosine residues. The results are from a representative of five experiments.
Synergistic concentrations of DEX and the EP3 PGE2R-specific agonist MB28.767 inhibit protein tyrosine phosphorylation events elicited by anti-CD3 mAb. Because the EP3 agonist MB28.767 mimics the synergistic action of PGE2 and DEX on T-cell proliferation, experiments were performed to determine whether MB28.767 and DEX alter anti-CD3 mAb-induced protein tyrosine phosphorylation. Figure 6 compares the effects of DEX and PGE2 with those elicited by MB28.767 and DEX on tyrosine phosphorylation of substrate protein in anti-CD3 mAb-stimulated T-cells. The results show that the addition of 1006 M MB28.767 concurrently with 1008 M DEX (lane 7) inhibits antiCD3 mAb-induced protein tyrosine phosphorylation and mimics the results obtained with 1008 M PGE2 and DEX (lane 6). DISCUSSION We have previously reported that low (1009 to 1008 M) concentrations of DEX and PGE2 , which individually are minimally immunosuppressive, act synergistically to inhibit anti-CD3 mAb-induced IL-2 secretion and proliferation of human T-cells (12). Although DEX and PGE2 mediate their actions via distinct signaling pathways, we have hypothesized that cross-talk between the signaling pathways induced by DEX, PGE2 ,
/ 6c0d$$8090
03-08-96 11:23:11
cias
and TCR/CD3 converges at the level of IL-2 gene regulation. Thus, the synergistic inhibitory effect of DEX and PGE2 on the induction of IL-2 mRNA in anti-CD3 mAb-stimulated T-cells could result from negative cross-talk between signaling events induced by these agents with those coupled to the TCR/CD3 complex. This would ultimately result in the failure of one or more important IL-2 gene transcription factors to be induced or appropriately modified. In the present paper, we have begun to dissect the biochemical mechanisms which are involved in the synergistic effects of DEX and PGE2 on T-cell activation. There is precedence for cross-talk between the distinct signaling cascades induced in T-cells by DEX, PGE2 , and anti-CD3 mAb. For example, costimulation of the PGE2R and the TCR/CD3 synergistically increases PGE2-induced accumulation of cAMP (24). Thus, signaling events initiated by ligands binding to these two receptors (PGE2R and TCR/CD3) can bidirectionally affect each other resulting in the downregulation of activation signals induced through stimulation of the TCR/CD3 complex. Additionally, high (1006 M) concentrations of PGE2 or DEX have been reported to inhibit tyrosine phosphorylation of a 100-kDa protein substrate elicited by stimulation of the TCR/CD3 complex (27, 28). These data suggest that signaling events initiated by ligands binding to either of these two receptors (PGE2R or GR) can directly interfere with activation signals induced through stimulation of the TCR/ CD3 complex. Finally, cross-talk may occur between the signaling pathways induced by DEX and PGE2 resulting in the amplification of PGE2- or DEX-induced second messengers which in turn inhibit TCR/CD3-coupled activation signals. Indeed, suprahigh concentrations (1004 M) of GS have been demonstrated to increase cAMP in PGE2-treated human lymphocytes (23). Although we have demonstrated that 1008 M DEX does not enhance cAMP accumulation in T-cells stimulated with 1008 M PGE2 (11), this does not preclude the possibility that DEX can amplify the cAMP response of Tcells concurrently stimulated with PGE2 and anti-CD3 mAb. Our data indicate that the synergistic immunosuppressive effect of DEX and PGE2 on anti-CD3 mAbinduced T-cell proliferation does not result from increased levels of cAMP. This is further supported by the results obtained when specific agonists for the EP2 or the EP3 subtype of the PGE2R were substituted for PGE2 . We demonstrate that the synergism observed with PGE2 and DEX is correlated with signaling events induced by stimulation of the EP3 subtype of the PGE2R which is not coupled to adenylyl cyclase and cAMP generation. Collectively, these data indicate that the synergistic inhibition of the proliferative response of Tcells to anti-CD3 mAb by low concentrations of DEX and PGE2 occurs through a cAMP-independent pathway. Additionally, the observed synergism of PGE2 and
AP: Cell Immuno
MODULATION OF T-CELL RECEPTOR SIGNALING
FIG. 6. EP3 agonist, MB28.767, and DEX inhibit protein tyrosine phosphorylation events induced in anti-CD3 mAb-stimulated T-cells. Postnuclear lysates from unstimulated human T-cells (lane 1) and human T-cells, stimulated as described under Materials and Methods with anti-CD3 mAb alone (lane 2) or in the presence of 1008 M DEX (lane 3), 1008 M PGE2 (lane 4), 1006 M MB28.767 (lane 5), 1008 M DEX / 1008 M PGE2 (lane 6), or 1008 M DEX / 1006 M MB28.767 (lane 7), were separated by SDS–PAGE (10% acrylamide). After transfer, tyrosine phosphorylated proteins were analyzed as described in the legend to Fig. 5. The results are from a representative of three experiments.
123
The binding of low concentrations of GS and PGE2 to their specific receptors on T-cells results in the induction of distinct signaling pathways which may go unnoticed unless an activation signal is concurrently received from the TCR/CD3 complex. The signaling pathways induced by low concentrations of GS or PGE2 individually do not interfere with T-cell activation events, but when the ligands simultaneously occupy their receptors, positive cross-talk between GR- and PGE2R-coupled signaling pathways results in an amplified signal which interferes with early transmembrane tyrosine phosphorylation events linked to stimulation of the TCR/CD3 complex. This effectively results in the ablation of downstream biochemical events in the activation cascade which are required for the induction of one or more important IL-2 transcription factors. While the signaling pathway induced by the stimulation of the GS and PGE2R on T-cells remains to be established, the data presented indicate that a cAMPindependent pathway is involved in the observed synergistic inhibition of anti-CD3 mAb-induced T-cell activation by low concentrations of DEX and PGE2 . ACKNOWLEDGMENTS
DEX on T-cell activation cannot be attributed to the induction of apoptosis by these agents. The earliest transmembrane signaling event elicited by the stimulation of the TCR/CD3 complex involves the induction of protein tyrosine kinase (PTK) activity (26) which is crucial for initiation of the subsequent signal transduction cascade and T-cell activation (29). Thus, inhibition of tyrosine kinase activity in anti-CD3 mAb-stimulated T-cells will result in the ablation of subsequent biochemical events required for the induction of a number of key IL-2 transcription factors (29). Data from the present study demonstrate that the addition of either DEX or PGE2 , even at 1006 M, does not significantly inhibit anti-CD3 mAb-induced protein tyrosine phosphorylation in T-cells. However, the concurrent addition of 1008 M of these two agents reduces tyrosine phosphorylation of substrate proteins in antiCD3 mAb-stimulated T-cells to that observed in resting cells. There are at least two possible explanations for this observation, based on the potential for positive cross-talk between PGE2- and DEX-induced signaling events. First, amplification of DEX- and PGE2-induced second messengers may uncouple the PTK(s) from the TCR/CD3 complex preventing subsequent phosphorylation events. Alternatively, the amplified second messengers linked to DEX and PGE2 may upregulate protein tyrosine phosphatases which counter the actions of PTKs. Experiments are ongoing in the laboratory to address these possibilities. Based on the data presented here together with those from previous studies (12), a model can be constructed.
/ 6c0d$$8090
03-08-96 11:23:11
cias
This work was supported by National Institutes of Health Grants MH50983, MH47679, and NS17423 and a Research Scientist Award (K05MH01069) to T. Roszman.
REFERENCES 1. Felten, D. L., Cohen, N., Ader, R., Felten, E., Carlson, S. L., and Roszman, T. L., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 2. Savino, W., and Dardenne, M., Immunol. Today 16, 318, 1995. 3. Keller, S. E., Schleifer, S. J., and Demetrikopolous, M. K., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 4. Kiecolt-Glaser, J. K., and Glaser, R., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 5. Munck, A., and Guyre, P. M., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 6. Ottaway, C. A., and Husband, A. J., Immunol. Today 15, 511, 1994. 7. Boumpas, D. T., Anastassiou, E. D., Older, S. A., Tsokos, G. C., Nelson, D. L., and Balow, J. E., J. Clin. Invest. 87, 1738, 1991. 8. Northrop, J. P., Crabtree, G. R., and Mattila, P. S., J. Exp. Med. 175, 1235, 1992. 9. Coleman, R. A., Smith, W. L., and Narumiya, S., Pharmacol. Rev. 46, 205, 1994. 10. Paliogianni, F., Kincaid, R. L., and Boumpas, D. T., J. Exp. Med. 178, 1813, 1993. 11. Elliott, L., Brooks, W., and Roszman, T., Cell. Mol. Neurobiol. 12, 411, 1992. 12. Elliott, L., Brooks, W., and Roszman, T., Cell. Mol. Neurobiol. 13, 579, 1993. 13. Elliott, L. H., Carlson, S. L., Morford, L. A., and McGillis, J. P., in ‘‘Methods in Neuroscience: Neuroimmunology,’’ Vol. 14, p. 115. Academic Press, San Diego, 1995.
AP: Cell Immuno
124
ELLIOTT ET AL.
14. Pollok, K. E., and Snow, E. C., Cell. Signaling 3, 435, 1991. 15. Chomcyznski, P., and Sacchi, N., Anal. Biochem. 162, 156, 1987. 16. Clark, C. C., Arya, S. K., Wong-Staal, F., Matsumoto-Kobayashi, M., Kay, R. M., Kaufman, R. J., Brown, E. L., Shoemaker, C., Copeland, T., Oroszlan, S., Smith, K., Sarngadharan, M. G., Lindner, S. G., and Gallo, R. C., Proc. Natl. Acad. Sci. USA 81, 2543, 1984. 17. Kyprianou, N., and Isaacs, J. T., Biochem. Biophys. Res. Commun. 165, 73, 1989. 18. Kanner, S. B., Deans, J. P., and Ledbetter, J. A., Immunology 75, 441, 1992. 19. Laemmli, U. K., Nature 227, 680, 1970. 20. Bansal, N., Houle, A., and Melnykovych, G., FASEB J. 5, 211, 1991. 21. Wesselborg, S., Janssen, O., and Kabelitz, D., J. Immunol. 150, 4338, 1993.
/ 6c0d$$8090
03-08-96 11:23:11
cias
22. Minakuchi, R., Wacholtz, M. C., David, L. S., and Lipsky, P. E., J. Immunol. 145, 2616, 1990. 23. Mendelsohn, J., Multer, M. M., and Boone, R. R., J. Clin. Invest. 52, 2129, 1973. 24. Roszman, T. L., and Carlson, S. L., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 25. Bartik, M. M., and Roszman, T. L., FASEB J. 8, A782, 1994. [Abstract] 26. June, C. H., Fletcher, M. H., Ledbetter, J. A., and Samelson, L. E., J. Immunol. 144, 1591, 1990. 27. Anastassiou, E. D., Paliogianni, F., Balow, J. P., Yamada, H., and Boumpas, D. T., J. Immunol. 148, 2845, 1992. 28. Paliogianni, F., Ahuja, S. S., Yamada, H., Balow, J. P., and Boumpas, D. T., Arthritis Rheum. 35, S127, 1992. [Abstract] 29. Weiss, A., Cell 73, 209, 1993.
AP: Cell Immuno
169, 117–124 (1996)
0099
Dexamethasone and Prostaglandin E2 Modulate T-Cell Receptor Signaling through a cAMP-Independent Mechanism LUCINDA H. ELLIOTT,1 AGATA K. LEVAY, BRAD SPARKS, MARIANNE MILLER,
AND
THOMAS L. ROSZMAN
Department of Microbiology and Immunology, College of Medicine, University of Kentucky, MS415 UKMC, Lexington, Kentucky 40536-0084 Received October 3, 1995; accepted December 7, 1995
One possible explanation for the link between stress and increased incidence of infection can be attributed to concomitant increases in levels of glucocorticoids (GS) and prostaglandin E2 (PGE2), both of which possess potent immunoregulatory activities. We have previously demonstrated that concentrations of PGE2 and the synthetic glucocorticoid, dexamethasone (DEX), which individually do not inhibit human T-cell responsiveness to anti-CD3 monoclonal antibody (mAb), act synergistically to inhibit IL-2 secretion and subsequent T-cell proliferation. In the present paper, we demonstrate that treatment of anti-CD3 mAb-stimulated T-cells with low (1008 and 1009 M) concentrations of DEX and PGE2 results in the inhibition of steadystate levels of IL-2 mRNA. Initial studies to elucidate the biochemical mechanisms involved indicate that the inhibitory effects of DEX and PGE2 cannot be correlated with increased levels of intracellular cAMP or the induction of apoptosis. However, the data indicate that DEX and PGE2 when added together interrupt anti-CD3 mAb-induced tyrosine phosphorylation of substrate proteins. Furthermore, the synergistic effect of DEX and PGE2 is mimicked by agonists for the cAMP-independent EP3 subtype of the PGE2 receptor. These data suggest that DEX and PGE2 elicit cAMPindependent signaling pathways which interact to inhibit the T-cell receptor-linked signal transduction cascade in anti-CD3 mAb-stimulated T-cells. q 1996 Academic Press, Inc.
INTRODUCTION There is compelling evidence to support bidirectional communication between the nervous and immune systems (1, 2). One example of this communication derives from studies which document stress-induced neural mediators inhibition of immune function (3, 4). Chief 1
To whom correspondence should be addressed.
among the neural mediators responsible for this immunosuppression are glucocorticoids (GS).2 They have been shown to have a profound effect on immunity including inhibition of lymphocyte proliferation (5), impairment of inflammatory cytokine production (5), and alterations in lymphocyte trafficking (5, 6). Collectively these studies suggest that one physiologic role of GS may be the homeostatic control of immune defense mechanisms and prevention of damage to normal tissue (5). Other physiologic modes of immunoregulation can be activated as a consequence of external stimuli. An example of this is a localized inflammatory response involving recruitment of phagocytic cells, including macrophages as well as immunocompetent T-cells to the site of antigen deposition. During this response another homeostatic mechanism becomes operant and involves the secretion of prostaglandin E2 (PGE2) by activated macrophages. As with GS, the immunomodulatory effects of PGE2 are evidenced by the inhibition of T-cell proliferation and cytokine production (7). The collective observations that GS and PGE2 possess potent immunoregulatory function and are concomitantly increased during stress and/or inflammation offer one hypothesis for the link between stress and increased susceptibility to infection. Hence normal and pathologic conditions may coexist where T-cells encounter immunomodulatory compounds which act synergistically to suppress their function. The immunoregulatory effects mediated by GS and PGE2 on human T-cells are correlated with negative regulation of IL-2 in stimulated T-cells (7, 8). Recent studies using the synthetic GS, dexamethasone (DEX), suggest that negative regulation of IL-2 gene transcription results from direct repression, via protein – protein interactions, of a key IL-2 DNA binding protein by ligand-activated glucocorticoid receptors 2 Abbreviations used: mAb, monoclonal antibody; IL-2, interleukin2; DEX, dexamethasone; PGE2 , prostaglandin E2 ; PGE2R, prostaglandin E2 receptor; GS, glucocorticoids; GR, glucocorticoid receptor; PTK, protein tyrosine kinase; TCR, T-cell receptor.
117
/ 6c0d$$8090
03-08-96 11:23:11
cias
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP: Cell Immuno
118
ELLIOTT ET AL.
(GR) (8). PGE2 mediates its effect by binding to a specific G-protein-linked seven transmembrane EP receptor, of which four subtypes (EP1 , EP2 , EP3 , and EP4) have been identified (reviewed in 9). Of these, only EP2 and EP3 are present on lymphoid cells. The inhibitory effect of PGE2 on IL-2 gene transcription has been attributed to increased levels of cAMP and the subsequent activation of protein kinase A (PKA) which interferes, via a phosphorylation event, with downstream signaling events coupled to the T-cell receptor (TCR/CD3) complex (10). However, these studies utilized high concentrations (1006 M ) of DEX or PGE2 , and hence may not reflect some in vivo situations. Because concomitant increased levels of GS and PGE2 can occur during stress and/or inflammation, we have previously determined the combined immunomodulatory effects of DEX, and PGE2 on Tcell activation. Thus, low concentrations (1009 to 1008 M ) of PGE2 and DEX, which individually do not inhibit T-cell responsiveness to anti-CD3 monoclonal antibody (mAb) act synergistically to inhibit T-cell proliferation (11). Furthermore, the synergistic antiproliferative effect of DEX and PGE2 is correlated with suppression of IL-2 synthesis by activated Tcells and can be reversed by the addition of recombinant IL-2 (12). In the present study, we have further investigated the mechanism(s) which contribute to the synergistic inhibitory effect of low concentrations of DEX and PGE2 on T-cell activation. The data indicate that addition of DEX and PGE2 to anti-CD3 mAb-stimulated T-cells are correlated with a synergistic decrease in steady-state levels of IL-2 mRNA. However, the synergistic antiproliferative effect of these compounds is not the result of apoptosis nor increased accumulation of cAMP. Rather, the data indicate that DEX and PGE2 when added together to anti-CD3 mAb-stimulated T-cells inhibit the induction of protein tyrosine phosphorylation events. Thus, these data indicate that low concentrations of DEX and PGE2 elicit biochemical events which synergistically interact to inhibit the TCR/CD3-stimulated signal transduction cascade. MATERIALS AND METHODS Isolation of purified resting T-cells. Peripheral blood leukocytes (PBL) were isolated from heparinized venous blood, obtained from healthy volunteers, by centrifugation over Ficoll–Hypaque gradients (13). Small resting T-cells were isolated on Percoll gradients (14) followed by rosetting with sheep erythrocytes (13). Percoll was diluted to a final density of 1.099 g/cm as follows. To prepare 100% Percoll solution, 716 ml of stock Percoll (Pharmacia, Piscataway, NJ) was mixed with 100 ml 1.5 M NaCl, 20 ml 1 M Na–Hepes, 164 ml tissue culture grade water, the pH adjusted to 7.45 and filter
/ 6c0d$$8090
03-08-96 11:23:11
cias
sterilized (personal communication from Dr. Carl June). The Percoll solution was further diluted in 11 RPMI 1640 to 80, 70, 60, and 50%. Percoll gradients were prepared by sequentially layering 3 ml of each Percoll dilution [80, 70, 60, 50%] into a 15-ml conical. The PBL (1 1 108) contained in 3 ml of RPMI were layered onto the top of the gradient and the gradients were centrifuged to equilibrium at 2000g for 30 min at 47C (14). The cells banding at the 60–70 and 80–90% interfaces were pooled, washed, and rosetted with sheep erythrocytes to obtain purified resting T-cells. The purity of these cells was greater than 95% as determined by immunofluorescence. Proliferation assays. Purified resting T-cells (1 1 105 cells/well) were stimulated in triplicate with immobilized anti-CD3 monoclonal antibody (mAb 500 ng/ well) in 96-well culture plates alone (control) and in the presence of low concentrations (1009 to 1008 M) of DEX (Sigma, St. Louis, MO) and/or PGE2 (Sigma) as previously described (11). In some of the wells the PGE2 receptor agonists, Butaprost (generously provided by Dr. Kluender, Miles Inc., West Haven, CT) and MB28.767 (generously provided by Rhone-Poulenc Rorer, Essex, England) at a final concentration of 1006 M were substituted for PGE2 . The plates were incubated for 72 hr in a moist 5% CO2 atmosphere, with 2.5 mCi of [3H]thymidine (specific activity 6.7 Ci/mM) added during the last 5 hr. The cells were harvested using a Skatron multiwell harvester (Skatron, Sterling, VA) and the cpm of [3H]thymidine was incorporated per well determined by scintillation counting. The percentage inhibition of the control proliferative response in each treatment group was determined as previously described (11). The synergistic index (SI) was calculated as previously described (11). Quantitation of steady-state levels of IL-2 mRNA. Purified T-cells (5 1 106 cells/well) were stimulated in 24-well culture plates with immobilized anti-CD3 mAb (7.5 mg/well) alone (control) and in the presence of low concentrations of DEX and/or PGE2 . The plates were incubated as previously described for 6 hr and the cells from five wells (2.5 1 107) were pooled from each treatment group for RNA extraction. Total cellular RNA was extracted by the method of Chomcyznski and Sacchi (15) and steady-state levels of IL-2 mRNA were quantitated by Northern analysis utilizing 40 – 60 mg of total RNA per treatment group. Briefly, the RNA was electrophoresed on a 1% agarose gel containing formaldehyde, transferred to nitrocellulose, and hybridized at 427C for 48 hr with the IL-2 cDNA clone pTCGF-11 (ATCC, Rockville, MD). The pTCGF-11 800-bp cDNA clone, which corresponds to the entire coding sequence of human IL-2 (16) was labeled with [a-32P]dCTP by random priming using a random-primed DNA labeling kit (Boeh-
AP: Cell Immuno
119
MODULATION OF T-CELL RECEPTOR SIGNALING
ringer Mannheim, Indianapolis, IN). The autoradiograms were analyzed using the Visage 2000 Bioimage Analyzer to determine the relative levels of IL-2 mRNA in each lane. The blots were stripped and hybridized with a 32P-labeled cDNA probe for ribosomal RNA (rRNA, generous gift from Dr. Jackie Featherstone, Univ. of Kentucky, Lexington, KY) to verify that similar amounts of total RNA were loaded per well. Qualitative analysis of DNA fragmentation. Purified T-cells (5 1 106 cells/well) were stimulated in 24well culture plates as described above. The plates were incubated for 24 hr and the cells from two wells pooled from each treatment group for DNA extraction. A modification of the method of Kyprianou and Isaacs (17) was used to assay for DNA fragmentation in each treatment group. Briefly, 1 1 107 cells from each treatment group were resuspended in 1 ml of lysis buffer (10 mM Tris– HCl, pH 8.0, 0.5% Triton X-100, 5 mM EDTA, 0.5% SDS, and 300 mg/ml proteinase K) and incubated for 18 hr in 377C water bath followed by 1 hr incubation at 377C with RNase A (40 mg/ml). The lysates were extracted sequentially, twice with equal volumes of buffer (10 mM Tris–HCl, pH 8.0)-saturated phenol, twice with phenol:chloroform:isoamyl alcohol (25:24:1), and twice with chloroform:isoamyl alcohol (24:1). The DNA was precipitated overnight at 0207C with 2 vol of absolute ethanol in the presence of 10 mM MgCl2 and 0.3 M sodium acetate, pH 5.2. Precipitated DNA was pelleted by centrifugation at 15,000g for 30 min at 47C, washed with 70% ethanol, and vacuum dried and the pellet resuspended in TE buffer (10 mM Tris–HCl, pH 8.0, 1 mM EDTA). The DNA concentration was determined spectrophotometrically and the DNA samples (10 mg/well) were analyzed by electrophoresis on a 1.6% agarose gel using TAE running buffer (40 mM Tris– acetate, 1 mM EDTA, pH 7.8). DNA fragmentation was visualized under a UV light after staining with ethidium bromide (0.5 mg/ml). Tyrosine phosphorylation and Western analysis. Resting T-cells (5 1 106/cells) in a final volume of 250 ml of RPMI were preincubated for 10 min on ice with soluble anti-CD3 mAb (10 mg) alone (control) and in the presence of low concentrations of DEX and/or PGE2 to allow ligand binding to come to equilibrium. The Eppendorf tubes were then transferred to a 377C water bath and the anti-CD3 mAb was cross-linked by the addition of 20 ml of goat anti-mouse IgG antibody (Organon Teknika Corp, Durham, NC). After the addition of cross-linking antibody, the cells were stimulated for 2 min and the reaction was stopped by the addition of 1/10 vol of 101 RIPA lysis buffer containing phosphatase and protease inhibitors (18) and incubation on ice for 15 min. After centrifugation at 12,000 rpm for 15 min, the postnuclear lysates (1 1 106 cell equivalents)
/ 6c0d$$8090
03-08-96 11:23:11
cias
were mixed with 51 Laemmli sample buffer (19) and boiled for 5 min and the proteins were separated by SDS–PAGE (10% acrylamide). The proteins were transferred to 0.2 nitrocellulose and the resulting blots were blocked for 2 hr with 20 mM Tris-buffered saline (TBS) containing 0.3% Tween 20 and 3% bovine serum albumin (BSA). The blots were incubated with gentle rocking at RT overnight with horseradish peroxidase (HRPO)-conjugated recombinant anti-phosphotyrosine mAb (RC-20, Transduction Laboratories, Lexington, KY) diluted according to manufacturer’s directions. Positive protein bands were detected by enhanced chemiluminescence (ECL; Amersham, Arlington Heights, IL) according to the manufacturer’s directions. RESULTS Low concentrations of DEX and PGE2 synergize to inhibit steady-state levels of IL-2 mRNA induced by anti-CD3 mAb stimulation of T-cells. We have previously reported that the synergistic antiproliferative effect of low concentrations (1009 to 1008 M) of DEX and PGE2 on anti-CD3 mAb-stimulated human T-cells is correlated with the inhibition of IL-2 synthesis (12). Experiments were performed to determine the individual and combined effects of 1009 M of DEX and PGE2 on anti-CD3 mAb-induced steady-state levels of IL-2 mRNA in T-cells. Results indicated that peak levels of IL-2 mRNA are observed in human T-cells 6 hr after stimulation with immobilized anti-CD3 mAb and decline thereafter over the next 18 hr (data not shown). Thus, IL-2 mRNA was quantitated 6 hr after anti-CD3 mAb stimulation of T-cells in subsequent experiments. The data indicate that while neither DEX nor PGE2 alone significantly inhibits the induction of IL-2 mRNA in anti-CD3 mAb-stimulated T-cells, simultaneous addition of these substances does inhibit the induction of IL-2 mRNA, with a synergistic index of 1.43 (Fig. 1). The synergistic inhibitory effect of DEX and PGE2 on anti-CD3 mAb induced activation of T-cells is not the result of apoptosis. DEX or anti-CD3 mAb stimulation has been shown to induce apoptosis in thymocytes, Tcell leukemia and T-cell clones (20). Although freshly isolated peripheral blood T-cells are largely resistant to the induction of apoptosis by anti-CD3 mAb (21), the possibility remains that apoptosis could occur as the result of the synergistic action of anti-CD3 mAb, DEX, and PGE2 on these cells. Furthermore, the observed reversible effect of exogenously added rIL-2 on the inhibitory actions of these agents could be attributed to the ability of IL-2 to rescue T-cells from apoptosis (12). Experiments were performed to determine whether concurrent stimulation of T-cells with anti-CD3 mAb, DEX, and PGE2 induced apoptosis, as evidenced by
AP: Cell Immuno
120
ELLIOTT ET AL.
FIG. 1. DEX and PGE2 synergistically inhibit steady-state levels of IL-2 mRNA in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 6 hr with immobilized anti-CD3 mAb alone (column 1) or in the presence of 1009 M DEX / PGE2 (column 2), 1009 M DEX (column 3), or 1009 M PGE2 (column 4). Steady-state levels of IL-2 mRNA were quantitated by Northern analysis utilizing 40–60 mg of total RNA from each sample. To normalize for RNA loading, the results are expressed as the ratio of the OD for IL-2 mRNA to 24S ribosomal RNA. The results are from a representative of three experiments.
Stimulation of the EP3 subtype of the PGE2 receptor mimics the synergy observed with PGE2 and DEX. Recent evidence from our laboratory suggests that human T-cells express at least two subtypes of receptors, EP2 and EP3 , for PGE2 (25). Functional studies utilizing specific agonists demonstrate that stimulation of the EP2 subtype of the PGE2 receptor (PGE2R) is correlated with a rise in intracellular cAMP, while stimulation of the EP3 subtype is not (9). However, stimulation of either PGE2R subtype with 1004 M of the appropriate agonists results in the inhibition of T-cell activation comparable to that observed with 1006 M PGE2 (25). Experiments were performed to determine whether the synergy observed with DEX and PGE2 is linked to a specific subtype of PGE2R human T-cells. Concentrations of specific agonists for the EP2 (Butaprost) and EP3 (MB28.767) subtypes of the PGE2R, which individually do not affect T-cell responsiveness to anti-CD3 mAb, were used in these studies. The data presented in Fig. 4 indicate that stimulation of the EP3 subtype of the PGE2R with the 1006 M of the EP3-specific agonist MB28.767 concurrently with 1008 M DEX results in the synergistic inhibition (SI Å 1.9) of anti-CD3 mAbinduced proliferation of T-cells which is comparable to that observed with 1008 M PGE2 (SI Å 2.0). However,
DNA fragmentation. As shown in Fig. 2 neither the stimulation of T-cells with anti-CD3 mAb alone (lane 5) nor that in the presence of 1008 M PGE2 (lane 4), 1008 M DEX (lane 3), or 1008 M PGE2 / 1008 M DEX (lane 2) induced DNA fragmentation. Mouse thymocytes stimulated for 24 hr with 1007 M DEX were used as a positive control in these experiments (lane 1). Identical results were obtained using T-cells stimulated with anti-CD3 mAb for 48 hr in the presence of as much as 1006 M DEX (data not shown). Increased cAMP accumulation is not responsible for the synergistic antiproliferative effect of DEX and PGE2 on anti-CD3 mAb-stimulated T-cells. It is generally accepted that high concentrations (1006 M) of PGE2 inhibit T-cell activation by increasing intracellular levels of cAMP (22). Although DEX mediates its effects via a cytosolic steroid receptor not linked to cAMP induction, suprahigh concentrations (1004 M) have been shown to upregulate PGE2-induced cAMP levels in human lymphocytes (23). Moreover, stimulation of T-cells with anti-CD3 mAb can amplify the cAMP response to high concentrations of PGE2 (24). Thus, experiments were performed to determine whether the concurrent stimulation of T-cells with low concentrations of DEX, PGE2 , and anti-CD3 mAb induced a synergistic rise in cAMP. The results presented in Fig. 3 show that 1008 M DEX does not amplify the cAMP response to 1008 M PGE2 in anti-CD3 mAb-stimulated T-cells.
/ 6c0d$$8090
03-08-96 11:23:11
cias
FIG. 2. DEX and PGE2 do not induce DNA fragmentation in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 24 hr with immobilized anti-CD3 mAb alone (lane 5) or in the presence of 1008 M PGE2 (lane 4), 1008 M DEX (lane 3), or DEX / PGE2 (lane 2) and the extracted DNA was analyzed for fragmentation on a 1.6% agarose gel. DNA extracted from mouse thymocytes stimulated with 1007 M DEX (lane 1) was used as a positive control for DNA fragmentation. The results are from a representative of three experiments.
AP: Cell Immuno
MODULATION OF T-CELL RECEPTOR SIGNALING
121
signaling events elicited by the EP2-specific agonist, butaprost, do not synergize with DEX.
FIG. 3. DEX does not increase PGE2-induced cytosolic levels of cAMP in stimulated T-cells. Human T-cells were stimulated as described under Materials and Methods for 10 min with biotinylated anti-CD3 mAb cross-linked with streptavidin alone (column 2) or in the presence of 1008 M DEX (column 3), 1008 M PGE2 (column 4), or 1008 M DEX / 1008 M PGE2 (column 5). Cytosolic levels of cAMP for resting T-cells are in column 1. The results are from a representative of two experiments.
Synergistic concentrations of DEX and PGE2 inhibit protein tyrosine phosphorylation in anti-CD3 mAbstimulated T-cells. One possible explanation for the observed inhibition of anti-CD3 mAb stimulation of Tcells by synergistic concentrations of DEX and PGE2 involves the potential for positive cross-talk between the distinct signaling cascades linked to DEX, and PGE2 , with those emanating from the TCR/CD3. The earliest known signaling event linked to the TCR/CD3 is the activation of protein tyrosine kinases (PTK) which occurs within seconds after receptor perturbation and is essential for the induction of the subsequent biochemical signals (26). Thus, experiments were performed to determine the effect of DEX and PGE2 on anti-CD3 mAb-induced tyrosine phosphorylation events. The data shown in Fig. 5 indicate that neither DEX nor PGE2 when added individually, even at high concentrations (1006 M), significantly alter anti-CD3 mAb-induced protein tyrosine phosphorylation (compare lanes 2–6). However, concurrent addition of low concentrations (1008 M) of DEX and PGE2 (lane 7) reduced anti-CD3 mAb-induced protein tyrosine phosphorylation to that observed in resting T-cells (lane 1).
FIG. 4. Stimulation of the EP3 subtype of the PGE2 R with specific agonist synergizes with DEX to inhibit anti-CD3 mAb-induced proliferation of T-cells. Human T-cells (5 1 105 cells/ml) were stimulated as described under Materials and Methods for 72 hr with immobilized anti-CD3 mAb alone (control) and in the presence of the indicated concentrations of DEX and/or PGE2 , MB28.767 (EP3 agonist), or butaprost (EP2 agonist). The percentage inhibition of the control response and SI were calculated as previously described (11). The results are from a representative of three experiments where the mean cpm { SEM of the control response was 59,433 { 11,088.
/ 6c0d$$8090
03-08-96 11:23:11
cias
AP: Cell Immuno
122
ELLIOTT ET AL.
FIG. 5. DEX and PGE2 inhibit protein tyrosine phosphorylation events induced in anti-CD3 mAb-stimulated T-cells. Postnuclear lysates were obtained from unstimulated human T-cells (lane 1) and human T-cells, stimulated as described under Materials and Methods for 2 min with anti-CD3 mAb alone (lane 2) or in the presence of 1008 M DEX (lane 3), 1006 M DEX (lane 4), 1008 M PGE2 (lane 5), 1006 M PGE2 (lane 6), or 1008 M DEX / 1008 M PGE2 (lane 7), were separated by SDS–PAGE (10% acrylamide). After transfer to nitrocellulose, tyrosine phosphorylated proteins were analyzed by ECL Western detection utilizing RC-20 mAb to phosphotyrosine residues. The results are from a representative of five experiments.
Synergistic concentrations of DEX and the EP3 PGE2R-specific agonist MB28.767 inhibit protein tyrosine phosphorylation events elicited by anti-CD3 mAb. Because the EP3 agonist MB28.767 mimics the synergistic action of PGE2 and DEX on T-cell proliferation, experiments were performed to determine whether MB28.767 and DEX alter anti-CD3 mAb-induced protein tyrosine phosphorylation. Figure 6 compares the effects of DEX and PGE2 with those elicited by MB28.767 and DEX on tyrosine phosphorylation of substrate protein in anti-CD3 mAb-stimulated T-cells. The results show that the addition of 1006 M MB28.767 concurrently with 1008 M DEX (lane 7) inhibits antiCD3 mAb-induced protein tyrosine phosphorylation and mimics the results obtained with 1008 M PGE2 and DEX (lane 6). DISCUSSION We have previously reported that low (1009 to 1008 M) concentrations of DEX and PGE2 , which individually are minimally immunosuppressive, act synergistically to inhibit anti-CD3 mAb-induced IL-2 secretion and proliferation of human T-cells (12). Although DEX and PGE2 mediate their actions via distinct signaling pathways, we have hypothesized that cross-talk between the signaling pathways induced by DEX, PGE2 ,
/ 6c0d$$8090
03-08-96 11:23:11
cias
and TCR/CD3 converges at the level of IL-2 gene regulation. Thus, the synergistic inhibitory effect of DEX and PGE2 on the induction of IL-2 mRNA in anti-CD3 mAb-stimulated T-cells could result from negative cross-talk between signaling events induced by these agents with those coupled to the TCR/CD3 complex. This would ultimately result in the failure of one or more important IL-2 gene transcription factors to be induced or appropriately modified. In the present paper, we have begun to dissect the biochemical mechanisms which are involved in the synergistic effects of DEX and PGE2 on T-cell activation. There is precedence for cross-talk between the distinct signaling cascades induced in T-cells by DEX, PGE2 , and anti-CD3 mAb. For example, costimulation of the PGE2R and the TCR/CD3 synergistically increases PGE2-induced accumulation of cAMP (24). Thus, signaling events initiated by ligands binding to these two receptors (PGE2R and TCR/CD3) can bidirectionally affect each other resulting in the downregulation of activation signals induced through stimulation of the TCR/CD3 complex. Additionally, high (1006 M) concentrations of PGE2 or DEX have been reported to inhibit tyrosine phosphorylation of a 100-kDa protein substrate elicited by stimulation of the TCR/CD3 complex (27, 28). These data suggest that signaling events initiated by ligands binding to either of these two receptors (PGE2R or GR) can directly interfere with activation signals induced through stimulation of the TCR/ CD3 complex. Finally, cross-talk may occur between the signaling pathways induced by DEX and PGE2 resulting in the amplification of PGE2- or DEX-induced second messengers which in turn inhibit TCR/CD3-coupled activation signals. Indeed, suprahigh concentrations (1004 M) of GS have been demonstrated to increase cAMP in PGE2-treated human lymphocytes (23). Although we have demonstrated that 1008 M DEX does not enhance cAMP accumulation in T-cells stimulated with 1008 M PGE2 (11), this does not preclude the possibility that DEX can amplify the cAMP response of Tcells concurrently stimulated with PGE2 and anti-CD3 mAb. Our data indicate that the synergistic immunosuppressive effect of DEX and PGE2 on anti-CD3 mAbinduced T-cell proliferation does not result from increased levels of cAMP. This is further supported by the results obtained when specific agonists for the EP2 or the EP3 subtype of the PGE2R were substituted for PGE2 . We demonstrate that the synergism observed with PGE2 and DEX is correlated with signaling events induced by stimulation of the EP3 subtype of the PGE2R which is not coupled to adenylyl cyclase and cAMP generation. Collectively, these data indicate that the synergistic inhibition of the proliferative response of Tcells to anti-CD3 mAb by low concentrations of DEX and PGE2 occurs through a cAMP-independent pathway. Additionally, the observed synergism of PGE2 and
AP: Cell Immuno
MODULATION OF T-CELL RECEPTOR SIGNALING
FIG. 6. EP3 agonist, MB28.767, and DEX inhibit protein tyrosine phosphorylation events induced in anti-CD3 mAb-stimulated T-cells. Postnuclear lysates from unstimulated human T-cells (lane 1) and human T-cells, stimulated as described under Materials and Methods with anti-CD3 mAb alone (lane 2) or in the presence of 1008 M DEX (lane 3), 1008 M PGE2 (lane 4), 1006 M MB28.767 (lane 5), 1008 M DEX / 1008 M PGE2 (lane 6), or 1008 M DEX / 1006 M MB28.767 (lane 7), were separated by SDS–PAGE (10% acrylamide). After transfer, tyrosine phosphorylated proteins were analyzed as described in the legend to Fig. 5. The results are from a representative of three experiments.
123
The binding of low concentrations of GS and PGE2 to their specific receptors on T-cells results in the induction of distinct signaling pathways which may go unnoticed unless an activation signal is concurrently received from the TCR/CD3 complex. The signaling pathways induced by low concentrations of GS or PGE2 individually do not interfere with T-cell activation events, but when the ligands simultaneously occupy their receptors, positive cross-talk between GR- and PGE2R-coupled signaling pathways results in an amplified signal which interferes with early transmembrane tyrosine phosphorylation events linked to stimulation of the TCR/CD3 complex. This effectively results in the ablation of downstream biochemical events in the activation cascade which are required for the induction of one or more important IL-2 transcription factors. While the signaling pathway induced by the stimulation of the GS and PGE2R on T-cells remains to be established, the data presented indicate that a cAMPindependent pathway is involved in the observed synergistic inhibition of anti-CD3 mAb-induced T-cell activation by low concentrations of DEX and PGE2 . ACKNOWLEDGMENTS
DEX on T-cell activation cannot be attributed to the induction of apoptosis by these agents. The earliest transmembrane signaling event elicited by the stimulation of the TCR/CD3 complex involves the induction of protein tyrosine kinase (PTK) activity (26) which is crucial for initiation of the subsequent signal transduction cascade and T-cell activation (29). Thus, inhibition of tyrosine kinase activity in anti-CD3 mAb-stimulated T-cells will result in the ablation of subsequent biochemical events required for the induction of a number of key IL-2 transcription factors (29). Data from the present study demonstrate that the addition of either DEX or PGE2 , even at 1006 M, does not significantly inhibit anti-CD3 mAb-induced protein tyrosine phosphorylation in T-cells. However, the concurrent addition of 1008 M of these two agents reduces tyrosine phosphorylation of substrate proteins in antiCD3 mAb-stimulated T-cells to that observed in resting cells. There are at least two possible explanations for this observation, based on the potential for positive cross-talk between PGE2- and DEX-induced signaling events. First, amplification of DEX- and PGE2-induced second messengers may uncouple the PTK(s) from the TCR/CD3 complex preventing subsequent phosphorylation events. Alternatively, the amplified second messengers linked to DEX and PGE2 may upregulate protein tyrosine phosphatases which counter the actions of PTKs. Experiments are ongoing in the laboratory to address these possibilities. Based on the data presented here together with those from previous studies (12), a model can be constructed.
/ 6c0d$$8090
03-08-96 11:23:11
cias
This work was supported by National Institutes of Health Grants MH50983, MH47679, and NS17423 and a Research Scientist Award (K05MH01069) to T. Roszman.
REFERENCES 1. Felten, D. L., Cohen, N., Ader, R., Felten, E., Carlson, S. L., and Roszman, T. L., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 2. Savino, W., and Dardenne, M., Immunol. Today 16, 318, 1995. 3. Keller, S. E., Schleifer, S. J., and Demetrikopolous, M. K., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 4. Kiecolt-Glaser, J. K., and Glaser, R., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 5. Munck, A., and Guyre, P. M., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 6. Ottaway, C. A., and Husband, A. J., Immunol. Today 15, 511, 1994. 7. Boumpas, D. T., Anastassiou, E. D., Older, S. A., Tsokos, G. C., Nelson, D. L., and Balow, J. E., J. Clin. Invest. 87, 1738, 1991. 8. Northrop, J. P., Crabtree, G. R., and Mattila, P. S., J. Exp. Med. 175, 1235, 1992. 9. Coleman, R. A., Smith, W. L., and Narumiya, S., Pharmacol. Rev. 46, 205, 1994. 10. Paliogianni, F., Kincaid, R. L., and Boumpas, D. T., J. Exp. Med. 178, 1813, 1993. 11. Elliott, L., Brooks, W., and Roszman, T., Cell. Mol. Neurobiol. 12, 411, 1992. 12. Elliott, L., Brooks, W., and Roszman, T., Cell. Mol. Neurobiol. 13, 579, 1993. 13. Elliott, L. H., Carlson, S. L., Morford, L. A., and McGillis, J. P., in ‘‘Methods in Neuroscience: Neuroimmunology,’’ Vol. 14, p. 115. Academic Press, San Diego, 1995.
AP: Cell Immuno
124
ELLIOTT ET AL.
14. Pollok, K. E., and Snow, E. C., Cell. Signaling 3, 435, 1991. 15. Chomcyznski, P., and Sacchi, N., Anal. Biochem. 162, 156, 1987. 16. Clark, C. C., Arya, S. K., Wong-Staal, F., Matsumoto-Kobayashi, M., Kay, R. M., Kaufman, R. J., Brown, E. L., Shoemaker, C., Copeland, T., Oroszlan, S., Smith, K., Sarngadharan, M. G., Lindner, S. G., and Gallo, R. C., Proc. Natl. Acad. Sci. USA 81, 2543, 1984. 17. Kyprianou, N., and Isaacs, J. T., Biochem. Biophys. Res. Commun. 165, 73, 1989. 18. Kanner, S. B., Deans, J. P., and Ledbetter, J. A., Immunology 75, 441, 1992. 19. Laemmli, U. K., Nature 227, 680, 1970. 20. Bansal, N., Houle, A., and Melnykovych, G., FASEB J. 5, 211, 1991. 21. Wesselborg, S., Janssen, O., and Kabelitz, D., J. Immunol. 150, 4338, 1993.
/ 6c0d$$8090
03-08-96 11:23:11
cias
22. Minakuchi, R., Wacholtz, M. C., David, L. S., and Lipsky, P. E., J. Immunol. 145, 2616, 1990. 23. Mendelsohn, J., Multer, M. M., and Boone, R. R., J. Clin. Invest. 52, 2129, 1973. 24. Roszman, T. L., and Carlson, S. L., in ‘‘Psychoneuroimmunology,’’ 2nd ed. Academic Press, San Diego, 1991. 25. Bartik, M. M., and Roszman, T. L., FASEB J. 8, A782, 1994. [Abstract] 26. June, C. H., Fletcher, M. H., Ledbetter, J. A., and Samelson, L. E., J. Immunol. 144, 1591, 1990. 27. Anastassiou, E. D., Paliogianni, F., Balow, J. P., Yamada, H., and Boumpas, D. T., J. Immunol. 148, 2845, 1992. 28. Paliogianni, F., Ahuja, S. S., Yamada, H., Balow, J. P., and Boumpas, D. T., Arthritis Rheum. 35, S127, 1992. [Abstract] 29. Weiss, A., Cell 73, 209, 1993.
AP: Cell Immuno