The T cell migration stop signal is a central step in T cell activation and inflammation; however, its regulatory mechanisms remain largely unknown. Using a live-cell, imaging-based, high-throughput screen, we identified the PG, PGE2, as a T cell stop signal antagonist. Src kinase inhibitors, microtubule inhibitors, and PGE2 prevented the T cell stop signal, and impaired T cell–APC conjugation and T cell proliferation induced by primary human allogeneic dendritic cells. However, Src inhibition, but not PGE2 or microtubule inhibition, impaired TCR-induced ZAP-70 signaling, demonstrating that T cell stop signal antagonists can function either upstream or downstream of proximal TCR signaling. Moreover, we found that PGE2 abrogated TCR-induced activation of the small GTPase Rap1, suggesting that PGE2 may modulate T cell adhesion and stopping through Rap1. These results identify a novel role for PGs in preventing T cell stop signals and limiting T cell activation induced by dendritic cells.

T cells are highly migratory cells that travel at speeds up to 30 μm/min (1) and during inflammation can arrest their migration in response to receptor-mediated signals (2). T cells receive a migration “stop signal” and can rapidly halt migration after TCR signaling and interaction with APCs. Transient T cell stopping can also be sufficient to induce T cell activation under some conditions (3, 4), and more prolonged interactions can be associated with the generation of CD8+ memory T cells (5). In any case, the T cell stop signal is essential for some types of immune synapse formation and T cell activation (6), and represents an attractive therapeutic target. The molecular mechanisms controlling the T cell stop signal are not well understood but likely involve signaling through one or more of the TCR proximal kinases Lck, Fyn, and ZAP-70, and activation of the small GTPase Rap1 (7). Inhibiting proximal TCR signaling through ligation of the endogenous receptor CTLA-4 can reverse the T cell stop signal (8).

Despite its importance we have limited understanding of the signaling pathways that regulate T cell arrest induced by TCR engagement. In this study, we describe a live-cell, imaging-based, high-throughput method to identify signaling pathways that control the T cell stop signal induced by TCR ligation. Using live imaging, we can assess the kinetics that accompany the decision to stop or go in response to TCR engagement and we can identify small molecules that modify the kinetics of T cell stopping, and thereby may impact duration of T cell–APC interactions. This is especially important in light of recent studies that demonstrate that the duration of T cell–APC contacts can determine T cell fate and the development of T cell activation or tolerance (9).

In this study, we identified PGE2 as a novel regulator of T cell stopping and T cell–dendritic cell (DC) interactions. PGs are bioactive lipids that have been implicated in inflammation and are targeted by cyclooxygenase (COX) inhibitors commonly used to treat inflammatory disease (10, 11). However, the precise mechanisms by which PGs control inflammation are not well understood, and recent studies have suggested that specific PGs, notably PGE2, may have anti-inflammatory effects (12, 13). Our findings provide novel insight into how PGE2 may limit T cell activation by impairing T cell arrest and inhibiting T cell–DC interactions. Moreover, our studies demonstrate that PGE2, unlike Src kinase inhibition, alters T cell stopping downstream of ZAP-70 and linker of activated T cells (LAT) phosphorylation at the level of Rap1 GTPase activation, indicating that the T cell stop signal can be decoupled from proximal TCR signaling.

OKT3 Ab was purified from a B cell hybridoma line (14), and ICAM-1–Fc was purified from transfected CHO cells (15) by affinity to protein G-Sepharose. Calcein-acetoxymethyl ester (AM) was obtained from Invitrogen (Carlsbad, CA). FTY720 was from Cayman Chemical (Ann Arbor, MI). PHA, PP2, U-73122, and PGE2 were obtained from Fisher Scientific.

Leukocytes were obtained from whole blood (16) using Lymphoprep and resuspended in fresh T cell media (RPMI 1640, 10% heat-inactivated FBS, 1× HEPES, pyruvate, nonessential amino acids, β-mercaptoethanol). Cells were stimulated with PHA and expanded in the presence of IL-2 (50 U/ml; Chiron) for 5–10 d. For conjugation and proliferation assays, fresh human T cells were magnetically purified by a negative selection method using a pan T cell isolation kit (Miltenyi Biotec, Auburn, CA).

DCs were obtained as described previously (17, 18). Monocytes were purified by magnetic sorting with anti-CD14 beads (Miltenyi Biotec) and differentiated 3 d in RPMI 1640 media supplemented with 2 mM l-glutamine, 100 μg/ml penicillin/streptomycin, and 10% FBS containing 300 U/ml recombinant human GM-CSF (Berlex Labs) and 200 U/ml recombinant human IL-4 (PeproTech, Rocky Hill, NJ).

The 384-well tissue culture-treated plates (BD Optilux) were coated by addition of ICAM (20 μl to each well) at 5 μg/ml in coating buffer (Tris pH 9.5). Plates were incubated at 37°C for 1 h, washed, and blocked by addition of 50 μl blocking buffer (2% BSA in 1× PBS) for 1 h at 37°C.

Cells were suspended at 500,000/ml and labeled by addition of 1 μM calcein-AM for 15 min. Cells were collected by centrifugation and washed twice with media. Cells were plated and centrifuged at 500 × g for 5 min, then incubated for 30 min at 37°C in the presence of inhibitors. A total of 1 μg/ml OKT3 for was added for 10 min. Cells were aspirated and washed three to six times with pre-equilibrated fresh media. Adherent cells were quantified by plate reading.

We established conditions (modified from Ref. 8) for modeling the T cell stop signal in a high-throughput assay (Supplemental Table I). Human peripheral blood T (HPBT) cells were labeled with calcein-AM (Invitrogen). A total of 50 μl HPBT cells at 1 × 106/ml (after 5–14 d of expansion) was added to each well of a 384-well plate. Cells were plated and centrifuged at 500 × g for 5 min, then incubated for 5 min at 37°C. For low-throughput experiments, cells were washed twice using pre-equilibrated culture media, and test compounds were added by pipetting. For high-throughput experiments, cells were mixed six times with a robotic pipette and media exchanged once. Test compounds (6–12 μM) were added using a 384-well pin transfer device. The T cell stop signal was induced with soluble OKT3 (1 μg/ml), a CD3 Ab known to stimulate TCR signaling (14), for 10 min after plating and washing the cells. Migration was monitored using a BD Pathway microscope. Two images were acquired under 10× magnification at an interval of ∼7.5 min. Cells in the t = 0 image were outlined with a three-pixel dilation width to define the region of interest (ROI) for each cell, and intensity within the ROI was calculated. Background intensity was subtracted, and the percentage remaining intensity for each individual ROI was calculated according to the equation (IfinalBG)/(IinitialBG) = Iremaining, where I is pixel intensity and BG is background intensity. ROIs (cells) with Iremaining values >0.5 were deemed stopped, whereas those with values <0.5 were migrating. Hits were defined by having a percentage migration >40%. Treatment with OKT3 caused rapid depolarization, stopped migration within minutes, and changed the distribution of the population of T cells from primarily migratory to stopped (Supplemental Video 1). Typically, 60–80% of control cells were migrating compared with 5–20% of the OKT3-treated cells. This approach yielded an average Z′ factor of 0.5 and a 6.7-fold difference between positive and negative controls.

A total of 50 μl HPBT cells (after 5–14 d of expansion) was added to each well of a 384-well plate. Cells were plated and centrifuged at 500 × g for 5 min, then incubated for 5 min at 37°C. Cells were washed twice using pre-equilibrated culture media, and test compounds were added by pipetting. Cells were incubated with compounds for 30 min at 37°C. Stop signal was induced by addition of OKT3 Ab for 10 min at 37°C. Cells were placed on movable microscope stage in a climate-controlled chamber set at 37°C. Images were obtained using Metamorph, and data were analyzed by Microsoft Excel. For typical experiments, images were acquired for 15 min at 30- or 45-s intervals simultaneously for 8–12 experimental conditions. Migration was quantified by cell tracking; typically, 30 cells picked at random per video were tracked.

DCs were stimulated for 8 h with 250 ng/ml LPS (Sigma-Aldrich, St. Louis, MO). DCs were labeled with DiD (Invitrogen). T cells were labeled with 2.5 μM CFSE. T cells (1 × 105 cells in 100 μl) were treated for 30 min with indicated concentration of compounds, then mixed at a 1:1 ratio with DCs in media. Mixture was immediately centrifuged at 800 × g for 3 min and incubated at 37°C for the indicated time. Mixture was vortexed for 30 s and analyzed by flow cytometry.

T cells were labeled with 2.5 μM CFSE and cultured at a 50:1 ratio with allogeneic LPS-matured DCs with inhibitors or DMSO. Proliferation was assessed on day 6 by flow cytometry. The percentage of live T cells that had undergone cell division was determined by gating on DAPI CD3+ cells and assessing the fraction that showed diminished CFSE fluorescence intensity.

T cells (days 5–10) at 2 × 107/ml in 0.5–1 ml media were incubated with compounds for 30 min. Cells were placed on ice for 5 min, then coated with 1 μg/ml OKT3 on ice for 20 min. Cells were suspended in 100 μl media containing goat anti-mouse F(Ab′)2 at 37°C for 3 min. Cells were lysed with 500 μl ice-cold radioimmunoprecipitation assay buffer (25 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, 0.1% SDS). Lysing buffer contained freshly added phosphatase inhibitor mixture (1:100 dilution, P-5726; Sigma) and protease inhibitor mixture (1:100 dilution, P-8340; Sigma). Proteins were resolved by SDS-PAGE on 10% gels, transferred to nitrocellulose, and blotted with p-ZAP-70–Y319 (Cell Signaling), p-LAT-Y191 (Cell Signaling), p-Lck-Y505 (Cell Signaling), pY (4G10), vinculin (Sigma), p-Src-416 (Cell Signaling), p-Src-529 (Cell Signaling), or Fyn (Santa Cruz) Abs. Detection was performed using Alexa Fluor 680 goat anti-mouse IgG (Molecular Probes) and IRDye 800CW goat anti-rabbit IgG (Rockland).

Activated HPBT cells (25 × 106) were suspended in 1 ml fresh T cell media in the presence or absence of test compounds for 30 min at 37°C and stimulated with OKT3 as described earlier. Cells were lysed in 500 μl Rap1 lysis buffer (1% Triton X-100; 50 mM Tris-HCl, pH 7.5; 100 mM NaCl; 10 mM MgCl2; 1 mM PMSF; 1 mM leupeptin; 0.5 mM aprotinin) (19). Lysates were cleared by centrifugation (16,000 rpm for 10 min) and incubated with glutathione S-transferase-RalGDS-Rap-binding domain (Millipore) for 1 h at 4°C with rotation. Beads were washed three times with lysis buffer and subjected to Western blot analysis with anti-Rap1 (Santa Cruz). Twenty-five microliters lysate was reserved to use as a loading control.

Activated HPBT cells (1.5 × 106) in 1 ml T cell media were added to 24-well plates that had been precoated for 2 h with 1 μg/ml OKT3 in PBS. Cells were stimulated for 5 h at 37°C in the presence of brefeldin A (eBioscience). Cells were pelleted by centrifugation and resuspended in FACS buffer (2% FBS in PBS). Cells were labeled with allophycocyanin-CD4 (eBioscience) according to manufacturer’s protocol, washed, and fixed overnight with 4% paraformaldehyde. Cells were resuspended in FACS buffer containing 0.1% saponin for 15 min. Cells were blocked for 15 min with Fc block (eBioscience) and labeled for 30 min with FITC–IL-2 (eBioscience). Cells were washed twice with FACS buffer and analyzed by flow cytometry.

All columns in bar graphs represent the mean of the indicated number of replicates. Error bars on graphs represent SEMs. ANOVA with Tukey’s post hoc testing was used to calculate statistical significance. Unless otherwise indicated, comparisons were done relative to the control. An α level of 0.05 was used as the level of significance.

To identify novel regulators of the T cell migration stop signal, we developed an image-based, high-throughput screen for small molecules that impair TCR-induced T cell arrest (Fig. 1, Supplemental Table I, Supplemental Video 1). The T cell stop signal was induced in primary human T cells by treatment with the anti-CD3 Ab OKT3 as described in 1Materials and Methods (Fig. 1A). Soluble OKT3 was sufficient to induce the T cell stop signal without TCR cross-linking or costimulation with anti-CD28 Ab, which is generally required to induce full T cell activation and proliferation, suggesting that early TCR engagement may be sufficient to induce the T cell stop signal (Fig. 1B, Supplemental Video 1).

FIGURE 1.

Src kinase inhibitors prevent the T cell stop signal. A, Schematic of methods used to distinguish between motile and nonmotile cells. B, The fraction of migratory cells significantly decreased with OKT3 treatment and was prevented by addition of 5 μM PP2 (representative data from >15 experiments, 32 replicates each, mean ± SD). C, Src kinase inhibitors reversed the OKT-3–mediated T cell stop signal. Cells were pretreated with 5 μM Src inhibitors PP1, PP2, SKI-1, and SU6656 (mean ± STDEV; n = 3; *p < 0.05 by ANOVA). D, SU6656 inhibits T cell adhesion to ICAM-1. Cells were incubated for 30 min with indicated concentrations (mean ± SEM; n = 3; *p < 0.05 by ANOVA). E, PP2 and SU6656, but not SKI-1, inhibited phosphorylation of ZAP-70 and LAT. Cells were incubated for 30 min with compounds and stimulated for 3 min with OKT3 as described in 1Materials and Methods. Representative blots from three independent experiments are shown. F, Quantification of Western blots (mean ± SEM; n = 3; *p < 0.05, ANOVA with Tukey’s post hoc test).

FIGURE 1.

Src kinase inhibitors prevent the T cell stop signal. A, Schematic of methods used to distinguish between motile and nonmotile cells. B, The fraction of migratory cells significantly decreased with OKT3 treatment and was prevented by addition of 5 μM PP2 (representative data from >15 experiments, 32 replicates each, mean ± SD). C, Src kinase inhibitors reversed the OKT-3–mediated T cell stop signal. Cells were pretreated with 5 μM Src inhibitors PP1, PP2, SKI-1, and SU6656 (mean ± STDEV; n = 3; *p < 0.05 by ANOVA). D, SU6656 inhibits T cell adhesion to ICAM-1. Cells were incubated for 30 min with indicated concentrations (mean ± SEM; n = 3; *p < 0.05 by ANOVA). E, PP2 and SU6656, but not SKI-1, inhibited phosphorylation of ZAP-70 and LAT. Cells were incubated for 30 min with compounds and stimulated for 3 min with OKT3 as described in 1Materials and Methods. Representative blots from three independent experiments are shown. F, Quantification of Western blots (mean ± SEM; n = 3; *p < 0.05, ANOVA with Tukey’s post hoc test).

Close modal

Lck and Fyn both play a critical role in proximal TCR signal transduction (20). To determine whether inhibition of proximal TCR signal transduction was sufficient to block the stop signal, we pretreated T cells with the Src kinase inhibitor PP2, which inhibits both Lck and Fyn activity (20). PP2 had no effect on T cell random motility on ICAM-1 but blocked T cell arrest induced by TCR ligation with OKT3 (Fig. 1B). In the presence of both OKT3 and PP2, primary human T cells on ICAM-1 demonstrated rapid, random motility comparable with untreated control T cells on ICAM-1. The findings demonstrate that Src kinases are required for the T cell stop signal but not T cell random motility.

To determine whether blocking proximal TCR signal transduction using other Src family kinase inhibitors was also sufficient to block the TCR-induced stop signal, we used a panel of other Src inhibitors including PP1, SU6656, or SKI-1 (21). PP1 and SU6656, like PP2, impaired TCR-induced T cell arrest (Fig. 1C). Pretreatment with PP1, SU6656, or SKI-1 had no effect on T cell random motility on ICAM-1. However, SU6656, but not SKI-1, blocked the TCR stop signal and impaired adhesion of T cells to ICAM-1–coated plates (Fig. 1D). Accordingly, we also found that SU6656, but not SKI-1, inhibited phosphorylation of ZAP-70 and LAT under the conditions of our assay (Fig. 1E, 1F). Taken together, we found that the three different Src kinase inhibitors that impaired proximal TCR signaling also blocked the TCR-mediated T cell stop signal.

To identify novel signaling pathways involved in T cell stop signals, we screened a library of 1600 compounds for effects on OKT3-induced T cell arrest as described in 1Materials and Methods and Supplemental Table I (Fig. 2, Supplemental Table II). Cells were plated on ICAM-1–coated 384-well plates in the presence or absence of test compounds. Approximately 60% of control cells (green line) were actively migrating, whereas stimulation with OKT3 induced a stop signal, decreasing migration to ∼10% (Fig. 2A, red line). The majority of test compounds did not alter the ability of T cells to stop migrating (Fig. 2A, gray dots). After confirmation, five compounds, 0.31% of the total screened, prevented the T cell stop signal (Fig. 2A). The positive compounds included two PGs (PGE1 and PGE2) and three microtubule-disrupting compounds (colchicine, albendazole, and nocodazole; Table I). Representative time-lapse images for PP2, PGE2, and colchicine are shown in Fig. 2B. Control cells that were actively migrating do not colocalize at times t = 0 and t = 7.5, whereas cells stopped with OKT3 colocalized. Treatment with PP2, PGE2, or colchicine was sufficient to reverse colocalization and block T cell stopping.

FIGURE 2.

A high-throughput assay for the HPBT cell migration stop signal identifies PGE2 and microtubule inhibitors as T cell stop signal antagonists. A, PGE1, PGE2, colchicine, nocodazole, and albendazole impaired TCR-induced T cell stopping. Calcein-labeled HPBT cells (50,000/well) were plated and screened as described in 1Materials and Methods. B, Representative images at t = 0 and t = 7.5 min from primary screen after 30-min treatment with selected compounds (PP2, PGE2, and colchicine; ∼6–12 μM) and 10-min treatment with OKT3 (1 μg/ml). Merged image shows overlapped cells in the presence of OKT3 alone that is reversed by PP2, PGE2, and colchicine. Scale bar, 100 μm.

FIGURE 2.

A high-throughput assay for the HPBT cell migration stop signal identifies PGE2 and microtubule inhibitors as T cell stop signal antagonists. A, PGE1, PGE2, colchicine, nocodazole, and albendazole impaired TCR-induced T cell stopping. Calcein-labeled HPBT cells (50,000/well) were plated and screened as described in 1Materials and Methods. B, Representative images at t = 0 and t = 7.5 min from primary screen after 30-min treatment with selected compounds (PP2, PGE2, and colchicine; ∼6–12 μM) and 10-min treatment with OKT3 (1 μg/ml). Merged image shows overlapped cells in the presence of OKT3 alone that is reversed by PP2, PGE2, and colchicine. Scale bar, 100 μm.

Close modal
Table I.
Confirmed hits from screening 1600 compounds of known bioactivity
NameStructure% Migration
Controls   
 Untreated Not applicable 72 
 CD3 Not applicable 15 
 PP2  68 
Prostanoids   
 Alprostadil (PGE1 65 
 Dinoprostone (PGE2 52 
Microtubules   
 Colchicine  58 
 Albendazole  46 
 Nocodazole  62 
NameStructure% Migration
Controls   
 Untreated Not applicable 72 
 CD3 Not applicable 15 
 PP2  68 
Prostanoids   
 Alprostadil (PGE1 65 
 Dinoprostone (PGE2 52 
Microtubules   
 Colchicine  58 
 Albendazole  46 
 Nocodazole  62 

To confirm these results, we performed time-lapse microscopy to track the kinetics of T cell motility in T cells stimulated with OKT3 in the presence and absence of test compounds (Fig. 3). As previously reported, primary human T cells were highly polarized and motile on ICAM-1–coated plates (Fig. 3A). Treatment with OKT3 induced a loss of T cell polarity and impaired migration (from 10 μm/min in control T cells to 3 μm/min with OKT3; Fig. 3). The effects of OKT3 on cell polarity and T cell motility were prevented by either treatment with PGE2 or the microtubule-disrupting compound, colchicine (Fig. 3B, 3C, Supplemental Videos 2–4). The other positive hits also increased T cell polarity and random motility in the presence of OKT3 (Supplemental Fig. 1A); however, PP2 was most effective at reversing T cell stopping. Taken together, these results identify both PGs (PGE2) and microtubule-disrupting agents (colchicine) as novel T cell stop signal antagonists. The identification of PGs as stop signal antagonists is particularly surprising because PGs have previously been reported to inhibit T cell-directed migration (22).

FIGURE 3.

PGE2 treatment impairs the TCR stop signal. A, Differential interference contrast microscopy images of HPBT cells migrating on ICAM-1 in the presence or absence of PGE2 (5 μM) and OKT3 (1 μg/ml) demonstrated that OKT3 induced a loss of cell polarity that is abrogated in the presence of both PGE2 and OKT3. Scale bar, 20 μm. B, Displacement of individual cells during a 15-min period. C, Migration speeds (in μm/min) expressed as mean ± SEM (n = 3; *p < 0.05, ANOVA).

FIGURE 3.

PGE2 treatment impairs the TCR stop signal. A, Differential interference contrast microscopy images of HPBT cells migrating on ICAM-1 in the presence or absence of PGE2 (5 μM) and OKT3 (1 μg/ml) demonstrated that OKT3 induced a loss of cell polarity that is abrogated in the presence of both PGE2 and OKT3. Scale bar, 20 μm. B, Displacement of individual cells during a 15-min period. C, Migration speeds (in μm/min) expressed as mean ± SEM (n = 3; *p < 0.05, ANOVA).

Close modal

Src kinase inhibition with PP2 prevents proximal TCR signal transduction and activation of ZAP-70. To determine whether PGE2 functions upstream of proximal T cell signaling to affect T cell arrest, we characterized the effects of PGE2 on phosphorylation of downstream targets (Fig. 4). In accordance with previous reports, stimulation of human T cells with OKT3 for 3 min induced an increase in total tyrosine phosphorylation, which was blocked by PP2 but was not altered by treatment with PGE2 or colchicine. To determine whether PGE2 affects ZAP-70 phosphorylation, we examined phosphorylation on tyrosine 319 (23). OKT3 strongly stimulated ZAP-70 Y319 phosphorylation, which was prevented by treatment with PP2. In contrast, PGE2 and colchicine, at concentrations that abrogate T cell arrest, had no effect on OKT3-induced ZAP-70 phosphorylation. PP2, but not PGE2 or colchicine, prevented OKT3-induced ZAP-70 activation and phosphorylation of LAT at Y191. Moreover, PP2, but not PGE2 or colchicine, decreased levels of the activating Fyn phosphorylation at Y420. Taken together, these results demonstrate that PGE2 and colchicine, in contrast with PP2, function as T cell stop signal antagonists downstream or independently of proximal T cell signal transduction. These results were surprising because PGE2 has been reported to inhibit Src kinase activity under some conditions (10, 24).

FIGURE 4.

PP2, but not PGE2 or colchicine, impairs TCR-induced phosphorylation of ZAP-70 and LAT. A, Western blot analysis of whole-cell lysate. Cells were treated with 5 μM compounds. OKT3 induced total tyrosine phosphorylation (4G10) and phosphorylation of ZAP-70 at Y319 and phosphorylation of LAT at Y191, which was prevented by treatment with PP2, but not PGE2 or colchicine. PP2, but not PGE2 or colchicine, also impaired phosphorylation of Fyn. For analysis of Fyn, lysates were immunoprecipitated with Fyn Ab and blotted with p-Src-418 Ab. One representative blot from three experiments is shown. B, Quantification of Western blots (mean ± SEM; n = 3; *p < 0.05, ANOVA).

FIGURE 4.

PP2, but not PGE2 or colchicine, impairs TCR-induced phosphorylation of ZAP-70 and LAT. A, Western blot analysis of whole-cell lysate. Cells were treated with 5 μM compounds. OKT3 induced total tyrosine phosphorylation (4G10) and phosphorylation of ZAP-70 at Y319 and phosphorylation of LAT at Y191, which was prevented by treatment with PP2, but not PGE2 or colchicine. PP2, but not PGE2 or colchicine, also impaired phosphorylation of Fyn. For analysis of Fyn, lysates were immunoprecipitated with Fyn Ab and blotted with p-Src-418 Ab. One representative blot from three experiments is shown. B, Quantification of Western blots (mean ± SEM; n = 3; *p < 0.05, ANOVA).

Close modal

TCR engagement induces LFA-1 activation and T cell adhesion to ICAM-1 (25). To determine whether PGE2 modulates TCR-mediated LFA-1 activation, we tested the effects of PGE2 on T cell adhesion to ICAM-1 in the presence of OKT3. Treatment with OKT3 induced a 3- to 6-fold increase in adhesion to ICAM-1 relative to untreated control cells (Fig. 5A). PP2 blocked TCR-induced adhesion of T cells to ICAM-1. PGE2 and colchicine also impaired T cell adhesion to ICAM-1 in the presence of OKT3. These results suggest that PGE2 limits T cell arrest by impairing TCR-mediated inside-out LFA-1 activation and adhesion to ICAM-1.

FIGURE 5.

PGE2 treatment impairs TCR-induced adhesion to ICAM-1 and activation of Rap1. A, Adhesion to ICAM-1. Cells were treated with 10 or 50 μM PP2, colchicine, or PGE2 and assessed for adhesion to ICAM-1 in the presence or absence of OKT3. B, Representative Western blot analysis of GTP-bound Rap1 after treatment with 10 μM PP2 or PGE2 and stimulation with OKT3. C, Quantification of rap1 GTP binding (mean ± SEM; n = 3; *p < 0.05, ANOVA).

FIGURE 5.

PGE2 treatment impairs TCR-induced adhesion to ICAM-1 and activation of Rap1. A, Adhesion to ICAM-1. Cells were treated with 10 or 50 μM PP2, colchicine, or PGE2 and assessed for adhesion to ICAM-1 in the presence or absence of OKT3. B, Representative Western blot analysis of GTP-bound Rap1 after treatment with 10 μM PP2 or PGE2 and stimulation with OKT3. C, Quantification of rap1 GTP binding (mean ± SEM; n = 3; *p < 0.05, ANOVA).

Close modal

TCR engagement induces activation of the small GTPase Rap1, which is required for inside out activation of LFA-1 and subsequent adhesion to ICAM-1 (19). To determine whether PGE2 modulates Rap1 activation, we tested the effects of PGE2 on Rap1 GTP binding after stimulation with OKT3. OKT3 induced an ∼8-fold increase in GTP-bound Rap1 relative to untreated control cells (Fig. 5B). Both PP2 and PGE2 blocked Rap1 activation (Fig. 5C). These results suggest that PGE2 may limit T cell arrest by preventing Rap1 GTPase activation and impairing LFA-1–mediated adhesion to ICAM-1.

Duration of T cell–DC interactions modulate T cell fate and activation (9). To determine whether PGE2 modulates human T cell–DC interactions, we characterized the effects of PGE2 on the interactions between T cells and allogeneic DCs using flow cytometry. DCs were derived from peripheral blood monocytes treated with GM-CSF and IL-4 for 3 d, followed by LPS for 8 h. Interactions with naive human T cells labeled with CFSE were performed using flow cytometry as described in 1Materials and Methods (17) (Fig. 6, Supplemental Fig. 1B). In the presence of vehicle control, the efficiency of T cell–DC conjugation was 22.3% (Fig. 6A). PP2 (50 μM) reduced the rate of conjugation to 9.5%, which was reported as TCR-dependent conjugation. Treatment of T cells with PGE2 (50 μM) also inhibited conjugation, with an approximate 60% decrease in TCR-dependent conjugation (Fig. 6B, 6C). To determine whether PGE2 modulates downstream T cell signaling, we tested the effects of PGE2 on TCR-stimulated IL-2 production (Fig. 6D). OKT3 induced a ∼15-fold increase in IL-2 production relative to untreated control cells. PP2, PGE2, and colchicine blocked TCR-induced IL-2 production.

FIGURE 6.

PGE2 inhibits T cell conjugation and proliferation with allogeneic DCs. A, CFSE-labeled T cells were mixed with allogeneic DiD-labeled DCs and assessed for conjugation in the presence or absence of PP2 (50 μM) or PGE2 (50 μM). Conjugation events as a percentage of total events for one representative experiment are shown. B, Percentage of conjugation from three independent experiments in the presence or absence of colchicine (50 μM), PP2 (50 μM), or PGE2 (50 μM). C, Fraction of TCR-dependent conjugation inhibited by PGE2 (mean ± SEM; n = 3; *p < 0.05, ANOVA). D, IL-2 production in T cells that had been pretreated with PP2, PGE2, or colchicine (mean ± SEM; n = 3; *p < 0.05, ANOVA). E, CFSE-labeled T cells were activated by DCs and analyzed for proliferation by CFSE dilution at 6 d poststimulation in the presence or absence of PGE2 (50 μM) or PP2 (50 μM). Data are representative from three independent experiments. F, Dose-response curves for allogeneic DC stimulated T cell proliferation in the presence or absence of PGE2, PP2, or colchicine (mean ± SEM; n = 3; *p < 0.05, ANOVA).

FIGURE 6.

PGE2 inhibits T cell conjugation and proliferation with allogeneic DCs. A, CFSE-labeled T cells were mixed with allogeneic DiD-labeled DCs and assessed for conjugation in the presence or absence of PP2 (50 μM) or PGE2 (50 μM). Conjugation events as a percentage of total events for one representative experiment are shown. B, Percentage of conjugation from three independent experiments in the presence or absence of colchicine (50 μM), PP2 (50 μM), or PGE2 (50 μM). C, Fraction of TCR-dependent conjugation inhibited by PGE2 (mean ± SEM; n = 3; *p < 0.05, ANOVA). D, IL-2 production in T cells that had been pretreated with PP2, PGE2, or colchicine (mean ± SEM; n = 3; *p < 0.05, ANOVA). E, CFSE-labeled T cells were activated by DCs and analyzed for proliferation by CFSE dilution at 6 d poststimulation in the presence or absence of PGE2 (50 μM) or PP2 (50 μM). Data are representative from three independent experiments. F, Dose-response curves for allogeneic DC stimulated T cell proliferation in the presence or absence of PGE2, PP2, or colchicine (mean ± SEM; n = 3; *p < 0.05, ANOVA).

Close modal

To determine whether PGE2 modulates DC-induced T cell proliferation, we analyzed proliferation of T cells using CFSE dilution at day 6 after stimulation with allogeneic DCs (Fig. 6E, Supplemental Fig. 1C). In the presence of DMSO only, 31% of cells were proliferative. PP2 and PGE2 blocked T cell proliferation at concentrations of 50 μM (0.6 and 0.5% of control, respectively). The inhibition by PGE2 was dose dependent (Fig. 6F), and a statistically significant decrease in proliferation was seen at concentrations as low as 2.5 μM. Growth inhibition was also observed on stimulation with CD3/CD28-coated beads in the presence of PGE2 (Supplemental Fig. 1D). Taken together, these data demonstrate that PGE2 impairs T cell–DC conjugation and DC-induced T cell proliferation.

The identification of small molecules that alter T cell interactions with APCs represents an intriguing therapeutic strategy for autoimmune diseases such as rheumatoid arthritis and systemic lupus erythematosus (SLE). Indeed, a recent study has highlighted the critical importance of T cell and APC contact duration in determining T cell fate in vivo and the development of T cell tolerance or activation (9). There are currently no known small molecules that reverse the T cell stop signal in clinical use, and the addition of such drugs to treat autoimmune diseases is particularly attractive given the high cost of biologic agents and the resultant burden on the healthcare system. In this study, we have identified at least three distinct classes of “reverse-stop” small molecules that impair TCR-induced T cell arrest but not random T cell motility: 1) Src family tyrosine kinase inhibitors, 2) microtubule depolymerizing agents, and 3) PGs. These compounds act in contrast with inhibitors of phospholipase C (U73122), which block both basal and activated T cell motility or sphingosine-1-phosphate analog FTY720 and the PI3K inhibitor LY-294002, which altered basal motility but did not affect adhesion or spreading induced by OKT3 (Supplemental Table III).

The requirement of Src family kinases for the TCR-induced T cell stop signal, but not for T cell random motility, indicates that Src inhibitors represent T cell stop signal antagonists. This is consistent with the model that proximal T cell signaling is necessary for TCR-induced T cell arrest. It is intriguing that not all Src kinase inhibitors, most notably SKII, are capable of reversing the T cell stop signal. The results suggest that the stop signal is dependent on an Src family kinase, which is preferentially targeted by PP1, PP2, and SU6656, but not SKI1.

Previous work has indicated that Src kinase activation is required for TCR-mediated polarization of the microtubule-organizing center toward the T cell–APC contact (26) (27). It is interesting that our data indicate that Src inhibitors and microtubule disruption impair T cell stopping and interactions with APC. This is, to our knowledge, the first report to show that microtubules are necessary for the T cell arrest induced by TCR ligation. In accordance with our findings, previous studies have reported that microtubule disruption induces random motility of neutrophils (28) and modulates T cell random migration through rho/ROCK signaling (29). However, ROCK inhibition did not affect TCR-induced T cell stopping in our system, suggesting that effects of microtubule inhibition on T cell arrest may be independent of Rho/ROCK signaling.

The finding that both microtubule polymerization inhibitors and PGs are capable of preventing the T cell stop signal without affecting ZAP-70 or LAT phosphorylation (i.e., proximal TCR signaling) is particularly interesting (Fig. 3). In fact, we had initially hypothesized that the screening results would yield molecules that work to disrupt proximal signaling, such as the Src inhibitors. Our findings suggest that it is possible to decouple proximal TCR signaling from the TCR stop signal. PGE2 had no effect on the phosphorylation of Lck or Fyn at concentrations that block T cell arrest, suggesting that PGE2 effects on T cell arrest are independent of its effects on Src kinase activity. Our findings identified a novel role for PGE2 in the regulation of the small GTPase Rap1, which is critical for TCR-induced inside-out activation of LFA-1.

To our knowledge, this is the first report to implicate PGE2 in regulating the T cell stop signal. In contrast, previous studies have reported that PGE2 stimulates the ability of DCs to induce T cell proliferation (30). The finding that PGE1 and PGE2 impair T cell migration stopping, as well as inhibit T cell proliferation (31), indicate there may be counteracting mechanisms in place. Therefore, the presence of PGs may both promote and block DC-dependent T cell activation depending on the context of exposure. In addition, although PGE2 has been largely thought to be proinflammatory, recent studies have suggested that PGE2 and PG analogs may be anti-inflammatory in cases of autoimmune diseases such as SLE (13), because of its effects on DC-mediated cytokine production and shifting immune response from a Th1 to Th2 profile. Inhibition of the TCR stop signal would provide an additional anti-inflammatory mechanism for PGE2.

Interactions between DCs and T cells play a central role in the pathogenesis of autoimmune diseases such as SLE and represent an important therapeutic target. In addition to affecting the TCR stop signal, we found that PGE2 significantly impaired T cell–DC interactions and DC-induced T cell proliferation (Fig. 4). PGE2 and certain PG analogs are Food and Drug Administration-approved agents, and the novel effects on T cell stop signal and interactions with DCs suggest they may have therapeutic benefit in patients with SLE. In support of this possibility is a recent article that suggests that PGE2 also inhibits IFN-α secretion by plasmacytoid DCs, key players in SLE pathogenesis (13). In addition, another report recently demonstrated that COX inhibitors disrupt resolution of inflammation that was dependent on PGE2 in a mouse arthritis model (12). Moreover, our results may help to explain why exacerbation of SLE-like symptoms has been reported in patients treated with COX inhibitors (13), which function to decrease PG synthesis.

In summary, we have identified small molecules that modulate the T cell stop signal using a novel image-based, high-throughput screen. Because the approach is activation based rather than inhibition based, there are likely to be fewer off-target hits. We have shown that Src kinase inhibitors potently block the T cell stop signal and impair T cell–DC interactions. Our findings suggest that compounds that function either downstream or independently of ZAP-70 and LAT are also capable of reversing the T cell stop signal. The ability of PGs to block TCR-induced Rap1 activation and antagonize the T cell stop signal is especially intriguing and supports the use of this class of compounds as therapeutic agents that may have benefit in autoimmune disease. Likewise, these results may help explain the surprising proinflammatory effects sometimes seen with COX-2 inhibitors. Taken together, the findings suggest that small molecules that reverse the migration stop signal in vitro may either impair proximal TCR signaling, inhibit signaling at the level of Rap1, or directly induce random motility, thereby limiting TCR-induced stopping and DC-induced T cell activation. This study illustrates that high-throughput imaging of primary human cells can effectively be used to identify small molecules that alter migration stopping, allowing for further understanding of the molecular mechanisms that regulate Ag-induced T cell arrest, and offering a new paradigm for drug discovery.

We thank Noel Peters and Song Guo at the University of Wisconsin Keck Small Molecule Screening Facility and Kathy Schell at the University of Wisconsin Flow Cytometry Facility for technical assistance. We thank Sarah Wernimont, Miriam Shelef, David Bennin, and Taylor Starnes for phlebotomy assistance.

This work was supported by National Institutes of Health/National Institute of Allergy and Infectious Diseases Grant R01 AI068062 (to A.H.) and the Burroughs Wellcome Foundation (A.H.). A.J.W. received postdoctoral support from a University of Wisconsin Institute on Aging Training grant (National Institutes of Health grant T32AG000213-17), Sanjay Asthana, principal investigator.

The online version of this article contains supplemental material.

Abbreviations used in this article:

AM, acetoxymethyl ester; COX

cyclooxygenase

DC

dendritic cell

HPBT, human peripheral blood T; LAT

linker of activated T cells

ROI

region of interest

SLE

systemic lupus erythematosus.

1
Friedl
P.
,
Weigelin
B.
.
2008
.
Interstitial leukocyte migration and immune function.
Nat. Immunol.
9
:
960
969
.
2
Dustin
M. L.
,
Bromley
S. K.
,
Kan
Z.
,
Peterson
D. A.
,
Unanue
E. R.
.
1997
.
Antigen receptor engagement delivers a stop signal to migrating T lymphocytes.
Proc. Natl. Acad. Sci. USA
94
:
3909
3913
.
3
Gunzer
M.
,
Schäfer
A.
,
Borgmann
S.
,
Grabbe
S.
,
Zänker
K. S.
,
Bröcker
E. B.
,
Kämpgen
E.
,
Friedl
P.
.
2000
.
Antigen presentation in extracellular matrix: interactions of T cells with dendritic cells are dynamic, short lived, and sequential.
Immunity
13
:
323
332
.
4
Gunzer
M.
,
Weishaupt
C.
,
Hillmer
A.
,
Basoglu
Y.
,
Friedl
P.
,
Dittmar
K. E.
,
Kolanus
W.
,
Varga
G.
,
Grabbe
S.
.
2004
.
A spectrum of biophysical interaction modes between T cells and different antigen-presenting cells during priming in 3-D collagen and in vivo.
Blood
104
:
2801
2809
.
5
Scholer
A.
,
Hugues
S.
,
Boissonnas
A.
,
Fetler
L.
,
Amigorena
S.
.
2008
.
Intercellular adhesion molecule-1-dependent stable interactions between T cells and dendritic cells determine CD8+ T cell memory.
Immunity
28
:
258
270
.
6
Mempel
T. R.
,
Henrickson
S. E.
,
Von Andrian
U. H.
.
2004
.
T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases.
Nature
427
:
154
159
.
7
Guy
C. S.
,
Vignali
D. A.
.
2009
.
Organization of proximal signal initiation at the TCR:CD3 complex.
Immunol. Rev.
232
:
7
21
.
8
Schneider
H.
,
Downey
J.
,
Smith
A.
,
Zinselmeyer
B. H.
,
Rush
C.
,
Brewer
J. M.
,
Wei
B.
,
Hogg
N.
,
Garside
P.
,
Rudd
C. E.
.
2006
.
Reversal of the TCR stop signal by CTLA-4.
Science
313
:
1972
1975
.
9
Katzman
S. D.
,
O’Gorman
W. E.
,
Villarino
A. V.
,
Gallo
E.
,
Friedman
R. S.
,
Krummel
M. F.
,
Nolan
G. P.
,
Abbas
A. K.
.
2010
.
Duration of antigen receptor signaling determines T-cell tolerance or activation
.
Proc. Natl. Acad. Sci. USA
107
:
18085
18090
.
10
Chemnitz
J. M.
,
Driesen
J.
,
Classen
S.
,
Riley
J. L.
,
Debey
S.
,
Beyer
M.
,
Popov
A.
,
Zander
T.
,
Schultze
J. L.
.
2006
.
Prostaglandin E2 impairs CD4+ T cell activation by inhibition of lck: implications in Hodgkin’s lymphoma.
Cancer Res.
66
:
1114
1122
.
11
Wymann
M. P.
,
Schneiter
R.
.
2008
.
Lipid signalling in disease.
Nat. Rev. Mol. Cell Biol.
9
:
162
176
.
12
Chan
M. M.
,
Moore
A. R.
.
2010
.
Resolution of inflammation in murine autoimmune arthritis is disrupted by cyclooxygenase-2 inhibition and restored by prostaglandin E2-mediated lipoxin A4 production.
J. Immunol.
184
:
6418
6426
.
13
Fabricius
D.
,
Neubauer
M.
,
Mandel
B.
,
Schütz
C.
,
Viardot
A.
,
Vollmer
A.
,
Jahrsdörfer
B.
,
Debatin
K. M.
.
2010
.
Prostaglandin E2 inhibits IFN-alpha secretion and Th1 costimulation by human plasmacytoid dendritic cells via E-prostanoid 2 and E-prostanoid 4 receptor engagement.
J. Immunol.
184
:
677
684
.
14
Van Wauwe
J. P.
,
De Mey
J. R.
,
Goossens
J. G.
.
1980
.
OKT3: a monoclonal anti-human T lymphocyte antibody with potent mitogenic properties.
J. Immunol.
124
:
2708
2713
.
15
Stanley
P.
,
Hogg
N.
.
1998
.
The I domain of integrin LFA-1 interacts with ICAM-1 domain 1 at residue Glu-34 but not Gln-73.
J. Biol. Chem.
273
:
3358
3362
.
16
Simonson
W. T. N.
,
Franco
S. J.
,
Huttenlocher
A.
.
2006
.
Talin1 regulates TCR-mediated LFA-1 function.
J. Immunol.
177
:
7707
7714
.
17
Hegde
S.
,
Jankowska-Gan
E.
,
Roenneburg
D. A.
,
Torrealba
J.
,
Burlingham
W. J.
,
Gumperz
J. E.
.
2009
.
Human NKT cells promote monocyte differentiation into suppressive myeloid antigen-presenting cells.
J. Leukoc. Biol.
86
:
757
768
.
18
Hegde
S.
,
Chen
X.
,
Keaton
J. M.
,
Reddington
F.
,
Besra
G. S.
,
Gumperz
J. E.
.
2007
.
NKT cells direct monocytes into a DC differentiation pathway.
J. Leukoc. Biol.
81
:
1224
1235
.
19
Katagiri
K.
,
Hattori
M.
,
Minato
N.
,
Kinashi
T.
.
2002
.
Rap1 functions as a key regulator of T-cell and antigen-presenting cell interactions and modulates T-cell responses.
Mol. Cell. Biol.
22
:
1001
1015
.
20
Salmond
R. J.
,
Filby
A.
,
Qureshi
I.
,
Caserta
S.
,
Zamoyska
R.
.
2009
.
T-cell receptor proximal signaling via the Src-family kinases, Lck and Fyn, influences T-cell activation, differentiation, and tolerance.
Immunol. Rev.
228
:
9
22
.
21
Blake
R. A.
,
Broome
M. A.
,
Liu
X.
,
Wu
J.
,
Gishizky
M.
,
Sun
L.
,
Courtneidge
S. A.
.
2000
.
SU6656, a selective src family kinase inhibitor, used to probe growth factor signaling.
Mol. Cell. Biol.
20
:
9018
9027
.
22
Mesri
M.
,
Liversidge
J.
,
Forrester
J. V.
.
1996
.
Prostaglandin E2 and monoclonal antibody to lymphocyte function-associated antigen-1 differentially inhibit migration of T lymphocytes across microvascular retinal endothelial cells in rat.
Immunology
88
:
471
477
.
23
Pelosi
M.
,
Di Bartolo
V.
,
Mounier
V.
,
Mège
D.
,
Pascussi
J. M.
,
Dufour
E.
,
Blondel
A.
,
Acuto
O.
.
1999
.
Tyrosine 319 in the interdomain B of ZAP-70 is a binding site for the Src homology 2 domain of Lck.
J. Biol. Chem.
274
:
14229
14237
.
24
Choudhry
M. A.
,
Uddin
S.
,
Sayeed
M. M.
.
1998
.
Prostaglandin E2 modulation of p59fyn tyrosine kinase in T lymphocytes during sepsis.
J. Immunol.
160
:
929
935
.
25
Abraham
C.
,
Griffith
J.
,
Miller
J.
.
1999
.
The dependence for leukocyte function-associated antigen-1/ICAM-1 interactions in T cell activation cannot be overcome by expression of high density TCR ligand.
J. Immunol.
162
:
4399
4405
.
26
Martín-Cófreces
N. B.
,
Sancho
D.
,
Fernández
E.
,
Vicente-Manzanares
M.
,
Gordón-Alonso
M.
,
Montoya
M. C.
,
Michel
F.
,
Acuto
O.
,
Alarcón
B.
,
Sánchez-Madrid
F.
.
2006
.
Role of Fyn in the rearrangement of tubulin cytoskeleton induced through TCR.
J. Immunol.
176
:
4201
4207
.
27
Gomez
T. S.
,
Billadeau
D. D.
.
2008
.
T cell activation and the cytoskeleton: you can’t have one without the other.
Adv. Immunol.
97
:
1
64
.
28
Xu
J.
,
Wang
F.
,
Van Keymeulen
A.
,
Rentel
M.
,
Bourne
H. R.
.
2005
.
Neutrophil microtubules suppress polarity and enhance directional migration.
Proc. Natl. Acad. Sci. USA
102
:
6884
6889
.
29
Takesono
A.
,
Heasman
S. J.
,
Wojciak-Stothard
B.
,
Garg
R.
,
Ridley
A. J.
.
2010
.
Microtubules regulate migratory polarity through Rho/ROCK signaling in T cells.
PLoS ONE
5
:
e8774
.
30
Scandella
E.
,
Men
Y.
,
Gillessen
S.
,
Förster
R.
,
Groettrup
M.
.
2002
.
Prostaglandin E2 is a key factor for CCR7 surface expression and migration of monocyte-derived dendritic cells.
Blood
100
:
1354
1361
.
31
Goodwin
J. S.
,
Bankhurst
A. D.
,
Messner
R. P.
.
1977
.
Suppression of human T-cell mitogenesis by prostaglandin. Existence of a prostaglandin-producing suppressor cell.
J. Exp. Med.
146
:
1719
1734
.

The authors have no financial conflicts of interest.

Supplementary data