- Email: [email protected]
Anti-CD2 Monoclonal Antibody-Induced Receptor Changes
CELLULAR IMMUNOLOGY
167, 249–258 (1996)
Article No. 0033
Anti-CD2 Monoclonal Antibody-Induced Receptor Changes II. Interaction of CD2 and CD3 JIXUN LIN,*,† KENNETH D. CHAVIN,†,‡ LIHUI QIN,* YAOZHONG DING,*,§
AND
JONATHAN S. BROMBERG*,§,1,2
Departments of *Surgery and †Microbiology and Immunology, University of Michigan, Ann Arbor, Michigan 48109; and Departments of §Surgery and ‡Microbiology and Immunology, Medical University of South Carolina, Charleston, South Carolina 29425 Received June 21, 1995; accepted September 14, 1995
Anti-CD3 monoclonal antibodies (mAbs) and antiCD2 mAbs each prolong allograft survival and cause transient downmodulation of homologous receptor expression. Anti-CD2 mAbs also act synergistically with anti-CD3 mAbs to prolong allograft survival and induce tolerance. The effect of combined anti-CD2 and anti-CD3 mAb treatment on receptor expression was further analyzed with an in vitro model. The antiCD2 mAb 12-15 caused CD2 expression on purified splenic T cells to decrease from 72.6% [mean channel fluorescence (MCF) 0.68] to 41.5% (0.45) total positive cells while CD3 expression remained unchanged [69.1% (3.47) to 76.4% (4.04)]. The anti-CD3 mAb 2C11 caused CD2 expression to increase from 72.6% (0.68) to 93.0% (1.74) while CD3 expression decreased from 69.1% (3.47) to 62.6% (2.15). The combination of antiCD2 plus anti-CD3 preserved CD2 expression (72.6 to 71.1%) while still decreasing CD3 expression [69.1% (3.47) to 69.9% (2.37)]. Modulation of CD2 and CD3 expression was similar on mixed splenic T lymphocytes and isolated CD4 and CD8 subsets. Modulation did not change with the addition of the cytokines IL-1, IL-2, IL-4, IL-6, IL-10, TNFa, or TGFb. Kinetic studies showed that modulation of CD2 was rapid, persistent, and of the same magnitude from Day 1 to Day 7 of culture while CD3 downmodulation was transient. The results of transcriptional analysis and receptor distribution suggested that downmodulation was due to receptor internalization while upmodulation was due to increased transcription. Analysis of expression of other adhesion molecules demonstrated that CD11a, CD18, CD44, CD45, CD48, CD54, and CD62L were significantly increased by either anti-CD2 or anti-CD3 mAbs while the combination was not synergistic. However, anti-CD3 significantly decreased VLA-4a (CD49d) expression and anti-CD2 enhanced 1 Supported by American Surgical Association Foundation Fellowship Award and NIH Grant AI32655. 2 To whom correspondence should be addressed at Department of Surgery, University of Michigan, 2926 Taubman Center, Box 0331, Ann Arbor, MI 48109-0331.
this decrease. Conversely anti-CD3 significantly increased IL-2R (CD25) expression and anti-CD2 profoundly inhibited the increase. These results show that anti-CD2 and anti-CD3 mAbs significantly modulate CD2 and CD3 expression on T cells and modulation is accompanied by changes in the array of other T cell surface receptors. Changes in cell surface receptor display may provide an additional explanation for the synergistic effect of anti-CD2 plus anti-CD3 in prolonging allograft survival. q 1996 Academic Press, Inc.
INTRODUCTION The cell surface molecule CD2 is widely distributed in the immune system. It is a 55kDa transmembrane glycoprotein found on T cells, B cells, NK cells, and some antigen-presenting cells (APCs)3 in mice (1, 2). CD2 functions as an adhesion receptor which strengthens the interaction between helper T cells and APCs or between effector T cells and target cells. CD2 also plays a role in T cell activation (3–5) and modulates CD3-related events (6). CD2 shares some common second messenger pathways of T cell activation with CD3 (7–9) and has also been demonstrated to regulate the function of a variety of other cell surface molecules (3– 7, 9, 10). CD2 therefore appears to play a major regulatory role in immunity. Anti-mouse CD2 mAbs can deliver a negative signal to T cells and inhibit CD3-related events in vitro under certain circumstances (11). Anti-CD2 mAbs administered in vivo also suppress multiple T cell-dependent immune responses (12–14). Associated with suppression of immunity is downmodulation of cell surface CD2 without cellular depletion (12, 13). Our previous studies also showed that anti-CD2 mAbs could prolong allograft survival and this immunosuppression was accompanied by the downmodulation of T cell CD2 expression 3 Abbreviations used: APC, antigen-presenting cell; mAb, monoclonal antibody; MCF, mean channel fluorescence.
249
/ m4726$8025
01-03-96 15:51:37
cial
0008-8749/96 $12.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP: Cell Immuno
250
LIN ET AL.
in vivo (12, 15). However, anti-CD2 mAb alone did not produce tolerance. Anti-CD3 mAb has also been used in transplantation to prolong allograft survival. Its use encompasses both prophylactic administration to prevent rejection (16) and therapeutic treatment of acute rejection (17, 18). Administration of anti-CD3 results in modulation of CD3 expression and that of a number of other surface receptors (19). While anti-CD3 mAb is a potent immunosuppressant, it fails to produce true, long-term tolerance (20) and has associated side effects, including a deleterious cytokine syndrome which is attributed to T cell activation and cytokine release (21–23). Therefore, the goal of tolerance without significant toxicity has not been met with this reagent. Since CD2 is functionally and physically associated with CD3 as determined by coprecipitation studies and demonstration of common signaling pathways (6–9), it was hypothesized that anti-CD2 could act synergistically with anti-CD3 in inducing immunosuppression. We therefore demonstrated that combined treatment with anti-CD2 plus anti-CD3 mAbs prolonged allograft survival, induced alloantigen-specific tolerance, and diminished the anti-CD3-induced cytokine syndrome (24, 25). We also noted that changes in cell surface receptor expression correlated exactly with both immunosuppression and tolerance in vivo. Previous studies also demonstrated that the extent of CD2 downmodulation produced by anti-CD2 mAb is directly related to its inhibitory effect on cell-mediated immunity in vivo (12) and on anti-CD3-driven T cell activation in vitro (26). There is a similar relationship between the immunosuppressive capacity of anti-CD3 mAbs and changes in receptor expression (17–20). These results suggest that antibody–ligand interactions and subsequent changes in receptor expression are important for the process of immunosuppression and tolerance induction. Therefore, we sought to define the determinants of receptor modulation by studying the effects of anti-CD2 and anti-CD3 mAbs on CD2 and CD3 expression on murine T cells with an in vitro model. Results show that anti-CD2 and anti-CD3 mAbs modulate CD2 and CD3 expression similarly on CD4/ and CD8/ T cell subpopulations. While anti-CD2 and anti-CD3 upmodulate most adhesion receptors, they downmodulate CD49d synergistically and act antagonistically on the expression of CD25. Various cytokines have no effect on CD2 and CD3 expression, showing that these receptor–ligand interactions do not affect CD2–anti-CD2 or CD3–anti-CD3 interactions and subsequent second signals. These studies suggest that one mechanism for the synergistic effect of antiCD2 plus anti-CD3 on allograft survival is a significant change in the array of cell surface receptors. MATERIALS AND METHODS Animals. BALB/cByJ (H-2d) female mice 8–10 weeks of age were purchased from Jackson Laboratory (Bar Harbor, ME).
/ m4726$8025
01-03-96 15:51:37
cial
Antibodies and other reagents. The rat IgG1 antimurine CD2 12-15 hybridoma (27) was a gift of Dr. P. Altevogt (Heidelberg, Germany); the 145-2C11 (antiCD3e) hybridoma (28) was a gift of Dr. J. A. Bluestone (University of Chicago), and the HM48-1 (anti-CD48) hybridoma (29) was a gift of Dr. H. Yagita (Tokyo, Japan). The GK1.5 (anti-CD4) (30), 53-6.72 (anti-CD8) (31), Y13-259 (anti-Ras) (32), 3C7 (anti-CD25) (33), KM201 (anti-CD44) (34), M1/89 (anti-CD45) (35), R12 (anti-CD49d) (36, 37), YN1/1.7.4 (anti-CD54) (38), M17/4.2 (anti-CD11a), M18/2.a.8 (anti-CD18) (39), H57597 (anti-CD3 a/b) (40), and MEL-14 (anti-CD62, L-selectin) (41) hybridomas were purchased from the American Type Culture Collection (Rockville, MD). All hybridomas were grown in culture and purified over protein G (Pharmacia, Piscataway, NJ). Phycoerythrin L3T4 (Becton–Dickinson, San Jose, CA), phycoerythrin–streptavidin (Becton–Dickinson), biotin–xxNHS (Calbiochem, San Diego, CA), affinity-purified F(ab*)2 FITC goat anti-rat IgG (TAGO, Inc., Burlingame, CA), and affinity-purified FITC goat antihamster IgG (Kirkegaard & Perry Lab,Inc., Gaithersburg, MD) were used as alternative or second-step reagents. Recombinant murine IL-1, rmIL-2, rmIL-4, and rmIL-6 (gifts from Dr. M. Ogawa, MUSC), murine IL10 COS cell supernatant with an activity of 3000 units/ ml (a gift of Dr. William Fanslow, Immunex, Seattle, WA), rmTNFa (a gift of Dr. Grace Wong, Genentech, San Francisco, CA), rmTGFb (Gibco BRL, Gaithersburg, MD), Con A (Sigma, St. Louis, MO), PHA (Sigma), or Con A supernatant was used as a source of either stimulation or growth factors in conditioned medium. Antibody conjugation. Purified mAbs were dialyzed against FITC labeling buffer (0.05 M boric acid, 0.2 M NaCl, pH 9.2) at 47C. Antibody concentrations were determined by A280. Twenty microliters of 5 mg/ml FITC in DMSO for each milligram of antibody in 1 ml labeling buffer was added and incubated for 2 hr at room temperature in the dark. Unbound FITC was removed by filtration over Sephadex G-25 and titer determined by flow cytometry. For biotin labeling, purified mAbs were dialyzed against biotin labeling buffer (0.1 M NaHCO3, 0.1 MNaCl, pH 7.4). Ten microliters of 10 mg/ml biotin in DMSO for each milligram of antibody in 1 ml labeling buffer was added and incubated for 1 hr at room temperature. Unbound biotin was removed by filtration over Sephadex G-25. Cell preparation. Mice were sacrificed, and spleens were removed and gently dissociated into a single cell suspension. Red blood cells were removed by TrisNH4Cl lysis. Cell suspensions were passed through nylon wool columns to enrich for T cells (42). These cells were routinely ú70% T cells. Cells were plated at 1 1 106 cells in 2-ml wells in complete RPMI medium (RPMI 1640 supplemented with 10% FCS, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 IU/ml penicillin, 100 mg/ml streptomycin, 1 1 nonessential amino
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
251
acids, and 2 1 1005 M 2-mercaptoethanol). Cultures were routinely supplemented with rmIL-2 (0.15 U/ml) as a source of growth factor. MAbs or other cytokines were added to the wells as indicated. Control groups received either control mAbs or no treatment. Cells were harvested 1–7 days later and stained, and flow cytometric analysis was performed. Separation of CD4 and CD8 subsets. T cell-enriched splenic lymphocytes isolated over nylon wool columns as described above were further negatively selected by antibody column depletion. Anti-CD4 and anti-CD8 columns were purchased (R & D Systems, Minneapolis, MN) and used according to manufacturer’s instructions. These columns removed not only the relevant T cell subsets but also B cells and macrophages. Flow cytometry. Cell washes and antibody dilutions were performed in phosphate-buffered saline plus 1% bovine serum albumin at 47C. CD2 was stained with the 12-15 mAb (1:30) followed by affinity-purified F(ab*)2 FITC goat anti-rat IgG (H / L) (1:100). CD4 was stained with PE-L3T4 (1:100). CD3, CD8, CD11a, CD18, CD25, CD44, CD45, CD49d, and MEL-14 were all stained directly by the FITC-labeled antibodies described above. Negative controls included second antibodies with or without irrelevant first mAbs in all experiments. Saturating concentrations of antibodies were used as determined by dose–response curves, which were repeated several times during the course of these studies. Flow cytometric analysis was performed on an Epics Elite flow cytometer (Coulter, Hialeah, FL) or a FACScan flow cytometer (Becton–Dickinson). Results are expressed as percentages of cells staining above background or changes in mean channel fluorescence on a logarithmic scale of relative cellular fluorescence. Northern hybridization. Cultured cells were harvested and whole-cell RNA was isolated using RNAzol B (Cinna/Biotecx, Houston, TX). Equal quantities of RNA were electrophoresed on 1% formaldehyde agarose gels and transferred to nylon membranes. The membranes were sequentially probed for CD2, CD3, and b-actin using 32P random primer (Stratagene, La Jolla, CA) labeled probes. The CD2 probe was the the 0.7-kb EcoRI– HpaI fragment from the murine CD2 cDNA clone pMCD2-2 (2). The CD3 probe was the 1.4-kb EcoRI fragment of CD3e (43). The b-actin probe was the 540-base pair PCR product between bases 25 and 565 generated from C57BL/6 genomic DNA. Radiolabeled hybridization products were analyzed by autoradiography and densitometry by NIH Image 1.49 software. RESULTS Modulation of CD2 and CD3 Expression by Anti-CD2 and Anti-CD3 mAbs Anti-CD2 mAb 12-15, anti-CD3 mAb 2C11, and/or the irrelevant IgG1 isotype control mAb 259 were
/ m4726$8025
01-03-96 15:51:37
cial
FIG. 1. Modulation of T cell surface CD2 and CD3 by anti-CD2 and anti-CD3 mAbs. Nylon wool-purified splenic T lymphocytes incubated with 5 mg/ml of control Y13-259 mAb or anti-CD2 mAb 12-15 or 0.5 mg/ml anti-CD3 mAb 2C11 for 3 days were analyzed for cell surface expression of CD2 and CD3. Numbers in upper right are mean channel fluorescence for gated population.
added to cultures of murine splenic T lymphocytes. As shown in Fig. 1, flow cytometric analysis demonstrated that anti-CD2 mAb caused lymphocytic expression of CD2 to decrease from 72.6% (MCF 0.68) to 41.5% (0.45) while expression of CD3 remained unchanged [69.1% (3.47) to 76.4% (4.04)] after 3 days of culture with antiCD2 mAb compared to the cells in the control group. Conversely anti-CD3 mAb caused CD2 expression to increase from 72.6% (0.68) to 93.0% (1.74) and CD3 expression to decrease from 69.1% (3.47) to 62.6% (2.15). The combination of mAbs caused CD2 expression to remain unchanged (72.6% (0.68) to 71.1% (0.72)) while CD3 expression still decreased (69.1% (3.47) to 69.9% (2.37)). Staining for CD3 a/b showed identical results as for CD3e, demonstrating that the entire CD3 complex was modulated. Because indirect immunofluorescent techniques were used, cell surface CD2 and CD3 receptors coated with mAb from culture, and CD2 and CD3 which remained unbound by mAb in culture, were both quantitated by the flow cytometric methods
AP: Cell Immuno
252
LIN ET AL.
ulating culture conditions nor influenced by interaction of multiple cytokines with their receptors. Modulation of CD2 and CD3 Expression on T Cell Subpopulations
FIG. 2. Modulation of CD2 and CD3 by anti-CD2 and anti-CD3 mAb is dose dependent. Nylon wool-purified splenic T lymphocytes incubated with the indicated doses of anti-CD2 mAb 12-15 or antiCD3 mAb 2C11 for 3 days were assessed for cell surface expression of CD2 and CD3.
employed here. Further, controls showed that these mAbs did not sterically hinder each other (not shown). These results demonstrate that the anti-CD2 mAb 1215 and the anti-CD3 mAb 2C11 could downmodulate CD2 and CD3 expression, respectively. Combined antiCD2 plus anti-CD3 could still downmodulate CD3, but CD2 expression was increased compared to anti-CD2 alone and decreased compared to anti-CD3 alone. Dose–response data showed that 12-15 at 5 mg/ml induced maximal downmodulation of CD2 expression and 2C11 at 0.5–5 mg/ml induced maximal downmodulation of CD3 and upmodulation of CD2 expression simultaneously (Fig. 2). In kinetic experiments, cells were harvested after 1 to 7 days of culture and examined. Flow cytometric analysis showed that modulation of CD2 and CD3 expression was rapid with downmodulation seen within 24 hr of culture. The effect was persistent for CD2 since the magnitude of modulation remained constant throughout 7 days of culture while CD3 was slowly reexpressed over time (not shown). The cultures shown here were supplemented with 0.15 U/ ml purified recombinant murine IL-2 as a T cell growth factor. Supplementation with different doses of rmIL2, Con A, PHA, or Con A supernatant as a source of conditioned medium resulted in identical results (not shown). Various cytokines were also used to supplement cultures and evaluated for their effect on CD2 and CD3 modulation. The results in Fig. 3 show that modulation of CD2 and CD3 does not change with the addition of IL-4 or IL-6. Identical results were obtained with IL-1, IL-10, TNFa, and TGFb (not shown). Therefore, modulation induced by anti-CD2 and anti-CD3 mAbs was neither dependent on a restricted set of stim-
/ m4726$8025
01-03-96 15:51:37
cial
Since murine CD2 is expressed not only on T lymphocytes but also on B cells, NK cells, and some APCs, and since CD3 is expressed exclusively on T cells, it was important to analyze CD2 and CD3 expression on more highly purified CD4/ and CD8/ subpopulations. AntiCD2 mAb caused downmodulation of CD2 expression and had no effect on CD3 expression on all populations tested. CD2 expression decreased from 81.9 to 31.9% on nylon wool-enriched T cells, from 93.9 to 35.6% on CD8-depleted (CD4/) T cells, and from 88.8 to 57.9% on CD4-depleted (CD8/) T cells. Anti-CD3 mAb caused upmodulation of CD2 expression on nylon wool-enriched T cells from 81.9% (0.76) to 94.3% (2.14), CD8depleted T cells from 88.8% (1.33) to 95.2% (2.30), and CD4-depleted T cells from 88.8% (1.85) to 93.1% (2.61). Anti-CD3 mAb also induced downmodulation of CD3 expression on nylon wool-enriched T cells from 87.2% (1.35) to 84.2% (0.99), CD8-depleted T cells from 95.2% (11.2) to 90.1% (6.34), and CD4-depleted T cells from 93.9% (11.7) to 92.6% (7.91). Combined anti-CD2 and -CD3 mAbs preserved CD2 expression but still downmodulated CD3 expression on all the populations (Fig. 4). The results demonstrate that anti-CD2 caused CD2 downmodulation, anti-CD3 caused CD2 upmodulation and CD3 downmodulation, and the combination of mAbs preserved CD2 expression while downmodulating CD3 on both CD4/ cells and CD8/ cells. It should
FIG. 3. Cytokines do not affect mAb-induced down modulation. IL-4 or IL-6 at 50 U/ml was added at the initiation of culture with 5 mg/ml 12-15 and 0.5 mg/ml 2C11. Cells were harvested and labeled after 3 days of culture.
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
253
FIG. 4. Anti-CD2 mAb 12-15 and anti-CD3 mAb 2C11 cause modulation of CD2 and CD3 expression on various T cell subtypes. The indicated cell types were cultured with 5 mg/ml 12-15 and/or 0.5 mg/ml 2C11 for 3 days and then labeled for flow cytometry.
be noted that the experiment in Fig. 4 is from a 3-day culture at which time CD3 is starting to be reexpressed. CD2 and CD3 Transcription and Distribution Previous work demonstrated that the major mechanism of CD2 (44) and CD3 (20–22) downmodulation was internalization of antigen–antibody complexes, rather than capping and shedding of the complex. Our previous work (26) also demonstrated that CD2 mRNA levels of anti-CD2-treated cells were unchanged compared to controls, suggesting that anti-CD2 mAb had little effect on the level of CD2 mRNA transcription, which supports the finding that antigen–antibody internalization is the major mechanism for downmodulation. It remained to be determined whether the alterations in expression of CD2 and CD3 by anti-CD2 plus anti-CD3 were due to changes at the transcription or internalization levels. RNA was isolated from control or treated cells and hybridized with probes for CD2, CD3, or b-actin. The results (Fig. 5) show that CD2 mRNA levels were increased at Day 2 and Day 4 as a result of the activational effects of anti-CD3 mAb, either alone or in combination with anti-CD2. CD3 mRNA levels were increased by anti-CD3 on Day 4 but not on Day 2. Neither anti-CD2 nor anti-CD3 mAb decreased the levels of CD2 or CD3 mRNA, and antiCD2 mAb alone had no significant effect on mRNA expression. To examine the effects of mAb treatment on internalization of receptors, splenic T lymphocytes were incubated with biotin-labeled anti-CD2 mAb, washed, then cultured with or without anti-CD3, and stained with phycoerythrin–streptavidin either before or after culture. In addition cells from some groups were reincubated with biotin–anti-CD2 at the end of culture and
/ m4726$8025
01-03-96 15:51:37
cial
then stained with phycoerythrin. In this fashion then the fate of CD2 originally present on the cells and the expression of new CD2 could be determined. The results (Fig. 6) demonstrate that cells stained with phycoerythrin at the initiation of culture and analyzed by
FIG. 5. Northern hybridization. Purified T cells were cultured with 12-15, Y13-259, and/or 2C11 for 2 or 4 days, and total RNA was isolated and blotted with probes for CD2, CD3, or b-actin. For CD2 a total of 3.8 mg RNA was loaded per lane; for CD3 and b-actin 3.0 mg RNA was loaded per lane. Lane 1, Y13-259; lane 2, 12-15; lane 3, 2C11; lane 4, 12-15 / 2C11. Numbers underneath blots represent ratios of arbitrary units of optical density normalized to lane 1 and to b-actin.
AP: Cell Immuno
254
LIN ET AL.
analysis showed a marked decrease which was true if the cells were incubated with or without anti-CD3 (curves 5 and 6). If cells from these last two groups were stained at the end of culture with additional biotin–anti-CD2 followed by phycoerythrin, then flow cytometry demonstrated that the anti-CD3-treated group had increased CD2 (curve 8) up to the level of the control (curve 2), while the group not treated with antiCD3 had only a minimal increase in CD2 (curve 7). The results in Figs. 5 and 6 taken together show that antiCD2 causes downmodulation of CD2 by internalization and anti-CD3 causes upmodulation of CD2 by transcriptional increase. This is further supported by the results in Figs. 6D and 6E which demonstrate that actinomycin D can block the majority of anti-CD3driven upmodulation of CD2. Expression of Other Cell Surface Receptors
FIG. 6. (A–C) Anti-CD2 causes internalization of CD2 and antiCD3 causes expression of new CD2 molecules. Splenic T lymphocytes were labeled with biotin–anti-CD2 prior to culture. Curve 1, negative control; 2, stained with phycoerythrin–streptavidin (PE) and analyzed prior to culture; 3, stained with PE prior to and analyzed after 24 hr of culture; 4, as for 3 but cultured with anti-CD3; 5, stained with PE and analyzed after culture; 6, as for 5 but cultured with anti-CD3; 7, stained with additional biotin–anti-CD2 and PE and analyzed, all after culture; 8, as for 7, but cultured with anti-CD3. (D and E) Upmodulation of CD2 by anti-CD3 is inhibited by transcriptional blockade. T lymphocytes were incubated with or without actinomycin D (50 mg/ml) or anti-CD3 mAb, harvested after 24 hr, and stained for CD2. Curve 1, negative control; 2, cells alone; 3, cells plus anti-CD3; 4, cells plus actinomycin D; 5, anti-CD3 plus actinomycin D.
flow cytometry either at that time or 24 hr later showed nearly identical staining (curves 2 and 3). This was also true if the cells were incubated with or without anti-CD3 (curves 3 and 4). If phycoerythrin staining was delayed until the end of culture, then fluorescent
/ m4726$8025
01-03-96 15:51:37
cial
Since anti-CD2 and anti-CD3 induce both CD2 and CD3 modulation and immunosuppression, and since CD2 and CD3 have physical and functional associations with other surface receptors, it was important to see if they affected the expression of other important cell surface receptors. Therefore, cells incubated in culture with the mAbs were subsequently directly labeled with a series of FITC-conjugated mAbs directed toward other cell surface adhesion receptors. Control experiments demonstrated that 12-15 and 2C11 did not sterically hinder the labeling by any of these other mAbs (not shown); therefore, anti-CD2 and anti-CD3 bound to the cell surface in culture would not interfere with subsequent labeling steps. The results in Table 1 show that CD4 remained unaffected by concomitant CD2 and CD3 modulation, and CD8 expression also remained stable with anti-CD2 treatment but increased slightly with anti-CD3 treatment. This demonstrates that antiCD2 and anti-CD3 did not cause a major preferential loss or growth of T cells or their subsets. Most other receptors, representing members of the selectin, integrin, and immunoglobulin gene superfamily, were upmodulated slightly by anti-CD2 and more significantly by anti-CD3, while the combination of mAbs was not different than anti-CD3 alone (Table 1). However, the effects on CD25 and CD49d expression were unique. CD25 expression increased significantly from 28.3% (2.38) to 95.9% (6.02) after treatment with anti-CD3, due to the activational properties of anti-CD3. While anti-CD2 alone had no effect on CD25 expression [28.3 to 24.3% (2.45)], anti-CD2 significantly inhibited antiCD3-driven CD25 expression from 95.9 to 63.8% (3.58) (Fig. 7). CD49d expression remained unchanged from 85.6% (0.95) to 90.2% (1.16) after anti-CD2 treatment, but decreased significantly from 85.6 to 57.4% (0.93) after anti-CD3 treatment. CD49d expression decreased further from 57.4 to 47.0% (0.77) with combined antiCD2 plus anti-CD3 treatment (Fig. 7). This suggests that immunologically important effects of anti-CD2
AP: Cell Immuno
255
INTERACTION OF CD2 AND CD3
TABLE 1 Anti-CD2 and Anti-CD3 mAbs Modulate Expression of Adhesion Receptors Cell surface receptor % positive cells (mean channel fluorescence) Treatment group Control Anti-CD2 Anti-CD3 Anti-CD2/anti-CD3
CD2 81.9 30.1 94.3 73.3
CD3e
(0.76) (0.52) (2.14) (0.81)
87.2 91.3 84.2 88.3
CD3a/b
(1.35) (1.67) (0.99) (1.10)
74.4 78.8 46.5 29.1
CD45 Control Anti-CD2 Anti-CD3 Anti-CD2/anti-CD3
86.8 92.2 95.6 98.8
(1.64) (1.97) (5.25) (3.98)
(0.75) (0.99) (0.69) (0.84)
CD48 31.9 28.3 33.3 35.4
(0.40) (0.43) (0.44) (0.50)
CD11a 36.9 42.8 92.8 91.3
(0.76) (1.36) (1.76) (1.43) CD49d
85.6 90.2 57.4 47.0
(0.95) (1.16) (0.93) (0.77)
CD18C 34.9 41.2 88.3 85.7
D25
(0.67) (1.27) (1.21) (0.96)
14.5 13.1 23.7 19.4 CD54
43.9 38.0 65.6 70.2
(0.92) (0.71) (0.67) (0.70)
(0.63) (0.71) (0.78) (0.88)
CD44 57.8 56.5 87.7 85.9
(1.56) (2.01) (3.71) (1.99)
CD62L 69.2 71.2 95.7 91.3
(2.90) (3.27) (14.9) (7.05)
Note. Purified T lymphocytes were incubated with control mAb Y13–259 or 12–15 at a 5 mg/ml or 2C11 at 0.5 mg/ml for 3 days in culture and evaluated by indirect or direct immunofluorescence for expression of the indicated receptors.
and anti-CD3 may be related not only to CD2 and CD3 modulation but also to changes in the entire cell surface receptor display. The effects of anti-CD2 mAb on the modulation of cell surface receptor expression induced by anti-CD3 mAbs can therefore be categorized into at least three patterns. First, anti-CD2 mAb had no effect on anti-CD3 mAb-induced upmodulation of an array of adhesion molecules; in fact, anti-CD2 alone partially upmodulated these receptors. Second, anti-CD2 mAb acted synergistically with anti-CD3 mAb to downmodulate CD49d expression. Third, anti-CD2 mAb inhibited
FIG. 7. Anti-CD2 antagonizes anti-CD3-driven upmodulation of CD25, and anti-CD2 plus anti-CD3 synergistically downmodulates CD49d expression. 12-15 at 5 mg/ml and/or 2C11 at 0.5 mg/ml were added to culture with T lymphocytes for 3 days. Il-2 was added at 125 U/ml to culture for CD25 measurements.
/ m4726$8025
01-03-96 15:51:37
cial
the upmodulation of CD25 expression induced by antiCD3 mAb. DISCUSSION Our previous work demonstrated that anti-CD2 mAb was effective in inducing immunosupression and in prolonging allograft survival (12, 15). The immunosuppression induced by anti-CD2 mAbs was accompanied by the concomitant downmodulation of CD2 expression in vivo (11). Our recent in vitro studies showed that anti-CD2 mAbs significantly and persistently downmodulated CD2 on various T cell populations in an epitope- and isotype-dependent fashion and CD2 downmodulation was accompanied by changes in other T cell surface receptors (26). Given that CD2 has close physical and functional associations with CD3, antiCD2 can act synergistically with anti-CD3 to prolong allograft survival and induce tolerance, and changes in the expression of these two cell surface molecules correlate with the induction of immunosuppression and tolerance; we therefore studied the determinants of receptor modulation in a more defined in vitro system. The results here demonstrate that the anti-CD2 mAb 12-15 significantly decreased CD2 expression with no change in CD3 expression on T lymphocytes. Anti-CD3 mAb 2C11 significantly decreased CD3 expression while increasing CD2 expression. The combination of 12-15 and 2C11 downmodulated CD3 expression and resulted in CD2 expression between that of anti-CD2 and anti-CD3 alone. The modulation of CD2 and CD3 was unlikely to be due to the depletion or expansion of certain cell subpopulations because flow cytometric analysis showed preservation of CD4/ and CD8/ T cell populations, and the same modulating effects were observed on isolated CD4 and CD8 populations. Furthermore, since the modulation of CD2 and CD3 could be observed after only 1 day of culture, it was unlikely
AP: Cell Immuno
256
LIN ET AL.
that major changes in cell populations occurred. AntiCD2 and anti-CD3 mAb induced CD2 and CD3 modulation in a wide variety of circumstances, regardless of growth factors, cytokines, or other cell types present. Therefore, this is a ubiquitous experimental result which was not dependent on specialized culture conditions. The transcriptional data showed that anti-CD2 alone did not significantly affect levels of CD2 or CD3 mRNA while anti-CD3 increased the levels of both messages and that CD2 upmodulation by anti-CD3 was dependent on RNA transcription. The biotin-labeling data showed that anti-CD2 caused internalization of CD2. Therefore, downmodulation of surface receptor expression represents antibody-induced internalization, while increased CD2 expression as a result of antiCD3 represents increased transcription. In the case of apparently stable CD2 expression in anti-CD2 plus anti-CD3 mAb-treated cells, it can be postulated that anti-CD2-induced downmodulation as a result of complex internalization is balanced by anti-CD3-induced upmodulation as a result of increased transcription. In the case of CD3 expression in anti-CD3 mAb-treated cells, receptor complex internalization occurs initially and accounts for receptor downmodulation. Later on, increased transcription causes reexpression of CD3. CD2 has been shown to associate with and modify the function of a variety of cell surface molecules including CD11a/CD18, CD26, CD28, CD38, CD44, CD45, and CD49d (3–7, 10, 45–49). CD2 also has physical and functional associations with the TCR/CD3 complex (6– 9, 49). We therefore analyzed the effect of CD2 and CD3 modulation on the expression of other cell surface molecules. The expression of CD4 and CD8 remained unchanged, showing that T cell subsets were preserved. Our previous studies showed that the expressions of the adhesion molecules LFA-1a (CD11a), LFA1b (CD18), Pgp-1 (CD44), CD45, and MEL-14 (L-selectin, CD62L) were all increased by anti-CD2 mAb alone while other receptors (CD25, CD48, ICAM-I) (CD54) were not increased (26). Similar results were also observed in vivo. The present studies demonstrate that the expressions of the adhesion molecules CD11a/ CD18, CD44, CD45, CD48, CD54, and CD62L are all increased by anti-CD3 mAb but are not further affected by the addition of anti-CD2 mAb. Since the TCR/CD3 complex is the major receptor involved in antigen recognition and CD2 is one of the coreceptors engaged during initial antigen recognition, the upregulation of these other receptors in response to engagement of CD2 or CD3 may represent the consequence of initial activational signals which would promote T–APC or T–target interactions. Since these other adhesion molecules represent members of several distinct molecular families (i.e., selectin, integrin, immunoglobulin), these CD3- and CD2-derived transcriptional signals are likely to be broad. Nonetheless, there is some specificity since not all receptors demonstrate upmodulation. It is considered that anti-CD3 mAb induces immunosup-
/ m4726$8025
01-03-96 15:51:37
cial
pression by first activating T cells and then inducing anergy by overstimulation or exhaustion (50). The upmodulation of an array of cell surface receptors is one aspect of T cell activation. We speculate that another explanation for the immunosuppressive effects of antiCD3 mAb is inappropriate display of other coreceptors. Inability to engage these receptors after upregulation may facilitate subsequent anergy induction (51). Contrary to the upmodulation of most adhesion molecules, anti-CD2 mAb acted synergistically with antiCD3 mAb to downmodulate CD49d expression. CD49d is an adhesion receptor for VCAM-1, fibronectin, MAdCAM-1, and thrombospondin. It is involved in lymphocyte trafficking, homing, and a variety of inflammatory reactions (52–54). CD49d expression is important for T lymphocyte interactions with vascular endothelium and egress from vascular spaces into peripheral tissues and may also contribute to the control of entry into allogeneic tissues. Anti-CD3 significantly decreased CD49d expression, which may prevent effector lymphocytes from transmigrating across vascular endothelium and into allogeneic tissues. Anti-CD2 further decreased CD49d expression when combined with anti-CD3 and thus may further inhibit lymphocyte–endothelium interactions. The synergistic downmodulation of CD49d by the combination of anti-CD2 plus anti-CD3 may contribute to their synergistic effect in allograft prolongation. By implication anti-CD49d mAbs should prolong graft survival and we have obtained evidence that these mAbs prolong graft survival in our cardiac model (Lin and Bromberg, unpublished results). CD25 or IL-2R is important for T cell activation and proliferation through autocrine and paracrine mechanisms and anti-CD3 mAb alone significantly increased CD25 expression. Increased expression of CD25 indicates a state of T cell activation and is mechanistically related to the subsequent release of other cytokines induced by anti-CD3 mAb (20–22). The cytokines induced by anti-CD3 mAb can cause significant clinical side effects. Anti-CD2 mAb administered alone had no effect on CD25 expression, but it significantly decreased anti-CD3-driven expression. This is likely to impede further T cell activation and suppress cytokine release. This finding also likely accounts for the ability of anti-CD2 to ameliorate the anti-CD3-associated cytokine syndrome (24, 25). It will be important to understand the interaction of CD2- and CD3-related second messenger pathways and how receptor-mediated signals influence the expression of other receptors. Preliminary investigations demonstrate that anti-CD2 activates cAMP and protein kinase A activity which impede both IL-2 and IL-2R expression driven by antiCD3 (Lin and Bromberg, unpublished results). One important function for CD2 is as a signal transducer in T cell activation. CD2 can deliver activating signals through combinations of anti-CD2 mAbs, antigen, or mitogens (55, 56), resulting in changes in calcium currents, inositol phosphate turnover, and IL-2
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
production (57–59). These pathways are also common to CD3-mediated T cell activation. Studies have shown a physical association between CD2 and CD3 (6– 9, 60, 61). More recent studies show that CD2 is physically associated with TCR z chain (9, 60) and that a CD2–TCR z complex is important for CD2 function. TCR z is also the portion of the CD3 complex responsible for TCR/CD3 signal transduction. This provides direct evidence that CD2 and CD3 not only interact with each other at the surface receptor level and but also share subsequent signal transduction pathways. The synergism in prolongation of allograft survival by antiCD2 plus anti-CD3 mAbs is probably related to the physical and functional association of the CD2 and CD3 receptors and the subsequent ability of CD2-mediated signaling to enhance or abrogate some aspects of CD3mediated signaling. Studies of anti-CD2-induced immunosuppression show that these mAbs alter immune competence in several different ways. We previously presented evidence that anti-CD2 causes receptor blockade in vitro during CTL–target interaction (62), induces anergy in vivo and in vitro to subsequent CD3associated signals (24, 25), and generates a population of TH2 suppressor cells in vivo (62). Importantly, antiCD2 does not cause T cell depletion. The results here delineate some additional mechanisms by which the mAbs may cause immunosuppression. A major alteration in the array of cell surface receptors induced by anti-CD2 plus anti-CD3 may alter the ability of T cells to traffic, respond to receptor engagement, interact with APC, and release lymphokines. ACKNOWLEDGMENTS We thank Drs. Altevogt, Bluestone, Ogawa, Wong, and Yagita for gifts of reagents and antibodies.
REFERENCES 1. Altevogt, P., Michaelis, M., and Kyewski, B., Eur. J. Immunol. 19, 1509, 1989. 2. Yagita, H., Okumura, K., and Nakauchi, H., J. Immunol. 140, 1321, 1988. 3. Dang, N. H., Torimoto, Y., Sugita, K., Daley, J. F., Schow, P., Prado, C., Schlossman, S. F., and Morimoto, C., J. Immunol. 145, 3963, 1990. 4. Cerdan, C., Martin, Y., and Brailly, H., J. Immunol. 146, 560, 1991. 5. Samelson, L. E., Fletcher, M. C., Ledbetter, J. A., and June, C. H., J. Immunol. 145, 2448, 1990. 6. Bockenstedt, L. K., Goldsmith, M. A., Dustin, M., Olive, D., Springer, T. A., and Weiss, A., J. Immunol. 141, 1904. 7. Holter, W., Majdic, O., Stockinger, H., Howard, B. H., and Knapp, W., J. Immunol. 140, 1043, 1988. 8. Kvanta, A., Gerwins, P., Van der Ploeg, I., Nordstedt, C., Jondal, M., and Fredholm, B. B., Eur. J. Pharmacol. 189, 363, 1990. 9. Gassman, M., Amrein, K. E., Flint, N. A., Schraven, B., and Burn, P., Eur. J. Immunol. 24, 139, 1994. 10. Moingeon, P. E., Lucich, J. L., and Stebbins, C. C., Eur. J. Immunol. 21, 605, 1991.
/ m4726$8025
01-03-96 15:51:37
cial
257
11. Ohno, H., Nakamura, T., Yagita, H., Okumura, K., Taniguchi, M., and Saito, T., J. Immunol. 147, 2100, 1991. 12. Bromberg, J. S., Chavin, K. D., Altevogt, P., Kyewski, B. A., Guckel, B., Naji, A., and Barker, C. F., Transplantation 51, 219, 1991. 13. Guckel, B., Berek, C., Lutz, M., Altevogt, P., Schirrmacher, V., and Kyewski, B. A., J. Exp. Med. 174, 957, 1991. 14. Jonker, M., Malissen, B., and Von Vreeswijk, W., Transplant. Proc. 35, 521, 1983. 15. Chavin, K. D., Lau, H. T., and Bromberg, J. S., Transplant 54, 286, 1992. 16. Vigeral, P., Chkoff, N., Chatenoud, L., Campos, H., Lacombe, M., Droz, D., Goldstein, G., Bach, J. F., and Kreis, H., Transplantation 41, 730, 1986. 17. Ortho Multicenter Transplant Study Group, N. Engl. J. Med. 313, 337, 1985. 18. Cosimi, A. B., Burton, R. C., Colvin, R. B., Goldstein, G., Delmonico, F. L., La, Q. M., Tolkoff, R. N., Rubin, R. H., Herrin, J. T., and Russell, P. S., Transplantation 32, 535, 1981. 19. Alberola-Ila, J., Places, L., de la Calle, O., Romero, M., Yague, J., Gallart, T., Vives, J., and Lozano, F., J. Immunol. 146, 1085, 1991. 20. Hirsch, R., Eckhaus, M., Auchincloss, H. J., Sachs, D. H., and Bluestone, J. A., J. Immunol. 140, 3766, 1988. 21. Alegre, M., Vandenabeele, P., Flamand, V., Moser, M., Leo, O., Abramowicz, D., Urbain, J., Fiers, W., and Goldman, M., Eur. J. Immunol. 20, 707, 1990. 22. Ferran, C., Sheehan, K., Dy, M., Schreiber, R., Merite, S., Landais, P., Noel, L. H., Grau, G., Bluestone, J., and Bach, J. F., Eur. J. Immunol. 20, 509, 1990. 23. Ferran, C., Dy, M., Merite, S., Sheehan, K., Schreiber, R., Leboulenger, F., Landais, P., Bluestone, J., Bach, J. F., and Chatenoud, L., Transplantation 50, 642, 1990. 24. Chavin, K. D., Qin, L., Lin, J., Yagita, H., and Bromberg, J. S., J. Immunol. 151, 7249, 1993. 25. Chavin, K. D., Qin, L., Lin, J., Woodward, J. E., Baliga, P., and Bromberg, J. S., Ann. Surg. 218, 492, 1993. 26. Lin, J., Yon, R. W., Chavin, K. D., Qin, L., Woodward, J., Ding, Y., Yagita, H., and Bromberg, J. S., Transplantation 59, 1162, 1995. 27. Altevogt, P., Kohl, U., Von Hoegen, P., Lang, E., and Schirrmacher, V., Eur. J. Immunol. 19, 341, 1989. 28. Leo, O., Foo, M., Sachs, D. H., Samelson, L. E., and Bluestone, J. A., Proc. Natl. Acad. Sci. USA 84, 1374, 1987. 29. Kato, K., Koyanagi, M., Okada, H., Takanashi, T., Wong, Y. W., Williams, A. F., Okumura, K., and Yagita, H., J. Exp. Med. 176, 1241, 1992. 30. Wilde, D. B., Marrack, P., Kappler, J., Dialynas, D. P., and Fitch, F. W., J. Immunol. 131, 2178, 1983. 31. Ledbetter, J. A., and Herzenberg, L. A., Immunol. Rev. 47, 63, 1979. 32. Furth, M. E., Davis, L. J., Fleurdelys, B., and Scolnick, E. M., J. Virol. 43, 294, 1982. 33. Gustavo, O. R., Robb, R. J., Shevach, E. M., and Malek T. R., J. Immunol. 133, 1970, 1984. 34. Miyake, K., Medina, K. L., Hayashi, S., Ono, S., Harnaoka, T., and Kincade, P. W., J. Exp. Med. 171, 477, 1990. 35. Springer, T., Galfre, G., Secher, D. S., and Milstein, C., Eur. J. Immunol. 8, 539, 1978. 36. Holzmann, B., McIntyre, B. W., and Weissman, I. L., Cell 56, 37, 1989. 37. Holzmann, B., and Weissman, I. L., EMBO J. 8, 1735, 1989. 38. Prieto, J., Takei, F., Gendelman, R., Christenson, B., Biberfeld, P., and Patarroyo, M., Eur. J. Immunol. 19, 1551, 1989.
AP: Cell Immuno
258
LIN ET AL.
39. Sanchez-Madrid, F., Simon, P., Thompson, S., and Springer, T. A., J. Exp. Med. 158, 586. 40. Kubo, R. T., Born, W., Kappler, J. W., Marrac, P., and Pigeon, M., J. Immunol. 142, 2736, 1989. 41. Holzmann, B., McIntyre, B. W., and Weissman, I. L., Cell 56, 37, 1989. 42. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3, 645, 1973. 43. Gold, D. P., Clevers, H., and Alarcon, B., Proc. Natl. Acad Sci. USA 84, 7649, 1987. 44. Abraham, D. J., Bou-Gharios, G., Beauchamp, J. R., PlaterZyberk, Maini, R. N., and Olsen, I., J. Leukocyte Biol. 49, 329, 1991. 45. Funaro, A., Spagnoli, G. C., Ausiello, C. M., Alessio, M., Roggero, S., Delia, D., Zaccolo, M., and Malavasi, F., J. Immunol. 145, 2390, 1990. 46. Conrad, P., Rothman, B. L., Kelley, K. A., and Blue, M. L., J. Immunol. 149, 1833, 1992. 47. Blue, M. L., Conrad, P., Davis, G., and Kelley, K. A., Cell Immunol. 138, 238, 1991. 48. Altin, J. G., Papler, E. B., and Parish, C. R., Eur. J. Immunol. 24, 450, 1994. 49. Dianzani, U., Redoglia, V., Malavasi, F., Bragardo, M. , Pileri, A., Janeway, C. A., and Bottomly, K., Eur. J. Immunol. 22, 365, 1992. 50. Anasett, C., Tan, P., Hansen, J. A., and Martin, P. J., J. Exp. Med. 172, 1691, 1990.
/ m4726$8025
01-03-96 15:51:37
cial
51. Lin, H., Bolling, S. F., Linsley, P. S., Wei, R.-Q., Gordon, D., Thompson, C. B., and Turka, L. A. J. Exp. Med. 178, 1801, 1993. 52. Paul, L. C., Davidoff, A., Benediktsson, H., and Issekutz, T. B., Transplant 55, 1196, 1993. 53. Orosz, C. G., Ohye, R. G., Pelletier, R. P., VanBuskirk, A. M., Huang, E., Morgan, C., Kincade, P. W., and Ferguson, R. M., Transplant 56, 453, 1993. 54. Turunen, J. P., Mattila, P., Halttunen, J., Hayry, P., and Renkonen, R., Transplant 54, 1053, 1992. 55. Moingeon, P., Chang, H. C., Sayre, P. H., Clayton, L. K., Alcover, A., Gardner, P., and Reinherz, E. L., Immunol. Rev. 111, 111, 1989. 56. Bierer, B. E., and Burakoff, S. J., J. Immunol. Rev. 111, 267, 1989. 57. Van Kooyk, Y., Van de Weil-Van Kemenade, P., Weder, P., Kuijpers, T. W., and Figdor, C. G., Nature 342, 811, 1989. 58. Monostori, E., Desai, D., Brown, M. H., Cantrell, D. A., and Crumpton, M. J., J. Immunol. 144, 1010, 1990. 59. Lorre, K., van Verwilghen, D. J., Baroja, M. J., and Ceuppens, J. L., J. Immunol. 144, 4681, 1990. 60. Moingeon, P., Lucich, J. L., McConkey, D. J., Letourneur, F., Malissen, B., Kochan, J., Chang, H. C., Rodewald, H. R., and Reinherz, E. L., Proc. Natl. Acad. Sci. USA 89, 1492, 1992. 61. Brown, M. H., Cantrell, D. A., Brattsand, G., Crumpton, M. J., and Gullberg, M., Nature 339, 551, 1989. 62. Chavin, K. D., Qin, L., Yon, R., Lin, J., Yagita, H., and Bromberg, J. S., J. Immunol. 152, 3729, 1994.
AP: Cell Immuno
167, 249–258 (1996)
Article No. 0033
Anti-CD2 Monoclonal Antibody-Induced Receptor Changes II. Interaction of CD2 and CD3 JIXUN LIN,*,† KENNETH D. CHAVIN,†,‡ LIHUI QIN,* YAOZHONG DING,*,§
AND
JONATHAN S. BROMBERG*,§,1,2
Departments of *Surgery and †Microbiology and Immunology, University of Michigan, Ann Arbor, Michigan 48109; and Departments of §Surgery and ‡Microbiology and Immunology, Medical University of South Carolina, Charleston, South Carolina 29425 Received June 21, 1995; accepted September 14, 1995
Anti-CD3 monoclonal antibodies (mAbs) and antiCD2 mAbs each prolong allograft survival and cause transient downmodulation of homologous receptor expression. Anti-CD2 mAbs also act synergistically with anti-CD3 mAbs to prolong allograft survival and induce tolerance. The effect of combined anti-CD2 and anti-CD3 mAb treatment on receptor expression was further analyzed with an in vitro model. The antiCD2 mAb 12-15 caused CD2 expression on purified splenic T cells to decrease from 72.6% [mean channel fluorescence (MCF) 0.68] to 41.5% (0.45) total positive cells while CD3 expression remained unchanged [69.1% (3.47) to 76.4% (4.04)]. The anti-CD3 mAb 2C11 caused CD2 expression to increase from 72.6% (0.68) to 93.0% (1.74) while CD3 expression decreased from 69.1% (3.47) to 62.6% (2.15). The combination of antiCD2 plus anti-CD3 preserved CD2 expression (72.6 to 71.1%) while still decreasing CD3 expression [69.1% (3.47) to 69.9% (2.37)]. Modulation of CD2 and CD3 expression was similar on mixed splenic T lymphocytes and isolated CD4 and CD8 subsets. Modulation did not change with the addition of the cytokines IL-1, IL-2, IL-4, IL-6, IL-10, TNFa, or TGFb. Kinetic studies showed that modulation of CD2 was rapid, persistent, and of the same magnitude from Day 1 to Day 7 of culture while CD3 downmodulation was transient. The results of transcriptional analysis and receptor distribution suggested that downmodulation was due to receptor internalization while upmodulation was due to increased transcription. Analysis of expression of other adhesion molecules demonstrated that CD11a, CD18, CD44, CD45, CD48, CD54, and CD62L were significantly increased by either anti-CD2 or anti-CD3 mAbs while the combination was not synergistic. However, anti-CD3 significantly decreased VLA-4a (CD49d) expression and anti-CD2 enhanced 1 Supported by American Surgical Association Foundation Fellowship Award and NIH Grant AI32655. 2 To whom correspondence should be addressed at Department of Surgery, University of Michigan, 2926 Taubman Center, Box 0331, Ann Arbor, MI 48109-0331.
this decrease. Conversely anti-CD3 significantly increased IL-2R (CD25) expression and anti-CD2 profoundly inhibited the increase. These results show that anti-CD2 and anti-CD3 mAbs significantly modulate CD2 and CD3 expression on T cells and modulation is accompanied by changes in the array of other T cell surface receptors. Changes in cell surface receptor display may provide an additional explanation for the synergistic effect of anti-CD2 plus anti-CD3 in prolonging allograft survival. q 1996 Academic Press, Inc.
INTRODUCTION The cell surface molecule CD2 is widely distributed in the immune system. It is a 55kDa transmembrane glycoprotein found on T cells, B cells, NK cells, and some antigen-presenting cells (APCs)3 in mice (1, 2). CD2 functions as an adhesion receptor which strengthens the interaction between helper T cells and APCs or between effector T cells and target cells. CD2 also plays a role in T cell activation (3–5) and modulates CD3-related events (6). CD2 shares some common second messenger pathways of T cell activation with CD3 (7–9) and has also been demonstrated to regulate the function of a variety of other cell surface molecules (3– 7, 9, 10). CD2 therefore appears to play a major regulatory role in immunity. Anti-mouse CD2 mAbs can deliver a negative signal to T cells and inhibit CD3-related events in vitro under certain circumstances (11). Anti-CD2 mAbs administered in vivo also suppress multiple T cell-dependent immune responses (12–14). Associated with suppression of immunity is downmodulation of cell surface CD2 without cellular depletion (12, 13). Our previous studies also showed that anti-CD2 mAbs could prolong allograft survival and this immunosuppression was accompanied by the downmodulation of T cell CD2 expression 3 Abbreviations used: APC, antigen-presenting cell; mAb, monoclonal antibody; MCF, mean channel fluorescence.
249
/ m4726$8025
01-03-96 15:51:37
cial
0008-8749/96 $12.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AP: Cell Immuno
250
LIN ET AL.
in vivo (12, 15). However, anti-CD2 mAb alone did not produce tolerance. Anti-CD3 mAb has also been used in transplantation to prolong allograft survival. Its use encompasses both prophylactic administration to prevent rejection (16) and therapeutic treatment of acute rejection (17, 18). Administration of anti-CD3 results in modulation of CD3 expression and that of a number of other surface receptors (19). While anti-CD3 mAb is a potent immunosuppressant, it fails to produce true, long-term tolerance (20) and has associated side effects, including a deleterious cytokine syndrome which is attributed to T cell activation and cytokine release (21–23). Therefore, the goal of tolerance without significant toxicity has not been met with this reagent. Since CD2 is functionally and physically associated with CD3 as determined by coprecipitation studies and demonstration of common signaling pathways (6–9), it was hypothesized that anti-CD2 could act synergistically with anti-CD3 in inducing immunosuppression. We therefore demonstrated that combined treatment with anti-CD2 plus anti-CD3 mAbs prolonged allograft survival, induced alloantigen-specific tolerance, and diminished the anti-CD3-induced cytokine syndrome (24, 25). We also noted that changes in cell surface receptor expression correlated exactly with both immunosuppression and tolerance in vivo. Previous studies also demonstrated that the extent of CD2 downmodulation produced by anti-CD2 mAb is directly related to its inhibitory effect on cell-mediated immunity in vivo (12) and on anti-CD3-driven T cell activation in vitro (26). There is a similar relationship between the immunosuppressive capacity of anti-CD3 mAbs and changes in receptor expression (17–20). These results suggest that antibody–ligand interactions and subsequent changes in receptor expression are important for the process of immunosuppression and tolerance induction. Therefore, we sought to define the determinants of receptor modulation by studying the effects of anti-CD2 and anti-CD3 mAbs on CD2 and CD3 expression on murine T cells with an in vitro model. Results show that anti-CD2 and anti-CD3 mAbs modulate CD2 and CD3 expression similarly on CD4/ and CD8/ T cell subpopulations. While anti-CD2 and anti-CD3 upmodulate most adhesion receptors, they downmodulate CD49d synergistically and act antagonistically on the expression of CD25. Various cytokines have no effect on CD2 and CD3 expression, showing that these receptor–ligand interactions do not affect CD2–anti-CD2 or CD3–anti-CD3 interactions and subsequent second signals. These studies suggest that one mechanism for the synergistic effect of antiCD2 plus anti-CD3 on allograft survival is a significant change in the array of cell surface receptors. MATERIALS AND METHODS Animals. BALB/cByJ (H-2d) female mice 8–10 weeks of age were purchased from Jackson Laboratory (Bar Harbor, ME).
/ m4726$8025
01-03-96 15:51:37
cial
Antibodies and other reagents. The rat IgG1 antimurine CD2 12-15 hybridoma (27) was a gift of Dr. P. Altevogt (Heidelberg, Germany); the 145-2C11 (antiCD3e) hybridoma (28) was a gift of Dr. J. A. Bluestone (University of Chicago), and the HM48-1 (anti-CD48) hybridoma (29) was a gift of Dr. H. Yagita (Tokyo, Japan). The GK1.5 (anti-CD4) (30), 53-6.72 (anti-CD8) (31), Y13-259 (anti-Ras) (32), 3C7 (anti-CD25) (33), KM201 (anti-CD44) (34), M1/89 (anti-CD45) (35), R12 (anti-CD49d) (36, 37), YN1/1.7.4 (anti-CD54) (38), M17/4.2 (anti-CD11a), M18/2.a.8 (anti-CD18) (39), H57597 (anti-CD3 a/b) (40), and MEL-14 (anti-CD62, L-selectin) (41) hybridomas were purchased from the American Type Culture Collection (Rockville, MD). All hybridomas were grown in culture and purified over protein G (Pharmacia, Piscataway, NJ). Phycoerythrin L3T4 (Becton–Dickinson, San Jose, CA), phycoerythrin–streptavidin (Becton–Dickinson), biotin–xxNHS (Calbiochem, San Diego, CA), affinity-purified F(ab*)2 FITC goat anti-rat IgG (TAGO, Inc., Burlingame, CA), and affinity-purified FITC goat antihamster IgG (Kirkegaard & Perry Lab,Inc., Gaithersburg, MD) were used as alternative or second-step reagents. Recombinant murine IL-1, rmIL-2, rmIL-4, and rmIL-6 (gifts from Dr. M. Ogawa, MUSC), murine IL10 COS cell supernatant with an activity of 3000 units/ ml (a gift of Dr. William Fanslow, Immunex, Seattle, WA), rmTNFa (a gift of Dr. Grace Wong, Genentech, San Francisco, CA), rmTGFb (Gibco BRL, Gaithersburg, MD), Con A (Sigma, St. Louis, MO), PHA (Sigma), or Con A supernatant was used as a source of either stimulation or growth factors in conditioned medium. Antibody conjugation. Purified mAbs were dialyzed against FITC labeling buffer (0.05 M boric acid, 0.2 M NaCl, pH 9.2) at 47C. Antibody concentrations were determined by A280. Twenty microliters of 5 mg/ml FITC in DMSO for each milligram of antibody in 1 ml labeling buffer was added and incubated for 2 hr at room temperature in the dark. Unbound FITC was removed by filtration over Sephadex G-25 and titer determined by flow cytometry. For biotin labeling, purified mAbs were dialyzed against biotin labeling buffer (0.1 M NaHCO3, 0.1 MNaCl, pH 7.4). Ten microliters of 10 mg/ml biotin in DMSO for each milligram of antibody in 1 ml labeling buffer was added and incubated for 1 hr at room temperature. Unbound biotin was removed by filtration over Sephadex G-25. Cell preparation. Mice were sacrificed, and spleens were removed and gently dissociated into a single cell suspension. Red blood cells were removed by TrisNH4Cl lysis. Cell suspensions were passed through nylon wool columns to enrich for T cells (42). These cells were routinely ú70% T cells. Cells were plated at 1 1 106 cells in 2-ml wells in complete RPMI medium (RPMI 1640 supplemented with 10% FCS, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 IU/ml penicillin, 100 mg/ml streptomycin, 1 1 nonessential amino
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
251
acids, and 2 1 1005 M 2-mercaptoethanol). Cultures were routinely supplemented with rmIL-2 (0.15 U/ml) as a source of growth factor. MAbs or other cytokines were added to the wells as indicated. Control groups received either control mAbs or no treatment. Cells were harvested 1–7 days later and stained, and flow cytometric analysis was performed. Separation of CD4 and CD8 subsets. T cell-enriched splenic lymphocytes isolated over nylon wool columns as described above were further negatively selected by antibody column depletion. Anti-CD4 and anti-CD8 columns were purchased (R & D Systems, Minneapolis, MN) and used according to manufacturer’s instructions. These columns removed not only the relevant T cell subsets but also B cells and macrophages. Flow cytometry. Cell washes and antibody dilutions were performed in phosphate-buffered saline plus 1% bovine serum albumin at 47C. CD2 was stained with the 12-15 mAb (1:30) followed by affinity-purified F(ab*)2 FITC goat anti-rat IgG (H / L) (1:100). CD4 was stained with PE-L3T4 (1:100). CD3, CD8, CD11a, CD18, CD25, CD44, CD45, CD49d, and MEL-14 were all stained directly by the FITC-labeled antibodies described above. Negative controls included second antibodies with or without irrelevant first mAbs in all experiments. Saturating concentrations of antibodies were used as determined by dose–response curves, which were repeated several times during the course of these studies. Flow cytometric analysis was performed on an Epics Elite flow cytometer (Coulter, Hialeah, FL) or a FACScan flow cytometer (Becton–Dickinson). Results are expressed as percentages of cells staining above background or changes in mean channel fluorescence on a logarithmic scale of relative cellular fluorescence. Northern hybridization. Cultured cells were harvested and whole-cell RNA was isolated using RNAzol B (Cinna/Biotecx, Houston, TX). Equal quantities of RNA were electrophoresed on 1% formaldehyde agarose gels and transferred to nylon membranes. The membranes were sequentially probed for CD2, CD3, and b-actin using 32P random primer (Stratagene, La Jolla, CA) labeled probes. The CD2 probe was the the 0.7-kb EcoRI– HpaI fragment from the murine CD2 cDNA clone pMCD2-2 (2). The CD3 probe was the 1.4-kb EcoRI fragment of CD3e (43). The b-actin probe was the 540-base pair PCR product between bases 25 and 565 generated from C57BL/6 genomic DNA. Radiolabeled hybridization products were analyzed by autoradiography and densitometry by NIH Image 1.49 software. RESULTS Modulation of CD2 and CD3 Expression by Anti-CD2 and Anti-CD3 mAbs Anti-CD2 mAb 12-15, anti-CD3 mAb 2C11, and/or the irrelevant IgG1 isotype control mAb 259 were
/ m4726$8025
01-03-96 15:51:37
cial
FIG. 1. Modulation of T cell surface CD2 and CD3 by anti-CD2 and anti-CD3 mAbs. Nylon wool-purified splenic T lymphocytes incubated with 5 mg/ml of control Y13-259 mAb or anti-CD2 mAb 12-15 or 0.5 mg/ml anti-CD3 mAb 2C11 for 3 days were analyzed for cell surface expression of CD2 and CD3. Numbers in upper right are mean channel fluorescence for gated population.
added to cultures of murine splenic T lymphocytes. As shown in Fig. 1, flow cytometric analysis demonstrated that anti-CD2 mAb caused lymphocytic expression of CD2 to decrease from 72.6% (MCF 0.68) to 41.5% (0.45) while expression of CD3 remained unchanged [69.1% (3.47) to 76.4% (4.04)] after 3 days of culture with antiCD2 mAb compared to the cells in the control group. Conversely anti-CD3 mAb caused CD2 expression to increase from 72.6% (0.68) to 93.0% (1.74) and CD3 expression to decrease from 69.1% (3.47) to 62.6% (2.15). The combination of mAbs caused CD2 expression to remain unchanged (72.6% (0.68) to 71.1% (0.72)) while CD3 expression still decreased (69.1% (3.47) to 69.9% (2.37)). Staining for CD3 a/b showed identical results as for CD3e, demonstrating that the entire CD3 complex was modulated. Because indirect immunofluorescent techniques were used, cell surface CD2 and CD3 receptors coated with mAb from culture, and CD2 and CD3 which remained unbound by mAb in culture, were both quantitated by the flow cytometric methods
AP: Cell Immuno
252
LIN ET AL.
ulating culture conditions nor influenced by interaction of multiple cytokines with their receptors. Modulation of CD2 and CD3 Expression on T Cell Subpopulations
FIG. 2. Modulation of CD2 and CD3 by anti-CD2 and anti-CD3 mAb is dose dependent. Nylon wool-purified splenic T lymphocytes incubated with the indicated doses of anti-CD2 mAb 12-15 or antiCD3 mAb 2C11 for 3 days were assessed for cell surface expression of CD2 and CD3.
employed here. Further, controls showed that these mAbs did not sterically hinder each other (not shown). These results demonstrate that the anti-CD2 mAb 1215 and the anti-CD3 mAb 2C11 could downmodulate CD2 and CD3 expression, respectively. Combined antiCD2 plus anti-CD3 could still downmodulate CD3, but CD2 expression was increased compared to anti-CD2 alone and decreased compared to anti-CD3 alone. Dose–response data showed that 12-15 at 5 mg/ml induced maximal downmodulation of CD2 expression and 2C11 at 0.5–5 mg/ml induced maximal downmodulation of CD3 and upmodulation of CD2 expression simultaneously (Fig. 2). In kinetic experiments, cells were harvested after 1 to 7 days of culture and examined. Flow cytometric analysis showed that modulation of CD2 and CD3 expression was rapid with downmodulation seen within 24 hr of culture. The effect was persistent for CD2 since the magnitude of modulation remained constant throughout 7 days of culture while CD3 was slowly reexpressed over time (not shown). The cultures shown here were supplemented with 0.15 U/ ml purified recombinant murine IL-2 as a T cell growth factor. Supplementation with different doses of rmIL2, Con A, PHA, or Con A supernatant as a source of conditioned medium resulted in identical results (not shown). Various cytokines were also used to supplement cultures and evaluated for their effect on CD2 and CD3 modulation. The results in Fig. 3 show that modulation of CD2 and CD3 does not change with the addition of IL-4 or IL-6. Identical results were obtained with IL-1, IL-10, TNFa, and TGFb (not shown). Therefore, modulation induced by anti-CD2 and anti-CD3 mAbs was neither dependent on a restricted set of stim-
/ m4726$8025
01-03-96 15:51:37
cial
Since murine CD2 is expressed not only on T lymphocytes but also on B cells, NK cells, and some APCs, and since CD3 is expressed exclusively on T cells, it was important to analyze CD2 and CD3 expression on more highly purified CD4/ and CD8/ subpopulations. AntiCD2 mAb caused downmodulation of CD2 expression and had no effect on CD3 expression on all populations tested. CD2 expression decreased from 81.9 to 31.9% on nylon wool-enriched T cells, from 93.9 to 35.6% on CD8-depleted (CD4/) T cells, and from 88.8 to 57.9% on CD4-depleted (CD8/) T cells. Anti-CD3 mAb caused upmodulation of CD2 expression on nylon wool-enriched T cells from 81.9% (0.76) to 94.3% (2.14), CD8depleted T cells from 88.8% (1.33) to 95.2% (2.30), and CD4-depleted T cells from 88.8% (1.85) to 93.1% (2.61). Anti-CD3 mAb also induced downmodulation of CD3 expression on nylon wool-enriched T cells from 87.2% (1.35) to 84.2% (0.99), CD8-depleted T cells from 95.2% (11.2) to 90.1% (6.34), and CD4-depleted T cells from 93.9% (11.7) to 92.6% (7.91). Combined anti-CD2 and -CD3 mAbs preserved CD2 expression but still downmodulated CD3 expression on all the populations (Fig. 4). The results demonstrate that anti-CD2 caused CD2 downmodulation, anti-CD3 caused CD2 upmodulation and CD3 downmodulation, and the combination of mAbs preserved CD2 expression while downmodulating CD3 on both CD4/ cells and CD8/ cells. It should
FIG. 3. Cytokines do not affect mAb-induced down modulation. IL-4 or IL-6 at 50 U/ml was added at the initiation of culture with 5 mg/ml 12-15 and 0.5 mg/ml 2C11. Cells were harvested and labeled after 3 days of culture.
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
253
FIG. 4. Anti-CD2 mAb 12-15 and anti-CD3 mAb 2C11 cause modulation of CD2 and CD3 expression on various T cell subtypes. The indicated cell types were cultured with 5 mg/ml 12-15 and/or 0.5 mg/ml 2C11 for 3 days and then labeled for flow cytometry.
be noted that the experiment in Fig. 4 is from a 3-day culture at which time CD3 is starting to be reexpressed. CD2 and CD3 Transcription and Distribution Previous work demonstrated that the major mechanism of CD2 (44) and CD3 (20–22) downmodulation was internalization of antigen–antibody complexes, rather than capping and shedding of the complex. Our previous work (26) also demonstrated that CD2 mRNA levels of anti-CD2-treated cells were unchanged compared to controls, suggesting that anti-CD2 mAb had little effect on the level of CD2 mRNA transcription, which supports the finding that antigen–antibody internalization is the major mechanism for downmodulation. It remained to be determined whether the alterations in expression of CD2 and CD3 by anti-CD2 plus anti-CD3 were due to changes at the transcription or internalization levels. RNA was isolated from control or treated cells and hybridized with probes for CD2, CD3, or b-actin. The results (Fig. 5) show that CD2 mRNA levels were increased at Day 2 and Day 4 as a result of the activational effects of anti-CD3 mAb, either alone or in combination with anti-CD2. CD3 mRNA levels were increased by anti-CD3 on Day 4 but not on Day 2. Neither anti-CD2 nor anti-CD3 mAb decreased the levels of CD2 or CD3 mRNA, and antiCD2 mAb alone had no significant effect on mRNA expression. To examine the effects of mAb treatment on internalization of receptors, splenic T lymphocytes were incubated with biotin-labeled anti-CD2 mAb, washed, then cultured with or without anti-CD3, and stained with phycoerythrin–streptavidin either before or after culture. In addition cells from some groups were reincubated with biotin–anti-CD2 at the end of culture and
/ m4726$8025
01-03-96 15:51:37
cial
then stained with phycoerythrin. In this fashion then the fate of CD2 originally present on the cells and the expression of new CD2 could be determined. The results (Fig. 6) demonstrate that cells stained with phycoerythrin at the initiation of culture and analyzed by
FIG. 5. Northern hybridization. Purified T cells were cultured with 12-15, Y13-259, and/or 2C11 for 2 or 4 days, and total RNA was isolated and blotted with probes for CD2, CD3, or b-actin. For CD2 a total of 3.8 mg RNA was loaded per lane; for CD3 and b-actin 3.0 mg RNA was loaded per lane. Lane 1, Y13-259; lane 2, 12-15; lane 3, 2C11; lane 4, 12-15 / 2C11. Numbers underneath blots represent ratios of arbitrary units of optical density normalized to lane 1 and to b-actin.
AP: Cell Immuno
254
LIN ET AL.
analysis showed a marked decrease which was true if the cells were incubated with or without anti-CD3 (curves 5 and 6). If cells from these last two groups were stained at the end of culture with additional biotin–anti-CD2 followed by phycoerythrin, then flow cytometry demonstrated that the anti-CD3-treated group had increased CD2 (curve 8) up to the level of the control (curve 2), while the group not treated with antiCD3 had only a minimal increase in CD2 (curve 7). The results in Figs. 5 and 6 taken together show that antiCD2 causes downmodulation of CD2 by internalization and anti-CD3 causes upmodulation of CD2 by transcriptional increase. This is further supported by the results in Figs. 6D and 6E which demonstrate that actinomycin D can block the majority of anti-CD3driven upmodulation of CD2. Expression of Other Cell Surface Receptors
FIG. 6. (A–C) Anti-CD2 causes internalization of CD2 and antiCD3 causes expression of new CD2 molecules. Splenic T lymphocytes were labeled with biotin–anti-CD2 prior to culture. Curve 1, negative control; 2, stained with phycoerythrin–streptavidin (PE) and analyzed prior to culture; 3, stained with PE prior to and analyzed after 24 hr of culture; 4, as for 3 but cultured with anti-CD3; 5, stained with PE and analyzed after culture; 6, as for 5 but cultured with anti-CD3; 7, stained with additional biotin–anti-CD2 and PE and analyzed, all after culture; 8, as for 7, but cultured with anti-CD3. (D and E) Upmodulation of CD2 by anti-CD3 is inhibited by transcriptional blockade. T lymphocytes were incubated with or without actinomycin D (50 mg/ml) or anti-CD3 mAb, harvested after 24 hr, and stained for CD2. Curve 1, negative control; 2, cells alone; 3, cells plus anti-CD3; 4, cells plus actinomycin D; 5, anti-CD3 plus actinomycin D.
flow cytometry either at that time or 24 hr later showed nearly identical staining (curves 2 and 3). This was also true if the cells were incubated with or without anti-CD3 (curves 3 and 4). If phycoerythrin staining was delayed until the end of culture, then fluorescent
/ m4726$8025
01-03-96 15:51:37
cial
Since anti-CD2 and anti-CD3 induce both CD2 and CD3 modulation and immunosuppression, and since CD2 and CD3 have physical and functional associations with other surface receptors, it was important to see if they affected the expression of other important cell surface receptors. Therefore, cells incubated in culture with the mAbs were subsequently directly labeled with a series of FITC-conjugated mAbs directed toward other cell surface adhesion receptors. Control experiments demonstrated that 12-15 and 2C11 did not sterically hinder the labeling by any of these other mAbs (not shown); therefore, anti-CD2 and anti-CD3 bound to the cell surface in culture would not interfere with subsequent labeling steps. The results in Table 1 show that CD4 remained unaffected by concomitant CD2 and CD3 modulation, and CD8 expression also remained stable with anti-CD2 treatment but increased slightly with anti-CD3 treatment. This demonstrates that antiCD2 and anti-CD3 did not cause a major preferential loss or growth of T cells or their subsets. Most other receptors, representing members of the selectin, integrin, and immunoglobulin gene superfamily, were upmodulated slightly by anti-CD2 and more significantly by anti-CD3, while the combination of mAbs was not different than anti-CD3 alone (Table 1). However, the effects on CD25 and CD49d expression were unique. CD25 expression increased significantly from 28.3% (2.38) to 95.9% (6.02) after treatment with anti-CD3, due to the activational properties of anti-CD3. While anti-CD2 alone had no effect on CD25 expression [28.3 to 24.3% (2.45)], anti-CD2 significantly inhibited antiCD3-driven CD25 expression from 95.9 to 63.8% (3.58) (Fig. 7). CD49d expression remained unchanged from 85.6% (0.95) to 90.2% (1.16) after anti-CD2 treatment, but decreased significantly from 85.6 to 57.4% (0.93) after anti-CD3 treatment. CD49d expression decreased further from 57.4 to 47.0% (0.77) with combined antiCD2 plus anti-CD3 treatment (Fig. 7). This suggests that immunologically important effects of anti-CD2
AP: Cell Immuno
255
INTERACTION OF CD2 AND CD3
TABLE 1 Anti-CD2 and Anti-CD3 mAbs Modulate Expression of Adhesion Receptors Cell surface receptor % positive cells (mean channel fluorescence) Treatment group Control Anti-CD2 Anti-CD3 Anti-CD2/anti-CD3
CD2 81.9 30.1 94.3 73.3
CD3e
(0.76) (0.52) (2.14) (0.81)
87.2 91.3 84.2 88.3
CD3a/b
(1.35) (1.67) (0.99) (1.10)
74.4 78.8 46.5 29.1
CD45 Control Anti-CD2 Anti-CD3 Anti-CD2/anti-CD3
86.8 92.2 95.6 98.8
(1.64) (1.97) (5.25) (3.98)
(0.75) (0.99) (0.69) (0.84)
CD48 31.9 28.3 33.3 35.4
(0.40) (0.43) (0.44) (0.50)
CD11a 36.9 42.8 92.8 91.3
(0.76) (1.36) (1.76) (1.43) CD49d
85.6 90.2 57.4 47.0
(0.95) (1.16) (0.93) (0.77)
CD18C 34.9 41.2 88.3 85.7
D25
(0.67) (1.27) (1.21) (0.96)
14.5 13.1 23.7 19.4 CD54
43.9 38.0 65.6 70.2
(0.92) (0.71) (0.67) (0.70)
(0.63) (0.71) (0.78) (0.88)
CD44 57.8 56.5 87.7 85.9
(1.56) (2.01) (3.71) (1.99)
CD62L 69.2 71.2 95.7 91.3
(2.90) (3.27) (14.9) (7.05)
Note. Purified T lymphocytes were incubated with control mAb Y13–259 or 12–15 at a 5 mg/ml or 2C11 at 0.5 mg/ml for 3 days in culture and evaluated by indirect or direct immunofluorescence for expression of the indicated receptors.
and anti-CD3 may be related not only to CD2 and CD3 modulation but also to changes in the entire cell surface receptor display. The effects of anti-CD2 mAb on the modulation of cell surface receptor expression induced by anti-CD3 mAbs can therefore be categorized into at least three patterns. First, anti-CD2 mAb had no effect on anti-CD3 mAb-induced upmodulation of an array of adhesion molecules; in fact, anti-CD2 alone partially upmodulated these receptors. Second, anti-CD2 mAb acted synergistically with anti-CD3 mAb to downmodulate CD49d expression. Third, anti-CD2 mAb inhibited
FIG. 7. Anti-CD2 antagonizes anti-CD3-driven upmodulation of CD25, and anti-CD2 plus anti-CD3 synergistically downmodulates CD49d expression. 12-15 at 5 mg/ml and/or 2C11 at 0.5 mg/ml were added to culture with T lymphocytes for 3 days. Il-2 was added at 125 U/ml to culture for CD25 measurements.
/ m4726$8025
01-03-96 15:51:37
cial
the upmodulation of CD25 expression induced by antiCD3 mAb. DISCUSSION Our previous work demonstrated that anti-CD2 mAb was effective in inducing immunosupression and in prolonging allograft survival (12, 15). The immunosuppression induced by anti-CD2 mAbs was accompanied by the concomitant downmodulation of CD2 expression in vivo (11). Our recent in vitro studies showed that anti-CD2 mAbs significantly and persistently downmodulated CD2 on various T cell populations in an epitope- and isotype-dependent fashion and CD2 downmodulation was accompanied by changes in other T cell surface receptors (26). Given that CD2 has close physical and functional associations with CD3, antiCD2 can act synergistically with anti-CD3 to prolong allograft survival and induce tolerance, and changes in the expression of these two cell surface molecules correlate with the induction of immunosuppression and tolerance; we therefore studied the determinants of receptor modulation in a more defined in vitro system. The results here demonstrate that the anti-CD2 mAb 12-15 significantly decreased CD2 expression with no change in CD3 expression on T lymphocytes. Anti-CD3 mAb 2C11 significantly decreased CD3 expression while increasing CD2 expression. The combination of 12-15 and 2C11 downmodulated CD3 expression and resulted in CD2 expression between that of anti-CD2 and anti-CD3 alone. The modulation of CD2 and CD3 was unlikely to be due to the depletion or expansion of certain cell subpopulations because flow cytometric analysis showed preservation of CD4/ and CD8/ T cell populations, and the same modulating effects were observed on isolated CD4 and CD8 populations. Furthermore, since the modulation of CD2 and CD3 could be observed after only 1 day of culture, it was unlikely
AP: Cell Immuno
256
LIN ET AL.
that major changes in cell populations occurred. AntiCD2 and anti-CD3 mAb induced CD2 and CD3 modulation in a wide variety of circumstances, regardless of growth factors, cytokines, or other cell types present. Therefore, this is a ubiquitous experimental result which was not dependent on specialized culture conditions. The transcriptional data showed that anti-CD2 alone did not significantly affect levels of CD2 or CD3 mRNA while anti-CD3 increased the levels of both messages and that CD2 upmodulation by anti-CD3 was dependent on RNA transcription. The biotin-labeling data showed that anti-CD2 caused internalization of CD2. Therefore, downmodulation of surface receptor expression represents antibody-induced internalization, while increased CD2 expression as a result of antiCD3 represents increased transcription. In the case of apparently stable CD2 expression in anti-CD2 plus anti-CD3 mAb-treated cells, it can be postulated that anti-CD2-induced downmodulation as a result of complex internalization is balanced by anti-CD3-induced upmodulation as a result of increased transcription. In the case of CD3 expression in anti-CD3 mAb-treated cells, receptor complex internalization occurs initially and accounts for receptor downmodulation. Later on, increased transcription causes reexpression of CD3. CD2 has been shown to associate with and modify the function of a variety of cell surface molecules including CD11a/CD18, CD26, CD28, CD38, CD44, CD45, and CD49d (3–7, 10, 45–49). CD2 also has physical and functional associations with the TCR/CD3 complex (6– 9, 49). We therefore analyzed the effect of CD2 and CD3 modulation on the expression of other cell surface molecules. The expression of CD4 and CD8 remained unchanged, showing that T cell subsets were preserved. Our previous studies showed that the expressions of the adhesion molecules LFA-1a (CD11a), LFA1b (CD18), Pgp-1 (CD44), CD45, and MEL-14 (L-selectin, CD62L) were all increased by anti-CD2 mAb alone while other receptors (CD25, CD48, ICAM-I) (CD54) were not increased (26). Similar results were also observed in vivo. The present studies demonstrate that the expressions of the adhesion molecules CD11a/ CD18, CD44, CD45, CD48, CD54, and CD62L are all increased by anti-CD3 mAb but are not further affected by the addition of anti-CD2 mAb. Since the TCR/CD3 complex is the major receptor involved in antigen recognition and CD2 is one of the coreceptors engaged during initial antigen recognition, the upregulation of these other receptors in response to engagement of CD2 or CD3 may represent the consequence of initial activational signals which would promote T–APC or T–target interactions. Since these other adhesion molecules represent members of several distinct molecular families (i.e., selectin, integrin, immunoglobulin), these CD3- and CD2-derived transcriptional signals are likely to be broad. Nonetheless, there is some specificity since not all receptors demonstrate upmodulation. It is considered that anti-CD3 mAb induces immunosup-
/ m4726$8025
01-03-96 15:51:37
cial
pression by first activating T cells and then inducing anergy by overstimulation or exhaustion (50). The upmodulation of an array of cell surface receptors is one aspect of T cell activation. We speculate that another explanation for the immunosuppressive effects of antiCD3 mAb is inappropriate display of other coreceptors. Inability to engage these receptors after upregulation may facilitate subsequent anergy induction (51). Contrary to the upmodulation of most adhesion molecules, anti-CD2 mAb acted synergistically with antiCD3 mAb to downmodulate CD49d expression. CD49d is an adhesion receptor for VCAM-1, fibronectin, MAdCAM-1, and thrombospondin. It is involved in lymphocyte trafficking, homing, and a variety of inflammatory reactions (52–54). CD49d expression is important for T lymphocyte interactions with vascular endothelium and egress from vascular spaces into peripheral tissues and may also contribute to the control of entry into allogeneic tissues. Anti-CD3 significantly decreased CD49d expression, which may prevent effector lymphocytes from transmigrating across vascular endothelium and into allogeneic tissues. Anti-CD2 further decreased CD49d expression when combined with anti-CD3 and thus may further inhibit lymphocyte–endothelium interactions. The synergistic downmodulation of CD49d by the combination of anti-CD2 plus anti-CD3 may contribute to their synergistic effect in allograft prolongation. By implication anti-CD49d mAbs should prolong graft survival and we have obtained evidence that these mAbs prolong graft survival in our cardiac model (Lin and Bromberg, unpublished results). CD25 or IL-2R is important for T cell activation and proliferation through autocrine and paracrine mechanisms and anti-CD3 mAb alone significantly increased CD25 expression. Increased expression of CD25 indicates a state of T cell activation and is mechanistically related to the subsequent release of other cytokines induced by anti-CD3 mAb (20–22). The cytokines induced by anti-CD3 mAb can cause significant clinical side effects. Anti-CD2 mAb administered alone had no effect on CD25 expression, but it significantly decreased anti-CD3-driven expression. This is likely to impede further T cell activation and suppress cytokine release. This finding also likely accounts for the ability of anti-CD2 to ameliorate the anti-CD3-associated cytokine syndrome (24, 25). It will be important to understand the interaction of CD2- and CD3-related second messenger pathways and how receptor-mediated signals influence the expression of other receptors. Preliminary investigations demonstrate that anti-CD2 activates cAMP and protein kinase A activity which impede both IL-2 and IL-2R expression driven by antiCD3 (Lin and Bromberg, unpublished results). One important function for CD2 is as a signal transducer in T cell activation. CD2 can deliver activating signals through combinations of anti-CD2 mAbs, antigen, or mitogens (55, 56), resulting in changes in calcium currents, inositol phosphate turnover, and IL-2
AP: Cell Immuno
INTERACTION OF CD2 AND CD3
production (57–59). These pathways are also common to CD3-mediated T cell activation. Studies have shown a physical association between CD2 and CD3 (6– 9, 60, 61). More recent studies show that CD2 is physically associated with TCR z chain (9, 60) and that a CD2–TCR z complex is important for CD2 function. TCR z is also the portion of the CD3 complex responsible for TCR/CD3 signal transduction. This provides direct evidence that CD2 and CD3 not only interact with each other at the surface receptor level and but also share subsequent signal transduction pathways. The synergism in prolongation of allograft survival by antiCD2 plus anti-CD3 mAbs is probably related to the physical and functional association of the CD2 and CD3 receptors and the subsequent ability of CD2-mediated signaling to enhance or abrogate some aspects of CD3mediated signaling. Studies of anti-CD2-induced immunosuppression show that these mAbs alter immune competence in several different ways. We previously presented evidence that anti-CD2 causes receptor blockade in vitro during CTL–target interaction (62), induces anergy in vivo and in vitro to subsequent CD3associated signals (24, 25), and generates a population of TH2 suppressor cells in vivo (62). Importantly, antiCD2 does not cause T cell depletion. The results here delineate some additional mechanisms by which the mAbs may cause immunosuppression. A major alteration in the array of cell surface receptors induced by anti-CD2 plus anti-CD3 may alter the ability of T cells to traffic, respond to receptor engagement, interact with APC, and release lymphokines. ACKNOWLEDGMENTS We thank Drs. Altevogt, Bluestone, Ogawa, Wong, and Yagita for gifts of reagents and antibodies.
REFERENCES 1. Altevogt, P., Michaelis, M., and Kyewski, B., Eur. J. Immunol. 19, 1509, 1989. 2. Yagita, H., Okumura, K., and Nakauchi, H., J. Immunol. 140, 1321, 1988. 3. Dang, N. H., Torimoto, Y., Sugita, K., Daley, J. F., Schow, P., Prado, C., Schlossman, S. F., and Morimoto, C., J. Immunol. 145, 3963, 1990. 4. Cerdan, C., Martin, Y., and Brailly, H., J. Immunol. 146, 560, 1991. 5. Samelson, L. E., Fletcher, M. C., Ledbetter, J. A., and June, C. H., J. Immunol. 145, 2448, 1990. 6. Bockenstedt, L. K., Goldsmith, M. A., Dustin, M., Olive, D., Springer, T. A., and Weiss, A., J. Immunol. 141, 1904. 7. Holter, W., Majdic, O., Stockinger, H., Howard, B. H., and Knapp, W., J. Immunol. 140, 1043, 1988. 8. Kvanta, A., Gerwins, P., Van der Ploeg, I., Nordstedt, C., Jondal, M., and Fredholm, B. B., Eur. J. Pharmacol. 189, 363, 1990. 9. Gassman, M., Amrein, K. E., Flint, N. A., Schraven, B., and Burn, P., Eur. J. Immunol. 24, 139, 1994. 10. Moingeon, P. E., Lucich, J. L., and Stebbins, C. C., Eur. J. Immunol. 21, 605, 1991.
/ m4726$8025
01-03-96 15:51:37
cial
257
11. Ohno, H., Nakamura, T., Yagita, H., Okumura, K., Taniguchi, M., and Saito, T., J. Immunol. 147, 2100, 1991. 12. Bromberg, J. S., Chavin, K. D., Altevogt, P., Kyewski, B. A., Guckel, B., Naji, A., and Barker, C. F., Transplantation 51, 219, 1991. 13. Guckel, B., Berek, C., Lutz, M., Altevogt, P., Schirrmacher, V., and Kyewski, B. A., J. Exp. Med. 174, 957, 1991. 14. Jonker, M., Malissen, B., and Von Vreeswijk, W., Transplant. Proc. 35, 521, 1983. 15. Chavin, K. D., Lau, H. T., and Bromberg, J. S., Transplant 54, 286, 1992. 16. Vigeral, P., Chkoff, N., Chatenoud, L., Campos, H., Lacombe, M., Droz, D., Goldstein, G., Bach, J. F., and Kreis, H., Transplantation 41, 730, 1986. 17. Ortho Multicenter Transplant Study Group, N. Engl. J. Med. 313, 337, 1985. 18. Cosimi, A. B., Burton, R. C., Colvin, R. B., Goldstein, G., Delmonico, F. L., La, Q. M., Tolkoff, R. N., Rubin, R. H., Herrin, J. T., and Russell, P. S., Transplantation 32, 535, 1981. 19. Alberola-Ila, J., Places, L., de la Calle, O., Romero, M., Yague, J., Gallart, T., Vives, J., and Lozano, F., J. Immunol. 146, 1085, 1991. 20. Hirsch, R., Eckhaus, M., Auchincloss, H. J., Sachs, D. H., and Bluestone, J. A., J. Immunol. 140, 3766, 1988. 21. Alegre, M., Vandenabeele, P., Flamand, V., Moser, M., Leo, O., Abramowicz, D., Urbain, J., Fiers, W., and Goldman, M., Eur. J. Immunol. 20, 707, 1990. 22. Ferran, C., Sheehan, K., Dy, M., Schreiber, R., Merite, S., Landais, P., Noel, L. H., Grau, G., Bluestone, J., and Bach, J. F., Eur. J. Immunol. 20, 509, 1990. 23. Ferran, C., Dy, M., Merite, S., Sheehan, K., Schreiber, R., Leboulenger, F., Landais, P., Bluestone, J., Bach, J. F., and Chatenoud, L., Transplantation 50, 642, 1990. 24. Chavin, K. D., Qin, L., Lin, J., Yagita, H., and Bromberg, J. S., J. Immunol. 151, 7249, 1993. 25. Chavin, K. D., Qin, L., Lin, J., Woodward, J. E., Baliga, P., and Bromberg, J. S., Ann. Surg. 218, 492, 1993. 26. Lin, J., Yon, R. W., Chavin, K. D., Qin, L., Woodward, J., Ding, Y., Yagita, H., and Bromberg, J. S., Transplantation 59, 1162, 1995. 27. Altevogt, P., Kohl, U., Von Hoegen, P., Lang, E., and Schirrmacher, V., Eur. J. Immunol. 19, 341, 1989. 28. Leo, O., Foo, M., Sachs, D. H., Samelson, L. E., and Bluestone, J. A., Proc. Natl. Acad. Sci. USA 84, 1374, 1987. 29. Kato, K., Koyanagi, M., Okada, H., Takanashi, T., Wong, Y. W., Williams, A. F., Okumura, K., and Yagita, H., J. Exp. Med. 176, 1241, 1992. 30. Wilde, D. B., Marrack, P., Kappler, J., Dialynas, D. P., and Fitch, F. W., J. Immunol. 131, 2178, 1983. 31. Ledbetter, J. A., and Herzenberg, L. A., Immunol. Rev. 47, 63, 1979. 32. Furth, M. E., Davis, L. J., Fleurdelys, B., and Scolnick, E. M., J. Virol. 43, 294, 1982. 33. Gustavo, O. R., Robb, R. J., Shevach, E. M., and Malek T. R., J. Immunol. 133, 1970, 1984. 34. Miyake, K., Medina, K. L., Hayashi, S., Ono, S., Harnaoka, T., and Kincade, P. W., J. Exp. Med. 171, 477, 1990. 35. Springer, T., Galfre, G., Secher, D. S., and Milstein, C., Eur. J. Immunol. 8, 539, 1978. 36. Holzmann, B., McIntyre, B. W., and Weissman, I. L., Cell 56, 37, 1989. 37. Holzmann, B., and Weissman, I. L., EMBO J. 8, 1735, 1989. 38. Prieto, J., Takei, F., Gendelman, R., Christenson, B., Biberfeld, P., and Patarroyo, M., Eur. J. Immunol. 19, 1551, 1989.
AP: Cell Immuno
258
LIN ET AL.
39. Sanchez-Madrid, F., Simon, P., Thompson, S., and Springer, T. A., J. Exp. Med. 158, 586. 40. Kubo, R. T., Born, W., Kappler, J. W., Marrac, P., and Pigeon, M., J. Immunol. 142, 2736, 1989. 41. Holzmann, B., McIntyre, B. W., and Weissman, I. L., Cell 56, 37, 1989. 42. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3, 645, 1973. 43. Gold, D. P., Clevers, H., and Alarcon, B., Proc. Natl. Acad Sci. USA 84, 7649, 1987. 44. Abraham, D. J., Bou-Gharios, G., Beauchamp, J. R., PlaterZyberk, Maini, R. N., and Olsen, I., J. Leukocyte Biol. 49, 329, 1991. 45. Funaro, A., Spagnoli, G. C., Ausiello, C. M., Alessio, M., Roggero, S., Delia, D., Zaccolo, M., and Malavasi, F., J. Immunol. 145, 2390, 1990. 46. Conrad, P., Rothman, B. L., Kelley, K. A., and Blue, M. L., J. Immunol. 149, 1833, 1992. 47. Blue, M. L., Conrad, P., Davis, G., and Kelley, K. A., Cell Immunol. 138, 238, 1991. 48. Altin, J. G., Papler, E. B., and Parish, C. R., Eur. J. Immunol. 24, 450, 1994. 49. Dianzani, U., Redoglia, V., Malavasi, F., Bragardo, M. , Pileri, A., Janeway, C. A., and Bottomly, K., Eur. J. Immunol. 22, 365, 1992. 50. Anasett, C., Tan, P., Hansen, J. A., and Martin, P. J., J. Exp. Med. 172, 1691, 1990.
/ m4726$8025
01-03-96 15:51:37
cial
51. Lin, H., Bolling, S. F., Linsley, P. S., Wei, R.-Q., Gordon, D., Thompson, C. B., and Turka, L. A. J. Exp. Med. 178, 1801, 1993. 52. Paul, L. C., Davidoff, A., Benediktsson, H., and Issekutz, T. B., Transplant 55, 1196, 1993. 53. Orosz, C. G., Ohye, R. G., Pelletier, R. P., VanBuskirk, A. M., Huang, E., Morgan, C., Kincade, P. W., and Ferguson, R. M., Transplant 56, 453, 1993. 54. Turunen, J. P., Mattila, P., Halttunen, J., Hayry, P., and Renkonen, R., Transplant 54, 1053, 1992. 55. Moingeon, P., Chang, H. C., Sayre, P. H., Clayton, L. K., Alcover, A., Gardner, P., and Reinherz, E. L., Immunol. Rev. 111, 111, 1989. 56. Bierer, B. E., and Burakoff, S. J., J. Immunol. Rev. 111, 267, 1989. 57. Van Kooyk, Y., Van de Weil-Van Kemenade, P., Weder, P., Kuijpers, T. W., and Figdor, C. G., Nature 342, 811, 1989. 58. Monostori, E., Desai, D., Brown, M. H., Cantrell, D. A., and Crumpton, M. J., J. Immunol. 144, 1010, 1990. 59. Lorre, K., van Verwilghen, D. J., Baroja, M. J., and Ceuppens, J. L., J. Immunol. 144, 4681, 1990. 60. Moingeon, P., Lucich, J. L., McConkey, D. J., Letourneur, F., Malissen, B., Kochan, J., Chang, H. C., Rodewald, H. R., and Reinherz, E. L., Proc. Natl. Acad. Sci. USA 89, 1492, 1992. 61. Brown, M. H., Cantrell, D. A., Brattsand, G., Crumpton, M. J., and Gullberg, M., Nature 339, 551, 1989. 62. Chavin, K. D., Qin, L., Yon, R., Lin, J., Yagita, H., and Bromberg, J. S., J. Immunol. 152, 3729, 1994.
AP: Cell Immuno