CELLULAR
IMMUNOLOGY
14,
31-41 (1992)
Monoclonal Antibodies to Common Epitopes of the Human a/(3 T-Cell Receptor Preferentially Activate CD45RA+ T-Cells’ REINHARD SCHWINZER,’HANS JORGENSCHLITT, AND KURT WONIGEIT~ Klinik fir Abdominal- und Transplantationschirurgie, Medizinische Hochschule Hannover, Hannover, Federal Republic of Germany Received June 3, 1991; accepted October 16, 1991 The murine monoclonal antibody BMA 031 (IgG2b) is directed to a monomorphic epitope on the human a/@T-cell receptor. In contrast to anti-CD3 antibodies of the IgG2b isotype, BMA 03 1 is able to induce a proliferative responsein T-cells from IgG2b low responders.This response occurs independently of cross-linking conditions indicating that the mode of activation differs from stimulation by the anti-CD3 antibody OKT3 (IgG2a) which strictly dependson cross-linking conditions. To further characterize the stimulatory potential of the two antibodies we studied the lymphocyte subsets responsive to stimulation by BMA 03 1 and OKT3. In CD45RA+ cells both antibodies exhibited similar effects. They induced weak expression of the 55kDa chain of the interleukin-2 receptor (CD25), virtually no interleukin-2 secretion, but neverthelessstrong proliferation. In CD45RO+ cells OKT3 and BMA 031 showed markedly different effects. OKT3 stimulated strong CD25 expression, strong interleukin-2 production, and marked proliferation. In contrast, CD45RO+ cells stimulated by BMA 031 showed only weak CD25 expression but neither interleukin-2 production nor proliferation. Thesedata suggestthat CD45RA+ and CD45RO+ cells differ in their capability to produce interleukin-2 upon stimulation via the CD3/T-cell receptor complex and also in the requirement for interleukin-2 to mount a proliferative response. The differential effect of OKT3 and BMA 03 1 in CD45RO+ cells probably results from the failure of BMA 031 to trigger interleukin-2 production which may be a consequence of its inability to induce CD3/T-cell receptor cross-linking in IgG2b low responders.BMA 03 1 is therefore a useful tool for the selective activation of CD45RA+ cells in these individuals. o 1992 Academic press, IX.
INTRODUCTION On the surface of mature T-cells the CD3 complex is expressedin close association with the T-cell receptor (TcR)~ (1, 2). Monoclonal antibodies (mAb) to the TcR and to CD3 are able to activate T-cells (3, 4). Since T-cell stimulation by soluble mAb usually requires cross-linking mediated by Fc receptors on accessorycells the stimulatory potential is dependent on the antibody isotype (5, 6). Anti-CD3 mAb of the ’ This work was supported by the Deutsche Forschungsgemeinschaft:“Forschergruppe Organtransplantation” (Pi 48/l 1; Dl/D4), and by the Jung Stiftung fur Wissenschaftund Forschung. ’ Present address: Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Denver, CO 80206. 3 To whom correspondenceshould be addressedat Klinik fur Abdominal- und Transplantationschirurgie, Medizinische Hochschule Hannover, Konstanty-Gutschow-Str. 8, D-3000 Hannover 61, FRG. 4 Abbreviations used: IL-2, interleukin-2; IL-2R, interleukin-2 receptor; mAb, monoclonal antibody(ies); PBMC, peripheral blood mononuclear cells; TcR, T-cell receptor. 31 000%8749192 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
32
SCHWINZER, SCHLITT, AND WONIGEIT
IgG2a isotype are stimulatory for all individuals whereas IgG2b mAb of similar specificity stimulate only T-cells of very few so-called IgG2b high responders. The failure to thesemAb to stimulate T-cells of the vast majority of individuals is usually explained by insufficient binding of IgG2b by Fc receptors. An important role of the fine specificity of a stimulating mAb is suggestedby data from our laboratory showing that mAb directed to the constant portion of the human cz//3TcR differ from typical anti-CD3 mAb in their stimulatory properties. For example, lymphocytes from IgG2b low responders which fail to respond to anti-CD3 mAb of the IgG2b &type mount a proliferative response to the IgG2b mAb BMA 031 that is directed to the cu/pTcR. The proliferative responsetriggered by BMA 03 1 does not depend on cross-linking conditions. Whereas the anti-CD3 mAb OKT3 induces interleukin-2 (IL2) secretion, expression of high density interleukin-2 receptors (IG2R) and cytotoxic activity in bulk peripheral blood mononuclear cell (PBMC) cultures, PBMC stimulated by BMA 03 1 expressonly low levels of CD25 but do neither secrete IL-2 nor develop cytotoxic activity. Taken together these data suggestthat OKT3 and BMA 03 1 induce different modes of T-cell activation (7, 8). The differential responseof bulk PBMC cultures to OKT3 and BMA 03 1 could be explained in two different ways. If an individual T-cell has the capacity to respond to stimulation by using different signaling pathways, OKT3 and BMA 03 1 might trigger different signaling mechanisms and the subsequent activation of different pathways. On the other hand, if a given T-cell is inherently programmed to elicit a certain response after stimulation regardless of the type of stimulus, the different responses in bulk cultures could indicate that OKT3 and BMA 031 activate discrete T-cell subsets.To address this issue, we have analyzed the stimulatory capacities of OKT3 and BMA 031 in different T-cell subsets.Clear-cut differences were found in the response patterns of CD45RA+ and CD45RO+ T-cells which are regarded to represent naive and memory T-cells, respectively (9-l 1). MATERIALS AND METHODS Antibodies For the lymphocyte activation experiments mAb OKT3 (anti-CD3, IgG2a), UCHTl (anti-CD3, IgG2b) (12) BMA 03 1 (ant&//3 TcR, IgG2b), and BMA 032 (anti-a/p TcR, IgG2b) ( 13, 14) were used. Antibodies of the BMA serieswere provided by Dr. R. Kurrle (Behringwerke, Marburg, FRG). OKT3 was provided by Cilag GmbH (Alsbath-Hgihnlein, FRG). UCHTl was kindly provided by Dr. P. C. L. Beverley (London, UK). mAb BMA 040 (antiCD4, Behringwerke Ag, Marburg, FRG), OKT8 (antiCD8, American tissue culture collection, Rockville, MD), UCHLl (anti-CD45R0, donated by Dr. P. C. L. Beverley, London, UK) (15), and various antiCD45R/ CD45RA mAb obtained from the Fourth International Workshop on leucocyte differentiation antigens (16) were used to isolate T-cell subsets by means of negative selection. The directly labeled mAb FITC-Leu2 (anti-CD8), FITC-Leu3 (anti-CD4), FITC-Leu 18 (antiCD45RA), PE-anti-Tat (antiCD25), and FITCXeuM3 (anti-CD 14) used for flow cytometry analysis were purchased from Becton-Dickinson (Heidelberg, FRG). FITC-UCHLl (antiCD45RO) was obtained from Dako (Hamburg, FRG). Cell Preparations In the described experiments only lymphocytes from IgG2b low responders as determined by their failure to respond to stimulation with mAb UCHTl (IgG2b) and
ENHANCED
CD45RA+
CELL
REACTIVITY
33
to low concentrations (15 rig/ml) of BMA 03 1 were analyzed. Two types of cell preparations were used. In a first set of experiments PBMC were isolated from heparinized peripheral blood by Ficoll gradient centrifugation and depleted of various T-cell subsets. No further depletion of monocytes/accessory cells was performed. In another set of experiments a population of small resting T-cells was prepared by E-rosetting. Usually >95% of E+ cells were CD3+ and 98-99% reacted with anti-CD2 mAb. Irradiated (50 Gy) E- cells were used as a source of monocytes/accessory cells. Phenotypic analysis of this fraction revealed that it usually contained 45-60% of monocytes (CD14+). Isolation of T-Cell Subsets Magnetic immunoselection was performed to negatively isolate various T-cell subsets.PBMC or E+ cells were incubated with saturating dosesof BMA 040 (anti-CD4), OKT8 (anti-CDS), UCHLl (anti-CD45RO), or anti-CD45RA mAb for 30 min at 4°C. Cells were washed twice, mixed with sheep anti-mouse Ig-coated beads (Dynabeads, Dianova, Hamburg, FRG), and incubated for 60 min at 4°C. Cells reactive with the primary mAb and thus bound by the magnetic beads were removed with a magnet. Nonmagnetized cells were collected, washed twice, and reanalyzed. Usually lessthan 6% of cells from the reciprocal subsetwere detectedin thesepreparations as determined by flow cytometry analysis. Proliferation Assays Unseparated PBMC or negatively selectedcell preparations were cultured in roundbottomed microtiter plates (Nunc, Roskilde, Dk) at 5 X 105/ml ( 1 X 10’ cells/well) in RPM1 1640 medium supplemented with 10% fetal calf serum, penicillin ( 100 U/ ml), and streptomycin (100 &ml). In some experiments 20 U/ml of purified human IL-2 (Lymphocult T-HP, Biotest, Frankfurt, FRG) was added to the cultures. All antiCD3/TcR mAb were used at concentrations determined for optimal proliferation of unseparated PBMC (OKT3, 15 rig/ml; BMA 03 1, BMA 032, 1.5 pg/ml; UCHT 1, 1: 100 dilution of culture supernatant). If T-cells were used, stimulation was performed in the presenceof 2.5 X lo4 irradiated (50 Gy) autologous E- cells per well as a source of monocytes/accessory cells. Proliferation was assessed by incorporation of [3H]thymidine ([3H]-dThd) added 16 hr prior to harvest. Induction of IL-2 Secretion and IL-2R Expression To induce IL-2 secretion, E+ cells were prepared and further divided into CD45RA+ and CD45ROf cells by means of negative selection. CD45RAf or CD45RO+ T-cells ( 10 X lO’/ml; 2 X 105/well) were stimulated in the presence of 2.5 X lo4 irradiated autologous non-T-cells for 24 hr with mAb OKT3 or BMA 03 1. Measurement of IL2 in the supematant of these cultures was performed using the IL-2-dependent rat Tcell line G2 produced in our laboratory. G2 cells ( 1 X 104)were cultivated for 48 hr with various concentrations of supematant from cell cultures. Defined serial dilutions of purified IL-2 were included in each experiment for control. The proliferation of G2 cells was determined by [3H]-dThd incorporation for another 16 hr. Proliferation induced by the supematants was then compared with the IL-2 titration curve to calculate units per milliliter. To induce IG2R expression, PBMC or negatively isolated T-cell subsets(5 X 105/ml) were cultured in a final volume of 200 ~1in round-bottomed
34
SCHWINZER, SCHLITT, AND WONIGEIT
microtiter plates with medium, OKT3 ( 15 rig/ml), or BMA 03 1 ( 1.5 pg/ml). After 48 hr cells were collected, washed, and prepared for flow cytometry analysis.
Flow Cytometry Analysis Reanalysis of negatively isolated T-cell subsetswas performed by direct immunofluorescence using FITC-conjugated mAb. Residual binding of the mAb used for separation which may block reactivity of the directly labeled mAb was always controlled by a FITC-conjugated goat anti-mouse IgG but was never observed. Cells were washed twice with staining buffer (PBS containing 0.5% BSA and 0.1% NaN3) and incubated for 30 min at 4°C with saturating concentrations of mAb. After two additional washes analysis of the stained cells was performed on a FACStar flow cytometer (BectonDickinson, Mountain View, CA). To detect the 55-kDa chain of the IL-2R separately on CD45RA+ and CD45RO+ cells, two-color flow cytometry analyseswere performed. Cultured cells were stained with the mAb combinations FITCLeu18 (antiCD45RA)/ PE-anti-Tat (antXD25) or FITC-UCHLl (anti-CD45RO)/PE-anti-Tat (antXD25). Appropriate gates were set to include only activated lymphoblasts as identified by increased forward and side scatter intensity. Results are shown as contour plots where the FITC fluorescence intensity is plotted versus the PE fluorescence intensity. RESULTS
Enhanced Proliferative Responseof CD45RA+ Cells to Stimulation with mAb BMA 031 The previously observeddifferencesin the responsepatterns of PBMC to stimulation with OKT3 or BMA 03 1 (7, 8) raised the possibility that different T-cell populations are preferentially activated by the two mAb. To analyze this question, PBMC and negatively isolated T-cell subsets were stimulated with the two mAb. In CD4+ and CD8 cells OKT3 and BMA 031 exhibited similar stimulatory capacities (data not shown). In contrast, cell preparations depleted either of CD45ROf or CD45RA+ cells exhibited a differential response pattern. Both were induced to proliferate by binding of OKT3 (Fig. 1). The CD45RA+ fraction (CD45RO depleted) showed an enhanced responseto BMA 03 1 stimulation whereas CD45RO+ cells (CD45RA depleted) were nearly unresponsive to BMA 03 1-triggered stimulation. This unusual reactivity pattern was also found using another anti-TcR mAb (BMA 032). Since both BMA 031 and BMA 032 are of the IgG2b isotype, control experiments were performed using the IgG2b anti-CD3 mAb UCHT 1. As expected from the IgG2b nonresponder status of the tested blood donors, this mAb triggered a significant proliferative response in neither CD45RO+ nor CD45RAf cells (data not shown) indicating that the specificity of the tested anti-TcR mAb contributes to their unusual stimulatory activity. The experiments summarized in Fig. 1 were performed using mAb concentrations inducing a maximal proliferative response in PBMC ( 15 r&ml OKT3 and 1.5 &ml BMA 03 1). To exclude the possibility that the differential pattern of OKT3- and BMA 031-induced activation observed in CD45RA+ and CD45RO+ cells may result from the different antibody amounts, various concentrations of OKT3 and BMA 03 1 were tested. In line with earlier observations (7) effective stimulation by BMA 03 1 required much higher mAb concentrations as compared with OKT3 (Table 1). Although the strong responding CD45RAf cells showed a small proliferative response when stim-
35
ENHANCED CD45RA+ CELL REACTIVITY
Cells PBMC
CD45RO depl.
CD45RA depl.
0
50
100
150
200
% relative
260
300
350
400
response
FIG. 1. Response patterns of cell population depleted of CD45RA+ and CD45ROCcells to stimulation with OKT3 and BMA 031. Unseparated PBMC, CD45RO- (CD45RA+), and CD45RA (CD45RO+) cell populations were stimulated with 15 rig/ml OKT3 (stippled bars) and 1.5pg/ml BMA 03 1 (solid bars). Since proliferation induced by OKT3 wasgenerally higher than that induced by BMA 03 1 the proliferative responses of the various cell populations are expressedasthe percentageof the responseobservedin PBMC to stimulation with the respective mAb. The results represent the means f SD obtained in four (CD45RO depleted) and five (CD45RA depleted) individual experiments.
mated with 15 rig/ml, a clear-cut responsecould only be achieved using high dosesof BMA 03 1 (1.5 pg/ml). CD45RO+ cells were nearly unresponsive to BMA 03 1 stimulation irrespective of the mAb concentration. In all cell preparations tested, OKT3 reached a maximal proliferative response already at 15 rig/ml. This result indicates that the differential response of CD45RAf and CD45RO+ cells to anti-CD3 and antiTcR stimulation did not result from different mAb concentrations used for stimulation. Monocytes expressCD45RO in high density whereas only low amounts of CD45RA molecules can be detected on their cell surface. Therefore, it is likely that negatively
TABLE 1 Proliferative Responsesof PBMC, CD45RA+, and CD45ROf Cells to Stimulation with Different Concentrations of OKT3 and BMA 03 1” BMA031”
OKT3 b Experiment
Cells
15
150
1500
15
150
1500
1
PBMC CD45RA+ PBMC CD45RO+
60.1 19.9 40.2 30.4
51.5 24.6 nd’ nd
50.0 22.7 32.8 22.6
1.8 4.6 2.8 1.8
2.7 6.8 8.9 1.6
9.1 29.8 11.2 1.9
2
’ Unseparated PBMC (1 X 10’) or PBMC enriched for CD45RA+ or CD45RO+ cells by depletion of CD45RO+and CD45RA+ cells, respectively, were stimulated for 3 days with OKT3 or BMA 03 1. Proliferation was determined during the last 16 hr of culture by [)H]dThd incorporation. Results are expressedas cpm X lo-’ of triplicate cultures corrected by subtraction of media controls. b rig/ml. ‘Not determined.
36
SCHWINZER, SCHLITT, AND WONIGEIT
selectedCD45RA+ and CD45RO+ PBMC contain a different proportion of monocytes/ accessorycells. To exclude the possibility that the differential response of CD45RA+ and CD45RO+ cells to anti-CD3 and anti-TcR stimulation results from differences in purity, a population of small resting T-cells (E-rosette positive) was prepared and further divided into CD45RA* and CD45RO+ cells. These cells were stimulated in the presence of equal numbers of autologous irradiated non-T-cells as a source of monocytes/accessory cells. In accordance with the experiments using PBMC (Fig. 1) a differential effect of OKT3 and BMA 03 1 was also observedin these cell preparations. OKT3 was able to induce a proliferative response in both CD45RA+ and CD45RO+ T-cells (Table 2). Although BMA 03 1 triggered a proliferative response in CD45RA+ T-cells this mAb was nearly unable to induce proliferation in CD45ROf T-cells. A proliferative response to stimulation with the recall antigen tetanus-toxoid was only observed in CD45RO+ T-cells whereas the CD45RA subset was completely unresponsive. The differential tetanus-toxoid responsesare in line with published data ( 15, 17) and indicate that CD45RA+ and CD45RO+ T-cells representthe naive and the memory T-cell pool, respectively.
Induction of IL-2 Responsivenessin BMA 031-Stimulated CD45RO’ T-Cells The low responsivenessof CD45RO+ T-cells to stimulation with BMA 03 1 raised the question whether these cells cannot be activated at all by BMA 031 or whether they achieve only an early state of activation. Since the induction of IL-2 responsiveness is regarded to be an early step in T-cell activation (18), we analyzed the effect of exogeneous IL-2 on BMA 031-induced activation of CD45RO+ cells. As shown in Table 3, BMA 03 1 was very effective in triggering proliferation of CD45RO+ T-cell if IL-2 was provided exogeneously. These data indicate that BMA 03 1 is able to activate CD45RO+ T-cells. This activation, however, seems to be incomplete since it only results in the induction of functional IL-2R but not in proliferation.
D#erences in IL-2 Secretion betweenCD45RA’ and CD45RO’ T-Cells Stimulated via the CD3/TcR Complex The observation that BMA 03 1 triggered proliferation of CD45ROt T-cells strictly depended on exogeneous IL-2 whereas CD45RA T-cells proliferated without the adTABLE 2 Proliferative Responsesof T-Cell SubsetsNegatively Selected for the Expression of CD45RA or CD45RO to Stimulation with CD3/TcR mAb and Tetanus-Toxoid” Proliferative Response(cpm X 10m3)to Experiment
Cells
Medium
OKT3
BMA 03 1
TT
1 2 3
CD45RA+ CD45RO+ CD45RA+ CD45RO+
0.6 0.7 0.9 0.6
29.5 39.6 17.8 14.4
28.0 2.1 25.3 3.2
0.9 8.7 ndb nd
’ E” cells were further divided into CD45RA+ and CD45RO+ T-cells by means of negative selection and stimulated ( 1 X IO5cells/well) with OKT3 (15 rig/ml), BMA 03 1 (1.5 &ml), or tetanus-toxoid (1:300 v/ v). Irradiated autologous E- cells (2.5 X 10“) were added to each well. Results are expressedas cpm X IO-’ of triplicate cultures. b Not determined.
37
ENHANCED CD45RA+ CELL REACTIVITY TABLE 3 Induction of IL-2 Responsivenessin CD45ROf Cells Stimulated by BMA 03 1”
Proliferative Response(cpm X 10d3)to Experiment
Cells
IL-2
Medium
OKT3
BMA 03 1
1
PBMC CD45RO+
2
E+ CD45RO+
3
E+ CD45RO+
+ + +
0.7 10.0 1.7 11.2 0.1 7.5
30.7 83.4 33.8 56.1 10.0 31.8
1.8 50.0 4.5 31.9 1.3 20.0
n PBMC (I X IO’) or T-cells (E+) enriched for CD45RO+ cells by depletion of the CD45RA+ population were stimulated for 3 days in the absence or in the presence of 20 U/ml IL-2 with OKT3 (15 rig/ml) or BMA 031 (1.5 pg/ml). Proliferation was determined during the last 16 hr of culture by [3H]-dThd incorporation. Results are expressedas cpm X IO-’ of triplicate cultures.
dition of IL-2 could mean that BMA 03 1 is able to induce IL-2 secretion in CD45RA+ but not in CD45ROt T-cells. On the other hand, it could also be possiblethat CD45RA+ and CD45RO+ T-celis have different requirements for IL-2 to mount a proliferative response.To distinguish between these alternatives, the amount of IL-2 was measured in the supematant of CD45RA+ and CD45ROf T-cells stimulated with OKT3 or BMA 03 1. After stimulation of purified CD45RAf T-cells with OKT3 for 24 hr, only the highest concentration of supematant induced a marginal proliferative responsein the IL-2-dependent rat T-cell line G2 indicating that there is only a small amount (co.5 U/ml) of IL-2 present in these cultures (Fig. 2). IL-2 was completely undetectable in the supematant of BMA 031~stimulated CD45RA+ cells. In the supematant of CD45RO+ T-cells stimulated with OKT3, IL-2 could readily be detected (4 U/ml). Again, stimulation by mAb BMA 031 did not result in measurable IL-2 release. Diferential IL-2 Receptor Expression Pattern after OKT3- and BMA 031-Induced Activation The induction of functional IL-2R by BMA 031 stimulation could be confirmed by flow cytometry analysis revealing low levels of the 55-kDa chain of the IG2R (CD25) in negatively isolated CD45RA+ and CD45RO+ T-cells. In contrast, OKT3 induced low levels of CD25 in CD45RA+ cells but high levels in CD45RO+ cells (data not shown). To exclude separation artifacts as a reason for these patterns, two-color analyses were performed in unseparated PBMC. In line with the experiments using isolated subsets there was a marked differential expression of CD25 in CD45RA+ (CD45RO-) and CD45RA- (CD45RO+) cells as a result of anti-CD3 (OKT3) stimulation (Fig. 3, upper panels). Whereasonly a small subsetofthe CD45RAf (CD45RO-) population expressed CD25, nearly the whole population of CD45RA- (CD45RO’) cells was CD25+. In addition, the concentration of CD25 expressedon the cell surface differed between CD45RA+ (CD45RO-) and CD45RX (CD45ROf) cells. CD45RA+ (CD45RO-) cells expressed much lower levels of CD25 (320 arbitrary fluorescence units) as compared to CD45RA- (CD45RO+) cells (670 fluorescence units). A quite different pattern was observed in anti-TcR (BMA 031)-stimulated cultures. In these
38
SCHWINZER, SCHLITT, AND WONIGEIT CD45RA+
30
-+
medium
-
OK13
-
EMA 031
26 m
“’
I +--, - ’ J
I:2
t:,
I:8
I:2
I:4
I:8
I:16
dilution
FIG. 2. Proliferative responsesof the IL-2dependent rat T-cell line G2 to various dilutions of culture supematant from stimulated CD45RA+ and CD45RO+ T-cells. E+ T-cells were divided into the subsets CD45RA and CD45ROand cultured in the presenceof irradiated autologous non-T-o&.. Cells were stimulated for 24 hr with OKT3 (15 r&ml) or BMA 03 1 (1.5 &ml). The supematants were collected and titrated on G2 cells (1 X 104/well).Proliferation of G2 cells was determined after 2 days. CD45RA+ T-cells proliferated to stimulation with both OKT3 and BMA 03 1 (53,336 and 38,916 cpm, respectively). The proliferative responsesof CD45RO+ cells were 39,578 cpm (OKT3) and 2127 cpm (BMA 031).
cultures, the expression of CD25 on CD45RA+ (CD45RO-) cells was comparable to that found after OKT3 stimulation. However, in contrast, the characteristic high expression density of CD25 found in OKT3-stimulated CD45RA- (CD45RO+) cells was never observed. In BMA 03 l-stimulated cultures also the CD45RA (CD45ROf) cells expressed only low levels of CD25. According to the largely nonoverlapping expression of CD45RA and CD45R0, reciprocal patterns were observed when the cultured cells were stained with the mAb combination anti-CD45RO/anti-CD25 (Fig. 3, lower panels). DISCUSSION This paper elucidates the cellular basis of the differential stimulating capacity of the mAb BMA 03 1 and OKT3. It demonstrates that OKT3 is able to stimulate proliferation of both CD45RAt and CD45RO+ cells whereas BMA 031 predominantly stimulates CD45RA+ cells. It has previously been shown that CD45RA+ and CD45RO+ cells which are regarded to represent naive and memory T-cells, respectively, have different stimulatory requirements to mount a proliferative response (19-22). Our results demonstrate that under the culture conditions used BMA 031 fulfills the requirements to induce proliferation in CD45RA+ but not in CD45RO+ cells. The comparison of the activation events induced in CD45RA+ and CD45RO+ cells revealed that in both populations CD25 is induced at low levels (Fig. 3) but that no measurableIL-2 can be demonstrated in the culture supematant (Fig. 2). Since addition of exogeneous IL-2 led to a proliferative response of CD45RO+ cells as well (Table 3) it can be concluded that a major causefor the differential responseof CD45RAf and CD45RO+ cells upon stimulation
ENHANCED CD45RA+ CELL REACTIVITY medium
CD45RA-FITC
-
CD46RO-FITC
-
anti-CD3 (OKT3)
39
anti-TcR (BMA 031)
FIG. 3. Different patterns of IL-2R expression in CD45RAf and CD45RO+cells stimulated by OKT3 and BMA 03 1. PBMC were cultured for 2 days with medium, OKT3 (15 @ml), or BMA 03 1 (1.5 &ml). The cells were harvested and stained with FITC-Let118 (CD45RA, upper panels) or FITC-UCHLI (CD45R0, lower panels) in combination with PE-anti-Tat (antiCD25). Data are displayed as contour plots where the FfTC fluorescenceintensity (x-axis) is plotted versus the PE fluorescenceintensity (y-axis) on a logarithmic scale. On the basis of control samples the plots were divided into quadrants: (1) Cells with PE fluorescence only; (2) cells with PE and FITC fluorescence;(3) unstained cells; and (4) cells with FITC fluorescenceonly. Analysis was selectively performed on blastoid cells as determined by an increased forward and side scatter intensity. Note that OKT3 induced high density expression of CD25 in CD45RA (CD45RO+) and low density in CD45RA+ (CD45RO-) cells whereas after BMA 03 1 stimulation both subsetsexpressed CD25 only at low levels.
with BMA 031 is a difference in the requirement for IL-2. This conclusion is also supported by the finding that isolated CD45RA+ cells responded to OKT3 stimulation by marked proliferation although no or only very little production of IL-2 could be demonstrated. In addition to the different requirement for IL-2 the different potential to produce this lymphokine appears to be another inherent functional discrepancy between CD45RA+ and CD45RO+ cells at least when antiCD3/TcR mAb are used for stimulation. This follows from the finding that OKT3, which was ineffective in eliciting IL-2 production in CD45RA+ cells, induced strong IL-2 secretion in CD45RO+ cells (Fig. 2). Since the expression of CD25 was also stronger on CD45RO+ cells (Fig. 3) it can be assumedthat there was substantial IL-2 consumption suggestingan even larger difference in the production rate than that indicated by the IL-2 measurement in the culture supematant. Although the response of CD45RA+ cells to OKT3 and BMA 03 1 was remarkably similar it is likely that different activation mechanisms are involved in this situation. Since the activation of unseparated PBMC by OKT3 is strictly dependent on crosslinking conditions this might also apply for isolated CD45RA+ cells. This conclusion is supported by the finding that the IgG2b variant of the anti-CD3 mAb UCHTl was not able to stimulate CD45RA+ cells of IgG2b low responder individuals lacking the
40
SCHWINZER, SCHLITT, AND WONIGEIT
appropriate Fc receptors on their accessorycells (data not shown). In contrast, BMA 03 1 and BMA 032 effectively stimulated CD45RA+ cells although both mAb are also of the IgG2b isotype. As previously postulated (7, 8) this points to a major effect of the epitope on the CD3/TcR complex to which the stimulating mAb is directed. In CD45RA+ cells only certain epitopes appear to be able to induce proliferation in the absenceof cross-linking. In contrast to CD45RAf cells, the CD45RO+ population required sufficient concentrations of IL-2 to mount a proliferative response. Thus, the differential effects of OKT3 and BMA 03 1 in CD45RO+ cells in all likelihood reflect a differential capacity of the two mAb to induce IL-2 in this subset. As already mentioned, mAb OKT3 but not BMA 031 can bind to Fc receptors on monocytes/accessory cells of IgG2b low responder cells. This binding enables a stimulating mAb to cross-link the CD3/TcR complex on the responding T-cell (23,24) but might also trigger cytokine releasefrom the monocytes (25, 26). The different IL-2 inducing capacity of the two mAb could therefore indicate that cross-linking of the CD3/TcR complex is a prerequisite for IL2 induction in CD45RO+ cells. On the other hand, it might be possible that IL-2 production in CD45RO+ cells in addition depends on monocyte-derived factors. Since stimulation of CD45ROf cells by artificially immobilized BMA 031 resulted in IL-2 production and subsequent proliferation in the absence of accessory cells (data not shown), it is likely that cross-linking of the CD3/TcR complex is sufficient to induce IL-2 synthesis in CD45RO+ cells. In the present study we have shown that the anti-CD3 mAb OKT3 triggers proliferation in both CD45RA+ and CD45RO+ cells whereas the anti-TcR mAb BMA 03 1 induces a proliferative responseonly in CD45RA+ T-cells. Since it is likely that OKT3 and BMA 03 1 trigger different modes of activation (7) the differential response of CD45RAf and CD45ROf cells might indicate that the two subsetsuse discrete activation pathways. In addition to these mechanistic aspects,the observation that BMA 03 1 induces proliferation only in CD45RA+ T-cells is also of practical value since the majority of polyclonal stimuli have been shown to preferentially activate CD45RO+ cells (27-29). ACKNOWLEDGMENTS We thank A. Jeremias for expert technical assistance.Drs. P. C. L. Beverley and R. Kurrle are thanked for donating mAb.
REFERENCES 1. Reinherz, E. L., Meuer, S. C., and Schlossman,S. F., Zmmunol. Rev. 74, 83, 1983. 2. Clevers, H., Alarcon, B., Wileman, T., and Terhorst, C., Annu. Rev. Zmmunol. 6,629, 1988. 3. Meuer, S. C., Fitzgerald, K. A., Hussey, R. E., Hodgdon, J. C., Schlossman, S. F., and Reinherz, E. L., J. Exp. Med. 157,105, 1983. 4. Van Wauwe, J. P., de Mey, J. R., and Goossens,J. G., J. Zmmunol. 124, 2708, 1980. 5. Tax, W. J. M., Willems, H. W., Reekers, P. P. M., Capel, P. J. A., and Koene, R. P. P., Nature 304, 445, 1983. 6. Smith, K. G. C., Austyn, J. M., Hariri, G., Reverley, P. C. L., and Monks, P. J., Eur. J. Zmmunol. 16, 478, 1986. 7. Schlitt, H. J., Kurrle, R., and Wonigeit, K., Eur. J. Zmmunol. 19, 1649, 1989. 8. Schlitt, H. J., Schwinzer, R., and Wonigeit, K., &and. J. Zmmunol. 32, 7 17, 1990. 9. Sanders,M. E., Makgoba, M. W., and Shaw, S., Zmmunol. Today 9, 195, 1988.
ENHANCED CD45RA+ CELL REACTIVITY
41
10. Sanders, M. E., Makgoba, M. W., Sharrow, S. O., Stephany, D., Springer, T. A., Young, H. A., and Shaw, S., J. Zmmunol. 140, 1401, 1988. 11. Merkenschlager, M., Terry, L., Edwards, R., and Beverley, P. C. L., Eur. J. Zmmunol. 18, 1653, 1988. 12. OFlynn, K., Zanders, E. D., Lamb, J. R., Beverley, P. C. L., Wallace, D. L., Tatham, P. E. R., Tax, W. J. M., and Linch, D. C., Eur. J. Zmmunol. 15, I, 1985. 13. Kurrle, R., Seyfert, W., Trautwein, A., and Seiler, F. R., In “Leucocyte Typing II” (E. L. Reinherz et a/., Eds.), pp. 137-142. Springer-Verlag, New York, 1986. 14. Borst, J., Van Dongen, J. M., De Vries, E., Comans-Bitter, W. M., Van Tol, J. D., Vossen, J. M., and Kurrle, R., Human Zmmunol. 29, 175, 1990. 15. Smith, S. H., Brown, M. H., Rowe, D., Callard, R. E., and Beverley, P. C. L., Immunology 58,63, 1986. 16. Schwinzer, R., In “Leucocyte Typing IV” (W. Knapp et al., Eds.), pp. 628-634. Oxford Univ. Press, London, 1989. 17. Morimoto, C., Letvin, N. L., Distaso, J. A., Aldrich, W. R., and Schlossman, S. F., J. Zmmunol. 134, 1508, 1985. 18. Ledbetter, J. A., June, C. H., Martin, P. J., Spooner, C. E., Hansen, J. A., and Meier, K. E., J. Zmmunol. 136,3945,
1986.
19. Byrne, J. A., Butler, J. L., and Cooper, M. D., J. Zmmunol. 141, 3249, 1988. 20. Dohlsten, M., Hedlund, G., Sjogren, H. O., and Carlsson, R., Eur. J. Zmmunol. 18, 1173, 1988. 21. Wasik, M. A., and Morimoto, C., J Zmmunol. 144, 3334, 1990. 22. Damle, N. K., and Doyle, L. V., J. Zmmunol. 143, 1761, 1989. 23. Geppert, T. D., and Lipsky, P. E., J. Zmmunol. 138, 1660, 1987. 24. Van Lier, R. A. W., Brouwer, M., Rebel, V. I., Van Noesel, S., and Aarden, L. A., Immunology 68,45, 1989. 25. Manger, B., Weiss, A., Weyand, C., Gorozny, J., and Stobo, J. D., J. Zmmunol. 135, 3669, 1985. 26. Scheurich, P., Ucer, U., Wrann, M., and Pfizenmaier, K., Eur. J. Zmmunol. 15, 1091, 1985. 27. Sanders,M. E., Makgoba, M. W., June, C. H., Young, H. A., and Shaw, S., Eur. J. Zmmunol. 19,803, 1989. 28. Horgan, K. J., van Seventer, G. A., Shimizu, Y., and Shaw, S., Eur. J. Zmmunol. 20, 1111, 1990. 29. Wallace, D. L., and Beverley, P. C. L., Immunology 69,460, 1990.
IMMUNOLOGY
14,
31-41 (1992)
Monoclonal Antibodies to Common Epitopes of the Human a/(3 T-Cell Receptor Preferentially Activate CD45RA+ T-Cells’ REINHARD SCHWINZER,’HANS JORGENSCHLITT, AND KURT WONIGEIT~ Klinik fir Abdominal- und Transplantationschirurgie, Medizinische Hochschule Hannover, Hannover, Federal Republic of Germany Received June 3, 1991; accepted October 16, 1991 The murine monoclonal antibody BMA 031 (IgG2b) is directed to a monomorphic epitope on the human a/@T-cell receptor. In contrast to anti-CD3 antibodies of the IgG2b isotype, BMA 03 1 is able to induce a proliferative responsein T-cells from IgG2b low responders.This response occurs independently of cross-linking conditions indicating that the mode of activation differs from stimulation by the anti-CD3 antibody OKT3 (IgG2a) which strictly dependson cross-linking conditions. To further characterize the stimulatory potential of the two antibodies we studied the lymphocyte subsets responsive to stimulation by BMA 03 1 and OKT3. In CD45RA+ cells both antibodies exhibited similar effects. They induced weak expression of the 55kDa chain of the interleukin-2 receptor (CD25), virtually no interleukin-2 secretion, but neverthelessstrong proliferation. In CD45RO+ cells OKT3 and BMA 031 showed markedly different effects. OKT3 stimulated strong CD25 expression, strong interleukin-2 production, and marked proliferation. In contrast, CD45RO+ cells stimulated by BMA 031 showed only weak CD25 expression but neither interleukin-2 production nor proliferation. Thesedata suggestthat CD45RA+ and CD45RO+ cells differ in their capability to produce interleukin-2 upon stimulation via the CD3/T-cell receptor complex and also in the requirement for interleukin-2 to mount a proliferative response. The differential effect of OKT3 and BMA 03 1 in CD45RO+ cells probably results from the failure of BMA 031 to trigger interleukin-2 production which may be a consequence of its inability to induce CD3/T-cell receptor cross-linking in IgG2b low responders.BMA 03 1 is therefore a useful tool for the selective activation of CD45RA+ cells in these individuals. o 1992 Academic press, IX.
INTRODUCTION On the surface of mature T-cells the CD3 complex is expressedin close association with the T-cell receptor (TcR)~ (1, 2). Monoclonal antibodies (mAb) to the TcR and to CD3 are able to activate T-cells (3, 4). Since T-cell stimulation by soluble mAb usually requires cross-linking mediated by Fc receptors on accessorycells the stimulatory potential is dependent on the antibody isotype (5, 6). Anti-CD3 mAb of the ’ This work was supported by the Deutsche Forschungsgemeinschaft:“Forschergruppe Organtransplantation” (Pi 48/l 1; Dl/D4), and by the Jung Stiftung fur Wissenschaftund Forschung. ’ Present address: Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Denver, CO 80206. 3 To whom correspondenceshould be addressedat Klinik fur Abdominal- und Transplantationschirurgie, Medizinische Hochschule Hannover, Konstanty-Gutschow-Str. 8, D-3000 Hannover 61, FRG. 4 Abbreviations used: IL-2, interleukin-2; IL-2R, interleukin-2 receptor; mAb, monoclonal antibody(ies); PBMC, peripheral blood mononuclear cells; TcR, T-cell receptor. 31 000%8749192 $3.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
32
SCHWINZER, SCHLITT, AND WONIGEIT
IgG2a isotype are stimulatory for all individuals whereas IgG2b mAb of similar specificity stimulate only T-cells of very few so-called IgG2b high responders. The failure to thesemAb to stimulate T-cells of the vast majority of individuals is usually explained by insufficient binding of IgG2b by Fc receptors. An important role of the fine specificity of a stimulating mAb is suggestedby data from our laboratory showing that mAb directed to the constant portion of the human cz//3TcR differ from typical anti-CD3 mAb in their stimulatory properties. For example, lymphocytes from IgG2b low responders which fail to respond to anti-CD3 mAb of the IgG2b &type mount a proliferative response to the IgG2b mAb BMA 031 that is directed to the cu/pTcR. The proliferative responsetriggered by BMA 03 1 does not depend on cross-linking conditions. Whereas the anti-CD3 mAb OKT3 induces interleukin-2 (IL2) secretion, expression of high density interleukin-2 receptors (IG2R) and cytotoxic activity in bulk peripheral blood mononuclear cell (PBMC) cultures, PBMC stimulated by BMA 03 1 expressonly low levels of CD25 but do neither secrete IL-2 nor develop cytotoxic activity. Taken together these data suggestthat OKT3 and BMA 03 1 induce different modes of T-cell activation (7, 8). The differential responseof bulk PBMC cultures to OKT3 and BMA 03 1 could be explained in two different ways. If an individual T-cell has the capacity to respond to stimulation by using different signaling pathways, OKT3 and BMA 03 1 might trigger different signaling mechanisms and the subsequent activation of different pathways. On the other hand, if a given T-cell is inherently programmed to elicit a certain response after stimulation regardless of the type of stimulus, the different responses in bulk cultures could indicate that OKT3 and BMA 031 activate discrete T-cell subsets.To address this issue, we have analyzed the stimulatory capacities of OKT3 and BMA 031 in different T-cell subsets.Clear-cut differences were found in the response patterns of CD45RA+ and CD45RO+ T-cells which are regarded to represent naive and memory T-cells, respectively (9-l 1). MATERIALS AND METHODS Antibodies For the lymphocyte activation experiments mAb OKT3 (anti-CD3, IgG2a), UCHTl (anti-CD3, IgG2b) (12) BMA 03 1 (ant&//3 TcR, IgG2b), and BMA 032 (anti-a/p TcR, IgG2b) ( 13, 14) were used. Antibodies of the BMA serieswere provided by Dr. R. Kurrle (Behringwerke, Marburg, FRG). OKT3 was provided by Cilag GmbH (Alsbath-Hgihnlein, FRG). UCHTl was kindly provided by Dr. P. C. L. Beverley (London, UK). mAb BMA 040 (antiCD4, Behringwerke Ag, Marburg, FRG), OKT8 (antiCD8, American tissue culture collection, Rockville, MD), UCHLl (anti-CD45R0, donated by Dr. P. C. L. Beverley, London, UK) (15), and various antiCD45R/ CD45RA mAb obtained from the Fourth International Workshop on leucocyte differentiation antigens (16) were used to isolate T-cell subsets by means of negative selection. The directly labeled mAb FITC-Leu2 (anti-CD8), FITC-Leu3 (anti-CD4), FITC-Leu 18 (antiCD45RA), PE-anti-Tat (antiCD25), and FITCXeuM3 (anti-CD 14) used for flow cytometry analysis were purchased from Becton-Dickinson (Heidelberg, FRG). FITC-UCHLl (antiCD45RO) was obtained from Dako (Hamburg, FRG). Cell Preparations In the described experiments only lymphocytes from IgG2b low responders as determined by their failure to respond to stimulation with mAb UCHTl (IgG2b) and
ENHANCED
CD45RA+
CELL
REACTIVITY
33
to low concentrations (15 rig/ml) of BMA 03 1 were analyzed. Two types of cell preparations were used. In a first set of experiments PBMC were isolated from heparinized peripheral blood by Ficoll gradient centrifugation and depleted of various T-cell subsets. No further depletion of monocytes/accessory cells was performed. In another set of experiments a population of small resting T-cells was prepared by E-rosetting. Usually >95% of E+ cells were CD3+ and 98-99% reacted with anti-CD2 mAb. Irradiated (50 Gy) E- cells were used as a source of monocytes/accessory cells. Phenotypic analysis of this fraction revealed that it usually contained 45-60% of monocytes (CD14+). Isolation of T-Cell Subsets Magnetic immunoselection was performed to negatively isolate various T-cell subsets.PBMC or E+ cells were incubated with saturating dosesof BMA 040 (anti-CD4), OKT8 (anti-CDS), UCHLl (anti-CD45RO), or anti-CD45RA mAb for 30 min at 4°C. Cells were washed twice, mixed with sheep anti-mouse Ig-coated beads (Dynabeads, Dianova, Hamburg, FRG), and incubated for 60 min at 4°C. Cells reactive with the primary mAb and thus bound by the magnetic beads were removed with a magnet. Nonmagnetized cells were collected, washed twice, and reanalyzed. Usually lessthan 6% of cells from the reciprocal subsetwere detectedin thesepreparations as determined by flow cytometry analysis. Proliferation Assays Unseparated PBMC or negatively selectedcell preparations were cultured in roundbottomed microtiter plates (Nunc, Roskilde, Dk) at 5 X 105/ml ( 1 X 10’ cells/well) in RPM1 1640 medium supplemented with 10% fetal calf serum, penicillin ( 100 U/ ml), and streptomycin (100 &ml). In some experiments 20 U/ml of purified human IL-2 (Lymphocult T-HP, Biotest, Frankfurt, FRG) was added to the cultures. All antiCD3/TcR mAb were used at concentrations determined for optimal proliferation of unseparated PBMC (OKT3, 15 rig/ml; BMA 03 1, BMA 032, 1.5 pg/ml; UCHT 1, 1: 100 dilution of culture supernatant). If T-cells were used, stimulation was performed in the presenceof 2.5 X lo4 irradiated (50 Gy) autologous E- cells per well as a source of monocytes/accessory cells. Proliferation was assessed by incorporation of [3H]thymidine ([3H]-dThd) added 16 hr prior to harvest. Induction of IL-2 Secretion and IL-2R Expression To induce IL-2 secretion, E+ cells were prepared and further divided into CD45RA+ and CD45ROf cells by means of negative selection. CD45RAf or CD45RO+ T-cells ( 10 X lO’/ml; 2 X 105/well) were stimulated in the presence of 2.5 X lo4 irradiated autologous non-T-cells for 24 hr with mAb OKT3 or BMA 03 1. Measurement of IL2 in the supematant of these cultures was performed using the IL-2-dependent rat Tcell line G2 produced in our laboratory. G2 cells ( 1 X 104)were cultivated for 48 hr with various concentrations of supematant from cell cultures. Defined serial dilutions of purified IL-2 were included in each experiment for control. The proliferation of G2 cells was determined by [3H]-dThd incorporation for another 16 hr. Proliferation induced by the supematants was then compared with the IL-2 titration curve to calculate units per milliliter. To induce IG2R expression, PBMC or negatively isolated T-cell subsets(5 X 105/ml) were cultured in a final volume of 200 ~1in round-bottomed
34
SCHWINZER, SCHLITT, AND WONIGEIT
microtiter plates with medium, OKT3 ( 15 rig/ml), or BMA 03 1 ( 1.5 pg/ml). After 48 hr cells were collected, washed, and prepared for flow cytometry analysis.
Flow Cytometry Analysis Reanalysis of negatively isolated T-cell subsetswas performed by direct immunofluorescence using FITC-conjugated mAb. Residual binding of the mAb used for separation which may block reactivity of the directly labeled mAb was always controlled by a FITC-conjugated goat anti-mouse IgG but was never observed. Cells were washed twice with staining buffer (PBS containing 0.5% BSA and 0.1% NaN3) and incubated for 30 min at 4°C with saturating concentrations of mAb. After two additional washes analysis of the stained cells was performed on a FACStar flow cytometer (BectonDickinson, Mountain View, CA). To detect the 55-kDa chain of the IL-2R separately on CD45RA+ and CD45RO+ cells, two-color flow cytometry analyseswere performed. Cultured cells were stained with the mAb combinations FITCLeu18 (antiCD45RA)/ PE-anti-Tat (antXD25) or FITC-UCHLl (anti-CD45RO)/PE-anti-Tat (antXD25). Appropriate gates were set to include only activated lymphoblasts as identified by increased forward and side scatter intensity. Results are shown as contour plots where the FITC fluorescence intensity is plotted versus the PE fluorescence intensity. RESULTS
Enhanced Proliferative Responseof CD45RA+ Cells to Stimulation with mAb BMA 031 The previously observeddifferencesin the responsepatterns of PBMC to stimulation with OKT3 or BMA 03 1 (7, 8) raised the possibility that different T-cell populations are preferentially activated by the two mAb. To analyze this question, PBMC and negatively isolated T-cell subsets were stimulated with the two mAb. In CD4+ and CD8 cells OKT3 and BMA 031 exhibited similar stimulatory capacities (data not shown). In contrast, cell preparations depleted either of CD45ROf or CD45RA+ cells exhibited a differential response pattern. Both were induced to proliferate by binding of OKT3 (Fig. 1). The CD45RA+ fraction (CD45RO depleted) showed an enhanced responseto BMA 03 1 stimulation whereas CD45RO+ cells (CD45RA depleted) were nearly unresponsive to BMA 03 1-triggered stimulation. This unusual reactivity pattern was also found using another anti-TcR mAb (BMA 032). Since both BMA 031 and BMA 032 are of the IgG2b isotype, control experiments were performed using the IgG2b anti-CD3 mAb UCHT 1. As expected from the IgG2b nonresponder status of the tested blood donors, this mAb triggered a significant proliferative response in neither CD45RO+ nor CD45RAf cells (data not shown) indicating that the specificity of the tested anti-TcR mAb contributes to their unusual stimulatory activity. The experiments summarized in Fig. 1 were performed using mAb concentrations inducing a maximal proliferative response in PBMC ( 15 r&ml OKT3 and 1.5 &ml BMA 03 1). To exclude the possibility that the differential pattern of OKT3- and BMA 031-induced activation observed in CD45RA+ and CD45RO+ cells may result from the different antibody amounts, various concentrations of OKT3 and BMA 03 1 were tested. In line with earlier observations (7) effective stimulation by BMA 03 1 required much higher mAb concentrations as compared with OKT3 (Table 1). Although the strong responding CD45RAf cells showed a small proliferative response when stim-
35
ENHANCED CD45RA+ CELL REACTIVITY
Cells PBMC
CD45RO depl.
CD45RA depl.
0
50
100
150
200
% relative
260
300
350
400
response
FIG. 1. Response patterns of cell population depleted of CD45RA+ and CD45ROCcells to stimulation with OKT3 and BMA 031. Unseparated PBMC, CD45RO- (CD45RA+), and CD45RA (CD45RO+) cell populations were stimulated with 15 rig/ml OKT3 (stippled bars) and 1.5pg/ml BMA 03 1 (solid bars). Since proliferation induced by OKT3 wasgenerally higher than that induced by BMA 03 1 the proliferative responses of the various cell populations are expressedasthe percentageof the responseobservedin PBMC to stimulation with the respective mAb. The results represent the means f SD obtained in four (CD45RO depleted) and five (CD45RA depleted) individual experiments.
mated with 15 rig/ml, a clear-cut responsecould only be achieved using high dosesof BMA 03 1 (1.5 pg/ml). CD45RO+ cells were nearly unresponsive to BMA 03 1 stimulation irrespective of the mAb concentration. In all cell preparations tested, OKT3 reached a maximal proliferative response already at 15 rig/ml. This result indicates that the differential response of CD45RAf and CD45RO+ cells to anti-CD3 and antiTcR stimulation did not result from different mAb concentrations used for stimulation. Monocytes expressCD45RO in high density whereas only low amounts of CD45RA molecules can be detected on their cell surface. Therefore, it is likely that negatively
TABLE 1 Proliferative Responsesof PBMC, CD45RA+, and CD45ROf Cells to Stimulation with Different Concentrations of OKT3 and BMA 03 1” BMA031”
OKT3 b Experiment
Cells
15
150
1500
15
150
1500
1
PBMC CD45RA+ PBMC CD45RO+
60.1 19.9 40.2 30.4
51.5 24.6 nd’ nd
50.0 22.7 32.8 22.6
1.8 4.6 2.8 1.8
2.7 6.8 8.9 1.6
9.1 29.8 11.2 1.9
2
’ Unseparated PBMC (1 X 10’) or PBMC enriched for CD45RA+ or CD45RO+ cells by depletion of CD45RO+and CD45RA+ cells, respectively, were stimulated for 3 days with OKT3 or BMA 03 1. Proliferation was determined during the last 16 hr of culture by [)H]dThd incorporation. Results are expressedas cpm X lo-’ of triplicate cultures corrected by subtraction of media controls. b rig/ml. ‘Not determined.
36
SCHWINZER, SCHLITT, AND WONIGEIT
selectedCD45RA+ and CD45RO+ PBMC contain a different proportion of monocytes/ accessorycells. To exclude the possibility that the differential response of CD45RA+ and CD45RO+ cells to anti-CD3 and anti-TcR stimulation results from differences in purity, a population of small resting T-cells (E-rosette positive) was prepared and further divided into CD45RA* and CD45RO+ cells. These cells were stimulated in the presence of equal numbers of autologous irradiated non-T-cells as a source of monocytes/accessory cells. In accordance with the experiments using PBMC (Fig. 1) a differential effect of OKT3 and BMA 03 1 was also observedin these cell preparations. OKT3 was able to induce a proliferative response in both CD45RA+ and CD45RO+ T-cells (Table 2). Although BMA 03 1 triggered a proliferative response in CD45RA+ T-cells this mAb was nearly unable to induce proliferation in CD45ROf T-cells. A proliferative response to stimulation with the recall antigen tetanus-toxoid was only observed in CD45RO+ T-cells whereas the CD45RA subset was completely unresponsive. The differential tetanus-toxoid responsesare in line with published data ( 15, 17) and indicate that CD45RA+ and CD45RO+ T-cells representthe naive and the memory T-cell pool, respectively.
Induction of IL-2 Responsivenessin BMA 031-Stimulated CD45RO’ T-Cells The low responsivenessof CD45RO+ T-cells to stimulation with BMA 03 1 raised the question whether these cells cannot be activated at all by BMA 031 or whether they achieve only an early state of activation. Since the induction of IL-2 responsiveness is regarded to be an early step in T-cell activation (18), we analyzed the effect of exogeneous IL-2 on BMA 031-induced activation of CD45RO+ cells. As shown in Table 3, BMA 03 1 was very effective in triggering proliferation of CD45RO+ T-cell if IL-2 was provided exogeneously. These data indicate that BMA 03 1 is able to activate CD45RO+ T-cells. This activation, however, seems to be incomplete since it only results in the induction of functional IL-2R but not in proliferation.
D#erences in IL-2 Secretion betweenCD45RA’ and CD45RO’ T-Cells Stimulated via the CD3/TcR Complex The observation that BMA 03 1 triggered proliferation of CD45ROt T-cells strictly depended on exogeneous IL-2 whereas CD45RA T-cells proliferated without the adTABLE 2 Proliferative Responsesof T-Cell SubsetsNegatively Selected for the Expression of CD45RA or CD45RO to Stimulation with CD3/TcR mAb and Tetanus-Toxoid” Proliferative Response(cpm X 10m3)to Experiment
Cells
Medium
OKT3
BMA 03 1
TT
1 2 3
CD45RA+ CD45RO+ CD45RA+ CD45RO+
0.6 0.7 0.9 0.6
29.5 39.6 17.8 14.4
28.0 2.1 25.3 3.2
0.9 8.7 ndb nd
’ E” cells were further divided into CD45RA+ and CD45RO+ T-cells by means of negative selection and stimulated ( 1 X IO5cells/well) with OKT3 (15 rig/ml), BMA 03 1 (1.5 &ml), or tetanus-toxoid (1:300 v/ v). Irradiated autologous E- cells (2.5 X 10“) were added to each well. Results are expressedas cpm X IO-’ of triplicate cultures. b Not determined.
37
ENHANCED CD45RA+ CELL REACTIVITY TABLE 3 Induction of IL-2 Responsivenessin CD45ROf Cells Stimulated by BMA 03 1”
Proliferative Response(cpm X 10d3)to Experiment
Cells
IL-2
Medium
OKT3
BMA 03 1
1
PBMC CD45RO+
2
E+ CD45RO+
3
E+ CD45RO+
+ + +
0.7 10.0 1.7 11.2 0.1 7.5
30.7 83.4 33.8 56.1 10.0 31.8
1.8 50.0 4.5 31.9 1.3 20.0
n PBMC (I X IO’) or T-cells (E+) enriched for CD45RO+ cells by depletion of the CD45RA+ population were stimulated for 3 days in the absence or in the presence of 20 U/ml IL-2 with OKT3 (15 rig/ml) or BMA 031 (1.5 pg/ml). Proliferation was determined during the last 16 hr of culture by [3H]-dThd incorporation. Results are expressedas cpm X IO-’ of triplicate cultures.
dition of IL-2 could mean that BMA 03 1 is able to induce IL-2 secretion in CD45RA+ but not in CD45ROt T-cells. On the other hand, it could also be possiblethat CD45RA+ and CD45RO+ T-celis have different requirements for IL-2 to mount a proliferative response.To distinguish between these alternatives, the amount of IL-2 was measured in the supematant of CD45RA+ and CD45ROf T-cells stimulated with OKT3 or BMA 03 1. After stimulation of purified CD45RAf T-cells with OKT3 for 24 hr, only the highest concentration of supematant induced a marginal proliferative responsein the IL-2-dependent rat T-cell line G2 indicating that there is only a small amount (co.5 U/ml) of IL-2 present in these cultures (Fig. 2). IL-2 was completely undetectable in the supematant of BMA 031~stimulated CD45RA+ cells. In the supematant of CD45RO+ T-cells stimulated with OKT3, IL-2 could readily be detected (4 U/ml). Again, stimulation by mAb BMA 031 did not result in measurable IL-2 release. Diferential IL-2 Receptor Expression Pattern after OKT3- and BMA 031-Induced Activation The induction of functional IL-2R by BMA 031 stimulation could be confirmed by flow cytometry analysis revealing low levels of the 55-kDa chain of the IG2R (CD25) in negatively isolated CD45RA+ and CD45RO+ T-cells. In contrast, OKT3 induced low levels of CD25 in CD45RA+ cells but high levels in CD45RO+ cells (data not shown). To exclude separation artifacts as a reason for these patterns, two-color analyses were performed in unseparated PBMC. In line with the experiments using isolated subsets there was a marked differential expression of CD25 in CD45RA+ (CD45RO-) and CD45RA- (CD45RO+) cells as a result of anti-CD3 (OKT3) stimulation (Fig. 3, upper panels). Whereasonly a small subsetofthe CD45RAf (CD45RO-) population expressed CD25, nearly the whole population of CD45RA- (CD45RO’) cells was CD25+. In addition, the concentration of CD25 expressedon the cell surface differed between CD45RA+ (CD45RO-) and CD45RX (CD45ROf) cells. CD45RA+ (CD45RO-) cells expressed much lower levels of CD25 (320 arbitrary fluorescence units) as compared to CD45RA- (CD45RO+) cells (670 fluorescence units). A quite different pattern was observed in anti-TcR (BMA 031)-stimulated cultures. In these
38
SCHWINZER, SCHLITT, AND WONIGEIT CD45RA+
30
-+
medium
-
OK13
-
EMA 031
26 m
“’
I +--, - ’ J
I:2
t:,
I:8
I:2
I:4
I:8
I:16
dilution
FIG. 2. Proliferative responsesof the IL-2dependent rat T-cell line G2 to various dilutions of culture supematant from stimulated CD45RA+ and CD45RO+ T-cells. E+ T-cells were divided into the subsets CD45RA and CD45ROand cultured in the presenceof irradiated autologous non-T-o&.. Cells were stimulated for 24 hr with OKT3 (15 r&ml) or BMA 03 1 (1.5 &ml). The supematants were collected and titrated on G2 cells (1 X 104/well).Proliferation of G2 cells was determined after 2 days. CD45RA+ T-cells proliferated to stimulation with both OKT3 and BMA 03 1 (53,336 and 38,916 cpm, respectively). The proliferative responsesof CD45RO+ cells were 39,578 cpm (OKT3) and 2127 cpm (BMA 031).
cultures, the expression of CD25 on CD45RA+ (CD45RO-) cells was comparable to that found after OKT3 stimulation. However, in contrast, the characteristic high expression density of CD25 found in OKT3-stimulated CD45RA- (CD45RO+) cells was never observed. In BMA 03 l-stimulated cultures also the CD45RA (CD45ROf) cells expressed only low levels of CD25. According to the largely nonoverlapping expression of CD45RA and CD45R0, reciprocal patterns were observed when the cultured cells were stained with the mAb combination anti-CD45RO/anti-CD25 (Fig. 3, lower panels). DISCUSSION This paper elucidates the cellular basis of the differential stimulating capacity of the mAb BMA 03 1 and OKT3. It demonstrates that OKT3 is able to stimulate proliferation of both CD45RAt and CD45RO+ cells whereas BMA 031 predominantly stimulates CD45RA+ cells. It has previously been shown that CD45RA+ and CD45RO+ cells which are regarded to represent naive and memory T-cells, respectively, have different stimulatory requirements to mount a proliferative response (19-22). Our results demonstrate that under the culture conditions used BMA 031 fulfills the requirements to induce proliferation in CD45RA+ but not in CD45RO+ cells. The comparison of the activation events induced in CD45RA+ and CD45RO+ cells revealed that in both populations CD25 is induced at low levels (Fig. 3) but that no measurableIL-2 can be demonstrated in the culture supematant (Fig. 2). Since addition of exogeneous IL-2 led to a proliferative response of CD45RO+ cells as well (Table 3) it can be concluded that a major causefor the differential responseof CD45RAf and CD45RO+ cells upon stimulation
ENHANCED CD45RA+ CELL REACTIVITY medium
CD45RA-FITC
-
CD46RO-FITC
-
anti-CD3 (OKT3)
39
anti-TcR (BMA 031)
FIG. 3. Different patterns of IL-2R expression in CD45RAf and CD45RO+cells stimulated by OKT3 and BMA 03 1. PBMC were cultured for 2 days with medium, OKT3 (15 @ml), or BMA 03 1 (1.5 &ml). The cells were harvested and stained with FITC-Let118 (CD45RA, upper panels) or FITC-UCHLI (CD45R0, lower panels) in combination with PE-anti-Tat (antiCD25). Data are displayed as contour plots where the FfTC fluorescenceintensity (x-axis) is plotted versus the PE fluorescenceintensity (y-axis) on a logarithmic scale. On the basis of control samples the plots were divided into quadrants: (1) Cells with PE fluorescence only; (2) cells with PE and FITC fluorescence;(3) unstained cells; and (4) cells with FITC fluorescenceonly. Analysis was selectively performed on blastoid cells as determined by an increased forward and side scatter intensity. Note that OKT3 induced high density expression of CD25 in CD45RA (CD45RO+) and low density in CD45RA+ (CD45RO-) cells whereas after BMA 03 1 stimulation both subsetsexpressed CD25 only at low levels.
with BMA 031 is a difference in the requirement for IL-2. This conclusion is also supported by the finding that isolated CD45RA+ cells responded to OKT3 stimulation by marked proliferation although no or only very little production of IL-2 could be demonstrated. In addition to the different requirement for IL-2 the different potential to produce this lymphokine appears to be another inherent functional discrepancy between CD45RA+ and CD45RO+ cells at least when antiCD3/TcR mAb are used for stimulation. This follows from the finding that OKT3, which was ineffective in eliciting IL-2 production in CD45RA+ cells, induced strong IL-2 secretion in CD45RO+ cells (Fig. 2). Since the expression of CD25 was also stronger on CD45RO+ cells (Fig. 3) it can be assumedthat there was substantial IL-2 consumption suggestingan even larger difference in the production rate than that indicated by the IL-2 measurement in the culture supematant. Although the response of CD45RA+ cells to OKT3 and BMA 03 1 was remarkably similar it is likely that different activation mechanisms are involved in this situation. Since the activation of unseparated PBMC by OKT3 is strictly dependent on crosslinking conditions this might also apply for isolated CD45RA+ cells. This conclusion is supported by the finding that the IgG2b variant of the anti-CD3 mAb UCHTl was not able to stimulate CD45RA+ cells of IgG2b low responder individuals lacking the
40
SCHWINZER, SCHLITT, AND WONIGEIT
appropriate Fc receptors on their accessorycells (data not shown). In contrast, BMA 03 1 and BMA 032 effectively stimulated CD45RA+ cells although both mAb are also of the IgG2b isotype. As previously postulated (7, 8) this points to a major effect of the epitope on the CD3/TcR complex to which the stimulating mAb is directed. In CD45RA+ cells only certain epitopes appear to be able to induce proliferation in the absenceof cross-linking. In contrast to CD45RAf cells, the CD45RO+ population required sufficient concentrations of IL-2 to mount a proliferative response. Thus, the differential effects of OKT3 and BMA 03 1 in CD45RO+ cells in all likelihood reflect a differential capacity of the two mAb to induce IL-2 in this subset. As already mentioned, mAb OKT3 but not BMA 031 can bind to Fc receptors on monocytes/accessory cells of IgG2b low responder cells. This binding enables a stimulating mAb to cross-link the CD3/TcR complex on the responding T-cell (23,24) but might also trigger cytokine releasefrom the monocytes (25, 26). The different IL-2 inducing capacity of the two mAb could therefore indicate that cross-linking of the CD3/TcR complex is a prerequisite for IL2 induction in CD45RO+ cells. On the other hand, it might be possible that IL-2 production in CD45RO+ cells in addition depends on monocyte-derived factors. Since stimulation of CD45ROf cells by artificially immobilized BMA 031 resulted in IL-2 production and subsequent proliferation in the absence of accessory cells (data not shown), it is likely that cross-linking of the CD3/TcR complex is sufficient to induce IL-2 synthesis in CD45RO+ cells. In the present study we have shown that the anti-CD3 mAb OKT3 triggers proliferation in both CD45RA+ and CD45RO+ cells whereas the anti-TcR mAb BMA 03 1 induces a proliferative responseonly in CD45RA+ T-cells. Since it is likely that OKT3 and BMA 03 1 trigger different modes of activation (7) the differential response of CD45RAf and CD45ROf cells might indicate that the two subsetsuse discrete activation pathways. In addition to these mechanistic aspects,the observation that BMA 03 1 induces proliferation only in CD45RA+ T-cells is also of practical value since the majority of polyclonal stimuli have been shown to preferentially activate CD45RO+ cells (27-29). ACKNOWLEDGMENTS We thank A. Jeremias for expert technical assistance.Drs. P. C. L. Beverley and R. Kurrle are thanked for donating mAb.
REFERENCES 1. Reinherz, E. L., Meuer, S. C., and Schlossman,S. F., Zmmunol. Rev. 74, 83, 1983. 2. Clevers, H., Alarcon, B., Wileman, T., and Terhorst, C., Annu. Rev. Zmmunol. 6,629, 1988. 3. Meuer, S. C., Fitzgerald, K. A., Hussey, R. E., Hodgdon, J. C., Schlossman, S. F., and Reinherz, E. L., J. Exp. Med. 157,105, 1983. 4. Van Wauwe, J. P., de Mey, J. R., and Goossens,J. G., J. Zmmunol. 124, 2708, 1980. 5. Tax, W. J. M., Willems, H. W., Reekers, P. P. M., Capel, P. J. A., and Koene, R. P. P., Nature 304, 445, 1983. 6. Smith, K. G. C., Austyn, J. M., Hariri, G., Reverley, P. C. L., and Monks, P. J., Eur. J. Zmmunol. 16, 478, 1986. 7. Schlitt, H. J., Kurrle, R., and Wonigeit, K., Eur. J. Zmmunol. 19, 1649, 1989. 8. Schlitt, H. J., Schwinzer, R., and Wonigeit, K., &and. J. Zmmunol. 32, 7 17, 1990. 9. Sanders,M. E., Makgoba, M. W., and Shaw, S., Zmmunol. Today 9, 195, 1988.
ENHANCED CD45RA+ CELL REACTIVITY
41
10. Sanders, M. E., Makgoba, M. W., Sharrow, S. O., Stephany, D., Springer, T. A., Young, H. A., and Shaw, S., J. Zmmunol. 140, 1401, 1988. 11. Merkenschlager, M., Terry, L., Edwards, R., and Beverley, P. C. L., Eur. J. Zmmunol. 18, 1653, 1988. 12. OFlynn, K., Zanders, E. D., Lamb, J. R., Beverley, P. C. L., Wallace, D. L., Tatham, P. E. R., Tax, W. J. M., and Linch, D. C., Eur. J. Zmmunol. 15, I, 1985. 13. Kurrle, R., Seyfert, W., Trautwein, A., and Seiler, F. R., In “Leucocyte Typing II” (E. L. Reinherz et a/., Eds.), pp. 137-142. Springer-Verlag, New York, 1986. 14. Borst, J., Van Dongen, J. M., De Vries, E., Comans-Bitter, W. M., Van Tol, J. D., Vossen, J. M., and Kurrle, R., Human Zmmunol. 29, 175, 1990. 15. Smith, S. H., Brown, M. H., Rowe, D., Callard, R. E., and Beverley, P. C. L., Immunology 58,63, 1986. 16. Schwinzer, R., In “Leucocyte Typing IV” (W. Knapp et al., Eds.), pp. 628-634. Oxford Univ. Press, London, 1989. 17. Morimoto, C., Letvin, N. L., Distaso, J. A., Aldrich, W. R., and Schlossman, S. F., J. Zmmunol. 134, 1508, 1985. 18. Ledbetter, J. A., June, C. H., Martin, P. J., Spooner, C. E., Hansen, J. A., and Meier, K. E., J. Zmmunol. 136,3945,
1986.
19. Byrne, J. A., Butler, J. L., and Cooper, M. D., J. Zmmunol. 141, 3249, 1988. 20. Dohlsten, M., Hedlund, G., Sjogren, H. O., and Carlsson, R., Eur. J. Zmmunol. 18, 1173, 1988. 21. Wasik, M. A., and Morimoto, C., J Zmmunol. 144, 3334, 1990. 22. Damle, N. K., and Doyle, L. V., J. Zmmunol. 143, 1761, 1989. 23. Geppert, T. D., and Lipsky, P. E., J. Zmmunol. 138, 1660, 1987. 24. Van Lier, R. A. W., Brouwer, M., Rebel, V. I., Van Noesel, S., and Aarden, L. A., Immunology 68,45, 1989. 25. Manger, B., Weiss, A., Weyand, C., Gorozny, J., and Stobo, J. D., J. Zmmunol. 135, 3669, 1985. 26. Scheurich, P., Ucer, U., Wrann, M., and Pfizenmaier, K., Eur. J. Zmmunol. 15, 1091, 1985. 27. Sanders,M. E., Makgoba, M. W., June, C. H., Young, H. A., and Shaw, S., Eur. J. Zmmunol. 19,803, 1989. 28. Horgan, K. J., van Seventer, G. A., Shimizu, Y., and Shaw, S., Eur. J. Zmmunol. 20, 1111, 1990. 29. Wallace, D. L., and Beverley, P. C. L., Immunology 69,460, 1990.
