Potent Apoptotic Signaling and Subsequent Unresponsiveness Induced by a Single CD 2 mAb ( BTI-322 ) in Activated Human Peripheral T Cells

Manipulation of CD2 molecules with CD2 mAb pairs has been shown to deliver apoptotic signals to activated mature T cells. We show that BTI-322, a CD2 mAb directed at a peculiar epitope of CD2, can trigger on its own the apoptotic death of IL-2-activated peripheral T cells and of OKT3-stimulated T cells, contrasting in this respect with a series of other mouse or rat CD2 mAb. F(ab *)2 fragments were as potent as the whole Ab. BTI-322-induced apoptosis proceeded in a few hours and was independent of the Fas/Fas ligand system. Less than 5 ng/ml of BTI-322, added at the begining of culture, were able to eliminate within 4 days most CD3 cells from OKT3and IL-2-stimulated lymphocytes, the only cells remaining being CD16 CD2 NK cells. T cell proliferative responses induced by a mitogenic CD2 mAb pair or by PHA-P (which mainly binds to CD2) were not inhibited by BTI-322. In this case, the apoptotic effect was successfully counteracted by simultaneous enhancement of T cell divisions. Thus, the killing effect of BTI-322 was most effective when T cells were exclusively stimulated through the CD3/TCR complex. Apoptosis of the responding T cells may explain why T cells recovered from a primary MLC performed in the presence of BTI-322 responded to third party cells but not to the primary stimulatory cells. These data constitute the rational basis for the use of BTI-322 for inducing tolerance in human allotransplantation. The Journal of Immunology,1998, 160: 3797–3804.

T he CD2 receptor is a transmembrane 50-kDa glycoprotein expressed early in ontogeny on the surface of human thymocytes, mature T cells, and NK cells, the ligands of which are CD58 (LFA-3) and CD59 (1)(2)(3), in addition to CD48, the murine analogue of CD58 (4).The extracellular region of the CD2 molecule comprises two Ig superfamily domains, the first domain binding CD58 and CD59, allowing CD2 to function as an adhesion molecule and to promote the cognate interactions between T cells and APCs (5) and between thymocytes and thymic epithelial cells (6).CD2 has also a large cytoplasmic domain with proline-rich sequences (7), which allows it to associate with the SH3 domain of p56 lck (8) and perhaps with the other src tyrosine kinase p59 fyn which has been shown to coimmunoprecipitate with CD2 (9,10).Cross-linking of CD2 with certain pairs of mAbs provides a potent stimulus which leads to the activation of p56 lck (11), to the tyrosine phosphorylation of numerous substrates (12,13), to the initiation of the phospholipase C pathway (14), and finally to IL-2 synthesis and T cell proliferation (15).Optimal activation of T cells in vitro is strongly dependent on CD2/CD58 interactions (16,17), and although single CD2 mAb are non mi-togenic by themselves, they can costimulate T cell activation via the TCR (18,19).In addition, the perturbation of CD2 can fully trigger the cytolytic machinery of CTL and NK cells (20).
Contrasting with these T cell activation events, perturbation of CD2 may represent a potential negative regulatory mechanism for T cell responses as initially suggested by the inhibitory effect of a single CD2 mAb directed at the CD58 binding site (21,22).Blocking cellular adhesion is not the sole mechanism by which CD2 mAb can induce T cell unresponsiveness.Indeed, inhibitory signals were shown to be transduced through CD2 during interaction with the dimeric form of a soluble LFA-3 (first domain)/IgG fusion protein (LFA3TIP), 3 independent of blocking CD2/CD58 interactions (23).This resulted in sustained specific T cell unresponsiveness to CD3/TCR-derived stimuli.The negative regulatory role of CD2 has also been documented in experimental models.The injection in mice of a nondepleting CD2 mAb was shown to induce a long-lasting state of T cell unresponsiveness, irrespective of CD2 modulation (24).The negative signaling capacity of CD2 has been exploited to induce Ag-specific anergy in experimental organ transplant models by administering CD2 and CD48 mAb, either separately or in combination (25)(26)(27).A single injection of CD2 mAb at cardiac engraftment induces indefinite donor-specific tolerance in transplanted rats (28), and administration of LFA3TIP significantly prolongs primate cardiac allograft survival (29).
Another regulatory mechanism of CD2 molecules might reside in their capacity to transduce apoptotic signals to activated T cells.Previous studies, including ours, have reported that CD2-induced apoptosis occurs provided T cells are exposed to two CD2 mAb directed at distinct epitopes (30 -33).In this report, we show that the rat CD2 mab BTI-322, originally described by Bazin and colleagues as Lo-CD2a (34), is capable on its own of delivering potent apoptotic signals to activated T cells.In primary T cells polyclonally stimulated through the CD3/TCR, this results in almost complete elimination of the responding CD3 ϩ cells.This may explain why BTI-322 has strong suppressive effects on in vitro T cell proliferative responses and cytokine synthesis, and why it induces T cell alloantigen hyporesponsiveness (35,36).Our data confirm and extend the idea that the manipulation of CD2 with certain CD2 mAb can have drastic inhibitory effects on T cell responses.The potential use of BTI-322 for the treatment of acute graft rejection and for the induction of tolerance in human allotransplantation is discussed.

Cell separation
Blood samples from healthy volunteers of both sexes were provided from the Blood Transfusion Center from Ho ˆpital Saint Louis, Paris, France.PBMC were isolated by Ficoll-Isopaque (Eurobio, Les Ulis, France) density (d ϭ 1.078) gradient centrifugation.Adherent cells were removed by incubation on plastic dishes for 30 min at 37°C, and the rest of the cells were fractionated on nylon wool columns.In some experiments, monocytes were recovered and mixed with T cells at a 1:10 ratio.

Flow cytometric analysis of cell death
Cells were stained with propidium iodide (5 g/ml) 10 min before examination to detect dead cells (FL3 positive) using a FACScan (Becton Dickinson, Mountain View, CA) or a Coulter Epics Profile II cytofluorometer (Coulter Electronics, Hialeah, FL).

Hypoploid cell assessment
5 ϫ 10 5 cells were washed twice in PBS with 5.5 mM glucose and fixed overnight in ethanol (70% in water, at 4°C).Cells were then resuspended in 0.5 ml of PBS containing 50 g/ml propidium, 100 U/ml RNase A (Sigma) and incubated for 30 min at room temperature under agitation.The DNA content of 10 4 cells was monitored by cytofluorometry using a Coulter Epics profile II analyzer.

Proliferation assays
Cells were incubated with 1 Ci/well [ 3 H]TdR (Amersham, Les Ulis, France), and the amount of radioactivity incorporated was determined after a 6-h pulse.

The inability of BTI-322 mAb to participate in a mitogenic CD2 mAb pair is correlated with its cell death-inducing capacity
In our initial observation (Fig. 1A), peripheral blood T cells exposed to the combinations of T11 1 ϩ D66, T11 1 ϩ GT2, and GT2 ϩ D66 CD2 mAb (1 g/ml) were activated to mount a proliferative response, whereas the combination of BTI-322 with T11 1 , GT2, or D66 triggered only little cell proliferation.While IL-2 further enhanced the proliferation of T cells exposed to mitogenic CD2 mAb pairs such as GT2 ϩ T11 1 , it did not overcome the failure of CD2 mAb pairs containing BTI-322 to trigger a proliferative response, as assessed in kinetic experiments (Fig. 2A).Note, however, that a small level of proliferation could be induced by the combination of BTI-322 ϩ D66, in accordance with the observation that engagement of CD2 with D66 mAb alone provides a potent activation signal, resulting in the tyrosine phosphorylation of the GTPase-activating protein-associated 62-kDa protein (13).
The introduction of BTI-322 within a CD2 mAb pair in the absence of IL-2, at the beginning of culture, was always correlated with the generation of propidium iodide-permeable dead cells (Fig. 1B).BTI-322 alone was unable to induce significant cell death in nonstimulated T cells, suggesting that its association with another CD2 mAb first resulted in T cell activation, then in cell death.Exogenous IL-2 did not correct the failure of BTI-322 to trigger substantial cell divisions when associated with another CD2 mAb.Instead, BTI-322 became capable of inducing within 6 days the FIGURE 1. BTI-322 mAb is unable to form a mitogenic pair with another CD2 mAb but rather induces cell death.A, 1 ϫ 10 5 nylon woolpurified T cells were stimulated for 4 days with 1 g/ml of the indicated CD2 mAb.[ 3 H]TdR incorporation corresponded to a 6-h pulse performed at the end of the culture period.B, 1 ϫ 10 5 T cells exposed for 6 days to the indicated CD2 mAb (1 g/ml).Cell viability was then estimated by flow cytometry after staining with propidium iodide to detect dead cells (red fluorescence).accumulation of up to ϳ60% cells exhibiting fragmented DNA regardless of whether it was added alone or in combination with another CD2 mAb (Fig. 2B).Since IL-2 can induce the in vitro proliferation of those peripheral T cells that constitutively express functional IL-2 receptors, provided monocytes are present (37,38), it is very likely that such cells were the targets of the lethal effect of BTI-322.

The inhibitory effect of BTI-322 on T cell proliferative responses is primarily based on its potent apoptotic activity
The relationship between the cell death-inducing capacity of BTI-322 and its effect on T cell proliferation was further assessed in a polyclonal proliferation system, namely nylon wool-purified T cells subjected to OKT3 and IL-2 stimulation.When added at the beginning of culture, BTI-322 (at 1 g/ml) inhibited cell division and at the same time induced cell death (see Fig. 3).At similar Ab concentrations, T11 1 and D66, as well as Lo-Tact-1 (a rat IgG2a anti-CD25), also exerted some inhibitory effect on T cell proliferation, but they did not significantly enhance the rate of cell death above the background level.Dose-response experiments showed that the killing effect of BTI-322, examined at day 4 of the culture period, was detectable at 1 ng/ml and plateaued at 10 ng/ml, yielding up to 70% dead cells (Fig. 3C).Cell death rate was strictly correlated with the inhibition of [ 3 H]TdR incorporation (not shown) and with the decrease in the relative numbers of CD3 ϩ cells.The small percentage of cells escaping the lethal effect of 10 ng/ml BTI-322 proved to be NK cells: after reculture for 6 days in IL-2-containing medium, they expanded into CD3-negative cells expressing CD16, a specific NK cell marker (Fig. 4).Those NK cells were negative for CD2 expression and remained so for an extended period of time (Ͼ12 days in IL-2-containing medium), despite regular washings to remove putative residual BTI-322 molecules, suggesting that a CD2-negative population of NK cells was surviving BTI-322 treatment.
Cell death induced by BTI-322 had the typical features of apoptosis.When preactivated T cells (that had been exposed to OKT3) were treated with 1 g/ml BTI-322, ϳ50% hypoploid (Ͻ2N DNA) cells were generated within 16 h, revealing in electron microscopic analysis condensed chromatin and loss of microvilli (Fig. 5).In contrast, whatever their concentration, T11 1 or D66 did not induce apoptotic cell death above the background level seen with control Lo-DNP 57 mAb, an irrelevant rat IgG2b.The same was true for 39C1.5 and CD2.6, two CD2 rat mAb that display an IgG2a isotype and are able to form mitogenic pairs with other CD2 mAb (39) (not shown).
A minimum of 25 ng/ml BTI-322 was required to induce maximum (Ն50%) apoptotic cell death (Fig. 6), while we repeatedly  observed, as previously reported (33), that at least a 1-g/ml Ab concentration of the T11 1 ϩD66 mAb pair was required to achieve this effect.BTI-322 was able to trigger the same rate of cell death in the presence of added monocytes (10%) or after elimination of remaining monocytes with CD14 mAb ϩ complement.F(abЈ) 2 fragments were also effective, arguing against the role of FcRbearing accessory cells in this phenomenon.In addition, humanized BTI-322 with a human IgG1 Fc had the same killing capacity as the nonmanipulated rat Ab.Maximum apoptosis was achieved in ϳ7 h (Fig. 6B) and did not involve the synthesis or the release of Fas ligand (Fas-L), as evidenced by the fact that cell death was not inhibited by soluble M3 mAb, an anti-Fas Ab which, unlike control anti-Fas M33, can prevent the interactions between Fas receptor and its ligand, Fas-L (40) (see Fig. 6C).

BTI-322 does not inhibit T cell proliferative responses directly stimulated through CD2
BTI-322 mAb is able to inhibit the formation of T cell conjugates with sheep erythrocytes (34), while D66 is unable to block the formation of such conjugates and GT2 displays only partial E rosette-blocking activity (41,42).The cytofluorometric analysis of Figure 7A indeed shows that BTI-322 did not compete with GT2 and D66 for CD2 binding sites under the conditions of our assays, suggesting that these three mAb are directed at distinct epitopes of the CD2 molecule.This allowed us to examine the effect of BTI-322 on CD2-induced T cell proliferative responses.T cells optimally stimulated to proliferate by GT2 ϩ D66 exhibited significant cell death when exposed to 1 g/ml BTI-322 from the beginning of culture (47% propidium iodide-permeable cells vs 21% in cells exposed to control Lo-DNP-57 mAb) (Fig. 7B).However, substantial numbers of activated cells resisted apoptosis, as judged by   forward scatter analysis, and by the enhancement of [ 3 H]TdR incorporation (Fig. 7, B and C).When T cells were stimulated by 2.5 g/ml PHA-P, a mitogenic lectin that binds directly to CD2 (43,44) but little to CD3⑀ (45), BTI-322 also favored T cell expansion, although a portion of the cells succumbed to apoptosis (Fig. 7C).T cell proliferation was not decreased in these conditions (not shown).The possibility still existed that PHA had induced resistance to BTI-322-induced apoptosis by binding simultaneously to the TCR and to costimulatory molecules such as CD28.Addition of anti-CD28 mAb during the primary activation of T cells via the CD3/TCR complex can in fact prevent activated T cells from undergoing apoptosis following re-cross-linking of this complex (46).However, adding an anti-CD28 mAb along with BTI-322 to anti-CD3-stimulated cultures did not rescue activated T cells from the lethal effect of BTI-322 (Fig. 7C).Thus, in accordance with the above mentioned studies (43)(44)(45), it is likely that PHA was mimicking an anti-CD2 stimulation.On the whole, our data indicate that apoptosis induced by BTI-322 in anti-CD2-stimulated T cells does not concern all activated cells and that it is efficiently counteracted by simultaneous enhancement of T cell divisions.Such is not the case for T cells activated through the CD3/TCR which exclusively respond to BTI-322 application by apoptotic death.

Removal of BTI-322 from mixed lymphocyte cultures allows the surviving T cells to respond to third party stimulation and to soluble Ag stimulation
Contrasting with the almost total elimination of CD3 ϩ cells induced by BTI-322 in OKT3-stimulated cultures, oligostimulation of T cell clones during an MLR performed in the presence of BTI-322 left most CD3 ϩ cells intact.(Fig. 8A).This is in agreement with the observation that unstimulated T cells are not susceptible to the lethal effect of BTI-322 (see above).Up to 100 ng/ml concentration, BTI-322 did not induce significant downregulation of CD2 molecules at the surface of resting T cells as judged from anti-T11 1 fluorometric staining (not shown).It has been reported that BTI-322 is capable of almost completely inhibiting the proliferation of T cells during a primary MLC and that the surviving cells manifest profound proliferative unresponsiveness on restimulation through the CD3/TCR (35).Given the potent apoptotic effect of BTI-322, it is likely that in these experiments, trace amounts of BTI-322 mAb (not detectable by FACS analysis) were still attached to the cells, inducing the apoptotic death of T cells, once they were activated.To prevent the carryover of BTI-322 mAb in such a system, we washed extensively the T cells recovered from a primary MLR performed in the presence of BTI-322 (100 ng/ml) and allowed them to rest for 2 days.Under these conditions, control T cells responded to the primary stimulatory FIGURE 7. BTI-322 cannot inhibit T cell proliferation induced by mitogenic CD2 mAb or PHA-P.A, Resting T cells were incubated with GT2 and D66 mAb for 0.5 h in ice and then subsequently incubated with 1 g/ml BTI-322 and FITC-labeled mAb (MARG 26 ) specifically recognizing rat IgG.B, 1 ϫ 10 5 T cells stimulated for 4 days with GT2 ϩ D66 (1 g/ml) in the presence or absence of 1 g/ml BTI-322 and then stained with propidium iodide to show the percentages of dead cells.C, [ 3 H]TdR incorporation of T lymphocytes stimulated for 4 days with 1 g/ml GT2 ϩ D66, or T cell numbers obtained after a 3-day stimulation with 2.5 g/ml PHA or after a 4-day stimulation with anti-CD3 (250 ng/ml) and anti-CD28 mAb (2 g/ml) plus IL-2.BTI-322 (1 g/ml) or 1 g/ml isotype-matched Lo-DNP-57 mAb (control cells) were added to the cultures.cells while BTI-322-treated T cells remained silent.Yet, they responded to third party stimulatory cells, although with a delayed kinetics and with decreased magnitude (Fig. 8B).They also responded to PPD or tetanus toxoid presented by autologous monocytes to the same extent as control T cells (Fig. 9), confirming that BTI-322 had specifically eliminated the alloreactive T cells proliferating in primary MLC.

Discussion
Previous studies have demonstrated that CD2 molecules are capable of transducing apoptotic signals to activated T cells, provided they are subjected to at least two CD2 mAb directed at distinct epitopes of the CD2 molecules (30 -32).It was shown that CD2mediated apoptosis of peripheral T cells results in the activation of p56 lck (33) and that CD2 must be in a CD2R ϩ conformation (i.e., express the CD2-restricted epitope) to function as an apoptosistransducing molecule (47), as is the case for the proliferation signals delivered by mitogenic CD2 mAb pairs (48,49).In the present study, we show that triggering of a single CD2 epitope recognized by BTI-322 mAb results in potent apoptotic signaling in T cells activated through the CD3/TCR, leading to subsequent T cell unresponsiveness.BTI-322 did not compete for CD2 binding with other mAb such as GT2 and D66 (this study) or with T11 1 mAb (not shown).The definition of the epitope it recognizes still awaits elucidation.However, unlike many other CD2 mAb tested thus far, which had to be in pairs and at high concentration (1 g/ml), less than 5 ng/ml of this single Ab was sufficient to almost completely eliminate T cells polyclonally stimulated with CD3 mAb when added at the beginning of culture.In experiments not shown, we noticed that after exposure to BTI-322, resting T cells, which were initially unsensitive to the lethal effect of BTI-322, underwent apoptosis on subsequent anti-CD3 stimulation (not shown).This is probably due to the few BTI-322 molecules, not detectable by FACS analysis, that were carried over by these cells (extensive washings followed by a 2-day rest indeed prevented them from undergoing apoptosis).Interestingly, the BTI-322 killing effect was most effective when T cell activation was induced through the CD3/TCR complex.In experiments in which BTI-322 was added to T cells in combination with a mitogenic CD2 mAb, the killing effect was successfully counteracted by simultaneous enhancement of T cell proliferation.The same was true when T cells were simultaneously exposed to BTI-322 and to PHA-P, a lectin that preferentially binds to CD2.The different outcomes of BTI-322 epitope binding, depending on whether the primary stimulus was derived from CD3 or from CD2, may rely in differences in the signal transduction pathways derived from these receptors.CD2 is structurally close to the CD3/TCR complex (50) and to other membrane molecules such as the CD45 phosphatase (51).Although it shares common signal transduction events with the CD3/TCR complex, including tyrosine phosphorylation of phospholipase C␥-1 and a strong calcium response (14,52), it does not induce the phosphorylation of the CD3 chain (53) and does not recruit ZAP-70 to CD3 (54).CD2 activates p56 lck (55) with which it is physically associated (8), as well as the GTPase-activating protein-associated p62 protein which might act as an anchor molecule (54).One can therefore conceive that the signaling pathway initiated by the binding of BTI-322 to its epitope may interfere with the signaling pathway initiated through the CD3/TCR, resulting exclusively in the activation of the apoptotic program of the cells.When interfering with mitogenic signaling through CD2, BTI-322 may enhance the proper association of the proline-rich cytoplasmic tail of CD2 with its associated molecules, perhaps by inducing some change in the conformation of CD2.This may enhance T cell proliferation or induce apoptosis, depending on the ratio of proapoptotic and antiapoptotic molecules such as those of the Bcl-2 family, contained in the cells.
CD2-negative NK cells were apparently spared by BTI-322.CD2 does not transduce apoptotic signals in peripheral NK cells while CD16 (an Fc␥RIII) is quite effective in this respect (56).CD2-positive NK cells might have killed each other by Ab-dependent cell-mediated cytotoxicity after engagement of CD16 by BTI-322 and/or might have been subjected to apoptosis after CD16 engagement.Whatever, CD2-negative NK cells with normal functions can be generated in vitro from CD34 ϩ hemopoietic progenitor cells contained in peripheral blood (57) or from a preexisting subset of CD16 ϩ CD2 Ϫ peripheral NK cells (58), provided IL-2 (which is supplied exogenously in our experiments) is present.
Preliminary studies have suggested that BTI-322 can induce in vitro specific alloantigen unresponsiveness (36).Our own study corroborates this finding, demonstrating that T cells surviving a primary MLR performed in the presence of BTI-322 no longer responded to the primary set of alloantigens while they proliferated in response to third party cells.In view of our results, one can assume that the primary alloreactive T cells had been eliminated by apoptosis.However, the surviving T cells proliferated significantly less to third party allogeneic stimulation than did control T cells, despite extensive washings of the cells.It is possible that primary and secondary stimulatory cells were sharing multiple alloantigens, resulting in the elimination of the corresponding T cell clones.This possibility is supported by the observation that T cells recovered from primary MLC responded to PPD or tetanus toxoid to the same extent as control T cells.Also, using more restricted proliferation systems, Xu et al. 4 have shown that BTI-322 specifically deletes T cells activated by anti-V␤8, leaving intact the proliferative responsiveness of V␤13.Earlier studies have established that nondepleting CD2 mAb can prolong allograft and xenograft survival in animal models (25,27,28,59), and it has been speculated that immunosuppression and tolerance induction may depend on the perturbation, linked to antigenic down-modulation of CD2, on the initial cellular interactions (25,60).True negative signals can be delivered in vitro to T cells by CD2 mAb (22,61) or by LFA3TIP (23) leading to the abrogation of Ag-specific responses in situations where T cell activation does not depend on CD2/CD58 interactions (23).Accordingly, in vivo prolongation of cardiac allografts has recently been obtained in primates by using LFA3TIP (29).The effects of this reagent did not rely on the apoptotic death of activated T cells (29).Our data suggest that manipulation of CD2 to trigger apoptosis in human activated T cells through BTI-322 epitope (the definition of which still awaits elucidation) is a promising strategy to target CD2 for induction of transplant tolerance or treatment of allograft rejection.BTI-322 has been recently utilized in a randomized phase II pilot study for the prevention of rejection in renal allograft recipients, with promising results (62).The preliminary results of this small study suggest that BTI-322 is a safe agent able to significantly reduce (by Ͼ50%) the incidence of rejection when included in a standard immunosuppression regimen, the median time for first rejection being twice as long in the group treated by BTI-322 than in controls.

FIGURE 2 .
FIGURE 2. BTI-322 mAb alone has potent T cell death-inducing capacity.A, DNA synthesis of 5 ϫ 10 4 T cells exposed during 6 days to the indicated CD2 mAb (1 g/ml) and to 100 U/ml IL-2.Values are means Ϯ SD of triplicates cultures.B, 1.5 ϫ 10 5 T cells exposed in culture to 1 g/ml CD2 mAb.Percentages of cells with fragmented DNA (Ͻ2N) were detected by flow cytometry after staining ethanol-permeabilized cells with propidium iodide.

FIGURE 3 .
FIGURE 3. Relationship between T cell death and inhibition of CD3induced T cell proliferation in BTI-322-treated T cells.T cells (1 ϫ 10 5 ) were stimulated for 4 days with soluble OKT3 (250 ng/ml) and 100 U/ml IL-2 in the presence of 1 g/ml CD2 mAb (D66 or T11 1 ), 500 ng/ml BTI-322, or 1 g/ml Lo-Tact-1 mAb.A, T cell proliferation as assessed by [ 3 H]TdR incorporation.B, percent cell death as estimated by trypan blue exclusion and cell morphology.C, Effect of graded concentrations of BTI-322 mAb; absolute numbers of viable cells and percent dead cells were estimated by trypan blue exclusion; the percentage of CD3 ϩ cells was assessed by cytofluorometry.

FIGURE 4 .
FIGURE 4. Phenotypic analysis of cells recovered from primary OKT3 ϩ IL-2-stimulated cultures performed in the presence of BTI-322.T cells were stimulated as in Figure3in the presence or absence of 10 ng/ml BTI-322, thoroughly washed, and recultivated for 6 days in the sole presence of 100 U/ml IL-2.

FIGURE 5 .
FIGURE 5. BTI-322 alone is able to trigger the apoptotic death of activated T cells.A, T cells were stimulated for 4 days with 250 ng/ml OKT3 and 100 U/ml IL-2 and then exposed for 16 h to the indicated mAb at 1 g/ml.Percentages of hypoploid cells were detected by flow cytometry after staining the DNA with propidium iodide (linear scales are represented).One experiment representative of four is shown.B, Ultrastructural changes associated with BTI-322-associated cell death.

FIGURE 6 .
FIGURE 6. BTI-322-induced apoptosis occurs at low concentrations and is independent of the Fas-Fas-L system.A, Nylon wool-purified T cells activated for 4 days with OKT3 ϩ IL-2 and then subjected for 16 h to graded doses of BTI-322 in the presence of 10% monocytes (ⅷ).⅜, Effect of BTI-322 on T cell preparations treated twice with anti-CD14 mAb ϩ complement.‚, T cells exposed to "humanized" BTI-322.OE, T cells exposed to F(abЈ) 2 fragments of BTI-322.*, T cells exposed to isotypematched Lo-DNP-57 mAb.B, Anti-CD3-activated T cells exposed to 100 ng/ml BTI-322.C, Anti-CD3-activated T cells preincubated during 1 h with 10 g/ml M3 or M33 anti-Fas mAb and then exposed for 16 h to 100 ng/ml BTI-322 or to 1/10 diluted yeast supernatant containing recombinant human Fas-L.In all these experiments, dead and apoptotic cells were enumerated by microscopic examination according to the criteria of trypan blue dye inclusion and cell morphology.Experiments are representative of four others (A) and of two others (B and C).

FIGURE 8 .
FIGURE 8. BTI-322 induces alloantigen-specific unresponsiveness.A, cells recovered from a 5-day MLC performed in the presence or absence of 100 ng/ml BTI-322.Expression of CD3 expression was assessed by double immunofluorescence staining.B, kinetic of the secondary alloproliferative responses of T cells recovered from a primary MLC performed in the presence of BTI-322 (100 ng/ml).

FIGURE 9 .
FIGURE 9. T cells cultured with BTI-322 in a primary MLC respond to soluble PPD and tetanus toxoid antigens in secondary cultures.Cells recovered from a 6-day MLC performed in the presence or absence of 100 ng/ml BTI-322 were washed, rested for 2 days, and then stimulated for 6 days in secondary cultures with PPD (25 g/ml) or with tetanus toxoid (1/250 dilution of the stock solution).The soluble Ags were presented by 10% autologous monocytes.Histograms represent the means Ϯ SD of triplicate values.As in Figure8, BTI-322-treated cells were unresponsive to the primary allogeneic stimulatory cells (not shown).