Interest in antilymphocyte sera (ALS) began at the turn of the century when Metchnikoff (1) and others described their anti-inflammatory properties. Their immunosuppressive activity was brought to light only half a century later by Woodruff and Forman (2), and their potential to prolong allograft survival or to induce some form of tolerance was extensively analyzed in rodent models [reviewed in (3, 4)]. ALS or antithymocyte globulin (ATG) preparations have been used in human transplantation since the late 1960s (5) for the following indications: prevention (induction therapy) and treatment of acute rejection of organ allografts, including steroid resistant rejection (6), treatment of graft-versus-host disease after bone marrow transplantation (7), therapy of aplastic anemia (8), and conditioning of recipients of bone marrow from unrelated HLA-matched (9) or haploidentical-related donors (10).
Despite extensive clinical use, the pharmacology and mechanisms of action of ATGs in vivo remain mostly unknown. Up to 23 antibody specificities have been identified in ATG (11–13) mostly against non-T cell-specific antigens. Our group (14) has demonstrated that ATG-induced lymphocytopenia may result from several mechanisms. In vitro, at high concentrations (>100 μg/ml), polyclonal ATGs trigger the classical complement pathway resulting in lymphocyte lysis. In contrast, even at low concentrations, ATGs induce Fas (CD95) and Fas-ligand expression resulting in Fas/Fas-L-mediated apoptosis of activated T cells (14). It has also been suggested by Merion et al. (15) that ATG treatment could lead to T-cell anergy and to the down-modulation of T cell functional molecules (15). However, clinically, the main documented effect of ATG is a massive T-cell depletion in the blood. This depletion may be long-lasting in organ transplantation, and a relative defect in CD4+ and increase in CD8+ CD57+ cells can be observed several years after ATG treatment (16). Thus, the possibility of an effect on thymus function may be considered. Possible depletion of T cells in peripheral lymphoid tissues had been documented in earlier studies in the mouse (17, 18) but has not been investigated in experiments conducted with ATG in a nonhuman primate model (19–21).
The present study was designed to investigate the changes in lymphoid tissues induced by rabbit ATG monotherapy in a nonhuman primate model (cynomolgus monkey). ATG was given at three dosages (low, high, and very high). In addition to the effect of ATG on skin and heart allograft rejection, access of ATG to peripheral lymphoid tissues and thymus, cell depletion and functional alterations of the remaining T cells were investigated. The data demonstrate a dose-dependent T-cell depletion in lymph nodes and spleens from treated animals without significant changes in the thymus. In lymph nodes, apoptosis was shown to contribute to this massive depletion.
MATERIALS AND METHODS
Male and female 2- to 3-year-old cynomolgus monkeys (Macaca fascicularis), weighing from 2.5 to 5 kg and being the offspring of wild-born parents, were used as recipients. All procedures were performed in accordance with the guide for the care and use of laboratory animals by competent persons in accredited facilities (Biomatech, Chasse sur Rhône, France). Each animal received individual care for the duration of the study and was treated with prophylactic antibiotics every 2 days from day 0 to day 14. Each animal manipulation (administration of ATG, blood or tissue sampling, surgery) was performed under anesthesia with Zoletil (zolazepam; 15 mg/kg; Virbac S.A., Cannes, France) + atropine (0.01 mg/kg; Aguettant, Lyon, France).
ATG preparation, doses injected, and treatment schedule.
The rabbit ATG preparation used in this study was Thymoglobuline (batch number 98 TMG 0550; IMTIX-SangStat, Lyon, France), a ready to use liquid solution stored at +2°C to +8°C containing 5.71 mg/ml of purified IgG.
Treatment was performed by intravenous infusions of 1 mg/kg (low dose: LOD), 5 mg/kg (high dose: HID), or 20 mg/kg (very high dose: VHID). Doses were calculated by considering the body surface area rather than the body weight, the ratio being 3/1 for human to macaque (3000 cm2/4 kg). They were calculated from the weight of the animal at day −1 and remained unchanged. Calculation of equivalence in humans is detailed in the discussion section. ATG diluted in 50 ml of Ringer’s lactate (Aguettant) was injected intravenously at a rate of 10 ml/kg/hr via the saphenous vein by means of a catheter and an automatic syringe for the first 3 doses. After this, intravenous injections were performed quickly (over 5 min) after dilution in 10 ml of Ringer’s lactate. The injection sites were shaved mechanically and disinfected with iodine alcohol. Animals in these experiments were distributed as follows. The control group comprised 10 animals. There were 12 monkeys in the LOD group and 16 in the HID group, which received intravenous injections scheduled at day −1/day 0 (the day of surgery)/day 1 and then according to the following schedule (day 3/day 6/day 8/day 10/day 13) corresponding to 8 doses within 2 weeks. The VHID group comprised seven animals and was subdivided in two. Three VHID monkeys were treated according to the above described long protocol (lVHID) and the remaining four animals were treated according to a short VHID protocol (sVHID) comprising two intravenous injections scheduled at day −1 and day 0. Fourteen of these animals (3 controls, 3 LOD, 3 HID, 2 sVHID, and the 3 lVHID) were killed at day 6 for the purpose of collecting their thymuses. All other monkeys were killed after the rejection of their allografts.
Skin grafting and heterotopic heart transplantation.
Animals were typed for MHC A/B/DR mismatching by serology at the Biomedical Primate Research Center (M. Jonker, BPRC, Rijswyk, Netherlands). The selection criteria were ABO compatibility for the heterotopic heart grafts but not for the skin grafts and at least 1 DR and 1 A or B locus mismatch between donor and recipient. Animals were infused with Ringer’s lactate glucose (20 ml of a 30% glucose solution in 500 ml) at a rate of 10 ml/kg/hr during surgery. Twenty animals (five controls, five LOD, eight HID, and two sVHID) received skin allografts. Each recipient received four skin allografts from two different donors according to Balner’s protocol (22). The fourteen animals killed at day 6 received skin allografts according to the same protocol. Eleven monkeys (two controls, four LOD, and five HID) received heterotopic heart implantations performed as already described (23, 24).
A complete follow-up of allograft survival was established for those 31 animals. The date of complete rejection of skin allografts was determined by the presence of a scabby aspect, or brown coloring with loss of flexibility, or complete necrosis of the epidermis. Heterotopic heart graft survival was determined by palpation of the abdomen and by recording an electrocardiogram (Windograf, Gould Instruments S.A., Longjumeau, France) every day. A non-beating heart, as determined by palpation and confirmed by electrocardiogram, was considered to be rejected.
FACS analyses on blood and tissue samples.
Blood samples were collected, before ATG injection, at the indicated times on the figures. On the day of death of the animals scheduled for thymus samplings, three axillary lymph nodes and a spleen punch were also collected on each of these animals.
To evaluate the relative quantity (μg/ml) of ATG bound to various cells, 50 μl of heparinized blood (or cells from lymphoid tissues) was incubated with 20 μl of appropriately diluted donkey anti-rabbit IgG (DARIG) coupled to fluorescein isothiocyanate (FITC) (Jackson Immunoresearch, West Grove, PA) for 30 min at 4°C. To determine the presence of rabbit antibodies, dilutions of monkey sera were applied on 50 μl of fresh control monkey blood for 30 min at 4°C. After washing, cell-bound antibodies were detected by incubation with DARIG-FITC. Following red blood cell lysis in a specific buffer (NH4Cl 1.53 M, EDTA 50 mM, KHCO3 0.1 M), the remaining leukocytes were fixed and subsequently analyzed by FACS analysis after electronic gating of lymphocytes, monocytes, and neutrophils windows. In parallel, following the same procedure, a mean standard curve was established on control monkey blood cells incubated in vitro with known concentrations of ATG as detailed in another report (25). Estimating the amount of bound and free ATG was then done by plotting the median of fluorescence intensity (MedFI) obtained from monkey experiments against that obtained with known concentrations of ATG. A similar procedure was applied for the determination of the ATG binding to erythrocytes and platelets.
To determine in vivo ATG-induced down-modulation of lymphocyte surface markers, monkey peripheral blood mononuclear cells were isolated by Ficoll gradient centrifugation. After extensive washing, cells were incubated for 30 min at 4°C with fluorochrome-coupled monoclonal antibodies (mAbs) directed against rhesus monkey CD3 (Clinisciences, Montrouge, France) and human CD2, CD4, CD8, CD20, CD56 (Becton Dickinson, Pharmingen, Le Pont de Claix, France) previously selected for their cross-reactivity with cynomolgus antigens. After fixation, cells were submitted to FACS analysis to measure the percentage of stained cells and the intensity (MedFI) of labeling.
Lymphoid tissues were pounded in a mortar in a determined volume of RPMI 1640 (ICN, Orsay, France) supplemented with 10% heat inactivated fetal calf serum (FCS; Dutcher, Brumath, France) and counts of viable cells were determined. The same analyses described above were performed on cells extracted from lymphoid tissues.
All FACS analyses were performed under identical settings of photomultipliers and electronic gating of each cell type after calibration with fluorescent beads (Fluorospheres DAKO, Copenhagen, Denmark).
Rabbit IgG (ELISA) and anti-rabbit IgG antibody response.
Rabbit IgG were detected by an immunoenzymatic assay. Ninety six-well microtiter plates were coated with 1 μg/ml of goat anti-rabbit IgG (CAPPEL-Flobio, Asnières, France). After incubation of monkey sera, immune complexes were revealed by alkaline phosphatase-conjugated goat anti-rabbit IgG F(ab′)2 (Immunotech, Luminy, France) diluted 1/10,000 followed by development with 1 mg/ml p-nitrophenyl phosphate (Sigma, La Verpillère, France) in a diethanolamine buffer (diethanolamine 1 M, MgCl2 0.05 mM, NaN3 15 mM; pH 9.8) for 15 min. Absorbance was measured at 405 nm using an ELISA plate reader. The same procedure was applied to detect anti-rabbit antibodies, except that plates were coated with diluted ATG and complexes were detected with peroxidase-conjugated goat anti-monkey IgG (Nordic Immunological Labs, San Clemente, CA) diluted 1/10,000.
Determination of apoptotic cell death in axillary lymph nodes.
Axillary lymph nodes were surgically sampled, 1 hr after the end of ATG infusion at day 3 on 15 animals (6 controls, 3 LOD and 6 HID) and at day 0 on 4 VHID monkeys. Lymph nodes were pounded in RPMI-10% FCS, filtered, and incubated for 30 min at 37°C on a nylon column followed by a 2-hr incubation in 100-mm dishes in the same culture medium to eliminate macrophages and monocytes. Cells were then further incubated for 4 hr at 37°C in a 5% CO2 atmosphere to allow cells that had received a death signal to enter apoptosis. Annexin V (Boerhinger Mannheim, Meylan, France) staining was then performed according to the manufacturer’s procedure followed immediately by FACS analysis (26, 27). TUNEL (terminal deoxynucleotide transferase-mediated dUTP nick-end labeling) analysis using the ApopTag Peroxidase kit (Appligene, Illkirch, France) was also performed after neutralization of endogenous peroxidase activity on paraffin sections of 3.7% formaldehyde-fixed axillary lymph nodes sampled under the conditions described above.
Samples of lymph nodes, spleen, and thymus were frozen embedded in Tissue-Tek (Sakura). Five-micrometer sections were performed using a cryomicrotome (Leica) and frozen (−20°C) after overnight drying. Before processing, they were fixed in acetone for 10 min. After rehydration in PBS, tissue sections were saturated with 2-10% human sera diluted in PBS-1% BSA. Slides were then incubated for 30 min with a biotinylated polyclonal anti-rabbit antibody (Jackson Immunoresearch), washed and further incubated with peroxidase labeled streptavidin (Vector Laboratory, Burlingame, CA). Labeling was then revealed with VIP substrate (Vector). Slides were counterstained with hematoxylin and lithium carbonate and mounted in glycerol gelatin mounting medium (Sigma, St. Louis, MO).
Mixed lymphocyte reactions.
Lymphocytes isolated from monkey lymph nodes as described above were cultured at 105 cells/well with mitomycin B-treated RPMI-1886 human B cells (104 cells/well). After 4 days, cells were pulsed with 0.5 μCi/well [3H] thymidine for 12 hr. Radioactive incorporation was measured by standard liquid scintillation counting and results are given in counts per minute.
The homogeneities of variances have been tested using the analysis of variance test with ATG treatment as the variable. Whenever the influence of ATG treatment was declared significant by this test, respective means were compared with each other by the Scheffé test (28). Graft survival times were compared by using the log-rank test.
Changes in peripheral blood cell counts.
No significant differences were observed between the different groups after ATG treatment concerning red blood cell counts whose decrease was probably due to the repeated blood samplings (Fig. 1a). Platelet counts remained stable, except in the long VHID treatment (Fig. 1b). Large variations in the number of neutrophils were observed in control monkeys but less in the LOD and HID groups. The short VHID protocol resulted in a neutrophil depletion reaching 70% at day 16 with a return to normal neutrophil counts by day 20 and the long VHID treatment in a manifest neutropenia (Fig. 1c). The number of blood monocytes seemed to increase in control monkeys until day 8 and this was further enhanced by the LOD treatment. Conversely, the short VHID treatment had a pronounced depletive effect on these cells at day 0 with a return to normal by day 7, whereas the depletion of blood monocytes was still visible at day 6 in the long VHID group (not shown).
ATG treatment induced a transient dose-dependent lymphocytopenia (Fig. 1d) with a drop in each lymphocyte subset analyzed (Fig. 1, e–i). Only CD4 (Fig. 1g) and CD56 (not shown) lymphocytes were not significantly decreased by the LOD treatment. Of interest, the peripheral blood T-cell counts of monkeys in the short and long VHID protocols dropped to zero within 2 hr after the first ATG infusion and remained low until day 3 for the animals included in the short VHID treatment and the day of death for those of the long VHID protocol. Only B cells were not totally depleted during this period of time. However, lymphocyte reconstitution seemed to be faster in the short VHID group compared with the HID group.
These results demonstrate that this ATG treatment resulted in a significant decrease of PBLs on a dose-dependent basis, without major depletion of other cell types, except for neutrophils and platelets in both VHID groups.
Kinetics of free antibodies and rabbit IgG in monkey sera.
The pharmacokinetics of ATG in monkey sera was analyzed by two different methods. Levels of rabbit IgG assessed by ELISA reached a maximum at day 6 (day 0 for the VHID) and then returned to undetectable levels by day 13, whatever the dose administered (Fig. 2A, a). This clearance was due to the immunization of the animals, as demonstrated by the presence of anti-rabbit IgG antibodies by day 8 in the sera of all treated animals (Fig. 2A, a). The total level of free specific antilymphocyte antibodies was assessed by FACS analysis. MedFIs obtained in vitro with known concentrations of ATG were used to estimate an equivalent concentration of these circulating antilymphocyte antibodies. Results show that the kinetics of specific antilymphocyte antibody levels was comparable to that of total rabbit IgG (Fig. 2A, b).
In vitro analysis of the binding activity of ATG revealed that antibodies contained within this reagent are directed against all blood cell types (Fig. 2B, a) with decreasing labeling intensities in the following order: T cells > B and natural killer (NK) cells (not shown) > monocytes and neutrophils > platelets > erythrocytes. By comparison, sera collected from HID-treated monkeys at day 6 contained relatively lower amounts of antibodies to neutrophils and monocytes, while anti-red blood cells (RBC) and anti-platelet antibodies had been totally absorbed in vivo (Fig. 2B, a and b). Furthermore, the same sera were relatively enriched in antibodies that bind to CD3+ cells and partially depleted from antibodies that stain CD20+ cells as compared with the original ATG preparation (data not shown).
Taken together, these data demonstrate that a strong anti-rabbit IgG response results in the clearance of free antibodies present in serum at days 8-11 and the neutralization of antibodies injected beyond this time. Before immunization, cross-reactive antibodies directed against neutrophils, platelets, and RBC were rapidly absorbed in vivo and could be demonstrated by DARIG-FITC staining on blood cells from treated monkeys (data not shown).
Lymphocyte depletion in peripheral lymphoid tissues.
Thymuses, axillary lymph nodes, and spleens were collected at day 6; lymphocytes extracted from these organs were counted and their phenotype analyzed by FACS. As shown in Figure 3 (a–d), there was a dose-dependent T-cell depletion in axillary lymph nodes and spleens. Depletion of B cells (Fig. 3e) and NK cells (Fig. 3f) occurred only in the short and long VHID protocols. No significant depletion of thymocytes was observed and the ratio of double-positive (CD4+CD8+) cells to that of single-positive cells remained unchanged (data not shown).
Determination of apoptotic cell death in axillary lymph nodes.
Earlier reports mentioned the occurrence of tingible bodies and pyknotic nuclei in lymph nodes from rodents treated with ALS (17, 18). To assess the contribution of apoptosis to lymphocyte depletion, cells from axillary lymph nodes obtained 1 hr after ATG infusion were processed for the determination of membrane phosphatidyl serine externalization by FACS using Annexin V as a specific marker. Figure 4A indicates that there was a dose-dependent increase in the percentage (12%, 30%, and 38% for LOD-, HID-, and VHID-treated monkeys, respectively) of apoptotic lymphocytes (Fig. 4B) No differences could be evidenced between the two VHID protocols. In parallel, TUNEL studies performed on paraffin-embedded sections of axillary lymph nodes also showed an increased number of cells with double-strand DNA breaks in the deep cortex (data not shown).
ATG binding to lymphoid cells in vivo.
The amount of rabbit antibody bound to lymphocytes from thymuses, lymph nodes, spleens, and peripheral blood was measured by FACS analysis (Fig. 5A). PBLs were the most importantly coated cells, with an antibody density equivalent to that achieved by 6 μg/ml of ATG in the HID protocol (data not shown). The level of cell-bound antibody paralleled that of free antibody in plasma in the LOD protocol, whereas an excess of free versus cell-bound antibodies was observed in the HID series. In both VHID-treated animal groups, the magnitude of PBL depletion precluded any analysis. Cell-bound antibodies dropped to baseline levels between day 8 and day 11 in parallel with the immune clearance of plasma antibodies after anti-ATG antibody response (Fig. 2A). Lymphocytes from lymph nodes and spleens were less coated than PBLs, except in short VHID-treated monkeys where the coating of spleen lymphocytes was important (Fig. 5A). In contrast, a very weak coating was observed on thymocytes, although the fluorescence intensity of these thymocytes was similar to that of PBL, lymph node, and spleen lymphocytes upon addition of a saturating dose of ATG (50 μg/ml) in vitro, thus excluding a global down-modulation of ATG binding epitopes (Fig. 5A). Immunohistological analyses confirmed these observations (data not shown).
Down-modulation of T-cell surface antigens.
The expression of many cell surface molecules is transiently down-modulated after incubation with a ligand or a specific antibody (29–33). In vitro, our ATG preparation induced the down-modulation of CD2, CD3, CD4, and CD8 with dose-dependent kinetics and magnitude (Fig. 6A). The occurrence of modulation in vivo was investigated on cells that had escaped depletion. Remaining circulating T cells from treated animals had down-modulated CD3, CD4, CD8, and CD2 surface antigens (Fig. 6B, a–d, respectively). This phenomenon was maximal around day 6 and it was, to some extent, correlated with the dose of ATG injected. In comparison, CD20 and CD56 molecules exhibited variations of their respective surface densities regardless of treatment (data not shown). Similarly to PBLs, lymphocytes from lymph nodes and spleens had also down-modulated the surface expression of CD3, CD4, CD8, and CD2 molecules in a dose-dependent manner at day 6 (Fig. 6C, a–d), whereas the density of CD20 and CD56 molecules was not modified (data not shown). No change of CD4 and CD8 surface expression was observed on thymocytes.
Immunosuppression by ATG treatment.
Rejection of both skin and heart allografts (Fig. 7A) was delayed in a dose-dependent manner. Skin allografts in nontreated monkeys had a median survival time of 9.25 days, whereas LOD- and HID-treated monkeys had an extended graft survival (median survival time: 13 and 22 days, respectively). The median survival of the HID group was significantly different from that of the control group (P <0.001) and that of the LOD series (P <0.05, log-rank test). In the short VHID-treated animals, skin grafts were rejected at 22 and 26 days. Heart allograft median survival time was 8.5 days in controls, 12.5 in LOD-treated, and 17 days in HID-treated monkeys. In this case, the median survival of the HID group was significantly different from that of the control group (P <0.01) (Fig. 7A). Allograft survival was correlated to the magnitude of blood lymphopenia during the first week of ATG treatment (Fig. 7B).
We further analyzed the functional consequences of ATG treatment on lymph node cells that had escaped depletion. Lymph node lymphocytes sampled at day 6 from HID-treated monkeys had a decreased ability to proliferate in mixed lymphocyte reaction against the RPMI-1886 human B cell line (Fig. 7C). ATG coating, decreased T/B ratios, and down-modulation of CD3, CD4, CD8, and CD2 may contribute to the decreased MLR responsiveness of the remaining lymphocytes.
The main objective of this study was to gain further insight into the mechanisms of action of ATG in a nonhuman primate model, with special emphasis on the access of antibodies to lymphoid tissues and the effect of ATG dosage. The dose equivalence between cynomolgus monkeys and humans was calculated (i) according to the threefold difference in body surface versus body weight as detailed in Materials and Methods and (ii) taking into account the level of antigenic cross-reactivity between lymphocytes from the two species. To this end, cynomolgus and human lymphocytes were coated with Thymoglobuline over a wide range of concentrations, and the amount of bound antibodies was determined by FACS analysis after addition of FITC-DARIG. Thymoglobuline concentrations required to achieve identical levels of binding were twice higher with cynomolgus than with human lymphocytes. From these data, we concluded that equivalent doses in humans were six times lower than those administered in the monkeys. The low dose (LOD, equivalent to 0.15–0.20 mg/kg/day in humans) is not currently used clinically, but it was sufficient to induce borderline immunosuppressive effects. The high dose (HID, equivalent to 0.8–1.0 mg/kg/day in humans) reflects the ATG dosage currently used in organ transplantation (34, 35). The very high dose (VHID, equivalent to 3.5 mg/kg/day) is comparable to that administered as part of a conditioning regimen in recipients of stem cell allografts from haploidentical familial donors (10, 36) or from HLA-matched unrelated donors (9). In the latter context, ATG is expected to contribute to T-cell depletion in the recipient, especially in protocols with decreased doses of total body irradiation or without irradiation, and also to deplete in vivo the donor cell inoculum from mature T cells. In renal transplantation, but not in other clinical applications, ATG dosage is routinely adjusted to CD2+ or CD3+ cell counts in peripheral blood (37).
In the absence of associated immunosuppressive drug, ATG induces a strong antibody response resulting in complete neutralization of ATG injected beyond day 8 (Fig. 2). Therefore the reported effects were achieved before the occurrence of immunization, that is after five ATG infusions in the LOD and HID protocols and two ATG infusions in the sVHID protocol.
The present study demonstrates differences in the consequences of ATG treatment between peripheral blood and lymphoid organs (Figs. 3 and 5). It is noteworthy that no significant alteration of thymocytes (cellularity, proportion of double-positive versus single-positive CD4+ or CD8+ cells) was observed whatever the dose of ATG administered, whereas the ATG coating of thymocytes remained minimal, except in the VHID protocol (Fig. 5). At marked difference with the thymus, T-cell depletion and ATG coating of the remaining T cells in peripheral lymphoid organs indicated a direct ATG T-cell depleting effect on spleen and on lymph node cells. From these data, one may assume that a dose-dependent dynamic process occurs, involving absorption of ATG antibodies on blood cells and diffusion of rabbit IgG and unabsorbed antibodies to the extravascular spaces. In the intravascular compartment, cells coated beyond a certain threshold of antibody density (equivalent to 6 μg/ml ATG in this study) would then disappear, whereas cells coated at lower antibody density (e.g., B and NK cells) would escape depletion. Of note, antibody coating remained homogenous in spleen and lymph nodes and lower than in peripheral blood, indicating that lymphocytes that circulate in the intravascular compartment no longer migrate to peripheral lymphoid organs once coated by ATG. In this respect, our data confirm that T-cell counts in peripheral blood cannot be regarded as a fair indicator of T-cell depletion in peripheral lymphoid tissues (38) because only the highest ATG dosage (VHID) associated with rapid and complete disappearance of all blood T cells induced maximal yet incomplete (about 85%) T-cell depletion in peripheral lymphoid tissues.
The mechanisms of peripheral T-cell depletion induced by ATG treatment are complex and still only partly deciphered. Complement-dependent lysis that occurs in vitro at ATG concentrations greater than 100 μg/ml (14) may transiently operate in blood during HID and VHID protocols. Opsonization with subsequent phagocytosis by liver, spleen, and lung macrophages may contribute to eliminate blood cells coated at intermediate ATG density. However, the most important and rather unexpected depleting mechanism demonstrated by this study is lymphocyte apoptosis in peripheral lymphoid tissues but not in the thymus (Fig. 4). Apoptosis was documented by an externalization of membrane phosphatidylserine (26, 27) and by the presence of DNA breaks in the deep cortex of lymph nodes. In vitro, ATG triggers Fas-ligand (CD95L) and Fas (CD95) expression both at submitogenic and mitogenic concentrations (14), but only preactivated T cells that have received an interleukin 2 signal are susceptible to this apoptotic pathway (39). Although all monkeys in these experiments received skin or heart allografts, a process of “activation-induced cell death”(40) restricted to donor-specific alloreactive T cells can be excluded in view of the high magnitude and early kinetics of the apoptotic response. Whether apoptosis is associated with or dependent on T-cell activation (41), whether it requires CD2, CD3, anti-TCR or CD5 antibodies, and to which extent it may be modulated by associated treatments (corticosteroids, calcineurin inhibitors...) deserves further investigations. The present data suggest that the magnitude of T-cell depletion in peripheral lymphoid tissues may be related to the peak levels rather than to the cumulative doses of ATG administered. These experimental data validate the use of high dose Thymoglobuline treatments (2.5 mg/kg/day x 4) for achieving T-cell depletion in peripheral lymphoid tissues before hematopoietic stem cell transplantation.
An obvious limitation to ATG dose escalation is the presence of cross-reacting antibodies that entail a risk of hemolytic anemia, thrombocytopenia, and neutropenia. In our experimental model, non-T cell-specific antibodies were rapidly cleared in vivo without obvious clinical consequences because of the very high antigenic mass (erythrocytes) or rapid renewal (neutrophils) of cells bearing cross-reactive or shared epitopes (Fig. 2B, a and b). However, using very high doses, only brief treatments (two infusions) were tolerated, whereas repeated injections resulted in thrombocytopenia and neutropenia. In organ transplantation, ATG-induced neutropenia is primarily of peripheral origin and reversible within 24 hr of ATG withdrawal. ATG induction treatments of short duration (e.g., 3–5 days) and high doses (2 mg/kg) may be considered in organ transplantation to achieve initial T-cell depletion. Dosage adjustment to lymphocyte blood cell counts does not seem useful, but doses should be reduced if marked neutropenia occurs.
In addition to T-cell depletion, ATG treatment was shown to induce functional alterations of the weakly covered remaining T cells. This was demonstrated by impaired proliferative responses of lymph node cells in mixed leukocyte reactions and down-modulation of T-cell surface signaling molecules such as CD2, CD3, CD4, and CD8. Such effects may contribute to the overall immunosuppressive activity of ATG treatment demonstrated by the significant prolongation of skin and heterotopic heart MHC-mismatched allografts in accordance with previously published studies (19, 20) and they may persist beyond treatment withdrawal in the absence of patient antibody response. Whether such functional effects could interfere with T-cell reconstitution after ATG therapy along with associated immunosuppressive treatments in organ transplantation deserves further study in other clinical and experimental models.
In conclusion, ATG monotherapy in cynomolgus monkeys resulted in a dose-dependent T-cell depletion in peripheral blood and to a lesser extent in peripheral lymphoid tissues but not in the thymus. The major mechanism of T-cell depletion is ATG-induced apoptosis in peripheral lymphoid organs. Depletion occurred rapidly (24 hr) and was maximal at very high doses (20 mg/kg). Nondepleted T cells were coated by ATG and functionally altered with decreased expression of several membrane receptors (TCR/CD3) and coreceptors (CD2, CD4, CD8), and decreased responsiveness in mixed leukocyte reaction. The relative contribution of T-cell depletion and functional impairment in the prolongation of skin and allograft survival cannot be precisely delineated. The data support the use of high ATG doses in clinical applications where peripheral T-cell depletion is the main therapeutic objective, but the immunosuppressive or tolerance-enhancing efficiency of such treatments in organ transplantation was not documented in the present study and deserves further investigation.
We are grateful to M. Moyne and J.P. Boutrand (Biomatech) and to J.C. Moulin (Aventis Pasteur). We particularly thank A. Oubenaissa for his surgical skill and help during heterotopic heart transplantations and J. Margonari for her excellent technical assistance in preparing tissue samples for TUNEL analyses. We thank Dr. R. Buffet, Dr. N. Whitaker, R. Buelow, and Pr. JP. Soulillou for critical reading of the manuscript.
1. Metchnikoff E. Etude sur la resorption des cellules. Ann inst Pasteur 1899; 13: 737.
2. Woodruff MFA, Forman B. Effects of antilymphocyte serum on suspension of lymphocytes in vitro. Nature 1951; 168: 36.
3. Taub RN. Biological effects of heterologous antilymphocyte serum. In: Kallos P, Helsinborg Waksman BH, eds. Progress in allergy, Vol 14. Basel: S. Karger AG, 1970: 208.
4. Lance ER, Medawar PB, Taub RN. Antilymphocyte serum. Adv Immunol 1973; 17: 2.
5. Starzl TE, Marchioro TL, Hutchinson DE, Porter KA, Cerilli GJ, Brettschneider L. The clinical use of antilymphocyte globulin in renal homotransplantation. Transplantation 1967; 5(4) (Suppl): 1100.
6. Cosimi AB. Antilymphocyte globulins and monoclonal antibodies. In: Morris PJ, ed. Kidney transplantation: principle and practice. Philadelphia: Saunders, 1988: 343.
7. Storb R, Gluckman E, Thomas ED, et al. Treatment of established human graft-versus-host disease by antithymocyte globulin. Blood 1974; 44 (1): 56.
8. Marsh J, Schrezenmeier H, Marin P, et al. Prospective randomized multicenter study comparing cyclosporin alone versus the combination of antithymocyte globulin and cyclosporin for treatment of patients with nonsevere aplastic anemia: a report from the European Blood and Marrow Transplant (EBMT) Severe Aplastic Anaemia Working Party. Blood 1999; 93 (7): 2191.
9. Ringden O, Remberger M, Carlens S, et al. Low incidence of acute graft-versus-host disease, using unrelated HLA-A-, HLA-B-, and HLA-DR-compatible donors and conditioning, including anti-T-cell antibodies. Transplantation 1998; 66 (5): 620.
10. Aversa F, Tabilio A, Velardi A, et al. Treatment of high-risk acute leukemia with T-cell-depleted stem cells from related donors with one fully mismatched HLA haplotype. N Engl J Med 1998; 339 (17): 1186.
11. Raefsky EL, Gascon P, Gratwohl A, Speck B, Young NS. Biological and immunological characterization of ATG and ALG. Blood 1986; 68 (3): 712.
12. Bonnefoy-Berard N, Vincent C, Revillard JP. Antibodies against functional leukocyte surface molecules in polyclonal antilymphocyte and antithymocyte globulins. Transplantation 1991; 51 (3): 669.
13. Rebellato LM, Gross U, Verbanac KM, Thomas JM. A comprehensive definition of the major antibody specificities in polyclonal rabbit antithymocyte globulin. Transplantation 1994; 57 (5): 685.
14. Genestier L, Fournel S, Flacher M, Assossou O, Revillard JP, Bonnefoy-Berard N. Induction of Fas (Apo-1, CD95)-mediated apoptosis of activated lymphocytes by polyclonal antithymocyte globulins. Blood 1998; 91 (7): 2360.
15. Merion RM, Howell T, Bromberg JS. Partial T-cell activation and anergy induction by polyclonal antithymocyte globulin. Transplantation 1998; 65 (11): 1481.
16. Muller TF, Grebe SO, Neumann MC, et al. Persistent long-term changes in lymphocyte subsets induced by polyclonal antibodies [see comments]. Transplantation 1997; 64 (10): 1432.
17. Gray JG, Monaco AP, Wood ML, Russell PS. Studies on heterologous anti-lymphocyte serum in mice. I. In vitro and in vivo properties. J Immunol 1966; 96 (2): 217.
18. Lance EM. The mechanism of action of anti-lymphocyte serum: studies of antibody eluate. J Exp Med 1969; 130 (1): 49.
19. Thomas JM, Carver M, Scott J, Williams E, Thomas F. Effect of rabbit anti-human thymocyte globulin on lymphocyte subpopulations and functions following allotransplantation in the rhesus monkey. Transplantation 1979; 27 (3): 163.
20. Thomas JM, Williams EA, Thomas FT, Millington GM. T cell-specific activity in rabbit anti-human thymocyte globulin: correlation with immunosuppressive activity in nonhuman primates. Transplantation 1978; 25 (3): 97.
21. Thomas JM, Carver FM, Foil MB, et al. Renal allograft tolerance induced with ATG and donor bone marrow in outbred rhesus monkeys. Transplantation 1983; 36 (1): 104.
22. Balner H, Dersjant H, Van Bekkum DW. Testing of antihuman lymphocyte sera in chimpanzees and lower primates. Transplantation 1969; 8 (3): 281.
23. Ono K, Lindsey ES. Improved technique of heart transplantation in rats. J Thorac Cardiovasc Surg 1969; 57 (2): 225.
24. Schmid C, Binder J, Heemann U, Tilney NL. Successful heterotopic heart transplantation in rat. Microsurgery 1994; 15 (4): 279.
25. Préville X, Nicolas L, Flacher M, Revillard J-P. A quantitative flow cytometry assay for the preclinical testing and pharmacological monitoring of rabbit antilymphocyte globulins (rATG). J Immunol Methods 2000; 245 (12):45.
26. Martin SJ, Reutelingsperger CP, McGahon AJ, et al. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J Exp Med 1995; 182 (5): 1545.
27. Verhoven B, Schlegel RA, Williamson P. Mechanisms of phosphatidylserine exposure, a phagocyte recognition signal, on apoptotic T lymphocytes. J Exp Med 1995; 182 (5): 1597.
28. Scheffé H. The analysis of variance. New York: Wiley, 1959.
29. Borroto A, Lama J, Niedergang F, Dautry-Varsat A, Alarcon B, Alcover A. The CD3epsilon subunit of the TCR contains endocytosis signals. J Immunol 1999; 163 (1): 25.
30. Lin J, Yon RW, Chavin KD, et al. Anti-CD2 monoclonal antibody-induced receptor changes: down modulation of cell surface CD2. Transplantation 1995; 59 (8): 1162.
31. Morel P, Vincent C, Wijdenes J, Revillard JP. Down-regulation of cell surface CD4 molecule expression induced by anti-CD4 antibodies in human T lymphocytes. Cell Immunol 1992; 145 (2): 287.
32. Morel P, Nicolas JF, Wijdenes J, Revillard JP. Down-regulation of lymphocyte CD4 antigen expression by administration of anti-CD4 monoclonal antibody. Clin Immunol Immunopathol 1992; 64 (3): 248.
33. Chatenoud L, Baudrihaye MF, Kreis H, Goldstein G, Schindler J, Bach JF. Human in vivo antigenic modulation induced by the anti-T cell OKT3 monoclonal antibody. Eur J Immunol 1982; 12 (11): 979.
34. Brennan DC, Flavin K, Lowell JA, et al. A randomized, double-blinded comparison of Thymoglobulin versus Atgam for induction immunosuppressive therapy in adult renal transplant recipients [published erratum appears in Transplantation 1999 May 27; 67(10): 1386]. Transplantation 1999; 67 (7): 1011.
35. Gaber AO, First MR, Tesi RJ, et al. Results of the double-blind, randomized, multicenter, phase III clinical trial of Thymoglobulin versus Atgam in the treatment of acute graft rejection episodes after renal transplantation. Transplantation 1998; 66 (1): 29.
36. Slavin S, Nagler A, Naparstek E, et al. Nonmyeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 1998; 91 (3): 756.
37. Abouna GM, al-Abdullah IH, Kelly-Sullivan D, et al. Randomized clinical trial of antithymocyte globulin induction in renal transplantation comparing a fixed daily dose with dose adjustment according to T cell monitoring. Transplantation 1995; 59 (11): 1564.
38. Westermann J, Pabst R. Lymphocyte subsets in the blood: a diagnostic window on the lymphoid system? Immunol Today 1990; 11 (11): 406.
39. Fournel S, Genestier L, Robinet E, Flacher M, Revillard JP. Human T cells require IL-2 but not G1/S transition to acquire susceptibility to Fas-mediated apoptosis. J Immunol 1996; 157 (10): 4309.
40. Kabelitz D, Pohl T, Pechhold K. Activation-induced cell death (apoptosis) of mature peripheral T lymphocytes. Immunol Today 1993; 14 (7): 338.
41. Kishimoto H, Sprent J. Strong TCR ligation without costimulation causes rapid onset of fas-dependent apoptosis of naive murine CD4+
T cells. J Immunol 1999; 163 (4): 1817.