Effect of combined cytostatic cyclosporin A and cytolytic ... - Nature

Gene Therapy (2002) 9, 201–207  2002 Nature Publishing Group All rights reserved 0969-7128/02 $25.00 www.nature.com/gt

RESEARCH ARTICLE

Effect of combined cytostatic cyclosporin A and cytolytic suicide gene therapy on the prevention of experimental graft-versus-host disease S Maury1, E Litvinova1, O Boyer1, L Benard2, S Bruel1, D Klatzmann1 and JL Cohen1 1 2

Biologie et The´rapeutique des Pathologies Immunitaires CNRS/UPMC ESA 7087, Hoˆpital Pitie´-Salpeˆtrie`re, Paris, France; and Service d’Anatomie Pathologique, Hoˆpital Pitie´-Salpeˆtrie`re, Paris, France

The immunosuppressive drug cyclosporin A (CsA) represents the standard preventive treatment of graft-versushost disease (GVHD), the main complication of allogeneic hematopoietic stem cell transplantation (HSCT). However, its efficacy is only partial and many patients develop lethal GVHD despite CsA. A strategy of genetic immunosuppression based on conditional elimination of donor T cells expressing the Herpes simplex type 1 thymidine kinase (TK) suicide gene was recently developed. In this system, ganciclovir (GCV) selectively kills dividing but not quiescent TK T cells. Since CsA is known to have a cytostatic effect on T cells, it could negatively interfere with the division-dependent TK gene therapy. We thus tested whether administration of

CsA would antagonize elimination of alloreactive donor TK T cells mediated by GCV in a murine model of GVHD. In vivo experiments revealed that, contrary to GCV, CsA only transiently controlled alloactivation-induced T cell proliferation, and likewise could not prevent lethal GVHD. When T cells resumed proliferation under CsA, they were however still sensitive to GCV. Survival, as well as immune reconstitution, was excellent in mice treated with GCV alone or in combination with CsA. These observations should help to design improved suicide gene therapy trials in the field of allogeneic HSCT. Gene Therapy (2002) 9, 201–207. DOI: 10.1038/sj/gt/3301637

Keywords: hematopoietic stem cell transplantation; graft-versus-host disease; suicide gene therapy; cyclosporin A; herpes simplex type-1 thymidine kinase; CFSE

Introduction Graft-versus-host disease (GVHD) represents the main and often lethal complication encountered after allogeneic hematopoietic stem cell transplantation (HSCT). The immunosuppressive drug cyclosporin A (CsA), administered during the first 3 to 6 months following HSCT, is part of the standard preventive treatment of GVHD.1 This immunosuppressive regimen is only partially efficient, and 15 to 60% of the patients indeed develop GVHD despite CsA.2 Recent introduction of new immunosuppressive drugs, such as FK506, has not reduced the frequency of severe GVHD.3 T cell depletion of the graft efficiently prevents GVHD but leads to impaired engraftment,4 prolonged immune deficiency causing an increased rate of infections,5 and impaired graft-versus-leukemia effects resulting in an increased frequency of leukemia relapse.6,7 Ex vivo transduction of donor T cells with a suicide gene encoding Herpes simplex type 1 thymidine kinase (TK) offers a new approach for controlling GVHD by treatment with the synthetic nucleoside analog ganciclovir (GCV). TK expression allows the transformation of

Correspondence: D Klatzmann, CNRS/UPMC ESA 7087, Hoˆpital Pitie´Salpeˆtrie`re, 83, bd de l’Hoˆpital, F-75651 Paris Cedex 13, France Received 13 November 2001; accepted 30 November 2001

GCV into a component highly toxic for dividing T cells, a metabolism not occurring in non-transduced cells. When TK T cells divide after alloantigen recognition, they become sensitive to GCV and are consequently killed.8 Thus, suicide gene therapy should permit the selective elimination of T cells that recognize recipient alloantigens, while preserving a large repertoire of T cells which were not dividing during GCV treatment. In preclinical experiments using TK T cells from transgenic mice, we and others have previously demonstrated the feasibility of preventing GVHD by this strategy,9,10 as well as the possibility of controlling ongoing GVHD, when GCV is initiated at the time of clinical signs of GVHD.9 In humans, suicide gene therapy of GVHD has been evaluated in two different clinical situations. One study concerned patients with relapsing hematological malignancy or EBV-induced lymphoproliferation after T celldepleted allogeneic HSCT who were treated with ex vivo TK-transduced donor lymphocyte infusions.11 In three patients who developed GVHD after TK T cell infusion, GCV administration induced two complete and one partial remission. The other study concerned patients receiving transduced T cells, together with a T cell-depleted allogeneic marrow graft.12 GVHD occurred in five patients and treatment with GCV was associated with complete or partial response in four and one patient(s), respectively. Other clinical trials now investigate the ability of the suicide gene system to cure,13 but also to prevent (our unpublished protocol) GVHD.

Cyclosporin A and suicide gene therapy of GVHD S Maury et al

202

Thus, CsA and TK/GCV are two GVHD treatments with different modes of action. Whether they should be used separately or in combination deserves to be analyzed. Actually, CsA could antagonize the therapeutic effect of GCV on TK T cells. Indeed, at least in vitro, CsA is known to inhibit T cell division.14,15 Thus, its association with the TK/GCV system could impair the cell division-dependent elimination of donor TK T cells mediated by GCV. Such an antagonism between a cytostatic agent – CsA – and a division-dependent cytolytic system – TK/GCV – would finally reduce or even suppress the therapeutic benefit of the TK/GCV suicide gene approach. Since CsA represents the standard preventive treatment of GVHD, we assume that for future protocols using suicide gene therapy as GVHD prophylaxis, it will be decided to include CsA. Therefore, in the present study, we investigated the combined effects of CsA and the TK/GCV suicide gene system in a murine model of GVHD. Our conclusions are of clinical relevance for the design of future trials involving the TK/GCV system in the field of allogeneic HSCT.

Results Effect of CsA and/or GCV on the T cell response after allogeneic stimulation in vitro We first investigated a potential antagonism between CsA and GCV by studying their separate and combined effects on allogeneic T cell proliferation in mixed lymphocyte cultures (MLC). For this, responder TK T cells were cultured for 4 days in the presence of allogeneic, irradiated stimulating cells and proliferation was measured by tritiated thymidine incorporation. CsA or GCV alone exerted dose-dependent inhibitory effects on T cell proliferation (Figure 1a and b, respectively). Complete inhibition of proliferation was observed at 1000 ng/ml of CsA and 10 ␮m of GCV. Fifty percent inhibition (IC 50%) was achieved at a concentration of CsA close to 100 ng/ml. For GCV, IC 50% was around 0.5 ␮m, a dosage at least 100-fold lower than that needed to obtain the same effect with non-transgenic T cells. However, it should be emphasized that the consequences of the two treatments were markedly different. When CsA was withdrawn from the culture medium at day 2, cells rapidly resumed proliferation revealing a cytostatic effect on T cells. In contrast, GCV withdrawal led to little or no cell proliferation, reflecting a cytotoxic effect on proliferating TK T cells (data not shown). We then analyzed the combined effect of CsA and GCV on transgenic TK T cells. When a high concentration of CsA (500 or 1000 ng/ml) was added to increasing doses of GCV, we observed an almost complete inhibition of proliferation, whatever the GCV concentration used (not shown). When using a partially inhibiting dose of CsA (100 ng/ml), the inhibition curve depicting the sensitivity of TK T cells to GCV was similar to that observed without CsA (Figure 1b). This indicates that GCV efficiently controlled T cells that still divided upon allogeneic stimulation despite the presence of CsA in the culture, with no detectable antagonism. Effect of CsA and/or GCV on T cell division after allogeneic activation in vivo We next investigated the capacity of GCV and CsA to control T cell division during allogeneic HSCT. For this, Gene Therapy

Figure 1 In vitro sensitivity of TK-expressing T cells to (a) increasing doses of CsA, and (b) increasing doses of GCV, in absence or in presence of CsA. Splenocytes from TK-transgenic B6 mice were stimulated in MLC with splenocytes from B6/D2 mice. The dashed lines indicate 50% inhibition of proliferation. (a) Results are expressed as percent of inhibition of proliferation in three independent experiments (mean ± s.d.). (b) Inhibition of proliferation of TK T cells exerted by GCV alone (䊏) is compared with the one obtained in presence of 100 ng/ml of CsA (䉬). Inhibition of GCV on proliferation of T cells from non-transgenic B6 mice is also showed (왖). Comparable results were obtained in three independent experiments.

TK T cells were obtained from transgenic mice also expressing a human CD4 (hCD4) transgene as a marker at the cell surface of both CD4+ and CD8+ T cells.16 They were labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE), before being injected together with T cell-depleted bone marrow (BM) cells into irradiated allogeneic recipient mice. At different time-points after HSCT, spleen cells were harvested from grafted mice and division of hCD4+ donor cells was shown by the sequential loss of CFSE fluorescence intensity. We first analyzed the effect of CsA on T cell division kinetics in the early post-graft period (Figure 2). When recipient mice did not receive any treatment, cell division could be observed in CD4+ and CD8+ T cells at day 2. In contrast, at the same time-point, no cell division was observed when recipient mice were treated with CsA, illustrating the CsA cytostatic effect on T cells. However,


contrast, in mice that had been, and were still, treated with CsA, the majority of CD4+ and CD8+ T cells had not divided. This indicates that CsA retains its cytostatic effect on the T cells that start dividing after GCV discontinuation.

Figure 2 CsA exerts a transient cytostatic effect on CD4+ and CD8+ cells and does not antagonize the effect of GCV on TK T cell proliferation after transfer in semi-allogeneic irradiated hosts. CFSE-labeled T cells from [TKxhCD4] double transgenic B6 mice were transferred into semi-allogeneic (B6/D2) lethally irradiated mice. Recipient mice were either not treated (control) or received (1) CsA i.p. daily from day ⫺2 preceding transplant; (2) five injections of GCV between day 1 and day 3 after transplant; or (3) the combination of CsA and GCV. Data show proliferation (as shown by CFSE dilution pattern) of donor (hCD4+ gated) CD4+ and CD8+ T cells recovered from spleen of recipient mice at three different time-points (day 2, day 5 and day 7) following transfer. Similar results were obtained for three allogeneic recipient mice, at each time-point.

at day 5, the vast majority of CD4+ and CD8+ T cells had divided in both untreated and CsA-treated animals, indicating that the cytostatic effect of CsA on T cells is only transient. The intensity and pace of alloantigen-induced T cell proliferation largely depends on differences in the genetic backgrounds of donor and recipient. The transient cystostatic effect of CsA was not due to our particular combination of donor and recipient since it was also observed in a different genetic background combination (FVB->B6, not shown). Since GCV toxicity depends on cell division, GCV should eliminate the dividing T cells, despite CsA. Thus, we next compared the effect of GCV alone, or combined with CsA, on T cell division kinetics (Figure 2). GCV was injected twice daily from day 1 to day 3 after HSCT, while CsA was injected once daily from day ⫺2 before HSCT until sacrifice. When recipient mice did not receive any treatment, the majority of CD4+ and CD8+ T cells had already undergone a large number of divisions (more than eight) at day 5 after transfer. In contrast, when recipient mice received GCV alone or in combination with CsA, no divided cells could be observed in either CD4+ or CD8+ T cells at day 5. Since CsA alone did not prevent T cell division at day 5, these results illustrate the cytolytic effect of GCV on dividing TK T cells and indicate that CsA does not interfere with it at this timepoint. At day 7, when GCV had been stopped for 4 days, part of CD4+ T cells and the majority of CD8+ T cells had divided several times in mice treated with GCV alone. In

203

Effect of CsA and/or GCV on the prevention of GVHD after HSCT The above in vivo and ex vivo results cannot accurately predict the combined effect of CsA and GCV on GVHD. We investigated this by inducing GVHD with donor TK T cells obtained from a TK-transgenic line developed in the B6 background.17 When these T cells were transferred to semi-allogeneic irradiated [B6xD2]F1 recipient mice, a lethal GVHD was constantly observed (Figure 3a). Five injections of GCV administered every 12 h and initiated at day 1 after HSCT led to 80% survival at day 70 after HSCT (Figure 3a) (P = 0.01 compared with untreated animals). CsA alone, administered from day ⫺2 to day 8 at a dose of 50 mg/kg/day, did not prevent GVHDrelated mortality, with a survival rate of 11% at day 70 (Figure 3a). The single mouse surviving after CsA prophylaxis showed typical clinical signs of GVHD, including hunching and weight loss. This lack of efficacy could not be due to insufficient CsA levels, since residual CsA blood levels were 1957 ± 395 ng/ml (mean ± s.d., n

Figure 3 CsA does not impair the protective effect of GCV on the occurrence of GVHD in mice receiving TK-transgenic T cells. Lethally irradiated B6/D2 recipient mice were grafted with B6 BM-cells plus TKtransgenic T cells. (a) GVHD-related mortality is compared between mice receiving no prophylactic treatment, GCV alone (five injections between day 1 and day 3 after HSCT), CsA alone (daily injection from day ⫺2 to day 8), or GCV plus CsA. (b) GVHD-related mortality is compared between mice receiving no prophylactic treatment (same control group than in a), GCV (one injection at day 1 after HSCT) alone or in combination with CsA (daily injection from day ⫺2 to day 8). Number of mice is indicated in parentheses. Gene Therapy


204

= 4) at day 2 after initiation of treatment, and remained stable over time. Importantly, when GCV and CsA were administered together, the survival rate was similar to that observed with GCV alone. Histopathologic examination of liver, a target organ of GVHD, confirmed the absence or only mild signs of GVHD in the GCV- or GCV + CsA-treated groups (not shown). The results obtained in these in vivo experiments may depend on the scheme of GCV administration used. Therefore, we also tested several other schedules, consisting of one daily injection beginning at day 1 after HSCT, and administered on 1, 2 or 3 successive days, either alone or in combination with CsA (n = 4 in each of these six groups). With only one injection of GCV, mice were efficiently protected from GVHD, with 75-100% survival at day 70 after HSCT (Figure 3b). When GCV was administered together with CsA, the survival rate was statistically similar to that observed with GCV alone (Figure 3b). Using two or three injections of GCV, survival rates were statistically similar to those observed with one injection, and co-administration of CsA did not impair GVHD prophylaxis (not shown). With or without CsA, and whatever the number of GCV injections used, histopathologic examination of spleen and liver of these animals confirmed the absence of GVHD (not shown). Effect of CsA and/or GCV on immune reconstitution and allogeneic T cell responses after prevention of GVHD Immune reconstitution depends in large part on cell proliferation that could be impaired by CsA. Also, GVHD is frequently associated with immune deficiency.18 Thus, besides survival, it is important to assess immune reconstitution under our different therapeutic schemes. Analysis of splenocytes of grafted mice showed good T and B cell reconstitution in GCV- and GCV + CsA-treated groups, indicating efficient GVHD prevention (Table 1). No statistical difference was observed between these two groups regarding the absolute or relative T and B cell counts, in spleen and blood. Additionally, more than 95% of donor T and B cells were of donor origin indicating efficient engraftment. Thus, in accordance with clinical and histopathologic findings, addition of CsA to the GCV regimen neither impaired nor improved immune reconstitution of grafted mice. We also tested the ability of T cells from animals treated with CsA, GCV or GCV + CsA to respond to

Table 1

alloantigens. In MLC, splenocytes from mice grafted with BM plus TK T cells and treated by either GCV or GCV + CsA did not proliferate upon stimulation with donor H-2b or recipient H-2d allogeneic cells, whereas they responded normally to third-party H-2k stimulation (Figure 4). Thus, GCV treatment, associated or not with CsA, leads to a specific hypo-responsiveness of T cells to recipient alloantigens.

Discussion Despite prophylaxis based on CsA, GVHD still represents a frequent life-threatening complication encountered after allogeneic HSCT. The TK suicide gene therapy strategy, which permits the elimination of dividing T cells without toxicity on non-dividing cells, provides a

Figure 4 GCV treatment, associated or not with CsA, leads to a specific hyporesponsiveness to recipient alloantigens. Splenocytes from grafted mice were collected at day 70 post-HSCT and cultured in MLC in presence of donor (B6, H-2b), recipient (B6/D2, H-2bd), or third party (C3H, H2k) irradiated splenocytes. Proliferative responses of splenocytes collected from animals treated byCsA alone (daily injection from day ⫺2 to day 8 GCV alone (five injections between day 1 and day 3 post-HSCT), or GCV plus CsA (daily injection from day ⫺2 to day 8) are compared. Comparable results were obtained in three GCV-treated and in three (GCV + CsA)treated animals.

T and B cell reconstitution after different prophylactic regimens of GVHDa Spleen

GCV (n = 4) GCV + CsA (n = 4) Controlb (n = 2) a

Blood

No. of cells (×106)

T (%)

B (%)

% Donor origin

T (%)

B (%)

% Donor origin

77 (±27)

11 (±1)

36 (±14)

>95%

11 (±2)

39 (±17)

>95%

70 (±20)

14 (±4)

50 (±4)

>95%

8 (±2)

52 (±8)

>95%

98 (±10)

11 (±2)

42 (±4)

—

17 (±1)

58 (±3)

—

GCV treatment consisted of five injections administered every 12 h and initiated at day 1 after BMT. CsA was administered daily from day ⫺2 to day 8. Mice were killed at day 70 post-BMT for reconstitution analysis. The frequencies of T and B cells are expressed as percent of total cells (mean ± s.d.). Donor type was determined by allotype-specific MHC class I expression on T and B cells. b B6/D2 ungrafted mice. Gene Therapy


new way of preventing GVHD. Indeed, in a murine model of GVHD, we have already shown that a 7-day GCV treatment initiated at the time of transplantation is fully protective against GVHD,9 and allows an efficient immune reconstitution.9,19–21 We also previously showed that the TK/GCV system more efficiently prevents than cures GVHD.20 Whether or not this preventive treatment should be associated with the classical preventive treatment based on CsA in future clinical trials was the aim of this study. Indeed, there was a theoretical possibility that the cytostatic effect exerted by CsA could antagonize the cell division-dependent destruction of alloreactive T cells mediated by the TK/GCV system. Here, in vitro and in vivo experiments were in agreement that CsA does not antagonize the TK/GCV suicide gene system for prevention of GVHD. Grafted mice treated by GCV alone or GCV + CsA showed the same survival after HSCT, as well as no difference in either histopathologic signs of GVHD or in biological parameters of immune reconstitution. In addition, the TK/GCV system, associated or not with CsA, provided a clearly better prevention of clinical GVHD in comparison with the individual CsA treatment. Although we employed only a short course of CsA in this control group, longer CsA regimens, resembling those used in humans, have also been reported to be only partially or not at all protective against lethal GVHD in rodents.22–25 Using the recently available CFSE staining of T cells before their infusion, we observed that CsA does not control T cell division for more than 3–4 days after HSCT. This is the first observation of a mere transient cytostatic effect of CsA. Importantly, it provides an explanation for the only partially efficient effect of CsA for GVHD prophylaxis. Additionally, the detection of donor T cell divisions despite CsA treatment may explain why GCV remained active for elimination of alloreactive T cells. One may argue that the cytostatic effect exerted by CsA might be more important in humans than in mice, and therefore antagonize GCV. The inhibitory effect of CsA on mouse T cell proliferation that we found in MLC was similar to the one previously reported with human T cells,14 with an IC 50% close to 100 ng/ml in both cases. To circumvent potential variations of T cell sensitivity to CsA between humans and mice in vivo, we intentionally used a saturating dose of CsA, providing CsA blood levels of nearly 2000 ng/ml. Since the therapeutic CsA blood level used for GVHD prophylaxis in humans is 100–250 ng/ml, it is unlikely that, at a significantly lower blood concentration, CsA would exert a more important cytostatic effect on human alloreactive T cells than that observed in our murine model. Thus, it is reasonable to conclude that CsA should not inhibit the cell divisiondependent deletion of alloreactive TK T cells mediated by GCV in humans. This conclusion is further supported by recent clinical results. In the study of Tiberghien et al,12 four out of five patients who developed GVHD after TK T cell infusion and therefore received GCV were still under CsA at the initiation of treatment. Three of these four patients showed complete responses to GCV alone and one patient showed a partial response, which necessitated the addition of glucocorticoids to achieve a complete response. Thus, at least in three of these four patients, CsA did not antagonize the expected effect of GCV on alloreactive T cells. Together, our results demonstrate that a CsA-based

GVHD prophylaxis regimen does not alter the alloreactive T cell depletion mediated by the TK/GCV suicide gene system. In the present study, we showed that shortening the GCV treatment, even to only one injection administered 1 day after transplantation, successfully prevented GVHD. We believe that these results are of clinical relevance for the design of future trials involving gene-modified donor T cells for modulation of alloreactivity. The final decision to use CsA or not in clinical trials should be discussed with regard to the clinical setting of HSCT. For instance, for non-malignant diseases, such as aplastic anemia or inherited immune disorders, the priority is to control GVHD. Therefore, CsA could be combined with GCV to ensure the most efficient prophylaxis of GVHD, especially for patients grafted with HLAmismatched donors, which are at high risk of GVHD. For malignancies, the priority is to maintain the graftversus-leukemia effect of HSCT. Previous studies have suggested that long-term immune suppression with CsA can alter the graft-versus-leukemia effect,26,27 or graft-versus-tumor28 effects. It should thus be beneficial in this setting to use TK/GCV alone, a treatment that efficiently prevents GVHD in our murine model. This should prevent prolonged exposure to the immunosuppressive CsA in patients. Additionally, avoiding CsA could lower the risk of secondary malignancies associated with prolonged immune suppression, such as solid cancers or EBV-related lymphoproliferative disease.29

205

Materials and methods Mice C57BL/6 (B6, H-2b), DBA/2 (D2, H-2d), [B6xD2]F1 (B6/D2, H-2bd) and C3H (H-2k) mice were obtained from Iffa Credo (L’Arbresle, France). Ep⌬TK line 6 (TK) transgenic B6 mice expressing the TK product in both CD4+ and CD8+ T cells were described previously.17 EpCD4 line 10 transgenic B6 mice express the human (h)CD4 protein selectively on CD4+ and CD8+ cells.16 Double transgenic [TKxhCD4] B6 mice were obtained by breeding line 6 TK and line 10 hCD4 transgenic mice in the animal facility of the Faculte´ de Me´ decine Pitie´ Salpeˆ trie`re (Paris, France). Mice were manipulated according to EU guidelines. Tracking of the hCD4 surface marker, rather than the intracellular TK molecule which is difficult to detect by available methods, readily allows monitoring of infused mature donor CD4+ T cells independently of newly produced thymus-derived hCD4- T cells originating from non-transgenic HSCs. This system also permits identification of infused donor T cells versus host T cells surviving after irradiation. Mixed lymphocyte culture (MLC) and in vitro sensitivity to CsA and GCV Responder splenocytes (2 × 105) were cultured in roundbottomed 96-well microplates in presence of 4 × 105 20Gy irradiated stimulating splenocytes from different mouse strains. Increasing concentrations of CsA (Sandimmun; Novartis, Basle, Switzerland) and/or of GCV (Cymevan; Roche, Neuilly-sur-Seine, France) were added in a final volume of 0.2 ml of RPMI 1640 medium (Flobio, Courbevoie, France) supplemented with 10% fetal calf serum (Flobio). 3H-TdR uptake was measured after 4 days of culture in triplicate wells. Gene Therapy


206

Experimental GVHD and administration of CsA and GCV Experiments were performed as described,9 except otherwise stated. Briefly, B6/D2 euthymic females (8–12 weeks of age) were lethally irradiated (11 Gy). TCD-BM cells (10 × 106) from B6 mice plus 10 × 106 T cells collected from spleen and lymph nodes of [TKxhCD4] transgenic B6 mice were injected i.v. 1 day after irradiation. CsA was administered i.p. daily from day ⫺2 before HSCT, at a dose of 50 mg/kg/day diluted in 200 ␮l PBS. Residual CsA blood levels were monitored by radioimmuno-assay. GCV was administered i.p. at a dose of 50 mg/kg diluted in 200 ␮l PBS. Different schemes of CsA and GCV administration were used. Assessment of donor TK T cell proliferation in vivo Spleen and lymph node cells of donor mice were incubated at a concentration of 107 cells/ml with CFSE (Molecular Probes Europe, Leiden, The Netherlands) at a final concentration of 1.5 ␮m in a RPMI 1640 medium, for 10 min at 37°C and 5% CO2. The labeling reaction was stopped by addition of fetal calf serum (20% of the final volume). Labeled cells were washed twice in phosphate buffer saline, numbered and then injected intravenously with BM cells in lethally irradiated mice. Mice were treated by CsA and/or GCV, or not treated, and splenocytes were collected at different time-points after transplantation. Cell division was shown by the sequential loss of CFSE fluorescence of the donor (hCD4+ gated) CD4+ and CD8+ populations. Immunocytofluorometry Staining reactions were performed on blood cells and on collagenase-digested spleen cells from grafted animals. Cells were first incubated with 2.4.G2 anti-Fc receptor mAb. For analysis of immune reconstitution and chimerism cells were stained with combinations of the following mAbs: phycoerythrin (PE)-labeled anti-CD3 (clone 1452c11, Pharmingen, San Diego, CA, USA), FITC-labeled anti-B220 (clone RA3-6B2, Caltag Laboratories, San Francisco, CA, USA), biotinylated anti-H-2Kd (clone SF1-1.1, Pharmingen), and PE-labeled anti-H-2Kb (clone CTKb, Caltag). Biotinylated mAbs were revealed with tricolorlabeled streptavidin (Caltag). For CFSE experiments, splenocytes were stained with the following mAbs: PElabeled anti-CD4 (clone RMH-5, Pharmingen), tricolorlabeled anti-CD8 (clone CT-CD8a, Caltag), and allophycocyanin-labeled anti-hCD4 (Caltag Laboratories). Events were acquired on a FACSCalibur (Becton Dickinson, San Jose, CA, USA) and analyzed using CellQuest software (Becton Dickinson). Histopathological examination Liver samples were prepared in Bouin’s fixative, embedded in paraffin and sections were stained with hematoxylin and eosin. Slides were evaluated by a pathologist unaware of treatment data. Severity of GVHD was scored as previously described.20 Statistical analysis Statview software (Abacus Concepts, Berkeley, CA, USA) was used for statistical analysis. Kaplan–Meier survival curves were established for each experimental group. Survival differences between two groups were determined using the log-rank test. Differences in lymphoid cell

Gene Therapy

counts between two groups were determined using the Student’s t test. P values are indicated only when differences between two groups were statistically significant.

Acknowledgements We thank G Gavory for excellent animal care, G Boisserie and F Baillet for the irradiation of mice, JJ Benoliel for CsA dosage, and B Salomon and C Frisen for critical reading of the manuscript. JLC is supported by the Fondation pour la Recherche Me´ dicale, EL is supported by the Ligue Nationale contre le Cancer. This work was supported by the University Pierre et Marie Curie, the Centre National de la Recherche scientifique and Ge´ nopoie´ tic.

References 1 Storb R et al. Methotrexate and cyclosporine versus cyclosporine alone for prophylaxis of graft-versus-host disease in patients given HLA-identical marrow grafts for leukemia: long-term follow-up of a controlled trial. Blood 1989; 73: 1729–1734. 2 Socie G, Cahn JY. Acute graft-versus-host disease. In: Barrett A, Treleaven J (eds). The Clinical Practice of Stem-Cell Transplantation. Isis Medical Media: Oxford, 1998, pp 595–618. 3 Ratanatharathorn V et al. Phase III study comparing methotrexate and tacrolimus (prograf, FK506) with methotrexate and cyclosporine for graft-versus-host disease prophylaxis after HLA-identical sibling bone marrow transplantation. Blood 1998; 92: 2303–2314. 4 Martin PJ et al. Effects of in vitro depletion of T cells in HLAidentical allogeneic marrow grafts. Blood 1985; 66: 664–672. 5 Ash RC et al. Successful allogeneic transplantation of T celldepleted bone marrow from closely HLA-matched unrelated donors. N Engl J Med 1990; 322: 485–494. 6 Horowitz MM et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood 1990; 75: 555–562. 7 Goldman JM et al. Bone marrow transplantation for chronic myelogenous leukemia in chronic phase. Increased risk for relapse associated with T cell depletion. Ann Intern Med 1988; 108: 806–814. 8 Tiberghien P et al. Ganciclovir treatment of herpes simplex thymidine kinase-transduced primary T lymphocytes: an approach for specific in vivo donor T cell depletion after bone marrow transplantation? Blood 1994; 84: 1333–1341. 9 Cohen JL et al. Prevention of graft-versus-host disease in mice using a suicide gene expressed in T lymphocytes. Blood 1997; 89: 4636–4645. 10 Helene M et al. Inhibition of graft-versus-host disease. Use of a T cell-controlled suicide gene. J Immunol 1997; 158: 5079–5082. 11 Bonini C et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science 1997; 276: 1719–1724. 12 Tiberghien P et al. Administration of herpes simplex-thymidine kinase-expressing donor T cells with a T cell-depleted allogeneic marrow graft. Blood 2001; 97: 63–72. 13 Link Jr CJ et al. Adoptive immunotherapy for leukemia: donor lymphocytes transduced with the herpes simplex thymidine kinase gene for remission induction. HGTRI 0103. Hum Gene Ther 1998; 9: 115–134. 14 Hess AD, Tutschka PJ. Effect of cyclosporin A on human lymphocyte responses in vitro. I. CsA allows for the expression of alloantigen-activated suppressor cells while preferentially inhibiting the induction of cytolytic effector lymphocytes in MLR. J Immunol 1980; 124: 2601–2608. 15 Bunjes D, Hardt C, Rollinghoff M, Wagner H. Cyclosporin A mediates immunosuppression of primary cytotoxic T cell responses by impairing the release of interleukin 1 and interleukin 2. Eur J Immunol 1981; 11: 657–661. 16 Salmon P et al. Characterization of an intronless CD4 minigene


17

18

19

20

21

22

23

expressed in mature CD4 and CD8 T cells, but not expressed in immature thymocytes. J Immunol 1996; 156: 1873–1879. Cohen JL, Boyer O, Thomas-Vaslin V, Klatzmann D. Suicide gene-mediated modulation of graft-versus-host disease. Leuk Lymphoma 1999; 34: 473–480. Dulude G, Roy DC, Perreault C. The effect of graft-versus-host disease on T cell production and homeostasis. J Exp Med 1999; 189: 1329–1342. Cohen JL, Boyer O, Klatzmann D. Suicide gene therapy of graftversus-host disease: immune reconstitution with transplanted mature T cells. Blood 2001; 97: 2071–2076. Cohen JL et al. Immunological defects after suicide gene therapy of experimental graft-versus-host disease. Hum Gene Ther 1999; 10: 2701–2707. Cohen JL et al. Preservation of graft-versus-infection effects after suicide gene therapy for prevention of graft-versus-host disease. Hum Gene Ther 2000; 11: 2473–2481. Borel JF, Feurer C, Gubler HU, Stahelin H. Biological effects of cyclosporin A: a new antilymphocytic agent. Agents Actions 1976; 6: 468–475. Blazar BR, Taylor PA, Fitzsimmons WE, Vallera DA. FK506 inhibits graft-versus-host disease and bone marrow graft rejection in murine recipients of MHC disparate donor grafts by

24

25 26

27

28

29

interfering with mature peripheral T cell expansion post-transplantation. Blood 1994; 84: 2044–2049. Townsend RM, Gilbert MJ, Korngold R. Combination therapy with a CD4-CDR3 peptide analog and cyclosporin A to prevent graft-versus-host disease in a MHC-haploidentical bone marrow transplantation model. Clin Immunol Immunopathol 1998; 86: 115–119. Tutschka PJ et al. Use of cyclosporin A in allogeneic bone marrow transplantation in the rat. Nature 1979; 280: 148–151. Collins Jr RH et al. Hematologic relapse of chronic myelogenous leukemia following allogeneic bone marrow transplantation: apparent graft-versus-leukemia effect following abrupt discontinuation of immunosuppression. Bone Marrow Transplant 1992; 10: 391–395. Kolb HJ et al. Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients. Blood 1990; 76: 2462–2465. Childs R et al. Regression of metastatic renal cell carcinoma after nonmyeloablative allogeneic peripheral blood stem cell transplantation. N Engl J Med 2000; 343: 750–758. Deeg HJ, Socie G. Malignancies after hematopoietic stem cell transplantation: many questions, some answers. Blood 1998; 91: 1833–1844.

207

Gene Therapy