A 20-day treatment with LF15–0195, a deoxyspergualine analog, induced long-term heart allograft survival in the rat without signs of chronic rejection. LF15–0195-treated recipients did not develop an anti-donor alloantibody response. Analysis of graft-infiltrating cells, IL10, TNFα, IFNγ mRNA and iNOS protein expression in allografts, 5 days after transplantation, showed that they were markedly decreased in allografts from LF15–0195-treated recipients compared with allografts from untreated recipients. Surprisingly, spleen T cells from LF15–0195 recipients, 5 days after grafting, were able to proliferate strongly in vitro, when stimulated with donor cells, but had reduced mRNA expression for IFNγ compared with spleen T cells from untreated graft recipients. Furthermore, when T cells from naive animals were stimulated in vitro, using anti-CD3 and anti-CD28, LF15–0195 also increased T-cell proliferation in a dose-dependent fashion; however, these cells expressed less of the Th1-related cytokines, IFNγ and IL2, compared with untreated cells, suggesting that LF15–0195 could act on T-cell differentiation. In conclusion, we show here that a short-term treatment with LF15–0195 induced long-term allograft tolerance, decreasing the in situ anti-donor response, and we illustrate evidence for the development of regulatory mechanisms.
Present therapy for human pathologies such as end-stage organ failure and leukemias includes organ and cell allotransplantations, respectively (1,2). However, genetic disparities between organ donor and recipient at the level of the major histocompatibility complex (MHC) can lead to acute or chronic rejection. The clinical use of potent new immunosuppressive drugs has helped to control allograft rejection and prolonged the survival of transplants. However, current immunosuppressive drugs need to be taken for the life of the graft, inducing a nonspecific immunosuppression, and are associated with several complications such as malignancies and opportunistic infections (3). The induction of donor-specific tolerance, defined as the acceptance of a graft after treatment discontinuation without signs of chronic rejection, would be the solution of choice for long-term function and improved recipient conditions.
In this study, we examined the effects of LF15–0195, in a fully MHC-mismatched LEW.1W to LEW.1A rat heart allograft model. LF15-0195 is an analog of deoxyspergualine (DSG) (4), a compound isolated from culture filtrates of Bacillus laterosporus (5). Since the discovery of DSG in the early 1980s, only a limited number of new analogs have been described. LF08-0299 was the first reported to have long-term stability and to be able to induce long-term immune tolerance in 75% of rat heart allografts, with the emergence of regulatory CD4+ T cells (6,7). DSG has been reported to prolong allograft survival (8), and to inhibit B-cell differentiation and antibody production (9–11), antigen processing (12), NO synthase induction in macrophages (13), and human lymphocyte proliferation in response to mitogens and allogeneic stimulation (14). DSG and LF08-0299 are known to bind to hsc70 (heat shock protein 70), but the mechanisms of its immunosuppressive properties are unclear (15,16).
Our results show that treatment for 20 days with LF15–0195 induced long-term allograft survival in more than 90% of recipients. The analysis of the anti-donor response showed that anti-donor alloantibody production was strongly inhibited. Moreover, graft-infiltrating cells (GICs) were reduced and expressed low levels of cytokines. In contrast, the anti-donor proliferative capacity of spleen T cells from LF15–0195-treated recipients was markedly increased. By in vitro analysis, we demonstrated that LF15–0195 had a direct effect on T-cell differentiation.
Materials and Methods
Animals and transplantations
Eight-week-old male LEW.1W (RT1.u) or LEW.1L (RT1.l) rats served as heart donors, and LEW.1A (RT1.a) rats as allograft recipients (Centre d'Elevage Janvier, Le Genest-Saint-Isle, France). Heterotopic heart grafts and a second graft in the neck were performed as previously described (17). The grafts were evaluated daily for function by palpation, and rejection was defined as the day of cessation of heart beat and confirmed by histology.
LF15–0195 (‘Laboratoires Fournier’, DAIX, France) was prepared in phosphate-buffered saline (PBS) and delivered to allograft recipients by intraperitoneal injection at 3 mg/kg for 20 days, starting the day of cardiac transplantation.
Antibodies used in this study
The following hybridomas (mouse IgG) were obtained from the European Collection of Animal Cell Culture (Salisbury, UK) and were used to phenotype rat leukocytes: Ox1 and Ox 30 (anti-CD45), R7-3 (anti-TCRαβ), ED1 (recognizing CD68 on monocytes, macrophages, granulocytes and dendritic cells), Ox33 (anti-CD45RA isoform present on B cells) W3/25 (anti-CD4) and Ox8 (anti-CD8α). The hybridoma J319 (anti-rat CD28) was kindly provided by Dr T. Hünig (University of Wurzburg, Germany). These antibodies were purified from hybridoma culture supernatants in our laboratory. W3/25 was coupled to Phycoerythrin (PE) (Bioatlantic, Nantes, France). The monoclonal antibody 3.2.3 (anti-NKR-P1 present on natural killer cells and a subpopulation of dendritic cells) and the polyclonal rabbit anti-inductible nitric oxide synthase (iNOS) antibody were purchased from Serotec (Oxford, UK). Secondary antibodies included biotin-conjugated anti-mice IgG, anti-rabbit IgG, horseradish peroxidase-conjugated streptavidin and Vector VIP, purchased from Vector laboratories (Vector Laboratories Inc., Burlingame, CA, USA). Anti-CD3 monoclonal antibody was purchased from Pharmingen (Becton Dickinson Co., Mountain View, CA, USA). Fluorescein isothiocyanate (FITC)-affinity pure F(ab′)2 fragment, mouse anti-rat IgG, Fc (gamma) fragment specific antibody was purchased from Jackson Immunoresearch Laboratories (West Grove, PA, USA). DB1-FITC (anti-rat IFNγ) was kindly provided by Dr A. Saoudi (INSERM U28, Toulouse, France).
Determination of anti-donor alloantibodies
Lewis 1W splenocytes (donor haplotype) were incubated with decomplemented sera, 1 : 4 diluted in phosphate-buffered saline (PBS) containing 0.5% (w/v) bovine serum albumin (BSA) (Sigma) and 0.02% (w/v) sodium azide (Az) (PBS-BSA-Az) from grafted untreated or LF15–0195-treated rats. To stain for IgG, cells were reacted with FITC-affinity pure F(ab′)2 fragment mouse anti-rat IgG, Fc (gamma) fragment specific (Jackson Immunoresearch Laboratories); cells were collected on a FACScan® and analyzed using the CellQuest software (Becton Dickinson).
Quantitative analysis of cellular infiltrate by immunohistology
Immunohistology was performed on hearts from untreated or LF15–0195-treated rats, harvested 5 days after transplantation. Heart fragments were snap-frozen, embedded in Tissue Teck (OCT compound, California, USA), cut into 5-μm sections and fixed in acetone. Tissue sections were labeled using a three-step indirect immunoperoxidase technique with Ox1/Ox30, ED1, R7–3, Ox33, 3.2.3 and iNOS as primary antibodies. Tissue sections were then incubated with corresponding biotin-conjugated anti-mice IgG, or anti-rabbit IgG, and then with horseradish peroxidase-conjugated streptavidin. They were then developed with Vector VIP. The area of each immunoperoxidase-labeled tissue section infiltrated by cells was determined by quantitative morphometric analysis as previously described (18). Results are expressed as the percentage of the area of the tissue section occupied by positive cells (± SD).
Purification of cells used in this study
Splenocytes: Cell suspensions from spleens were prepared as described previously (19). Cells were resuspended in RPMI 1640 (Gibco, Sigma Chemical Co., St Louis, MO, USA) supplemented with 2 mm l-glutamine, 5 × 10−5 M 2-mercaptoethanol, 1 mm sodium pyruvate (Gibco), 1% (w/v) nonessential amino acids (Gibco), 100 U/mL penicillin (Gibco); 0.1 mg/mL streptomycin (Gibco) and 10% (v/v) heat-inactivated fetal calf serum (Gibco).
Donor dendritic cells: Dendritic cells (DC) were enriched (40%) from spleen as previously described (20).
T lymphocytes: Total or CD4+ T cells were purified from splenocytes by negative selection. Briefly, spleen cells were incubated with a cocktail of mouse anti-rat antibodies: Ox6 (anti class II), ED3 (recognizing macrophages) and Ox33 (anti-CD45 present on B cells). Cells were then incubated with superparamagnetic beads with affinity-purified Goat-anti-mouse IgG covalently bound to the surface (Dynal, Oslo, Norway), and removed with a magnet. For the purification of CD4+ T cells, Ox8 (anti-CD8) monoclonal antibody was added to the cocktail. The purity of the collected T cells was controlled by FACS analysis (FACScan®, Becton Dickinson Co.) with an anti-TCR αβ monoclonal antibody (R7.3) staining for total T cells and with an anti-CD4 (W3/25) for CD4+ T cells (purity > 95%).
One-way mixed lymphocyte reaction (MLR): Low-density cells corresponding to DC-enriched cell populations from donor-type LEW.1W (RT1.u) rats were irradiated and served as stimulator cells, while T cells served as responder cells. Responder (2 × 105 cells/well) and stimulatory cells (5 × 104 cells/well) were plated in 96-well round-bottomed plates in triplicate in a final volume of 200 μL.
Anti-CD3 plus anti-CD28 stimulation: Purified T cells from naive rats (5 × 104 cells/well) were stimulated in a 96-well flat-bottom plate (NUNCTM, Merck, Eurolab, France) coated with anti-CD3 (0.75 μg/mL) (Pharmingen, Becton Dickinson Co.) and with addition of soluble anti-CD28 (0.6 μg/mL) and with or without LF15–0195 (2 μg/mL).
The cultures were incubated at 37 °C, in 5% (v/v) CO2 and pulsed for the last 8 h with 0.5 μCi [3H]TdR (Amersham, Les Ulis, France). Cells were then harvested on glass fiber filters and [3H]TdR incorporation was measured using standard scintillation procedures (Packard Institute, Meriden, CT, USA).
Heart samples at 5 days after transplantation were immediately frozen in liquid nitrogen and stored at − 80 °C until RNA extraction. Total RNAs from whole allografts were extracted according to the technique of Chirgwin (21). Total RNAs from MLR assay cells were extracted using the technique of Chomczynsky and Sacchi (22). The RNAs were quantified by UV absorbance at 260 nm.
Quantitative RT/PCRs were performed on the ABI Prism 7700 (PE-Biosystems) using the TaqMan chemistry (PE-Biosystems under license from Roche Molecular Systems Inc). This TaqMan system performed real-time kinetic PCR and true quantitative gene analysis. The sequences of the gene-specific primers are given in Table 1. Standards were prepared by PCR amplification of each target sequence using these primers. PCR products were extracted and the OD260 allowed the quantification of the template in the standards. The standards were diluted in order to load 107−102 copies per well. Total RNAs from grafts or from MLR were reverse-transcribed using oligo-dT, as previously described (23). A constant amount of cDNA corresponding to the reverse transcription of 100 ng of total RNA, or each dilution of the standard, were amplified using the SYBR Green PCR Core Kit (PE Biosystems) containing the primers for HPRT, IFNγ IL10, IL2, IL13, TNFα or TGFβ. The PCR efficiencies of all the standards were greater than 99% and the correlation index between the input copy numbers and the fluorescence was always greater than 0.95. Data were expressed as ratios of the number of copies of the specific gene to the number of copies of the HPRT gene.
Table 1. : Oligonucleotide sequences used for quantitative RT/PCR
5′ to 3′ oligonucleotidessequences
Supernatants from cultures were harvested 72 h after stimulation. IFNγ, IL10 and IL2 were measured using Pharmingen OptEIATM ELISA and TGFβ1 was measured by the TGFβ Immunoassay system (Promega Corporation). ELISAs were performed according to the manufacturer's instructions.
Flow cytometric analysis of IFNγ intracellular staining
Two hours before cell harvest, 10 μg/mL of Brefeldin A (Sigma) was added. Cells were harvested, washed in the presence of Brefeldin A (5 μg/mL) and stained using PE-W3/25 (anti-CD4) in PBS-BSA-Az (1 μg/mL). Labeled cells were then fixed with 2% (v/v) paraformaldehyde for 20 min at 4 °C. After being washed with PBS-BSA-Az and incubated for 10 min in saponin medium [PBS-BSA-Az, 0.1% (v/v) saponin], cells were incubated for 30 min at room temperature with the FITC-DB1 (anti-rat IFNγ) (1.25 μg/mL). After three washes in PBS-BSA-Az-0.1% saponin and one additional wash without saponin to allow membrane closure, the cells were resuspended in 150 μL of PBS-BSA-Az. Cells were collected on a FACScan® and analyzed using the CellQuest software (Becton Dickinson).
Statistical evaluation was performed using Student's t-test for unpaired data, and results were considered significant if p-values were < 0.05.
Data were expressed as mean ± SE.
Specific donor LEW.1W (RT1.u) allograft tolerance in LEW.1A (RT1.a) recipients after short-term treatment with LF15–0195
As shown in Table 2, untreated LEW.1A recipients (RT1.a) rejected LEW.1W (RT1.u) cardiac allografts in about 7 days (n = 12), whereas a treatment for 20 days with LF15–0195 (3 mg/kg, i.p. daily) prolonged graft survival to > 200 days in about 90% of recipients (n = 18) (p < 0.001). Acceptance by long-term survivor (LTS) of donor LEW.1W (RT1.u) (> 200 days) (n = 3) but not third-party LEW.1L (RT1.l) second heart allografts in the neck (15 days) (n = 2) demonstrated donor-specific allograft tolerance.
Table 2. : LEW1.W donor-specific tolerance to cardiac allografts in LEW.1 A recipients after a short-term treatment with LF15–0195. LEW.1A recipients were grafted with LEW.1W heart and were either untreated or treated with LF15–0195 (3 mg/kg, i.p.) for 20 days, starting the day of transplantation. Long-term survivor (LTS) LF15–0195-treated recipients were grafted with a donor (RT1.u) or third-party (RT1.l) second heart in the neck
Heart allograft emplacement
Graft survival days
Twenty-day treatment with LF15-0195 at 3 mg/kg i.p.
Absence of anti-donor antibodies response in LF15–0195-treated recipients
Sera from untreated or LF15–0195-treated recipients were tested for anti-donor IgG antibodies response at 5, 7 and 11 days after grafting for four animals in the two groups. As shown in Figure 1, untreated recipients developed a strong anti-donor alloantibody response, whereas LF15–0195-treated recipients did not develop any anti-donor IgG (p < 0.001). LF15–0195 totally inhibited the development of an anti-donor alloantibody response.
Immunohistological analysis of graft-infiltrating leukocytes
Phenotypic analysis of cellular infiltrate was performed in allografts from untreated or LF15–0195-treated recipients 5 days after transplantation, for four animals in each group. Tissue cryostat sections were labeled with Ox1/Ox30 (anti-leukocyte common antigen CD45), R7-3 (anti-TCRαβ), ED1 (recognizing CD68 on monocytes, macrophages, granulocytes and dendritic cells), Ox33 (anti-CD45 isoform present on B cells), and 3.2.3 (anti-NKR-P1 receptor present on natural killer cells and a subpopulation of dendritic cells). Representative micrographs of graft tissue cryostat sections labeled with Ox1/Ox30 and ED1 are shown in Figure 2. Allograft from untreated recipients (Figure 2A) presented an intense leukocyte infiltrate, which was significantly decreased in an allograft from LF15–0195 recipient (Figure 2B). This decrease was dramatic for the CD68-positive cells, the major cell subset infiltrating allografts (Figure 2C vs. Figure 2D).
As shown in Figure 3, total leukocytes (CD45-positive cells) represented 29% of area infiltrate in allografts from untreated recipients (n = 4), compared with 11% in allografts from LF15–0195-treated recipients (n = 4) (p < 0.001). The graft infiltrating cells (GIC) reduction was significant for CD68-positive cells (15% vs. 4%, p < 0.001), TCRαβ-positive cells(5% vs. 2%, p < 0.001), and NKR–P1-positive cells (1.5% vs. 0.8%, p < 0.02). The minor CD45–RA-positive cell subset was not significantly modified between the two groups (0.53% vs. 0.27%).
These results indicate a reduced infiltration of effector cells in allografts from LF15–0195-treated recipients at day 5 after transplantation.
Immunohistological analysis of iNOS protein expression in allografts
Allografts from untreated or LF15–0195-treated recipients (n = 4) were harvested 5 days after transplantation, and cryostat graft sections were labeled with anti-iNOS polyclonal antibody. The two representative micrographs show that iNOS protein was highly expressed in allografts from untreated recipients, whereas it was expressed by a lower number of cells in allografts from LF15–0195-treated recipients (Figure 4A). This decrease was confirmed by counting positive cells (13% area infiltrate in allografts from untreated recipients vs. 5% in allografts from LF15–0195-treated recipients, p < 0.003, Figure 4B).
The decrease in cells expressing iNOS protein correlated with the decrease in GICs and, in particular, in CD68-positive cells observed in allografts from LF15–0195-treated recipients.
Quantitative analysis of cytokine mRNA expression in allografts
Cytokine mRNA expression analysis were performed on allografts from untreated or LF15–0195-treated recipients harvested 5 days after transplantation (n = 4). As shown in Figure 5, the IL10 mRNA expression was markedly decreased by 9-fold in allografts from LF15–0195-treated recipients compared with allografts from untreated recipients (p < 0.008). To a lesser extent, IFNγ and TNFα mRNA expressions were significantly decreased 2-fold (p < 0.04) and 1.6-fold (p < 0.05), respectively. IL2 mRNA expression was also decreased, but the difference did not reach significance. The mRNA expression of the Th2-related cytokines IL13 and IL4 (data not shown) and of the immunosuppressive cytokine TGFβ was not significantly different between the two groups.
The decrease in IFNγ and TNFα mRNA expression could be correlated with the significant decrease of GICs observed in allografts from LF15–0195-treated recipients. The strong decrease in IL10 mRNA expression could suggest an inhibition of the activation state of GICs, particularly macrophages. In contrast, despite the decrease in GICs, the mRNA expression of the Th2-related cytokines IL13 and IL4 and of the immunosuppressive cytokine TGFβ was preserved.
Anti-donor response of spleen cells from grafted animals
MLR: Total spleen cells (n = 2) or purified spleen T cells (n = 4) from untreated or LF15–0195-treated recipients were purified 5 days after transplantation and were stimulated for 3 days with an irradiated donor dendritic cells-enriched cell population, as described in Materials and Methods. Surprisingly, as shown in Figure 6(A), the proliferative response of spleen cells from LF15–0195-treated recipients was 3-fold higher than that of spleen cells from untreated recipients (p < 0.05). Moreover, the proliferative response of spleen-purified T cells was also 2-fold higher than that of spleen T cells from untreated recipients (p < 0.05).
Analysis of cytokines expressed by proliferating T cells stimulated by donor ApC (MLR): We then performed quantitative RT/PCR analysis with RNAs extracted from MLR assays cells to analyze the cytokine expression of T cells from untreated (n = 2) or LF15–0195-treated recipients (n = 4) stimulated with the irradiated donor dendritic cell-enriched cell population at 72 h of stimulation. As shown in Figure 6(B), the level of IL2 mRNA expression was not significantly increased in MLR with T cells from LF15–0195-treated recipients, despite the strong proliferation. Surprisingly, mRNA expression of IFNγ was significantly reduced by 2.4-fold (p < 0.05) in MLR with T cells from LF15–0195-treated recipients. The mRNA expressions of IL10 and TGFβ were similar between the two groups. The expressions of IFNγ, IL10 and TGFβ had been confirmed by ELISA on culture supernatant at 72 h (data not shown).
Therefore, despite a significant increase in the anti-donor proliferative response of spleen T cells from LF15–0195-treated recipients compared with that of untreated recipients, IFNγ expression was decreased, whereas IL10 and TGFβ expressions were preserved.
Effect of LF15–0195 on purified T cells stimulated in vitro with anti-CD3 plus anti-CD28
LF15–0195 increased the proliferation: To determine if LF15–0195 acted directly on T cells, we performed in vitro experiments by stimulating purified total or CD4+ T cells from naive rats with anti-CD3 plus anti-CD28, and analyzed the proliferation by thymidine incorporation. Total T cells were stimulated for 72 h with coated anti-CD3 (0.75 μg/mL) plus anti-CD28 (0.6 μg/mL) with serial dilutions of LF15–0195 (0.01–10 μg/mL). LF15–0195 increased the proliferation of T cells in a dose-dependent manner (Figure 7A).
LF15–0195 inhibited the production of the Th1 IFNγ and IL2 related cytokines by CD4+ stimulated T cells: To analyze the production of IFNγ by CD4+ or CD8+ T cells, we performed CD4 and IFNγ intracellular staining of total T cells stimulated for 72 h with anti-CD3 plus anti-CD28 with an optimal dose of LF15–0195 (2 μg/mL). Figure 7(B) illustrates that a decreased percentage of CD4+ T cells expressed IFNγ when LF15–0195 was added (26% vs. 16%), whereas the percentage of CD4– T cells secreting IFNγ was not modified (7% vs. 9%).
To analyze whether expression of other cytokines by CD4+ T cells was modified, purified CD4+ T cells were stimulated for 72 h with anti-CD3 plus anti-CD28, and we tested supernatants of culture by ELISA for IL2, IFNγ, IL10 and TGFβ.
As shown in Figure 8, the levels of IL2 and IFNγ in supernatants of culture at 72 h of stimulation were significantly decreased (by 5.6-fold and 2-fold, respectively) when LF15–0195 was added, despite the increase of proliferation observed (data not shown). The level of IL10 in supernatants was slightly decreased and stimulation did not induce secretion of TGFβ. These results were confirmed by mRNA quantitative analysis (data not shown).
In a fully MHC-mismatched rat allograft model, a short-term treatment with LF15–0195, a DSG analog, induced long-term tolerance. This tolerance was donor specific, since a second donor heart allograft was accepted whereas a third-party second heart was rejected.
In this paper, in an attempt to try to define the mechanisms that may control allograft rejection and induction of tolerance, we analyzed the anti-donor alloantibody response, the phenotype and the activation state of graft-infiltrating cells, the anti-donor spleen T-cell proliferative response and cytokine production. Moreover, to confirm a direct effect of LF15–0195 on T cells, we performed in vitro analysis of the effect of LF15–0195 in the stimulation (anti-CD3 plus anti-CD28) of purified T cells from naive rats.
We demonstrated that the anti-donor alloantibody response was absent in LF15–0195-treated recipients. In the same combination LEW.1W to LEW.1A, injection of sera from ‘rejecting’ animals was sufficient to induce rejection of allograft from tolerant recipients, suggesting an important involvement of alloantibodies in acute rejection (24). Moreover, alloantibody response leads to vessel damage that could contribute to chronic rejection (25,26). It has been reported that DSG suppressed the immunoglobulin synthesis of B lymphocyte in both T-cell-dependent and T-cell-independent systems in vivo and in vitro (9,27,28). DSG has been reported to block an early stage of preB cell differentiation, the A–C to C′ transition, controlled by expression of a pre-receptor complex containing the Ig heavy chain, but not the light chain (11). Moreover, DSG inhibits kappa light chain expression in 70Z/3 preB cells by blocking lipopolysaccharide-induced NF-kappa B activation (10).
We observed that at 5 days after transplantation, LF15–0195 reduced by half the total leukocyte infiltration in allografts. This decrease was dramatic for CD68-positive cells (monocytes, macrophages, granulocytes and dendritic cells). Moreover, expression of cytokines expressed by graft-infiltrating macrophages (iNOS and TNFα) was also reduced 5 days after transplantation in allografts from LF15–0195-treated recipients, and correlated with the decrease of GICs. IL10 mRNA expression was strongly reduced by 9-fold in allografts from LF15–0195-treated recipients. This decrease can be explained not only by the 2-fold decrease in GICs, but also by an inhibition of the activation state of the GICs. In the same LEW.1W to LEW.1A allograft combination, tolerance by blood transfusion prior to engraftment was also characterized by inhibition of IL10 mRNA expression in allografts, in contrast to rejected allografts from untreated recipients which strongly expressed IL10 (23). However, always in the same combination, intragraft expression of IL10 by rat adenovirus transfection significantly prolongs allograft survival (29). Indeed, IL10 has opposite effects in allograft survival. In addition, treatment with IL10 has been reported to accelerate islet allograft rejection (30), cardiac allograft rejection (31), and to decrease survival of bone marrow grafts (32). In contrast, administration of viral IL10 has been reported to prolong allograft survival in rat (33) and in mice (34). Viral IL10 does not possess the T-cell costimulatory activities of authentic cellular IL10, and this could explain the differences observed in these models (35). Moreover, these different effects of IL10 on allograft survival could be due to the different kinetics of expression of IL10. It could also depend on the cells that produce IL10 (APCs, T cells). Indeed, IL10 is strongly expressed by activated macrophages (36), but also by Th2 T cells, which inhibit both Th1-cell proliferation and production of IL2 and IFNγ (37). Moreover, IL10 is strongly expressed by regulatory cells (38). We hypothesize that the decrease in IL10 mRNA expression observed in allografts from LF15–0195-treated recipients at 5 days after transplantation could be related to the inhibition of the activation state of macrophages, but we do not exclude the possibility that IL10-producing T cells could appear after the phase of induction of tolerance.
The inducible form of NOS (iNOS) secreted by activated macrophages was also strongly inhibited in allografts from LF15–0195-treated recipients. iNOS is important in acute rejection (39–41), and treatment of allografts with a selective iNOS inhibitor prolongs allograft survival (42). Moreover, TNFα expression by activated macrophages, NK and T cells (considered as a marker for acute allograft rejection) (43) was inhibited in allografts from LF15–0195-treated recipients. Macrophages represented half of the GICs in untreated recipients and played an early role in cell-mediated immunity by secreting inflammatory cytokines, triggering recruitment of effector cells and cellular damages. They are important in acute rejection (44–46). Nash demonstrated that treatment of animals with anti-macrophage agents prolonged survival of transplanted islet cells (47).
In monocytes, it has been reported that DSG inhibits IL1 production, class I (48) and class II (49) induction, antigen processing (12), and NO synthase induction (13). Macrophage activation is also inhibited by DSG. We suggest that treatment with LF15–0195 dramatically inhibited monocytes/macrophages activation and local recruitment. This could be a direct target, and crucial inhibition step, to inhibit rejection.
IFNγ mRNA expression was reduced in allografts from LF15–0195-treated recipients 5 days after transplantation. The Th2-related cytokines IL13 and IL4 and the immunosuppressive cytokine TGFβ mRNA expressions were not markedly modified. Moreover, the in vitro study (MLR) of the spleen purified T cells revealed that, in spite of an increased proliferative response against donor antigens, T cells from LF15–0195 decreased the expression of IFNγ, whereas expression of IL10 and TGFβ was preserved.
T cells are important mediators of graft rejection by recognizing foreign histocompatibility antigens expressed on donor APC (direct presentation) or presented by recipient antigen presenting cells (APC) (indirect presentation) (50). Activated CD4+ T lymphocytes could be differentiated in Th1- or Th2-type cells (51–54). Th1-type cells secrete cytokines (IL2, IFNγ), triggering effector mechanisms such as augmentation of MHC class II expression, activation of macrophages (secretion of TNFα, iNOS), activation and differentiation of B cells to secrete IgG2a, and generation of cytotoxic T cells (Fas-L, perforin, granzyme B) (55–57). Th1-type cell responses are described in various models of allograft rejection (58,59). In contrast, Th2-type cells secrete IL4, IL5, and IL13, which down-regulate Th1 cytokine expression and favor production of IgE and IgG1 by B cells (60). Many studies have suggested that tolerance induction may be linked to inhibition of Th1-type cells in transplantation (61–63). Moreover, regulatory cells have been reported to be involved in several models of allograft tolerance maintenance (38). Regulatory cells have been described to express IL10 (64), TGFβ (65) or IL4 (66,67). The fact that we did not observe increased expression of IL10, TGFβ or IL4 in T cells from LF15–0195-treated recipients at 5 days after transplantation does not exclude the possibility that this compound may generate regulatory cells which require a long time to expand (68,69). In addition, IL10 mRNA expression after in vitro donor stimulation was preserved in spleen T cells from LF15-0195-treated recipients, suggesting that the striking decrease in IL10 expression in allografts could be due to an inhibition of activation of macrophages.
Few studies about the effect of DSG on T cells have been published. The results of in vitro studies remain controversial. Indeed, in in vitro experiments, fetal calf serum contained high amounts of polyamine oxidase, which hydrolyzes DSG to toxic aldehydes and modulates the suppressive activity (70). However, Kerr and Atkins demonstrated that DSG inhibits lymphocyte proliferation in response to mitogenic and allogeneic stimulation (14). In contrast, other studies described no effect of DSG on proliferation and activation of T cells (11,12).
This discrepancy may be observed because LF15–0195 has been demonstrated to be very stable in solution and may be less toxic in vitro than DSG (4). We show in this study that LF15–0195 increased the proliferation of T cells stimulated in vitro by anti-CD3 plus anti-CD28 at the low dose of 0.01 μg/mL, and was not toxic at the high dose of 10 μg/mL. This effect seems to be independent of two T-cell growth factors: IFNγ and IL2 (71). At 72 h of stimulation, LF15–0195 decreased the IL2 and IFNγ mRNA and protein expression of CD4+ T cells. This suggests that LF15–0195 can directly modulate the T-cell proliferation and the Th1-related cytokine expression by a mechanism which is not yet identified. Further experiments will determine if this strong proliferation observed in the first days of treatment, and if this modulation of differentiation of T cells may be linked to the emergence of regulatory T cells. Indeed, in this model, we demonstrated the presence of regulatory CD4+ T cells in tolerant long-term grafted recipients able to transfer the tolerance to naive secondary recipients, suggesting that LF15–0195 could directly act on T-cell function (Chiffoleau et al. manuscript submitted). Moreover, allograft tolerance induced by another analog of DSG, LF08-299, is also maintained by regulatory CD4+ T cells (6).
Molecular mechanisms of action of LF15–0195 have yet to be elucidated, but this study demonstrates that LF15–0195 may have specific effects related to long-term tolerance induction in blocking several effector mechanisms and modulating T-cell differentiation.
We thank Jean-Marie Heslan for primer choice and Taq-Man technology, Helga Smith for grafts, Cécile Guillot and Sophie Brouard for graft advertising, and Joanna Ashton for reading the manuscript.
This work was supported by FOURNIER laboratories (DAIX, France).