A NOVEL CLINICALLY RELEVANT APPROACH TO TIP THE BALANCE TOWARDS REGULATION IN STRINGENT TRANSPLANT MODEL

Abstract

Background

Regulatory T cells (T_reg) actively regulate alloimmune responses and promote transplantation tolerance. Thymoglobulin^®, a rabbit polyclonal anti-thymocyte globulin (ATG), is a widely used induction therapy in clinical organ transplantation that depletes peripheral T cells. However, resistance to tolerance induction is seen with certain T cell depleting strategies and is attributed to alterations in the balance of naïve, memory and T_regs. The exact mechanism of action of ATG and its effects on the homeostasis and balance between T_regs and T-effector-memory cells (T_em) are unknown.

Methods

A novel antibody reagent, rabbit polyclonal anti-murine thymocyte globulin (mATG), generated by the same process utilized to manufacture Thymoglobulin^®, was used alone or in combination with CTLA4Ig and/or sirolimus (SRL) in a stringent fully MHC-mismatched murine skin allograft model to study graft survival and mechanisms involved.

Results

mATG depletes T cells but preferentially spares CD25⁺ natural T_regs which limit skewing of T cell repertoire toward T_em phenotype among the recovering T cells. T cell depletion with mATG combined with CTLA4Ig and SRL synergize to prolong graft survival by tipping the T_reg/T_em balance further in favor of T_regs by preserving T_regs, facilitating generation of new T_regs by a conversion mechanism and limiting T_em expansion in response to alloantigen and homeostatic proliferation.

Conclusions

Simultaneous T-cell depletion with ATG and costimulatory blockade, combined with SRL synergize to promote regulation and prolong allograft survival in a stringent transplant model. These results provide the rationale for translating such novel combination therapy to promote regulation in primate and human organ transplantation.

INTRODUCTION

Allograft tolerance may be viewed as a balance between alloreactivity and regulatory mechanisms, where different subsets of T cells can interact to create the appropriate environment for long-term graft acceptance (1). T_regs, a T cell subset involved in modulating the immune response (2), have been demonstrated to play a critical role in preventing autoimmune diseases (3), regulating immune response during infectious diseases (4), maintaining feto-maternal tolerance (5), and also preventing allograft rejection in transplantation (6). Consequently, there is great interest in harnessing the power of T_regs to therapeutic advantage, and indeed ex-vivo expanded natural CD4⁺CD25⁺ T_reg therapy has been used to prevent experimental acute allograft rejection (7). On the other hand T-effector cells are thought to play a major part in mediating acute and chronic allograft rejection (8). It has been proposed that tolerance may be viewed not as an “all or none” phenomenon but rather as a balance between these different subsets of T cells, which can interact to create the appropriate environment for long-term graft acceptance under a variety of immunosuppressive treatments (9).

Various preparations of anti-thymocyte globulin (ATG), a polyclonal T cell depleting agent, have been used clinically since the early 80’s as induction therapy and for treatment of acute rejection in solid organ transplantation (10, 11). ATG consists of purified IgG fraction of sera from rabbits or horses. Genzyme’s Thymoglobulin^®, produced by immunizing rabbits with fresh human thymocytes, is probably the most widely used and studied polyclonal ATG preparation. Both experimental and clinical studies of Thymoglobulin^® report profound T cell depletion through complement-dependent lysis or activation associated apoptosis, which is believed to be the key mechanism of its action by reducing the alloreactive T cell repertoire and frequency (11–15). Other potential mechanisms of action include modulation of surface adhesion receptor molecules or chemokine receptor expression (16). However the exact mechanisms of its action are still unknown and an ever-expanding array of target antigens is being identified in ATG preparations (17). We and others have reported that besides profound peripheral T cell depletion, Thymoglobulin^® modulates the immune response through its effects on the function of many other immune cell types, including, B cells, monocytes/macrophages, dendritic cells (DC) and T_regs (17–19). Further, anti-murine thymocyte globulin (mATG) has also been shown to bind to DC and deplete CD8⁺DC in the NOD mouse (20). However, resistance to tolerance induction is seen with certain T cell depleting strategies and is attributed to alterations in the balance of naive, memory and regulatory T cells (21). Our group has previously demonstrated that Thymoglobulin^® can expand T_regs ex vivo, inviting a fresh look at the tolerogenic potential of this T cell depletion strategy (19). Similarly, SRL, another widely used immunosuppressive agent in clinical transplantation, is thought to expand naturally occurring T_regs (22, 23). Further, strategies to block T cell costimulatory pathways prolong allograft survival in rodents and primates through expansion of CD4⁺CD25⁺ regulatory T cells in certain situations, and are currently undergoing clinical development in kidney transplantation (24–28) providing the rationale to combine ATG with SRL and CTLA4Ig to promote regulation and long-term allograft survival. We therefore tested the effect of combined mATG, SRL and B7 T cell costimulatory blockade in vivo in a fully allogeneic mouse skin allograft model. Our experiments confirm the previous report that mATG preserves natural T_regs(29). Further, we report for the first time that these preserved CD25⁺ natural T_regs limit skewing of T cell repertoire toward T-effector-memory cells (T_em) phenotype among the recovering T cells. Furthermore, simultaneous T-cell depletion combined with SRL and costimulatory blockade synergize to prolong allograft survival in a stringent transplant model by further limiting the alloreactive T_em cell pool and that this effect is dependent on CD25⁺ T_regs. These data emphasize the advantage of a novel strategy combining mATG with T cell costimulatory blockade to preserve T_regs and limit homeostatic proliferation of effector and memory T cells and provide the rationale to translate such a strategy to promote regulation in human organ transplantation.

METHODS

Mice

C57BL/6 (B6, H-2b), BALB/c (H-2d), B6.PL-Thy1^a/CyJ (Thy1.1 B6) and B6.C.H-2 bm12 (bm12) mice were purchased from Jackson Laboratory. ABM-Foxp3GFP^rep mice were generated by inter-breeding ABM and Foxp3GFP^rep mice, both on the B6 background and bearing the Thy 1.2 allele. All mice were housed in accordance with institutional guidelines.

Generation and characterization of mATG

mATG was generated in a manner analogous to the commercial ATG product (Thymoglobulin^®) by immunizing rabbits with a mixture of thymocytes from 8 different strains of mice (C57BL/6, BALB/c, DBA/2, 129, C3H, SJL, Swiss Webster, ICR) as described previously (30). Thymocyte suspensions were prepared from thymi extracted from the various donor mice and at least 25 New Zealand rabbits were immunized twice, 2 weeks apart, and terminally bled 2 weeks following the second immunization (Millbrook Immunoserv). Total rabbit IgG from the resulting serum was pooled and purified with a process analogous to Thymoglobulin^®. Control rabbit IgG was similarly purified from whole normal rabbit serum. mATG contains cytotoxic antibodies directed against antigens expressed on murine T and non-T lymphocytes as well as a wide variety of antigens as described previously (13, 18, 31, 32).

Treatment Protocols

mATG or control IgG was administered at a dose of 0.5 mg on day 0 and day 4 after transplantation as induction therapy; CTLA4Ig (Abatacept) is a human IgG1 fusion protein obtained from Bristol-Myers Squibb and was administered at a dose of 0.5 mg on day 0 and 0.25 mg on days 2, 4, 6, 8 and 10. Sirolimus was a generous gift of Wyeth Pharmaceuticals and administered at a dose of 0.3 mg/kg/day on days 0, 2, 4, 6, 8, and 10. CD25⁺ T cell depletion was achieved by treating mice pre-operatively with 0.5 mg of anti-CD25 mAb (PC61) on days 6 and 1 before transplantation. All treatments were administered by intraperitoneal injection.

Skin Transplantation

Full-thickness trunk skin grafts (1 cm²) harvested from BALB/c donors were transplanted onto the flank of B6 recipient mice, sutured with 6.0 silk, and secured with dry gauze and a bandage for 7 days. Skin graft survival was monitored daily thereafter, and rejection was defined as complete graft necrosis. For mechanistic studies bm12 skin was harvested and transplanted onto the flank of Thy 1.1 B6 recipients mice with the same technique.

Flow Cytometry Analysis

Recipient lymphocytes were isolated from the spleens and draining lymph nodes 7 days after transplantation. Cells were stained with anti-CD3 FITC, anti-CD19 PE, anti-CD8 APC and anti-CD4 PerCP to assess the depletion effect of mATG; T_em cells were determined by staining with anti-CD4 PerCP, anti-CD8 FITC, anti-CD62L APC, and anti-CD44 PE (all from BD Pharmingen) and analyzed for cells bearing the CD44^hiCD62L^lo phenotype. Cells were also stained with NK1.1 PE, anti-CD11c APC and anti-CD80 PE, anti-CD11b FITC and anti-F4/80 PE to identify NK cells, dendritic cell and macrophages respectively. T_regs were detected by staining with anti-CD4 FITC, anti-CD25 PE and intracellular staining with Foxp3 APC (eBioscience) as per manufacturer’s instructions. Flow cytometry was performed using a FACSCalibur^™ flow cytometry system and analyzed using CellQuest^™ software (BD Biosciences). Results are representative of at least three independent experiments.

ELISPOT assay

Immunospot plates (Cellular Technology, Cleveland) were coated with anti-mouse IFN-γ or IL-4. Recipient splenocytes plus irradiated (30 Gy) allogeneic splenocytes (0.5×10⁶ each) were cultured for 24 h (IFN-γ) or 48 h (IL-4), incubated with biotinylated anti-mouse IFN-γ or IL-4 and streptavidin horseradish peroxidase (DAKO), and developed with 3-amino-9-ethylcarbazole (Sigma-Aldrich). The resulting spots were counted on a computer-assisted immunospot image analyzer (T Spot Image Analyzer, Cellular Technology). mAbs were from Pharmingen.

Suppression Assay

Spleens and lymph nodes were harvested from B6 recipients of BALB/c skin 7 days post-transplantation, pooled single-cell leukocyte suspensions were prepared. CD4⁺ T cells were purified by magnetic bead negative selection (Miltenyi) with >90% purity. Then, CD25⁺ T cells were isolated from the enriched CD4⁺ according to the manufacturers protocol (Miltenyi). The purity was consistently >95% for CD4⁺CD25⁻ and CD4⁺CD25⁺ T cell preparations. Isolated 0.4×10⁶ CD4⁺CD25⁻ T cells (T_eff) were incubated with 0.4×10⁶ irradiated (30 Gy) BALB/c allogeneic splenocytes (stimulators) in the IFN-γ ELISPOT assay cultures as described above. Decreasing numbers of sorted CD4⁺CD25⁺ T cells (T_regs) starting at 0.4×10⁶ from mATG or Control Ig treated recipients were added to the cultures in 1:1, 1:1/4, 1:1/16 T_eff/T_reg ratios. After 24 hours, frequency of IFN-γ producing cells was determined by counting spots. The percentage inhibition of IFN-γ production was calculated by applying the following formula: $[({\mathrm{T}}_{\text{eff}}\text{IFN}-\mathrm{\gamma }\text{spots})-({\mathrm{T}}_{\text{eff}}+{\mathrm{T}}_{\text{reg}}\text{IFN}-\mathrm{\gamma }\text{spots})]/({\mathrm{T}}_{\text{eff}}\text{IFN}-\mathrm{\gamma }\text{spots})\times 100$.

Adoptive transfer of Thy1.2⁺Foxp3GFP^rep TCR-tg T cells

Adoptive transfer of Thy1.2⁺Foxp3GFP^rep TCR-tg T cells was performed as previously described in the ABM TCR-tg system (33). Briefly, spleens and lymph nodes were harvested from ABM-Foxp3GFP^rep mice and pooled single-cell leukocyte suspensions were prepared. CD4⁺ T cells were purified by magnetic bead positive selection (Miltenyi) with > 90 % purity. ABM-Foxp3GFP^rep CD4⁺ T cells express CD90.2 (Thy 1.2⁺) on their surface. More than 90 % of CD4⁺ T cells expressed the tg TCR, and approximately 2% of CD4⁺ T cells were GFP⁺. CD4⁺ isolated cells were then flow-sorted for GFP^+/− cells with a > 98% purity. 0.3 × 10⁶ Thy1.2⁺Foxp3GFP positive or negative cells were then injected intraperitoneally into Thy 1.1 B6 mice one day before skin transplantation (day −1). Recipients were sacrificed on day 7 following transplantation and draining axillary lymph nodes and spleens were collected, single-cell leukocyte suspensions were prepared and Thy1.2⁺Foxp3GFP^+/− TCR-tg T cells were identified by flow cytometry.

Immunohistochemistry staining

Skin graft samples were obtained at each time point in each treatment group, fixed in formalin 10% and stained with H&E for evaluation of cellular infiltration.

Statistics

Graft survival was expressed graphically by the Kaplan-Meier method, and statistical differences in survival between groups were assessed by the log-rank test. Student t test was used for comparison of means. P<0.05 was considered statistically significant.

RESULTS

mATG prolongs allograft survival by depleting naïve T cells and preserving natural T_regs

We first examined the depletion effect of mATG in naïve mice. As expected, mATG caused profound (> 85%) depletion of CD3⁺ (0.2×10⁶ vs. 2.8×10⁶; p < 0.00002), CD4⁺ (0.15×10⁶ vs. 1.8×10⁶; p < 0.00001) and CD8⁺ (0.05×10⁶ vs. 0.88×10⁶; p < 0.00002) cells in the lymph nodes compared to control immunoglobulin (Ig) treated mice. Similar changes were seen in the spleen [CD3⁺ (1.8×10⁶ vs. 19.8×10⁶; p < 0.00001), CD4⁺ (1.3×10⁶ vs. 12.9×10⁶; p < 0.00002) and CD8⁺ (0.5×10⁶ vs. 6.2×10⁶; p < 0.000003)]. There was no decrease in the number of CD19⁺ cells with mATG in the LN (4.2×10⁶ vs. 3.5×10⁶; p = NS) or spleen (29×10⁶ vs. 34×10⁶; p = NS). Next, we examined the kinetics of depletion and recovery of lymphocyte populations following mATG in the full MHC-mismatched skin transplant recipients treated with or without mATG induction therapy. We noted profound (> 80%) and persistent depletion of CD3⁺, CD4⁺ and CD8⁺ T cells in the graft draining lymph nodes particularly at day 7, but also at 14 and 28 days after transplantation compared to control Ig treated allograft recipients (Figure 1A). Similar changes were seen in spleen (Figure 1A). Despite the profound depletion of T cells, we noted relative preservation of number of Foxp3⁺ natural T_regs in the lymph nodes (LN) of naïve mice at day 7 post-mATG induction therapy (0.17×10⁶ vs. 0.14×10⁶; p = NS) and throughout the time-course in mATG treated skin allograft recipients compared to control Ig treated recipients, resulting in significantly higher percentage of T_regs at days 7 and 14 post-transplantation; and in due course declining to levels seen in control recipients by day 28 post-transplantation (Figure 1B). Similar changes were seen in spleen (Figure 1B). Next, we assessed the function of these preserved T_regs in vitro and noted that T_regs from mATG treated mice suppress alloantigen-specific IFN-γ production more efficiently compared to T_regs from Control Ig treated mice at each of the T_eff/T_reg ratios tested (Figure 1C). This preservation of functional T_regs with mATG however, led to only modest prolongation of the stringent full MHC-mismatched skin allograft survival in recipients treated with mATG induction therapy compared to Control Ig treated recipients [median graft survival time (MST) = 15.5 days vs. 9 days respectively; p < 0.0004] (Figure 1D).

Figure 1. mATG depletes naïve T cells, preserves functional natural T_regs and prolongs allograft survival.

A. Line graphs depicting absolute numbers of CD3⁺, CD4⁺, CD8⁺ and CD19⁺ cells in LN (left column) and spleen (right column) on days −1, 7, 14 and 28 in relation to transplantation with mATG induction or control Ig treatment. B. Line graphs depicting absolute numbers and percentage of T_regs in LN (left column) and spleen (right column) on days −1, 7, 14 and 28 in relation to transplantation with mATG induction or control Ig treatment. C. Bar graph depicting suppression of IFN-γ production, from a maximum of 140 IFN-γ spots by T_eff from transplanted untreated recipients, by T_regs from mATG or Control Ig treated recipients. D. Kaplan-Meier graphs depicting full MHC-mismatched skin allograft survival with mATG induction or control Ig treatment.

mATG synergizes with CTLA4Ig and SRL to prolong allograft survival

In an attempt to overcome the rejection of skin allograft despite profound T cell depletion with mATG induction therapy, we initially combined CTLA4Ig with mATG and noted significant prolongation of full MHC-mismatched skin allograft survival in recipients treated with mATG induction therapy and short course CTLA4Ig compared to recipients treated with mATG induction therapy alone (MST 36 days vs. 15.5 days respectively; p < 0.003) (Figure 2A). Next, we examined the effect of adding short-course sirolimus (SRL) to this regimen. Interestingly, there was significant prolongation of graft survival in recipients treated with combined mATG induction, short-course CTLA4Ig and SRL therapy compared to mATG induction and short-course CTLA4Ig (MST 58 vs. 36 days respectively; p < 0.001) or short-course CTLA4Ig and SRL therapy without mATG induction (MST 58 vs. 17 days respectively; p < 0.0007) (Figure 2A). We then examined the allospecific cytokine production, by splenocytes from full MHC-mismatched skin allograft recipients, 7 days after transplantation. The inhibition of graft rejection with mATG induction therapy alone or in combination with short-course CTLA4Ig, SRL, or CTLA4Ig + SRL was associated with almost complete inhibition of IFN-γ and significant inhibition of IL-4 by splenocytes from these recipients (Figure 2B). Histological examination showed well-preserved grafts with no evidence of rejection in grafts harvested from recipients treated with combination therapy compared to control Ig or mATG induction therapy alone (Figure 2C).

Figure 2. Combination therapy synergizes to prolong graft survival, inhibit allospecific cytokine production and preserves graft histology.

A. Kaplan-Meier graphs depicting full MHC-mismatched skin allograft survival with mATG in combination with CTLA4Ig and/or SRL or mATG induction alone. B. Bar graphs depicting production of allospecific cytokine IFN-γ (left panel) and IL-4 (right panel), by ELISPOT assay with splenocytes from full MHC-mismatched skin allograft recipients, 7 days after transplantation with various treatments. C. H and E stained light micrographs depicting features of acute rejection in grafts harvested 7 days after transplantation from recipients treated with control Ig (left upper panel); minimal infiltration of grafts harvested 7 days after transplantation (right upper panel) compared to features of acute rejection in grafts harvested 14 days after transplantation from recipients treated with mATG induction therapy alone (left lower panel); well-preserved graft with no evidence of rejection in grafts harvested from recipients treated with combination therapy (right lower panel).

Beneficial effect of mATG alone or in combination with CTLA4Ig and/or SRL on allograft survival is dependent on CD25⁺ T cells

To determine if prolonged allograft survival with mATG alone or in combination with CTLA4Ig and/or SRL is dependent on T_regs, we administered anti-CD25 mAb to the recipients prior to full MHC-mismatched skin transplantation. CD25⁺ T cell depletion prior to transplantation led to abrogation of the beneficial effect of mATG induction therapy alone or in combination with CTLA4Ig and/or SRL on allograft survival (Figure 3A), indicating that T_regs are required for the graft prolonging effect of mATG. Next, to investigate whether T_reg generation by the thymus or peripheral conversion of naïve T cells to T_regs and/or expansion of preexisting T_regs in the periphery is responsible for the beneficial effects of mATG, we performed thymectomy in the recipients 3 weeks prior to transplantation. Interestingly, thymectomy prior to transplantation did not affect the graft prolonging effect of mATG induction therapy whilst CD25⁺ T cell depletion in thymectomized recipients did abrogate the graft prolonging effect of mATG (Figure 3B), indicating that thymic export of T_regs is not required, whilst the presence of peripheral T_regs (expansion of preexisting T_regs and/or peripheral conversion of naïve T cells to T_regs) is required for the beneficial effect of mATG on allograft survival.

Figure 3. Prolonged allograft survival with mATG alone or in combination with CTLA4Ig and/or SRL is dependent on CD25⁺ T cells.

A. Kaplan-Meier graphs depicting full MHC-mismatched skin allograft survival with CD25 pre-depletion and mATG in combination with CTLA4Ig and/or SRL or mATG induction alone. B. Kaplan-Meier graphs depicting full MHC-mismatched skin allograft survival in thymectomized or non-thymectomized recipients with or without CD25 pre-depletion and mATG induction therapy.

CTLA4Ig and Sirolimus inhibit effector-memory differentiation and promote regulation when combined with mATG

To further understand the mechanisms of graft prolongation with mATG-based combination therapy, we utilized the TCR-transgenic-reporter mouse model. 0.3×10⁶ CD4⁺Thy1.2⁺ABMFoxp3GFP⁺ or CD4⁺Thy1.2⁺ABMFoxp3GFP⁻ cells were injected in Thy1.1⁺C57BL/6 mice with or without bm12 skin transplantation. In these mice we first studied the effect of mATG induction therapy on antigen-specific T_reg (Thy1.2⁺Foxp3GFP⁺) population by analyzing the adoptively transferred CD4⁺Thy1.2⁺ABMFoxp3GFP⁺ cells. In keeping with our previous in vitro observations with Thymoglobulin (19), in the case of CD4⁺Thy1.2⁺ABMFoxp3GFP⁺ cell-adoptive transfer, we found mATG induction therapy preserved Thy1.2⁺Foxp3GFP⁺ cells and promoted their expansion in vivo compared to control Ig treatment (0.4×10⁶ vs. 0.04×10⁶ respectively; p < 0.03) (Figure 4A). Further, in the case of CD4⁺Thy1.2⁺ABMFoxp3GFP⁻ cell-adoptive transfer, mATG significantly promoted conversion of Thy1.2⁺Foxp3GFP⁻ cells to Thy1.2⁺Foxp3GFP⁺ cells compared to control Ig treatment (0.53×10⁶ vs. 0.26×10⁶ respectively; p < 0.04) (Figure 4B). Despite the preferential preservation of Thy1.2⁺Foxp3GFP⁺ cells and conversion of Thy1.2⁺Foxp3GFP⁻ cells to Thy1.2⁺Foxp3GFP⁺ cells with mATG, the effect of mATG induction therapy on skin allograft survival was modest (Figure 1D). We reasoned that rejection might be occurring due to an increase in the T_em (Thy1.2⁺Foxp3GFP⁻CD44^hiCD62L^lo) populations in response to the alloantigen and/or homeostatic proliferation in the aftermath of T cell depletion with mATG induction therapy. Interestingly, combining mATG induction therapy with short-course CTLA4Ig and SRL either individually or together affords significant synergy in promoting allograft survival (Figure 2A). To understand the mechanisms of this synergy, we next studied the effect of mATG induction therapy in combination with CTLA4Ig and SRL individually or together on antigen specific T_reg and T_em populations and evaluated the T_reg/T_em ratio in various treatment groups 7 days after bm12 skin transplantation utilizing the ABM-Foxp3GFP^rep model. Interestingly we found that combining CTLA4Ig or SRL with mATG induction therapy led to not only an increase in Thy1.2⁺Foxp3GFP⁺ cells by preservation and conversion but also a significant inhibition of generation of Thy1.2⁺Foxp3GFP⁻ cells with the T_em phenotype resulting in a favorable T_reg/T_em ratio (Figure 4C).

Figure 4. CTLA4Ig and Sirolimus inhibit effector-memory differentiation and promote favourable T_reg/T_em balance when combined with mATG.

A. Bar graphs depicting absolute numbers of Thy1.2⁺Foxp3GFP⁺ or Thy1.2⁺Foxp3GFP⁻ populations 7 days after transplantation, in both lymph nodes and spleen from recipients adoptively transferred with ABM TCR-tg Thy1.2⁺Foxp3GFP⁺ cells and treated with mATG induction therapy alone, in combination with CTLA4Ig and/or SRL or control Ig. Representative histograms of Thy1.2⁺Foxp3GFP⁺ cells from recipients in each treatment group are shown below respective bar graphs. There is significant preservation of Thy1.2⁺Foxp3GFP⁺ T_reg cells and inhibition of Thy1.2⁺Foxp3GFP⁻CD44^hiCD62L^low T_em cells in recipients treated with either mATG induction therapy alone or in combination with SRL, CTLA4Ig or SRL+CTLA4Ig (*/#p < 0.05 for each group compared with Control Ig treatment). There is no significant difference in either the T_reg or T_em cells between Control Ig, CTLA4Ig and SRL alone treatment groups (p = NS between each group). B. Bar graphs depicting absolute numbers of Thy1.2⁺Foxp3GFP⁺ or Thy1.2⁺Foxp3GFP⁻ populations 7 days after transplantation, in both lymph nodes and spleen from recipients adoptively transferred with ABM TCR-tg Thy1.2⁺Foxp3GFP⁻ cells and treated with mATG induction therapy alone, in combination with CTLA4Ig and/or SRL or control Ig. Representative histograms of Thy1.2⁺Foxp3GFP⁺ cells from recipients in each treatment group are shown below respective bar graphs. There is significant conversion of Thy1.2⁺Foxp3GFP⁻ cells to Thy1.2⁺Foxp3GFP⁺ T_reg cells in recipients treated with either mATG induction therapy alone or in combination with SRL or CTLA4Ig (*p < 0.02 for each group compared with Control Ig treatment). There is significant inhibition of Thy1.2⁺Foxp3GFP⁻CD44^hiCD62L^low T_em cells in recipients treated with mATG alone, or in combination with CTLA4Ig or SRL+CTLA4Ig (#p < 0.02 for each group compared with Control Ig treatment). There is no significant difference in either the T_reg or T_em cells between Control Ig, CTLA4Ig, SRL alone treatment groups (p = NS between each group). C. Scatter plots depicting T_reg/T_em ratios 7 days after transplantation, in both lymph nodes and spleen from recipients adoptively transferred with ABM TCR-tg Thy1.2⁺Foxp3GFP⁺ (left panel) or Thy1.2⁺Foxp3GFP⁻ (right panel) cells and treated with mATG induction therapy alone, in combination with CTLA4Ig and/or SRL or control Ig.

mATG preserves T_regs which limit homeostatic proliferation of memory cells following T cell depletion and the prolonged allograft survival is dependent on favorable T_reg/T_em balance

We investigated whether the T_regs spared by mATG limit homeostatic proliferation of T_em cells and affect graft survival, by pre-depleting CD25⁺ cells and then transplanting B6 mice with full MHC-mismatched BALB/c skin grafts or not and treated with mATG induction therapy alone or in combination with CTLA4Ig+SRL. We found that T cells that emerge in the aftermath of depletion with mATG are more skewed towards a T_em (CD44^hiCD62L^lo) phenotype in both naïve and transplanted mice treated with mATG or mATG+CTLA4Ig+SRL when CD25⁺ cells are pre-depleted compared to controls (with no CD25⁺ cell depletion) (Table 1). Consequently, T_reg/T_em balance in naïve and transplanted mice is switched in favor of T_em cells with CD25 pre-depletion and results in accelerated allograft rejection, most dramatically in the mATG+CTLA4Ig+SRL group (Table 1).

Table 1.

mATG preserves T_regs which limit homeostatic proliferation of memory cells following T cell depletion

Therapy	Transplant	CD25 pre-depletion	Treg (n)	Teff (n)	Treg/Teff Ratio	MST (days)
mATG	No	No	166500±61	267365±2	0.6	NA
mATG	No	Yes	41585±6 p<0.03	378833±25 p<0.007	0.1; p<0.02	NA
mATG	Yes	No	98214±3	140884±5	0.7	15.5
mATG	Yes	Yes	17377±6 p<0.02	379366±12 p<0.03	0.05; p<0.00001	12.5; p<0.003
mATG+CTLA4Ig+SRL	Yes	No	26441±2	31518±14	0.9	58
mATG+CTLA4Ig+SRL	Yes	Yes	14615±4 p<0.01	200110±60 p<0.009	0.07; p<0.0007	25; p<0.001

All p values are in comparison with no CD25 depletion in the respective group

DISCUSSION

mATG, like the other preparations of anti-thymocyte globulins, depletes T cells effectively both in naïve and transplanted animals (Figure 1A). In keeping with our previous studies with Thymoglobulin (19), T_regs were preferentially spared from depletion with mATG (Figure 1B). Minamimura et al. reported that naïve CD25⁺ T cells are resistant to depletion by anti-lymphocyte serum and this is thought to be due to their higher expression of anti-apoptotic gene Bcl-X_L. Interestingly, on further analysis, Bcl-X_L protein expression by flow cytometry appeared to be highest in Foxp3⁺ cells regardless of IL-2R α-chain (CD25) expression (34), explaining the relative sparing of Foxp3⁺ cells from the depletion effect of mATG in our studies (Figure 1B).

The fate of the alloimmune response is dependent on the balance between alloreactive T_em and T_regs in vivo. mATG prolonged allograft survival by profound T cell depletion and preferential sparing of T_regs (Figure 1B and 1C) and this effect was independent of recent thymic immigrants but dependent on expansion of preexisting T_regs and/or peripheral conversion of naïve T cells to T_regs (Figure 3B). Of note, the prolongation of graft survival with mATG treatment was modest. This is of particular clinical relevance since homeostatic proliferation is thought to be a barrier to transplantation tolerance induction in both reconstituted scid mice and WT mice treated with T cell-depleting antibodies (35). Further, in an appropriate environment, Foxp3⁺ T_regs can differentiate into effector cells (36–38). Recently it was demonstrated that Foxp3⁺ cells consists of a heterogeneous population and Foxp3⁺CD25⁻ cells retain developmental plasticity and preferentially proliferate into cells with effector phenotype in a lymphopenic or inflammatory environment but maintain some regulatory function (39). On the other hand, it has been noted that naïve T cells readily express Foxp3 in response to alloantigen (40). In keeping with this we noted significant conversion of adoptively transferred T_regs into T_em cells and vice versa (Figure 4A and 4B) in the context of transplantation and treatment with Control Ig or mATG. However, our data suggests that mATG, unlike Control Ig and other T cell depleting therapies (21), preserves T_regs and limits conversion to/expansion of T_em cells (Figure 1B and 4). Further, these preserved T_regs themselves undergo homeostatic proliferation as suggested by an elegant study that showed CD4⁺CD25⁺ T_reg cells are also capable of undergoing homeostatic proliferation in a lymphopenic environment and acquire memory-like phenotype (CD44^highCD45RB^low expression) with loss of CD25 and CD62L expression but exhibit substantially augmented suppressive function (41). Furthermore, T_regs are thought to play a major role in regulating lymphopenia-induced T cell proliferation and acquisition of T_em phenotype (42). Moreover, ex-vivo expanded antigen-specific natural T_regs synergize with host T cell depletion and promote long-term allograft acceptance (43). In keeping with these, our data suggests that mATG preserves CD25⁺ natural T_regs which limit skewing of T cell repertoire toward T_em phenotype among the recovering T cells by inhibiting both lymphopenia-induced and alloantigen-induced proliferation of T cells (Table 1).

B7/CD28 blockade with CTLA4Ig, on the other hand, has also been demonstrated to be effective in controlling the homeostasis of T_em cells by reducing expansion of central memory cells and migration of activated memory CD4 cells to the periphery (44). Further, costimulation requirement of alloreactive T cells is critically dependent on their precursor frequency, where recipients with low CD4⁺ and CD8⁺ donor-reactive T cell frequencies exhibited long-term skin graft survival upon CD28/CD154 blockade, whereas simultaneously raising the frequency of CD4⁺ T cells to 0.5% and CD8⁺ T cells to 5% precipitated graft rejection despite CD28/CD154 blockade (45). These, taken together with our data (Figure 2A and 4), indicate that T cell depletion with mATG may limit the precursor frequency of alloreactive T cells and render them more susceptible to costimulation blockade. Further, adding SRL to costimulation (B7-CD28 and CD40-CD40L pathways) blockade prolonged skin and heart graft survival in murine models by two main mechanisms: inhibition of proliferation of alloreactive T cells and induction of apoptosis of dividing alloreactive T cells (46). A more recent study in human renal transplant recipients demonstrated that combined treatment with ATG, SRL and mycophenolate mofetil was associated with a progressive increase in T_reg population both in the periphery and in the graft infiltrating cells with concomitant decrease of donor responsiveness in vitro (47). Moreover, both CTLA4Ig and SRL inhibit lymphopenia-induced and alloantigen-induced proliferation of T cells (48, 49). Consistent with these data, we observed significant synergy on combining mATG with CTLA4Ig and SRL individually or together (Figure 2A and 4). The beneficial effects of this combination therapy were partly dependent on CD25⁺ T_regs indicating that T_regs are required for the graft prolonging effect of combination therapy involving mATG (Figure 3A). These data emphasize the advantages of this novel strategy of combining mATG with T cell costimulatory blockade and SRL to preserve T_regs and limit homeostatic proliferation of effector and memory T cells to promote regulation in human organ transplantation.

In summary, our data demonstrate that the combination of mATG with costimulation blockade and sirolimus represents an effective strategy in preventing allograft rejection and promoting graft acceptance in a full MHC-mismatched stringent skin transplant model. The effectiveness of this combination strategy depends not only on the depletion effects of mATG but also on generating a favourable T_reg/T_em balance. These data will serve as the basis of further studies in non-human primates to develop a clinically applicable protocol to promote regulation in transplantation.