mTOR Inhibition Suppresses Posttransplant Alloantibody Production Through Direct Inhibition of Alloprimed B Cells and Sparing of CD8+ Antibody-Suppressing T cells

Abstract

Background: De novo alloantibodies (donor-specific antibody) contribute to antibody-mediated rejection and poor long-term graft survival. Because the development of donor-specific antibody is associated with early graft loss of cell transplants and reduced long-term survival of solid organ transplants, we hypothesized that conventional immunosuppressives, calcineurin inhibitors (CNi), and mammalian target of rapamycin inhibitors (mTORi), may not be as effective for suppression of humoral alloimmunity as for cell-mediated immunity.

Methods: Wild-type or CD8-depleted mice were transplanted with allogeneic hepatocytes. Recipients were treated with mTORi and/or CNi and serially monitored for alloantibody and graft survival. The direct effect of mTORi and CNi on alloprimed B cell function was investigated in Rag1 mice adoptively transferred with alloprimed IgG1 B cells. The efficacy of mTORi and/or CNi to suppress CD8-mediated cytotoxicity of IgG1 B cells was evaluated in in vitro and in vivo cytotoxicity assays.

Results: Mammalian target of rapamycin inhibitors, but not CNi, reduced alloantibody production in transplant recipients, directly suppressed alloantibody production by alloprimed IgG1 B cells and delayed graft rejection in both low and high alloantibody producers. Combination treatment with mTORi and CNi resulted in loss of the inhibitory effect observed for mTORi monotherapy in part due to CNi suppression of CD8 T cells which downregulate alloantibody production (CD8 TAb-supp cells).

Conclusions: Our data support that mTORi is a potent inhibitor of humoral immunity through suppression of alloprimed B cells and preservation of CD8 TAb-supp cells. In contrast, alloantibody is readily detected in CNi-treated recipients because CNi does not suppress alloprimed B cells and interferes with downregulatory CD8 TAb-supp cells.

Figure 1. Superior efficacy of mTORi compared to CNi for suppression of posttransplant alloantibody production and prolongation of allograft survival

A) C57BL/6 (H-2^b, wild-type; WT) mice were transplanted with allogeneic FVB/N (H-2^q) hepatocytes. On days 0-14 posttransplant cohorts of recipients were treated with 1 mg/kg/d mTORi or CNi or a 5% DMSO-vehicle control. Alloantibody was quantified by analyzing recipient serum by antibody titer on day 14 posttransplant. mTORi-treated recipients had significantly reduced alloantibody (n=4, titer=15±3) compared with DMSO-treated recipients (n=4, titer=113±31, *p<0.001) and CNi-treated recipients (n=4, titer=94±21, ^†p=0.001). CNi-treated and control recipients had comparable alloantibody titers. B) Recipient graft survival was assessed through serum hA1AT levels. Treatment with either mTORi (MST=24.5 days, *p=0.005) or CNi (MST=26 days, ^†p=0.006) resulted in a significant prolongation of hepatocyte survival compared with control recipients (MST=12 days). C) CD8-depleted (days -2, -1) recipients (high-alloantibody producers) were similarly transplanted with allogeneic hepatocytes and treated with mTORi, CNi, or a DMSO-vehicle control on days 0-14 posttransplant. mTORi-treated recipients had significantly reduced levels of alloantibodies (n=4, titer=28±8) compared to the DMSO-treated (n=4, titer=263±55, *p<0.001) and CNi-treated recipients (n=5, titer=220±49, ^†p<0.001). In contrast, CNi-treated and control recipients had comparable alloantibody titers. D) mTORi (MST=25.5 days) prolonged hepatocyte survival in high alloantibody producers compared to DMSO-treated recipients (MST=14 days, *p=0.02). CNi-treated (MST=12 days) and control recipients had similar graft survival. Data is representative of triplicate experiments.

Figure 2. mTORi (but not CNi) treatment suppresses in vivo alloantibody production by alloprimed IgG1⁺ B cells

Rag1 KO mice were adoptively transferred (AT) with alloprimed IgG1⁺ B cells (10×10⁶ cells i.v.) and injected with FVB/N splenocytes as an antigen source (20×10⁶ cells, i.p.). As a control, a cohort of mice received AT of only naive B cells and exhibited low levels of alloantibody (data not shown). On day 1 relative to AT, mice were given anti-CD40 mAb (200 mg) and IL-4 (0.625 mg) to mimic CD4⁺ T cell “help” normally absent in these mice. Cohorts of mice were treated with DMSO-vehicle control, mTORi (1 mg/kg/d), or CNi (1 mg/kg/d) on days 1-7. Alloantibody was quantified by analyzing recipient serum by antibody titer on day 7 posttransplant. AT of primed B cells resulted in high levels of alloantibody production in Rag1 KO mice (n=4, 500±100). Treatment with mTORi resulted in significant inhibition of in vivo alloantibody production by alloprimed B cells (n=5, titer=70±12) compared to DMSO-treated (*p<0.001) and CNi-treated mice (n=6, titer=500±100, ^†p<0.001). CNi-treated recipients exhibited no reduction in alloantibody compared with control mice. Data was combined from duplicate experiments.

Figure 3. mTORi-mediated suppression of in vivo alloantibody production is negated when combined with CNi treatment

WT mice received hepatocyte transplants from FVB/N donors. Recipients were treated on days 0-14 with mTORi (0.25 mg/kg/d), CNi (1 mg/kg/d), a combination of both mTORi and CNi, or DMSO vehicle control. Some recipients were CD8-depleted (d-2,-1) and treated with DMSO control, mTORi (1 mg/kg/d), CNi (1 mg/kg/d), combined mTORi (1 mg/kg/d) and CNi (1 mg/kg/d). A) Alloantibody was quantified by analyzing recipient serum alloantibody titer on day 14 posttransplant. In both WT and CD8-depleted recipients CNi-treatment did not result in a reduction of alloantibody (WT: n=4, 94±21; CD8-depleted: n=5, 220±49) compared with the control (WT: n=4, titer=113±31; CD8-depleted: n=4, titer=263±55). In both WT and CD8-depleted recipients mTORi-treated recipients had significantly reduced alloantibody (WT: n=4, titer=25±5; CD8-depleted: n=4, titer=28±8) compared to the control (p=0.002, p=0.0001 respectively). Unexpectedly, combination treatment with both mTORi and CNi resulted in significantly higher levels of alloantibody in both WT (n=6, titer=75±11) and CD8-depleted recipients (n=3, titer=133) compared to recipients treated with mTORi alone (*p=0.005, ^†p=0.003 for WT and CD8-depleted groups respectively). B) Recipient graft survival was assessed during treatment with CNi and/or mTORi and for an additional two weeks following cessation of immunosuppressive therapy. While mTORi treatment significantly prolonged graft survival (MST=24.5 days, *p=0.005) in WT recipients, combination treatment with mTORi and CNi did not (MST=12.5 days, p=ns) when compared with DMSO-treated control recipients (MST=12 days). C) In contrast, in CD8-depleted recipients both treatment with mTORi alone (MST=25.5), or combination treatment with mTORi and CNi (MST=35.5 days) significantly prolonged hepatocyte survival compared to DMSO-treated controls (MST=14 days, *p=0.02, ^†p=0.002 respectively). Data was combined from triplicate experiments.

Figure 4. CNi (but not mTORi) inhibits CD8-mediated killing of alloprimed B cells in vitro

WT and CD8 KO mice were transplanted with FVB/N hepatocytes. Naive and alloprimed CD8⁺T cells and B cells were isolated (day 7 posttransplant for alloprimed cells). CD8⁺ T cells and B cells were co-cultured at a 10:1 ratio for 4h with mTORi or CNi (10 nM) or 5% DMSO vehicle control. Prior to co-culture, B cells were stained with CFSE. Propidium iodide staining was used to determine apoptosis within CFSE-stained B cells (percent PI positive by flow cytometry). A) Bulk alloprimed CD8⁺T cells (“aCD8”) caused significant apoptosis of alloprimed (IgG1⁺) B cells (n=3, 11.4±1.2%) compared to naive CD8⁺ T cells (“nCD8”, n=3, 3.5±0.2%). mTORi treatment did not inhibit CD8-mediated apoptosis of primed B cells (n=3, 11.1±2.2%). However, CNi inhibits CD8-mediated cytotoxicity to B cells (n=3, 3.5±0.8%) compared to DMSO-treated (*p=0.012) and mTORi-treated co-cultures (^†p=0.02). Data is representative of triplicate experiments. B) Representative flow cytometry plots are shown. To analyze for apoptotic B cells in the co-cultures, cells were first gated on lymphocytes and then subsequently on the CFSE⁺ labeled-B cells. Uptake of propidium iodide (PI) by CFSE⁺ labeled B cells was analyzed.

Figure 5. CNi (but not mTORi) inhibits CD8-mediated killing of alloprimed B cells in vivo

WT transplant recipients were treated with mTORi (1 mg/kg/d), CNi (1 mg/kg/d), combined mTORi (1 mg/kg/d) and CNi (1 mg/kg/d), or 5% DMSO vehicle control on days 1-7 posttransplant. In vivo cytotoxicity of B cells was determined by transferring equal numbers of CFSE-labeled alloprimed IgG1⁺ (CFSE^hi) and naïve (CFSE^lo) B cells into WT mice 7 days after transplant. The spleen of recipient mice was collected 18 hours later and evaluated for recovery of both naive and alloprimed IgG1⁺ B cells to calculate the specific lysis of alloprimed IgG1⁺ B cells. A) WT transplant recipient mice treated with DMSO vehicle control (n=4, 61.5±5.3%) or mTORi (n=5, 65.2±7.4%) exhibited relatively high cytotoxicity, while those receiving CNi (n=5, 27.5±5.4%) exhibited significantly less in vivo cytotoxicity of alloprimed IgG1⁺ B cells compared to DMSO-treated (*p=0.009) and mTORi-treated recipients (^ǂp=0.005). Combined mTORi (1 mg/kg/d) and CNi (1 mg/kg/d) treatment (n=3, 31.3±6.1%) had similar in vivo cytotoxicity of alloprimed IgG1⁺ B cells compared to CNi-treated recipients. Data was combined from duplicate experiments. B) Representative flow cytometry plots are shown. To analyze the proportion of CFSE-labeled alloprimed (CFSE^hi) and naïve (CFSE^lo) B cells remaining 18 hours after adoptive transfer, splenocytes from recipient mice were analyzed by first gating on lymphocytes and then subsequently on the CFSE⁺ labeled-B cells.