Donor Requirements for Regulatory T Cell Suppression of Murine Graft-versus-Host Disease

Abstract

Adoptive transfer of freshly isolated natural occurring CD4(+)CD25(+)Foxp3(+) regulatory T cells (Treg) prevents graft-versus-host disease (GVHD) in several animal models and following hematopoietic cell transplantation (HCT) in clinical trials. Donor-derived Treg have been mainly used, as they share the same MHC with CD4(+) and CD8(+) conventional T cells (Tcon) that are primarily responsible for GVHD. Third party-derived Treg are a promising alternative for cellular therapy, as they can be prepared in advance, screened for pathogens and activity, and banked. We explored MHC disparities between Treg and Tcon in HCT to evaluate the impact of different Treg populations in GVHD prevention and survival. Third-party Treg and donor Treg are equally suppressive in ex vivo assays, whereas both donor and third-party but not host Treg protect from GVHD in allogeneic HCT, with donor Treg being the most effective. In an MHC minor mismatched transplantation model (C57BL/6 → BALB/b), donor and third-party Treg were equally effective in controlling GVHD. Furthermore, using an in vivo Treg depletion mouse model, we found that Treg exert their main suppressive activity in the first 2 d after transplantation. Third-party Treg survive for a shorter period of time after adoptive transfer, but despite the shorter survival, they control Tcon proliferation in the early phases of HCT. These studies provide relevant insights on the mechanisms of Treg-mediated protection from GVHD and support for the use of third-party Treg in clinical trials.

Figure 1. Donor derived or third-party Treg equally suppress Tcon proliferation in vitro

A: C57BL/6 Tcon were cultured with BALB/c irradiated splenocytes and C57BL/6 or FVB/N Treg at 1:1, 1:2, 1:4, 1:8 and 1:16 Treg/Tcon ratios. B: Either BALB/c Tcon or FVB/N Tcon were cultured with C57BL/6 irradiated splenocytes and BALB/c Treg at a ratio of 1:2. Normalized histograms are reported to allow comparison between different Tcon proliferations. For statistical analysis 2-tailed student t test was used, * p < 0.05, ** p < 0.01. One representative experiment is shown. Experiments were repeated three times.

Figure 2. Donor type and third-party Treg reduce Tcon proliferation in vivo

Transplantation scheme is shown as described in the text (A). In vivo BLI data are presented from representative animals at day +7 (B), day +15 (C) and day +29 (D) after transplantation. Each data point is the measurement of luminescence from one mouse (photons/sec/cm²). Mice that received only C57BL/6 luc+ Tcon ( ), Tcon and donor type C57BL/6 Treg (◆), Tcon and third-party FVB/N Treg (◇) and Tcon and host type BALB/C Treg (△) are shown. E: Images of representative animals from each group (minimum 5 mice/group) are shown. For statistical analysis 2-tailed student t test was used, * p < 0.05, *** p < 0.001, ns = not significant. Data are representative of 3 independent experiments.

Figure 3. Third-party Treg survive for a shorter period of time compared to donor-derived Treg following in vivo transfer

The absolute numbers of H2K^b+CD45.1+CD4+FoxP3+ donor type (●), H2K^q+CD45.1+CD4+FoxP3+ third-party (□) or H2K^d+Thy1.1+CD4+FoxP3+ host type (◆) reisolated Treg are shown in lymph nodes (A), spleen (B), peripheral blood (C) and liver (D) at day 6 and at day 12 after transplantation. No H2K^d+Thy1.1+CD4+FoxP3+ host type Treg are shown at day 12 as only few mice survived. For statistical analysis third-party Treg and host Treg have been individually compared with donor Treg (2-tailed student t test, * p < 0.05, ns = not significant). Pooled data from two consecutive experiments are shown.

Figure 4. Infiltration of effector memory CD4+CD62L-CD44+ donor T cells is reduced after donor type or third-party Treg adoptive transfer

A: Numbers of live H2K^b+CD4+CD62L-CD44+ cells found in liver of BALB/c mice that received only donor C57BL/6 Tcon (white), Tcons and host type BALB/c Treg (black), Tcon and donor type C57BL/6 Treg (grey) and Tcon and third-party FVB/N Treg (white dotted) are shown after analysis of the samples at day 6 after transplantation. A representative gating strategy of CD4+CD62L-CD44+ cells is also reported. B: H2K^b+CD4+CD62L+ percentage over H2K^b+CD4+ cells is shown in lymph nodes. Histograms of CD62L expression on H2K^b+CD4+ cells in lymph nodes of mice that received only donor C57BL/6 Tcon (no dashing), Tcon and host type BALB/c Treg (dotted), Tcon and donor type C57BL/6 Treg (short dashed) and Tcon and third-party FVB/N Treg (long dashed) and a representative gating strategy of CD4+CD62L+ cells are also reported. C: LAG3⁺CD4⁺FoxP3⁺ cell percentage over CD4⁺FoxP3⁺ cells is reported in lymph nodes and spleen after analysis of the samples at day 6 after transplantation is reported. D: IFNγ and IL-10 concentrations (pg/ml) in the sera collected at day 6 after transplantation are shown. White histograms are representative of mice that received only donor type Tcon, black histograms are representative of mice that received Tcon and host type Treg, grey histograms are representative of mice that received Tcon and donor type Treg and white dotted histograms are representative of mice that received Tcon and third-party Treg. For statistical analysis 2-tailed student t test was used, *** p < 0.0005, ** p < 0.005, * p < 0.05, ns p=not significant. Pooled data from two consecutive experiments are shown.

Figure 5. Donor and third-party but not host Treg prevent GvHD in allogeneic transplantation

Survival (A), weight variation (B) and GvHD score (C) of allogeneic transplanted BALB/c recipient mice after injection of only C57BL/6 Tcon ( ), Tcon and donor type C57BL/6 Treg (◆), Tcon and third-party FVB/N Treg (◇) and Tcon and host type BALB/c Treg (△) are shown. Mice that received TCD BM only (dashed line, ■) and mice that were lethally irradiated but not transplanted (dotted line, ●) were used as controls. For statistical analysis of mouse survival Kaplan-Meier test was used, for weight variation and GvHD score 2-way ANOVA test was used, * p < 0.05, ** p < 0.01, *** p < 0.001. Data is representative of one of three experiments (minimum 5 mice/group).

Figure 6. Treg equally protect from GvHD lethality in a MHC minor mismatched mouse model

Transplantation scheme (A). Survival (B), weight (C) and GvHD score (D) of MHC minor mismatched transplanted BALB/b recipient mice after injection of C57BL/6 Tcon ( ), Tcon and donor type C57BL/6 Treg (◆) and Tcon and third-party FVB/N Treg (◇) are shown. Mice that received TCD BM only (dashed line, ■) and mice that were lethally irradiated but not transplanted (dotted line, ●) were used as controls. For statistical analysis of mouse survival Kaplan-Meier test was used, for weight variation and GvHD score 2-way ANOVA test was used, * p < 0.05, ns = not significant. Pooled data from two experiments are shown.

Figure 7. Treg limit GvHD lethality in the very early phase of transplantation

Transplantation scheme is shown as described in the text (A). Treg were efficiently depleted after DT treatment. Percentages of GFP⁺ Treg in spleen and peripheral blood of mice that did not receive Treg transfer (Tcon), that received Treg transfer and DT at day −2 and −1 (Tcon + Treg DT −2), that received Treg transfer and DT at day 0 and +1 (Tcon + Treg DT 0) and that received Treg transfer with no DT treatment (Tcon + Treg no DT) are shown. Representative samples are reported. GFP⁺ Treg were detectable only in the last group (red arrows, B). Images of representative mice that received Tcon + luc⁺ Treg + DT at day −2 and −1 (Tcon + Treg DT −2) or Tcon + luc⁺ Treg + DT at day 0 and +1 (Tcon + Treg DT 0) or only Tcon + luc⁺ Treg (Tcon + Treg no DT) are reported. Images were taken at day 0 and at day +6. Data demonstrate that Treg were effectively depleted first in mice that received DT at day −2 and −1 and successively in mice that received DT at day 0 and +1, while Treg were in vivo proliferating in mice that did not receive DT treatment (C). Survival (D), weight (E) and GvHD score (F) of transplanted BALB/c recipient mice after injection of C57BL/6 Tcon ( ), Tcon + C57BL/6 albino FoxP3^DTR/GFP/luc Treg + DT at day −2 and −1 (▽), Tcon + Treg + DT at day 0 and +1 (◇) and Tcon + Treg with no DT (◆) are shown. Mice that received TCD BM only (dashed line, ■) and mice that were lethally irradiated but not transplanted (dotted line, ●) were used as controls. For statistical analysis of mouse survival Kaplan-Meier test was used, for weight variation and GvHD score 2-way ANOVA test was used, * p < 0.05, ns = not significant. Pooled data from two experiments are shown.