Instability of Foxp3 expression limits the ability of induced regulatory T cells to mitigate graft versus host disease

Abstract

Purpose: Graft versus host disease (GVHD) is the major complication of allogeneic bone marrow transplantation (BMT) and limits the therapeutic efficacy of this modality. Although the role of natural T-regulatory cells (nTreg) in attenuating GVHD has been extensively examined, the ability of induced T-regulatory cells (iTreg) to mitigate GVHD is unknown. The purpose of this study was to examine the ability of in vitro and in vivo iTregs to abrogate GVHD.

Experimental design: We examined the ability of in vitro differentiated and in vivo iTregs to reduce the severity of GVHD in a clinically relevant mouse model of BMT. The effect of blockade of interleukin (IL) 6 signaling on the efficacy of these Treg populations was also studied.

Results: In vitro differentiated iTregs fail to protect mice from lethal GVHD even when administered at high Treg:effector T-cell ratios. Lack of GVHD protection was associated with loss of Foxp3 expression and in vivo reversion of these cells to a proinflammatory phenotype characterized by secretion of IFN-γ. Phenotypic reversion could not be abrogated by blockade of IL-6 signaling or by in vitro exposure of iTregs to all-trans retinoic acid. In contrast, the in vivo induction of iTregs was significantly augmented by IL-6 blockade and this resulted in reduced GVHD.

Conclusion: Instability of Foxp3 expression limits the utility of adoptively transferred iTregs as a source of cellular therapy for the abrogation of GVHD. Blockade of IL-6 signaling augments the ability of in vivo iTregs to prevent GVHD but has no effect on in vitro differentiated iTregs.

Figure 1. Induced Tregs are equipotent to nTregs in the suppression of alloreactive T cell responses.

(A). Sorted CD4⁺ EGFP⁻ T cells from Foxp3^EGFP mice were cultured with immobilized anti-CD3 (2.5 μg/ml), soluble anti-CD28 (1 μg/ml), and IL-2 (100 U/ml) with or without TGF-β (10 ng/ml). The percentage of CD4⁺ EGFP⁺ T cells after three days in culture with or without TGF-β is shown in the dot plots. (B). Cell surface expression of EGFP, CD25, CD44, CD103, CD62L, and GITR on in vitro-differentiated iTregs and nTregs obtained from the spleen and lymph nodes of Foxp3^EGFP mice. (C). Intracellular cytokine staining for IFN-γ and IL-17 secretion in iTregs after culture in anti-CD3, anti-CD28 and TGF-β for 3-4 days. IFN-γ secretion from naïve CD4⁺ T cells (Tconv) stimulated with anti-CD3 and anti-CD28 antibodies is shown as a control. (D). Purified B6 Thy1.2⁺ T cells (1 × 10⁵) were cultured with Balb/c CD11c⁺ dendritic cells (5 × 10⁴) in the presence of varying ratios of iTregs (black bars) or nTregs (white bars) for five days in triplicate wells. Control wells are depicted as hatched bars. Data are presented as mean cpm ± SEM and are representative of one of three experiments with similar results. Statistics: * p ≤ 0.05, ** p < 0.01.

Figure 2. In vitro generated iTregs fail to protect from lethal GVHD.

(A,B). Lethally irradiated (900 cGy) Balb/c mice were transplanted with TCD B6 BM alone (8 × 10⁶) (●, n=11) or together with B6 spleen cells adjusted to yield a T cell dose of 0.4-0.6 × 10⁶ αβ T cells. Animals transplanted with adjunctive spleen cells received either no additional cells (■, n=13) or in vitro-differentiated Tregs in a 1:1 ratio with naïve αβ T cells (□, n=11). Overall survival (A) and the percentage of initial body weight over time (B) are depicted. Data are cumulative results derived from three independent experiments. (C). Pathological damage in the colon, liver, and lung at day 24 post transplantation using a semi-quantitative scale as described in “Histologic analysis.” Mice (n= 8-10/group) were similarly transplanted as in A with the exception that that the αβ T cells dose was 0.5 × 10⁶. Data are presented as the mean ± SEM. (D). Lethally irradiated Balb/c mice were transplanted with TCD B6 BM alone (●, n=8) or together with B6 spleen cells adjusted to yield a T cell dose of 0.6 × 10⁶ αβ T cells. Animals transplanted with adjunctive spleen cells received either no additional cells (■, n=8) or in vitro-differentiated Tregs at a 1:1 (□, n= 8) or 5:1 (▼, n= 8) ratio with naïve αβ T cells. Data are cumulative results of two independent experiments. (E). Lethally irradiated Balb/c mice were transplanted with B6 BM alone (●, n=15), or B6 BM plus spleen cells adjusted to yield a dose of 0.6 × 10⁶ αβ T cells. Animals transplanted with adjunctive spleen cells received no additional cells (■, n= 16) or 0.6 × 10⁶ nTregs (□, n= 16). Overall survival is depicted. Data are cumulative results of four experiments. Statistics: * p ≤ 0.05, ** p < 0.01.

Figure 3. Foxp3-expressing iTregs have limited in vivo persistence in GVHD recipients.

(A-C) Lethally irradiated Balb/c mice were transplanted with TCD B6 BM, B6 spleen cells (adjusted to yield 0.6 × 10⁶ αβ T cells), and 0.6 × 10⁶ iTregs. Mice (n=8-12/group) were sacrificed at 7, 10, or 14 days post transplantation. (A). Representative dot plots gated on CD4⁺ cells depicting EGFP⁺ iTregs in the specified tissue sites. (B,C). Relative and absolute number of iTregs in the spleen, liver, lung and colon at the defined time points post transplantation. Data are presented as the mean ± SEM and are the cumulative results from three independent experiments. (D,E). Lethally irradiated Balb/c mice were transplanted with TCD B6 BM, B6 spleen cells (adjusted to yield 0.6 × 10⁶ αβ T cells), and 0.6 × 10⁶ nTregs. Mice (n = 4-8/group) were sacrificed at 24 or 60 days post transplantation. (D). Representative dot plots gated on CD4⁺ cells depicting EGFP⁺ nTregs in the specified tissue sites. (E). The relative percentage of nTregs in the spleen, liver, lung and pooled colon samples is shown. Data are presented as the mean ± SEM and are the cumulative results from two experiments.

Figure 4. Lack of protection is attributable to reversion of iTregs to a proinflammatory phenotype in vivo.

(A-E) Lethally irradiated Balb/c mice (n=12) were transplanted with 8 × 10⁶ B6.PL (Thy1.1) BM and spleen cells (adjusted to yield a dose of 0.6 × 10⁶ αβ T cells) with 0.6 × 10⁶ iTregs (Thy1.2). Spleen, liver, lung, and pooled colon samples were analyzed for Thy1.2⁺ cells to determine EGFP expression. (A) Representative dot plots gated on CD4⁺ T cells depicting the percentage of Thy1.2⁺ H-2K^b+ cells that expressed EGFP 10 days after transplantation. (B,C) The relative percentage of CD4⁺ Thy1.2⁺ EGFP⁻ cells is shown in B, and the absolute number of EGFP⁺ and EGFP⁻ cells in the specified tissue sites is depicted in C. Data are presented as the mean ± SEM and are the cumulative results of three independent experiments. (D) Representative dot plots and (E) percentages of CD4⁺Thy1.2⁺EGFP⁻ cells that secreted IFN-γ or IL-17A as determined by intracellular cytokine staining. (F) Lethally irradiated Balb/c mice were transplanted with B6.PL BM, B6.PL spleen cells (adjusted to yield a dose of 0.6 × 10⁶ αβ T cells) and 1.8 × 10⁴ B6 Thy1.2⁺ CD4⁺ EGFP⁻ cells. Representative dot plots gated on CD4⁺ T cells showing Thy1.2⁺ expressing cells in the spleen, liver, lung, and colon from animals 10 days after transplantation. Statistics: * p<0.05, **p<0.01.

Figure 5. All-trans retinoic acid does not prevent iTreg reversion.

(A) CD4⁺EGFP⁻ cells were converted to iTregs in the presence or absence of 10 nM ATRA. The percentage of CD4⁺ EGFP⁺ T cells after three days in culture with (RA-iTregs) or without (iTregs) ATRA is shown. (B) B6 Thy1.2⁺ T cells (1 × 10⁵) were cultured with Balb/c CD11c⁺ dendritic cells (5 × 10⁴) in the presence of varying ratios of iTregs that had been converted in the absence (black bars) or presence (white bars) of ATRA for five days. Controls are depicted as hatched bars. Data are presented as mean cpm ± SEM and are representative of one of three experiments with similar results (C-E) Lethally irradiated Balb/c mice (n=8) were transplanted with B6.PL BM, B6.PL spleen cells (adjusted to yield 0.6 × 10⁶ αβ T cells) and 0.6 × 10⁶ B6 iTregs or RA-iTregs. (C) Representative dot plots gated on CD4⁺ T cells showing the percentage of Thy1.2⁺ cells in spleen, liver, lung, and pooled colon 10 days after transplantation. (D) Absolute numbers of CD4⁺Thy1.2⁺ cells that were EGFP⁺ and EGFP⁻ cells in the specified tissue sites. (E) Percentage of CD4⁺ Thy1.2⁺ EGFP⁻ cells that secreted IFN-γ derived from mice transplanted with RA-iTregs. Data are cumulative results of two experiments. Statistics: * p<0.05, **p<0.01.

Figure 6. Blockade of IL-6R signaling does not prevent iTreg reversion but increases Thy1.2⁺ cell numbers.

(A-C) Lethally irradiated Balb/c mice were transplanted with B6.PL BM, B6.PL spleen cells (adjusted to yield 0.6 × 10⁶ αβ T cells) and 0.6 × 10⁶ B6 in vitro-differentiated iTregs. Mice were then administered anti-IL-6R (n=11) or isotype control (n=9) antibody on days 0 and 7 as described in Methods. Data are cumulative results of three experiments. (A) Representative dot plots gated on CD4⁺ T cells showing the percentage of Thy1.2⁺ cells that expressed EGFP in spleen, liver, lung, and colon samples 10 days after transplantation. (B) Absolute number of CD4⁺ Thy1.2⁺ EGFP⁺ and EGFP⁻ cells in the specified tissue sites from animals treated with isotype (black bars) or anti-IL-6R antibody (white bars). (C) Percentage of CD4⁺ Thy1.2⁺ EGFP⁻ cells that secreted IFN-γ in each tissue. (D) Representative histograms showing IL-6R expression on naïve CD4⁺ T cells (solid line) and in vitro-differentiated iTregs (dashed line) prior to transplantation. Isotype control is in gray. (E). IL-6R expression on iTregs obtained from the spleen of mice ten days post transplantation. Isotype control is in gray. Statistics: * p<0.05, **p<0.01.

Figure 7. Antibody blockade of the IL-6R augments generation of iTregs in vivo and attenuates GVHD.

Lethally irradiated Balb/c mice were transplanted with B6 Rag-1 BM (8 × 10⁶) alone (n= 18) (hatched bars) or together with sorted CD4⁺ EGFP⁻ T cells (0.2 × 10⁶) from Foxp3^EGFP mice. Cohorts of mice transplanted with CD4⁺ EGFP⁻ T cells were then administered rat IgG isotype control (n= 17) (black bars) or anti-IL-6R antibody (n= 20) (white bars) once weekly for four weeks as described in Methods. Mice in both groups were sacrificed 27-28 days post transplantation. Data are cumulative results from four independent experiments. (A). Pathological damage in the colon, liver and lung using a semiquantitative scoring system as detailed in Methods. (B). Representative dot plots showing percentage of in vivo-induced iTregs in the gated CD4⁺ T cell population from transplant recipients treated with either isotype control or anti-IL-6R antibody. (C). Percentage and (D) absolute number of iTregs in the spleen, liver, lung and colon of animals administered Rag-1 BM alone (hatched bars, n=14), or Rag-1 BM and CD4⁺ EGFP⁻ T cells and then treated with control (black bars, n=14) or anti-IL-6R antibody (white bars, n=16). Data are presented as the mean ± SEM and are the cumulative results from three experiments. Statistics: * p ≤ 0.05, ** p < 0.01.