Indoleamine 2,3-dioxygenase is a critical regulator of acute graft-versus-host disease lethality

Abstract

Graft-versus-host disease (GVHD) is initiated after activation of donor T cells by host antigen-presenting cells (APCs). The immunosuppressive enzyme indoleamine 2,3-dioxygenase (IDO) is expressed by APCs and parenchymal cells and is further inducible by inflammation. We investigated whether lethal conditioning and GVHD induce IDO and if IDO prevents tissue injury by suppressing immune responses at the induction site. We determined that IDO is a critical regulator of GVHD, most strikingly in the colon, where epithelial cells dramatically up-regulated IDO expression during GVHD. IDO(-/-) mice died more quickly from GVHD, displaying increased colonic inflammation and T-cell infiltration. GVHD protection was not mediated by control of T-cell proliferation, apoptosis, or effector mechanisms in lymphoid organs, nor did it require donor T regulatory cells. Instead, T cells in IDO(-/-) colons underwent increased proliferation and decreased apoptosis compared with their wild-type counterparts. This evidence suggests that IDO can act at the site of expression to decrease T-cell proliferation and survival, diminishing colonic inflammation and reducing disease severity. These studies are the first to identify a function for IDO in GVHD lethality and indicate that modulation of the IDO pathway may be an effective strategy for treatment of this disease.

Figure 1

Donor T cells cause accelerated GVHD in IDO^−/− versus wt recipients. (A) B6 or IDO^−/− mice were lethally irradiated (9.0 Gy) and infused with 10⁷ MHC-mismatched Balb/c non–T cell–depleted BM with or without 3 × 10⁶ Balb/c CD4⁺ T cells. Survival plots of B6 versus IDO^−/− recipients are shown. ● indicates BM→B6; ■, CD4⁺ T cells→B6; ○, BM→IDO^−/−; □, CD4⁺ T cells→IDO^−/− (P < .03). Data are from one experiment; n = 8 mice/group. (B) B6 or IDO^−/− mice were lethally irradiated and given Balb/c TCD BM with or without 3 × 10⁶ donor CD4⁺ and CD8⁺ T cells. Survival plots of B6 vs IDO^−/− recipients are shown. ● indicates BM→B6; ■, CD4⁺ and CD8⁺ T cells→B6; ○, BM→IDO^−/−; □, CD4⁺ and CD8⁺ T cells→IDO^−/− (P < .04). Data are from one experiment; n = 8 mice/group. (C) Balb/c or IDO^−/− mice were lethally irradiated (7.0 Gy) and infused with 10⁷ B6 TCD BM with or without 2 × 10⁶ CD4⁺ and CD8⁺ T cells. Survival plots of Balb/c versus IDO^−/− recipients are shown. ● indicates BM→Balb/c; ■, CD4⁺ and CD8⁺ T cells→Balb/c; ○, BM→IDO^−/−; □, CD4⁺ and CD8⁺ T cells→IDO^−/−. Data are pooled from 2 experiments; n = 16 mice/group (P < .001).

Figure 2

Donor T cells up-regulate host IDO expression and worsen GVHD in IDO^−/− vs wt recipients. (A) Wt or IDO^−/− mice were lethally irradiated and infused with 3 × 10⁶ CD4⁺ and CD8⁺ T cells and killed on day 7. Cryosections of colon (left) and small intestine (right) were stained by immunohistochemistry for IDO. (i,v) Wild-type BM only. (ii,vi) Wild-type GVHD. (iii,vii) IDO^−/− BM only. (iv,viii) IDO^−/− GVHD. (B) Hematoxylin and eosin staining of wt and IDO^−/− colons (i-iv) and spleens (v-viii) 27 days after transplantation with 3 × 10⁶ CD4⁺ and CD8⁺ Balb/c T cells. (i,v) Wild-type BM only. (ii,vi) Wild-type GVHD. (iii,vii) IDO^−/− BM only. (iv,viii) IDO^−/− GVHD. (C) IHC staining of tissues in B for CD4. Cell numbers in colon are quantitated below (*P < .002). (D) IHC staining of tissues in B for CD8. Cell numbers in colon are shown below (*P < .02). Original magnification ×200, except spleens, which were ×100. Images are representative of 3 to 5 mice per group.

Figure 3

IDO-mediated Inhibition of GVHD lethality occurs independently of the presence of donor Tregs. (A) Balb/c TCD BM with or without 3 × 10⁶ Balb/c CD25-depleted CD4⁺ and CD8⁺ T cells were infused into lethally irradiated B6 or IDO^−/− recipients. Survival plots are shown. ● indicates BM→B6; ■, CD25-depleted T cells→B6; ○, BM→IDO^−/−; □, CD25-depleted T cells→IDO^−/−. T cell groups of B6 vs IDO^−/− (P < .005). Data are from one experiment; n = 8 per group. (B) Balb/c TCD BM with or without 2 × 10⁶ Balb/c CD4⁺CD25⁻ T cells were infused into lethally irradiated B6 or IDO^−/− recipients. Survival plots of B6 vs IDO^−/− are shown. ● indicates BM→B6; ■, CD4⁺CD25⁻ T cells→B6; ○, BM→IDO^−/−; □, CD4⁺CD25⁻ T cells→IDO^−/−. T cell groups (P < .02). Data are from one experiment; n = 8 mice/group. (C) Balb/c TCD BM with or without 1 × 10⁶ Balb/c CD4⁺ T cells or CD4⁺CD25⁻ T cells were infused into IDO^−/− recipients. Survival plots of CD25-depleted versus nondepleted are shown. ○ indicates BM→IDO^−/−; □, CD4⁺ T cells→IDO^−/−; △, CD4⁺CD25⁻ T cells→IDO^−/−. T cell groups (P < .02). Data are from one experiment; n = 8 mice/group.

Figure 4

Host IDO induction and IDO-mediated spleen and colon protection do not require donor Tregs. (A) IHC for IDO in colon 7 days after transplantation with Balb/c TCD BM with or without 2 × 10⁶ CD4⁺CD25⁻ T cells. (B) Hematoxylin and eosin staining of colon (i-iv) and spleen (v-viii) 7 days after transplantation with Balb/c TCD BM with or without 2 × 10⁶ CD4⁺CD25⁻ T cells. (i,v) Wild-type BM only. (ii,vi) Wild-type GVHD. (iii,vii) IDO^−/− BM only. (iv,viii) IDO^−/− GVHD. (C) Tissues from panel B stained for CD4 by IHC. Original magnification ×200 for colons, ×100 for spleens.

Figure 5

Proliferation, survival, and effector function of allogeneic T cells in secondary lymphoid organs are not increased in IDO^−/− mice. (A) 60 × 10⁶ CFSE-labeled Balb/c splenocytes were transferred into lethally irradiated B6 or IDO^−/− recipients. Mice were killed on day 4, and spleen was examined by flow cytometry for CFSE dilution and annexin V positivity. Spleen data shown are representative of 4 mice/group per day. Cells were gated on CD4⁺ or CD8⁺, H-2K^b-negative events. Numbers indicate percentage of annexin V–positive cells. (B) Lethally irradiated B6 or IDO^−/− mice were infused with Balb/c TCD BM and 3 × 10⁶ CD4⁺ and CD8⁺ T cells. Drinking water was supplemented with BrdU at days 15 to 20 after transplantation. Spleen and mesenteric LN were harvested and analyzed by flow cytometry for BrdU incorporation. Data are pooled from 2 identical experiments; n = 3 or 4 mice/group. No significant differences were observed. Cells were gated on CD4⁺ or CD8⁺, H-2K^b–negative events. (C) Lethally irradiated B6 or IDO^−/− mice were infused with Balb/c TCD BM and 4 × 10⁶ CD4⁺ and CD8⁺ T cells. Seven days after transplantation, spleens were harvested and analyzed by flow cytometry for FasL, granzyme B, and IFN-γ expression; n = 4 mice/group. Cells were gated on CD4⁺ or CD8⁺, H-2K^b-negative events (*P < .05).

Figure 6

Effector phenotype of donor T cells in target organs is not changed by IDO. B6 or IDO^−/− mice were lethally irradiated and infused with Balb/c TCD BM and 3 × 10⁶ CD4⁺ and CD8⁺ T cells. At 21 days after transplantation, T cells were isolated from colons and livers, stimulated with plate-bound anti-CD3 for 4 hours, and stained for IFN-γ, IL-4, granzyme B, and IL-10; n = 4 mice/group. No significant differences were observed.

Figure 7

T-cell proliferation and survival are increased in colons of IDO^−/− recipients. (A) B6 or IDO^−/− mice were lethally irradiated and infused with Balb/c TCD BM and 3 × 10⁶ CD4⁺ and CD8⁺ T cells. At 21 days after transplantation, T cells were isolated from colons and livers and counted; n = 4 mice/group (*P < .04). (B) T cells isolated from colon and liver were stained for annexin V; n = 4 mice/group (*P < .01). (C) Day 27 tissues from the transplant in Figure 2 were stained for CD4 in Cy5 (shown in red) and Ki-67 in FITC (shown in green), a nuclear marker of proliferation. Proliferating CD4⁺ cells are yellow. Original magnification ×400. Images are representative of 2 similar experiments, 2 to 5 fields/mouse, 3 to 5 mice/group.