Abstract 688

Several lines of evidence show that donor derived mature CD4+CD25+Foxp3+ regulatory T cells (Tregs) suppress experimental GVHD. The mechanism of GVHD suppression by donor Tregs is, however, not well understood. Recent observations have brought in a renewed focus on the role of professional antigen presenting cells (APCs) in the induction and maintenance of GVHD by alloreactive T cell effectors (Teffs). But the role of APCs in modulating the responses of Tregs after allogeneic BMT is not known. We first tested the requirement of host APCs in Treg mediated regulation of GVHD. We utilized a clinically relevant CD8+ T cell dependent MHC matched but miHA disparate C3H.SW (H-2b) → wild type (wt) or Class II deficient Abb (II-/-) B6 (H-2b) model of GVHD because host APCs and target tissues from the Abb animals do not express class II and as such donor CD4+CD25+ Tregs will not directly interact with the host tissues while alloreactive CD8+ T cells could still respond to miHA allo-antigens presented by the intact class I on host APCs. The recipient Abb (II-/-) and wt B6 animals were lethally irradiated and transplanted with 2 × 105 CD8+ T cells along with or without CD4+CD25+ Tregs at 1:2 ratio from either syngeneic B6 or allogeneic C3H.SW animals. The wt recipients that received Tregs showed significantly better survival compared with the wt animals that did not receive any Tregs (P< 0.01) while the class II-/- animals showed similar GVHD mortality regardless of Treg infusion (P>0.8). To confirm whether the lack of Treg mediated protection was only due to the absence of interaction with host type APCs and also to exclude the possibility of development of Tregs from the infused BM we thymectomized wt B6 animals and then generated [B6 B6] controls and the [Abb B6] chimeras. These chimeric animals were used as recipients in a second BMT and transplanted with CD8+ Teffs and Tregs from allogeneic C3H.SW mice. Tregs reduced GVHD mortality in the [B6 B6] (P<0.01) but not in the [Abb B6] animals (P>0.7). We next evaluated whether host APC expression of allo-antigens alone was sufficient for Treg mediated GVHD protection in the absence of class II expression on target tissues by generating [B6 B6] and [B6 Abb] chimeras and found that Tregs demonstrated equivalent GVHD protection even when the class II allo-antigens were expressed only on the host APCs. Mechanistic studies demonstrated that Tregs significantly inhibited the expansion of CD8+ Teffs on days +10 and 17 after BMT in the spleens of the WT recipients (P<0.05) but not in the class II-/- animals. However, infused Tregs demonstrated reduced expansion in the class II-/- animals only early after BMT (on day +10) but was equivalent at later time-point (days 17 and 29) to the WT recipients. We further determined the mechanisms by which host APCs might contribute to Treg mediated protection. To this end we used IL-10-/-, indoleamine 2, 3 dioxygenase (IDO)-/- deficient animals and generated [IL-10-/- B6] and [IDO-/- B6] animals as recipients. Tregs mitigated GVHD mortality regardless of the ability of the host APCs to express IL-10 or IDO. We next determined whether Tregs suppressed Teffs in their activation phase at the level of their interaction with host APCs or in the effector phase. C3H.SW CD8+ T cells were primed (both in vivo and ex vivo with B6 allo-antigens) and then infused into the [β2mg-/- B6] animals such that pre-activated CD8 Teffs would still be able to initiate GVHD without the need for host APCs for their activation. Infusion of donor Tregs into [β2mg-/- B6] animals that were transplanted with the pre-activated Teffs mitigated GVHD severity demonstrating that Tregs, once activated by host APCs, were capable of suppressing Teff cells in their effector phase. Collectively our data show (a) host APCs are critical (b) expression of allo-antigens on host target tissues is not obligatory (c) host derived IL-10 and IDO are not critical for Treg mediated GVHD protection and (d) Tregs can mitigate GVHD by suppressing alloreactive Teffs in the effector phase even after they have been activated.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution