Allospecific Effector Memory T Cells Are Able to Mediate Second-Set Skin Graft Rejection but Unable to Induce Graft-Versus-Host Disease.

Abstract 232

Several different groups have independently demonstrated that non-allospecific effector memory T cells do not induce graft-versus-host disease (GVHD). Limited data are available regarding the ability of allospecific effector memory T cells to induce GVHD. We first studied this question in the C57BL/6 into BALB/c model. Similar to the data previously published by other groups, purified CD62L- effector memory T cells isolated from donors, who were primed with the host antigens 8 weeks earlier, had decreased ability to induce acute GVHD compared with unseparated and CD62L+ T cells. Similar results were observed when the parous female mice, who were sensitized to the host antigens during pregnancy, were used as memory T cell donors. In order to study this question more definitely and to understand the mechanisms underlying these surprising observations, we further studied the ability of allospecific effector memory T cells to induce GVHD using a novel GVHD model mediated by transgenic TEa T cells. All TEa T cells are CD4⁺ and recognize the same peptide in the context of I-A^b. This peptide corresponds to positions 52-68 from the alpha-chain of I-E class II molecules and is expressed in all antigen presenting cells from H-2^b/I-E⁺ strains such as CB6F1 mice. To generate memory T cells, naïve TEa cells were first parked in Rag1−/− mice and then immunized with irradiated CB6F1 spleen cells. More than eight weeks later, ∼98% of TEa cells obtained a memory phenotype (CD44high, CD45RB-.CD127+,CD11a bright, FasL bright, Ki67-, CD28-, KLRG-). Of them, about 93% were CD62L-CD44high effector memory T cells and 7% were CD62L+CD44high central memory T cells. These Rag1−/− mice that contained memory phenotype TEa cells rejected CB6F1 skin grafts much faster than naïve TEa mice did (median survival time: 6.5 vs. 13 days, P=0.01), suggesting that the memory phenotype T cells contained in these mice are functional. Moreover, CD62L-CD44high TEa cells purified from these mice mediated faster and stronger in vitro proliferative responses against alloantigens than naïve TEa cells did, further demonstrating that they are true functional effector memory T cells. We next tested the ability of these effector memory TEa cells to induce GVHD. Effector memory TEa cells were obtained after depletion of CD62L⁺ cells using magnetic beads and the purity was more than 99%. Transplantation of 1′10⁵ TEa naïve T cells together with 1′10⁷ T cell depleted bone marrow cells into lethally irradiated CB6F1 recipients induced lethal GVHD in all recipients and all animals in this group died within 35 days after transplantation. In contrast, none of the effector memory TEa cell recipients developed GVHD and all of them survived more than 100 days post transplantation (P<0.01, compared with naïve T cell control). To understand the mechanisms underlying these observations, we studied the kinetics of TEa proliferative responses upon challenge with alloantigens. The data indicated that effector memory TEa cells reached the peak responses faster than naïve TEa cells did. CFSE tracking experiment further confirmed this observation. Simultaneous staining with anti-Anexin V antibody and 7-AAD demonstrated that effector memory TEa cells undergone apoptosis and died faster than naïve T cells did. In conclusion, these data underscore the fundamental difference of alloresponses mediated by antigen-specific effector memory T cells in graft rejection and GVHD settings. The TEa transgenic T cell skin graft and GVHD models would allow further understanding of the unique alloresponses mediated by allospecific memory T lymphocytes in GVHD.