Abstract 687

Graft-versus-host disease (GVHD) remains a major cause of morbidity and mortality in allogeneic stem cell transplantation. We previously showed that recipient antigen-presenting cells (APCs) are required for CD8-dependent GVHD, while donor APCs promote GVHD in a MHC matched (C3H.SW (right arrow) B6; both H-2b) model (Shlomchik et al, Science 1999; Matte et al., Nat Med 2004). However, how donor APCs promote maximal GVHD was not addressed. The LTFNYRNL peptide from H60 is a dominant minor histocompatibility antigen (miHA) presented by H-2Kb. To study cross-presentation of H60, we crossed B6 mice congenic for H60 (CH60; hematopoietically restricted) or transgenic for H60 driven by an actin promoter (actH60; H60 is ubiquitously expressed) with B6 Kb-/- mice. These mice express H60 but cannot directly present it to donor CD8 cells as they do not express Kb. CH60C*Kb-/-and actH60*Kb-/- were irradiated and reconstituted with C3H.SW (H60-) bone marrow,106(6 superscript) CD8 T cells and 2*105( 5 superscript) CD4 (to promote the CD8 response to H60). Using H60-MHC tetramers, we detected H60-specific CD8 T cell expansion as early as day 8, with a peak at day14, demonstrating cross-priming by donor C3H.SW APCs. Intracellular IFN-γ staining and an in vivo CTL assay showed that these cross-primed CD8 T cells had effector functions. Surprisingly, accumulation of H60-tetramer+ cells was greater when it was exclusively cross-presented. SIINFEKL, a peptide derived from ovalbumin (OVA), is also presented by Kb. Therefore to confirm our findings we used B6 mice transgenic for ovalbumin crossed to Kb-/- mice (ova*Kb-/-) as recipients in the same model. SIINFEKL-tetramer+ T cells expansion was also observed in ova*Kb-/- recipients, demonstrating cross-priming. The source of miHA did not affect the cross-priming as similar SIINFEKL-reactive T cell expansion occurred in retransplanted (right arrow)ova*Kb-/-, ova*Kb-/-(right arrow) Kb-/- bone marrowγKb-/- chimeras. Cross-priming of SIINFEKL-reactive CD8 cells even occurred when BALB/c mice transgenic for OVA (BALB/c-ova; (H-2d)) were transplanted with B6 BM and a mix of B6 CD4 and CD8 cells. SIINFEKL-reactive cells produced IFN-γ and killed SIINFEKL-pulsed B6 cells in vivo. Because of the availability of knockout/transgenic mice backcrossed to B6, we used this system to explore mechanisms of cross-presentation. Donor CD11c+ dendritic cells (DCs) were required as cross-priming was abrogated when BALB/c-OVA mice were transplanted with BM from mice constitutively lacking CD11c+ DCs (Birnberg et al, Immunity 2008). CD4 help has been reported to be important for cross-priming. Surprisingly, however, cross-priming by donor APCs was unaffected when BALB/c-OVA mice were transplanted with B6 MHCII-/- BM but was greatly reduced in recipients of B6 CD40-/- BM. Thus, while CD40L activation of cross-priming DCs is important, CD4 cells which are likely the source of the CD40L need not actually make T cell receptor:MHC contacts with the cross-presenting DC. CD40L-conditioning of donor APCs is not required to cross-prime memory cells, as sort-purified memory CD8 cells from SIINFEKL-vaccinated mice expanded robustly in actOVA*Kb-/- but not Kb-/- mice. Cross-priming also occurred in recipients of IFNAR1-/- BM, indicating that Type I IFN activation of donor APCs is not required as has been reported in nontransplant settings. Taken together, our data demonstrate that cross-presentation by donor DCs occurs in MHC-matched and -mismatched transplants, and this cross-presentation likely explains the reduced GVHD we observed in recipients of MHCI- donor bone marrow. That T cells can be cross-primed to nonhematopoietic antigens provides a basis for persistent GVHD and for the generation of CD8 responses against antigens not initially targeted. We also found transplantation to be a permissive environment for cross-priming in that CD4 help could be delivered in trans, type I IFN APC activation was not required and memory cells could be activated without CD4 help. These data provide further rationale for targeting donor DCs and pathways required for cross-presentation to prevent and treat GVHD.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution