M. Oberbarnscheidt¹, J. Walch², Q. Li¹, J. Walters¹, A. Williams¹, G. Camirand¹, F. Lakkis^1
1Starzl Transplantation Institute, Dept. Of Surgery, University of Pittsburgh Medical Center, Pittsburgh/PA/UNITED STATES OF AMERICA, ²Starzl Transplantation Institute, Dept. Of Surgery, University of Pittsburgh Medical Center, Pittsburgh/UNITED STATES OF AMERICA

Body: Introduction: Unlike their naïve counterparts, memory T cells migrate to allografts and effect rejection without the need for priming in secondary lymphoid tissues. The mechanisms by which memory T cells migrate to transplanted organs, however, are not known. Here we tested whether CXCR3 and CCR5, known to direct the migration of effector T cells to sites of inflammation, are also required for the migration of memory T cells to a vascularized graft. Methods: Alloreactive memory T cells were generated in WT or CXCR3^-/- B6 mice by i.p. immunization with BALB/c splenocytes. CD4⁺ and CD8⁺ CD44^hi T cells were high-speed sorted >8 wks after immunization. To study migration, WT (Thy1.1) and CXCR3^-/- (Thy1.2) memory T cells were CFSE-labeled and co-transferred to B6 WT (CD45.1) or splenectomized B6 aly/aly (CD45.1) mice 2 days after transplanting BALB/c hearts (n = 6/grp). Graft infiltration with transferred cells was quantitated by flow cytometry 20 hrs and 3 days later. To study rejection, WT and CXCR3^-/- memory T cells were transferred separately into splenectomized aly/aly recipients of BALB/c hearts (n = 4/grp). These mice do not reject allografts unless effector or memory T cells are transferred, thus providing an in vivo model to study memory function. Results: WT CD4⁺ and CD8⁺ memory T cells expressed higher levels of CXCR3 and CCR5 than naïve T cells. CXCR3^-/- did not express CXCR3 but had similar CCR5 expression and produced similar IFNγ levels in response to allostimulation as WT memory T cells. These findings were confirmed in tetramer⁺ H60-specific CD8⁺ memory T cells. WT and CXCR3^-/- CD4⁺ and CD8⁺ memory T cells were recovered in equal numbers from cardiac allografts at 20 hrs and 3 days after transfer to either WT or splenectomized aly/aly recipients. On day 3, both WT and CXCR3^-/- cells had begun proliferating in the graft. Likewise, WT and CXCR3^-/- memory T cells precipitated equal rejection of heart allografts in splenectomized aly/aly recipients (MST=18 vs 20d). Treatment of mice adoptively transferred with CXCR3^-/- memory T cells with anti-CCR5 mAb (100 μg x 14 days) did not delay allograft rejection (MST=10d). Conclusion: The data demonstrate that CXCR3 and CCR5 are not critical for memory T cell migration to and rejection of vascularized cardiac allografts. This suggests that the requirements for memory T cell migration to sites of inflammation differ from those proposed for effector T cells, possibly due to inherent differences between the two cell types.

Disclosure: All authors have declared no conflicts of interest.