Memory T cells migrate to and reject vascularized cardiac allografts independent of the chemokine receptor CXCR3

Abstract

Background: Memory T cells migrate to and reject transplanted organs without the need for priming in secondary lymphoid tissues, but the mechanisms by which they do so are not known. Here, we tested whether CXCR3, implicated in the homing of effector T cells to sites of infection, is critical for memory T-cell migration to vascularized allografts.

Methods: CD4 and CD8 memory T cells were sorted from alloimmunized CXCR3 and wildtype B6 mice and cotransferred to congenic B6 recipients of BALB/c heart allografts. Graft-infiltrating T cells were quantitated 20 and 72 hr later by flow cytometry. Migration and allograft survival were also studied in splenectomized alymphoplastic (aly/aly) recipients, which lack secondary lymphoid tissues.

Results: We found that polyclonal and antigen-specific memory T cells express high levels of CXCR3. No difference in migration of wildtype versus CXCR3 CD4 and CD8 memory T cells to allografts could be detected in wildtype or aly/aly hosts. In the latter, wildtype and CXCR3 memory T cells precipitated acute rejection at similar rates. Blocking CCR5, a chemokine receptor also upregulated on memory T cells, did not delay graft rejection mediated by CXCR3 memory T cells.

Conclusions: CXCR3 is not critical for the migration of memory T cells to vascularized organ allografts. Blocking CXCR3 or CXCR3 and CCR5 does not delay acute rejection mediated by memory T cells. These findings suggest that the mechanisms of memory T cell-homing to transplanted organs may be distinct from those required for their migration to sites of infection.

Figure 1. Phenotype and ex vivo function of wt and CXCR3^−/− memory T cells

Splenic T cells were harvested from wt (a) or CXCR3^−/− (b) B6 mice > 6 wks after immunization with BALB/c splenocytes. Chemokine receptor and integrin (VLA-4) expression was analyzed by flow cytometry after gating on the CD4⁺ and CD8⁺ naïve (CD44^lowCD62L^high) and polyclonal memory (CD44^high) T cell populations, or on the CD8⁺H60⁺ CD44^high antigen-specific T_EM (CD62L^low) and T_CM (CD62L^high) memory subsets. Plots shown are representative of 3 experiments. (c) IFNγ production by wt and CXCR3^−/− polyclonal CD8⁺ memory T cells 16 hours after ex vivo restimulation with allogeneic (allo) or syngeneic (syn) splenocytes (n = 3, mean ± SD). (d) T_CM and T_EM phenotype of wt and CXCR3^−/− CD8⁺H60⁺ antigen-specific memory T cell populations. Plots shown are representative of 3 experiments. (e) Memory CD4 and CD8 T cells were sorted from B6 mice > 6 wks after immunization with BALB.B splenocytes and restimulated in vitro with either syngeneic (B6) or allogeneic (BALB.B) splenocytes for 24 hours. Chemokine receptor expression was analyzed by flow cytometry after gating on polyclonal CD4⁺ and antigen-specific CD8⁺H60⁺ CD44^high memory T cells. Stimulators were gated out by use of congeneic markers or CellTracker Violet labeling.

Figure 2. Migration of wt and CXCR3^−/− memory T cells to cardiac allografts

Sorted wt (CD45.2, Thy1.1) and CXCR3^−/− (CD45.2, Thy1.2) polyclonal CD4 and CD8 memory T cells were co-transferred to congenic (CD45.1, Thy1.2) recipients 2 days after cardiac allograft transplantation. Allografts were harvested at 20 and 72 hours after cell transfer. Co-transferred wt and CXCR3^−/− memory T cells that infiltrated the graft were identified and quantitated by flow cytometry according to the gating strategy shown (a). (b) Quantitation of transferred wt and CXCR3^−/− memory T cells recovered from allografts removed from wt recipients (2 independent experiments; n = 3 mice/experiment; mean ± SD). Proliferation of recovered CD8 memory T cell populations determined by CFSE dilution is shown in the histograms. (c) Immunofluorescence staining of cardiac allograft tissue demonstrating the presence of CXCR3^−/− (CD45.2) memory T cells within the myocardium 6 days after transfer to congenic wt (CD45.1) recipients. Transferred T cells appear green (CD45.2⁺), endothelial cells red (CD31⁺), and nuclei blue (DAPI). White bar = 50 µm (magnification = 60x). (d) Quantitation of transferred naïve wt T cells as well as wt and CXCR3^−/− memory T cells as described under (b) except that recipients were aly/aly-spleen mice (2 independent experiments; n = 3 mice/experiment; mean ± SD). n.s. = not significant

Figure 3. Cardiac allograft rejection by memory T cells is independent of CXCR3 and CCR5

CD4 and CD8 memory T cells were transferred to aly/aly-spleen recipients 2 days after cardiac transplantation and allograft survival was determined by palpation (a). Control mice did not receive any exogenous T cells or received naïve T cells. Rat anti-mouse CCR5 antibody was administered daily for 15 days starting on the day prior to transplantation. No significant difference in allograft survival was observed among the groups that received memory T cells. (b) Representative cardiac allograft histology (H&E staining) from the indicated groups showing extensive cellular infiltrate and myocyte destruction with active arteritis (magnification = 30x).