Figure 3.

GVH Recognition by Passenger Donor CD4 T Cells Accelerates Heart Allograft Rejection but Is Dependent upon the Host T-B Cell Axis

(A and B) Compared to B6 recipients of T cell-replete bm12.K^d.IE heart allografts, allograft vasculopathy was significantly less severe in allografts from T cell-depleted donors (A), with all hearts beating strongly at harvest (B); vasculopathy was restored by adoptive transfer of purified donor CD4 T cells at transplant (CD4 reconstitution; A).

(C and D) Transplantation of B6 recipients with heart allografts from bm12.K^d.IE donors challenged with a B6 skin graft (challenged donor) provoked stronger (C) anti-nuclear IgG autoantibody and (D) anti-K^d IgG alloantibody responses, with heart grafts rejected more rapidly (B).

(E) Allograft rejection is dependent upon host T and B cells (A), with the kinetics of rejection and the development of allograft vasculopathy in the different experimental groups mirroring recipient splenic GC activity.

(F) Host B cells are not required for optimal GVH activation of donor CD4 T cells, because upon transfer of CFSE-labeled bm12.K^d.IE CD4 T cells, the alloreactive fraction (boxed) divided similarly robustly in wild-type and B cell-depleted B6 recipients.

(G and H) Compared to wild-type B6 recipients, transplantation of bm12.K^d.IE allografts into T cell-deficient Tcrbd^−/− B6 recipients provoked similar anti-nuclear IgG autoantibody responses (G) but weak and transient anti-K^d IgG alloantibody with higher levels of alloantibody than observed in control naive serum achieved at week 1 (inset; H), without development of splenic GC activity (E).

(I) Whereas adoptively transferred bm12.K^d.IE CD4 T cells are readily detectable 7 days after transfer into Rag-2^−/− B6 mice, they were undetectable after transfer into wild-type B6 or Tcrbd^−/− mice.

^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001 (Mann-Whitney test in A, E, and H, inset [comparisons in A and E are to the bm12.K^d.IE group]; log rank [Mantel-Cox] test in B; two-way ANOVA in C, D, G, and H). Data are representative of one experiment (A–F and I; mean and SEM of n = 4 mice per group in A–F or n = 3 mice per group in I) or two independent experiments (G and H; mean and SEM of n = 4 mice per group).