The Early Rejection of Allogeneic BMC by Macrophages or NK Cells Is TRIF Signaling Dependent

Abstract 1883

The barrier for rejection of allogeneic bone marrow cells (BMC) has been attributed primarily to adaptive immunity, especially T cell immune responses. Significant progress has been made in developing immune-based nonmyeloablative conditioning strategies to achieve mixed chimerism in bone marrow transplantation (BMT) with targeting of T cells. The role of the innate immune system in BMC allorejection has not been adequately addressed. The fact that when T cells are targeted, alloengraftment requires additional conditioning from nonspecific reagents, such as irradiation and immunosuppressive drugs, suggesting the existence of another barrier. As humoral immunity is unlikely a barrier for BMC in unprimed naïve recipients, the innate immune system is most likely another barrier in BMT. The present study focused on a role for components of the innate immune response in allogeneic BMT. Using T cell-deficient (TCR-β/δ^−/−) mice, we found that rejection of transplanted allogeneic BMC occurred very early, well before the time required for T cell activation and was T cell independent, suggesting an effector role for innate immune cells in BMC rejection.

How the innate immune system recognizes or responds to allogeneic BMC remains unknown. Toll-like receptors (TLR) are a class of proteins that play a key role in the innate immune system. The signaling function of TLR depends on intracellular adaptors of MyD88 and/or TRIF. We have demonstrated that TRIF signaling is the innate immune signaling in BMC allorejection by showing superior engraftment in mice deficient in TRIF but not MyD88. To further determine the cell populations of innate immunity in allogeneic BMC rejection mediated through TRIF signaling, TRIF deficient (TRIF^Lps2/Lps2) mice were used as recipients for in vivo cytotoxicity assays after adoptive transfer of wildtype innate immune cells: macrophages or NK cells. Wildtype F4/80⁺ macrophages were sorted from B6 spleens and peritoneal cavities and NK1.1⁺ NK cells from B6 spleens. The doses of transferred NK1.1⁺ and F4/80⁺ cells were 370,000 and 140,000 per recipient, respectively. One day after transfer, 20 × 10⁶ CFSE-labeled BALB/c target (high intensity) and internal control B6 (low intensity) BMC were injected. TRIF^Lps2/Lps2 mice that did not receive transferred cells and wildtype B6 mice treated with saline served as controls.

As expected, donor cells were rapidly eliminated in control wildtype B6 mice and rejection was complete by day 3. The rejection of donor cells was significantly less in TRIF^Lps2/Lps2 mice without receiving adoptively transferred cells compared with wildtype B6 controls, from marginal significance (P = 0.04) at 3 hr to the highest significance (P = 0.0001) at day 3 after cell infusion. At day 3, the killing rates were 91.4 ± 1.4% in TRIF^Lps2/Lps2 mice without transferred cells and 97.6 ± 1.5% in wildtype B6 controls. The eliminating rates of donor cells were increased in TRIF^Lps2/Lps2 recipients that received either F4/80⁺ or NK1.1⁺ cells, and the kinetics of elimination of donor cells was shifted to resemble B6 controls with no significant difference between them at these 3 time points (P values: 0.11 to 0.87). The cytotoxicity of donor cells was significantly increased in TRIF^Lps2/Lps2 recipients adoptively transferred with F4/80⁺ or NK1.1⁺ cells when compared with TRIF^Lps2/Lps2 controls at all time points (P values: 0.038 to 0.002), except the one at 3hr when compared between TRIF^Lps2/Lps2 recipients received NK1.1⁺ cells and TRIF^Lps2/Lps2 controls (P = 0.058). At day 1, the killing percentages of CFSE labeled BALB/c cells were 71.0 ± 5.1%, 69.1 ± 3.9%, or 61.1 ± 5.8% in TRIF^Lps2/Lps2 recipients that received either F4/80⁺ cells, NK1.1⁺ cells, or none, respectively. Taken together, the restored cytotoxicity in TRIF-deficient recipients transferred with wildtype F4/80⁺ or NK1.1⁺ cells suggests that TRIF signaling is essential for macrophage- and NK cell-mediated early rejection of allogeneic BMC, and that both cell types function as non-redundant effector cells in BMC rejection.