Aggressive conditioning prior to hematopoietic cell transplants (HCT) reduces tumor burden, inhibits immune mediated resistance and provides space for the engraftment of hematopoietic stem cells. However, intense myeloablative conditioning also results in severe immune deficiency post-HCT. In addition to thymic reconstitution, T cells introduced at the time of transplant or early following HCT can undergo homeostatic expansion in the lymphopenic HCT recipient. Antigen (ag) -specific CD8 cells introduced early post-HCT can be beneficial, providing immediate immune function against viral infection or residual tumor. Memory CD8 cells (TM) are particularly attractive for restoring immune function because of their capacity to respond rapidly and generate effector activity in response to low ag. concentration. Recent reports, describing the phenotypic and functional conversion of transferred TN in lymphopenic recipients in the absence of foreign ag, led us to hypothesize that if ag-specific TN co-infused during HCT convert to memory-like cells in lymphopenic HCT recipients, these cells could subsequently provide important - and potentially equivalent - immune function to that by TM in reconstituting recipients. A syngeneic murine HCT model was utilized to investigate the homeostatic expansion, phenotype, and effector function of ag-specific naVve vs. memory CD8 populations following transplant. The TM population was generated in vitro from OT-I-Rag1−/ − spleen cells cultured with rmIL-2 and OVA peptide for 3 days followed by 2 days in rmIL-15. Resultant OT-I TM (CD44hi/Ly6C+/CD62Lhi/CD25lo) exhibited a central memory phenotype. The OT-I TN population (CD44lo/Ly6Clo/CD62Lhi/CD25lo) was positively selected from OT-I spleen and LNC using Miltenyi CD8 beads. Either 1.5x106 TM or TN, with 2x106 T cell-depleted B6 BM, were transplanted into 9.0-Gy conditioned syngeneic recipients. Homeostatic proliferation was assessed by OT.I cell numbers (CD8+/Vβ 5+/Vα 2+ ) in recipient spleens post-HCT. Both TN and TM expanded almost immediately post-transplant. However, by day 7 post-HCT, TN exhibited greater expansion resulting in 2x numbers of splenic OT.I cells compared to recipients of TM. TM underwent homeostatic expansion for the first 14 days following HCT, whereas TN continued to expand for another 1–2 wks. Notably, TN ultimately achieved higher overall numbers as a consequence of homeostatic expansion. Cell surface analysis of TM demonstrated these cells retained their memory phenotype >7 months post-HCT. In contrast, the initially naVve phenotype of transplanted TN began to convert towards a memory-like phenotype as early as 14 days post-HCT and by 42 days had fully converted. Paralleling the phenotypic conversion, effector activity of transplanted TN became memory-like by day 42 post-HCT. IFNγ production and lytic activity by transplanted TN was indistinguishable vs. TM. These findings demonstrate that the conditions for homeostatic expansion of both naVve and memory donor CD8 T cells are present in recipients immediately following ablative conditioning and transplant. Moreover, the higher homeostatic ‘set-point’ numbers of naVve vs. memory CD8 T cells suggest that conditions may favor the former early post-transplant. In total, the results suggest that homeostatic expansion and conversion of transplanted TN to memory-like cells demonstrate effector function indistinguishable from TM which may provide more effective immune function in reconstituting, ablatively conditioned HCT recipients than previously considered.

Author notes

Corresponding author