Impairment of BM hematopoietic niches during GVHD. (A-B) Reconstitution of leukocyte subsets at 2 weeks after retransplantation. (A) To evaluate the competency of hematopoietic cells, lethally irradiated BDF₁ mice received TCD BM from either BMT or GVHD recipient mice of the [B6→BDF₁] model (^[B6→BDF1]TCD BM→BDF₁). (B) To evaluate the competency of BM stroma, TCD BM from normal C57BL/6.SJL donors was retransplanted into either BMT or GVHD recipient mice in the [B6→BDF₁] GVHD model on day 14, after TBI of 2 Gy on day 13 (TCD BM→[B6→BDF₁]). In the retransplantation, we performed nonmyeloablative irradiation (2 Gy) because recipient mice, especially GVHD mice, were susceptible to irradiation. We analyzed the donor BM–derived hematopoiesis based on the congenic marker CD45.1. Data are mean and SE (n = 3). *P < .0001. Experiments were performed at least twice. (C) In the long-term competitive repopulation analysis, we first transferred TCD BM cells from B6.SJL (CD45.1⁺) with or without T cells from B6 (CD45.2⁺) donor mice into CD45.2⁺ BDF₁ recipient mice, generating BMT or GVHD group, respectively. On day 14 after transplantation, we killed these recipient mice and purified CD45.1⁺ BM cells. We transferred 1:1 mixture of these CD45.1⁺ BM cells from BMT or GVHD group and untreated normal TCD BM cells from B6 donor mice (CD45.2⁺) into lethally irradiated B6 recipient mice (CD45.2⁺). Graphs are the summary of flow cytometric analysis of hematopoietic recovery. Data are mean and SE (n = 3). MOs indicates monocytes; PMNs, polymorphonuclear leukocytes; KSL, c-Kit⁺Sca-1⁺Lin⁻ cells; CD4⁺SP, CD4⁺ single-positive cells (CD4⁺CD8⁻); CD8⁺SP, CD8⁺ single-positive cells (CD4⁻CD8⁺); DP, double-positive cells (CD4⁺CD8⁺); DN, double-negative cells (CD4⁻CD8⁻); BMT/UT, the ratio of untreated B6-derived cells to BMT-derived cells (CD45.1/CD45.2); and GVHD/UT, the ratio of untreated B6-derived cells to GVHD-derived cells (CD45.1/CD45.2). Experiments were performed twice.