Fig. 1

Decreased frequencies of Tregs in BM and PB from aGVHD patients. a Representative flow cytometric analysis of Treg frequencies in (i) healthy donor BM, (ii) non-GVHD patient BM, and (iii) aGVHD patient BM. Within the CD4⁺ T cell gate, Tregs were identified as CD25^high Foxp3⁺. b Frequencies of CD4⁺ CD25^high Foxp3⁺ Tregs in BM from non-GVHD (n = 19) and aGVHD patients (n = 12) and healthy controls (n = 9). c The percentages of Tregs in the BM of aGVHD I° (n = 5) and aGVHD II–IV° patients (n = 6). d Frequencies of CD4⁺ CD25^high Foxp3⁺ Tregs in PB and BM from healthy controls (n = 5), non-GVHD patients (n = 5) and aGVHD patients (n = 5). e Reversed ratios of the frequencies of Tregs in BM and PB in acute GVHD patients (n = 4) compared with those in healthy donors (n = 4) and non-GVHD patients (n = 4). f The percentages of RTEs in each CD4⁺ subset in BM from aGVHD patients (n = 3) and healthy donors (n = 7) are shown. g The numbers of CXCR4⁺ cells are shown in PB from patients with (n = 8) or without aGVHD (n = 18). h, i Correlation and linear regression of the Treg proportions compared with the numbers of blood platelets and megakaryocytes in bone marrow smears from patients with (n = 7) or without (n = 6) aGVHD. Each dot represents an independent core, with a two-tailed Pearson’s correlation statistical analysis including six patients in each group. j The numbers of CD34⁺ cells are shown in BM from patients with or without aGVHD (n = 6). *P < 0.05; **P < 0.001