In vitro RIBE lead to defects of human HSPCs. (A) Experimental design of in vitro RIBE model. CB-CD34⁺ cells were cocultured with irradiated or nonirradiated human bone marrow (BM) cells via transwell system. After 17 hours, CD34⁺ cells were collected for subsequent experiments. (B-C) Collected human CD34⁺ cells were transplanted into NOG mice. The frequency of human cell engraftment (B) and the lineage differentiation potential (C) were assessed in BM 20 weeks later (RIBE vs control [Ctrl]: 9.69% ± 1.903% vs 18.50% ± 3.301%, P = .023; n = 31 to 32 per group, 3 independent experiments). (D) Mean human cell engraftment in the BM of the secondary mice (RIBE vs Ctrl, 3.87 ± 1.494 vs 14.35 ± 4.430, P = .025; n = 15-17 per group). (E) Frequency of human cells in the recipient mice measured by LDA (n = 5-9 per group). (F) Number of hematopoietic colonies formed by human CD34⁺ cells in each group (n = 5 per group). (G) Proportion of BrdU⁺ cells of the human CD34⁺ cells from each group. BrdU was incubated with human CD34⁺ cells for 4 hours (n = 5 per group). (H) Frequency of early apoptotic cells (Annexin V⁺7AAD⁻) and late apoptotic cells (Annexin V⁺7AAD⁺) of human CD34⁺ cells from each group (n = 5 per group). (I) Fold change of SA-β-gal activity in human CD34⁺ cells was analyzed using C₁₂FDG as a substrate from each group (n = 5 per group). (J) Difference of SA-β-gal activity of human CD34⁺ cells from each group was confirmed by the SA-β-gal enzyme activity assay as shown in the microscopic images, and bars represent the proportion of blue stained cells. (K) Flow cytometric analysis of fold change in mitochondrial ROS levels by MitoSOX staining in human hematopoietic cells (n = 5 per group). (L-M) Fold change of mitochondrial membrane potential of human CD34⁺ cells determined by flow cytometry with TMRE (L) and DilC1(5) staining (M) (n = 5 per group). (N) Transmission electron microscopy (TEM) analysis of the mitochondria damage in human CD34⁺ cells of in vitro RIBE model. (O-Q) The energy phenotype of CD34⁺ cells in the in vitro RIBE model (O), the oxygen consumption rate (OCR) (P), and extracellular acidification rate (ECAR) (Q) were assayed by Seahorse assay (n = 5 per group). (R) Relative ATP levels in human CD34⁺ cells of in vitro RIBE model (n = 5 per group) (*P < .05; **P < .01; ***P < .001).