Expression of CEACAM1 associates with LT-HSC activity in expanded UCB cells. (A) Schematic representation of experimental strategy for in vivo studies. Data presented as mean ± SD for 2 independent biological replicates. (B) Long-term engraftment of cells from indicated populations in NSG mice 3, 12, and 24 weeks after transplantation. Each dot represents 1 mouse from 2 independent experiments. Black bars indicate median engraftment values. **P = .0018, ***P = .0003 (Mann-Whitney U test, 2-tailed). Cells were transplanted at cell doses reflecting population frequency at sort to compare the contribution of each subset to engraftment. (C) Lineage potential of indicated populations assessed for each engrafted mouse of the experiment presented in panel B. The percentage of human CD45⁺ myeloid (CD33⁺), B lymphoid (CD19⁺), T lymphoid (CD3⁺), and primitive (CD34⁺) cells for each condition is shown. Each dot represents 1 mouse. ***P = .0007 (CD19⁺), ***P = .0009 (CD33⁺), ***P = .0001 (CD34⁺), **P = .0035 (CD3⁺) (Mann-Whitney U test, 2-tailed). (D) LT-HSC frequency of CEACAM1⁺ and CEACAM1^- populations sorted from CD34⁺CD45RA^-CD90⁺EPCR⁺ITGA3⁺ cells. Estimated frequencies (red lines) presented as 1 per number of sorted cells and 95% CIs (gray boxes) are shown. P = .05 (Mann-Whitney U test, 1-tailed). The objective of these experiments was to determine if the LT-HSC frequency of the CEACAM1⁺ITGA3⁺ population is higher than that of the CEACAM1^-ITGA3⁺ subset. The hypothesis assumed for the test is unidirectional, supporting the use of a 1-tailed test. (E) Human engraftment in secondary NSG mice 18 weeks after transplantation. Primary NSG mice transplanted with CEACAM1⁺ITGA3⁺ and CD90⁺EPCR⁺ parental cells were sacrificed 24 weeks after transplantation, and 80% of BM was collected and transplanted; n = 2 biological replicates.