GATA-2^+/− marrow GMPs display impaired functionality in vitro and in vivo. Bone marrow–nucleated cells from each genotype were lineage depleted by using magnetic bead separation, and lineage-negative–enriched cells were stained with various cell surface antibodies to enable isolation of immunophenotypically defined progenitor subsets by cell sorting. LMPP, CMP, GMP, and MEP populations from each genotypic group were plated in colony-forming medium containing myeloid and erythroid growth factors. Three replicates were used per population per genotype in each experiment. The cumulative data for multiple experiments are shown: LMPP (A) (n = 2; P = not determined), CMP (B) (n = 5; G, P = .87; M, P = .61; GM, P = .93; E, P = .74; Meg, P = .45; E/Meg, P = .13; Mix, P = .49), GMP (C) (n = 5; G, P = .58; M, P = .6; GM, P = .03), MEP (D) (n = 5; E, P = .28; Meg, P = .3; E/Meg, P = .82), and Gr-1⁺Mac-1⁺ (E) (n = 3; G, P = .33; M, P = .28; GM, P = .37). In vivo functionality was assessed by using competitive transplantation of 10 000 GATA-2^+/− or GATA-2^+/+ GMP (B6SJL-CD45.1) and 150 C57BL/6-CD45.2 Lin-Sca-1⁺CD117⁺CD34⁻ competitor stem cells into C57BL/6-CD45.2–irradiated recipients. Eight days after transplantation, the peripheral blood of recipient mice were analyzed for the contribution of donor CD45.1 within the Gr-1⁺Mac-1⁺ compartment by flow cytometry (F) (n = 5; P = .037). ** indicates statistical significance. Error bars indicate SEM. Statistical analysis was performed by using the paired Student t test.