The desire to improve engraftment following transplantation of limited numbers of hematopoietic stem cells (HSC) has spurred the investigation of ex vivo stem cell expansion techniques. While surrogate outcomes, such as an increase in SCID-repopulating cells, suggest successful stem cell expansion in some studies, it is not clear that such assays predict outcomes using a more clinically relevant approach (e.g., myeloablation). We have addressed this by testing three cytokine combinations for their ability to increase the radioprotective and long-term marrow reconstitution capacity of hematopoietic cells cultured ex vivo. Low numbers of light-density (LD) mouse bone marrow (BM) cells or their expanded product were injected into lethally irradiated (9 Gy) congenic recipients. Survival rates and percent donor engraftment were compared at 2, 5, and 7 months post-transplant. The three cytokine combinations used were: (i) kit-ligand (L), thrombopoietin (Tpo), Flt-3 L; (ii) cytokines in (i) plus interleukin-11 (IL-11); (iii) cytokines in (ii) plus IL-3. At 7 months post-transplant, LD cell doses of 10(4), 2-2.5 x 10(4), and 0.5-1.0 x 10(5) gave predictable survivals of 20-30%, 40-70%, and 100%, respectively. Mean percent donor engraftments were 54.9% (SEM 36%), 55.7% (SEM 36%), and 76.3% (SEM 21%), respectively. When cells expanded for 3 or 5-7 days with the various cytokine combinations were transplanted into different groups of mice, survival rates and percent donor engraftment were almost uniformly poorer than results obtained with unmanipulated cells, and cells expanded for 5-7 days led to poorer outcomes than cells expanded for 3 days. Overall, ex vivo expansion of LD BM cells with the cytokine combinations chosen failed to improve transplant outcomes in this model.