We have examined the effect of the macrocyclic lactone PK-C activator, bryostatin 1, on the in vitro radioprotective capacity of the GM-CSF/IL-3 fusion protein, PIXY 321, toward normal committed myeloid progenitors (day 14 CFU-GM). Preincubation of CD 34+ cells for 24 h with 10 ng/ml PIXY 321 exerted significant radioprotective effects on these progenitors, (D = 1.403 vs 0.715 for controls), which were at least as great as those previously reported for higher concentrations (e.g., 50 ng/ml) or rGM-CSF. In contrast to the results of earlier studies involving rGM-CSF, preincubation of cells with both PIXY 321 and 10 nM bryostatin 1 did not lead to an increase in radioprotective effect when the total number of day 14 colonies was assessed. However, combinations of PIXY 321 and bryostatin 1 (or the tumour-promoting PK-C activator, PDBu) significantly increased the relative percentage and absolute number of surviving non-eosinophilic colonies (e.g., pure neutrophil, pure monocyte-macrophage, or mixed neutrophil-macrophage) at each radiation dose level. A similar pattern of response was noted in cells irradiated without a preconditioning interval, and in cells exposed to divided radiation doses. These results indicate that the GM-CSF/IL-3 fusion protein PIXY 321 exhibits significant in vitro radioprotective effects toward normal human bone marrow myeloid progenitors, and that co-administration of PK-C activators such as bryostatin 1 of PDBu selectively augments the radioprotective capacity of this hybrid cytokine toward non-eosinophilic elements.