(A) Significant difference in amplitude of total erythroid output in liquid erythroid culture between normal (□, n = 7) and DBA (▩ , n = 10) cultures, at limiting (0.2 U/mL) and standard (2 U/mL) concentrations of Epo, and in response to dexamethasone (dex) at both Epo levels (normal versus DBA, P < .001 for each Epo/dex combination). Number of DAF⁺ cells × 10⁵ (mean ± SEM) generated after 7 days in phase 2 of liquid serum-free culture with or without 10^-7 M dexamethasone in both phases, expressed per 10⁵ total nonadherent cells plated in phase 2. (B) Similarity between the number of colonies in DBA and normal cultures, in contrast to difference in total erythroid output. There was also no significant difference between the number of colonies at limiting (0.2 U/mL) and standard (2 U/mL) concentrations of Epo (P > .05). Dexamethasone significantly increased colony number at both Epo concentrations (P < .05 at 0.2 U/mL Epo; P < .02 at 2 U/mL Epo). Number of colonies counted at day 7 in clonogenic phase 2 culture for normal (□,n = 8) and DBA (▩ ,n = 9), expressed as colonies per 10⁴ nonadherent cells plated in phase 2 (mean ± SEM). Serum-free culture (liquid in phase 1 and semisolid in phase 2), with IL-3/SCF with or without 10^-7 M dexamethasone in both phases. (C) Colony numbers at day 7 of phase 2 clonogenic assays of enriched erythroid cells, expressed as a percentage of enriched cells plated in phase 2 (mean ± SEM; normal, □,n = 5; DBA, ▩ ,n = 4). There was no significant difference in colony number between normal and DBA except at 0.2 U/mL without added dexamethasone. However, DBA colonies were very small under these conditions, and colonies of fewer than 20 cells were not counted. Serum-free culture (liquid in phase 1 and semisolid in phase 2), with IL-3/SCF with or without 10^-7 M dexamethasone in both phases.