Effects of low oxygen tension and antioxidants on human erythropoiesis in vitro

K. Ono, B. P. Alter

Research output: Contribution to journalArticlepeer-review

11 Scopus citations


Cultured cells grow better in low oxygen than in room air, and the addition of a sulfhydryl-reducing agent is usually routine. It has been thought that such reagents protect cells from free-radical toxicity in atmospheric oxygen, which is higher than in physiological oxygen tension. Previous data demonstrating the efficacy of the usual antioxidant, 2-mercaptoethanol (2ME), were obtained in murine studies. We have compared the effect of 2ME and other antioxidants on human erythropoiesis using neonatal (cord) and adult blood. Without antioxidants, erythroid colony numbers were higher in 5% oxygen than in air, while colony formation could be improved even in 5% oxygen by the addition of antioxidants. In dose-response curves of each drug, the maximum increase in colony number was achieved with 0.1 pM 2ME, 10 μM reduced glutathione (GSH), 1 U/mL catalase (CAT), and 1 U/mL superoxide dismutase (SOD). Erythroid colony growth in air and in 5% oxygen was identical with 10 μM GSH and higher than in any other conditions, including with combinations of GSH and any other antioxidant. For human erythroid progenitor cell cultures, 10 μM GSH appears to be more appropriate than the usual 100 μM 2ME. Antioxidants are necessary, however, even in low oxygen tension. The mechanism of the GSH action is incompletely understood.

Original languageEnglish (US)
Pages (from-to)1372-1377
Number of pages6
JournalExperimental Hematology
Issue number13
StatePublished - 1995
Externally publishedYes


  • antioxidants
  • glutathione
  • human erythroid progenitor cells
  • low oxygen tension

ASJC Scopus subject areas

  • Cancer Research
  • Cell Biology
  • Genetics
  • Hematology
  • Oncology
  • Transplantation


Dive into the research topics of 'Effects of low oxygen tension and antioxidants on human erythropoiesis in vitro'. Together they form a unique fingerprint.

Cite this