Severe combined immunodeficiency disease (SCID) in patients with adenosine deaminase (ADA) deficiency is thought to result from increased levels of purine metabolites. We attempted to immunosuppress a patient with ADA deficiency and SCID using a continuous infusion of deoxyadenosine to obtain engraftment of a T cell-depleted haplocompatible parental bone marrow graft. Before administering the drug in vivo, we investigated hematopoietic colony formation in two children with ADA deficiency (including the potential recipient), the obligate heterozygote donor (father), and normal controls using deoxyadenosine and erythro-9-(2-hydroxy-3-nanyl)adenosine (EHNA), and inhibitor of ADA. Deoxyadenosine alone in concentrations as high as 100 microM had no significant affect on erythroid (BFU-E) or myeloid (CFU-c) colony formation. However, in the presence of EHNA there was a significant reduction in BFU-E and CFU-c growth in all subjects and controls. Increasing doses of deoxyadenosine were given to one patient with ADA deficiency and SCID as a continuous 24-hr intravenous infusion. We found that there was a linear relationship between the dose administered and the plasma level; however, doses greater than 100 mg/day were required to increase erythrocyte dATP levels. We were able to raise intracellular dATP levels to more than three times baseline with doses of deoxyadenosine of 200 mg/day. However, there were no significant effects on the absolute lymphocyte counts or the lymphocyte responses to mitogen or alloantigen, and the haploidentical marrow failed to engraft. Our results suggest that the bone marrow of ADA-deficient patients is normal with respect to standard colony formation, that inhibitors of ADA do not adequately model the deficient state, and that the immunodeficiency in ADA deficiency is not proportionately related to either the deoxyadenosine or dATP levels, both of which were significantly elevated at the time of transplantation.