The inherited deficiency in adenosine deaminase (ADA), which results in severe combined immunodeficiency, is generally regarded as an optimal model for the development of human somatic gene therapy. The ideal target for the correction of ADA deficiency and other lympho-hematopoietic disorders would be the hematopoietic stem cell. We have used a combination of recombinant human interleukins-3 and -6 to stimulate the proliferation of primitive human hematopoietic progenitor cells during a period of co-cultivation with irradiated cells producing high titers of an ADA-transducing retroviral vector packaged in amphotropic particles. In a series of nine experiments, an average of 83% of the clonogenic progenitors (CFU-E and CFU-GM) were found to have acquired the transferred sequence as determined by polymerase chain reaction analysis. In addition, in two experiments, 24-44% of the clonogenic progenitors derived from long-term myeloid cultures 9 weeks post-transduction were found to contain vector sequence. The latter cells are derived from so-called "long-term culture-initiating cells" (LTC-IC), which are primitive cells probably related to hematopoietic stem cells. Moreover, the transduced ADA enzyme was found to be expressed in both normal and ADA-deficient erythroid colonies, and in the nonadherent cells of long-term bone marrow culture for at least 2 weeks at levels that approximate the endogenous ADA levels of normal erythroid cells. These results indicate that the ADA coding sequence can efficiently be introduced by retroviral gene transfer into both committed and primitive human hematopoietic progenitor cells, and that this will result in adequate expression of the transduced enzyme in the progeny of committed hematopoietic progenitors.