Moloney Murine Leukemia Virus (MoMuLV) causes T cell neoplasms in rodents but is not known to be a pathogen in primates. The core protein and enzyme genes of the MoMuLV genome together with an amphotropic envelope gene are utilized to engineer the cell lines that generate retroviral vectors for use in current human gene therapy applications. We developed a producer clone that generates a very high concentration of retroviral vector particles to optimize conditions for gene insertion into pluripotent hematopoietic stem cells. This producer cell line also generates a much lower concentration of replication-competent virus that arose through recombination. Stem cells from rhesus monkeys were purified by immunoselection with an anti-CD34 antibody, incubated in vitro for 80-86 h in the presence of retroviral vector particles with accompanying replication-competent virus and used to reconstitute recipients whose bone marrow had been ablated by total body irradiation. The retroviral vector genome was detected in circulating cells of five of eight transplant recipients of CD34+ cells and in the circulating cells of two recipients of infected, unfractionated bone marrow mononuclear cells. Three recipients of CD34+ cells had a productive infection with replication-competent virus. Six or seven mo after transplantation, each of these animals developed a rapidly progressive T cell neoplasm involving the thymus, lymph nodes, liver, spleen, and bone marrow. Lymphoma cells contained 10-50 copies of the replication-competent virus, but lacked the retroviral vector genome. We conclude that replication-competent viruses arising from producer cells making retroviral vectors can be pathogenic in primates, which underscores the importance of carefully screening retroviral producer clones used in human trials to exclude contamination with replication-competent virus.

Retroviral-mediated gene transfer into primitive hematopoietic cells has been difficult to achieve in largeanimal models. We have developed an amphotropic producer clone that generates' >1010 recombinant retroviral particles (colony-forming units) per ml of culture medium. Autologous rhesus monkey bone-marrow cells were cocultured with either high (2 X 1010 colony-forming units/ml) or low (5 X 106 colony-forming units/ml) titer producer clones for 4-6 days and reinfused into sublethally irradiated animals. The proviral genome was detected in blood and bone-marrow cells from all three animals reconstituted with cells cocultured with the high-titer producer cells. In contrast, three animals reconstituted with bone marrow cocultured with the low-titer producer clone exhibited no evidence of gene transfer.Retroviral-mediated gene transfer (1,2) has been successfully applied to the insertion and expression of various genes in murine stem cells, including the bacterial neomycin resistance (neo) (3-5), human dihydrofolate reductase (6), murine interleukin 3 (IL-3) (7), murine granulocyte-macrophage colony stimulating factor (GM-CSF) (8), human adenosine deaminase (ADA) (9-11) and human P-globin genes (12)(13)(14) (17,18).The efficiency of retroviral-mediated gene transfer into murine stem cells is influenced by the titer ofthe recombinant virus (9), preconditioning of the donor marrow with 5-fluorouracil (5-FU) (3,4,6) and the addition of the hematopoietic growth factors IL-3 (3, 4) and interleukin 6 (IL-6) (13). (22) and GP+E86 (23) cells were maintained in Dulbecco's modified Eagle's medium (DMEM; Biofluids) with 10% newborn calf serum (NCS; GIBCO). A pA317 cell line producing 5 x 106 infectious particles per ml of the N2 (24) virus was generated by standard techniques (25). This cell line was cotransfected with the plasmids pXM-gIL-3 (26), pXM-IL-6 (27), and pMohgr, which contain the gibbon IL-3, the human IL-6, and the hygromycin resistance genes, respectively. 3738The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Low-level retroviral transduction and engraftment of hematopoietic long-term repopulating cells in large animals and humans remain primary obstacles to the successful application of hematopoietic stem cell (HSC) gene transfer in humans. Recent studies have reported improved efficiency by including stromal cells (STR), or the fibronectin fragment CH-296 (FN), and various cytokines such as flt3 ligand (FLT) during ex vivo culture and transduction in nonhuman primates. In this work, we extend our studies using the rhesus competitive repopulation model to further explore optimal and clinically feasible peripheral blood (PB) progenitor cell transduction methods. First, we compared transduction in the presence of either preformed autologous STR or immobilized FN. Long-term clinically relevant gene marking levels in multiple hematopoietic lineages from both conditions were demonstrated in vivo by semiquantitative PCR, colony PCR, and genomic Southern blotting, suggesting that FN could replace STR in ex vivo transduction protocols. Second, we compared transduction on FN in the presence of IL-3, IL-6, stem cell factor (SCF), and FLT (our best cytokine combination in prior studies) with a combination of megakaryocyte growth and development factor (MGDF), SCF, and FLT. Gene marking levels were equivalent in these animals, with no significant effect on retroviral gene transfer efficiency assessed in vivo by the replacement of IL-3 and IL-6 with MGDF. Our results indicate that SCF/G-CSF-mobilized PB CD34+ cells are transduced with equivalent efficiency in the presence of either STR or FN, with stable long-term marking of multiple lineages at levels of 10-15% and transient marking as high as 54%. These results represent an advance in the field of HSC gene transfer using methods easily applied in the clinical setting.