Optimization of retroviral-mediated gene transfer to human NOD/SCID mouse repopulating cord blood cells through a systematic analysis of protocol variables

Abstract

Retroviral transduction of human hematopoietic stem cells is still limited by lack of information about conditions that will maximize stem cell self-renewal divisions in vitro. To address this, we first compared the kinetics of entry into division of single human CD34 + CD38 − cord blood (CB) cells exposed in vitro to three different flt3-ligand (FL)-containing cytokine combinations. Of the three combinations tested, FL + hyperinterleukin 6 (HIL-6) yielded the least clones and these developed at a slow rate. With either FL + Steel factor (SF) + HIL-6 + thrombopoietin (TPO) or FL + SF + interleukin 3 (IL-3) + IL-6 + granulocyte-colony-stimulating factor (G-CSF), >90% of the cells that formed clones within 6 days undertook their first division within 4 days, although not until after 24 hours. These latter two, more stimulatory, cytokine combinations then were used to assess the effect of duration of cytokine exposure on the efficiency of transducing primitive CB cells with a gibbon ape leukemia virus-pseudotyped murine retroviral vector containing the enhanced green fluorescent protein (GFP) cDNA and the neomycin resistance gene. Fresh lin − CB cells exposed once to medium containing this virus plus cytokines on fibronectin-coated dishes yielded 23% GFP + CD34 + cells and 52–57% G418-resistant CFC when assessed after 2 days. Prestimulation of the target cells (before exposing them to virus) with either the four or five cytokine combination increased their susceptibility. In both cases, the effect of prestimulation assessed using the same infection protocol was maximal with 2 days of prestimulation and resulted in 47–54% GFP + CD34 + cells and 67–69% G418-resistant CFC. Repeated daily addition of new virus (up to three times), with assessment of the cells 2 days after the last addition of fresh virus, gave only a marginal improvement in the proportion of transduced CD34 + cells and CFC, but greatly increased the proportion of transduced LTC-IC (from 40% to >99%). Transplantation of lin − CB cells transduced using this latter 6-day protocol into NOD/SCID mice yielded readily detectable GFP + cells in 10 of 11 mice that were engrafted with human cells. The proportion of the regenerated human cells that were GFP + ranged from 0.2–72% in individual mice and included both human lymphoid and myeloid cells in all cases. High-level reconstitution with transduced human cells was confirmed by Southern blot analysis. These findings demonstrate that transplantable hematopoietic stem cells in human CB can be reproducibly transduced at high efficiency using a 6-day period of culture in a retrovirus-containing medium with either FL + SF + HIL-6 + TPO or FL + SF + IL-3 + IL-6 + G-CSF in which virus is added on the third, fourth, and fifth day.