The hematopoietic stem cell is the ultimate precursor of all types of cells found in the peripheral blood. In addition to the proliferation and differentiation into these cell types, the hematopoietic stem cell retains the capacity to self renew without differentiation, allowing for bone marrow transplantation. These qualities make the hematopoietic stem cell an ideal target for gene therapy. The delivery of a gene to hematopoietic stem cells and the subsequent transplantation into a recipient would ensure a lifetime supply of corrected blood cells of all lineages. Although mouse stem cells can be transduced efficiently with mouse specific ecotropic retroviruses, a major obstacle to performing gene therapy for hematopoietic diseases has been the low transduction efficiency of human hematopoietic stem cells using amphotropic retroviruses. We hypothesized that this might be due to differences in the level of expression of the receptors for the retroviruses on stem cells. We compared the levels of ecotropic and amphotropic receptor (eco R; ampho R) mRNA in RNA extracted from mouse (Lineage negative [Lin-], c-kit HI) and human (Lin-, CD34+, CD38-) stem cells. Eco R mRNA was present at levels comparable to that of unfractionated marrow or cultured cells. In contrast, ampho R mRNA was present at very low levels in stem cells. Similarly, the Gibbon Ape Leukemia Virus (which can transduce human cells) receptor (GALV R) mRNA was present at extremely low levels in stem cells. Using counter flow centrifugal elutriation, which separates cells on the basis of size, we found that a subpopulation of stem cells representing 20% of the total number , expressed higher levels of Ampho R, while the remaining 80% were negative for ampho R mRNA expression. We were able to show in a mouse model that the stem cell expressing ampho R mRNA were transduced at similar efficiencies with amphotropic and ecotropic retroviruses, while only ecotropic retrovirus transduction was observed in ampho R negative stem cells. We conclude that the efficiency of transduction correlates with the level of expression of the mRNA encoding the receptor. We also predict that by using only those human stem cells which express higher levels of ampho or GALV R, gene transfer into human stem cells can be improved to clinically useful levels. One approach would be to induce ampho or GALV R mRNA expression in stem cells, a hypothesis we are testing in cultured cells. alternatively, populations of stem cells other than steady state marrow cells might have higher levels of ampho R mRNA. We compared ampho R mRNA expression in untreated bone marrow cells to ampho R mRNA expression in """"""""cytokine primed"""""""" bone marrow cells collected from mice 14 days after treatment with G-CSF and SCF. Following cytokine treatment greater than 80% of the """"""""cytokine primed"""""""" stem cells expressed ampho R mRNA, as opposed to 20% of the untreated marrow stem cells. We tested whether this would increase gene transfer efficiency in a Rhesus monkey model. Previous work had shown that gene transfer into primate stem cells from untreated bone marrow was inefficient (less than 1%), detectable only by CPR. When marrow from """"""""cytokine primed"""""""" donors was used, gene transfer efficiencies were improved to approximately 10%, a level which was easily detected by Southern Blot analysis. We intend to extend these studies into gene marking experiments in human patients.
