Efficient Gene Transfer to Hematopoietic Repopulating Cells Using Concentrated RD114-Pseudotype Vectors Produced by Human Packaging Cells

The use of alternative viral envelope proteins to pseudotype oncoretroviral vectors has contributed substantially to recent improvements in retroviral gene transfer to hematopoietic stem cells. The envelope protein of endogenous feline leukemia virus (RD114) has emerged as a particularly attractive candidate in this context. Initial studies of RD114-pseudotype vectors produced by FLYRD18 packaging cells have shown promise, but it has also been demonstrated that FLYRD18 produces soluble factors that interfere with stem cell maintenance. We hypothesized that novel production methods for RD114-pseudotype vector preparations could overcome this limitation and lead to consistent high-level gene transfer in a canine model of stem cell gene therapy. We have developed a new retroviral packaging cell line based on human 293T cells (Phoenix-RD114). Virally conditioned medium derived from this cell line was volume reduced by centrifugation and resuspension in 1% of the original volume. Here we show that vectors produced by this method led to highly efficient transduction of canine hematopoietic repopulating cells. These results suggest that Phoenix-RD114-derived vectors may allow for more consistent transduction of hematopoietic repopulating cells than FLYRD18-derived vectors. An important obstacle to achieving clinically relevant levels of gene transfer to stem cells has been the low expression of the amphotropic viral receptor on the most immature hematopoietic cell populations [1]. This has prompted the development of alternative envelope proteins for pseudotyping. The receptor for the feline endogenous retrovirus RD114 is a neutral amino acid transporter that has been shown to be highly expressed on a number of human tissues, including CD34CD38 hematopoietic cells [2]. High expression of the receptor for RD114 in these very primitive hematopoietic cells provides a good rationale for the use of the RD114 pseudotype in the context of clinical stem cell gene transfer. Consistent with this rationale, RD114-pseudotype vectors have been shown to transduce efficiently canine hematopoietic repopulating cells [3] and human NOD/SCID repopulating cells [2,4–6]. Their stability is a particularly attractive feature of RD114-pseudotype vector particles, allowing for volume reduction by centrifugation [2,6]. Centrifugation-mediated concentration of vector particles offers two important advantages. First, higher vector titers achieved by centrifugation and resuspension in a smaller volume could potentially lead to higher titers, saturating available receptors on stem cells, thereby overcoming limitations imposed on transduction by low receptor expression [1,7,8]. Second, the concentration and resuspension significantly deplete potentially unwanted (e.g., FBS) as well as toxic byproducts present in the virally conditioned supernatant harvested from producer cell lines. In this context it is important to note that the original RD114-producer cell line, FLYRD18, was reported to have a differentiating activity on repopulating hematopoietic cells [4]. Production of transduction-inhibiting factors has been described for NIH3T3-derived producer clones [9], but not for producer cell lines based on human 293 cells. Recently, we were able to demonstrate an advantage of packaging cell lines derived from human 293T cells (Phoenix-GALV) compared to murine NIH3T3-derived PG13 packaging cells. This combination has led to gene transfer levels of up to 25% in a nonhuman primate model of stem cell gene therapy [10]. Based on the encouraging results obtained with Phoenix-GALV-producer cells, we established a 293T-based RD114-pseudotype packaging cell line (Phoenix-RD114)