Abstract

Retroviral-induced insertional mutagenesis has generated renewed concern due to the occurrence of three cases of T cell lymphoproliferative disease in the otherwise successful French SCID-X1 trial, as well as one instance of myeloid sarcoma in our rhesus macaque transplantation model. Recent genome-wide analysis of retroviral insertion sites for MLV and HIV-1 based vectors has shown that these retroviral vectors prefer to integrate into regions adjacent to (MLV) or within (HIV-1) genes, which likely results in an increased risk of insertional mutagenesis. Thus there is interest in the development of alternative vector systems, particularly for gene transfer to hematopoietic stem cells (HSCs). Intriguingly, the avian sarcoma leukosis virus (ASLV) has been reported to have an unbiased integration pattern in experiments done in a human cell line, but no studies have been done to determine the integration patterns of ASLV in primary cells that would be used in gene therapy, such as HSCs, and this vector has not previously been investigated as a gene transfer modality in primary hematopoietic cells. We tested the gene transfer efficiency and the integration pattern of an ASLV vector using our well-established non-human primate autologous transplantation model. ASLV vectors pseudotyped with an amphotropic MLV envelope and expressing marker genes were produced by transfection of vector DNA into the chicken cell line, DF-1. The resulting RCAS (Replication Competent ALV LTR with A Splice acceptor) vector is able to stably infect human and other mammalian cells but is not replication-competent in these cells. We obtained vector titers (measured on human HEK-293 cells) in the range of 2 to 9 x 106 cfu/ml for vectors carrying either GFP or a neomycin-resistance marker gene. We found that these ASLV vectors can efficiently and stably transduce rhesus macaque as well as human CD34+ hematopoietic progenitor and stem cells at levels up to 33% and 40%, respectively. We analyzed the proviral insertion sites by an optimized linear-amplification mediated PCR (LAM-PCR). 21.43% of the insertions we identified in CD34+ cells prior to transplantation are within gene-coding regions of the genome, which is comparable to random integration data and statistically significant different from MLV and HIV integration patterns we previously reported in primary hematopoietic cells. We are using these ASLV-transduced rhesus macaque CD34+ cells for in vivo autologous transplants to assess long-term stem cell gene transfer efficiency as well as the integration pattern in long-term repopulating cells. Two animals have been transplanted and will be analyzed in detail. We will compare the pattern of integration found in the transplanted cells with that seen prior to transplantation to ask whether the integration pattern is preserved after engraftment. These results, together with our previous findings showing distinctive integration pattern for MLV and HIV based vectors, further reinforce the idea that specific integration patterns of different retroviruses may arise because the different integration complexes bind to distinct cellular components.

Author notes

Corresponding author