In Vivo Selection of Murine Long-Term Repopulating Cells Transduced with a Foamy Virus Vector That Expresses MGMTP140K.

Recombinant foamy virus vectors transduce noncycling and cycling cells, are stable episomally, and integrate into the host genome during cell division. Due to the cytoplasmic stability of this vector, a substantial lag period between transduction and cell division required for provirus integration is possible. Therefore, in transplantation studies that use minimally stimulated hematopoietic stem and progenitor cells (HSC), integration of the foamy virus vector in HSC may occur once HSC divide post-transplantation. We used a foamy virus vector, MD9-P140K-EGFP, that co-expresses a mutant form of O⁶-methylguanine DNA methyltransferase (MGMTP140K) and the enhanced green fluorescent protein (EGFP) to test the hypothesis that HSC could be transduced with a foamy virus vector and selected in vivo by alkylator-based chemotherapy. We also compared foamy virus transduction and selection to our previously optimized strategy using an oncoretrovirus vector to express MGMTP140K (SF1-P140K-IRES-EGFP). Lineage-depleted bone marrow (BM) from C57BL/6 mice was transduced for 10-16 hours with the foamy virus vector or transduced following a 2-day prestimulation with the oncoretrovirus vector. Data presented are from three primary transplant experiments analyzed over 6 months and one secondary transplantation experiment analyzed for 6 months. The bulk transduction efficiency using the foamy virus vector ranged from 12–25% and the CFU transduction efficiency was 55–57%. Transductions with the oncoretrovirus vectors resulted in similar bulk and CFU transduction efficiencies (55–60%). Similar numbers of progenitor colonies (oncoretrovirus vs. foamy virus) were observed. MGMT activity in pooled progenitor colonies was ~10-fold higher in EGFP⁺ versus EGFP⁻ colonies. Although similar numbers of CFU were transduced using the two vector systems, significantly different levels of in vivo selection were obtained in primary recipient mice. Consistent with previous studies, selection of oncoretroviral vector-transduced cells resulted in high and sustained levels of EGFP⁺ cells in the PB and BM in primary and secondary recipient mice (80–99% EGFP⁺ with 2–3 cycles of 6BG/BCNU). For primary transplants using cells transduced with the foamy virus vector, EGFP expression in the PB peaked at 3 months post-treatment (26.2+/−4.0%) which represented a 4–6 fold increase compared to vehicle-treated mice. However, by 6 months EGFP expression dropped by 3-fold (9+/−1%). Western analysis of MGMT protein levels found in the BM at 6 months post-transplantation also showed a 3-fold decline in expression. In secondary reconstitution experiments, however, flow cytometry and Western analysis of MGMT expression indicated that EGFP expression in foamy virus-transduced HSC no longer correlated with MGMT expression. In fact MGMT expression levels following drug treatment were similar to those found in secondary recipient mice transplanted with oncoretroviral vector-transduced cells, suggesting that stem cells expressing MGMT were selected over time. These data demonstrate that although foamy virus transduction is not as efficient as the more commonly studied oncoretrovirus transduction strategy, a simple overnight protocol can be used to transduce minimally stimulated HSC with a foamy virus vector. These cells can be selected in vivo, can reconstitute mice for up to one year, and can maintain high levels of MGMT expression.