Search

Compared to other integrating viral vectors, foamy virus (FV) vectors have several distinct advantages as a gene transfer tool, including: the ability to carry larger expression cassettes, the increased stability due to a DNA genome formed in developing vector particles, and the random integration pattern without preference for promoter regions or active genes as is the case with gamma-retroviral and lentiviral vectors. In addition, the FV envelope has tropism for most cell types and is not associated with disease in humans. Combined, these properties make FV vectors the ideal candidate for gene therapy application. Proof of principle was provided by Bauer et al. (Nat. Med. 2008) showing cure of canine Leukocyte Adhesion Deficiency (LAD) using autologous CD34+ cells transduced with FV vector carrying the CD18 gene, without short or long-term adverse effects. Here we demonstrate the optimization of large scale manufacturing of FV vector carrying the human CD18 gene for the treatment of human LAD. As compared to the methods used in the canine LAD study, a 60-fold increase in FV-CD18 titer was observed using a combination of polyethylenimine (PEI) transfection and use of a codon-optimized gag, from approximately 1.70E+04 to 1.00E+06 Infectious Units/mL (IU/mL) on Raw 264.7 cells. An evaluation of purification strategies showed efficient vector purification and concentration of FV vector through a combination of heparin-affinity chromatography on the AKTA ready, a disposable fluid path chromatography unit, tangential flow filtration and ultracentrifugation, with an overall 5000-fold concentration and 20-30% recovery. This approach was compatible with large scale manufacturing in compliance with Good Manufacturing Practices (GMP). Although 5% DMSO is needed for optimal recovery of infectious FV vector from -80°C storage, highly concentrated vector can be diluted to reduce the DMSO concentration and toxicity to CD34+ cells, while maintaining effective transduction rates of up to 50%. This process is currently being finalized for clinical manufacturing in support of a Phase I trial for the treatment of LAD.Note: MN and DL equally contributed to this study.

Compared to other classes of retroviral vectors used for HSPC gene therapy, foamy viral vectors (FVV) have distinct advantages, including their non-pathogenicity, a safer integration profile, and a high-efficiency transduction of quiescent cells. Proof-of principle of its therapeutic safety and efficacy in HSPCs was provided with the correction of canine Leukocyte Adhesion Deficiency (CLAD). Dogs with CLAD and individuals with LAD type 1 (LAD-1) suffer from recurrent and life-threatening infections caused by mutations in the β2 integrin CD18 subunit. In this study, G-CSF-mobilized CD34+ HSPCs were isolated from a 19 YO male with severe LAD-1 due to homozygous deletions in CD18. CD34+ cells were transduced for 16 hours at MOI of 0, 5, 10, and 20 with a FVV expressing a human codon optimized CD18 transgene. Flow cytometry of CD34+ cells cultured for 3 days after transduction demonstrated CD18+ cell surface expression in 40-45% of cells. Thirty NSG mice were transplanted with CD18-FVV-transduced human LAD-1 HSPCs (~1.2 × 105 cells/mouse), and human cell engraftment was measured in murine BM 5 months after transplantation using flow cytometry. Human CD45+ cells were detected in all mice (average ~1%). Mice transplanted with mock-transduced (MOI=0) LAD-1 CD34+ cells showed a 3.4-fold, 2-fold and 1.4-fold lower engraftment compared to mice injected with CD34+ cells transduced with FVV at MOI 5 (p 0.05), respectively, suggesting a selective homing/engraftment or survival/proliferative advantage of CD18+ cells. The inverse relationship between engraftment levels and MOI correlated with a gradual decrease in cell survival with increasing MOI, most likely due to toxicity from DMSO (required for FVV cryopreservation) during transduction; cell viabilities of 91%, 84%, 82% and 69% were obtained at MOI 0, 5, 10 and 20, respectively, indicating that further increase in MOI would lead to increasing toxicity. High-level, clinically relevant gene marking levels were obtained; the percentages of human cells expressing CD18 in the murine BM 5 months post-transplantation were 36.0 ± 3.9%, 33.9 ± 5.1%, and 44.5 ± 1.6% at MOI 5, 10 and 20, respectively. Quantitative PCR analysis of vector integrants within engrafted human cells indicated a single integration event occurred in the majority of long-term repopulating HSPCs at all MOI tested. Flow cytometry-based lineage analysis of bone marrow from mice transplanted with FVV-transduced LAD-1 CD34+ cells revealed human CD18+ cells in all lineages, with a predominance in the CD13+ myeloid compartment (88.3 ± 4.5%) compared to other lineages, including CD20+ lymphoid (9.1 ± 3.9%), CD235a+ erythroid (0.2 ± 0.1%) and CD41+ megakaryocytic (0.6 ± 0.2%) lineages. Interestingly, human myeloid engraftment was superior in recipient mice engrafted with human CD18+ cells (81.5 ± 4.3%) compared to animals transplanted with non-transduced (CD18-) LAD-1 cells (65.3 ± 11.3%). Integration site analysis of engrafted human cells is ongoing. Thus, FVV-mediated transduction of human LAD-1 CD34+ cells leads to clinically significant levels of CD18 expression, supporting the use of this CD18-expressing FVV in a human clinical trial.