Abstract

Retroviral gene transfer has successfully cured patients with X-linked severe combined immunodeficiency (X-SCID). Unfortunately, three of the treated patients developed leukemia at least in part due to insertional activation of a neighboring gene. Further advances of this exciting therapeutic modality will require safer viral vectors. Genotoxicity experiments in animals, although an important preclinical study, are not practical for screening large numbers of vectors. We have developed a rapid plasmid-based assay to compare downstream gene activation by retroviral vectors. We constructed a plasmid, pACT, that contains a minimal CMV promoter, an internal ribosome entry site (IRES) and a luciferase reported gene followed by a polyadenylation site. Four different proviral constructs containing identical internal PGK-EGFP expression cassettes were inserted upstream of the minimal promoter and nucleofected (Amaxa Biosystems) into K562 cells. The relative luciferase activation of each provirus was determined and compared to cells nucleofected with the basic pACT plasmid. The relative luciferase activation by a plasmid containing a Moloney murine leukemia virus (MLV) vector provirus with full length long terminal repeats (LTRs) was 144, 1230, and 4525 at 1, 2 and 4 days post nucleofection, respectively. Changing the orientation of the proviral DNA did not significantly change the results, consistent with an enhancer effect from the proviral LTRs. In comparison, the relative luciferase activation by a plasmid containing a self inactivating (SIN) MLV vector provirus with LTR deletions was only 59, 90 and 170, respectively, and was reduced by greater than 90% with the reverse orientation consistent with residual activation by read-through transcription. For plasmids containing the HIV SIN vector provirus, the relative luciferase activation was 48, 45 and 52, respectively at 1, 2 and 4 days post nucleofection, and decreased by >90% with the inverse orientation. The relative luciferase activation for plasmids containing the Foamy Virus (FV) vector provirus, also a “SIN” vector, was the least, only 3, 2 and 6, respectively, at 1, 2 and 4 days post nucleofection. Therefore, all three SIN vectors (MLV, HIV and FV) produced significantly less downstream gene activation than a conventional MLV vector, with the FV vectors being the lowest. Additional studies comparing various internal promoters and vectors will also be presented. These results demonstrate the usefulness of the pACT assay system to identify safer retroviral vectors.

Corresponding author