471. Chimeric Lentiviral Vectors with Safety Insurance

Abstract

Lentiviral vectors as presently designed are considered to be safe. No evidence for the generation of replication competent viruses has emerged for these vectors when transduced in vitro into cultured cells. Their behavior in vivo remains to be ascertained. Their propensity to cause insertional mutagenesis, if any, is also unknown. One major safety concern is the recombination between the components of the vector system during vector production in the laboratory and with the resident virus in the clinic during therapy of individuals with apparent or non-apparent HIV infection. As an additional safety insurance, we have created chimeric or hybrid lentiviral vectors derived from HIV-1 and HIV-2. This was done for both transfer vectors and packaging constructs. Our premise was that HIV-1 and HIV-2 are dissimilar enough in sequence to curtail recombination, but similar enough to functionally complement. Our initial studies were done with HIV-2 vectors lacking the central polypurine track (PPT) and the woodchuck post-transcriptional enhancer element (WPRE), two regulatory elements reported to enhance transduction efficiency of HIV-1 vectors. This study includes HIV-2 vectors with these elements. A non-reciprocal interaction between conventional HIV-1 and HIV-2 vectors was observed in cross-packaging experiments. The titer of HIV-2 vectors was roughly the same whether packaged with the HIV-1 or HIV-2 packaging construct when titrated in cells in culture. In contrast, the titer of the HIV-1 vectors was twenty to thirty folds lower when packaged with HIV-2 than with the HIV-1 packaging construct. This could be due to the difference in the abundance of the transfer vector RNAs in the packaging cells, or due to the particular specificities of recognition and encapsidation. It was possible that the titer of HIV-1 vector was low when packaged with HIV-2 packaging construct because the Tat provided by the HIV-2 construct was poor in transactivating HIV-1 vector RNA. However, provision of additional HIV-1 or HIV-2 Tat did not change the outcome. This might imply that the reason for differential titer or non-reciprocacity lies in the interaction of the packaging signal and the gag determinants. With the chimeric packaging construct where the 'gag-pol' of the HIV-2 packaging constructs was replaced with the 'gag-pol' of HIV-1, the titer of the HIV-1 vector increased and that of the HIV-2 vector decreased relative to homologous packaging. This was not the case for the 'tat-rev' chimeric packaging constructs. However, the exchange of 'leader-gag' of HIV-1 vector with that of the HIV-2 vector in transfer vector chimeras by itself was detrimental for vector titer in both cases. These results support the possibility of creating chimeric vectors with added safety features and without sacrificing efficiency.