Molecular Therapy Volume 9, Supplement 1, May 2004 Copyright © The American Society of Gene Therapy S29 RNA VIRUS VECTORS I 70. Stable Packaging Cell Lines Development for the Production of Lentiviral Vectors Sophie Broussau, 1 Nadine Jabbour, 1 Guillaume Lachapelle, 1 Yan Xu, 1 Alaka Mullick, 1 Renald Gilbert, 1 Bernard Massie. 1,2,3 1 Genomics & Gene Therapy Vectors, Biotechnology Research Institute, NRC, Montréal, QC, Canada; 2 INRS-IAF, Université du Québec, Laval, QC, Canada; 3 Département de Microbiologie et Immunologie, Faculté de Médecine, Université de Montréal, Montréal, QC, Canada. Our ultimate goal is to develop an optimal lentivector (LV) backbone in a self-inactivated (SIN) configuration, in which the most efficient promoter controls transgene expression, flanked by adequate insulators. Although SIN vectors are ideal to minimize silencing and to insure proper control of regulated promoters, in this configuration, the production and recovery of LV are inefficient. We have worked towards the development of a new packaging cell line for the efficient production of regulated conditional SIN LV (cSIN). Our design involved the use of two gene-switches to independently regulate the packaging elements and the transgene. As the initial step we have established a stable 293SF (adapted to serum-free culture) cell line expressing both the cumate repressor (CymR) and the reverse tetracycline transctivator (rtTA). Our best 293SF-CymR/ rtTA clones were shown to maintain optimal expression level of both CymR and rtTA without selective pressure for at least ten weeks in culture. The generation of LV packaging cell lines from 293SF-CymR/rtTA then proceeded in two steps. First, the selection of clones stably expressing the LV Gag-Pol and Rev. We demonstrated that for all the clones selected, the production of LV was improved by the addition of Rev, suggesting that the expression of Rev was rate limiting. We hypothesized that high-level stable expression of either Rev or the Gag-Pol was not well tolerated by the cells. At the second step, we therefore co-transfected the best clones with the VSV-G envelope along with a modified Rev vector (regulated with CymR) to finally yield the 293SF-PacLV packaging cell lines. The selection of the best packaging cell lines is currently underway and preliminary results indicate that the best clones produce LV at higher level that transient transfection. The approach of producing the packaging cells line by selecting first the best clone expressing Gag-Pol and Rev will open the possibility to generate a collection of packaging cell lines using various envelope genes to pseudotype the LV, which could ultimately be useful to produce LV with improved gene delivery efficiency in stem cells. 71. Chimeric Lentiviral Vectors Derived from HIV-1 and HIV-2 Suresh K. Arya, 1 Jenice D’Costa, 2 Geetanjali Sachdeva, 1 Jean Cho, 1 Kritika Kachapati. 1 1 NIH; 2 Johns Hopkins. While a source of great misery, lentiviruses can be modified and channeled toward the useful purpose of developing vectors for gene transfer in gene therapy, functional genomics, and animal transgenesis. Being retroviruses and by virtue of their integration into the target cell genome, they provide for long term transgene expression. The added advantage of lentiviral vectors is their ability to affect gene transfer in non-dividing as well as dividing cells, without any anticipated immune reaction. To make a compelling case for their use, specially in gene therapy, three interrelated issues need to be addressed - targeting, regulation and safety. To enhance their safety or to minimize the chance of recombination, we are creating chimeric vectors derived from HIV-1 and HIV-2. This is being done both for transfer vectors and packaging constructs. It is based on the premise that there is sufficient sequence diversity between HIV-1 and HIV- 2 to curtail recombination. Vectors were produced by transient transfection of human embryo kidney 293T cells, using matching sets of transfer vector and packaging construct DNAs, and pseudotyped with VSV-G. For ease of detection, the transfer vectors carried indicator GFP gene. The resulting supernatants were titrated on 293T cells and their content of the core antigen (p24 or p27) determined by Eliza assays. Preliminary results show that the titer for the homologously packaged HIV-1 vector [HIV-1 x HIV-1p] was about 2.7 x 10e3 TU/ng and that of the homologously packaged HIV-2 vector [HIV-2 x HIV-2p] was about 0.2 x 10e3 TU/ng. The cross-packaged transfer vectors, HIV-1 x HIV-2p and HIV-2 x HIV- 1p, had titer of about 0.12 x 10e3 TU/ug and 0.16 x l0e3 TU/ug, respectively. For the chimeric transfer vectors, the titer of the HIV- 1 vector with the replacement of the HIV-1 leader-gag segment with that of the HIV-2 leader-gag segment was roughly the same as for the wild type vector, both in homologous and hetrologous packaging reaction. Initial experiments with the reverse transfer vector chimera gave titers lower than that for the corresponding wild type vector. Two of the packaging construct chimeras also have been tested. These chimeras consist of the exchanges of 5' halves (‘gag-pol’) and 3' halves (‘tat-rev-nef’) genes. When tested with HIV-2 transfer vectors, 3' half chimeras appeared to be more like the wild type constructs than the 5' half chimeras. This may reflect specificity of interaction between the leader sequence and the gag protein. Additional chimeric vectors are being created to test this hypothesis. The results show that the creation of chimeric vectors between HIV-1 and HIV-2 is feasible and suggest that an optimal combination of chimeric transfer vectors and packaging constructs with enhanced safety and minimal loss of efficiency can be designed. 72. Position and Copy Number Dependence of the Central Polypurine Tract (cPPT) Function in Simple and Complex Lentiviral Vectors Alex H. Chang, 1 Leszek Lisowski, 1 Matthias Stephan, 1 Isabelle Rivière, 1 Michel Sadelain. 1 1 Department of Medicine, Gene Transfer and Somatice Cell Engineering Facility, Memorial Sloan-Kettering Cancer Center, New York, NY. Recombinant lentiviral vectors derived from human immunodeficiency virus-1 are promising vectors for gene therapy applications. This is due to their ability to transduce nondividing cells, their genomic stability, their relatively large genomic capacity, as well as their ability to stably maintain long term transgene expression. Recent studies have uncovered that the central polypurine tract (cPPT) plays an important role in increasing lentivirus-mediated gene transfer efficiency. The cPPT participates in lentiviral reverse transcription by initiating the synthesis of a downstream plus strand cDNA. Together with the upstream plus strand initiated at the 3' polypurine tract, it forms a central DNA flap that bears in its center a short plus strand overlap. A possible role in nuclear import enhancement of the viral preintegration complex has been suggested (Zennou, Cell, 2000), but the mechanism of cPPT function is still not fully elucidated. It has also been reported that the incorporation of cPPT in lentiviral vectors can increase particle infectivity (Follenzi, Nature Genetics, 2000). However, a systematic evaluation of the cPPT in complex lentiviral vectors has not been reported to date. To further investigate the effect of the cPPT on transduction efficiency, we introduced one or two copies of cPPT at different locations in two lentiviral vectors. The first vector, CPG, has a small genome (3409 nucleotides) encoding the PGK promoter and GFP gene. The other, CMPD, is a complex vector with a large viral genome (9268 nucleotides) that encodes the human β-globin gene, together with a 3.2 kb locus control region, the PGK promoter and a drug resistance gene. The cPPT element was cloned at different positions near the 5'-, middle, or 3'- of the vectors. In order to rigorously compare their infectivity, concentrated vector stocks were quantified using P24 ELISA, RT assay and/or RNA dotblot. Hela