469. Improved Protocol and Use of Mitochondrial DNA as Reference for qPCR Quantification of Integrase-Defective Lentiviral Vectors (IDLVs)

Abstract

Integrase defective lentiviral vectors (IDLVs) are attractive tools for genetic manipulation and are increasingly being tested as recombinant viral vaccines against infectious pathogens and cancer. IDLVs remain mostly as extrachromosomal/episomal DNA circles in the infected cells, hence with reduced risk of genotoxicity mediated by insertional mutagenesis. We have demonstrated in our previous work the production and characterization of a tricistronic IDLV (packaged with the D64V mutation in the integrase and co-expressing GM-CSF, IFN-alpha and the CMV pp65 antigen) and IDLV-mediated monocyte transduction achieved under Good Manufacturing Process (GMP) compliant conditions (Sundarasetty et al., JTM 2015). One of the main criteria for batch release of the cell vaccine is to confirm and quantify IDLV copies in the thawed cell product. Currently used DNA extraction methods are not efficient in isolating small molecular weight episomal DNA from cells. In addition, genomic DNA loci used as reference controls are not informative regarding quantification of episomal DNAs. Mitochondrial DNA (mtDNA) is an episomal small molecular weight circular DNA (16.5 kb) that can serve as a reference for IDLV quantification. In order to maximize recovery of episomal DNAs, we explored the total DNA (tDNA) extraction described for mtDNA isolation (Badralmaa et al., J Vir. Meth. 2013), based on dehydration and precipitation of proteins and subsequent tDNA precipitation from the supernatants by isopropanol. As a reference for the qPCR quantification, we constructed and validated a plasmid containing a sequence homologous to a regulatory region of transfer vector (wPRE), a sequence for a genomic house-keeping gene (PTBP2) and a sequence for a mitochondrial house-keeping gene (Cytochrome B). The amplification of three target regions was validated by generating a standard curve with the reference plasmid ranging from 5 x10 to 5×105 copies (n=3). As a reference, we used an in-house generated 293T cell line (B5) containing three LV genomic copies. Total DNA extracted form B5 was serially diluted in non-transduced 293T DNA in order to result into 0.25, 0.5, 1, 2, 3 LV copies and the assay linearity was assessed in three independent runs. The reliable detection limit of the qPCR assay was 0.5 LV copies/cell and 0.9 copies/ng of genomic DNA. Having assessed the linearity, we assessed if this method could be used to quantify IDLV in transduced monocytes. The tricistronic IDLV was used to transduce monocytes of different donors in triplicates at increasing multiplicities of infection (MOI: 1, 2.5, 5 and 10). After thawing each cryopreserved batch, tDNA was extracted and analyzed. Our results showed a direct correlation between the IDLV copies per cell and ng of DNA used for the assay and MOI, whereas detection of the genomic and mtDNA references remained constant. Thus, this simple methodological adaptation provided an internal mtDNA control to better quantify episomal vector copies, which can be also used for qPCR quantification of other types of non-integrating vectors.