254. New Chimeric Gene Therapy Vectors Based on Four Different Mammalian Bocaviruses

Parvoviruses have long been developed as safe, efficient and versatile DNA delivery vectors for gene therapy applications. In particular adeno-associated viral (AAV) vectors have emerged as lead candidates owing to their apathogenicity and amenability to genetic modifications. Yet, a drawback is their limited cargo capacity of 4.9 kb, which is insufficient to accommodate larger genes. Intriguingly, it was shown recently that oversized single-stranded AAV genomes can be packaged into capsids of human bocavirus 1 (HboV1, also a parvovirus), yielding HboV1/AAV chimeras that specifically and efficiently transduce human airway epithelia (HAE). Motivated by this pioneering work, we aimed to expand the repertoire of HboV/AAV chimeras by vectorizing four additional primate bocaviruses known to infect the gastrointestinal (GI) tract. We thus assembled and cloned the VP1/VP2 ORFs from three human variants (HboV2-4) and gorilla bocavirus into a helper plasmid derived from HboV1, carrying the genes required for HboV1 replication and packaging. To assess viral particle assembly, HEK293T cells were transfected with three plasmids: (i) one of our new HboV helpers; (ii) a self-complementary AAV-YFP vector; and (iii) pDG, a plasmid encoding all genes for AAV packaging and replication. In all cases, correct expression of VP1/VP2 proteins was confirmed by Western blot analyses of cell lysates. Also, following large-scale production and iodixanol gradient purification, qPCR analyses of the 40% phase showed the presence of DNase-resistant particles for all five bocaviral serotypes. Most importantly, titration of these particles revealed comparable quantities, demonstrating that expression of NS and NP1 proteins from HboV1 supports assembly of the four other bocaviruses. Analysis in primary HAE showed YFP transgene expression for all chimeric vectors except for HboV2/AAV, congruent with prior detection of the cognate wild-type viruses in nasopharyngeal aspirates. Further in line with epidemiological data, we noted a marked difference in infectivity, from 15% for HboV1, to below 1% for the others. In looming experiments, the new vectors will be studied in primary epithelial cells from the GI tract, which are the putative natural target cells for HboV2-4 and gorilla bocavirus. Interestingly, infectivity of all chimeras could be boosted by adding proteasome inhibitors, reminiscent of data with AAV. Hence, to enhance escape from the proteasome degradation pathway and improve transduction, we mutated surface tyrosines in the VP2 protein of HboV1. Functional assessment of the resulting mutants is currently ongoing. Collectively, the large capacity, unique cell specificities and ability to cross-package AAV DNA make this novel vector set highly attractive for human gene therapy applications. In the future, it should be moreover rewarding to attempt molecular evolution of bocavirus capsids, taking advantage of the profound experience with AAV vectors.