Exploiting the adenoassociated virus Rep protein to mediate site-specific integration into the human genome and optimisation of Ad/AAV vector design

Avolio, Fabio (2008). Exploiting the adenoassociated virus Rep protein to mediate site-specific integration into the human genome and optimisation of Ad/AAV vector design. PhD thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000fa68

Abstract

Gene therapy is an approach to treating diseases in which an exogenous gene is introduced to correct for a defective or missing protein or to affect a biochemical pathway. Few successes have been reported in humans (0), as several technical issues limit its broader application. One question is how to deliver DNA to the appropriate cells. Nature provides one solution in the form of viruses, which are in essence protected gene delivery packages with native ability to introduce their genomes into cells. Once the desired gene is delivered to target cells, another issue that arises is the fate of the DNA. Some strategies rely on long-term expression from extra-chromosomal DNA, but there are cases, such as dividing cells, where it would be highly beneficial to permanently insert the gene into chromosomes. Certain viral genomes can be integrated into host DNA by non-homologous recombination or, in the case of retroviruses, by virally encoded integrases. While integration seems to be not dependent on target sequence, in vivo, retroviruses, such as HIV and murine leukemia virus, integrate preferentially into active genes (1, 2), introducing the possibility of insertional mutagenesis. The theoretical danger inherent in retrovirus-based gene therapy has been concretely demonstrated in a recent clinical trial in which the modified retrovirus integrated into the LMO2 locus, causing leukemia in three of the patients (4). A powerful system to circumvent this critical issue could be to develop a system to site-specific integrate the exogenous gene into a safety zone into the genome. To date, only one animal virus, the adenoassociated virus (AAV), has been identified that integrates its genome into a particular location into human chromosomal DNA. When cells are infected in the absence of helper virus, AAV establishes a latent infection in which the AAV genome integrates into a locus known as AAVS1 on the q arm of chromosome 19 (4). Recombinant AAV (rAAV) vectors too have a series of limitations as gene therapy vectors: they can accommodate only small genes and moreover eliminating most of wild-type AAV (wt AAV) sequences they have lost almost all the site-specific integration capability. On the basis of AAV site-specific integration machinery a series of effort to re-introduce wt AAV’s integration efficiency have been done. Different viruses have been engineered using the AAV’s integration machinery to transform them to target and integrate site-specifically large genes into the chr 19. Development of a maximized integrating, large capacity DNA viral vector is still an unmet goal of gene transfer technology. The initial aim of this project was to characterize the combination of the attributes of both the AAV and adenovirus (Ad) gene therapy vectors to develop an Ad/AAV hybrid virus system able to target site-specifically a large fragment of DNA into the host cell genome. In executing our experimental strategy, we found that, in addition to the known incompatibility of Rep expression and Ad growth, an equally large obstacle was presented by the inefficiency of the integration event when using traditional rAAV integrating elements. The finding that traditional rAAV plasmid vectors lack integration potency compared to wt AAV plasmid constructs led recently to the discovery of an AAV integration enhancer sequence element which functions in cis to an AAV inverted terminal repeat-flanked target gene. This study has addressed both of these problems. Moreover the project aimed also the capability of such vectors to target a large integrating cassette and the differences in the system integration efficiency compairing the dimension of the integrating cassettes. We demonstrated that an Ad can be generated that expresses Rep proteins and that Rep-mediated AAV persistence can occur in the presence of Ad vectors. We exploited the size limit capability introducing a large integrating cassette (12 kb) into the Ad/AAV vector and we obtained a good level of integration into the human genome. Specifically we succeeded in integrating it without any recombination event and in a site-specific fashion at good level. The model we extensively tested in human cell lines was also used successfully into human primary cells where we obtained site-specific integration into human chromosome 19 as expected. Another problem analysed was the flexibility of Ad vector system. Adenoviral vectors maintain the cellular specificity of adenoviruses from which they derive. A good gene therapy vector should have a broad tropism to ensure a good transduction efficiency, and moreover should be able to transduce cells of interest, such as CD34⁺ cells and other primary cells. We tried to expand Ad tropism engineering the commonly used Ad 5 vector (not able to transduce very well CD34⁺ and hematopoietic cells), transforming it to an Ad 5/35 modified vector. This approach permitted us to have better CD34⁺ cell transduction and a very good efficiency with hematopoietic cell lines.