Preview

PDF, English Download (4MB) | Terms of use

Abstract

Adeno-associated viruses (AAVs) present powerful vectors for human gene therapy and biomedical research. They enable the delivery of transgenes to a broad range of target tissues. This allows persistent expression of reporter genes, therapeutic gene replacement and the delivery of control elements for manipulating endogenous gene expression. Transduction efficiency and specificity for on-target over off-target cells are critical factors driving vector safety and applicability. In this study, I engineered AAV capsids, promoters and knockdown tools with the goal of generating efficient and specific vector components for the next generation of cell-type-specific gene therapy vectors. In the first part of this doctoral work, I utilized Cas13d (CasRx) and short-hairpin (sh)RNA effectors to assess AAV-induced RNA degradation (knockdown). While CasRx showed promising knockdown of a Renilla luciferase reporter target, it failed to silence endogenous CD44, a potential driver of metabolic (non-alcoholic) steatohepatitis. shRNA effectors, however, allowed robust target knockdown of cellular RNAs (CD44 and ACE2) as well as SARS-CoV-2 viral RNA. Direct targeting of SARS-CoV-2 genomic RNA triggered the evolution of escape mutations within the viral target sites. This mutational escape was efficiently suppressed by multiplexing of three shRNAs in a single AAV vector, thereby allowing a sustained suppression of SARS-CoV-2 infection in Vero E6 cells. In the second part, I focused on improving screening conditions for promoters and AAV capsids. By assessing eYFP reporter expression in vivo for four promoter constructs individually, I could validate the findings of a previous promoter screen. This screen had utilized high-throughput barcode sequencing for parallel readout of a library of AAV-promoter constructs. I then applied this barcoding technique to dissect the activity of the GFAP promoter and truncated versions thereof. The GFAP promoter has previously mostly been used to induce astrocyte-specific transgene expression in the central nervous system. Strikingly, though, my results demonstrate a highly efficient GFAP promoter-driven transgene expression in human and murine hepatocytes. To optimize the directed evolution of AAV capsids, I modified conventional capsid library screening by altering selection parameters. This was achieved by (i) introducing a Cas9-based negative selection for the removal of unwanted variants from the capsid library, and (ii) by exploring and applying RNA-based functional selection. I could generate an RNA-based screening platform by driving the expression of cap from the ubiquitous CMV promoter instead of the endogenous p40. Both screening approaches proved applicable in cell culture settings. As RNA-driven selection offers a functional readout from both on- and off-target cells, I applied this approach for in vivo screening of an AAV6 peptide display library in mouse non-parenchymal liver cells. CMV promoter-driven cap expression enabled RNA-based readout of variant enrichment for on- and off-target cell-types. This demonstrated improved selectivity for RNA- over conventional DNA-based screening and facilitated the identification of functional, cell-type-specific capsid candidates. In conclusion, my results show the development of efficient combinatorial shRNA-based knockdown vectors for inhibiting CD44 expression or SARS-CoV-2 infection in vitro. Furthermore, I could implement improvements in capsid and promoter screening. This allowed the detection of highly functional variants with potential future applications in the development of novel gene therapy vectors.