title: Secrets to finding the ideal mate:  New insights into parameters that govern successful Adeno-associated virus (AAV) vector evolution  
creator: Kienle, Eike Christoph
subject: 500
subject: 500 Natural sciences and mathematics
subject: 570
subject: 570 Life sciences
subject: 600
subject: 600 Technology (Applied sciences)
description: The success of human gene therapy - the treatment of hereditary or acquired diseases with a genetic cause - depends on potent, safe and specific gene delivery vectors. Amongst the large variety of currently available non-viral or viral vectors, those derived from Adeno-associated viruses (AAV) are particularly attractive due to a unique combination of assets: the virus/vector is apathogenic, poses little risk of insertional mutagenesis, has a broad host and cell range, and can easily be modified and produced in large quantities. However, no natural AAV fulfills all the requirements for clinical use in humans, raising a need for new technologies and strategies to engineer synthetic “designer” vectors.   Accordingly, the central aim of this work was to improve and apply two fundamental methods for molecular AAV vector evolution, DNA family shuffling and peptide display. The first technology relies on fragmentation and homology-based recombination of capsid genes from closely related AAV serotypes, resulting in libraries of AAV chimeras from which capsids with desired properties can be enriched via subsequent selection. Here, we initially assessed and optimized the key steps for AAV shuffling, culminating in a robust and standardized new protocol for AAV library generation. Next, we used this protocol to shuffle AAV2, 8 and 9, and exploited the resulting library to analyze two major parameters for AAV selection - helper virus and anti-AAV-antibodies. Interestingly, we found that AAV capsids selected in the presence of a helper virus give stronger gene expression, implying that their intracellular processing is enhanced. Moreover, comparative analyses of AAVs isolated under various conditions with human antisera showed that the degree of negative selection pressure determines the balance between infectivity and immunity of the viral particles. Finally, we also performed a helper virus-free in vivo biopanning with a library comprising AAV1, 5, 6, 8 and 9 in murine pancreas. Intriguingly, gene expression from the single clone emerging after three selection rounds was low, supporting our conclusion that the presence of a helper virus is key for potent AAV vector evolution.    Unlike shuffling, AAV vector improvement through peptide display starts with a single viral serotype (traditionally AAV2) whose capsid is expanded by insertion of 7-9 aa into an exposed loop, with the aim to alter vector tropism towards desired target cells. Here, we extended this strategy to 11 alternative AAV serotypes and demonstrate their tremendous, previously unrecognized potential as scaffolds for viral peptide display. Therefore, we first implemented a simple PCR protocol for rapid cloning of peptide-encoding sequences into AAV capsid genes, which replaces the original stepwise mutagenesis. We then used our new approach to insert 6 distinct peptides into all 12 AAV serotypes, resulting in a collection of 84 YFP-encoding vectors (12 wildtypes & 72 mutants). While screening this panel in a large array of human and non-human cell lines and primary cells, we made three important observations: i) AAV vector transduction is not determined by the peptide alone but largely depends on the capsid context; ii) alternative AAV serotypes with certain peptides frequently outperform the AAV2 prototype and its peptide derivatives; and iii) some serotype-peptide combinations even allow to transduce cells previously considered refractory to AAV infection. Subsequent analysis of further peptides showed that a common motif, NxxRxxx, is enriched in the best performing candidates and particularly enhances AAV1, 7-9 and rh.10. Based on these findings, we assembled a “Master panel” of vectors including the superior serotype-peptide combinations from our various screens, and used it in collaborative studies to select potent new AAV vectors in clinically relevant cell types, such as myeloid cell lines, primary human myeloma cells, or dorsal root ganglia and proprioceptive neurons.  As a whole, the work in this thesis makes a number of essential contributions to the fields of AAV biology, AAV vector development and human gene therapy. First, it resulted in optimized protocols and tools that substantially simplify, standardize and accelerate the future generation of tailored AAV capsids for vector engineering. Second, it also yielded a wide variety of new AAV variants - either in the form of shuffled libraries or as panels of specific capsid-peptide combinations - that can now be screened in further cell lines, primary cells or even directly in vivo. Third, the data obtained in this thesis with the various wildtype, shuffled or peptide-modified capsids greatly improve our knowledge of fundamental steps in cellular AAV infection. Most importantly, our results consistently exemplify that the function of AAV capsids is not determined by single residues, but rather results from very complex interactions of different regions that are dispersed throughout the viral shell. Altogether, the present thesis fuels the belief that AAV is one of the most versatile, powerful and robust viral vector systems available today and that it provides unique benefits for clinical translation in humans.   
date: 2014
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserverhttps://archiv.ub.uni-heidelberg.de/volltextserver/17851/1/Dissertation_E_Kienle_v1.pdf
identifier: DOI:10.11588/heidok.00017851
identifier: urn:nbn:de:bsz:16-heidok-178516
identifier:   Kienle, Eike Christoph  (2014) Secrets to finding the ideal mate: New insights into parameters that govern successful Adeno-associated virus (AAV) vector evolution.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/17851/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng