Search

Bioengineering of viral vectors for therapeutic gene delivery is a pivotal strategy to reduce doses, facilitate manufacturing, and improve efficacy and patient safety. Here, we engineered myotropic adeno-associated viral (AAV) vectors via a semirational, combinatorial approach that merges AAV capsid and peptide library screens. We first identified shuffled AAVs with increased specificity in the murine skeletal muscle, diaphragm, and heart, concurrent with liver detargeting. Next, we boosted muscle specificity by displaying a myotropic peptide on the capsid surface. In a mouse model of X-linked myotubular myopathy, the best vectors-AAVMYO2 and AAVMYO3-prolonged survival, corrected growth, restored strength, and ameliorated muscle fiber size and centronucleation. In a mouse model of Duchenne muscular dystrophy, our lead capsid induced robust microdystrophin expression and improved muscle function. Our pipeline is compatible with complementary AAV genome bioengineering strategies, as demonstrated here with two promoters, and could benefit many clinical applications beyond muscle gene therapy. ispartof: SCIENCE ADVANCES vol:8 issue:38 ispartof: location:United States status: published

Over the last decade, the field of gene and cell therapy has experienced a major turning point and has finally begun to fully realize its potential as a very attractive, versatile and innovative platform for the development of gene-based drugs. Gene therapy encompasses a spectrum of approaches, ranging from supplying missing genes to the correction of diseases at their molecular level, that all have in common the need of a vehicle (“vector") for specific and efficient delivery of therapeutic DNA or RNA. One branch of small viruses - the parvoviruses - have gained increasing attention as such vectors due to their non-pathogenicity, ease of engineering and low genotoxicity. Particularly, the adeno-associated virus (AAV) emerged as a top candidate, culminating in the authorization of three AAV-based gene therapy products, Glybera, Luxturna and Zolgensma. However, despite all the successes using recombinant (r)AAVs, there is still a demand for more specific vectors with larger DNA cargo capacity and lower immunogenicity. This need defined the scope of this doctoral thesis, which aimed at the construction and evaluation of new parvoviral vectors (derived from bocaviruses [BoVs]) and to increase the safety of vector application in humans. The first part of this work was fueled by a seminal study by Ziying Yan and colleagues in 2013, who used parvovirus cross-genera pseudotyping to combine an oversized rAAV2 genome of 5.5 kilobases (kb) with the capsid of the human bocavirus 1 (HBoV1). As reported, the rAAV2/HBoV1 vector could be produced efficiently and potently transduced primary human airway epithelial cells (pHAE). Here, we have validated and expanded on these intriguing findings by more comprehensively exploring the upper DNA packaging limit of the HBoV1 capsid. Notably, we found that up to 6.2 kb single-stranded (ss) - or 3.2 kb self-complementary (sc) - AAV genomes can be efficiently packaged into the HBoV1 capsid, as compared to only 5.1 (ssAAV) and 2.8 kb (scAAV) for AAV2, which has important ramifications for the delivery of complex rAAV vector DNA. Next, we further expanded this system to other primate BoV serotypes - three from humans (HBoV2, 3 and 4) and one from Gorilla (GBoV) - that have not been studied as vectors before. To this end, we successfully assembled the capsid genes of HBoV2-4/GBoV and produced chimeric rAAV/BoV vectors of all studied serotypes. With the help of reporter genes, we subsequently started to study and unravel the so-far unknown tropism of the new viral vectors. Strikingly, our screens on various primary cells and cell lines revealed that BoVs (especially GBoV) have a much wider tropism in vitro than previously anticipated. We found a wide range of primary and therapeutically relevant cells to be amenable to BoV infection, including human hepatocytes, T-cells and skeletal muscle cells. In addition, we obtained the first evidence that pseudotyped rAAV/BoV vectors also differ in their reactivity to pooled human antibodies (intravenous immunoglobulin, IVIg), which implies the possibility of vector re-dosing in rAAV/BoV-treated human gene therapy patients. Finally, we aimed to increase the fitness of BoV vectors and therefore employed a high-throughput diversification method called DNA family shuffling (DFS), to create the first library of chimeric BoV capsids. As hoped for, the library was packaging-competent, increased in titer over selection rounds and acquired a unique footprint when cycled in pHAE. Despite an excellent safety record of rAAV vectors, undesirable toxicity resulting from permanent gene expression represents a clinical concern. So far, the ensuing need to gain temporal control over vector persistence or expression has been addressed by using Cre recombinase or inducible systems that necessitate complex vector re-engineering. Thus, in the second part of this work, we aimed to overcome these limitations by introducing novel rAAV vectors that harbor a kill-switch (KS) based on the bacterial CRISPR II system (clustered regularly interspaced short palindromic repeats). This approach has two major components: (i) a (g)uide RNA expressed from the rAAV vector and (ii) the CRISPR/Cas9 endonuclease, which is supplied in trans and directed by the gRNA to a target site in the vector/transgene itself. We tested our KS system extensively in vitro and show a 10- to 100-fold reduction in transgene expression (Firefly luciferase) after supplying Cas9 in trans using ss and scAAV vectors for the expression of full-length and split Cas9, respectively. Moreover, we expanded our study to an in vivo application in mice, where we could recapitulate our findings in cell culture and trigger an up to 50% reduction in transgene expression. Finally, we devised a universal approach to inactivate any rAAV vector without further modifications. Therefore, we developed and experimentally validated self-inactivating (SIN) CRISPR vectors based on split Cas9 and ssAAVs that harbor the anti-target and anti-Cas9 gRNA and hence allow concurrent targeting of both. Moreover, we utilized different RNA polymerase III promoters (Pol III) to study and eventually optimize the effect of differential gRNA expression on the kinetics of both processes. Collectively, this work has yielded original BoV helper constructs and chimeras that represent valuable new tools to investigate fundamental and applied aspects of bocaviral biology, from the discovery of antigenic domains to the construction of designer viral vectors. Concomitantly, we have implemented and validated novel concepts to increase the safety of recombinant vectors including rAAV KS or SIN constructs that can be harnessed in future work, either alone or in combination with BoV capsids, to form the next generation of parvoviral vectors.

Anti-CRISPR proteins are powerful tools for CRISPR-Cas9 regulation; the ability to precisely modulate their activity could facilitate spatiotemporally confined genome perturbations and uncover fundamental aspects of CRISPR biology. We engineered optogenetic anti-CRISPR variants comprising hybrids of AcrIIA4, a potent Streptococcus pyogenes Cas9 inhibitor, and the LOV2 photosensor from Avena sativa. Coexpression of these proteins with CRISPR-Cas9 effectors enabled light-mediated genome and epigenome editing, and revealed rapid Cas9 genome targeting in human cells.