Browsing by Author "Asokan, Aravind"

Introduction

Methods

Results

Discussion

Therapeutic gene transfer and genome editing require effective delivery of genetic cargo to target cells and tissues. Recombinant adeno-associated viral (AAV) vectors are a promising delivery platform, but ongoing clinical trials continue to highlight a relatively narrow therapeutic window. Efforts to optimize vector dosing or engineer improved vectors are confounded, at least in part, by differences in AAV biology across animal species. Here, we present a broadly applicable, cross-species evolution approach to tackle this challenge. Specifically, I iteratively cycled AAV libraries administered intravenously and amplified isolates from CNS tissue in pigs, mice, and non-human primates to generate cross-species compatible AAVs (ccAAVs). By sequentially evolving AAV libraries in three different species, we discover a highly potent variant (AAV.cc47) that demonstrates improved attributes benchmarked against AAV serotype 9 (AAV9). Increased potency of AAV.cc47 is evidenced through robust reporter gene expression as well as Cre-mediated recombination and CRISPR/Cas9-mediated genome editing in a fluorescent reporter mouse model. Enhanced transduction efficiency of AAV.cc47 vectors is further corroborated in macaques and pigs, providing a strong rationale for potential clinical translation into human gene therapies. Lastly, we report increased expression of a therapeutic acid alpha-glucosidase (GAA) transgene in a mouse model of Pompe disease and enhanced restoration of dystrophin through CRISPR/Cas9 gene editing in the mdx mouse model of Duchenne Muscular Dystrophy using AAV.cc47 vectors. We discovered another cross-species compatible variant (AAV.cc84) with enhanced CNS transduction and de-targeted from the liver in vivo compared to AAV9. We demonstrate improved targeting of neurons in the brain following both systemic and intracerebroventricular (ICV) injection of reporter vectors in mice. Reporter gene expression and vector genome biodistribution reveal a liver detargeted phenotype with AAV.cc84 compared to AAV9. Lastly, enhanced neuronal transduction was confirmed in the pig CNS following intrathecal infusion of AAV.cc84 reporter vectors. Taken together, we envision that ccAAV vectors can potentially improve predictive modeling in preclinical studies as well as clinical translatability by broadening the therapeutic window of AAV based gene therapies.

Adeno-Associated Virus (AAV) is a gene therapy vector with immense clinical importance. However, its use as a template for homology directed repair has come under greater examination particularly for the generation of site-specific recombined Chimeric antigen receptor or CAR T-cells. This is because traditional CAR T-cells generated from retro- or lentiviral vectors have risks for insertional oncogenesis or exhaustion from tonic signaling due to use of a constitutively active promoter, both problems which AAV directed knockins may overcome. In fact, the use of site-specific knockin CAR T-cells have now entered clinical trials for the treatment of CD19+ blood-borne cancers. While the use of these next-gen AAV generated CARs have excelled for liquid tumors, their use for the treatment of solid tumors has lagged. This is due to poor preclinical modeling for solid-tumor directed CARs which take place in immunocompromised mouse models that do not fully recapitulate the tumor microenvironment known to be problematic for infiltrating lymphocytes. Thus, there is a clear need to evaluate these therapeutics in immunocompetent hosts, however, there exist no known AAV serotype that can effectively target murine T-lymphocytes to generate these site-specific knockins. To ameliorate this problem, we employed a capsid evolution from the AAV6 background to generate a murine T-lymphocyte tropic AAV variant dubbed Ark313. Ark313 is vastly superior to the parent serotype in transducing, gene editing and site-specific knockins in murine T-cells. To characterize how this was happening, we employed a genome wide CRISPR knockout screen in murine primary T-cells to reveal the essential factor for Ark313 transduction to be Qa-2, a non-classical MHC-1b molecule. Due to the restricted tissue expression of the Qa-2 antigen, we injected mice systemically with Ark313 and saw it could transduce up to 25% of spleen resident T-cells including naïve/memory/effector subsets when using a self-complementary transgene. Additionally, Ark313 displayed a liver de-targeted tropism reducing potential off target tissue transduction when employing an ubiquitous promoter. Together we have generated a novel tool for the facile genetic manipulation of murine T-cells both ex and in vivo. We believe Ark313 will be a fundamental reagent to employ when interrogating T-lymphocyte immunotherapeutic questions and for investigating immune basic biology. This work lays the groundwork for the development of human lymphocyte targeting AAVs for generating CARs to combat liquid and solid tumors via systemic dosing.

Recombinant adeno-associated virus (AAV) has emerged as a leading platform for therapeutic gene transfer. Despite significant progress and a string of FDA approvals, significant challenges that prevent the broader application of AAV gene therapy exist. High systemic dosing of AAV vectors in the clinic poses potential risk of severe and adverse side-effects due to anti-capsid immunity. While different immune modulatory regimens (IMR) are being explored, there is an urgent need for continued development of effective strategies to improve the safety of AAV gene therapies. In chapter 2 of this dissertation, I describe the discovery of a novel enzyme (IceM) that cleaves human IgM, a key trigger in the anti-AAV immune cascade. We then engineer a fusion enzyme (IceMG) with dual proteolytic activity against human IgM and IgG. IceMG cleaves B cell surface antigen receptors (BCR) and inactivates phospholipase gamma signaling in vitro. Importantly, IceMG is more effective at inhibiting complement activation compared to an IgG cleaving enzyme alone. Administration of IceMG in rhesus macaques enables robust and reversible clearance of both circulating IgM and IgG. Antisera from macaques treated with IceMG shows significantly decreased AAV neutralization as well as complement activation. Finally, we demonstrate that pre-treatment with IceMG restores AAV vector transduction in mice passively immunized with human antisera containing anti-AAV neutralizing antibodies. Thus, IgM cleaving enzymes show promise in simultaneously addressing multiple aspects of anti-AAV immunity mediated by B cells, circulating antibodies and complement. These studies have implications for improving safety of AAV gene therapies and more broadly, for use in organ transplantation and autoimmune diseases. In parallel, unraveling the biology of AAV entry and trafficking is central to developing improved AAV vectors. In chapter 3 of this dissertation, I delve into the biology of host factors that limit AAV transduction. We identify furin as a host factor that significantly restricts the transduction of AAV4-like serotypes. Through the interrogation of different steps and attachment factors in the AAV infectious pathway, we demonstrate that AAV cellular binding and uptake are significantly increased in a sialic acid-dependent manner. We postulate that furin likely plays a key role in regulating expression of cellular sialoglycans, which in turn, can influence permissivity to AAVs and possibly other viruses.

The type VI-Cas13 Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) enzymes of the bacterial adaptive immune system have been repurposed as programmable RNA guided, RNA targeting nucleases for eukaryotic RNA editing1–4. In these systems, the Cas ribonucleoprotein (RNP) is guided to its target RNA transcript by a single CRISPR RNA (crRNA), which it cleaves following satisfactory RNA base-pairing. The RNase domain of the Cas13 protein may be inactivated to serve as a RNA guided RNA binding protein with high affinity for its RNA target5–8. This has expanded the utility of Cas13 beyond programable RNA knockdown. The catalytically dead Cas13 (dCas13), now, can serve as a chassis upon which effector domains have been added to catalyze site-targeted single base edits, demethylation, RNA cleavage and alternative splicing with improved specificity and efficiency relative to existing antisense RNA technologies5,9,10. However, the ability to edit large stretches of mRNA transcripts remains a significant challenge.RNA splicing is well a conserved process in higher eukaryotes that joins the protein coding RNA regions (exons) along the same transcript via a dual trans-esterification reaction in cis11. The intergenic region (intron) between these two exons choreographs this reaction by recruiting the spliceosome to conserved molecular signatures. In this work, type VI CRISPR-Cas13 systems are leveraged to facilitate RNA rewriting by hijacking the cell’s RNA splicing machinery via an approach referred to herein as CRISPR Assisted RNA Fragment Trans-splicing (CRAFT). To achieve such, a Cas13 crRNA was linked with a sequence containing these intronic molecular signatures (hemi-intron) and one or more exons in a single recombinant trans-splicing RNA (rcRNA). This chimeric RNA was co-expressed with a catalytically inactive cognate dCas13 nuclease. This RNP employs CRISPR to target nascent pre-mRNA species, while the conserved molecular signals of the hemi-intron mediate incorporation of the linked exon(s) into the mature transcript in trans. Using split reporter-based assays, we evaluate orthogonal Cas13 systems, optimize guide RNA length and position for optimal trans-splicing across a range of intronic targets. We achieve markedly improved editing of large 5’ and 3’ segments in different endogenous mRNAs across various mammalian cell types compared to other spliceosome-mediated trans-splicing methods. Additionally, we demonstrate that CRAFT can serve as a versatile platform for attachment of protein tags, studying the impact of multiple mutations/single nucleotide polymorphisms, modification of untranslated regions (UTRs) or replacing large segments of mRNA transcripts. We validate CRAFT across different transformed and primary cell lines, multiple endogenous transcripts, develop a novel barcoded approach for guide selection to facilitate mRNA trans-splicing and affirm that CRAFT can serve as a promising tool for gene therapy and studying RNA biology.