Delivery of gene and protein therapy to the brain has traditionally been extremely limited. Using a directed evolution approach to viral capsid engineering and selection, the Gradinaru group at Caltech identified specific peptides that, when displayed on the surface of modified capsids, enhanced neuronal transduction compared to the conventionally-used adeno-associated virus AAV9, following intravenous (IV) injection in mice. The brain uptake of these novel AAVs studied here by PET imaging by the Ferrara and Gambhir groups at Stanford reaches an extraordinary temporal-peak spatial-maximum of ~35% ID/cc at 4 h post-injection in mouse models. Our quantitative analysis yields a 50-fold enhancement in the brain receptor affinities as compared with earlier AAVs. Recent selections at Caltech have provided additional capsids that transduce the mouse brain with reduced expression in the liver, spleen, kidneys, and lungs. However, strain and species differences in the blood-brain barrier (BBB) and transport of these capsids (e.g. high in most mouse and rat strains but low in BALB/c mice) raise questions as to the nature of transport in primates (including humans). In order to understand species/strain differences and facilitate future translational development of these therapies, we propose novel combined positron emission tomography (PET) imaging techniques that non-invasively assess the pharmacokinetics of the AAV over the first days after injection and the resulting gene expression over months or potentially years. We plan to address key issues by assessing receptor binding, transcytosis and neutralizing antibody (NAb) effects across species. There are several innovative aspects to our approach. First, the radioactive tag used for tracking the capsid is based on a multichelator, increasing the signal-to-noise ratio and allowing us to assess binding to the brain endothelium (key for effective BBB crossing) over the first minutes and hours after injection. Second, PET analysis of AAVs as nanometer-scale therapeutics allows us to non-invasively estimate accumulation and clearance. Our data suggest receptor-mediated accumulation of the engineered AAVs on the brain endothelium over the first few minutes after injection. Third, for real-time reporting on gene transduction, we include a dual reporter system. The HSV1-sr39tk reporter gene has low background in the peripheral tissues with the reporter probe [18F]FHBG. The pyruvate kinase M2 (PKM2) reporter gene has a low background level in the brain and is imaged with [18F]DASA-23, a tracer that freely crosses the BBB. For preliminary studies in rodents we add optical capsid tags and reporter genes to assess cell-specific uptake. Our resulting specific aims are to: 1) validate and apply imaging to assess receptor binding affinity, transcytosis, clearance and transduction across organ systems and 2) image new variants of AAV9 across species to gain insight into the impact of capsid structure. We hypothesize that changes in capsid structure impact both affinity and endothelial transcytosis and that quantifying the binding affinity and transcytosis of multiple variants will inform future analyses of key structural components.
Novel adeno-associated viruses show the potential to enhance gene delivery and transfection by ~50 fold in mice, potentially facilitating urgently needed treatments. This proposal seeks to understand the mechanistic basis for this enhancement and to determine whether the methods can be extended to other species. We develop novel imaging methods required to understand these mechanisms and translate the therapies to large animals and, in the future, to human medicine.
