Many cell types together assemble the functional circuitry of the human brain. For over a century, neuroscientists have categorized brain cell types by their features, including shape, position, physiology, molecules, and function. Single cell transcriptomics studies are now defining molecular cell types at a resolution not previously possible, uncovering a taxonomy of hundreds to thousands of brain cell types. These studies have also revealed dramatic differences in molecular signatures of homologous cell types across species, showing decisively that the difference between mouse and human brain is not simply the total number of neurons. However, the function of each cell class or type in brain circuitry, and dysfunction in disease, is only beginning to be evaluated. To characterize the roles of human brain cell classes in normal function and disease, it is critical that tools be developed to allow genetic access to cell classes in vivo. Such tools would enable precise therapeutic gene delivery to brain cell classes, permitting targeted treatment for class-specific etiologies like some epilepsies. Few genetic tools are available to mark and manipulate cell classes and types in non-genetically tractable species like human and non-human primate (NHP). Viruses including adeno-associated viruses (AAVs), containing cell class and type selective enhancers can be leveraged to gain genetic access to, and drive gene expression in specific brain cell classes in these species. We have initiated a project through the BRAIN Initiative to generate and validate reporter AAVs to mark specific cell classes in the mouse cortex in vivo and in human neocortical tissue ex vivo. Our groups have engineered AAV vectors and optimized capsids to access neurons and express transgenes in many discrete cell classes and types in mouse and primate. New and improved AAV tools promise to fuel human brain scientific discovery and clinical progress, but one impediment has been the costly and time-consuming process of validating new vectors in primates. We present three Aims to translate these promising new AAV vectors into a high-value set of primate-optimized tools that could eventually be used for gene therapies in humans. First, we will develop a platform for screening AAV vectors in NHP ex vivo brain slices, followed by individual validation of promising vectors in NHP in vivo and human ex vivo brain slice cultures. Second, we will identify optimal AAV capsids to: a) support widespread NHP neuronal transduction in vivo when applied intravenously or to cerebrospinal fluid (CSF), two preferred routes of delivery for human CNS gene therapy, and b) support AAV transduction of human primary brain tissue ex vivo. Third, we will perform proof-of-concept experiments using cell class-selective vectors to express a therapeutic transgene in defined classes to treat a severe and intractable form of childhood epilepsy called Dravet syndrome (DS). These experiments represent a significant step towards converting cell class-selective AAVs into first-in-class viral tools optimized for in vivo NHP brain studies and human gene therapy applications.
The human brain is composed of a remarkable diversity of cell types and classes, and at present, we have limited knowledge about the distinct contributions of each cell type and class to normal brain function. Furthermore, brain disorders are generally regarded as the consequence of cell type and circuit dysfunction, but the dearth of well-validated molecular genetic tools available to directly explore this relationship in large primate brains has constrained progress, especially for translational research. Our proposal aims to remedy this by engineering novel AAV capsids to allow enhanced neuronal transduction of primate brain, developing a suite of optimized AAV vectors for more precise cell type and cell class-selective delivery of transgenes, and testing if virus-mediated restoration of sodium channel function in forebrain inhibitory neurons can correct the devastating neurological deficits of Dravet syndrome.