AAV-mediated gene replacement is a powerful approach to treat genetically defined disease. It is often believed that there is a straightforward path to the clinic once gene replacement efficacy is shown in preclinical models. However, major obstacles limit safe and effective gene therapy for many diseases, particularly in systemically delivered contexts. Some diseases are not yet tractable because therapeutic gene expression in one tissue might be toxic in another, or because the therapeutic window of cargo expression is too narrow to safely administer. Toxicity in liver and dorsal root ganglia have been observed in several gene therapies, regardless of capsid or cargo. More versatile, fined-tuned, and self-regulated gene expression cassettes are required for safer and more effective gene therapies. Many approaches exist to regulate gene therapy cargoes, including synthetic promoters and tissue-specific microRNA binding sites. However, despite the ubiquity of RNA processing in the genome, few efforts have incorporated alternative splicing into gene therapies. In Aim 1, we will leverage tissue- specific splicing patterns to generate cargoes facilitating expression in certain tissues but not others. In a first example, we will use develop methods to restrict cargo expression to skeletal muscle and de-target the heart. We have already incorporated muscle-specific exons not expressed in heart into AAV and tested their activity in vivo; we will further optimize these cassettes by testing hundreds of splice site and cis-element motif variations. In a second example, we will identify and test exons that de-target liver and dorsal root ganglion but preserve expression in muscle, heart and/or central nervous system tissues. We will individually validate “winner” cassettes for both examples. We term this approach Tissue-specific Alternative splicing to Restrict Globally Expressed Therapeutic-AAV (TARGET-AAV). In Aim 2, we will re-purpose naturally occurring auto-regulatory cassettes to design and test gene therapies that can sense and regulate their own expression levels. We will use RNA binding proteins mutated in motor neuron disease as test cases, given that RNA binding proteins are well established to regulate their own expression. We will incorporate intronic miRNAs knocking down the corresponding endogenous proteins in the same cassette as an auto-regulated, healthy version of the same RNA binding protein. Similar to Aim 1, we will optimize auto-regulatory behavior by making alterations to intronic and exonic sequences. We will establish proof-of-concept for this approach in cell culture and in iPSC-based models of these disease. We call this approach Biologically Regulated Interchangeable Tuneable Elements (BRITE). Completion of this work will provide guiding principles for the field of gene therapy to incorporate alternative splicing into gene therapy cargoes.

The proposed research is relevant to public health as it develops and establishes alternative splicing as a powerful and viable method to control regulation of gene therapies. Current gene therapies do not take advantage of this ubiquitous regulatory mechanism, and we propose to use it to achieve tissue-specific and self-regulatory control over AAV-delivered cargoes. Findings from this study will constitute new compositions and guiding principles for the development of gene therapy cargoes.