Abstract
Recombinant adeno-associated virus (AAV) is one of the leading vectors being tested as a gene delivery vehicle for the treatment of many diseases, both in preclinical studies and clinical trials. AAV vectors deliver genes to various tissues following systemic injection and provide robust and long-term gene expression, often without evoking detectable immune responses. While many advances have been made in AAV-mediated cardiac gene therapy, a variety of challenges remain, three of which we investigated. First, although cardiac-specific transgene expression has been demonstrated, gene knockdown has yet to be achieved selectively in the heart. Second, because it has a lag phase before reaching maximal expression, AAV is delivered prior to myocardial infarction (MI) in most preclinical studies of heart failure, but would be more valuable in the clinical setting if it could be administered post-MI. Third, fibroblasts account for more than half of the cells in the heart and would be a viable target for many cardiac-directed therapies, but so far have not been effectively targeted by AAV.
We proposed several lines of investigation designed to address these challenges by bioengineering AAV vectors for cardiac-targeted gene therapy. First, to assess the potential for cardiac-selective gene knockdown, we developed an AAV9 vector to deliver short hairpin RNA against GFP to inhibit gene expression in transgenic ubc-GFP mice. We then tested the hypothesis that single-stranded AAV9, but not double-stranded AAV9, could provide cardiac-selective knockdown due to inefficient conversion of single-stranded AAV genomes into double-stranded DNA in the liver. Second, our lab has demonstrated that AAV9 provides robust and accelerated expression in cardiomyocytes after ischemia and reperfusion, but the mechanisms behind this ischemic induction effect are unknown. We attempted to elucidate these mechanisms, which we hypothesized were related to vascular permeability and AAV9 receptor availability. Finally, we developed an AAV expression cassette with the aim of targeting expression to cardiac fibroblasts and set out to determine the best serotype and time-point after myocardial infarction for targeting gene therapy to cardiac fibroblasts.
In the first aim, to provide cardiac-selective knockdown, we were successful in restricting knockdown primarily to cardiac tissue, with slight mRNA knockdown but no protein reduction in liver and skeletal muscle. However, our hypothesis regarding low liver knockdown – inefficient conversion of single-stranded AAV genomes into double-stranded DNA in the liver – was not supported by the data, with cardiac and liver knockdown being comparable between single- and double-stranded AAV. Results from the second aim, to elucidate the mechanisms behind ischemic induction of AAV9, were less clear due to unanticipated experimental confounds. Injection of several compounds to the myocardium revealed that tissue injury likely leads to an increase in AAV transduction primarily through post-entry cellular processing of AAV rather than through increased viral uptake by the cells, as hypothesized. Finally, our attempts to transduce cardiac fibroblasts with AAV instead led to gene expression in cells that express markers for hematopoietic stem cells but not fibroblasts. We obtained higher expression in these cells with AAV9 than with AAV1 or AAV6, and found increased expression when mice were injected 2 days post-myocardial infarction compared to mice injected at reperfusion.
