Recombinant adeno-associated viral vectors (rAAV) are showing promising clinical results in a few diseases including hemophilia B, hereditary blindness and lipoprotein lipase deficiency. Even with the early limited success in treating simple disease entities, there are a number of limitations with the current vectors. For example, in the hemophilia B trials, AAV8-hFIX transduced patient livers after IV infusion and an improvement in the bleeding diathesis was demonstrated, yet the following difficulties remain: 1) some patients develop a transient transaminitis thought to be due to a T-cell response directed against AAV8 capsids; 2) transgene expression based on vector dose per body mass is far less than predicted from mouse studies (~10x) and non- human primate studies (~3-5x), suggesting transduction in human hepatocytes is not optimal; 3) pre-existing neutralizing anti-AAV8 antibodies are found in about 1/3 of patients; 4) re-administration of a different vector maybe required; 5) a limitation i the size of AAV genome packaging makes it difficult or even impossible to treat other similar diseases (e.g. hemophilia A, the more common form of hemophilia). Due to the fact that significant differences in rAAV transduction are dictated by small variations in capsid amino acid sequences, we have pursued molecular shuffling of capsid genes to create AAV libraries with extensive sequence variants (~10e7). These novel capsids replace the wild-type capsid in a viral plasmid and are used to produce infectious virus in the presence of a helper virus. Replicating AAV is grown under selective conditions that we vary according to the desired outcome, and those with a selective advantage are isolated and their sequences determined. These capsids can then be used to package therapeutic genes and their transduction determined. We and others have selected AAV capsids with new properties, many of which have become very useful to the gene therapy community. However, we feel the full utility of replicating capsid libraries has yet to be reached. We plan to assemble new more complex libraries and select for enriched capsids after serial passage in primary human hepatocytes maintained in a chimeric mouse-human liver model under different selection pressures. We will make vectors from the most selected capsids and study their transduction in vitro and in vivo. Finally, we will use these libraries in two high-risk, high-payoff screens: 1) select capsids with novel cell specificity (e.g. an AAV that only transduces hepatocytes infected with hepatitis- C virus); 2) select capsids that can package larger genomes, thus expanding the utility of AAV vectors for diseases whose expression cassette just exceeds current packaging limits (e.g. cystic fibrosis, Duchenne muscular dystrophy and hemophilia A,). We believe this work will not only provide new AAV vectors for expanded use in research and clinical gene therapy, but will also provide important quantitative guidelines to make AAV capsid shuffling more efficient for all groups pursuing this line of research.
Recombinant adeno associated viral vectors (rAAV) show great potential in early clinical trials and preclinical studies. However, there are still significnt limitations and challenges with the current available vectors. The goal of this work is to generate new AAV capsid libraries and perform high throughput screens to isolate viruses with unique infection/transduction properties. These capsids will then be used in creating new AAV vectors and evaluated for their potential in preclinical animal models.