advertisement
Mechanisms of AAV Transduction in Glaucoma-Associated Human Trabecular Meshwork Cells
Terete Borrás, winner WGA Award 2006
The research of our laboratory involves the search for alternative treatments of glaucoma by the use of gene transfer. Our goal is to use genes as drugs. Genes that, upon delivered to the trabecular meshwork of the eye, will have longer duration of action and less toxic effects than conventional drugs. Because glaucoma affects primarily the older population, and compliance is a major issue in the treatment of the disease, the use of a gene-drug, which could be delivered once every few months, would be of great advantage.
Recombinant viral vectors constitute the best delivery system to carry a gene
inside the human cells. There are several types of viruses developed for the use
of gene therapy, which are mainly constructed by removing their viral master genes
and substituting them with therapeutic genes. Each of these viral vectors has different
tropisms for different cell types. For the past few years our laboratory has been
successful in delivering functional genes to the trabecular meshwork by the use
of Adenoviral vectors (Ad). We have ac-
complished positive gene transfer in the eyes from post-mortem human donors, living
rats and, together with our collaborators at University of Madison (WI) in nonhuman
primates. These Ad vectors are very efficient and have a high tropism for the cells
of the trabecular meshwork. However, we also found that this type of vectors could
elicit an inflammatory response that in most cases could have been responsible for
the short expression of the transgene product.
Another type of viral vector, Adeno-associated vector (AAV), was known to transduce genes for longer periods of time in the retina with undetectable inflammation or immune response. In fact, an AAV vector has been reported to restore vision in dogs, in a case of a congenital Leber Amaurosis with no apparent deleterious effects. The delivered copy of the corrected gene in the dog was showing expression for over two years. Despite this success in the cells of the retina, AAV vectors were unable to transduce the trabecular meshwork. All our trials in cell cultures, organ cultures and living animals were negative. The first logical interpretation for this negative finding was that AAV lacked tropism for the trabecular meshwork and was thus unable to enter the cell.
Because inability of cell entry is normally due to the lack of an appropriate receptor in the receiving cell, it occurred to us that changing the coat of the virus by genetic engineering would result in successful transgene transduction. Since trabecular meshwork cells are rich in integrin receptors and have been shown to bind to extracellular proteins containing RGD peptides, we reasoned than a RGD-modified AA vector would facilitate entry of the AAV virus by specifically using the RGD-integrin interactions. Much to our dismay, the new RGD-pseudotyped AAV vector was equally unable to transduce the trabecular meshwork.
Our continued search for the mechanism responsible for this lack of AAV transduction took us to test a co-infection procedure where a very small amount of an Ad vector carrying no gene was mixed with AAV before adding it to the cells. This time, the AAV transduction was drastically induced. Believing that Ad had helped the AAV virus entering the cell, we designed experiments to measure the levels of AAV genomes inside the trabecular meshwork, both with and without co-infection. Our big surprise came when we stained the gels and observed that the levels of AAV genomes inside the cells were identical in all cases; in those cells where we have had transduction (co-infection) and in those where we have had none. We so realized that the AAV virus had been `fooling' us all the time and making us believe that it was unable to enter the cell. AAV had no problems entering the trabecular meshwork cell. The limitations of transduction seemed to occur later...
To get a global insight in what was happening inside the cells under conditions of AAV transgene transduction/no transduction, we decided to compare the trabecular meshwork cell transcriptome under both conditions. We isolated their RNA and used high density oligonucleotide gene chips to find differentially expressed genes. After all the bioinformatics analysis, we observed that many genes which were encoding cell proliferation and cell replication functions were downregulated under the non/transduction condition. Back to the drawing board, and after discussions with our collaborators experts in AAV biology, we learned that after AAV cell entry, the virus needs to convert its single stranded DNA genome into double stranded before it is able to transcribe the transgene. We immediately suspected that the results of the gene chips were telling us that the transduction rate limiting step was probably going to be at the level of DNA replication. Somehow the trabecular meshwork cell, in contrast to that from the retina, was not able to provide the necessary enzymes to convert the AAV single-stranded DNA to double-stranded.
A clever design by one of our collaborators conceived a second generation AAV whose genome (devoid of all viral genes) contains in tandem both the negative and positive strand of the transgene. Upon cell entry, the single-stranded DNA of such a virus, named self-complementary AAV (scAAV) would induce a pairing of both transgene strands and produce a double-stranded molecule without the need for endogenous replication enzymes. This double-stranded molecule could now go on to transcribe and translate the transgene. Thus, we predicted, a scAAV would then be able to bypass the rate-limiting step thought to occur in the trabecular meshwork.
Since the paper was published, we have been able, in collaboration with Paul Kaufman, to show scAAV transduction in living animals, both rats and monkeys, where expression continues to be seen by gonioscopy, in a quiet eye after several months. As the investigation continues, we have great hopes that these vectors carrying a therapeutic gene could be useful one day for a gene transfer/gene therapy treatment of glaucoma.
Note. This study has been the result of a friendly and open collaboration of all authors, especially those of the Gene Therapy Center at the University of North Carolina, Chapel Hill.
Co-authors: Wei Xue, Vivian W. Choi, Jeffrey S. Bartlett, Guorong Li, Richard J. Samulski and Sarah S. Chisolm
