Gene therapy to correct disease has intuitive appeal—if a disease is caused by a genetic defect, then repairing the gene should correct it. But this is easier said than done.
Mark A. Kay, M.D., Ph.D., of Stanford University presented a seminar at the Bay Area Scientific Discussion Group (South San Francisco, Calif., October 22, 2015) on the past, present and future of gene therapy. He defined gene therapy as the introduction of nucleic acids into cells to correct or prevent a pathological process. Conceptually, genes could be added or repaired, or DNA could silence a gene that controls a process.
The challenge of gene therapy is transporting an active pharmaceutical ingredient (API) to cells and then delivering the API to the right part of the cell so that it can modulate activity. Since the process of locating a tissue target is similar to a viral infection, many researchers choose to mimic a natural viral infection mechanism to deliver the API to the cell. The alternative is to administer free API (DNA, lipids or polymers, etc.), which often cannot be targeted to specific locations and may be toxic.
Nature provides a wide range of potential viral vectors, such as adenoviruses, retroviruses, plasmids and poxviruses. A virus has a capsid coat that encompasses the infectious agent. For gene therapy, the API is encapsulated with the capsid coat, which can be of nonliving origin. This reduces the risk of infection.
Dr. Kay’s research focuses on using adeno-associated virus (AAV), a member of the Parvoviridae family. The therapeutic gene material is synthesized in vitro using a range of molecular biology reactions, with the finished construct encapsulated in the coat. Vector purity and the full-to-empty capsid ratio are important considerations.
Dr. Kay and his team tried a construct AAV-Factor IX that was successful in rodents and dogs. In contrast to the animal tests, the expression in humans was temporary. According to Kay, “No matter how good the animal models, one cannot predict the outcome until you try it in humans.”
Other constructs were attempted, but problems were encountered with preexisting immunity of some patients to the capsid. In addition, the dose needed to be more than 10 times higher than that administered to the animals.
Also concerning is whether expression declines with time. The capsid shell may induce a long-term immune response, so a booster using the same or similar capsid might not work. Then it would be necessary to develop another capsid coating for a booster. Developing a new capsid would be akin to developing a new drug.
Dr. Kay described molecular shuffling of the capsid sequences to produce new viruses. The capsid sequences from a selection of wild-type AAVs were cut out of the AAV’s DNA and were fragmented, ligated and reinserted into the AAV DNA, producing a library of mutated viruses. The shuffling process was repeated several times to generate viruses with the lowest expression of AAV proteins. This produced new capsids that could be used as vectors, but there were still questions about the suitability of the approach. For example, there is risk of insertional mutagenesis, preexisting immunity or cell-mediated immunity.
The viral capsid approach suffers from the response idiosyncrasies of the human immune system and the confounding results obtained from the animal models. The latter may be avoided with a human tissue-on-a-chip approach that would bypass animals and produce results more quickly and directly.
Robert L. Stevenson, Ph.D., is Editor Emeritus, American Laboratory/Labcompare; e-mail: [email protected].