Abstract
As recombinant adeno-associated virus (rAAV) vectors have become the vector of choice for in vivo gene therapy, many investigators have focused on optimizing the AAV capsid as the vehicle for gene delivery. The emergence of nucleotide barcodes in combination with next-generation sequencing (NGS) has allowed for the screening of high complexity capsid libraries generated using any of a variety of synthetic and natural evolutionary approaches. The basic concept was that as a library of different AAV capsid variants was generated, each distinct rAAV variant would package a DNA sequence with a unique DNA barcode sequence. Once this library was injected into the experimental animal, the rAAV vector DNA from the target organ of interest would be sequenced using NGS. A higher abundance of a barcode DNA sequence would be taken to indicate greater efficiency of gene transfer by the corresponding AAV capsid. 1
However, there are limitations to previously published DNA barcode NGS techniques. A protocol that relies solely on vector DNA barcoding and sequencing may give an accurate assessment of the efficiency of AAV capsid binding to cells within the target organ, but it would not give any indication of the efficiency of subsequent steps of cellular entry, nuclear translocation, and expression, all of which are required for effective gene therapy. Approaches that have combined DNA barcoding with expression of a reporter transgene product, such as green fluorescent protein (GFP), do indeed indicate both DNA transfer and expression but could be limited in their effectiveness by immunologic responses to foreign protein. Such immune responses to reporter proteins may not accurately reflect how such rAAV vectors would perform with a therapeutic transgene product, which could be considerably less immunogenic than GFP, for example.
An innovative solution to this problem is described in an original article in this issue from the Gao and Tai laboratories. 2 These investigators have employed the tough decoy (TuD) RNA technology previously developed as a means to inhibit cellular miRNAs. 3 TuD RNA is a highly stable non-coding RNA with a pseudo-hairpin structure, which renders it highly resistant to degradation. There is a sufficient span of variable sequence within the TuD structure to insert a nucleotide barcode, generating what the investigators call a barcoded TuD (bcTuD). In their experimental approach, the rAAV genome delivers a cassette with a U6 promoter to drive expression the bcTuD. A stuffer DNA fragment is inserted downstream of the U6-bcTuD to provide a unit length rAAV genome suitable for efficient ss-AAV packaging. The in vivo screen is performed by injecting the rAAV-bcTuD library into the animal and, at the appropriate time point, harvesting organs for next generation RNA sequencing. Only rAAV vectors capable of both gene transfer and expression would demonstrate a high abundance of the corresponding barcode RNA sequences. In the paper, the team shows a highly reproducible, dose-dependent quantitative reflection of vector delivery and expression in experimental mice.
The rAAV gene therapy system is being rapidly deployed to address a wide range of genetic diseases affecting diverse cell types within the retina, central nervous system, muscle, and liver. 4 Nonetheless, the efficiency of this system remains suboptimal, and the selectivity of different rAAV capsid variants for different target cells is modest at best. These factors necessitate higher doses of vector and lead to greater immunologic responses to vector capsid antigens. In many circumstances, directed physical delivery methods, such as subretinal injection or direct central nervous system delivery are employed when selective vector targeting is desired. Thus, there is much room for improvement in the rAAV capsid, and capsid library screening is likely to remain an important strategy as new applications are pursued. The use of the bcTuD could provide an important improvement to this approach. In that way, the bcTuD technology may help to advance and disseminate the application of rAAV gene therapy further to treat many more of the estimated 7,000 human genetic diseases than are currently being addressed.