A recent paper in Nature by Lisowski et al., from mark Kay’s research laboratory has made some very interesting discoveries into the importance of different aspects of adeno-associated virus (AAV) biology and its application to human gene therapy. As has become very clear over the past decade, AAV vectors have become very favourable for gene therapies studies in experimental models, and also in human clinical studies, notably haemophilia (for example, see Nathwani et al.,). Notably, the discovery and characterisation of AAV vectors based on alternate strains (called serotypes) aside from the well characterised AAV-2 have incentivised many research efforts due to the alternate gene transfer profiles that these different vector systems offer. However, for successful in vivo gene transfer, it is not only the initial binding of the virus to the cell that defines it “infectivity profile” or “tropism”, but a series of sequential events including cell binding, internalisation, trafficking and endosome escape, uncoating, nuclear entry and maintenance that define the ability of any given vector system to show utility at the clinical level. An additional important consideration is species differences in virus:host interactions that govern gene transfer efficiency. Further, for clinical studies, AAV vectors are sensitive to neutralisation in human blood due to prior exposure of the human population to parental viruses through natural infections. This further defines utility, and often can limit recruitment into clinical studies (for example, see Jessup et al.,).
In a pivotal recent trial in haemophilia B patients, AAV vectors based on serotype 8 (AAV-8) were used as a gene therapy agent to induce long term correction through AAV-induced production of coagulation factor IX from the liver. This was achieved through high dose delivery of AAV-8 through a peripheral vein (AAV-8 accumulates preferentially in the liver). In this dose escalation study, it was shown that high doses of AAV-8 were associated with higher levels of FIX production in patients, leading to beneficial clinical outcome. Although many studies in rodent and non-rodents have characterised and documented gene transfer efficiencyof AAV vectors, there is limited data in human. It would be clearly advantageous to achieve clinical benefit with reduced AAV vector doses. To address this, Lisowski and colleagues used an important model system. They repopulated immune deficient mouse livers in vivo with primary human liver cells (hepatocytes) and subjected mice to detailed in vivo AAV gene transferstudies, through their ability to track differential analysis of AAV uptake and gene transfer into mouse and human cells from the liver. They used a number of AAV vectors as well as an AAV-shuffled library that contained variants of many different AAV vectors, the latter to identify AAV variants with improved kinetics for human hepatocyes.
Their key findings were:
- That AAV-2 showed equal efficiency of uptake into human and mouse hepatocytes (both relatively inefficient).
- That AAV-8 showed much higher efficiency for mouse hepatocytes compared to human.
- That a novel variant (AAV-LK03) identified from the in vivo library screening, creatively developed for assessing selective uptake and productive infection of AAV into human hepatocytes in vivo, showed high propensity for infection of human hepatocytes over mouse cells, showed resistance to neutralisation by pooled components of human sera and used the c-met receptor for virus-cell attachment. Thereby, the authors identified an exciting new AAV vectors for clinical development.
These studies highlight the importance of species differences in gene transfer when using recombinant AAV vectors, and thereby the possible limitations that can govern clinical utility of existing AAV vectors. Although it is clearly important to further develop AAV-8 vectors following the clear early success of the haemophilia B trial (Nathwani et al.,), new AAV variants that offer enhanced gene transfer in to human cells may clearly be very effective clinically. Further, for different applications in gene therapy, it will be important to bring the advances brought by this paper into preclinical studies in order to identify and develop the most efficient and effective vectors. It will be very interesting to see how the AAV-LK03 vector is developed into clinical studies for liver-directed gene therapy.
Jessup, M., Greenberg, B., et al. (2011). Calcium Upregulation by Percutaneous Administration of Gene Therapy in Cardiac Disease (CUPID): a phase 2 trial of intracoronary gene therapy of sarcoplasmic reticulum Ca2+-ATPase in patients with advanced heart failure. Circulation 124(3): 304-313.
Lisowski, L., Dane, A. P., et al. (2013). Selection and evaluation of clinically relevant AAV variants in a xenograft liver model. Nature [Epub ahead of print].
Nathwani, A. C., Tuddenham, E. G., et al. (2011). Adenovirus-associated virus vector-mediated gene transfer in hemophilia B. New England journal of medicine 365(25): 2357-2365.