SelectiveIn VivoTargeting of Human Liver Tumors by Optimized AAV3 Vectors in a Murine Xenograft Model

Abstract

Current challenges for recombinant adeno-associated virus (rAAV) vector-based cancer treatment include the low efficiency and the lack of specificity in vivo. rAAV serotype 3 (rAAV3) vectors have previously been shown to be ineffective in normal mouse tissues following systemic administration. In the present study, we report that rAAV3 vectors can efficiently target and transduce various human liver cancer cells in vivo. Elimination of specific surface-exposed serine and threonine residues on rAAV3 capsids results in further augmentation in the transduction efficiency of these vectors, without any change in the viral tropism and cellular receptor interactions. In addition, we have identified a potential chemotherapy drug, shikonin, as a multifunctional compound to inhibit liver tumor growth as well as to significantly enhance the efficacy of rAAV vector-based gene therapy in vivo. Furthermore, we also document that suppression of tumorigenesis in a human liver cancer xenograft model can be achieved through systemic administration of the optimized rAAV3 vectors carrying a therapeutic gene, and shikonin at a dose that does not lead to liver damage. Our research provides a novel means to achieve not only targeted delivery but also the potential for gene therapy of human liver cancer.

Efficient and Targeted Transduction of Nonhuman Primate Liver With Systemically Delivered Optimized AAV3B Vectors

“…Additionally, we have generated tyrosine-mutant rAAV3B serotype vectors, and identified an optimized vector that efficiently transduces human liver tumours in a murine xenograft model in vivo 13 . In an attempt to further enhance the transduction efficiency of rAAV3B vector, we evaluated AAV3B vectors A c c e p t e d m a n u s c r i p t human liver tumor cells and primary human hepatocytes in vitro 14,15 . Moreover, the use of the rAAV3B-ST vectors not only led to targeted delivery and suppression of human liver tumorigenesis in a murine xenograft model, in our preliminary studies, it also resulted in efficient transduction of primary human humanized murine livers in vivo 14 .…”

“…All rights reserved(S663V+T492V) in the AAV3B capsid that significantly augmented the transduction efficiency of these vectors in liver cancer cells in a xenograft mouse model in vivo13,15 in primary human hepatocytes in vitro; and in humanized mouse and NHP models in vivo, it should be noted that this most efficient mutant was identified from among 10 different permutations and combinations using a single human cancer cell line, Huh7, in vitro 15 . Thus, it remains possible that one or more of the mutant AAV3B vectors will be optimal in transducing primary human hepatocytes in humanized mouse and NHP models in vivo.…”

Development of Optimized AAV Serotype Vectors for High-Efficiency Transduction at Further Reduced Doses

“…In our previously published studies, we identified the S663V+T492V doublemutant (DM) AAV3 vector as the most efficient vector in transducing human liver cancer cell lines. 25 Thus, the following sets of DM-ssAAV3-Fluc vectors were generated: DM-ssAAV3-WT-Fluc; DM-ssAAV3-LC1-Fluc and DM-ssAAV3-LC2-Fluc vectors was increased by up to 6-fold at such a low MOI.…”

Development of novel AAV serotype 6 based vectors with selective tropism for human cancer cells

“…To date only one member of the family, AAV serotype 3, has been found to have specificity for human hepatocytes and human liver cancer cells. AAV3 vectors have been shown to be able to discriminate between human liver cancer cells and mouse hepatocytes in xenograft animal model in vivo (25,30,40).…”

“…Genetic modifications of these regions increases the efficiency of transduction of AAV vectors in different tissues and cells types, most likely by preventing phosphorylation of these residues, and subsequent degradation of the vectors by host proteasome machinery. These studies have led to the development of several AAV vectors with higher activity in a variety of cell and animal models (17, 20, 22-26, 28, 30 Priority was given to the positions which are conserved among various serotypes, and have previously shown a better performance for other AAV vectors (17,22,23,28,30) (Table 1).…”