548. Recombinant AAV3 Serotype Vectors Generated by Using AAV3 ITRs and AAV3 Rep Proteins Efficiently Transduce Human Liver Cancer Cell Lines In Vitro and Human Liver Tumors in a Murine Xenograft Model In Vivo

Abstract

Our initial observation of the selective tropism of AAV3 serotype vectors for human liver cancer cell lines and primary human hepatocytes (Mol Genet Metabol., 98: 289-299, 2009; Hum Gene Ther., 21: 1741-1747, 2010; Gene Ther., 19: 375-84, 2012), led to renewed interest in this serotype, since AAV3 vectors and their variants have recently proven to be extremely efficient in targeting human and non-human primate hepatocytes in vitro as well as in vivo (Nature, 506: 382-386, 2014; Hum Gene Ther., 25: 1023-1034, 2014; Mol Ther, 23: 1867-1876, 2015; Mol Ther., 23: 1877-1887, 2015). Our previous studies also documented that the combined use of AAV3 ITRs, AAV3 Rep proteins, and AAV3 capsids led to the production of recombinant AAV3 vectors with higher titers than those produced using AAV2 ITRs, AAV2 Rep proteins, and AAV3 capsids. We also observed that the transduction efficiency of Rep3/ITR3 AAV3 vectors was ~4-fold higher than that of Rep2/ITR2 AAV3 vectors in a human hepatocellular carcinoma (HCC) cell line, Huh7, under identical conditions. In the present studies, we extended these observations to include capsid-optimized AAV3 vectors in which two surface-exposed residues (Serine 663 and Threonine 492) were mutated to generate a S663V+T492V double-mutant vector to examine whether the transduction efficiency of the Rep3/ITR3-S663V+T492V-AAV3 vectors could be further augmented. To this end, two human HCC cell lines, Huh7 and LH86, were transduced with WT-AAV3 and S663V+T492V-AAV3 vectors generated with ITR2/Rep2 and ITR3/Rep3, respectively, under identical conditions. Consistent with our previously published studies, the transduction efficiency of the S663V+T492V-AAV3 vectors was ~10-fold higher than its WT counterpart, and interestingly, there was a further 2-fold increase in the transduction efficiency of both vectors generated with ITR3/Rep3, compared with those generated with ITR2/Rep2 (Fig. 1AFig. 1A). We next evaluated the transduction efficiency of these vectors in a murine xenograft model bearing human liver tumors. AAV3-EGFP-Neo vectors generated with either ITR2/Rep2, or with ITR3/Rep3, were injected intra-tumorally at a dose of 1×1011 vgs/tumor. Forty-eight hrs post-vector administrations, transgene expression was determined in each tumor by Western blot assays. These results indicated that AAV3 vectors generated with ITR3/Rep3 transduced human liver tumors in vivo ~2-fold more efficiently than those generated with ITR2/Rep2 (Fig. 1BFig. 1B). It is anticipated that the combined use of ITR3/Rep3, and S663V+T492V-AAV3 capsids will further increase the transduction efficiency of these vectors. Taken together, our data suggest the transduction efficiency of AAV3 vectors can be significantly improved both by using homologous Rep proteins and ITRs as well as by capsid-optimization. The use of these modified AAV3 vectors should prove useful in liver-directed human gene therapy.View Large Image | Download PowerPoint Slide