819. Hyperactive Transposase Mutants of the Sleeping Beauty Transposon

Abstract

The Sleeping Beauty (SB) transposon system mediates stable genomic integration by a cut-and-paste transposition process and has shown tremendous success as a therapeutic vehicle for in vivo gene delivery. Nevertheless, the overall rate of transposition with SB is still insufficient for many clinical applications. Herein, we generated 141 missense mutants for the SB transposase and screened each for altered gene transfer activity in human cells using a genetic assay. We show that despite equivalent steady-state levels of wild-type (WT) and mutant transposases in transfected cells, many mutations introduced into the N-terminal DNA-binding domain (n=10) and C-terminal catalytic core domain (n=6) resulted in 2- to 4-fold higher transpositional efficiencies compared to WT. In order to better understand the transposition reaction for potential future development, we compared the functional activities of these hyperactive recombinases to that of the WT transposase using a combination of PCR and EMSAs. Results of these studies identified a number of pathways that contribute to improved transposase activity, including an altered affinity of the transposase for transposon ends, transposase conformational changes, enhanced donor cleavage activity, and increased rates of strand transfer. Importantly, many of these hyperactive SB (HSB) mutations functioned synergistically and, when combined with a hyperactive transposon, elevated transposition by 14-fold compared to the first-generation transposon system. Moreover, although WT and mutant transposases were negatively regulated at high enzyme concentrations, the HSB variants supported ~10-fold higher transpositional activity at all experimental doses, suggesting that lower doses could be administered to further improve the safety profile for SB in vivo. In addition, the use of HSB instead of the WT transposase dramatically improved the integration frequency of larger-sized elements, including ones greater than 14 kb in length. These HSB mutants should greatly extend the carrying capacity for SB, thus overcoming one of this system's major limitations. Finally, we studied the long-term in vivo persistence of two transposon reporters, β-galactosidase and human coagulation factor IX (hFIX), in the livers of adult mice following a single administration of WT- or hyperactive transposase-expressing plasmids. Results of these analyses demonstrated that these mutated enzymes could also greatly enhance SB's gene transfer capabilities in vivo, supporting stable integration in an estimated 57% of transfected mouse hepatocytes, compared to ~8% for the WT. This level of transposition was sufficient to maintain ~7-fold higher serum hFIX levels (955 ng/ml ± 241 ng/ml) than achieved in SB-expressing animals (140 ng/ml ± 56 ng/ml) for a period of 4 months (length of study). This level of hFIX obtained with the HSB mutant is ~19% of normal human levels and is well within a curative range capable of converting a severely affected hemophilia B patient to one with a much milder phenotype. Therefore, our work demonstrates the potential for improved in vivo gene delivery through directed transposase evolution, and provides a number of important insights into Sleeping Beauty transposon biology that could be exploited for future development.