| In the post-genome era, one of the most important topics of research is to edit or program genomic sequences and to generate desired phenotypes. Although virus-based strategies have long been developed to for efficient gene insertion, the random or semi-random integration can disrupt certain endogenous genes and cause unpredictable phenotypes. In contrast, targeted genome editing enables researchers to tailor genomic loci in a specific manner. Applications include studying gene functions, engineering microbes for industrial fermentation, improving traits in crop plants and livestock, treating human diseases, etc. This thesis describes my efforts on engineering transcription activator-like effector (TALE) nucleases (TALENs) as an efficient tool for targeted genome editing. Targeted genome engineering relies on the introduction of a site-specific double-strand break (DSB) in a pre-determined genomic locus by a rare-cutting DNA endonuclease. Subsequent repair of this DSB by non-homologous end joining or homologous recombination generates the desired genetic modifications such as gene disruption, gene insertion, gene correction, etc. For this purpose, I have constructed TALEN architecture by fusing the DNA binding domain of TALE and a FokI non-specific DNA cleavage domain. TALEs are isolated from the plant pathogenic bacteria from the genus Xanthomonas and their DNA binding domains are composed of a series of tandem repeats. Each repeat comprises 33--35 amino acids and recognizes a single nucleotide. The DNA recognition specificity is conferred by the highly variable amino acids at positions 12 and 13 (e.g., NI recognizes adenine, HD recognizes cytosine, NG recognizes thymine, and NN recognizes guanine and adenine). This simple code and independent DNA binding of the repeat units enable TALEs to bind to any custom-designed DNA sequence. The fusion of a FokI cleavage domain makes TALENs a new class of artificial DNA endonucleases which serve as a powerful tool for targeted genome editing. To monitor in vivo activities of TALEN, I have constructed various reporter systems constructed in yeast and human cells. To maximize the genome editing efficiency of TALENs, I have optimized the TALEN scaffold by the truncation of N- and C-termini of TALEs. Two TALEN scaffolds were identified with efficient activity in modifying both yeast and human genomes. To further improve TALEN platform, I have constructed a high-throughput screening system and identified the SunnyTALEN architecture through directed evolution. Compared with the existing TALEN platform, SunnyTALEN shows significantly increased genome editing efficacy in both yeast and human cells. To demonstrate the application of TALEN technology in human therapeutics, I have corrected the sickle cell disease mutation in patient-derived induced pluripotent stem cells. The corrected stem cells can serve as a regenerative medicine for the treatment of human genetic disorders. Lastly, I have created a novel single-chain TALEN architecture, which can be used to decrease the payload for efficient TALEN delivery. |
