|Ph.D Student||Kaufmann Beate|
|Subject||Deciphering Triplex Formation using a Synthetic|
Biology-Inspired and Deep-Sequencing Approach
|Department||Department of Biotechnology and Food Engineering||Supervisor||Professor Roee Amit|
|Full Thesis text|
Nature provides a tremendously rich toolbox of dynamic nucleic acid structures that are wide-spread in cells and affect multiple biological processes. With technological advances in deep-sequencing and DNA synthesis, non-canonical structures gained renewed scientific as well as biotechnological interest. One particularly intriguing form of such structures is the formation of triplexes. It involves three nucleic acid strands and a mix of Watson&Crick as well as Hoogsteen hydrogen bonds. By applying low-throughput approaches in vitro, progress in the field has been made, but until today the underlying rules for triplex formation remain debated and evidence for such triplexes in vivo (e.g. in form of RNA*DNA-DNA triplexes) is circumstantial. In this PhD project, I applied a combined strategy of synthetic biology-inspired circuit designs in bacterial and mammalian cells and the development of multiple deep-sequencing and DNA synthesis-based platforms to systematically refine the triplex code. I started with the design of synthetic long non-coding RNAs (slncRNAs) from the bottom-up and tested them in an enhancer-based circuit in bacteria and a gene-activation platform in mammalian cells. In both systems a non-specific and inconsistent up-regulatory effect of a reporter gene in presence of slncRNA molecules was observed, but yielded overall inconclusive results. The challenges I faced using the synthetic biology designs, prompted me to build several next-generation sequencing platforms to study triplex formation in vitro and in cells. To do so, I designed large libraries of short, single-stranded oligos containing putative triplex-forming sequences. Following transfection of the libraries into cells, or incubation with double-stranded DNA in vitro, a subset of oligos binds the double-stranded DNA via triplex formation, is selectively enriched (Triplex-Seq), or ligated to genomic DNA in close proximity (Triloci-Seq), and subsequently analyzed using next-generation sequencing. By applying the Triplex-Seq approach in vitro and in cells, I identified that triplexes are preferably formed in neutral compared to acidic pH, and G-rich oligos as well as G-rich double-stranded DNA form stable and highly specific triplexes. Furthermore, a minimal length of 7-10 nucleotides was shown to be sufficient for stable triplex formation. To identify genomic target sites to which the oligos were bound, I employed Triloci-Seq. Using this approach, I found putative genomic, GAArich motifs that exhibited a 3 nt periodicity in the sequence reads and have been predicted to form triplexes with GAArich and TTCrich oligos that were used in this approach. These results, together with the complementary Triplex-Seq data, refine the sequence context required for triplex formation thus establishing a powerful tool to further explore unusual nucleic acid interactions. I believe that my results demonstrate the power of deep-sequencing and synthetic biology platforms to explore triplex formation and build upon a growing interest in using DNA structures for bio- and nanotechnological applications.