Towards the In Vitro Selection of Therapeutic Aptamers

Aptamers are a class of nucleic acid molecules that mimic antibodies by folding, in a sequence specific manner, into 3D structures that can bind a target. Vast random libraries, routinely around 1015 nucleic acid sequences undergo iterative cycles of binding their target, stringent washing to remove weak and non-binders, elution, and amplification to become enriched in sequences with high binding affinity to the target. Aptamers have many benefits over other drug archetypes, such as their ease of discovery, reproducibility due to their being chemically synthesizable, reversible folding and thermal stability, and low cost relative to antibodies. This makes them attractive drug candidates, but aptamers have been largely relegated to diagnostic settings due to the poor biological stability of natural nucleic acids.The field has long been dominated by RNA and DNA aptamers, which boast an ease of discovery born of their origins in nature. However, recent advancements in polymerase engineering have enabled the discovery of polymerases with the ability to transcribe DNA into xeno nucleic acids (XNAs), synthetic nucleic acids with backbones unrelated to DNA or RNA. Although these engineered variants function with reduced activity when compared to their natural substrate, the ability to transfer information from one genetic polymer to another has had enabled the selection of XNA aptamers, among other technologies. One such XNA, threose nucleic acid or TNA, is unique in that it can form stable antiparallel duplexes with itself or DNA, while being completely recalcitrant to nuclease digestion. This make TNA well suited for therapeutic use, as it will be more stable in the body. In this dissertation, we first developed a method to allow the selection of TNA aptamers covalently linked to their encoding DNA using an engineered TNA polymerase. Next, aptamer chemical diversity was expanded with the addition of base-modified TNA nucleotides. Finally, the scalability of TNA aptamer production was explored through TNA ligation with natural ligases. Chapter 1 is a review of literature pertaining to natural nucleic acid and xeno nucleic acid aptamers, outlining what has been achieved and what remains to be overcome. Specifically, we focus on TNA aptamer selection, and describe a foundational approach to TNA aptamer selection, DNA display, which is utilized in Chapters 2-3. This new strategy borrows from previous work on mRNA display to overcome one of the biggest stumbling blocks to XNA aptamer selection: library regeneration after selection. The resultant TNA aptamers are characterized to reveal picomolar binding affinity, a magnitude previously not known to be achievable for TNA aptamers. The next chapter details the discovery of a TNA aptamer to HIV RT with picomolar affinity. While TNA’s biological stability was previously known, this TNA aptamer demonstrates high thermal stability of TNA molecules that is expected from DNA and RNA. The use of an engineered TNA polymerase and a display strategy that provides a powerful genotype−phenotype linkage allowed the discovery of aptamers ranging in affinity from ~0.4-4 nM. The TNA aptamers remain intact and capable of binding in the presence of nucleases. The combined properties of biological stability, high binding affinity, and thermal stability make TNA aptamers a powerful system for the development of diagnostic and therapeutic agents. Chapter 3 describes work to inject greater chemical diversity into aptamer selections by incorporating amino-acid-like side chains onto standard uracil bases. This strategy parallels work done by Mayer, Liu, and others to expand chemical diversity in aptamers. The base and backbone modified aptamers are referred to as ‘threomers’, which bind with lower KDs and slower off rates than their unmodified TNA counterparts. Kinetic measurements reveal that side chain modifications are critical for generating threomers with slow off-rate binding kinetics. These findings expand the chemical space of evolvable non-natural genetic systems to include functional groups that enhance protein target binding by mimicking the structural properties of traditional antibodies. A detailed investigation of the impact benzyl and indole displaying bases have on selection outcomes are reported here. Chapter 4 describes initial work to increase the scale of synthesizable TNA through a combination of solid phase TNA synthesis and ligation with natural ligases. A thorough exploration of ligases and reaction conditions with TNA-TNA and DNA-TNA ligations led to the identification of T3 DNA ligase as the most active on TNA substrates. This knowledge could serve as a starting point for future directed evolution of a TNA ligase