The Logic of Aptamer Isolation

Summary My goal in this collection of projects is to establish a unified framework to generate oligonucleotide-based receptors (aptamers) for small molecules based on concepts from organic and medicinal chemistry, with the focus on those targets that are otherwise difficult to isolate through standard protocols. In the process, I will generate a practical understanding of relationships between the complexity of the interactions in receptor space and the chemical forces (packing) that define this complexity, and build algorithms that can help us to access new properties, with possible applications beyond the aptamer isolation. These receptors will then enable rapid quantification, frequent monitoring, and multiplexing of challenging clinical analytes in complex mixtures. The experimental work in this program will specifically address the following two questions: 1. Can DNA receptors serve as general, off-the-shelf parts for biosensors and their arrays, interacting even with anions, small polar molecules, and, at the other extreme, hydrophobic molecules with complex shapes? 2. What are broadly applicable heuristics (suitable levels of abstraction) that could guide isolation of these receptors from the vast space of random oligonucleotides, when standard methods fail? The first question has direct, broad implications for our ability to use of the aptameric receptors in monitoring human health. It can be answered only empirically, by isolating receptors for difficult and disparate targets, ranging from phosphate, a small anion, to cyclosporine A, a poorly soluble, non-planar, immunosuppressant drug. Traditionally, the failures to isolate aptamers for such targets were explained by mismatches with functional groups on DNA; in contrast, I will pursue this as a problem of a combinatorial difficulty. All isolated receptors will be subsequently validated, within sensing devices, against gold-standard methods and in relevant matrices. The second question is then an approach to address a reformulated big challenge of finding macromolecules with complex functions in populations of random oligomers. This is a problem that has no direct optimal solutions, because of the combinatorial explosion of all possible species. Instead, we will use our understanding of general principles of ligand-receptor interactions to develop simple rules, which will be akin to individual steps in organic synthesis. When combined, they will allow us to implement multi-step procedures to isolate very rare functional oligonucleotide sequences. We have the option to keep adding new rules, while constantly validating the emerging heuristics empirically, through a series of practical implementations while pursuing the first question. Indeed, our principal practical goal, and the main criterion for success, will be to present multiple examples of tailored isolation procedures that would lead to identification of clinically useful DNA receptors with complexities that would otherwise bar their isolations in traditional single-step protocols.