Using capillary electrophoresis for high-throughput molecular screening.

摘要

Advances in the development of DNA-encoded chemical libraries have necessitated rapid selection strategies to aid in molecular discovery. Encoded chemical libraries are composed of thousands of unique (potential) drug molecules, each encoded with a corresponding unique DNA sequence, allowing for the simultaneous screening of ligands against an identified biomolecular target. The basic principles driving DNA-encoded drug selection are mimicked by aptamer selection principles, which involve the partitioning of target-bound DNA from unbound DNA and the subsequent amplification of aptamers by PCR (polymerase chain reaction). Several screening methodologies have been proposed, including systematic evolution of ligands by exponential enrichment (SELEX) and nonequilibrium capillary electrophoresis of equilibrium mixtures (NECEEM). However, these stand-alone techniques require multiple rounds of selection and produce a limited number of each unique aptamer. SELEX has the added disadvantage of heavy resource consumption. Thus, it still remains to develop a fast, efficient method for high-throughput screening of encoded chemical libraries.;Ultimately, the goal of this work is to address the need for novel screening methods by developing a selection platform utilizing tools from chemistry, physics, and cancer biology. As an initial step towards realizing this goal, methods were developed for the screening of oligonucleotide libraries. The present work demonstrates significant improvements in capillary electrophoresis (CE) separation, detection, and collection techniques for selection of potential therapeutic agents from oligonucleotide libraries and for compatibility with next generation ("lab-on-bead") sequencing technologies. Oligonucleotide libraries containing many thousands of potential aptamers were mixed with protein targets of interest and a single NECEEM experiment was used to isolate only those ligands that strongly bound the target. Automated CE fraction collection was subsequently used to isolate binding ligands and to obtain a significant enough concentration of binding ligands to pass along to the next generation sequencing (NGS) stage. In a proof-of-principle experiment, the successful coupling of CE and NGS methods was demonstrated, with CE selection resulting in library enrichment of 40x. A subsequent study involved selection of a known thrombin protein binding aptamer from a library of 32,768 oligonucleotides screened against thrombin protein followed by NGS analysis. Methods optimized for aptamer selection were then applied for the selection of encoded small molecules from a library of 13,824 DNA-encoded macrocycles screened against Src kinase, of which 15 macrocycles were enriched using CE-based selection.;To couple CE preselection methods with subsequent lab-on-bead drug discovery methods requires the production of nanobeads functionalized with multiple copies of CE-preselected oligonucleotides via a process called em-PCR (emulsion-PCR). To better assess this process, CE methods were also developed to characterize and distinguish between functionalized polymer nanoparticles of different sizes and different surface modifications, and to separate polymeric nanoparticles with and without DNA modification. The extension of these CE-based nanoparticle characterization methods to the separation of protein-bound and unbound aptamer-conjugated nanoparticles, and to a study of nanoparticles as buffer additives to improve CE separations was also undertaken, thus illustrating the versatility of CE as a tool for both analysis and materials characterization.