A genome-wide approach to aptamer discovery

Abstract

Aptamers are short, single-stranded DNA or RNA oligonucleotides that bind with high specificity and affinity to a target molecule due to their unique, sequence-dependent secondary structure. They have found applications as therapeutic agents, in targeted drug delivery, and as affinity reagents in biological and chemical sensing, separation and purification.

Here, Nucleolin serves as a model target and the G-quadruplex as a model motif, albeit with potential for discovery of an anti-Nucleolin aptamer. The Genome-Wide Approach to Aptamer Discovery could be extended to identify aptamers of any computationally detectable secondary structure against any chromatin-associating protein in any genome.

This dissertation describes A genome-wide approach to aptamer discovery, where the entire human genome is explored for discovery of DNA G-quadruplex aptamers against the protein Nucleolin. To identify potential aptamers, we used a near single-nucleotide resolution Chromatin Immunoprecipitation technique to map Nucleolin’s binding sites across the entire genome of a human breast cancer cell line. We computationally searched these binding sites for sequences with G-quadruplex-forming potential. The dissociation constants (Kd’s) of 12 potential G-quadruplex-forming sequences that were enriched in the Chromatin Immunoprecipitation results were measured against recombinant Nucleolin, with Kd’s ranging from ~25-2000 nM. We compared our dissociations constants for the genome-derived sequences to the anti-Nucleolin aptamer, AS1411, that failed at Phase II clinical trials. We also measured the Kd’s of several genomic G-quadruplexes whose Kd’s appear in the literature, for which our values are in good agreement.

Aptamer discovery is currently limited by its conventional discovery process, the Systematic Evolution of Ligands by Exponential Enrichment (SELEX), where a large combinatorial library of “random” sequences is screened through repeated selection rounds that reduce the pool of sequences until a single aptameric sequence against the desired target is identified. SELEX underrepresents a specific secondary structure, the G-quadruplex, at the library design and selection steps. To overcome this deficit in SELEX, our group has utilized a Genome-Inspired Reverse Selection Approach to Aptamer Discovery that makes use of the abundance of G-quadruplex motifs in the human genome. G-quadruplex-forming sequences are picked from promoter regions, typically oncogenes, are incubated with nuclear or cytoplasmic extracts from human cancer cell lines. Proteins are captured from the extracts and screened for selective binding. Selective binders are identified, and their interaction with the G-quadruplex aptamer is interrogated in situ. While this process overcomes the limitations of bias against G-quadruplexes in SELEX, the reverse selection process precludes selection of an aptamer to a predesignated target.

One outcome of the investigations of the Genome-Inspired Reverse Selection Approach was the discovery of the affinity binding of the protein Nucleolin in vitro and in situ with all tested oncogene promoter G-quadruplexes. Cell-surface Nucleolin is a biomarker and therapeutic target for human cancers, as it is overexpressed on malignant cell surfaces but absent in most nonmalignant tissues.