Steven Ochoa

Advisor: Valeria Milam

will defend a doctoral thesis entitled,

Efficient Exploration of Aptamer Sequence Space through Rationally Designed Screening Libraries and the CompELS Discovery Platform

On

Thursday, November 20th at 2 p.m.
EBB Krone 2209

Or

virtually via MS Teams: Link

Committee
            Dr. Valeria Milam - School of Materials Science and Engineering (advisor)
            Dr. Loren Williams - School of Chemistry and Biochemistry
            Dr. Philip Santangelo - School of Biomedical Engineering

      Dr. Vladimir Tsukruk - School of Materials Science and Engineering
            Dr. Meisha Shofner - School of Materials Science and Engineering 

Abstract

    Aptamers are short, single-stranded nucleic acid sequences that fold into complex three-dimensional structures capable of binding molecular targets with high affinity and specificity. Their stability, ease of synthesis, and chemical tunability make aptamers attractive alternatives to antibodies for diagnostics and therapeutics. Aptamer discovery, however, remains inefficient because sequences cannot yet be designed de novo and must instead be identified through high-throughput screening of large (~1014 unique sequences) random nucleic acid libraries. Traditionally, Systematic Evolution of Ligands by Exponential Enrichment (SELEX) has dominated as the aptamer screening platform, but it is a stochastic process that relies on enriching or amplifying the copy number of aptamer candidates each screening round. This amplification-dependent process can lead to low success rates and unintended sequence bias during selection. Moreover, conventional random sequence libraries often lack the structural diversity needed to yield functional target binders. This dissertation establishes a framework to address these limitations in both the design of the screening library and the aptamer screening platform.

     First, to address the inefficiency of conventional random screening libraries, three distinct DNA screening libraries were rationally designed to better sample more promising domains of sequence space. These three libraries consisted of the following sequence populations: a GT-rich library, based on the compositional analysis of validated aptamers; a pre-structured library, embedded with thermostable DNA tetraloop motifs to promote self-folding; and a standard random sequence library as a control. Computational modeling supported by experimental analyses confirmed these libraries exhibit distinct thermodynamic profiles and structural characteristics, validating their compositional design principles and their compatibility with high-throughput screening. 

      Second, an alternative aptamer discovery platform was developed that employs a novel enzymatic elution strategy to recover aptamer candidates and amplification-free screening strategy called Competition-Enhanced Ligand Selection (CompELS). CompELS was directly compared to SELEX in separate parallel screening runs using a model fluorescent protein target called mCherry. This comparative screening study comprehensively demonstrated that the CompELS platform more efficiently identified the same top aptamer candidates as SELEX. 

      This comparative selection study incorporated all three designer screening libraries and revealed that the empirically biased GT-rich library consistently outperformed both the random and tetraloop-embedded libraries. Next generation sequencing analysis confirmed that the most high-affinity aptamer sequences were GT-rich and primarily originated from the empirically biased library. Together, these findings validate that the rational library design approach coupled with a competition-based selection platform served as an effective strategy for both enriching aptamer screening libraries with functional sequence candidates and for accelerating aptamer discovery.