DNA polymerases are central biocatalysts in the life cycle of all organisms. Structural transitions of this enzyme during the incorporation of deoxynucleoside triphosphate (dNTP) substrates govern the accurate replication of DNA. Biophysical tools have emerged as useful techniques for understanding these structural transitions, and are continually being improved. Electronic-based techniques, in particular, provide many advantages over other biophysical techniques, including long duration monitoring and enhanced time resolution. For example, recently developed single-walled carbon nanotube field-effect transistors that measure DNA polymerase I Klenow Fragment (KF) activity serve as a valuable method for label-free detection of biocatalysis. In my Ph.D. work, I have sought to develop this technique further to better understand enzyme activity and take advantage of our observations for biotechnological applications.

Incorporation of dNTP analogs with KF-functionalized nanocircuits challenged the molecular recognition of these substrates by KF. For some dNTP analogs, an alternative motion of KF during dNTP incorporation was observed and may be valuable in the discrimination of DNA bases in a template. The nanocircuits also accurately identified DNA template lengths when such templates were composed of highly repetitive sequences. Accumulation of several data sets further enabled precise measurement of DNA template lengths. Initial experiments of dNTP incorporation with a KF nanocircuit measured at high bandwidths demonstrated that the accurate measurement of DNA template lengths required the lower time resolution. The higher bandwidth experiments revealed previously unobserved KF motions that lasted for short durations, making this technique unsuitable for counting DNA bases. However, improvement to KF nanocircuits measured at high bandwidth can provide a powerful tool for understanding transient KF dynamics.

To better understand and improve KF nanocircuits, a variety of KF variants were designed and generated. Many of these variants exhibited polymerase activity at the single-molecule level, and are in various stages of progression for their measurement with polymerase-functionalized nanocircuits. The experiments necessary to determine their viability in improving and understanding the electronic signals generated by KF were outlined for future Weiss lab members.

Membrane proteins are a class of biomolecules that also benefit from the improved understanding of their molecular mechanisms. In vitro evolution of high affinity binding partners with stable, rigid phage-displayed protein libraries were attempted for the Helicobacter pylori inner membrane protein Ure-I, but were unsuccessful due to Ure-I’s rapid degradation. An alternative method for in vitro evolution for high affinity binding partners to the Shigella dysenteriae outer membrane protein ShuA was optimized to maintain ShuA in a native membrane environment.

Molecular motions of proteins and their flexibility induce conformational states required for enzyme catalysis, signal transduction, and protein-protein interactions. However, the mechanisms for protein transitions between conformational states are often poorly understood, especially in the millisecond to microsecond range where conventional optical techniques and computational modeling are most limited. To investigate the microsecond dynamics of enzymes, single-walled carbon nanotube – field-effect transistors (SWCNT-FETs) were used as single-molecule biosensors. The SWCNT-FETs have sufficient sensitivity and bandwidth to monitor the conformational motions and processivity of an individual cAMP-dependent protein kinase A (PKA) molecule. Protein attachment is accomplished by functionalizing a SWCNT-FET device with a single protein and measuring the conductance versus time through the device as it is submerged in an electrolyte solution. PKA-functionalized nanocircuits elucidated the time PKA spent traveling between three different conformational motions upon binding of its two substrates.

To generalize this approach for the study of a wide variety of proteins at the single-molecule level, this dissertation investigates the bioconjugation process to determine and isolate the key parameters required for functionalizing a SWCNT with a single protein. Further analysis into the parameters implicated in the thiol-maleimide bioconjugation step of the SWCNT-FET device fabrication proved an exhausting effort. We developed improved linker molecules, anthracene maleimide, for attachment of three different proteins to the SWCNT and a new derivative of maleimide providing a solution to overcome side reactions associated with the reducing agent, tris(2-carboxyethyl)phosphine (TCEP). Protein purity and presence of detergent proved to play key roles in preventing nanocircuit surface fouling. An alternative bioconjugation method, utilizing an azide-functionalized hen egg-white lysozyme (HEWL) to tether the enzyme to the SWCNT through an alkyne linker molecule was developed. The work described here explores the current SWCNT attachment method as well as a new strategy to apply SWCNT-FET devices to study novel proteins. The results demonstrate SWCNT-FET as a sensitive technique for studying conformational dynamics of biological molecules in the microsecond range.

The sensitive detection of cancer biomarkers in urine could revolutionize cancer diagnostics and treatment. Such detectors must be inexpensive, easy to interpret, and sensitive. M13 bacteriophage provides one such highly inexpensive, and easily functionalizable platform for the development of such detectors. My work focuses on genetically and chemically modifying M13 viruses towards biological recognition of prostate-specific membrane antigen (PSMA), a prostate cancer biomarker. Elevated PSMA levels have been observed in prostate cancer patients’ urine, and also found on the surface of prostate cancer cells. For early diagnosis of prostate cancer, I have developed the chemistry necessary for the detection of clinically relevant concentrations of PSMA (<0.25 nM) in synthetic urine for direct application to patients’ samples. High sensitivity to PSMA results from the synergistic binding by two different ligands to PSMA on the same phage particle. The primary ligand is genetically encoded, and the secondary recognition ligand is chemically synthesized to electrostatically wrap around the phage surface. The concept of ‘phage wrapping’ utilizes the electrostatic attraction between the negatively charged phage surface and the positively charged oligolysine peptide ‘wrapper.’ Furthermore, the dual ligands result in bidentate binding with high copy, dense ligand display for enhanced PSMA detection through a chelate-based, avidity effect. Such dual modified viruses integrated into electrically conductive poly(3,4-ethylenedioxythiophene (PEDOT) films, act as the bioaffinity matrix for the electrochemical sensing of PSMA. The capture and binding of PSMA results in a change in the virus-PEDOT film’s resistance, which is also proportional to the PSMA concentration. Biosensing with films provides a 100 pM limit of detection for PSMA in synthetic urine without requiring enzymatic or other amplification. I further utilized this concept to reduce the non-specific adhesion between viruses and prostate cancer cells through wrapping the phage surface with polyethylene glycol. Additionally, I developed orthogonal wrapping techniques to selectively attach ligands and polyethylene glycol on the phage surface in desired ratios and architectures for the capture of PSMA-positive prostate cancer cells for metastasis detection. The combination of orthogonal chemically-modified viruses and biosensors holds the potential for the development of a real-time and reagent-free point-of-care device for prostate cancer detection.

The caveolins are a family of human membrane proteins implicated in cancer and other diseases. In Chapter 1, I briefly explain some of the basic biochemistry of the caveolins, and why they are not only a promising target for research and therapeutics but also a particularly challenging target. The novel peptide WL47, the development of which is described in later chapters, is introduced as a caveolin-1 ligand. WL47 will allow new avenues of investigation for this challenging target.

In Chapter 2, I review, compare, and contrast some of the myriad techniques available to quantify binding affinity for high affinity binding interactions such as the interaction between caveolin-1 and WL47. As I discovered in my own attempts to measure the dissociation constant for this interaction, high affinity is a boon for future use of a ligand probe, but it can be challenging to quantify initially. Because a review of techniques specifically suited to high affinity measurement was not available at the time I needed it, I have written this chapter to fill this gap in the literature.

In Chapter 3, I discuss the development of peptide ligand WL47 through an iterative process of library design, synthesis, screening, and data analysis. High affinity binding, target selectivity, and activity as an inducer of deoligomerization of oligomeric caveolin-1 is presented. Beyond providing a novel tool for probing the function of caveolin-1 oligomerization, this chapter is also a roadmap for the creation of future peptide ligands through this generalizable, iterative process

Chapter 4 details the progress that has been made toward understanding the importance of WL47 dimerization through a disulfide bridge. This chapter also covers initial efforts towards designing a redox-stable version of WL47 that will remain functional in conditions like those found in the interior of a live cell.

Chapter 5 concludes the dissertation with a discussion of the current state of the caveolin ligand project, including detailed descriptions of the experiments that will be necessary for future researchers to bring this project to its conclusion.

Numerous factors, such as genetics, medical history, environment, and lifestyle contribute to patient responses to disease conditions. Approaches in precision medicine focus on examining disease in the context of overall patient health, so that prevention and effectively tailored treatments result in improved outcomes. This thesis describes investigations of disease characterization and tool development geared towards uses in precision medicine. The COVID-19 pandemic has opened new avenues of investigation for the scientific community. The most important of which is understanding why patients have such diverse responses to infection. Here, we develop an M13-bacteriophage display-based screening platform to understand how antibody responses play a role in foreshadowing and driving disease severity in a subset of patients. We further characterize disease response in the context of co-morbidities and viral infection history and find that both factors can negatively impact disease severity in patients with a specific antibody response. Additionally, sensitivity of these patients to rising levels of cytokines suggest specific routes of pre-emptive treatments that may be beneficial. We, additionally, develop a platform for rapid (<5 mins) point-of-care detection, which can be used to assist in multi-analyte testing or in urgent triaging. Precision medicine is also important in addressing changes in disease response over time. Real-time automated monitoring of disease markers can help improve disease management and reduce adverse side effects. Using techniques in protein engineering, we harness the specific binding interactions of native proteins to develop biosensors for managing highly dynamic disease conditions. Firstly, we engineer the human insulin receptor towards developing an optical insulin biosensor to improve automated type-1-diabetes disease management. The investigation focuses on the development of soluble, compact insulin-binding variants. We further test various assay designs for detecting insulin-dependent optical signaling. Secondly, we engineer a photo-stable version of a previously developed FRET-based calcium sensor, Twitch 2B. The new version of this sensor demonstrates improved stability and performance within an optical-fiber-based implantable device. Such a sensor can be applied to early diagnosis of hypocalcemia during blood transfusions. Taken together, these investigations explore a wide range of protein engineering techniques across multiple expression platforms. By harnessing native protein-protein interactions, we build prognostic and disease management tools that can contribute to advancing the field of precision medicine. Additionally, we develop optical detection platforms and rapid antibody screening methods that can be used to dynamically monitor and explore crucial molecular interactions in disease.

Filamentous bacteriophage have proven to be a powerful biotechnology tool. A direct genotype-to-phenotype relationship makes phage easily genetically modified to display peptide or proteins on their solvent-facing surface. Thus, phage can function as pseudo-antibodies with similar binding affinity and target specificity. Furthermore, rational design of peptide or protein libraries allows for discovery of new target-binders via biopanning. Additionally, phage are increasingly gaining traction as a biomaterial due to an inherent monodispersity and ease of chemical modification (Chapter 1). In this dissertation, I will present how I have used phage display to validate a new immunoblot assay technique (Chapter 3), to identify a COVID-19 antibody that correlates with disease severity and investigate the origin of this antibody (Chapter 4, Chapter 5), and to detect valuable biomarkers with novel biosensing platforms (Chapter 6, Chapter 7, Chapter 8). Additionally, I have contributed new knowledge to the field of phage display by developing a novel method to directly select for noncompetitive binding peptides and antibody fragments for the more sensitive and specific sandwich ELISA (Chapter 2).