UC Merced

A central question in DNA biosensors is how the surface structure impact the molecular recognition, and ultimately the figures of merit of device function, such as sensitivity and reproducibility. Here is a quote from seminal work by Herne and Tarlov (1997): “surprisingly little is known about the surface structures of bound probes and the impact of the surface on hybridization reaction.” Two decades later, this remark is still valid, notwithstanding the exponentially increasing number of publications in DNA-based biosensors. The probe surface density is extensively investigated and used to the optimize the sensing properties. These studies must rely on the assumption that the probe distribution is homogeneous and the targets can access the immobilized probes with identical binding energies, and thus that the binding affinity and kinetics can be described by the ensemble average quantities. While such studies have revealed general trends of target recognition, these observables are far from adequate descriptors of the surface structures or complex surface interactions, especially in light of growing evidence that a realistic biosensor surface may be highly heterogeneous. In this dissertation, I performed single-molecule atomic force microscope (SM-AFM) imaging to resolve the closely-spaced individual probes as well as hybridization event on a functioning DNA biosensor surface. Chapter 1 provides the background and challenges in characterizing the surface structures of DNA biosensors, as well as an introduction to SM-AFM imaging. In Chapter 2, I applied new spatial statistical tools to characterize the spatial patterns of probe distribution and correlated these patterns to interfacial molecular recognition, which provided new insights on molecular crowding. In Chapter 3, I modeled the surface hybridization kinetics based on spatial statistical information extracted from the AFM database. This raises the intriguing prospect that the spatial patterns of biosensor surface can be rationally tailored to improve the performance. In Chapter 4, AFM is used to generate single molecule nanoarrays with well-controlled chemical and morphological heterogeneities, which serve as a model system for studying the DNA-surface interactions. And finally, in Chapter 5, I summarized the main findings, and elaborated the future research topic on surface patterning of probes with DNA nanostructures.