Each moment of each day countless millions of proteins circulate through the body, performing their tasks, recognizing and binding to their targets, all with minimal crosstalk between the countless other proteins circulating alongside them. It is no wonder then, that, when faced with problems such as detecting rare molecular targets within complex backgrounds, technology developers often employ biological molecules as the foundation with which to build sensors (“biosensors”) and develop processes (“bioassays”). But that is not to say that sensors and assays based on biorecognition are perfect. Bioassays, for example, are generally complex, time-consuming processes and thus they provide actionable information only after a significant time lag. Biosensors (such as blood glucose meters), in contrast, can provide real-time information, but those few that work in realistically complex sample matrices invariably rely on the specific chemical reactivity of their targets, greatly limiting the range of molecules they can detect. Motivated by these concerns, I have been exploring biosensors and bioassays that, unlike existing approaches, are simultaneously general and capable of rapidly returning answers.
The first strategy I have explored utilizes microfluidics to adapt existing, multi-step bioassays into a rapid-and-convenient point-of-care device. Specifically, taking advantage of the inherent automation potential of microfluidics I have shown it possible to automate and speed up an established, bench-top bioassay. Once fully automated, the assay can even be used to perform continuous measurements, tracking changes in the concentration of a target protein in a clinical sample stream in real-time.
The second strategy that I explored was the development of a single-step, reagentless biosensor platform termed electrochemical DNA-based (E-DNA) sensors. This broad class of electrochemical sensors utilizes binding-induced changes in the structure or dynamics of an electrode-bound biomolecule to reagenltessly and continuously report on the concentration of its target. Specifically, this class of sensors uses DNA either as a scaffold upon which to display a recognition element (scaffold-sensors), such as an antigen, or as the recognition element itself (aptamer-based sensors). Using Monte-Carlo simulations I have examined how changing the molecular weight of a recognition element affects the performance of scaffold sensors and used this information to develop a point-of-care serological assay to help rapidly diagnose syphilis. Using a similar Monte-Carlo model I likewise optimized the performance of aptamer-based sensors by examining how modifications around the aptamer’s binding pocket affect sensor performance. Together these studies are helping to bridge the gap between complex (but generalizable) bioassays and simple to use biosensors (such as blood glucose meters) which cannot readily be adapted to other targets.