Search for a command to run...
Biosensing platforms that are able to continuously measure concentrations of biomolecules are needed to study and control dynamic biochemical processes in areas such as biology, patient monitoring and bio based industrial processes. To be effective in real world use, these sensors must deliver strong analytical performance, have short time to results relative to the monitored process and maintain stability over long periods of time. Recently, sensors with single molecule resolution have emerged, that are able to track individual binding and unbinding events in real time. The discrete nature of the signals results in a measurement precision that is theoretically limited by counting statistics. In this thesis, we study how the access to single molecule resolution im-pacts the analytical properties of continuous biosensors. We use Biosensing by Particle Motion (BPM) as a model system, a biosensing platform that relies on optically tracking the motion of biofunctionalized particles on a biofunctionalized surface. By investigating how the single molecule resolution of BPM can be harnessed, we demonstrate its impact on sensor sensitivity, response time and long term performance. In Chapter 1, we introduce the field of continuous biosensing. We discuss the state of the art and strategies currently employed to achieve measurement precision, robustness and time resolution. Moreover, we discuss modeling strategies to investigate the analytical performance of biosensors. In Chapter 2, we present the development of a competition based BPM sensor for the detection of glycoalkaloids (GAs), small molecules that play a role in potato protein processing. We discuss the selection of GA specific anti-bodies via phage display and BPM. We demonstrate continuous monitoring over 20 hours, measurements in complex matrices and show how the analytical performance of BPM changes over time. In Chapter 3, we expand on the topic of long term signal changes and variability, by investigating intra and inter particle heterogeneities over long time spans. Individual particles in a BPM sensor for GA monitoring were studied over a period of 25 hours. By exposing the sensor to multiple series of concentrations, we uncover large differences in individual concentration dependent responses. Monte Carlo modeling shows that these differences are related to fluctuations in the number of accessible binder molecules. This insight was then used to clarify the origin of observed changes in the BPM signal over long time spans. In Chapter 4, we perform a study on the analytical performance limits of continuous biosensors with single molecule resolution. We develop a generic mathematical model that describes the switching behavior of reversible nanoswitches: molecular constructs that have distinct states controlled by affinity based interactions. We investigate via analytical and Monte Carlo modeling how molecular design parameters and sensing acquisition parameters control the sensor response, and we validate the observed scaling laws with a sandwich BPM sensor for ssDNA. We finally show that reliable, precise measurements in the low picomolar range are possible in measurement time scales of minutes. Chapter 5 summarizes the main findings in this thesis, describes opportunities for further research and discusses how single molecule resolution can be used to further improve the performance of continuous biosensors.