Quantifying molecular interactions is critical to the understanding of many biological processes and drug screening. To date, various detection techniques have been developed to determine the binding kinetics. However, because most of the mainstream detection technologies detect signals that scale with the mass of ligands bond to the sensor surface, it is still challenging to quantify the binding kinetics of small molecules. To address this problem, two different detection technologies, charge-sensitive optical detection (CSOD) and critical angle reflection (CAR), are developed for label-free detection of molecular interactions with the ability to detect a wide range of molecules including small molecules. In particular, CSOD technique detects the charge rather than the mass of a molecule with an optical fiber. However, the effective charge of a molecule decreases with the buffer ionic strength. For this reason, the previous CSOD works with diluted buffers, which could affect the measured molecular binding kinetics. Here a technique capable of detecting molecular binding kinetics in normal ionic strength buffers is presented. An H-shaped sample well was developed to overcome this problem. With this new design, the binding kinetics between G-protein-coupled receptors (GPCRs) and their small molecule ligands were measured in normal buffer. To further improve the signal-to-noise ratio of CSOD and move it toward high-throughput detection, CSOD was implemented with a quadrant-cell detector to achieve detection in higher frequency range and decrease low-frequency noise.This improved CSOD technique is capable for direct quantification of binding kinetics of phage-displayed peptides to their target protein using the whole phages. CAR imaging can be performed on surface plasmon resonance (SPR) imaging setups. It was shown that CAR is capable of measuring molecular interactions including proteins, nucleic acids and cell-based detections. In addition, it was shown that CAR can detect small molecule bindings and intracellular signals beyond SPR sensing limit. CAR exhibits several distinct characteristics over SPR, including tunable sensitivity and dynamic range, deeper vertical sensing range, and fluorescence compatibility. CAR is anticipated to have the ability to expand SPR capability in small molecule detection, whole cell-based detection, simultaneous fluorescence imaging, and broader conjugation chemistry.

My research focuses on studying the interaction between spatiotemporally encoded electric field (EF) and living cells and biomolecules. In this thesis, I report two projects that I have been working on to address these questions. My first project studies the EF modulation of the extracellular-signal-regulated kinase (ERK) pathway. I demonstrated modulation of ERK activities using alternative current (AC) EFs in a new frequency range applied through high-k dielectric passivated microelectrodes with single-cell resolution without electrochemical process induced by the EF stimulation. Further experiments pinpointed a mechanism of phosphorylation site of epidermal growth factor (EGF) receptor to activate the EGFR-ERK pathway that is independent of EGF. AC EFs provide a new strategy to precisely control the dynamics of ERK activation, which may serve as a powerful platform for control of cell behaviors with implications in wide range of biomedical applications. In the second project, I used solid-state nanopore system as the base platform for single molecule experiments, and developed a scalable bottom-up process to construct planar nanopore devices with self-aligned transverse tunneling junctions, all embedded on a nanofluidic chip, based on feedback-controlled reversible electrochemical deposition in a confined nanoscale space. I demonstrated the first simultaneous detection of translocating DNA molecules from both the ionic channel and the tunneling junction with very high yield. Meanwhile, the signal amplitudes from the tunneling junction are unexpectedly high, indicating that these signals are probably dominated by transient currents associated with the fast motion of charged molecules between the transverse electrodes. This new platform provides the flexibility and reproducibility required to study quantum-tunneling-based DNA detection and sequencing. In summary, I have developed two platforms that engineer heterogenous EF at different length scales to modulate live cells and single biomolecules. My results suggest that the charges and dipoles of biomolecules can be electrostatically manipulated to regulate physiological responses and to push detection resolution to single molecule level. Nevertheless, there are still many interesting questions remain, such as the molecular mechanism of EF-protein interaction and tunneling signal extraction. These will be the topics for future investigations.

Antibiotic resistance is a very important issue that threatens mankind. As bacteria

are becoming resistant to multiple antibiotics, many common antibiotics will soon

become ineective. The ineciency of current methods for diagnostics is an important

cause of antibiotic resistance, since due to their relative slowness, treatment plans

are often based on physician's experience rather than on test results, having a high

chance of being inaccurate or not optimal. This leads to a need of faster, pointof-

care (POC) methods, which can provide results in a few hours. Motivated by

recent advances on computer vision methods, three projects have been developed

for bacteria identication and antibiotic susceptibility tests (AST), with the goal of

speeding up the diagnostics process. The rst two projects focus on obtaining features

from optical microscopy such as bacteria shape and motion patterns to distinguish

active and inactive cells. The results show their potential as novel methods for AST,

being able to obtain results within a window of 30 min to 3 hours, a much faster

time frame than the gold standard approach based on cell culture, which takes at

least half a day to be completed. The last project focus on the identication task,

combining large volume light scattering microscopy (LVM) and deep learning to

distinguish bacteria from urine particles. The developed setup is suitable for pointof-

care applications, as a large volume can be viewed at a time, avoiding the need

for cell culturing or enrichment. This is a signicant gain compared to cell culturing

methods. The accuracy performance of the deep learning system is higher than chance

and outperforms a traditional machine learning system by up to 20%.

Cytometry is a method used to measure and collect the physical and chemical characteristics of a population of cells. In modern medical settings, the trend of precision and personalized medicines has imposed a need for rapid point-of-care diagnostic technologies. A rapid cytometric method, which aims at detecting and analyzing cells in direct patient samples, is therefore desirable. This dissertation presents the development of light-scattering-based imaging methods for detecting and analyzing cells and applies the technology in four applications. The first application is tracking phenotypic features of single particles, thereby differentiating bacterial cells from non-living particles in a label-free manner. The second application is a culture-free antimicrobial susceptibility test that rapidly tracks multiple, antimicrobial-induced phenotypic changes of bacterial cells with results obtained within 30 – 90 minutes. The third application is rapid antimicrobial susceptibility testing (AST) of bacterial cell growth directly in-patient urine samples, without a pre-culture step, within 90 min. This technology demonstrated rapid (90 min) detection of Escherichia coli in 24 clinical urine samples with 100% sensitivity and 83% specificity and rapid (90 min) AST in 12 urine samples with 87.5% categorical agreement with two antibiotics, ampicillin and ciprofloxacin. The fourth application is a multi-dimensional imaging cytometry system that integrates multiple light sources from different angles to simultaneously capture time-lapse, forward scattering and side scattering images of blood cells. The system has demonstrated capacity to detect red blood cell agglutination, assess red blood cell lysis, and differentiate red and white blood cells for potential implementation in clinical hematology analyses. These large-volume, light-scattering cytometric technologies can be used and applied in clinical and research settings to study, detect, and analyze cells. These studies developed rapid point-of-care diagnostic and imaging technologies for collectively advancing modern medicine and global health.

Quantifying molecular interactions is pivotal for understanding biological processes at molecular scale and for screening drugs. Although various detection technologies have been developed, it is still challenging to quantify the binding kinetics of small molecules because the sensitivities of the mainstream technologies scale down with the size of the molecule. To address this problem, two different optical detection methods, charge sensitive optical detection (CSOD) and virion
ano-oscillators, are developed to measure the binding-induced charge change instead of the mass change, which enables quantification of the binding kinetics for both large and small molecules.

In particular, the nano-oscillator approach provides a unique capability to image individual nanoparticles and measure the size and charge of each nanoparticle simultaneously. This approach is applied to measure one of the smallest biological particles - single protein molecules. By tracking the oscillation of each protein molecule, the size, charge, and mobility are measured in real-time with high precision. This capability also allows to monitor the conformation and charge changes of single protein molecules upon ligand binding. Measuring the size and charge of single proteins opens a new revenue to protein analysis and disease biomarker detection at the single molecule level.

The virion
ano-oscillators and the single protein approach employ a scheme where a particle is tethered to the surface with a polymer molecule. The dynamics of the particle is governed by two important forces: One is entropic force arising from the conformational change of the molecular tether, and the other is solvent damping on the particle and the molecule. The dynamics is studied by varying the type of the tether molecule, size of the particle, and viscosity of the solvent. The findings provide insights into single molecule studies using not only tethered particles, but also other approaches, including force spectroscopy using atomic force microscopy and nanopores.

Proteins play a central role to human body and biological activities. As powerful tools for protein detections, many surface plasmon resonance based techniques have been developed to enhance the sensitivity. However, sensitivity is not the only final goal. As a biosensor, four things really matter: sensitivity, specificity, resolution (temporal/spatial) and throughput.

This dissertation presents several works on developing novel plasmonic based techniques for protein detections on the last two aspects to extend the application field. A fast electrochemically controlled plasmonic detection technique is first developed with the capability of monitoring electrochemical signal with nanosecond response time. The study reveals that the conformational gating of electron transfer in a redox protein (cytochrome c) takes place over a broad range of time scales (sub-µs to ms). The second platform integrates ultra-low volume piezoelectric liquid dispensing and plasmonic imaging detection to monitor different protein binding processes simultaneously with low sample cost. Experiment demonstrates the system can observe binding kinetics in 10×10 microarray of 6 nL droplet, with variations of kinetic rate constants among spots less than ±5%. A focused plasmonic imaging system with bi-cell algorithm is also proposed for spatial resolution enhancement. The two operation modes, scanning mode and focus mode, can be applied for different purposes. Measurement of bacterial aggregation demonstrates the higher spatial resolution. Detections of polystyrene beads binding and 50 nm gold nanoparticles oscillation show a high signal to noise ratio of the system.

The real properties of protein rely on its dynamic personalities. The above works shed light upon fast and high throughput detection of protein kinetics, and enable more applications for plasmonic imaging techniques. It is anticipated that such methods will help to invoke a new surge to unveil the mysteries of biological activities and chemical process.

Measuring molecular interaction with membrane proteins is critical for understanding cellular functions, validating biomarkers and screening drugs. Despite the importance, developing such a capability has been a difficult challenge, especially for small molecules binding to membrane proteins in their native cellular environment. The current mainstream practice is to isolate membrane proteins from the cell membranes, which is difficult and often lead to the loss of their native structures and functions. In this thesis, novel detection methods for in situ quantification of molecular interactions with membrane proteins are described.

First, a label-free surface plasmon resonance imaging (SPRi) platform is developed for the in situ detection of the molecular interactions between membrane protein drug target and its specific antibody drug molecule on cell surface. With this method, the binding kinetics of the drug-target interaction is quantified for drug evaluation and the receptor density on the cell surface is also determined.

Second, a label-free mechanically amplification detection method coupled with a microfluidic device is developed for the detection of both large and small molecules on single cells. Using this method, four major types of transmembrane proteins, including glycoproteins, ion channels, G-protein coupled receptors (GPCRs) and tyrosine kinase receptors on single whole cells are studied with their specific drug molecules. The basic principle of this method is established by developing a thermodynamic model to express the binding-induced nanometer-scale cellular deformation in terms of membrane protein density and cellular mechanical properties. Experiments are carried out to validate the model.

Last, by tracking the cell membrane edge deformation, molecular binding induced downstream event – granule exocytosis is measured with a dual-optical imaging system. Using this method, the single granule exocytosis events in single cells are monitored and the temporal-spatial distribution of the granule fusion-induced cell membrane deformation are mapped. Different patterns of granule release are resolved, including multiple release events occurring close in time and position. The label-free cell membrane deformation tracking method was validated with the simultaneous fluorescence recording. And the simultaneous cell membrane deformation detection and fluorescence recording allow the study of the propagation of the granule release-induced membrane deformation along cell surfaces.

Signal transduction networks comprising protein-protein interactions (PPIs) mediate homeostatic, diseased, and therapeutic cellular responses. Mapping these networks has primarily focused on identifying interactors, but less is known about the interaction affinity, rates of interaction or their regulation. To better understand the extent of the annotated human interactome, I first examined > 2500 protein interactions within the B cell receptor (BCR) signaling pathway using a current, cutting-edge bioluminescence-based platform called “NanoBRET” that is capable of analyzing transient and stable interactions in high throughput. Eighty-three percent (83%) of the detected interactions have not been previously reported, indicating that much of the BCR pathway is still unexplored. Unfortunately, NanoBRET, as with all other high throughput methods, cannot determine binding kinetics or affinities. To address this shortcoming, I developed a hybrid platform that characterizes > 400 PPIs quantitatively and simultaneously in < 1 hour by combining the high throughput and flexible nature of nucleic programmable protein arrays (NAPPA) with the quantitative abilities of surface plasmon resonance imaging (SPRi). NAPPA-SPRi was then used to study the kinetics and affinities of > 12,000 PPIs in the BCR signaling pathway, revealing unique kinetic mechanisms that are employed by proteins, phosphorylation and activation states to regulate PPIs. In one example, activation of the GTPase RAC1 with nonhydrolyzable GTP-γS minimally affected its binding affinities with phosphorylated proteins but increased, on average, its on- and off-rates by 4 orders of magnitude for one-third of its interactions. In contrast, this phenomenon occurred with virtually all unphosphorylated proteins. The majority of the interactions (85%) were novel, sharing 40% of the same interactions as NanoBRET as well as detecting 55% more interactions than NanoBRET. In addition, I further validated four novel interactions identified by NAPPA-SPRi using SDS-PAGE migration and Western blot analyses. In one case, we have the first evidence of a direct enzyme-substrate interaction between two well-known proto-oncogenes that are abnormally regulated in > 30% of cancers, PI3K and MYC. Herein, PI3K is demonstrated to phosphorylate MYC at serine 62, a phosphosite that increases the stability of MYC. This study provides valuable insight into how PPIs, phosphorylation, and GTPase activation regulate the BCR signal transduction pathway. In addition, these methods could be applied toward understanding other signaling pathways, pathogen-host interactions, and the effect of protein mutations on protein interactions.

Antibiotic resistant bacteria are a worldwide epidemic threatening human survival. Antimicrobial susceptibility tests (ASTs) are important for confirming susceptibility to empirical antibiotics and detecting resistance in bacterial isolates. Current ASTs are based on bacterial culturing, which take 2-14 days to complete depending on the microbial growth rate. Considering the high mortality and morbidity rates for most acute infections, such long time frames are clinically impractical and pose a huge risk to a patient's life. A faster AST will reduce morbidity and mortality rates, as well as help healthcare providers, administer narrow spectrum antibiotics at the earliest possible treatment stage.

In this dissertation, I developed a nonculture-based AST using an imaging and cell tracking technology. I track individual Escherichia coli O157:H7 (E. coli O157:H7) Uropathogenic Escherichia Coli (UPEC) cells, widely implicated in food-poisoning outbreaks and urinary tract infections respectively. Cells tethered to a surface are tracked on the nanometer scale, and phenotypic motion is correlated with bacterial metabolism. Antibiotic action significantly slows down motion of tethered bacterial cells, which is used to perform antibiotic susceptibility testing. Using this technology, the clinical minimum bactericidal concentration of an antibiotic against UPEC pathogens was calculated within 2 hours directly in urine samples as compared to 3 days using current gold standard tools.

Such technologies can make a tremendous impact to improve the efficacy and efficiency of infectious disease treatment. This has the potential to reduce the antibiotic mis-prescription steeply, which can drastically decrease the annual 2M+ hospitalizations and 23,000+ deaths caused due to antibiotic resistance bacteria along with saving billions of dollars to payers, patients, and hospitals.