In this work, we aim to observe and study the physics of bacteria and cancer cells pearl chain formation under dielectrophoresis (DEP). Experimentally, we visualized the formation of Bacillus subtilis bacterial pearl chain and human breast cancer cell (MCF-7) chain under positive and negative dielectrophoretic force, respectively. Through a simple simulation with creeping flow, AC/DC electric fields, and particle tracing modules in COMSOL, we examined the mechanism by which bacteria self-organize into a pearl chain across the gap between two electrodes via DEP. Our simulation results reveal that the region of greatest positive DEP force shifts from the electrode edge to the leading edge of the pearl chain, thus guiding the trajectories of free-flowing particles toward the leading edge via positive DEP. Our findings additionally highlight the mechanism why the free-flowing particles are more likely to join the existing pearl chain rather than starting a new pearl chain. This phenomenon is primarily due to the increase in magnitude of electric field gradient, and hence DEP force exerted, with the shortening gap between the pearl chain leading edge and the adjacent electrode. The findings shed light on the observed behavior of preferential pearl chain formation across electrode gaps.

No abstract

In this thesis, our focus is the development of indirect binding models to predict the chemotactic response of E. coli to the signaling molecule AI-2. Our first model is developed from first-order binding kinetics to represent the interactions between the attractant molecule and the periplasmic binding protein, as well as between the ligand-binding protein pair and the chemoreceptor. The model accounts for the effects of the dissociation constants characterizing each binding interaction, as well as the crucial effect of the inclusion of the periplasmic binding protein concentration. The first-order model is validated through simulations of the capillary assay popularized by Adler and compared with existing experimental results in the literature. The predicted chemotactic response from our model is first compared with experimental reports of maltose chemotactic response as the maltose response is well-studied and there exists a wealth of data in the past literature. Our model achieves good agreement with experimental data for maltose at various levels of periplasmic binding protein concentration as well as extracellular attractant concentration. Moreover, the model is consistent with experimental results of multiple studies for one universal set of kinetic and chemotaxis parameters, lending further support to the accuracy of predictions made.Our next study involves the parametric analysis of the first-order model to better understand strategies pertaining to the indirect binding mechanism. It appears that the advantage of the indirect binding mechanism can be observed in nutrient-scarce conditions, as a bacterium is predicted to be able to ramp up the magnitude of chemotactic response and shift the sensitivity profile towards dilute attractant concentrations by raising the concentration of periplasmic binding protein moderately. In the third and last study presented in this thesis, we extend the indirect binding model to develop a higher-order model based on specific considerations for the AI-2 system. As the concentration of periplasmic binding protein is much lower than receptor concentration in the case of AI-2, there is a high possibility of ligand depletion, making it necessary to utilize the exact equilibrium balance for the total periplasmic binding protein. The higher-order model exhibits greater agreement with existing experimental data as compared to the earlier first order model. Moving forward, we believe it necessary to develop experimental approaches to aid in model validation, identifying microfluidics as a possible future direction.

No abstract