Abstract
Bioremediation is a clean-up technology that has been used to degrade hydrocarbon contaminants in soil and sediments. However, bioremediation in the ocean in the context of an oil spill is not well studied. The heterogeneous and ephemeral hydrocarbon distribution increases the challenges for marine bacteria to navigate the ocean in search of favorable locations for survival. Therefore, transport properties, such as motility and chemotaxis provide marine bacteria a way to locate and swim preferentially toward the hydrocarbons. Studies have shown that chemotaxis has the potential to increase bioavailability and enhance biodegradation efficiency. However, most studies have focused on bacteria chemotaxis to a single stimulus, while bacteria are commonly exposed to multiple stimuli with competing effects. Bacteria chemotactic responses to multiple stimuli are not fully understood. A mathematical model can give us insights into the chemosensory mechanism that bacteria use to integrate their overall response to multiple inputs. In this work, the chemotactic responses of the well-studied bacterium Escherichia coli (E. coli) were first measured to validate a mathematical model. Then, the model was adapted and used to study marine bacteria Halomonas sp. chemotactic responses.
The mathematical model for E. coli chemotactic responses to multiple stimuli was evaluated with experimental data. In the model, E. coli chemotactic velocities were derived from the transport equations for an attractant α-methylaspartate alone, repellent nickel ion alone, and several combinations of the two. The multi-scale model related the individual bacterium response to the population level response. At the individual level, the model incorporated the signal transduction mechanism as the stimuli bind to the receptor that crosses the cell membrane and the subsequent signaling reactions inside the cell. Values for the chemotactic parameters (stimuli sensitivity coefficient σ, signaling efficiency γ, and repellent sensitivity coefficient κ) were obtained by fitting the model to experimental results. The experimental data were collected using a microfluidic device designed to create a constant concentration gradient. For E. coli, the model correctly predicted the overall attraction or repulsion outcomes of the mixtures. However, quantitatively the prediction showed a slightly greater repellent response of the mixture than the experimental results.
The mathematical model was then used to evaluate marine bacteria Halomonas sp. 10BA chemotactic responses to multiple stimuli. It was assumed that Halomonas sp. have the same run-and-tumble motility pattern as E. coli because they have the same flagella arrangement. However, published studies suggested that the chemotactic mechanism for marine bacteria is different from E. coli. This difference was also supported from the experimental results because direct application of the model used for E. coli failed to correctly capture the chemotactic response of Halomonas sp. Since marine bacteria exhibit faster chemotactic responses than E. coli, instead of using only a single receptor for sensing a stimulus, we assumed that Halomonas sp. can use multiple receptors. The model was then updated using two chemotactic receptors for sensing attractant decane and one chemotactic receptor for sensing repellent copper. This change resulted in Halomonas sp. responding more quickly to stimuli. The updated model predicted Halomonas sp. response to multiple stimuli well at the low repellent Cu concentration, while it overpredicted bacteria repulsion at high Cu concentration of 2 mM. This discrepancy may be due to a high association constant for the reaction of Cu bound receptor and the phosphate group at high Cu concentration, altering the value for the repellent sensitivity coefficient κ. This underperformance may also possibly result from the increased swimming speed of marine bacteria at a higher repellent concentration. This increased swimming speed further changes the strength of chemotactic responses to both repellent and attractant. This work suggests that the signaling pathway adapted from E. coli can be applied to qualitatively describe marine bacteria chemotactic response to multiple stimuli. However, further modification should be applied to the model to accurately predict the response quantitatively. Regardless, this model can provide qualitatively estimation on the naturally occurring marine bacteria respond to the oil given the oil composition. This can provide some information on whether bioremediation should be considered or the conventional interventions should be enforced in oil spill cleanup.
