A modelling and computational study of biofilm dynamics: Exploring the role of initial attachment and horizontal gene transfer
Loading...
Repository DOI
Publication Date
2024-12-03
Type
doctoral thesis
Downloads
Citation
Abstract
Biofilms are ubiquitous in nature and one of the most successful lifestyles on earth. This success is largely due to the high level of interactions within biofilms, leading to the specialisation of subpopulations and subsequent heterogeneity. Biofilms have gathered research attention because of their potential use in engineering, but also because of the need to control their formation in man-made systems. However, due to their complexity and heterogeneity, biofilm systems can be cumbersome to analyse. In this context, model biofilm systems, whether they are experimental or mathematical, offer a reduction of the system to a more observable one through simplifying assumptions and hence allow easier analysis. This thesis aims to propose innovative biofilm models, focusing on diverse key aspects of biofilm ecosystems.
More specifically, this dissertation is divided between an original experimental model system, proposed to investigate cross-kingdom interactions in anaerobic communities during biofilm formation and four novel mathematical models, introduced to describe the main regulatory effects of trace metals on microbial biofilms. To this aim, novel mathematical functions are introduced to describe key phenomena linked to trace metals in biofilms dynamics: bacterial attachment (and subsequent biofilm initiation) and horizontal gene transfer, one of the drivers of genotypic diversity and antibiotic resistance.
The experimental model system is an undefined, engineered model system mimicking growth conditions in an Upflow Anaerobic Sludge Blanket reactor. It was studied with a sampling campaign during the first five days of the biofilm establishment, followed by biofilm characterisation with fluorescence microscopy and DNA sequencing at different time points. To complement DNA sequencing, a network-based analysis was carried out to investigate correlations between fungal and bacterial populations in different samples. This exploratory study into an unknown biofilm type brought to light the importance of the interactions between the bacterial microbiome and its fungal counterpart.
The mathematical models in this thesis are continuum models, formulated as systems of non-linear partial differential equations. Non-linear hyperbolic PDEs govern the advective transport and growth of the solid-phase species forming the biofilm, while parabolic quasi-linear PDEs model the diffusion-reaction of soluble substrates and bacteriophages. The first model presented focuses on the influence of ionic strength on bacterial attachment and subsequent biofilm formation, and is formulated as a 1D free boundary problem. This model focuses on drinking water distribution systems biofilms and their invasion by pathogenic bacteria Legionella pneumophila, with a focus on the necrotrophic metabolism of the latter, which gives it the ability to persist in biofilms. The second model is formulated as a free-boundary problem describing the impact of conjugation on plasmid spread in biofilm communities. More specifically, conjugation is modelled as a mass-action kinetics process subsequent to gene expression, modelled as a nonlocal term to account for recipient-sensing. The existence and uniqueness of the solutions are proved using the method of characteristics and the fixed point theorem. The third model is formulated as a multidimensional problem, and proposes a modelling framework to natural transformation in biofilms, modelled as a frequency-dependent process. It investigates the comparative influence of conjugation and transformation on the spread of antibiotic resistance and biofilm compartmentalisation due to differences in metabolisms and sensitivities to toxic stressors. Finally, the last model describes the interaction between bacteriophages and biofilm communities and includes generalised transduction. This model is the first biofilm model that includes the three main horizontal gene transfer mechanisms: conjugation, transformation and transduction.
All models are integrated numerically through the implementation of original code in MatLab and Comsol Multiphysics. Numerical simulations allow investigating the behaviour of the models, which are able to describe and predict key phenomena of biofilm dynamics. The results of the experimental section demonstrate the adequacy of model systems for investigating biofilm formation. The mathematical models can reproduce crucial elements of biofilm ecology, namely initial bacterial attachment and the main aspects of plasmid spread, such as horizontal gene transfer, the impact of selective pressure on vertical gene transfer or bacteriophage activity.
Publisher
University of Galway
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International