Modelling biofilm structure and biofilm-micropollutant interactions using an individual-based Model

Broughton, James Kuan (2019-04)

Thesis (MEng)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: Micropollutants are emerging contaminants that have received considerable attention in recent years due to environmental concerns. Diclofenac is a nonsteroidal anti-inflammatory drug that is commonly detected in South African surface waters and has been shown to be persistent and harmful to the environment. Biofilms, which are composed of sessile, mixed microbial communities, play a key role in wastewater treatment and functioning of natural ecosystems. There is, however, a poor understanding of biofilm-micropollutant interactions in this context. Biofilm structure is a useful indicator of the effect of micropollutant exposure on a biofilm and its activity. Recently, individual-based models have been developed that can simulate biofilm structure from first principles and show potential as a tool to provide mechanistic understanding of these interactions. The aim of the project was to develop and validate an individual-based model capable of describing biofilm structural development of a sample, as well as reproducing observed effects of diclofenac exposure on biofilm structure. A further aim was to investigate the feasibility of observed biofilm structure for validation of the biofilm model. Lumped biokinetic parameters of an environmental sample were estimated using batch respirometric methods and regression of model parameters. Heterotrophic growth was found to dominate in the sampled culture. The estimated parameters were thus used to calibrate lumped heterotrophic growth in the biofilm model. Mixed-species biofilms were cultivated in flow cells under control conditions and exposed to diclofenac at 0.1 and 10 mg.L-1. Confocal laser scanning microscopy was used to examine morphological changes in biofilm structure over time. Biofilm structural parameters were derived from microscopy data and compared to the simulation output. Observed structure of biofilms were successfully used to validate the proposed biofilm model. Experimental results indicated a dose-dependent response to diclofenac. Diclofenac at 10 mg.L-1 significantly inhibited biofilm growth over the exposure period compared to a control. Exposure at 0.1 mg.L-1 resulted in an increase in biofilm growth after 24 h. The biofilm model successfully reproduced observed trends in structure at 10 mg.L-1 and 0.1 mg.L-1. More work is required to elucidate the complex microbial interactions of diclofenac in the μg.L-1 range. This study showed that an individual-based model can reproduce in vitro biofilm structure development based on emergent structural parameters. Individual-based models coupled with comparative experimental methods show potential as a tool for investigating biofilm interactions and improving model development. However, areas of improvement were identified including model parameter uncertainties, limitations in the biofilm model, and reproducibility in experimental methods.

AFRIKAANSE OPSOMMING: Mikrobesoedelstowwe is opkomende kontaminante wat aansienlike aandag gekry het in die laaste jare as gevolg van bekommernisse oor die omgewing. Diclofenac is ’n nie-steroïedale anti-inflammatoriese geneesmiddel wat algemeen in Suid-Afrikaanse oppervlakwater waargeneem word en is bewys om volhoudend en beskadigend vir die omgewing te wees. Biofilms, wat bestaan uit sessiel, gemengde mikrobiese gemeenskappe, speel ’n belangrike rol in afvalwaterbehandeling en funksionering van natuurlike ekosisteme. Daar is egter gebrekkige insig oor biofilm-mikrobesoedelstof-interaksies in hierdie konteks. Biofilmstruktuur is ’n nuttige indikator van die effek van mikrobesoedelstof blootstelling op ’n biofilm en sy aktiwiteit. Onlangs is individu-gebaseerde modelle ontwikkel wat biofilmstruktuur kan simuleer vanuit eerste beginsels en wys potensiaal as ’n hulpmiddel om meganistiese insig in hierdie interaksies te verskaf. Die doel van hierdie projek was om ’n individu-gebaseerde model te ontwikkel geskik om die biofilm strukturele ontwikkeling van ’n steekproef te beskryf, sowel as om waargenome gevolge van diclofenac-blootstelling op biofilmstrukture te reproduseer. ’n Verdere doel was om die uitvoerbaarheid van waargenome biofilmstrukture vir die validering van die biofilmmodel te ondersoek. Gekonsentreerde biokinetiese parameters van ’n omgewingsproefsteek is beraam deur lot respirometriese metodes en regressie van modelparameters te gebruik. Dis gevind dat heterotrofiese groei in die proefsteekkultuur domineer. Die beraamde parameters is dus gebruik om gekonsentreerde heterotrofiese groei in die biofilmmodel te kalibreer. Gemengde-spesies-biofilm is gekweek in vloeiselle onder gekontroleerde toestande en blootgestel aan diclofenac by 0.1 en 10 mg.L- 1. Konfokale-laserskanderingmikroskopie is gebruik om morfologiese veranderinge in biofilmstrukture oor tyd te ondersoek. Biofilm strukturele parameters is afgelei uit mikroskopie data en vergelyk met die simulasie uitsette. Die waargenome struktuur van biofilms het die voorgestelde biofilmmodel suksesvol gevalideer. Eksperimentele resultate het ’n dosis-afhanklike respons op diclofenac aangedui. Diclofenac by 10 mg.L-1 het biofilm groei aansienlik onderdruk oor die periode van blootstelling in vergelyking met ’n kontrole. Blootstelling by 0.1 mg.L-1 het na ’n verhoging van biofilmgroei gelei na 24 uur. Die biofilmmodel het waargenome neigings in struktuur by 10 mg.L-1 en 0.1 mg.L-1 suksesvol gereproduseer. Meer werk word benodig om die komplekse mikrobiese interaksies van diclofenac in die μg.L-1 bestek te verklaar. Hierdie studie het gewys dat ’n individu-gebaseerde model in vitro biofilmstruktuurontwikkeling, gebaseer op opkomende strukturele parameters, kan reproduseer. Individu-gebaseerde modelle wat gepaard gaan met vergelykende eksperimentele metodes, wys potensiaal as ’n hulpmiddel om biofilminteraksies te ondersoek en om model ontwikkeling te verbeter. Areas van verbetering is egter geïdentifiseer, insluitend model parameter onsekerhede, beperkinge op die biofilmmodel, en herhaalbaarheid in eksperimentele metodes.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/105955
This item appears in the following collections: