The internal structure of biofilms is crucial for their growth and development. In
particular, the network of internal channels recently discovered in Escherichia coli
functions as a nutrient uptake and diffusion system. While mesoscopic imaging
of these channels has proven to be a powerful visualisation tool, the lack of a
specialised image analysis pipeline has so far prevented an accurate quantitative
characterisation of channel morphology. The genetic determinants of channel
formation also remain unknown.
This thesis describes the imaging and quantification of the network of nutrienttransporting
channels in E. coli biofilms through open-source image analysis
methods, and presents an experimental framework to perform gene deletions leading
to a change in cell shape for the E. coli strain JM105 mini-Tn7-gfp.
The morphology of bacterial biofilms is strongly affected by environmental growth
conditions, with mechanical and biochemical composition of the growth substrate
contributing to the formation of complex three-dimensional patterns. Thanks to
the combination of mesoscopic fluorescence imaging and open-source image analysis,
the width of individual nutrient-transporting channels was measured for
biofilms formed by the E. coli strain JM105 mini-Tn7-gfp, revealing a strong
dependence of channel width on both spatial location inside the biofilm and nutrient
availability within the substrate. The mechanism of nutrient transport
within channels was proposed to follow fluid dynamic behaviour, which would lead to increased nutrient flow towards the centre of the biofilm, where channels
are smaller in diameter.
The use of fractal geometry tools for the quantification of biofilm morphology and
expansion patterns is well documented. The network of intra-colony channels in
E. coli was also originally predicted to exhibit a fractal morphology, and this was
verified in this work through fractal analysis of mature E. coli biofilm images. A
dependence of channel architecture on E. coli cell shape was hypothesised due
to channel formation being an emergent property of biofilm formation, and this
was investigated with four cell shape mutants of E. coli obtained from the Keio
collection, which is a single-gene knockout library derived from the laboratory
strain BW25113. The complexity of internal channels was found to be comparable
to that of computer-generated fractals for all strains grown on both rich and
minimal medium substrates, though cell shape was not identified as a unique
channel morphology descriptor.
The characterisation of nutrient-transporting channels in E. coli has so far been
performed on the laboratory strain JM105 mini-Tn7-gfp thanks to its compatibility
with Mesolens imaging, which provides subcellular resolution across whole,
multi-millimetre sized biofilms. While the results presented in Chapter 2 of
this thesis were obtained with the same strain, the fractal analysis of biofilms
formed by cell shape mutants described in Chapter 3 was performed on the strain
BW25113, the parental strain for the Keio collection. This was due to the rapid
biofilm disruption to planktonic state exhibited by BW25113 during immersion
with liquid mounting medium, which was necessary to match the refractive index
of the Mesolens used in water immersion mode. A genetic engineering protocol
based on Lambda Red recombination was hence designed in order to circumvent
this problem and carry out the inactivation of genes involved in the regulation of cell shape in E. coli JM105 mini-Tn7-gfp. While testing this method, the resistance
of JM105 mini-Tn7-gfp to the antibiotic ampicillin was discovered and
characterised, leading to a proposed modification of the genetic engineering protocol
using traditional cloning methods.
| Date of Award | 7 Mar 2024 |
|---|
| Original language | English |
|---|
| Awarding Institution | - University Of Strathclyde
|
|---|
| Sponsors | EPSRC (Engineering and Physical Sciences Research Council) & University of Strathclyde |
|---|
| Supervisor | Gail McConnell (Supervisor) & Paul Hoskisson (Supervisor) |
|---|