Atterbury, Robert J
(2024)
Phylogenetic Investigation of Population Changes when using Bacteriophage to Control Salmonella Typhimurium and Escherichia coli infections in Pigs.
MRes thesis, University of Nottingham.
Abstract
Diarrhoeal disease is the third largest cause of mortality in children under 5 years old, leading to approximately half a million deaths per annum globally. Bacterial infections are thought to be responsible for just under half of these deaths, with Enterotoxigenic Escherichia coli (ETEC) and Nontyphoidal Salmonella (NTS) among the most frequently cited causes. Coinfection with these pathogens has been detected in up to half of infant diarrhoea cases, and this may be associated with more severe disease. ETEC and Salmonella Typhimurium also cause severe diarrhoea in piglets, leading to morbidity, mortality, significant economic loss and the risk of disease transmission in the human food chain. Hence, pigs have been used as a model of bacterial diarrhoea, both for veterinary research and as a proxy for human infant infection. Antimicrobial resistance (AMR) in both ETEC and Salmonella is a significant and growing global challenge in both humans and animals and alternative treatments are urgently needed. Bacteriophage (viruses which specifically infect bacteria) have been mooted as one such alternative.
In this thesis, I present genomic and phenotypic analysis of a collection of bacteriophage which infected either Salmonella Typhimurium or ETEC. I used the results of this analysis to select the best phage candidates to use in a therapeutic trial in pigs coinfected with both pathogens. I then used shotgun metagenomic analysis of pig faeces from control and phage-treated pigs taken at time points throughout this trial to determine the effect of both the pathogens and bacteriophage on the gut microbiome.
Genomic analysis revealed that all of the 96 phage used in this study belonged to the Caudoviricetes order of dsDNA tailed phage, and all but one were predicted to exhibit a lytic lifecycle. Most phage genomes were predicted to be circularly permuted, linear and use headful packaging of their DNA. None of the phage genomes carried known AMR or virulence genes. These initial findings suggested that all but one of these phage may have potential as therapeutic candidates.
I then performed hierarchical clustering of all the phage genomes based on amino acid similarity. These genome clusters where then used together with phenotypic data on the host ranges of these phage to perform two-dimensional hierarchical clustering based on genome similarity and host range across different bacterial serotypes. This was done to determine whether phage with similar genomes exhibited a similar host range. The results suggested that host range was strongly associated with genome cluster for some phage but not for others.
Whole genome sequencing was performed for two phage which infected the broadest range of E. coli and Salmonella (P433.1 and phi 10 respectively). Coliphage 433.1 belonged to the Autographiviridae family and had a 44 kb genome whereas Salmonella phage Phi10 belonged to the Ackermannviridae family and had a 159 kb genome. One step growth curves revealed that P433.1 had a significantly larger burst size (241) and shorter latent period (12 min) than Phi10 (14 and 43 min respectively). The lifecycles of both phage were predicted to be exclusively lytic and they formed clear plaques on agar overlays seeded with susceptible bacteria.
High titre (1010 PFU mL-1) suspensions of each phage were used to treat 21-day-old pigs which had been challenged with a 109 CFU mL-1 suspension containing antibiotic resistant variants of enteropathogenic E. coli P433 and Salmonella Typhimurium 4/74. Phage treatment was associated with significant reductions (>2.0 log10 CFU, p < 0.05) in both E. coli and Salmonella as well as reduced clinical scores compared with infected but untreated control animals. This was accompanied by a contemporaneous and sustained increase in the numbers of both phage, indicating active viral replication in the gut.
Metagenomic analysis of gut samples and faeces throughout the trial revealed variability between individual samples, even those belonging to the same treatment group. However, Non-Metric Multidimensional Scaling (NMDS) analysis of between-sample diversity scores revealed that the phage-treated and challenged groups clustered together, whereas the untreated control group clustered separately. This suggested that the gut microbiota of the pigs was somewhat disrupted following challenge with the pathogens, and that phage treatment neither worsened nor ameliorated this disruption. However, further analysis with PERMANOVA found that these differences were not statistically significant (p = 0.09).
In summary, this study demonstrates that phage therapy can effectively reduce colonisation of pigs co-challenged with two bacterial pathogens, and was associated with significant improvements in clinical scores of disease. It further demonstrates that phage therapy was not associated with significant disruptions to the gut microbiota following pathogen challenge, suggesting it may be preferable to the use of broad spectrum antimicrobials for the selective targeting of pathogenic bacteria.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Robert Atterbury - 20310865 - MRes thesis - Deposited Version.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Robert Atterbury - 20310865 - MRes thesis - Deposited Version.pdf")