Use this URL to cite or link to this record in EThOS:
Title: The impact of Schistosoma mansoni infection and treatment on the gut microbiota of school-aged children in Uganda
Author: Carruthers, Lauren Vera
ISNI:       0000 0005 0286 8978
Awarding Body: University of Glasgow
Current Institution: University of Glasgow
Date of Award: 2021
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Thesis embargoed until 04 Feb 2024
Access from Institution:
Schistosomiasis is a neglected tropical disease infecting millions of people globally, mainly in developing countries. In Mayuge District, Uganda, Schistosoma mansoni infection remains prevalent, with high infection intensities still common, despite over a decade of annual mass drug administration (MDA) interventions. The gut microbiome has become increasingly topical as studies continue to demonstrate correlations in microbial structure, with healthy or diseased microbiome states associated with a diverse range of medical conditions. Research shows that gut dwelling helminths can influence bacterial gut microbiota composition. Less, however, is known about the impact that helminths, which reside in the bloodstream but rely on egg passage into the gut to be excreted, may have on the gut community. The microbiome can also positively and negatively influence drug treatment though it is unclear whether there are any associations between the microbiome and combined praziquantel and albendazole treatment, standard of annual MDA in Uganda. It has also yet to be established if the microbiome influences S. mansoni clearance outcomes after administration of these drugs. The research in this thesis addresses these points across a longitudinal treatment study in the communities of Bugoto, Mayuge District, Uganda, with three key aims: 1) To identify if the host gut microbiome changes with co-administered anti-helminthic praziquantel and albendazole treatment; 2) To explore relationships between S. mansoni and the host gut microbiome and determine the impact of other prevalent co-infections, specifically hookworm and malaria, and host demographics on the gut microbiome; 3) To determine if microbiome composition can predict S. mansoni clearance. Prior to commencement of the longitudinal treatment study it was first necessary to determine the optimal approach to collect and store stool for microbiome analyses in the field. Initially, stool specimens were collected from three children. Inner and outer regions from each stool were stored raw, in RNAlater, and in ethanol and then frozen within 30 min of collection (0 h). Homogenised samples were also stored in each preservative and then frozen at 0 h and at doubling incremental time-points from 1 – 32 h. The impact of preservative, time-to-freezing, stool homogeneity and each child on microbial diversity and abundance were assessed using V1 – V2 region 16S rRNA sequencing. Data were analysed using QIIME and statistical analyses performed in R. Each child had a distinct microbiome signature that was characteristic from the phylum to genus taxonomic levels, regardless of the preservative, stool region and time-to-freezing. Preservation method significantly influenced species richness (number of species/operational taxonomic units), with each preservative also having a distinctive microbiome profile and biomarkers. Each preservative however, performed well with little variation between samples stored from each child within each preservative. Time-to-freezing and stool region were found to have a minor impact on the microbiome diversity and composition. Sampling and storage consistency were determined to be important for reliable microbiome comparisons to be made across the longitudinal microbiome study. Raw stool storage was used for the three key aims as this was deemed the most practical for the study. Two hundred and seven school-aged children, across two cohorts, were enrolled into the longitudinal treatment study. Children from Cohort 1 (aged six - 14 years-old) provided stool samples for microbiome analysis at study baseline in September 2017 and at 24 h, 3 wk and 22 wk post-treatment with praziquantel and albendazole. Children from Cohort 2 (praziquantel naïve, aged six-years-old) provided samples at baseline in March 2018 and 3 wk post-treatment. The impact of praziquantel and albendazole treatment on microbial diversity and abundance was assessed using V1 – V2 region 16S rRNA sequencing. Data were analysed using QIIME2 and additional statistical analyses performed in R. A significant shift in microbiome populations was associated with treatment, with a reduction in bacterial diversity and a significant increase in the relative abundance of the Prevotellaceae family by 24 h post-treatment. This microbiome disruption however appeared to be short-lived, with microbiomes recovering to baseline populations by 3 wk post-treatment. Treatment appeared to have a negligible impact in the long-term, with changes at 22 wk post-treatment considered more likely to be due to seasonal variation. Baseline and 22 wk post-treatment data were combined to determine the impact of S. mansoni prevalence and intensity, hookworm infection, malaria, anaemia, age, sex, household location and sample collection time-point on microbiome diversity and composition. Two separate analyses were undertaken for S. mansoni using two diagnostic methods; Kato-Katz thick-smear and point-of-care circulating-cathodic-antigen (POC-CCA) test which detect eggs and adult worm antigens respectively. S. mansoni infection intensity, collection time-point, age and anaemia were the variables identified by linear mixed effect modelling, found to influence microbiome diversity, as measured by one or more of the alpha and phylogenetic diversity measures. The microbiome diversity profiles between groups within each infection and demographic variable, were similar as measured by beta diversity metrics. Of the few significant beta diversity differences identified, these were not consistent across time-points. A few bacterial biomarkers were identified to be associated with S. mansoni infection however, there was no consistency across the diagnostic methods. No biomarkers were identified for the other variables. To identify if the microbiome could predict treatment success, the microbiomes of individuals who cleared their S. mansoni infections by 3 wk post-treatment compared to those who did not clear, were compared at baseline, at 24 h posttreatment and at 3 wk post-treatment. Analyses were first undertaken using Kato-Katz data as a measure of S. mansoni infection and then repeated using POC-CCA test data. Alpha, phylogenetic and beta diversity were similar between cleared and infected individuals across all three time-points. Bacterial abundance patterns between infected and cleared children also generally appeared similar. Bacterial biomarkers associated with clearance status were identified at every time-point however, there was no consistency across the diagnostic methods with individual variation appearing to account for at least some of the differences. The impact of findings are discussed in the context of S. mansoni infection. Challenges associated with microbiome analysis and the limitations of parasitic diagnostics are also highlighted. Limitations associated with individual variation also proved a recurring theme. Finally, suggestions for future studies to improve finding reproducibility and overcome challenges are described. Overall this research contributes to our understanding of schistosome – microbiome – treatment relationships.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral