Using shotgun metagenomics to explore the effects of HIV infection on the longitudinal nasopharyngeal microbiome of Malawian adults

Busby, Joseph (2022) Using shotgun metagenomics to explore the effects of HIV infection on the longitudinal nasopharyngeal microbiome of Malawian adults. MSc(R) thesis, University of Glasgow.

Abstract

People living with HIV are more likely to suffer from respiratory tract infections, including bacterial pneumonia, which can occur either on its own or as a secondary infection following respiratory viral infections. The nasopharynx is home to a microbial community collectively referred to as the nasopharyngeal microbiome (NPM). Many potential pathogens are carried asymptomatically in the nasopharynx in a proportion of the population, but their overgrowth and spread to the lungs can cause pneumonia. Therefore, the NPM can either repel pathogens or act as a reservoir for them to proliferate and spread to the lungs. We proposed that there may be differences in the NPM of HIV-infected and HIV-uninfected individuals, and that this might contribute to the increased risk of bacterial pneumonia associated with HIV infection.

To characterise the NPM, we carried out DNA and RNA shotgun sequencing on nasopharyngeal swab samples and processed them through a custom metagenomics pipeline that calculated the relative abundance of microbial species in each sample. We analysed the NPM of 10 HIV infected individuals (cases) and 6 HIV uninfected-individuals (controls) at 3 timepoints: baseline, 1 month and 9 months. This study represented the first attempt to study the NPM in HIV-infected individuals.

Analysis with PERMANOVA showed that on average, NPM composition was significantly different depending on HIV status. We found that the relative abundance of the genera Cutibacterium and Pahexavirus were significantly lower in cases than controls, and that this was driven by bacteriophages from the Pahexavirus genus and their bacterial host Cutibacterium acnes. Diversity analysis identified that the NPM of cases was less stable over time, had lower viral richness and higher bacterial evenness. These diversity measurements could all be at least partially attributed to differing abundance of Pahexavirus and Cutibacterium between cases and controls. We proposed that C. acnes might act as an immunomodulator in the NPM, however this would require further study to confirm.

This study also served as a proof of concept of using shotgun metagenomics to profile the NPM, where the microbial community is much less dense than in the gut. Contaminant human sequences were a major issue for the DNA and RNA datasets that limited the scope of our analysis, even when steps were taken to deplete human sequences prior to sequencing. The compositional nature of sequence data also caused issues for analysis. The detection and quantification of relationships between microbes can be improved by quantifying the absolute abundances of microbes, instead of using their relative abundances. We proposed steps that future studies could take to further reduce human contamination and to quantify the absolute abundances of microbial species in their samples.

• Item Type: Thesis (MSc(R))
• Qualification Level: Masters
• Subjects: Q Science > QR Microbiology
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Infection & Immunity
• Supervisor's Name: Ho, Dr. Antonia and Robertson, Professor David
• Date of Award: 2022
• Depositing User: Theses Team
• Unique ID: glathesis:2022-83430
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 14 Feb 2023 14:22
• Last Modified: 17 Feb 2023 08:47
• Thesis DOI: 10.5525/gla.thesis.83430
• URI: https://theses.gla.ac.uk/id/eprint/83430