Analysis of the sinonasal microbiome during health and chronic rhinosinusitis

Abstract

Chronic rhinosinusitis (CRS) is a common, debilitating disease that affects about 5% of the general population, and is characterised by long-term mucosal inflammation of the sinonasal cavity. The pathogenic role of the sinonasal microbiome, as well as the contribution of individual bacterial strains, in the aetiology of this condition, remains uncertain. This is due at least in part to a paucity of research describing the diversity, stability and functional capability of the sinonasal microbiota. This thesis seeks to elucidate microbial community patterns associated with diseased and healthy sinus mucosa by the application of molecular, cultivation and bioinformatic approaches. A meta-analysis based on bacterial community sequencing data revealed that increased dispersion of bacterial communities, lower bacterial diversity, and increased abundance of the genus Corynebacterium were associated with CRS. Bacterial network cooccurrences and fragmentation, which measures the network topography, were conducted. In silico removal of Burkholderia and Propionibacterium from healthy communities was correlated with a significant increase in fragmentation, suggesting their presence may be important in maintaining a stable sinonasal bacterial community structure. One of the first descriptions of predicted metagenomic functional differences between healthy and CRSaffected individuals is presented, with subsequent application of a metagenomic shotgun sequencing approach to validate those differences. Metagenomic shotgun sequencing revealed a variety of viruses, bacteriophages, bacteria, fungi and archaea that exist in the sinuses of CRS patients. The longitudinal stability of the bacterial and fungal communities in healthy subjects was studied over a two-year period. The results indicate that a small bacterial core community exists concurrently with a more transient fungal community. Taxa that exhibited seasonal and annual changes were identified, and some of these changes were correlated with local meteorological conditions. Finally, the broad, community-based view was narrowed to focus on strain-level diversity of Staphylococcus aureus in CRS patients. This research contributes to our current understanding of the role of the microbiome in CRS at both the microbial community and strain-levels, provides evidence for bacterial dysbiosis in CRS, and presents some insight into the functional capabilities of the CRS microbiome.