Abstract:
Chronic obstructive pulmonary disease (COPD) is a leading cause of death and is highly prevalent in South Africa (19% in adults over the age of 40 years). Inflammation of the lungs in COPD impairs the immune response and allows colonisation and infection with bacteria and viruses, that may cause exacerbations of the disease.
Culture-independent technologies have greatly increased the understanding of the lung microbiome. The most widely used method for targeted metagenomics is 16S rRNA sequencing. The IS-Pro (intergenic spacer profiling) method provides an alternative targeted metagenomics approach; however, the two methods have not been compared. There is limited data on the microbiome in the lungs of COPD patients in Africa. Due to local environmental conditions, immunological differences and clinical comorbidities, such as HIV, the microbiome may be different from that reported in studies from other countries. The purpose of this study was to identify the lung microbiome and lung virome in COPD patients in South Africa and to determine if the COPD disease states result in differences in its composition. Next-generation sequencing was used to determine the microbiome and virome of COPD patients from hospitals in Pretoria, South Africa and the IS-Pro method was compared to targeted metagenomics.
Twenty-four patients over the age of 40 years with a confirmed COPD diagnosis and no Mycobacterium tuberculosis infection were included; eighteen were in the stable state of diseases and six were in the exacerbation state of disease. Sputum specimens were collected from all consenting participants and DNA and RNA were extracted directly from the specimens using commercial kits. The extracted bacterial DNA was sent for targeted metagenomics and the IS-Pro method and the extracted viral DNA and RNA were sent for shotgun metagenomics sequencing.
The lung of the COPD participants showed a diverse microbiome with over 77 genera identified and the Firmicutes phylum predominating. When the stable and exacerbation states of COPD disease were compared, no significant differences in the alpha and beta diversity between the disease states were observed. However, during exacerbation state of the disease, the abundance of key phyla had decreased. Analysis of the virome showed a high prevalence of BeAn 58058, a close relative of the smallpox virus, with bacteriophages being the second most prevalent viruses.
When comparing the IS-Pro method to targeted metagenomics, an increased relative abundance of Proteobacteria with the IS-Pro method was observed, which was attributed to known lung pathogens, such as Burkholderia. The IS-Pro method was able to classify more operational taxonomic units (OTUs) to a species level, however, the unclassified OTUs from the IS-Pro method could only be classified to a phylum level.
To conclude, a diverse COPD microbiome was observed, with a virome that was dominated by the BeAn 58058 virus. The COPD disease states showed no variations in terms of diversity, however, the relative abundances of key phyla differed between disease states for the bacterial microbiome. Future studies should focus on longitudinal studies of the sputum microbiome in an African setting as well as functional metatranscriptomics studies with a focus on antibiotic resistance and virulence factors.