Nasopharyngeal microbiota is influenced by agricultural air pollution in individuals with and without COPD

Abstract

Respiratory health in chronic obstructive pulmonary disease (COPD) is influenced by environmental factors such as air pollution, yet the role of the airway microbiota in this relationship remains unclear. We investigated the association between exposure to air pollution from livestock farms and the nasopharyngeal microbiota in individuals with COPD compared to healthy control subjects. The study included nasopharyngeal swabs from 186 currently non-smoking participants in the Netherlands, including 65 individuals with COPD and 121 without from a regional rural cohort. Additionally, 116 individuals from a population-wide cohort were included as national controls. Samples were taken at three time points over 12 weeks. The nasopharyngeal microbiota was studied using 16 S rRNA gene-based sequencing for all baseline samples and a random selection of 6-weeks and 12-weeks samples. Dispersion models were used to determine the average concentrations of livestock-related PM₁₀, endotoxin, and ammonia at the participants' home addresses. Individuals with COPD had a higher absolute abundance of anaerobic bacteria, such as Peptoniphilus, Anaerococcus, Finegoldia magna, and Prevotella. Importantly, residential exposure to ammonia was identified as the most important driver of the microbial community composition, explaining 6.6% of the variation in nasopharyngeal microbiota in individuals with COPD. Higher ammonia concentrations were associated with decreased levels of key commensals and increased abundance of anaerobic bacteria. Furthermore, individuals living in areas with high livestock density exhibited greater microbial diversity compared to the broader national population. The study highlights the influence of residential exposure to livestock-related air pollution, particularly ammonia, on nasopharyngeal microbiota composition in individuals with COPD. Our findings suggest that environmental factors significantly impact microbial communities and underscore the potential role of anaerobic bacteria in COPD pathology. Future research should further investigate the mechanisms by which environmental air pollutants affect microbial communities and explore potential interventions to mitigate their effects on respiratory health.

Keywords: 16S rRNA amplicon sequencing; Ammonia exposure; Anaerobic bacteria; COPD; Livestock-related air pollution; Nasopharyngeal microbiota.

Fig. 1

Stability of the nasopharyngeal microbiota over time. Bray-Curtis dissimilarities between samples collected at baseline (T0), 6 (T1) and 12 weeks (T2) “within” and “between” individuals. Boxplots represent the 25th and 75th percentiles (lower and upper boundaries of boxes, respectively), the median (middle horizontal line), and measurements that fall within 1.5 times the interquartile range (IQR; distance between the 25th and 75th percentiles; whiskers). Mean Bray-Curtis dissimilarities are indicated by the asterisks. P-values determined by implementing the Wilcoxon signed-rank test. PERMANOVA analysis was used to determine the inter-individual and temporal variation.

Fig. 2

Beta and alpha diversity of the nasopharyngeal microbiota of COPD cases and controls living in proximity to livestock farms. (A) Explained variance of COPD status, livestock exposure-related characteristics and population (individuals living in a livestock dense area – VGO controls vs. the broader national population – Pienter3 controls) on nasopharyngeal microbial community composition, determined by multivariable PERMANOVA analyses based on Bray–Curtis dissimilarities with Hellinger transformation and Aitchison dissimilarities. (B) Linear regression results showing the relationship of bacterial density, observed ASVs and Shannon index (outcome) with COPD status, livestock exposure-related characteristics and population. Models were adjusted for season, age, smoking status, gender, education, atopy and antibiotics use. Coloured by group (VGO overall, VGO cases, VGO controls and Pienter3 controls). q-values were calculated using the Benjamini-Hochberg method to correct for multiple testing.

Fig. 3

Differential abundance of the most abundant ASVs associated with COPD and livestock farming. Association between the absolute abundance of the 29 most abundant ASVs (> 1% relative abundance in at least 5% of the samples) and (A) COPD status (VGO cases vs. controls), (B) more than 2 farms within 500 m of residence, (C) high NH₃ exposure, and (D) population (VGO controls vs. Pienter3 controls) as determined by ANCOM-BC2 - A and MaAsLin2 - M. The plot shows all the ASVs with at least one significant comparison. The colour corresponds to the direction of the association (pink for positive, blue for negative). Models were adjusted for season, age, smoking, gender, education, atopy and antibiotics use. q ≤ 0.25; *, q ≤ 0.1; **, q ≤ 0.05; ***. q-values were calculated using the Benjamini-Hochberg method to correct for multiple testing.