Insights into respiratory microbiome composition and systemic inflammatory biomarkers of bronchiectasis patients

Abstract

The human microbiomes, including the ones present in the respiratory tract, are described and characterized in an increasing number of studies. However, the composition and the impact of the healthy and/or impaired microbiome on pulmonary health and its interaction with the host tissues remain enigmatic. In chronic airway diseases, bronchiectasis stands out as a progressive condition characterized by microbial colonization and infection. In this study, we aimed to investigate the microbiome of the lower airways and lungs of bronchiectasis patients together with their serum cytokine and chemokine content, and gain novel insights into the pathogenesis of bronchiectasis. The microbiome of 47 patients was analyzed by sequencing of full-length 16S rRNA gene using amplicon sequencing Oxford Nanopore technologies. Their serum inflammatory mediators content was quantified in parallel. Several divergently composed microbiome groups were identified and characterized, the majority of patients displayed one dominant bacterial species, whereas others had a more diverse microbiome. The analysis of systemic immune biomarkers revealed two distinct inflammatory response groups, i.e., low and high response groups, each associated with a specific array of clinical symptoms, microbial composition, and diversity. Moreover, we have identified some microbiome compositions associated with high inflammatory response, i.e., high levels of pro- and anti-inflammatory cytokines, whereas other microbiomes were in correlation with low inflammatory responses. Although bronchiectasis pathogenetic mechanisms remain to be elucidated, it is clear that addressing microbiome composition in the airways is a valuable resource not only for diagnosis but also for personalized disease management.

Importance: The population of microorganisms on/in the human body resides in distinct local microbiomes, including the respiratory microbiome. It remains unclear what defines a healthy and a diseased respiratory microbiome. We investigated the respiratory microbiome in chronic pulmonary infectious disease, i.e., bronchiectasis, and researched correlations between microbiome composition, systemic inflammatory biomarkers, and disease characteristics. The bronchoalveolar microbiome of 47 patients was sequenced, and their serum inflammatory mediators were quantified. The microbiomes were grouped based on their content and diversity. In addition, patients were also grouped into low- and high-response groups according to their inflammatory biomarkers' levels. Certain microbiome compositions, mainly single-species dominated, were associated with high levels of inflammatory cytokines, whereas others correlated with low inflammatory response and remained diverse. We conclude that respiratory microbiome composition is a valuable resource for the diagnostics and personalized management of bronchiectasis, which may include preserving microbiome diversity and introducing possible probiotics.

Keywords: bronchiectasis; lower respiratory tract microbiome; respiratory microbiome; systemic inflammatory biomarkers.

Fig 1

Lung microbiome diversity in bronchiectasis patients. The heatmap illustrates the relative abundance of 41 bacterial species, derived by selecting the seven most abundant species from each sample across diverse samples and grouped into four distinct phyla: Proteobacteria, Firmicutes, Bacteroidetes, and Fusobacteria. Each row symbolizes a unique bacterial species, whereas each column indicates a different sample. Based on clustering indices, an optimal number of 15 clusters was identified, four of which comprised more than two study samples. Below, the Shannon diversity scatterplot depicts the species diversity within each sample.

Fig 2

Analysis of inflammatory response and clinical metrics. (A) Box plots display the relative protein levels of various inflammatory markers between the LRG and HRG. Provided cytokines and chemokines classification based on their role in inflammation: pro-inflammatory (red) and anti-inflammatory (green). Significant differences were observed in several markers, including IFNα, IL-1β, IFNγ, IL-4, IL-17A, IL-12p70, IL-13, MCP-1, TNFα, CD62E, and MIP-1α, all of which were higher in the HRG (P < 0.001). IL-1α and ICAM-1 were also significantly higher in the HRG (P < 0.01 and P < 0.05, respectively). Markers such as IP-10, MIP-1β, and CD62P showed no significant differences between the groups (P = 0.059, P = 0.179, and P = 0.459, respectively). (B) The PCA plot distinguishes between the LRG and HRG based on the inflammatory response, with PC1 (38.8% variance) and PC2 (15.3% variance). The LRG (green) and HRG (red) are clearly separated, indicating significant differences in inflammatory protein expression profiles. (C) Box plots compare clinical metrics between the LRG (green) and HRG (red). (D) The heatmap visualizing relative serum concentrations of inflammatory mediators with color intensity indicating protein level. Significance is indicated as *P < 0.05, **P < 0.01, ***P < 0.001.

Fig 3

Microbiome analysis and inflammatory response. (A) Box plots show the alpha diversity (Shannon index) of the microbiome between the LRG and HRG. The Shannon index is slightly higher in LRG compared to HRG (P = 0.09), indicating a trend toward greater microbial diversity in LRG. (B) Box plots display the relative abundance of Betaproteobacteria and Negativicutes between LRG and HRG. Betaproteobacteria are significantly more abundant in LRG (P < 0.05), whereas Negativicutes are significantly more abundant in HRG (P < 0.05). (C) Box plots show the relative abundance of specific bacterial genera between LRG and HRG. Genera such as Gemella, Neisseria, Prevotella, Selenomonas, and Veillonella are significantly more abundant in LRG than in HRG (P < 0.05 for all). (D) The Sankey diagram provides a visual representation of the relationship between inflammatory response groups (LRG and HRG), microbiome clusters, and alpha-diversity groups (Shannon index >1 in high-alpha-diversity group; Shannon index <1 in low-alpha-diversity group). It effectively shows the distribution of LRG and HRG across different microbiome clusters, highlighting the connections between specific clusters and overall microbial diversity. Significance is indicated as *P < 0.05.

Fig 4

Differential expression of inflammatory markers across microbiome clusters. Box plots depict the relative abundance (%) of inflammatory markers (CD62E, IL-1β, IL-12p70, IL-17A, MIP-1ɑ, and MIP-1β) across major microbiome clusters: Firmicutes, Haemophilus influenzae, and Pseudomonas aeruginosa. Asterisks indicate significant differences between clusters: * represents a P value < 0.05, and ** denotes a P value < 0.01. Notably, MIP-1 beta exhibits a near-significant trend with a P value of 0.069.