Background
Chronic obstructive pulmonary disease (COPD), one of the leading causes of morbidity and mortality worldwide, is characterized by irreversible airflow limitation due to airway and/or alveolar abnormalities [
1]. Acute exacerbation of COPD (AECOPD) is defined as an acute worsening of respiratory symptoms, such as increased cough, purulent sputum production, and dyspnoea, which requires a change in treatment [
2]. Frequent acute exacerbations lead to an increase in mortality as well as socioeconomic loss and health expenditures [
3].
Bacteria-induced respiratory infection is the major cause of acute exacerbations of COPD [
4]. Alterations in the airway microbiome are associated with decreased lung function and enhanced airway inflammation [
5,
6]. Clinic indices such as spirometry indices are the indicators of the severity of airflow limitation and key to the diagnosis of COPD [
7]. Inflammatory biomarkers can indicate bacterial exacerbation in COPD patients [
8]. These clinical indices can be utilized to accurately profile disease severity and prognosis in COPD patients and may be influenced by microorganisms [
9‐
11]. Recent evidence shows that different bacterial taxa identified in sputum samples from AECOPD patients are associated with different clinical outcomes [
12]. However, the manner in which these detected bacteria are involved in pathogenesis and how they ultimately affect the clinical indices of the host remain unclear. Thus, it is necessary to investigate the correlation between the relating microbiota composition and the pathogenesis of acute exacerbation and clinical indices [
13].
Traditional culture techniques have significant limitations regarding unculturable bacteria, as well as a poor sensitivity to detecting low-abundance bacteria [
14]. In some situations, which mainly occurs for patients chronically colonized by
Pseudomonas aeruginosa, chronic colonizing bacteria (rather than pathogenic bacteria) grow easily from the analysed sample by culturing, which can interfere with the potential cause of the exacerbations [
15]. In recent years, advances in diagnostic technology, such as the 16S ribosomal RNA (rRNA) gene sequencing, have now allowed for the assessment of microbiome features with unprecedented personalization and precision, thus demonstrating alterations in the composition and relative abundance of bacteria related to COPD patients [
16‐
18]. However, the exact relationship between changes in the sputum microbiome during AECOPD and clinical indices has not yet been elucidated by sequencing the 16S rRNA gene.
In this study, we investigated the correlation between key bacteria and certain clinical indices in COPD exacerbations by conducting 16S rRNA gene sequencing. We revealed that Streptococcus is the most dominant genus of AECOPD and found that Veillonella and Staphylococcus correlated with FEV1/FVC and C-reactive protein, respectively. These bacteria may serve as potential biomarkers and therapeutic targets for AECOPD.
Discussion
In this study, we investigated the characteristics of sputum microbiota in patients with acute exacerbation and other disease states of COPD patients via cross-sectional observation by using high-throughput 16S rRNA sequencing. It was found that the lung microbiome of AECOPD and recovery patients demonstrated lower bacterial richness, as well as significantly different compositions, compared to those of the stable COPD and healthy control group subjects. We also found alterations in metabolic pathways, and common clinical indices were associated with AECOPD microbial communities. Collectively, these differences in our findings indicate that microbial dysbiosis may contribute to AECOPD and highlight its correlation to clinical indices, which could be considered microbial biomarkers used in potential targets for therapeutic interventions.
Our results demonstrated that microbiome diversity was reduced in AECOPD patients, compared with stable COPD patients and healthy controls. Samples collected from the recovery group had the lowest alpha diversity measures, which may likely be due to antibiotic exposure during clinical treatments. A large longitudinal study showed an overall reduction in microbial diversity during COPD exacerbations compared to samples from stable patients [
25]. These results were consistent with the present study, thus suggesting that microbial diversity may be a biological indicator of AECOPD.
In our study, the most dominant phylum in sputum samples of AECOPD patients was Proteobacteria, which was consistent with the previous reports that the occurrence of AECOPD was related to the reduction in microbiome diversity and increased proportion of Proteobacteria [
25,
26]. Proteobacteria and its members such as
Haemophilus were associated with the onset of AECOPD [
27]. In the present study, the most dominant phylum in sputum samples from the stable COPD group was Firmicutes, consistent with previous findings [
28,
29].
At the genus level, the top 5 dominant genera present in all sputum samples were
Streptococcus,
Neisseria,
Prevotellaceae,
Haemophilus, and
Veillonella, which has been reported in a previous study [
28] except
Prevotellaceae. In the present study,
Streptococcus and
Prevotellaceae were the most dominant genus in AECOPD and stable COPD groups respectively.
Streptococcus has been reported as the most common genus among all samples in COPD patients [
29,
30]. Currently,
Prevotellaceae and
Streptococcus, which both belong to the Firmicutes, are considered members of the core pulmonary microbiome [
31].
Interestingly, compared to the stable COPD group, the AECOPD group had a higher level of Actinobacteria and the associated genera
Corynebacteriaceae and
Rothia. The AECOPD group also had a higher level of Proteobacteria and the associated genera
Stenotrophomonas and
Haemophilus, which are common COPD-related pathogens reported in the previous study [
32,
33]. The level of Bacteroidetes and the associated genus
Prevotella and
Porphyromonas were markedly lower in the AECOPD group than in the stable COPD and healthy controls groups. It has been reported that the proportion of
Prevotella of Bacteroidetes in the airways of asthma patients was lower than that of the normal population [
34]. These results may indicate that alteration of normal flora distribution may lead to enhanced inflammation and increased exacerbation risk.
Functional prediction showed that the transporters, peptidases, purine/pyrimidine, and nucleotide metabolic pathways altered in the AECOPD group. For example, ABC transporters and the secretion system were enriched in AECOPD groups compared to the controls. ABC transporters and their role in substrate transport across the bacterial membrane have been associated with antibiotic resistance [
35]. The bacterial secretion system contributes to secrete virulence factors for host invasion [
36]. These pathways are important for the survival of pathogenic bacteria. Furthermore, AECOPD groups exhibited decreased glycan biosynthesis and metabolism relative to the stable COPD group and controls. Similar to previous findings, impaired glucose metabolism was also found in COPD patients [
37]. The above results further indicate that the change in microbiome function may influence the occurrence or development of AECOPD.
In this study, our results indicated that the sputum microbiome was associated with disease indices.
Veillonella was found to have a significant positive association with FEV1% (
p < 0.05) and FEV1/FVC (
p < 0.001). Additionally,
Staphylococcus had a highly significant correlation with the inflammatory index CRP (
p < 0.01). The proportion of
Veillonella was slightly decreased in the AECOPD group (6.37%) when compared to that of the stable COPD group (7.38%) and healthy controls (9.00%) in our cohort. Simultaneously, the proportion of
Staphylococcus was slightly increased in the AECOPD group (0.79%) compared to that of the stable COPD group (0.03%) and healthy controls (0.04%) in our cohort. Interestingly, Filho et al. found that a higher abundance of
Staphylococcus was associated with decreased 1-year mortality while the survivors were abundant in
Veillonella [
12]. The results of our study may explain the reason for this result.
Veillonella is a gram-negative anaerobic bacterium that belongs to normal oral commensals [
38].
Veillonella species were observed to be greater in the healthy populations than in the COPD groups, thus suggesting a beneficial role for symbiosis in health and/or disease states [
39].
Staphylococcus, a group of gram-positive pathogens that cause infections in the lungs [
40], has been shown to result in the need for more antibiotics and longer hospitalizations [
41].
Staphylococcus aureus in particular has been reported to directly trigger the formation of neutrophil extracellular traps (NETs), which may consequently influence the cycle of inflammation in COPD [
42]. Taken together, these results indicate that
Veillonella and
Staphylococcus may be involved in COPD pathogenesis through their effects on lung function and local inflammatory responses, respectively. AECOPD patients may suffer a lung microecological imbalance, thereby aggravating the inflammatory responses and airflow limitations, which eventually increases the risk of readmission and mortality. Therefore, the sputum microbiome may be used to identify the clinical outcome and prognosis of COPD patients.
There were several limitations to our study. First, samples were cross-sectional and collected from different patients with different disease states, and the interference of confounding factors was unavoidable. Second, in addition to bacteria, viruses and fungi are also involved in the pathogenesis of acute exacerbations of COPD [
33]. However, due to the lack of these data, we did not take them into account in the present study. Finally, our results were mainly restricted to the urban area of Beijing. Future large-scale multicentre studies with different biogeographical backgrounds are required to validate our findings.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.