Comprehensive analysis of the gut microbiota in patients with chronic obstructive pulmonary disease of varying severity-A prospective, observational study

Abstract

Accumulating evidence has highlighted the influence of the gut microbiota on lung immunity. We examined the effects of changes in intestinal microecology on the development of Chronic Obstructive Pulmonary Disease (COPD) and identified microbial biomarkers for acute exacerbations of COPD (AECOPD). Fecal samples were collected from 30 patients with stable COPD, 30 patients with AECOPD, and 10 healthy individuals. Fecal microbiological profiles were analyzed using 16S rRNA gene sequencing. The results showed a distinct difference in the bacterial community composition between the AECOPD, COPD, and healthy control groups. The COPD and AECOPD groups had higher levels of Firmicutes but lower levels of Bacteroidetes compared to the healthy control group at the phylum level. At the genus level, there was an increased abundance of Lachnoclostridium, Alistipes, Streptococcus, and Prevotella in COPD and AECOPD patients. Increasing levels of Lachnoclostridium and Prevotella may indicate an acute exacerbation of COPD. This study identified specific microbial biomarkers associated with AECOPD and characterized the composition of gut microbiota in patients with AECOPD.

Keywords: AECOPD; Chronic obstructive pulmonary disease; Gut microbiota.

Fig. 1

Flow chart of the study.

Fig. 2

Gut microbiota composition from the healthy, stable COPD patients, and AECOPD patients in Genus (A) and Species (B) level. Community heatmap analysis from the healthy, stable COPD patients, and AECOPD patients in Genus (C) and Species (D) level.

Fig. 3

Gut microbiota alterations in the healthy control, stable COPD patients and AECOPD patients. (A) PCoA on Genus level; (B) Box-plot of PCoA; (C) PLS-DA analysis of the three groups; (D) Typing analysis on Species level.

Fig. 4

Specific microbial signatures of stable COPD patients and AECOPD patients. (A) Inter-group difference significance test.(B) Lefse multilevel species difference analysis.

Fig. 5

Relationship between differences in Gut Microbiota and Clinical Factors by group. (A) Canonical Correlation Analysis and distance-based redundancy analysis. CCA analysis(B) and db-RDA (C) analysis was used to determine the important clinical drivers affecting the distribution of the samples with a view to finding key species associated with the disease.

Fig. 6

Correlation Analysis and Model Prediction Based on Gut Microbiome Sequencing Results Using Random Forest. (A) The distribution and classification of Random Forest on Genus level; (B)Barplot of variable importance on Genus level. (C) Model evaluation by using top important features. (D) The ROC curve on Genus level.