Real-World Study on Upper/Lower Respiratory Tract Microbiome Changes in AECOPD and COPD

Background: Microbiome residing in the respiratory tract has emerged as an important player in the etiology and progression of COPD, but results are conicting regarding the features of respiratory tract microbiome in COPD and at exacerbations and it is unknown whether these features differ by ethnicity and geography. To address these questions, we enrolled healthy individuals and patients with COPD, including healthy-COPD pairs from same households, from four geographical regions of Yunan province, representative of different ethnicities and/or environmental exposures. Sputum and oropharyngeal swabs were collected from these healthy individuals and from COPD patients at stable state (COPD) or exacerbations (AECOPD) and subjected to 16S amplicon sequencing. Results: We found that both COPD disease status and region had an impact on alpha-diversity of sputum and oropharyngeal microbiomes, with AECOPD having the lowest microbiome diversity. Shifts in the relative abundance ( ≥ 1.5 fold, adj.p < 0.05) of microbes at healthy, exacerbation and stable COPD. Microbes enriched at exacerbation COPD were primarily Proteobacteria and Firmicutes phylum in upper respiratory tract. In the lower respiratory tract, population-based study did not nd any statistical differential abundance of microbe, however, paired-based study showed phylum of Proteobacteria, Chloroexi, Bacteroidota, Acidobacteriota, Desulfobacterota, Firmicutes and Verrucomicrobiota enriched in exacerbation COPD. Conclusions: This study was the rst attempt to combine the population-based and core-family-based to investigate the microbiome proling in the upper and lower respiratory tract in COPD patients, and specic microbial ora characteristics may inform future study on the pathogenesis or management of COPD patients.


Real-World Study on Upper/Lower Respiratory Tract Microbiome Changes in AECOPD and COPD
Microbes enriched at exacerbation COPD were primarily Proteobacteria and Firmicutes phylum in upper respiratory tract. In the lower respiratory tract, population-based study did not nd any statistical differential abundance of microbe, however, paired-based study showed phylum of Proteobacteria, Chloro exi, Bacteroidota, Acidobacteriota, Desulfobacterota, Firmicutes and Verrucomicrobiota enriched in exacerbation COPD.
Conclusions: This study was the rst attempt to combine the population-based and core-family-based to investigate the microbiome pro ling in the upper and lower respiratory tract in COPD patients, and speci c microbial ora characteristics may inform future study on the pathogenesis or management of COPD patients.

Background
Chronic obstructive pulmonary disease (COPD) is a heterogeneous complex, progressive respiratory disease featured by persistent symptoms, occasional exacerbation episodes, and chronic airway in ammation, the latter of which may lead to pathological changes in the respiratory tract including parenchymal destruction and air ow limitation. The mechanisms underlying overreactive immunity in the lungs of COPD patients are as yet unclear, which are considered to be an interplay between environmental and exogenous factors. Ethnicity may be a factor associated with the risk and outcome of COPD. Studies have found that the prevalence of COPD differs among the black, white, and Asian populations, and the black group seemed to have more severe exacerbations leading to hospitalization and are at increased risk of mortality [1,2]. In addition, geographic region may affect individual's predisposition for COPD, evidenced by the geographic disparity in the prevalence of COPD in China.
Microorganisms are increasingly recognised as an important player in the pathogenesis and progression of COPD. Childhood infection in the respiratory tract has been linked to increased risk for COPD in adulthood [3]. A variety of pathogenic bacteria were isolated in the lower respiratory tract of patients with stable COPD. Even more, pathogen infection was identi ed to account for 50% of exacerbation events.
However, this pathogenic view is being reconsidered as it is now appreciated that even the healthy lower respiratory tract is not sterile but colonized with numerous commensal microbes. Instead of being caused by speci c pathogens, it is hypothesized that dysbiosis of respiratory tract microbiome may be involved in COPD development and exacerbation events. Respiratory tract microbiome in COPD has been the subject of several studies, but results were con icting regarding whether there were marked changes in the respiratory tract microbiome of COPD patients at stable or exacerbation state [4][5][6][7]. this regard and considering the variability of COPD risks by ethnicity and geography, a question that remains to be answered is whether a common respiratory tract microbiome signature featuring COPD exists.
In this study, we aimed to identify the features of respiratory tract microbiome in COPD and reveal the impact of ethnicity and geography on the respiratory tract microbiome. To this aim, we enrolled patients in the core family as the smallest unit in Yunan province of China. The core families have similar genetic backgrounds and living habits, so as to maximize the control of the in uence of these factors on the respiratory tract microbiome. The complex topography in Yunan province gives rise to varied climates, ranging from frigid, temperate, to tropic zone climate. There is considerable ethnical diversity in this province, where people of one ethnicity usually reside in one region and have similar lifestyle. As such, it is anticipated that respiratory tract microbiomes of individuals from different regions of Yunan have been shaped by different genetic and/or environmental factors. In addition, we enrolled healthy-COPD pairs from the same households to control non-disease related in uences on the microbiome.

Study Design
In this study, we took nasopharyngeal swabs or sputum samples from enrolled samples or took both sample types from the same enrolled sample; Sampling was carried out in the same way under the condition of COPD; during the entire enrollment and sampling process, we faithfully carried out enrollment based on the normal clinical procedural. We divided all samples into groups according to their physical condition and sampling type. Each sample type is divided into three groups for pairwise comparison; at the same time, according to the core family, for the same sample type from the same family members, form a paired sample, then performed pairwise comparison. Among the sputum samples, there were only 6 core families including one patient with two COPD stages and corresponding healthy family members. Among the swab samples, such core families were only 10. Therefore, this study did not conduct a three-group pairwise comparison analysis. We incorporated these samples into speci c pairwise comparisons based on their characteristics. The detailed study design and sample enrollment have been shown in Table 1.

DNA extraction and microbiome sequencing
Samples were pre-treated, depending on the clinical sample type. Total DNA was extracted from swab The raw fastq les were trimmed adapter before starting the subsequent analysis. The operational taxonomic units (OTUs) classi cation analysis was done using published 16S microbial analysis with mothur pipeline [9,10]. The raw sequences data were merged into contigs, followed by ltering and trimming of reads based on quality score and several other metrics. The ltered sequences data were aligned to the reference Silva v138 [11]. The chimera sequences were removed using the vsearch algorithm [12]. Finally, the cleaned sequences data were assigned to taxonomy database for classi cation, here we used Bayesian classi er and a mothur pre-formatted Silva v138 taxonomy database as reference taxonomy [11]. The corresponding OTU and species annotations in this study can be found in the supplementary table 2.

Statistical Analyses
Statistical analyses were undertaken using R (v4.0.0). P value less than 0.05 was considered statistically signi cant. Core microbiota was de ned as those found in more than 10 samples and the relative abundance was more than 0.05%. The nal outputs from mothur were exported as count table and related taxonomy table, then import these les and clinical meta table into Phyloseq (v1.32.0) package to generate S4 object for further analysis [13].
Alpha diversity was calculated on the basis of the taxonomy pro le for each sample based on Shannon.
Alpha diversity estimates were computed using Microbiome (v2.  [14] for the microbial ecological association networks analysis with meinshausen-buhlmann's neighborhood selection method, the networks were plotted using ggnet2 (v0.1.0) with setting the minimum degree (nodes) value to 1.
The Phyloseq object data was converted to the relevant DESeqDataSet object with phyloseq_to_deseq2 function. The pair wised OTU differential abundance analysis was done using the DESeq2 (v1.28.1) [15] negative binomial Wald test.

Results
Overview of sputum and oropharyngeal microbiome bacterial families in both sputum and oropharyngeal microbiomes, whereas Porphyromonadaceae and Micrococcaceae were also common in sputum and oropharyngeal microbiomes, respectively (Fig. 1).

Alteration of sputum and oropharyngeal microbiome by COPD and resident regions
Although sputum microbiome vs oropharyngeal swab microbiome were not compared within individuals, between-group comparison showed that sputum samples had slightly higher alpha-diversity than oropharyngeal swabs (Fig. 2a, c). The alpha diversity of oropharyngeal swab microbiome was lowest in AECOPD (AECOPD vs COPD, p < 0.0001; AECOPD vs healthy control, p < 0.01; COPD vs healthy control, p < 0.01), and a similar trend was observed in sputum microbiome (healthy control vs COPD, p < 0.05; AECOPD vs healthy control, p < 0.001). Compared with healthy controls, alpha diversity was increased in oropharyngeal swab microbiome of COPD patients but decreased in their sputum microbiome. Differences in alpha-diversity of microbiome were observed across different geographical locations, with individuals from Lijiang consistently had the least diverse microbiome in both sputum and oropharyngeal swabs (Fig. 2b, d).
The distinction in overall microbiome composition was evaluated by principal coordinate analysis (PCoA) based on weighted UniFrac distance metrices (Fig. 3). The rst and second coordinates captured about 50.3% and 48.3% of variations in the composition of oropharyngeal microbiome and sputum microbiome, respectively. However, both oropharyngeal microbiome and sputum microbiome could not be discriminated by disease group or region. Nevertheless, PERMANOVA showed that both COPD disease status and resident region had an impact on microbiome compositions, suggesting the existence of differences in relative abundances of certain microbial taxa (Supplementary Table 3).

Respiratory Tract Microbiome Taxa Associated With Copd
In order to identify which bacterial clades in the oropharyngeal and sputum microbiome were different in relative abundances between healthy vs COPD stable state vs exacerbations, we performed negative binomial Wald test using DESeq2. A total of 35 OTUs were signi cantly different in relative abundances with a fold change ≥ .1.5 and an adjusted p-value ≤ 0.05 in either group comparisons or withinhousehold pair-wise comparisons ( Table 2, Supplementary Table 4). We considered differences revealed by pair-wise comparisons more likely to re ect disease effects on microbiome. Compared with healthy controls, oropharyngeal microbiome under COPD stable state had relative higher relative abundances of Streptococcus, Actinomyces, Actinobacteria spp., Rothia, and Veillonella. During COPD exacerbations compared with healthy controls, oropharyngeal microbiome consisted of higher relative abundances of Raoultella, Actinobacillus, Enterobacteriaceae spp., and Haemophilus, and sputum microbiome comprised higher relative abundances of Anaerolineaceae spp., Desulfobacterota spp., Anaerolineae spp., Pedosphaeraceae spp., Rhodobacteraceae spp., gammaproteobacterial spp., and alphaproteobacteria spp.. No taxa showed signi cant differences in relative abundances between COPD stable state and exacerbations in pair-wise comparisons using above mentioned cutoffs.

Discussion
Based on respiratory tract microbiome pro les of a cohort of healthy individuals and COPD patients from different geographical regions, representative of diverse ethnicities and environmental exposures, we found that both COPD disease status and geography had an impact on the respiratory tract microbiome. The alterations in respiratory tract microbiome are potentially related to the pathological state or pathogenesis of COPD.
The predominant microbial families in sputum and oropharyngeal swabs are largely the same, indicating a continuum of microbiome composition along the respiratory tract. Previous studies showed that in healthy individuals Bacteroidetes and Firmicutes are two predominant phyla, whereas at the genus level, Prevotella, Veillonella, and Streptococcus are highly abundant in sputum microbiome [16][17][18]. The sputum microbiome is more diverse than the oropharyngeal swab, consistent with the fact that sputum microbiome is a mixture of oral and upper/lower respiratory tract microbiomes.
Reduced diversity of sputum microbiome under COPD stable state compared with healthy state has been reported in previous studies [6]. Moreover, the diversity of sputum microbiome has been found to be further decreased at exacerbations[6]. We found a consistent decreasing pattern in sputum microbiome; however, the oropharyngeal swab microbiome showed an increase in COPD compared with healthy control, which needs to be con rmed in further studies with balanced sample distribution between groups.
Pair-wise comparison of patients and healthy controls who shared the same household help reduce confounding factors, such as diet and genetic factors, and may reveal true microbiome alterations associated with diseases. Identi ed enriched or depleted taxa within COPD respiratory microbiome may underlie COPD pathogenesis or be a re ection of the pathological status. Some of the taxa found to be increased under COPD stable state have been reported to be associated with COPD before, such as Actinomyces, Veillonella, and Rothia [19]. In addition, Actinobacillus was also markedly enriched under COPD stable state, which was discovered as an aggressive pathogen related to periodontitis. Previous studies have shown that the COPD respiratory tract microbiome became more abundant in Proteobacteria bacteria such as Pseudomonas spp. and Haemophilus spp. [20][21][22]. In our study, many bacterial clades under Proteobacteria were observed to be increased in abundance in COPD patients during exacerbations compared with healthy individuals, including Enterobacteriaceae spp., Gammaproteobacteria spp., Comamonadaceae spp., Alphaproteobacteria spp., Rhodobacteraceae spp., Raoultella, and Haemophilus. Raoultella is considered as opportunistic pathogens that may result in pneumonia in individuals with abnormal immunity [23]. Haemophilus contains the well-known opportunistic pathogen Haemophilus in uenzae, and respiratory infection by this bacterium causes airway in ammation and is linked to respiratory diseases including COPD [24]. Though still a conjecture, these clades may contribute to episodic exacerbations in COPD patients.
This study was initially based on the results of the population study, compared with the results generated from the core-family study, and the follow-up microbe's biological function veri cation study can be focused on the identical microbes. However, there are still many limitations. First, sample sizes across different groups were imbalanced, due to di culty in collecting respiratory microbiome samples. Second, the sample sizes for intra-household pairwise comparisons of healthy control vs COPD and healthy control vs AECOPD were quite limited, which still help eliminate confounding factors related to microbiome compositions. Third, few individuals had both sputum and oropharyngeal swab samples available, not allowing for comparing upper and lower microbiome within individuals. Fourth, due to small sample size at one geographical location, whether patients of different ethnics/locations in healthy or COPD respiratory tract microbiome between cannot be investigated.

Conclusions
This study was the rst attempt to combine the population-based and core-family-based to investigate the microbiome changes in the upper and lower respiratory tract in COPD patients. COPD is associated with speci c changes in upper and lower respiratory tract microbiomes, whereas exacerbation events confer additional alterations in respiratory tract microbiome compositions. These speci c microbiome changes may inform future study on the pathogenesis or management of COPD patients.  Figure 1 Flow chart of the study sample enrollment. Detailed comparisons between groups are shown in the Table  1.    Upper respiratory tract microbial community structure in different groups. Here, we only plotted those microbes with degree > 1 in groups, including Healthy, COPD and AECOPD groups, which were conclusively associated with certain clinical status. Lower respiratory tract microbial community structure in different groups. Here, we only plotted those microbes with degree > 1 in groups, including Healthy, COPD and AECOPD groups, which were conclusively associated with certain clinical status.

Figure 7
Identical and unique core microbiota in different groups with microbial community structure's degree > 1. Dash line represented core microbiome pro le from upper respiratory tract, and solid line for lower respiratory tract; The different color represented different groups, green for Healthy group, yellow for COPD group and pink for AECOPD group.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.