Of Mucus and Microbes: The Sticky Issue of Mucin–Microbiome Interactions in Chronic Obstructive Pulmonary Disease

Intervention on factors that contribute to lung function decline in chronic obstructive pulmonary disease (COPD) has long been an important goal. A subset of patients will lose lung function at a faster pace, and addressing this disheartening trajectory is critical (1). The pathobiological differences that drive this trajectory remain unclear, and elucidating the mechanisms and interactions involved should enlighten interventional strategies beyond encouraging smoking cessation.

In addition to the excess production and secretion of mucins, pathologic mucus, reflective of altered biochemical and biophysical properties, is a key feature of obstructive lung diseases, including COPD (2). Most recognized clinically in patients with symptoms of cough and sputum production, the consequences of pathologic mucus are increasingly clear, even in nonsevere disease (3–5). Mucus plugs visualized in chest computed tomography correlate with airflow obstruction independent of emphysema (3) and with all-cause mortality (4). In tobacco-exposed persons at risk for or with nonsevere COPD, higher total mucin concentrations—of MUC5AC, in particular—are linked to clinical markers of COPD and predictive of lung function decline (5). However, pathologic mucus alone may not be fully explanatory. Reported symptoms and clinical outcomes can vary greatly among patients, despite evidence of excess mucus (3, 5), and we still do not fully comprehend what dynamics in the airway milieu buy the unfortunate patient a ticket on the express train of lung function decline.

Clinically, links between culturable bacterial pathogens and severity of airway disease are well recognized. Recent studies have molecularly identified additional members of the airway microbiota associated with COPD morbidity or mortality, including lung function decline (6–8). The postulate that interactions between mucus and microbes are important in driving airway disease progression is, thus, intuitive. However, until now, no study of COPD has integrated the detailed study of pathologic mucus properties and the airway microbiome to discern how their relationships may differ in patients with more rapid lung function decline. In a study from the London COPD cohort in this issue of the Journal, Meldrum and colleagues (pp. 298–310) analyzed longitudinal sputum samples from participants with accelerated FEV₁ decline (n = 30) compared with those with no decline (n = 28) (9). Elegant analyses of mucus biochemical and biophysical properties were performed, and the sputum microbiome was profiled using shotgun metagenomics, an approach that allows for more detailed characterization of member species and functions. Mucus properties clearly became more altered in the accelerated-decline group. This was driven mostly by biochemical features such as the percentage of mucus solids and concentrations of total mucins and of MUC5AC, which displayed large increases in some participants. Biophysical measures of elasticity and viscosity also generally increased at follow-up in the accelerated-decline group. Between-groups differences in sputum bacterial composition were apparent by supervised partial least square-discriminant analysis, which further identified coassociations of Achromobacter and Klebsiella, as well as concentrations of MUC5AC and MUC5B with the accelerated-decline group. When grouped, instead, by the most recent GOLD classification criteria, the participants in Group E (frequent exacerbations regardless of symptom severity) displayed elevated levels of MUC5AC and MUC5B together with representative genera of potential pathogenic organisms, including Haemophilus, Moraxella, and Pseudomonas. By contrast, those with no FEV₁ decline or who belonged to GOLD Group A displayed a lower percentage of mucus solids and MUC5AC with an associated enrichment of Veillonella, Prevotella, Gemella, and Rothia.

Sputum can be a challenging biospecimen but is also of great value, particularly in large cohort studies. If it can be collected and is of acceptable quality, then efforts should be maximized whenever possible to merge adjacent scientific interests that could yield integrated insights into airway disease biology. The study by Meldrum and colleagues is a prime example of this and is furthermore unique in its integration of expert analysis of pathologic mucus properties with the use of shotgun metagenomics, a step up from 16S rRNA gene sequencing, to characterize the microbiome. Further, this approach was applied to the analysis of paired longitudinal samples, which is not always possible for various reasons. The findings of the study are limited by the small sample size in each group, which is due, in part, to sample availabilities from the represented extremes of lung function trajectory in the cohort. Elapsed time between initial and follow-up visits also varied between participants, but it was adjusted for in the analyses. The accelerated-decline group also had a higher proportion of current smokers. In theory, active smoking could be a major contributor to the outcome differences, given its known effects on mucus homeostasis (10). However, the changes in specific mucus measures did not differ between current and former smokers, which supports the notion that additional factors likely contribute to determining the pace of lung function decline. Despite having metagenomic data, the study did not call out specific species of bacteria implicated in the findings, nor did it explore the role of fungi or microbial functions that may further support the progressive loss of lung function. Chronic hypoxia near the epithelium in the vicinity of mucus plugs induces transcriptional changes and mucus hyperconcentration (11). These microenvironment changes can further shape host–pathogen interactions, including microbiome composition and functions (12). Thus, the intriguing findings from this study raise additional questions that merit further research beyond its present scope.

The accelerated-decline group had a higher baseline lung function than the no-decline group (mean FEV₁, 63% vs. 43% predicted), reflecting “room to fall” in the former group. Thus, given this timeframe window into COPD at spirometric GOLD Stage 2 or 3, insights from this study invite the following questions: When, in the course of COPD development, does the mucus–microbiome relationship shift from homeostasis to a pathologic dynamic that may (or may not) be reversible? Does this begin even before COPD is diagnosed clinically by current criteria? A large global registry study of COPD recently reported that lung function decline was highest in pre-COPD and GOLD Stage 1 and 2 groups (13). What are the critical factors that tip this balance such that the homeostatic roles of mucus and microbial mutualism become pathologic entities that, in some patients, potentially drive the express train of lung function decline? With the arrival of new cohort studies focused on early COPD, such as BEACON (14) and SOURCE (15), come also potential opportunities to build on the foundational findings by Meldrum and colleagues.