Host–Microbial Interactions: Idiopathic Pulmonary Fibrosis in Technicolor

Idiopathic pulmonary fibrosis (IPF) is a complex disease that is characterized by progressive declines in lung mechanics and gas exchange, which ultimately lead to respiratory failure and death. The complexity of IPF is exemplified by genetic and environmental contributions that are thought to culminate in recurrent “microinjuries” to alveolar epithelial cells, which lead to epithelial exhaustion and chronic disrepair. To date, the origins of this unremitting injury have remained elusive. Several recent studies proposed a role for impaired or dysregulated host defense and immune signaling in IPF pathogenesis (1–7). Exogenous stimuli, including microbes, are plausible environmental factors that contribute to organ fibrosis (8, 9). Specific microbiota in the lungs of patients with IPF predict disease progression, and the fibrotic environment of the IPF lung harbors a significantly greater burden of bacteria than chronic obstructive pulmonary disease and healthy lungs (10, 11). The stage is therefore set to determine whether the microbes residing in the lower airways conceivably drive recurrent injury and disrepair in the lungs of patients with IPF, particularly in the context of dysregulated host defense.

It is in this context that Molyneaux and colleagues (pp. 1640–1650) in this issue of the Journal (12) build on their previous novel contributions to the field of respiratory microbiota in IPF (11). This group applied 16S ribosomal RNA gene sequencing to baseline-acquired bronchoalveolar lavage (BAL) fluid from 60 patients with IPF and 20 matched control subjects to describe the respiratory microbiome. They reported significant differences in the total abundance and relative abundance of certain microbial species, including a Haemophilus sp., a Neisseria sp., a Streptococcus sp., and a Veillonella sp. compared with age-matched control subjects. Although these results are not novel to the field, the authors also acquired corresponding peripheral blood samples prospectively at 1, 3, 6, and 12 months from patients with IPF and control subjects to explore potential associated host responses. More than 1,300 transcript clusters were differentially expressed in IPF compared with controls, and the most enriched biological processes in these clusters were “host defense” and “stress” related (based on gene ontology). Host gene expression in peripheral blood cells was examined in a weighted gene coexpression network analysis, and five gene clusters (or modules) were reported. Gene expression modules, assigned arbitrary colors, which predicted poor prognosis, were enriched for innate defense transcripts, and were associated with peripheral neutrophil counts (brown and blue). However, a module that predicted favorable outcomes (turquoise) was enriched for lymphocytic transcripts and was associated with increased lymphocyte counts in the peripheral blood. Increased expression of the blue module transcripts was associated with worse survival, greater decline in lung function, a higher bacterial burden, and lower abundance of a Neisseria sp. and BAL neutrophilia in patients with IPF. Unlike previous work (10), no association was found between this module and disease progression and/or Staphylococcus or Streptococcus loads. This blue module contained several overexpressed host defense genes and enriched biological processes, including “defense response,” “response to bacterium,” and “immune response.” The green module was strongly associated with BAL neutrophilia, peripheral blood neutrophilia, and a higher abundance of Veillonella operational taxonomic units in BAL. The most enriched biological process within the green module was “response to bacterium.” The turquoise module, enriched for T-cell–associated transcripts, was associated with longer survival, reduced pulmonary function decline, and death, a finding that agreed with the observation that T-cell costimulatory protein expression in IPF correlates with prognosis (5). However, there was no association between the mucin 5B rs35705950 and toll-interacting protein rs3750920 and rs5743890 single-nucleotide polymorphisms and host gene expression in the current study. The other two modules reported interesting findings, but contributed less to the main hypothesis of the paper.

As a notable strength, this was the first study to postulate and integrate links between host gene transcription and respiratory microbiota in a well-characterized cohort of patients with IPF with serial follow-up and acquisition of large volumes of data. This cohort was enriched for disease progression, with 24 deaths on follow-up and another 13 patients with IPF who experienced declines in lung function that met the standard criteria for progression. Although the authors accounted for comorbidities within their statistical modeling and excluded patients with acute infection, there were caveats and limitations to this study. These included: patient numbers were limited and without validation; integration of large data sets was fraught with difficulty and might be subject to overinterpretation; and lack of longitudinal surveillance data on respiratory microbiota might weaken the results of the study.

The current paradigm of IPF centers on the alveolar epithelium, which is subject to recurrent unknown microinjuries. In conjunction with recent clinical observations of reduced mortality in patients treated with antimicrobials (13) and the known deleterious clinical outcomes of patients with IPF who are treated with immunosuppression (14), the findings of this study are provocative. Although this study provides further evidence for dysregulated host defense responses in IPF and novel associations between host response and lower airway microbiota, key questions remain. Could it be that altered microbial cues, in the setting of dysregulated alveolar epithelial repair, drive recurrent pattern recognition receptor activation and signaling? Could the fibrotic lung environment be responsible for physiological changes that promote dysbiosis, which, in turn, drives a dysregulated host response and bystander injury? Could a dysregulated host response predispose to recurrent pulmonary infection, which generates dysbiotic lower airways and contributes to disease progression through yet unknown mechanisms? Specifically, it is perplexing to note that several potent antimicrobial factors were up-regulated in IPF concomitant with dysbiosis in this disease, which suggests that these factors disproportionately target probiotic species, thereby favoring pathogenic species. Alternatively, it remains plausible that inhaled particulates and/or refluxed gastric acid might also contribute to epithelial injury in the lung. Further work is needed to address these exciting questions.

IPF remains an irreversible and devastating disease that portends a dismal prognosis. This intriguing work by Molyneaux and colleagues adds to other seminal studies and sheds additional light on the putative role of the respiratory microbiota in IPF pathogenesis. If future work strengthens the putative mechanistic links between respiratory dysbiosis and disease progression in IPF, then we may usher in an era of precision medicine for the respiratory microbiome.