Lung Microbiome in Autoimmune-Associated Interstitial Lung Disease

The lung microbiome is a diverse mucosal environment that has been shown to be implicated in the pathogenesis of various chronic lung diseases including insterstitial lung diseases (ILD) such as idiopathic pulmonary fibrosis (IPF). ILD is a well-established manifestation of several types of autoimmune diseases. This review will highlight recent work exploring the role of the lung microbiome in the pathogenesis of autoimmune-related ILD.

Key points

•

The microbiome of oropharyngeal and gastrointestinal tracts has been implicated in the pathogenesis of autoimmune diseases although the role of the lung microbiome is unknown.
•

Interstitial lung disease (ILD) is one of the most common manifestations of autoimmune diseases although the pathogenesis remains poorly understood and more reliable biomarkers are needed.
•

Prior studies in idiopathic pulmonary fibrosis have shown the association of lung microbiome dysbiosis with lung disease progression, exacerbations, and mortality.
•

There is emerging evidence that suggests the role of microbial dysbiosis of the lung in driving the development of ILD in various autoimmune diseases.

Abbreviations

ASS	antisynthetase syndrome
BALf	bronchoalveolar lavage fluid
CAP	community acquired pneumonia
COPD	chronic obstructive pulmonary disease
CTD	connective tissue disease
FT	fecal transplantation
GI	gastrointestinal
ILD	interstitial lung disease
IPF	idiopathic pulmonary fibrosis
IM	inflammatory myopathies
PPI	proton pump inhibitor
RA	rheumatoid arthritis
rRNA	ribosomal RNA
SjS	Sjogren’s syndrome
SSc	systemic sclerosis
UIP	usual interstitial pneumonia

Introduction

Interstitial lung disease (ILD) is one of the most common systemic manifestations of connective tissue diseases (CTDs) and when present is associated with increased morbidity and mortality. ^, The pathogenesis of lung-related manifestations such as ILD are likely multifactorial and the mechanisms driving the onset and progression of disease are poorly understood. Enhanced biomarkers are needed to better understand the natural history of ILD to allow for optimal screening and treatment strategies. A potential candidate pathway of interest is the lung microbiome. With the use of culture-independent molecular techniques, such as 16s ribosomal RNA (rRNA) gene sequencing and metagenomics, analysis of bronchoalveolar lavage fluid (BALf) has instilled a clear transformation of our understanding of the lung microbiome. What was once thought to be sterile, the lung microbiome is in fact a distinctly diverse and highly dynamic microenvironment influenced by microbial and host interactions. The most predominant organisms in the healthy lung include Prevotella , Streptococcus , Veronococcus , and Haemophilus species. In comparison to the gastrointestinal (GI) tract, the respiratory microbiome has a nearly 3-fold decrease in overall biomass, ^, although the oropharyngeal space and upper airways are known to have a higher abundance and overall biodiversity compared with the distal airways. The homeostasis of the lung microenvironment is thought to be controlled through 3 distinct mechanisms: (1) immigration from the oropharyngeal tract to the distal airways, (2) replication within the origin of the lung, and (3) emigration primarily driven through host mucociliary defense mechanisms. This paradigm can be heavily impacted by several factors including genetics, environmental exposures, infections, and disease state of the lung architecture. Other mucosal environments can impact the homeostasis of the lung microenvironment as well. Microbial dysbiosis within the GI tract can result in the production of immunomodulatory metabolites and inflammatory signals such as short fatty chain acids which are thought to drive pathology in tissue within a separate environment like the lung highlighting the principle of the gut-lung axis. It remains to be seen how these metabolites are directly contributing to immunomodulatory elements of the lung or have a secondary impact on microbial dysbiosis.

The majority of prior work studying the lung microbiome in chronic lung diseases has focused on chronic obstructive pulmonary disease (COPD), cystic fibrosis, and bronchiectasis. Over the past decade, there has been increased interest in exploring these mechanisms in ILD, specifically idiopathic pulmonary fibrosis (IPF) which is defined by the fibrotic subtype of ILD known as usual interstitial pneumonia (UIP). It is well established that in IPF the lung microbiome is characterized by increased bacterial burden and a less diverse microbial community which is associated with lung disease severity, progression, and mortality. ^, Decreased microbial diversity is also associated with the production of lung-derived inflammatory cytokines. The exact mechanism by which lung microbiome dysbiosis contributes to lung fibrosis in IPF is postulated to be driven through the initiation of a local inflammatory response either through molecular mimicry or direct stimulation of immunologic pathways such as the T help 17 cellular (Th17) pathway.

The microbiome has been implicated in the pathogenesis of several autoimmune diseases although the majority of work has focused on the GI tract. Given the prior work in IPF, this review will highlight more recent studies that have begun investigating the role of the lung microbiome in autoimmune diseases and the associated lung-based manifestations.

Disease-specific alterations of the lung microbiome

Rheumatoid Arthritis

Rheumatoid arthritis (RA) is associated with the presence of various autoantibodies, namely rheumatoid factor and anticyclic citrullinated peptide antibodies, and the primary clinical manifestation is a polyarticular inflammatory arthritis. However, it is clear that the lung has a unique role in the pathogenesis of RA. Multiple studies support the lung as an originating site of inflammation and autoantibody production in the preclinical phase. Additionally, it is clear that the lung can be a downstream target of the dysregulated immune response in RA. RA-associated ILD is clinically apparent in up to 10% of RA patients ^, and importantly, it is associated with substantial morbidity and mortality. ^,

Several prior studies support the role of the oropharyngeal and GI microbiome as it relates to the pathogenesis of RA with and without ILD. Periodontal disease is known to be increased in patients with RA and specific oral bacteria have been strongly implicated with the pathogenesis of RA, specifically Porphyromonas gingivalis and Aggregatibacter actinomycetemcomitans through mechanisms including intrinsic bacteria-mediated citrullination. ^,

There are very few studies that have investigated the lung microbiome in RA-ILD. Scher and colleagues compared BALf of healthy controls compared with individuals with early RA without prior exposure to immunosuppressive therapies. While no RA patients in this study carried an established diagnosis of ILD, 60% were noted to have abnormalities on chest imaging as defined by lung nodules, opacities, fibrosis, or emphysema. The authors identified decreased alpha diversity and bacterial burden in RA patients compared with controls, and they found a microbial community that was similar to a separate cohort of patients with clinically established sarcoidosis. Additionally, the abundance of several taxa was found to have associations with articular disease activity, systemic autoantibody levels, and BALf cellular profiles.

RA-ILD is unique in that a majority of patients exhibit a UIP pattern of ILD as opposed to other autoimmune-related ILDs which more commonly manifest as nonspecific interstitial pneumonia or organizing pneumonia. As stated previously, UIP is characteristic of IPF and while RA-ILD and IPF are thought to be separate disease entities, they share many overlapping clinical and epidemiologic features including shared genetic risk factors. The presence of the minor allele (T) at rs35705950 within the promoter region of the mucin-5B ( MUC5B ) gene confers one of the strongest risk factors of developing IPF and ILD in RA. ^, MUC5B is the predominant mucin expressed in the distal airways, and prior studies have shown that the presence of the at-risk minor allele correlates with increased mucin production. A prior study by Molyneaux and colleagues found that in a group of IPF patients, the at-risk allele was associated with decreased bacterial burden. Future studies are needed to better understand the complex interplay between RA-related risk factors and the subsequent effects on the lung microbiome and development of lung-based manifestations.

Inflammatory Myositis

Inflammatory myopathies (IMs) are a heterogenous group of autoimmune diseases highlighted by inflammation of the proximal skeletal musculature in addition to other organs including the joints, gut, heart, and lungs. ILD is a well-established manifestation, and it is thought to develop in 40% to 100% of patients with a higher incidence observed in specific disease phenotypes, such as those with antisynthetase syndrome (ASS) and amyopathic disease, or in the presence of various myositis-related autoantibodies including anti-Jo-1, anti-Ro-52, and anti-MDA5. It is well established that there is a significant association with seasonality and disease onset of dermatomyositis, particularly in anti-MDA5 positive individuals who often develop a rapidly progressive ILD associated with substantial mortality at disease onset. ^, In a juvenile dermatomyositis cohort, 71% of patients had signs or symptoms suggestive of an infection within 6 months of the initial diagnosis. The majority (80%) of infections were felt to be respiratory in nature. This would strongly suggest the potential role of respiratory infections as triggers to the onset of disease.

There are several studies that have explored differences in the lung microbiome in IM-associated ILD. Quintero-Puerta and colleagues collected BALf of patients with ASS and compared the lung microbiome in those with and without anti-Jo-1 autoantibodies. The most common genera identified across both groups were Streptococcus followed by Sediminibacterium and Veillonella. While there were no differences in alpha diversity, beta diversity was significantly different between groups. Additionally, Jo-1 positive individuals were found to have a significantly lower abundance of Veillonella species which itself was associated with differences in BALf cellular profiles.

Zhang and colleagues compared BALf of IM-ILD patients with CTD-ILD as well as patients with active community acquired pneumonia (CAP). The highest abundance genus level organism in IM-ILD was Prevotella . In comparison to CAP, IM-ILD exhibited significantly higher abundance of Pseudomonas and Corynebacterium at the genus level and Pseudomonas aeruginosa at the species level which was also associated with increased local production of krebs von den Lungen-6 (KL-6), a well-established biomarker associated with ILD severity and progression, as well as pro-inflammatory cytokines such as interleukin 2 (IL-2) and interleukin 8 (IL-8). ^,

Lastly, Lou and colleagues used 16s rRNA sequencing to compare BALf of IM-ILD patients with healthy controls. Investigators found increased bacterial burden in IM-ILD and genus level taxa most discriminatory of IM-ILD included Corynebacterium, Aeromonas , and Achromobacter . Ongoing future studies will be needed to better understand the mechanistic role of the lung microbiome in both the development and progression of ILD in IM.

Systemic Sclerosis

Systemic sclerosis (SSc) is a rare autoimmune disease that primarily causes enhanced tissue fibrosis and vasculopathic changes of the skin as well as other internal organs including the lung. ILD is one of the most common manifestations in SSc with the prevalence ranging between 50% and 90% depending on disease phenotype. ILD has a significant impact on a patient’s quality of life and is associated with increased morbidity and mortality ^, despite the use of various immunosuppressive and antifibrotic therapies. While there are very few prior studies evaluating both the upper airway and lung microbiome in SSc-associated ILD, several studies have shown the connection of GI microbial dysbiosis specifically with ILD in SSc which can better inform future studies focused on understanding lung-based manifestations.

Prior studies have shown a decrease in GI commensal bacteria in SSc patients compared with age-matched controls in addition to distinct differences in beta diversity when comparing SSc individuals with and without ILD. ^, Fecal calprotectin levels, which could be reflective of secondary inflammatory changes from microbial dysbiosis, in a large cohort of SSc patients were shown to be independently associated with the risk of developing ILD. Fretheim and colleagues completed a double blind randomized control trial evaluating changes in the GI microbiome after fecal transplantation (FT) in SSc. While the primary outcomes were clinical efficacy and safety, exploratory secondary endpoint analyses showed a lower decline in % diffusion capacity of carbon monoxide (DLCO) in the FT group compared with controls over the study period of 16 weeks.

Finally, SSc is unique compared with other forms of CTD given the high prevalence of upper GI dysmotility. This can lead to recurrent microaspiration which is known to be associated with the development of IPF and can contribute to translocation of both oropharyngeal and GI microorganisms into the distal airways resulting in more distal airway dysbiosis.

Sjogren’s Syndrome

The primary manifestation of Sjogren’s syndrome (SjS) is autoimmune-mediated damage of various exocrine tissues, including the lacrimal, parotid, and minor salivary glands, resulting in profound xerophthalmia and xerostomia. It is well established that these inflammatory changes of the salivary glands in SjS play a significant role in altering the surrounding microenvironment particularly given the increased risk of dental carries and periodontal disease. ^, Decreased flow of saliva has been shown to heavily impact the makeup of the oral microbiome, although there is conflicting data on differences in the oral microbiome of SjS patients compared with controls in part due to variability of salivary gland dysfunction among patients within cohorts. Several studies have investigated SjS compared with non-SjS patients with sicca symptoms and found no difference in the microbiome, whereas others found distinct taxa differences between groups which are associated with SjS-related disease activity. ^, Additionally, in a group of SjS patients treated with hydroxychloroquine, the post-treatment oral microbiome was significantly different than the pretreatment composition but did not normalize to that of controls. Interestingly, in those who were identified as hydroxychloroquine responders, the post-treatment microenvironment was more similar to controls compared with hydroxychloroquine nonresponders.

Extraglandular manifestations are common including ILD which has been reported to occur in roughly 10% to 20% of patients. ^, While there are no established studies to date evaluating the lung microbiome in SjS-related ILD, Jia and colleagues analyzed the GI microbiome in SjS patients without prior treatment compared with healthy controls and found increased abundance of several taxa with Lactobacillus being the most differentiating organism between groups particularly in SjS patients with ILD. Authors also identified augmentation of the phenylalanine synthetic pathway within the SjS-associated ILD group.

Controversies in studying the lung microbiome

There are several confounders that limit our ability to more reliably characterize the direct role of the lung microbiome in driving CTD-associated pulmonary manifestations. First, the impact of systemic medications and therapies on the microenvironment of the lung that are universally prescribed to patients with CTD is poorly understood. The effect of therapies, such as chronic immunosuppression, including systemic glucocorticoids, disease modifying antirheumatic drugs, biologics, and antibiotics, is poorly understood. ^, Chronic inhaled glucocorticoids have been shown to affect the airway microbiome in patients with asthma. Additionally, proton pump inhibitors (PPIs) are used frequently in patients with CTD to either control GI-related manifestations or to mitigate side effects from other therapies such as glucocorticoids. It is well known that PPIs have a significant impact on the gut microbiome. ^, Future studies will need to explore the effect of these therapies on the lung microbiome given our understanding of the influence of microaspirations on ILD progression as well as the established relationship between mucosal environments through the oral-gut-lung axis.

Second, it is well established that several environmental exposures associated with the development of autoimmune disease have a direct impact on microbial dysbiosis and, therefore, could confound disease-specific alterations. Air pollution, which has been associated with the development of RA and progressive ILD in SSc, has been shown to correlate with lung dysbiosis and increased microbial pathogen colonization of pathogenic taxa such as Pseudomonas and Acinetobacter . ^, Cigarette smoke exposure is a shared risk factor for the development of ILDs including IPF and RA-ILD. Tobacco is thought to mediate pulmonary fibrosis through a variety of mechanisms including direct epithelial cell injury, oxidative stress, M2 macrophage polarization in addition to lung microbial dysbiosis which leads to decrease in overall diversity of the lower respiratory tract. Silica exposure has also been linked to RA and specifically in males with SSc is associated with ILD. ^, Several studies have identified an association of silica-induced microbial dysbiosis and resultant contribution to pulmonary fibrosis through effects on metabolite production and dysregulated TLR4 signaling. ^,

Finally, isolated and recurrent infections assuredly have a direct impact on the lung microenvironment, and these changes can be transient or persist for an extended period of time thus limiting our ability to extrapolate findings from cross-sectional analyses. Of particular interest since 2020 is the potential effect of COVID-19 on the lung microbiome. Wei and colleagues investigated the oral microbiome of health care workers after recovery of COVID-19 and showed differential abundance of several bacterial and fungal taxa which persisted up to 12 months after resolution of the initial infection. It will be critical in future studies to control for covariates such as medication use, environmental/occupational exposures, and respiratory infection history so as to more accurately define the relationship of a specific disease state with lung dysbiosis.

Future considerations

Tissues used for lung microbiome analysis have largely focused on lung biopsies or BALf, both of which require invasive procedures that carry significant risk particularly in those with ILD. While induced sputum is most reflective of the upper airway microenvironment, studies have shown that it can be a valid surrogate for the composition of distal airways, and it is significantly less invasive and safer to collect than BALf. Additionally, analysis of the upper airway microbiome has been validated to serve as an informative biomarker in other forms of chronic lung disease, such as cystic fibrosis, COPD, as well as fibrotic ILD. Future studies investigating induced sputum could provide the platform for the development of a more readily obtainable lung-derived biomarker.

There are several approaches to the design of future research that will be important to implement given the complexities of studying and analyzing the lung microbiome. Studies should seek to simultaneously compare the microbiome at several mucosal sites to more clearly define this relationship and how each influences the other particularly in a diseased state. Prospective, longitudinal studies are critical to understanding the relationship between microbiome and progression of disease given the inherent variability of an individual’s microbiome on a day-to-day basis. Finally, the question will remain as to whether lung microbiome dysbiosis itself drives ILD pathogenesis or is simply a secondary bystander to the disruption of normal lung architecture or host immunoregulatory dysfunction. Mechanistic studies which can be modeled based on prior work in the oropharyngeal and GI tracts will need to be implemented and explored to better inform the causality of lung microbial dysbiosis in lung fibrosis.

Summary

In conclusion, studies of the lung microbiome over the past decade have begun to illustrate its potential role in driving both autoimmune diseases and the resultant pulmonary manifestations including ILD. Future large scale, prospective studies are needed to characterize the specific pathogenic mechanisms of the lung microbiome in CTD-ILD which would provide the necessary insight into the disease pathogenesis with the potential to enhance screening strategies and lung-targeted therapies.

Clinics care points

•

In idiopathic pulmonary fibrosis which shares significant clinical overlap with rheumatoid arthritis (RA)-interstitial lung disease (ILD), lung microbiome bacterial burden and decreased diversity is associated with more severe ILD and increased mortality.
•

There is decreased lung microbial alpha diversity in RA, and it is associated with various markers of disease activity.
•

Respiratory viral infections which likely have a direct impact on the lung microbiome have been associated with the development of inflammatory myopathies-associated ILD.
•

While the oropharyngeal and gastrointestinal (GI) microbiome have been implicated in systemic sclerosis-associated and Sjogren’s syndrome-associated ILD, additional focus will be needed in regards to the direct impact of the lung microbiome.
•

Perturbations within the oropharyngeal microbiome are likely to impact the lung microbiome through direct translocation.
•

Perturbations within the GI microbiome are likely to impact the lung microbiome from microaspiration and communication through the gut-lung axis.
•

There are several limitations to studying the lung microbiome in connective tissue disease (CTD)-ILD which include the confounding effects of genetics, medication use, environmental exposures, and recurrent infections.
•

Prospective and mechanistic studies are needed to better characterize the direct implications of lung dysbiosis on lung-related manifestations in CTD.