Emerging Pathogenic Links Between the Microbiota and the Gut–Lung Axis

PERSPECTIVES

Microbiota of the healthy gut and lungs MICROBIOME — OPINION The GIT remains, by far, the best-studied host-associated microbial ecosystem, partly Emerging pathogenic links between owing to its abundance of microorganisms and partly because the microbiota can be profiled through faeces, which is easily microbiota and the gut–lung axis obtainable. Both the abundance and diversity of the commensal microbiota Kurtis F. Budden, Shaan L. Gellatly, David L. A. Wood, Matthew A. Cooper, generally increase along the GIT, and there Mark Morrison, Philip Hugenholtz and Philip M. Hansbro are site-specific variations in the mucosa and the lumen13,14. These differences are Abstract | The microbiota is vital for the development of the immune system and governed by the prevailing environment, homeostasis. Changes in microbial composition and function, termed dysbiosis, in including pH, the concentration of bile the respiratory tract and the gut have recently been linked to alterations in immune acids, digesta retention time, mucin responses and to disease development in the lungs. In this Opinion article, we properties and host defence factors15. review the microbial species that are usually found in healthy gastrointestinal and Despite these variations, the GIT is dominated by members of only four respiratory tracts, their dysbiosis in disease and interactions with the gut–lung axis. bacterial phyla, Firmicutes, Bacteroidetes, Although the gut–lung axis is only beginning to be understood, emerging evidence Proteobacteria and Actinobacteria; indicates that there is potential for manipulation of the gut microbiota in the with lesser and sporadic representation treatment of lung diseases. of other phyla, including Fusobacteria, Verrucomicrobia and Spirochaetes. This ‘core’ gut microbial community comprises Chronic lung diseases, such as asthma which lack an appropriately developed up to 14 bacterial genera and 150 bacterial and chronic obstructive pulmonary disease immune system and show mucosal ‘species’, many of which have not yet (COPD), are common and often occur alterations, both of which can be restored been cultured16–18. together with chronic gastrointestinal through colonization with gut microbiota6,7. We are beginning to understand the tract (GIT) diseases, such as inflammatory The microbiota changes over time lung microbiota through programmes bowel disease (IBD) or irritable bowel from birth, to adulthood and into old such as the Lung HIV Microbiome Project, syndrome (IBS)1,2. Up to 50% of adults age, and in response to environmental which is a multi-centre study that examines with IBD and 33% of patients with IBS factors, such as diet, and drug and both individuals who are infected with have pulmonary involvement, such as environmental exposures8. HIV and uninfected individuals who inflammation or impaired lung function, In this ever-expanding field, researchers have varying histories of lung and/or although many patients have no history are now investigating how the local respiratory disease19. The lungs have a of acute or chronic respiratory disease3,4. microbiota influences immunity at distal large surface area with high environmental Furthermore, patients with COPD are sites, in particular how the gut microbiota exposure and are equipped with effective 2–3 times more likely to be diagnosed influences other organs, such as the brain, antimicrobial defences. Healthy lungs were with IBD4. Individuals with asthma have liver or lungs. This has led to the coining long considered to be sterile; however, the functional and structural alterations in of terms such as the ‘gut–brain axis’ and advent of culture-independent approaches their intestinal mucosa, and patients with ‘gut–lung axis’. For example, antibiotic- for microbial community profiling has COPD typically have increased intestinal induced alterations in the gut microbiota resulted in the detection of microbial DNA permeability2,5. Although the mature in early life increase the risk of developing in the lungs of healthy individuals19,20. GIT and respiratory tract have different allergic airway disease9–12; such findings These microorganisms probably reached environments and functions, they have the add to our understanding of the links the lungs from the oral cavity through same embryonic origin and, consequently, between exposure to microorganisms and microaspiration, as the taxonomic profiles have structural similarities. Thus, it is not allergy and autoimmunity (BOX 1). The of the two sites were similar19,20. Compared surprising that the two sites might interact mechanisms by which the gut microbiota with surrounding sites, the lungs had a in health and disease (FIG. 1); however, affects the immune responses in the lungs, decreased abundance of Prevotella-affiliated the underlying mechanisms are not and vice versa, are being uncovered, but taxa and an enrichment of Proteobacteria, well understood. many questions remain. In this Opinion specifically Enterobacteriaceae, An emerging area of intense interest article, we summarize the emerging role Ralstonia spp., and Haemophilus spp.19, is the influence of the host-associated of the microbiota in the gut–lung axis, which may have resulted from host microbiota on local and systemic immunity. highlighting gaps in our knowledge and the immunity and environment, such as redox This is exemplified in germ-free mice, potential for therapeutic intervention. state and oxygen availability. The lung

PERSPECTIVES

microbiota might not be resident in the lungs22. Technical challenges, such as the gut and respiratory mucosa provide healthy individuals, but rather transiently low microbial biomass and bronchoscope a physical barrier against microbial recolonized through microaspiration and contamination, constant seeding from penetration, and colonization with the breathing. The lungs have a comparatively oral and GIT sites, and mucociliary and normal microbiota generates resistance low microbial biomass and remarkably immune clearance, have hindered the to pathogens; for example, through the similar composition to adjacent sites, identification of a viable and resident, production of bacteriocins15. Furthermore, even though the lungs are continuously or a transiently recolonizing, microbiota a rapidly expanding collection of exposed to entering microorganisms and in the lungs, as well as further research commensal gut bacteria, including their environmental conditions differ vastly into host–microorganism interactions. segmented filamentous bacteria (SFB), from other body sites. These observations Novel methods of sampling tissue with Bifidobacterium spp. and members of support the hypothesis that entry and minimal contamination23, longitudinal the colonic Bacteroides genus, induce the selective elimination of a transient studies to identify temporal changes in production of antimicrobial peptides, microbiota is the major determinant of the microbiota, and the increasing use secretory immunoglobulin A (sIgA) microbial composition in the lungs, rather of metagenomic analyses to facilitate the and pro-inflammatory cytokines. than resident and viable microorganisms. cultivation of fastidious bacterial species24, Non-pathogenic Salmonella strains This does not negate the importance of will provide a clearer picture of the role of downregulate inflammatory responses host–microorganism interactions in the the respiratory microbiota and enable the in GIT epithelial cells by inhibiting lungs, as evidenced by correlations better design of interventional studies to the ubiquitylation of nuclear factor-κB between the composition of microbial develop a more complete understanding (NF-κB) inhibitor-α (IκBα)25, whereas communities and pulmonary inflammation of host–microorganism interactions in some Clostridium spp. promote and disease21. Rather, it highlights the the lungs. anti-inflammatory regulatory T cell 26 delicate balance between microbial (Treg cell) responses . In the respiratory exposure and elimination; the possibility Interactions between the gut and lungs tract, Streptococcus pneumoniae and of dysbiosis at oral sites preceding and/or Interactions of microorganisms between Haemophilus influenzae synergistically causing dysbiosis in the lungs and the sites. The epithelial surfaces of the activate host p38 mitogen-activated contributing to disease pathogenesis19; GIT and respiratory tract are exposed to a protein kinase (MAPK) in a Toll-like and the importance in distinguishing wide variety of microorganisms; ingested receptor (TLR)-independent manner to whether microbial DNA that is detected microorganisms can access both sites amplify pro-inflammatory responses27. by culture-independent techniques is and the microbiota from the GIT can Conversely, non-pathogenic S. pneumoniae truly representative of viable bacteria in enter the lungs through aspiration. Both and other bacteria and their components

Health Disease Inhalation or swallowing Triggers of dysbiosis • Infection • Antibiotics • Cigarette smoke

• Healthy lung environment • Chronic lung disease • Eﬀective clearance of infection • Poor clearance of infection

Immune modulation Dysfunctional immune modulation

Gastrointestinal • Bacterial ligands (for example, LPS) disease or symptoms • Bacterial metabolites (for example, SCFAs) Intestinal Dysbiosis of • Migrating immune cells microbiota microbiota

Figure 1 | Principles of gut–lung crosstalk in health and disease. A healthy system is still developing, this disturbance can substantially alter the way in Nature Reviews | Microbiology intestinal microbiota maintains homeostatic local immune responses through which the immune system perceives its surroundings in later life, which can the exposure of structural ligands (for example, lipopolysaccharide (LPS) and/ lead to the development of chronic inflammatory disorders in the gut and lung. or peptidoglycan) and secreted metabolites (for example, short-chain fatty In adulthood, dysbiosis of the gut microbiota, for example, through exposure acids (SCFAs)). Invading microorganisms and absorbed metabolites influence to cigarette smoke, can cause systemic inflammation and an outgrowth of circulating lymphocytes and contribute to the regulation of systemic immune opportunistic pathogens, which can lead to chronic inflammation at distal responses. When the gut microbiota is disturbed, for example, during infection sites. Although the specific taxa, ligands, metabolites and/or host responses or antibiotic exposure, the normal microbiota-derived signals are altered, may differ in specific disease situations, these broad principles outline the role which leads to a changed immune response. In early life, when the immune of the microbiota in gut–lung crosstalk.

PERSPECTIVES

can suppress allergic airway disease by Box 1 | The hygiene hypothesis and microbiota 28–31 inducing Treg cells . In recipients of lung transplants, the microbiota of the In 1989, after observing an inverse correlation between the occurrence of hay fever and number of siblings, David Strachan coined the term ‘hygiene hypothesis’ (REF. 116). He proposed that respiratory tract alters immunity in decreases in the incidence of infections during childhood altered the development of the immune the lungs. Firmicutes-dominated and system, leading to an increased risk of allergic disease. Subsequent studies showed that growing Proteobacteria-dominated dysbiosis up on a farm117, attending child care118 and exposure to dog-associated house dust44 all provided were associated with the expression protection against the development of asthma. This hypothesis was later modified to state that of inflammatory genes in pulmonary microbial exposure from commensal bacteria that had co‑evolved with humans (rather than faster leukocytes, whereas Bacteroidetes- evolving viruses) was necessary to correctly mature the immune system119. Both hypotheses have dominated dysbiosis was linked to a gene since been used to explain the increase in various autoimmune and allergic diseases that correlates expression profile that is characteristic of well with the decrease in infectious diseases in affluent countries. This is now supported by 116 120 tissue remodelling32. In both cell culture32 substantial epidemiological evidence for asthma, hay fever , atopic dermatitis , type 1 diabetes 121 122 and animal models33, the inflammatory mellitus and multiple sclerosis . Expansion of the gut microbiota begins immediately after birth and is heavily influenced by response that is induced by pathogenic environmental factors, with species in the phylum Actinobacteria often dominating during species is larger than the response induced infancy123,124. In this early period of life, changes in the microbiota may be linked to the by commensal microorganisms, which development of chronic lung disorders in later life. The decrease in exposure to infectious agents indicates that a diverse lung microbiota and changes in the microbiota have many causes, including improved hygiene and sanitation protects against pathology by ‘diluting’ practices, the provision of clean water, pasteurization and vaccination. Antibiotics directly cause the more pro-inflammatory stimuli dysbiosis and in infants this may increase susceptibility to chronic inflammatory diseases in later of pathogens. Although the transfer of life125. Furthermore, the modern diet, which is high in processed foods, also affects the gut microorganisms from faecal suspensions microbiota and may have a major, but currently undefined, role in these processes. has been used to determine the role of the In addition, in areas in which helminth infection is rare, the incidence of allergic disease is high. gut microbiota, such techniques have not Individuals with chronic helminth colonization show antigen-specific immune hyporesponsiveness, with increased levels of interleukin‑10 (IL‑10) and suppressive regulatory T cells (T cells). In yet been used to transfer the respiratory reg addition, helminths can influence the differentiation of B cells, immunoglobulin E (IgE) responses, microbiota between animals, which limits natural killer T cell activity and macrophage function, to downregulate immune responses and our understanding of their roles. thereby protect the host against allergic disease126. Several studies have shown the effects of GIT colonization with orally administered bacteria on lung function. Oral gavage microorganisms between sites, although microbiota enriched with oral-related of faecal suspensions in mice that were the translocation of bacteria from the GIT taxa, such as Prevotella spp., Rothia spp. first treated with antibiotics provided to the lungs has been observed in sepsis and Veillonella spp., was associated with short-term improvements in some, but not and acute respiratory distress syndrome, in TH17‑mediated immunity in the lungs of all, measures in an S. pneumoniae infection which barrier integrity is compromised40. healthy human hosts21. Whether these links model34.Although the nature of this ‘gut– In addition, some environmental factors, are correlative or causative remain unclear. lung axis’ has been challenged owing to such as dietary fibre, can produce similar Exposure of mice to dog-associated house the potential confounding effects of faecal changes in the microbiota of the GIT dust altered the caecal microbiota, and, administration by oral gavage and antibiotic and the lungs39. Whether these result in particular, increased the abundance use35, the concept warrants systematic from diet-driven changes in microbial of Lactobacillus johnsonii and other and controlled evaluation. In infants, the metabolites, changes in innate immune Firmicutes-related lineages, such as species composition of the gut microbiota and responses or a combination of both remains in the Peptococcaceae and Lachnospiraceae caesarean section have been linked to to be determined. families44. Mice that were either exposed atopic manifestations, and colonization by to dog dust or inoculated with L. johnsonii Clostridium difficile at one month of age Microbial species-specific effects on host showed decreases in respiratory tract was associated with wheeze and eczema immunity. The crucial role of the microbiota TH2 cytokine responses, and L. johnsonii throughout early life, and with asthma after in lung homeostasis and immunity is treatment protected against exposure to 6–7 years36. Positive associations between demonstrated by the poor outcomes of respiratory syncytial virus and allergens the presence of ‘beneficial’ bacteria, such germ-free mice that were exposed to acute such as ovalbumin. Other examples of as Bifidobacterium longum, in the gut and infections41 and their susceptibility to microbial influences on host immunity a lower incidence of asthma have also allergic airway disease42. Current research is include the ability of various Bacteroides 37 been identified , although larger and assessing the effects of selected members of spp. to expand Treg cell populations or longer studies are required to evaluate the commensal gut microbiota on systemic bias the TH1/TH2 phenotype in either these associations. immunity, including in the lungs, as well direction in a strain-specific manner, or Considerable evidence suggests that as the use of probiotics and prebiotics the suppression of host inflammatory host epithelia and other structural and to prevent and treat acute and chronic responses by the common bacterial immune cells assimilate information pulmonary disease (FIG. 3). For example, metabolites short-chain fatty acids (SCFAs), directly from microorganisms and from SFB in the gut, when present naturally which act through free fatty acid (FFA) the concomitant local cytokine response or introduced by probiotic dosing or receptors and/or epigenetic regulation of to adjust inflammatory responses, and that co‑housing of mice, stimulated pulmonary immune cells45. this shapes immune responses at distal T helper 17 (TH17) responses and In related human studies, seropositivity sites, such as the lungs38,39 (FIG. 2). There protection from S. pneumoniae infection to the gut-specific pathogen Helicobacter is less evidence of the direct transfer of and mortality43. Intriguingly, a respiratory pylori, in particular, cytotoxin-associated

PERSPECTIVES

Intestinal tract O2

Inﬂuences Mucus • Diet • Ingested or aspirated bacteria • Cigarette smoke and pollution • Prebiotics or probiotics

Respiratory tract

Surfactant O2 sIgA

O2 Progenitor Goblet Progenitor cells cell cell Paneth cells Lymphocyte migration Prevotella spp. Bacteroides spp. Peptococcus spp. Veillonella spp. Dendritic cell Biﬁdobacterium spp. Peptostreptococcus spp. Streptococcus spp. Eubacterium spp. Ruminococcus spp. Pseudomonas spp. Clostridium spp.

Bacterial seeding through reﬂux or aspiration

Figure 2 | Structural and functional similarities and differences structure vary along the gastrointestinal tractNature (GIT). Reviews By contrast, | Microbiology the res- between the gut and lung. The gut and respiratory tract epithelia have piratory tract and alveoli are oxygen-rich and movement is bidirectional substantial differences in functional purpose and exist in different (inhalation and exhalation). The temperature in the gut is relatively environments; however, they retain some anatomical similarities. Both are uniform at 37 °C, whereas the temperature in the respiratory tract differs derived from the endoderm and consist of columnar epithelial cells with depending on the proximity to the pharynx. Thus, it is unsurprising that projections of microvilli (gut) or cilia (respiratory tract) that function as a the microbial life in each environment is distinct. Changes in diet and physical barrier and as sentinels for the immune system in conjunction exposure to therapeutics and environmental particulates can directly with associated lymphoid tissue. Both secrete mucus through goblet cells affect the composition of the microbiota. Both the gut and lungs are able as well as secretory immunoglobulin A (sIgA; although less in the lung). to influence each other’s immune responses. Dendritic cells in the intes- The alveoli that are found in terminal airways in the lungs differ substan- tines and respiratory tract, and macrophages in the lungs, sample anti- tially, consisting of squamous epithelial cells that either secrete surfactant gens in the lumen. Lymphocytes in the associated lymphoid tissues (type 2 alveolar cells) or function in gas exchange (type 1 alveolar cells). circulate through the lymphatic system to affect systemic immunity. However, the similarities end here: the intestinal lumen is an oxygen-poor Bacteria from the gut can travel to the lungs through aspiration of vomit environment and functions to digest food and absorb nutrients. The or oesophageal reflux. During times of dysbiosis, disturbed epithelial movement of matter is unidirectional (mouth to anus), with the exception integrity may enable bacteria and their components and metabolites to of reflux or vomiting. Furthermore, the pH, enzyme presence and enter the circulatory system, which can cause systemic inflammation. gene A positive (cagA+) strains, has long Clearly, the incredible diversity and and that this oral tolerance to microbial been linked with decreased incidence of abundance of species in the gut microbiota components was due to interleukin‑10 asthma and allergy46–48. Conversely, two results in many immunomodulatory (IL‑10)-mediated hypo-responsiveness; recent meta-analyses suggest that infection signals, which have considerable combined however, subsequent exposure to LPS with H. pylori is positively associated with effects on host health. Although much was no longer tolerated and the immune increased incidence of COPD and other has been uncovered about the activity of response became similar to that seen chronic bronchial diseases49. Although these specific bacterial species, current research in conventional mice51,52. Furthermore, differences might be partly attributable has only just begun to assess the structure– a robust response to LPS by colonic to genetic, environmental and lifestyle function relationships of the gut and lung macrophages could be restored by factors, these findings raise the possibility microbiota with host immunity. commensal microorganisms53. that systemic immune responses that are Bacterial components can also have triggered by H. pylori might have different Components and metabolites of the anti-inflammatory effects, which attenuate roles in the aetiology of different lung gut microbiota that influence the lung. GIT pathology. Polysaccharide A (PSA) disorders. Strain variations, in addition to Early studies showed that germ-free mice from Bacteroides fragilis induces the the expression of cagA, might also affect have reduced responsiveness to lipo production of IL‑10 by T cells and protects 50 Treg cell responses . polysaccharide (LPS)-induced pathology against intestinal inflammatory disease

PERSPECTIVES

caused either chemically or by infection N‑oxide in atherosclerosis, which further were not consistent across all studies. with Helicobacter hepaticus54. Sphingolipids, highlights their importance in extra- In addition, although models of allergic which are naturally occurring cell membrane intestinal environments55. In studies of other airway disease support the existence of a components of many anaerobic gut genera diseases, Bacteroides spp. were associated critical developmental window early in including Bacteroides, decrease the number with early-onset autoimmune diseases, life42,62, only one study has provided direct of invariant natural killer T cells in the colon which may be a consequence of potent evidence that restoring the altered gut — cells that have been implicated in the activation of immunity by LPS produced microbiota through probiotic treatment can development of colitis55. The best-studied by these bacteria58. decrease susceptibility to asthma12. metabolites, SCFAs, are by‑products of the Similarly, in adults, the overall microbial fermentation of dietary fibre, have Gut microbiota and lung diseases composition of the faecal microbiota in anti-inflammatory properties, are a source Asthma. An increased risk of asthma has individuals with allergic asthma does not of energy for colonocytes, and regulate fatty been connected to the disruption of the differ from healthy individuals63,64. There acid and lipid synthesis in the host56. gut microbiota in early life (BOX 1), and are taxa-specific differences, such as the Much less is known about the influence several studies have sought to characterize enrichment of Bifidobacterium adolescentis, of microbial components and metabolites at the precise microbial constituents that are which negatively correlated with the time other sites, including the lungs. Decreases associated with the development of the since asthma diagnosis63. Interestingly, in Faecalibacterium spp., Lachnospira spp., disease in infants. heat-inactivated Bifidobacterium spp. that Veillonella spp. and Rothia spp. in the gut, The overall composition of the gut were isolated from infants with allergic and the urine levels of some microbial microbial community is not altered in asthma induced larger pro-inflammatory bile acid metabolites correlate with the infants at risk of the development of asthma, responses than those isolated from development of atopic wheeze in children, but subtle transient changes in select taxa healthy individuals65. although whether they are a cause or a can be detected in the first few months of There are several proposed mechanisms consequence of wheeze is not known12. life12,59. Increased risk of asthma has been through which the gut microbiota can Oral administration of SCFAs has been associated with an increase in the abundance attenuate the risk of developing asthma. shown experimentally to alleviate allergic of B. fragilis and total anaerobes in early Infants who were at risk of developing airway disease39,57. Microbial components life60, as well as reduced microbial diversity59 asthma had decreased levels of LPS in and metabolites have been implicated and decreases in Escherichia coli 61 and the their faeces12, whereas PSA from B. fragilis in other disorders, such as tryptophan relative abundances of Faecalibacterium spp., protected against the development of allergic in brain health, PSA in disorders of the Lachnospira spp., Rothia spp. and airways disease in mice by inducing IL‑10 central nervous system and trimethylamine Veillonella spp.12, although these findings responses in T cells66. H. pylori alleviated

H. pylori Biﬁdobacterium spp. B. fragilis Structural components Lactobacillus spp. (LPS, ﬂagella, LTA)

UreB VacA NAP GGT SCFAs PSA

Macrophage Neutrophil T cell B cell Dendritic cell Increased IL-10 Reduced inﬂammatory production cytokine responses

Treg cell Increased recruitment, Enhanced bacterial killing Increased cytokine Increased and cytokine humoral Tolerogenic reprogramming responses recruitment production immunity Figure 3 | Immune system programming by microbiota. Secreted and can downregulate host immune responses, whereas structural compo- structural components of microorganisms can influence the host immune nents from commensal bacteria can influenceNature inflammatory Reviews | Microbiology responses response both locally and at distal sites. Microbial metabolites, such as through the activation of pattern recognition receptors. GGT, γ-glutamyl short chain fatty acids (SCFAs), bind to free fatty acid receptors or pro- transpeptidase; IL‑10, interleukin‑10; LPS, lipopolysaccharide; LTA, mote epigenetic changes in host leukocytes, which induce anti-inflam- lipoteichoic acid; NAP, neutrophil-activating protein; PSA, poly matory responses and decrease inflammation. Virulence factors from saccharide A; Treg cell, regulatory T cell; UreB, urease subunit-β; VacA, pathogenic bacteria, such as Helicobacter pylori or Bacteroides fragilis, vacuolating cytotoxin A.

PERSPECTIVES

murine allergic airway disease in several COPD. Respiratory microbiota research are true and operate simultaneously or ways, namely through the direct activation of in COPD has assessed changes in the disease at different stages of disease. Exposure to 67 Treg cells by neutrophil-activating protein , state and with smoke exposure, which is environmental stimuli and the onset of or indirectly through urease subunit-β, a major risk factor for the development disease cause dysbiosis, which, in turn, which promotes tolerogenic reprogramming of this disease. Interestingly, although probably contributes to disease progression. of dendritic cells68. In addition, γ‑glutamyl the lung microbiota is similar in healthy Moreover, defined probiotic use may transpeptidase and vacuolating cytotoxin smokers and non-smokers, the oral benefit patients with COPD, particularly from H. pylori altered dendritic cell function, microbiota differs substantially between if used as an early preventive intervention. 19 but did not require Treg cells to alleviate the two groups . As enrichment of the lung Oral administration of Lactobacillus casei symptoms69. Commensal bacteria can microbiota with taxa from the oral cavity is improved the previously defective function also influence the development of asthma associated with increased inflammation in of peripheral natural killer cells in adult male through the production and secretion of smokers75, it is plausible that changes in the smokers90, whereas Bifidobacterium breve metabolites, specifically SCFAs. The risk oral microbiota and a failure to effectively and Lactobacillus rhamnosus reduced lung of asthma in infants was associated with a clear aspirated microorganisms contribute to pathology in a mouse model of COPD91 decrease in the concentration of acetate in disease development, and may help explain and reduced inflammatory responses in faeces12 and inversely correlated with serum why only a subset of smokers develop macrophages that were exposed to cigarette acetate concentrations in their mothers COPD. Moreover, there are substantial smoke extract in vitro92. Similarly, a diet that when they were pregnant57. A high-fibre differences between the lung microbiota of increased the production of SCFAs protected diet, which increased levels of SCFAs in patients with COPD compared with ‘healthy’ against elastase-induced inflammation and serum and faeces, protected mice against smokers76,77, which led to the proposal that emphysema93. Although a causal relationship the development of asthma symptoms, the respiratory microbiota may be useful between SCFAs and protection in this study a phenomenon that could be replicated in the early diagnosis of COPD. By contrast, was not confirmed, both cigarette smoke94 through the direct administration of acetate no study to date has investigated changes in and environmental particulate matter95 or propionate before disease onset to the gut microbiota of patients with COPD. decreased SCFA concentrations in rodents, promote tolerogenic immune responses in Nevertheless, in ‘healthy’ smokers, the faecal and cigarette smoke condensate decreased the 39,57 89 dendritic cells and Treg cells . The benefits microbiota is characterized by an increase in production of SCFAs in vitro . Furthermore, of a high-fibre diet were associated with a the abundance of Bacteroides–Prevotella78, increased intestinal translocation of bacteria decreased ratio of Firmicutes/Bacteroidetes and a decreased Firmicutes/Bacteroidetes and their products occurred after exposure and an enrichment of Bacteroidaceae in ratio79 compared with non-smokers. to particulate matter or the development both the faeces and lungs, which highlights These changes in the composition of the of COPD2,95,96. Bacterial toxins, such as the necessity of investigating microbial faecal microbiota have been associated enterotoxin97 or LPS98, can contribute to the communities at several body sites for a with intestinal inflammation and IBD80,81. pathogenesis of COPD and microbiota-asso- complete understanding of the influence Smokers also have a decreased abundance of ciated intestinal inflammation may become of microorganisms on host health. These Bifidobacterium spp.79,82, and hence may lose systemic and also contribute. The potential of studies did not directly explore the the anti-inflammatory effects that are often SCFAs to improve intestinal barrier function relationship between the composition associated with this genus. may account for their benefits in animal of the microbial communities at the two The causes of smoking-associated models of COPD99,100, although this is yet to sites, or the relative importance of the gut changes in the composition of the gut be explored in clinical studies. or lung microbiota in protection against microbiota are probably a combination of disease. Such studies would be valuable in environmental, host and microbial changes, Respiratory infections. The gut microbiota determining which body site to target with such as intestinal and immune disruption, is broadly protective against respiratory therapeutic interventions. An important but impaired clearance of pathogens83,84, infection, as its depletion or absence in understudied area is the role that interactions acidification of gastric contents85 and mice leads to impaired immune responses between microorganisms have in the ingestion of bacteria that are present in and worsens outcomes following bacterial development of asthma. For example, the cigarettes86. Furthermore, cigarette smoke or viral respiratory infection34,41,101–103. loss of intestinal bacteria and the subsequent can directly affect the virulence of bacteria87 Administration of SFB improved resistance outgrowth of commensal fungi triggered and fungi88, as well as alter the growth and to Staphylococcus aureus pneumonia43 and prostaglandin E2-induced changes in alveolar exopolysaccharide structure of known gut Bifidobacterium spp. protected against macrophages and increased allergic airway bacteria, such as Bifidobacterium animalis89, both bacterial104 and viral pulmonary inflammation70. Furthermore, gut helminth which may contribute to dysbiosis. Even infection in mice103,105. Lactobacillus spp. and infection protected mice against allergic after the cessation of smoking, many of Bifidobacterium spp.-based probiotics also airway disease, which was associated with an the changes that cause dysbiosis persist for improved the incidence and outcomes of increase in the abundance of Lachnospiraceae prolonged periods of time, and thus any respiratory infections in humans106–109. and other Clostridiales members, the therapeutic intervention to restore the gut Several aspects of experimental design production of SCFAs and subsequent robust microbiota may potentially require repeated influence the results of infection studies, 71 Treg cell responses in the lungs . Although the administration to prevent relapse. including the route of administration of 102,110 Treg cell-promoting capability of Clostridium In the absence of longitudinal or bacterial ligands , the facility from which spp. has previously been demonstrated in the interventional studies, it is difficult to research animals are sourced43, the type colon26,72, it is increasingly being explored for ascertain whether changes in the gut or of antibiotic used for the depletion of the the treatment of diseases at other body sites, respiratory microbiota are a cause or a microbiota62,102 and the infecting pathogen. including asthma73,74. consequence of COPD. Most likely, both For example, herpes simplex virus type 2 or

PERSPECTIVES

Legionella pneumophila do not seem to be Box 2 | Future directions influenced by antibiotic-mediated depletion of the microbiota102. Therapeutic efforts that involve the microbiota and that are focused on gastrointestinal disorders Nevertheless, several important are further advanced than endeavours that target the gut–lung axis, or indeed that target the lungs in general. Whereas initial research has focused on associative studies between mechanisms by which the gut microbiota pathophysiology and the composition of the microbiota, the next step is a shift to causal links, promotes the clearance of pathogens have which will then indicate interventional strategies for microbiota-modifying or immunomodulatory been identified. Innate immune responses therapeutics. A recent survey of the microbiota intellectual property landscape127 showed that to bacteria in the lungs are greatly enhanced patent filings (dominated by food and nutraceutical companies and smaller biotechnology by exposure to nucleotide-binding start-ups) were directed towards the treatment of infectious diseases (for example, infection with oligomerization domain (NOD)-like receptor Clostridium difficile), digestive and metabolic disorders (for example, inflammatory bowel disease and TLR agonists in the GIT, which include (IBD) and type 1 diabetes), and, to a lesser extent, inflammatory and/or immune disorders. peptidoglycan, LPS, lipoteichoic acid and Products that are currently in development encompass faecal transplants and ‘cocktails’ of live CpG DNA41,101,110. Similarly, stimulation of microorganisms. In addition, there is interest in microbial metabolites and related designer small TLRs by cell wall components and flagellin molecules to beneficially modulate host immune responses. For example, there are several patents filed on small-molecule agonists of free fatty acid receptor 2 (FFAR2), which is the host receptor for of gut bacteria is necessary for effective 128–130 102,111 short-chain fatty acids (SCFAs) . FFAR2 is a G protein-coupled receptor, a class that is adaptive immune responses to influenza , considered to be inherently ‘druggable’, that is expressed on neutrophils, eosinophils and other whereas the anti-inflammatory effects of oral immune cells, and has been linked to exacerbated or unresolved inflammation in animal models SCFA administration are linked to reduced of colitis, arthritis and asthma131. This provides a link between the SCFAs from fermentable pulmonary pathology following both dietary fibre and beneficial effects in inflammatory diseases such as asthma57. Receptor-targeted bacterial104,112 and viral113 infection in mice. approaches such as this may be complementary therapies to more traditional corticosteroids and However, the microbiota can also drive gut cytokine-directed treatments for pulmonary disorders. pathology in pulmonary infection. Influenza virus infection in mice increased the number of lung-derived CC-chemokine receptor enough to affect health and disease. It is other microorganisms and in the metabolites 9 positive (CCR9+)CD4+ T cells, which unclear whether changes in the microbiota that they produce. Thus, next-generation preferentially migrate to the GIT under the at one site affect many distal sites equally, ‘omics’ approaches are required to identify guidance of C-C motif ligand 25 (CCL25) or whether these systemic effects might be functional guilds to aid in defining how the expressed on intestinal epithelial cells114. This specific to certain tissues. To date, no such microbiota of the gut and the microbiota resulted in the outgrowth of E. coli and the broad study investigating these systemic of the lungs interact with each other and induction of aberrant TH17 responses and widespread effects has been carried out. influence health and disease. intestinal damage. Thus far, gut–lung microbiota studies In summary, the lung microbiota in a have had two major limitations: the first is healthy individual may be transient and Conclusions and perspectives discerning causative over correlative effects constantly re‑seeded from the environment Many studies have identified the presence and the second is timing. Most studies have and cleared by the immune system, but of a lung microbiota in health and disease. been associative. Furthermore, culture- may still influence health and disease. However, we believe that the healthy lung independent identification of microorganisms In respiratory diseases the lung microbiota microbiota may be transient and best has not yet replaced the need to isolate and probably becomes persistent and may be both described as a progression of taxa that are culture suspected opportunistic pathogens a cause and a consequence of the disease, influenced by adjacent body sites and the or probiotics to study their effects, and many forming a pathogenic feedback loop. It is clear external environment, rather than an actively members of the microbiota cannot be easily that bacterial components and metabolites reproducing core resident community. cultured. Thus, it is typically unclear whether in the gut and the lungs have the capacity This is not down-playing the importance changes that are observed in the microbiota to modulate systemic and local immunity, of a transient microbiota in healthy lungs, are the cause or effect of disease. In the case with specific taxa able to influence the which could still have important roles in of timing, most experimental data have pathogenesis of respiratory diseases, such as inflammatory responses whether viable or described the effect of the gut microbiota asthma, COPD and respiratory infections. not. By contrast, the microbiota is more on the development of lung disease and Such relationships have been identified in likely to be persistent and resident in the not on established lung disease. Longitudinal other respiratory diseases, such as cystic respiratory tract and lungs of individuals studies in humans and animals that associate fibrosis115, which, as a genetic disease, is a with respiratory disease, although whether changes in the microbiota with the severity special case. Respiratory challenges with it is a cause or a consequence of disease of established chronic lung disease are environmental factors such as pollution, remains to be elucidated. Furthermore, the required. Research into manipulations of the cigarette smoke, antibiotics and diet, influence lung microbiota could affect, or be affected microbiota during lung disease is necessary disease risk and probably drive pathogenesis by, microorganisms or immune responses at to improve our understanding and inform the through their ability to modulate the distal sites. development of novel therapies (BOX 2). composition of the microbiota, although The crosstalk between microorganisms Increasingly, microbiota research is the mechanisms of these effects remain and the host is complex and our current moving towards defining functional guilds. unknown. Further longitudinal studies and understanding of these interactions is in its As taxonomic variation between sites and improved interventional experiments will infancy. It is unlikely that any one of these individuals is so large, and the microbiota help to elucidate the role of the microbiota interactions is solely responsible for the consists of thousands of species, it is highly and gut–lung crosstalk in respiratory disease, functions of the microbiota, and alterations likely that there is redundancy between and will potentially lead to the identification in any part of these relationships may be species in terms of their interactions with of new and effective avenues for treatment.

PERSPECTIVES

Kurtis F. Budden, Shaan L. Gellatly and Philip 17. Qin, J. et al. A human gut microbial gene catalogue 40. Dickson, R. P. et al. Enrichment of the lung microbiome M. Hansbro are at the Priority Research Centre for established by metagenomic sequencing. Nature 464, with gut bacteria in sepsis and the acute respiratory 59–65 (2010). distress syndrome. Nat. Microbiol. 1, 16113 (2016). Healthy Lungs, University of Newcastle and Hunter 18. Ormerod, K. L. et al. Genomic characterization of the 41. Fagundes, C. T. et al. Transient TLR activation restores Medical Research Institute, Newcastle, uncultured Bacteroidales family S24‑7 inhabiting the inflammatory response and ability to control New South Wales 2305, Australia. guts of homeothermic animals. Microbiome 4, 36 pulmonary bacterial infection in germfree mice. (2016). J. Immunol. 188, 1411–1420 (2012). David L. A. Wood is at the Australian Centre for 19. Morris, A. et al. Comparison of the respiratory 42. Olszak, T. et al. Microbial exposure during early life Ecogenomics, School of Chemistry and Molecular microbiome in healthy nonsmokers and smokers. has persistent effects on natural killer T cell function. Biosciences, The University of Queensland. Am. J. Respir. Crit. Care Med. 187, 1067–1075 (2013). Science 336, 489–493 (2012). 20. Bassis, C. M. et al. Analysis of the upper respiratory 43. Gauguet, S. et al. Intestinal microbiota of mice Matthew A. Cooper is at the Institute for Molecular tract microbiotas as the source of the lung and gastric influences resistance to Staphylococcus aureus Bioscience, The University of Queensland. microbiotas in healthy individuals. mBio 6, e00037 pneumonia. Infect. Immun. 83, 4003–4014 (2015). (2015). 44. Fujimura, K. E. et al. House dust exposure mediates Mark Morrison is at The University of Queensland 21. Segal, L. N. et al. Enrichment of the lung microbiome gut microbiome Lactobacillus enrichment and airway Diamantina Institute, Translational Research Institute. with oral taxa is associated with lung inflammation of immune defense against allergens and virus infection. a TH17 phenotype. Nat. Microbiol. 1, 16031 (2016). Proc. Natl Acad. Sci. USA 111, 805–810 (2014). Philip Hugenholtz is at the Australian Centre for 22. Rogers, G. B. et al. Assessing the diagnostic 45. Samuelson, D. R., Welsh, D. A. & Shellito, J. E. Ecogenomics, School of Chemistry and Molecular importance of nonviable bacterial cells in respiratory Regulation of lung immunity and host defense by the infections. Diagn. Microbiol. Infect. Dis. 62, 133–141 intestinal microbiota. Front. Microbiol. 6, 1085 (2015). Biosciences, and the Institute for Molecular Bioscience, (2008). 46. Chen, Y. & Blaser, M. J. Inverse associations of The University of Queensland, Brisbane, Queensland 23. Shanahan, E. R., Zhong, L., Talley, N. J., Morrison, M. Helicobacter pylori with asthma and allergy. 4072, Australia; and The University of Queensland & Holtmann, G. Characterisation of the Arch. Intern. Med. 167, 821–827 (2007). Diamantina Institute, Translational Research Institute, gastrointestinal mucosa-associated microbiota: a 47. Reibman, J. et al. Asthma is inversely associated with novel technique to prevent cross-contamination during Helicobacter pylori status in an urban population. Woolloongabba, Queensland 4102, Australia. endoscopic procedures. Aliment. Pharmacol. Ther. 43, PLoS ONE 3, e4060 (2008). Correspondence to P.M.H. 1186–1196 (2016). 48. Chen, Y. & Blaser, M. J. Helicobacter pylori colonization 24. Pope, P. B. et al. Isolation of Succinivibrionaceae is inversely associated with childhood asthma. [email protected] implicated in low methane emissions from Tammar J. Infect. Dis. 198, 553–560 (2008). wallabies. Science 333, 646–648 (2011). 49. Wang, F., Liu, J., Zhang, Y. & Lei, P. Association of doi:10.1038/nrmicro.2016.142 25. Neish, A. S. et al. Prokaryotic regulation of epithelial Helicobacter pylori infection with chronic obstructive Published online 3 Oct 2016 responses by inhibition of IκB-α ubiquitination. Science pulmonary disease and chronic bronchitis: a meta- 289, 1560–1563 (2000). analysis of 16 studies. Infect. Dis. (Lond.) 47, 597–603

1. Roussos, A., Koursarakos, P., Patsopoulos, D., 26. Atarashi, K. et al. Treg induction by a rationally selected (2015). Gerogianni, I. & Philippou, N. Increased prevalence of mixture of Clostridia strains from the human 50. Hussain, K. et al. Helicobacter pylori-mediated irritable bowel syndrome in patients with bronchial microbiota. Nature 500, 232–236 (2013). protection from allergy is associated with asthma. Respir. Med. 97, 75–79 (2003). 27. Ratner, A. J., Lysenko, E. S., Paul, M. N. & IL‑10‑secreting peripheral blood regulatory T cells. 2. Rutten, E. P., Lenaerts, K., Buurman, W. A. Weiser, J. N. Synergistic proinflammatory responses Front. Immunol. 7, 71 (2016). & Wouters, E. F. Disturbed intestinal integrity in induced by polymicrobial colonization of epithelial 51. McGhee, J. R. et al. Lipopolysaccharide (LPS) regulation patients with COPD: effects of activities of daily living. surfaces. Proc. Natl Acad. Sci. USA 102, 3429–3434 of the immune response: T lymphocytes from normal Chest 145, 245–252 (2014). (2005). mice suppress mitogenic and immunogenic responses 3. Yazar, A. et al. Respiratory symptoms and pulmonary 28. Preston, J. A. et al. Inhibition of allergic airways to LPS. J. Immunol. 124, 1603–1611 (1980). functional changes in patients with irritable bowel disease by immunomodulatory therapy with whole 52. Michalek, S. M., Kiyono, H., Wannemuehler, M. J., syndrome. Am. J. Gastroenterol. 96, 1511–1516 killed Streptococcus pneumoniae. Vaccine 25, Mosteller, L. M. & McGhee, J. R. Lipopolysaccharide (2001). 8154–8162 (2007). (LPS) regulation of the immune response: LPS influence 4. Keely, S., Talley, N. J. & Hansbro, P. M. Pulmonary– 29. Thorburn, A. N., Foster, P. S., Gibson, P. G. on oral tolerance induction. J. Immunol. 128, intestinal cross-talk in mucosal inflammatory disease. & Hansbro, P. M. Components of Streptococcus 1992–1998 (1982). Mucosal Immunol. 5, 7–18 (2012). pneumoniae suppress allergic airways disease and 53. Ueda, Y. et al. Commensal microbiota induce LPS 5. Vieira, W. A. & Pretorius, E. The impact of asthma on NKT cells by inducing regulatory T cells. J. Immunol. hyporesponsiveness in colonic macrophages via the the gastrointestinal tract (GIT). J. Asthma Allergy 3, 188, 4611–4620 (2012). production of IL‑10. Int. Immunol. 22, 953–962 (2010). 123–130 (2010). 30. Thorburn, A. N. & Hansbro, P. M. Harnessing 54. Mazmanian, S. K., Round, J. L. & Kasper, D. L. 6. Wymore Brand, M. et al. The altered Schaedler flora: regulatory T cells to suppress asthma: from potential A microbial symbiosis factor prevents intestinal continued applications of a defined murine microbial to therapy. Am. J. Respir. Cell Mol. Biol. 43, 511–519 inflammatory disease. Nature 453, 620–625 (2008). community. ILAR J. 56, 169–178 (2015). (2010). 55. Sharon, G. et al. Specialized metabolites from the 7. Al‑Asmakh, M. & Zadjali, F. Use of germ-free animal 31. Preston, J. A. et al. Streptococcus pneumoniae microbiome in health and disease. Cell Metab. 20, models in microbiota-related research. J. Microbiol. infection suppresses allergic airways disease by 719–730 (2014). Biotechnol. 25, 1583–1588 (2015). inducing regulatory T‑cells. Eur. Respir. J. 37, 53–64 56. den Besten, G. et al. The role of short-chain fatty acids 8. Quercia, S. et al. From lifetime to evolution: timescales (2011). in the interplay between diet, gut microbiota, and host of human gut microbiota adaptation. Front. Microbiol. 32. Bernasconi, E. et al. Airway microbiota determines energy metabolism. J. Lipid Res. 54, 2325–2340 5, 587 (2014). innate cell inflammatory or tissue remodeling profiles (2013). 9. Noverr, M. C., Falkowski, N. R., McDonald, R. A., in lung transplantation. Am. J. Respir. Crit. Care Med. 57. Thorburn, A. N. et al. Evidence that asthma is a McKenzie, A. N. & Huffnagle, G. B. Development of http://dx.doi.org/10.1164/rccm.201512-2424OC developmental origin disease influenced by maternal allergic airway disease in mice following antibiotic (2016). diet and bacterial metabolites. Nat. Commun. 6, 7320 therapy and fungal microbiota increase: role of host 33. Larsen, J. M. et al. Chronic obstructive pulmonary (2015). genetics, antigen, and interleukin‑13. Infect. Immun. disease and asthma-associated Proteobacteria, but 58. Vatanen, T. et al. Variation in microbiome LPS 73, 30–38 (2005). not commensal Prevotella spp., promote Toll-like immunogenicity contributes to autoimmunity in 10. Russell, S. L. et al. Early life antibiotic-driven changes receptor 2‑independent lung inflammation and humans. Cell 165, 842–853 (2016). in microbiota enhance susceptibility to allergic pathology. Immunology 144, 333–342 (2015). 59. Abrahamsson, T. R. et al. Low gut microbiota diversity asthma. EMBO Rep. 13, 440–447 (2012). 34. Schuijt, T. J. et al. The gut microbiota plays a in early infancy precedes asthma at school age. 11. Russell, S. L. et al. Perinatal antibiotic-induced shifts protective role in the host defence against Clin. Exp. Allergy 44, 842–850 (2014). in gut microbiota have differential effects on pneumococcal pneumonia. Gut 65, 575–583 60. Vael, C., Nelen, V., Verhulst, S. L., Goossens, H. inflammatory lung diseases. J. Allergy Clin. Immunol. (2016). & Desager, K. N. Early intestinal Bacteroides 135, 100–109 (2015). 35. Dickson, R. P. & Cox, M. J. The premature invocation fragilis colonisation and development of asthma. 12. Arrieta, M. et al. Early infancy microbial and of a ‘gut–lung axis’ may obscure the direct effects of BMC Pulm. Med. 8, 19 (2008). metabolic alterations affect risk of childhood asthma. respiratory microbiota on pneumonia susceptibility. 61. Orivuori, L. et al. High level of faecal calprotectin at age Sci. Transl. Med. 7, 307ra152 (2015). Gut http://dx.doi.org/10.1136/gutjnl-2016-311823 2 months as a marker of intestinal inflammation 13. Aguirre de Carcer, D. et al. Numerical ecology (2016). predicts atopic dermatitis and asthma by age 6. validates a biogeographical distribution and 36. van Nimwegen, F. A. et al. Mode and place of delivery, Clin. Exp. Allergy 45, 928–939 (2015). gender-based effect on mucosa-associated bacteria gastrointestinal microbiota, and their influence on 62. Russell, S. L. et al. Perinatal antibiotic treatment affects along the human colon. ISME J. 5, 801–809 asthma and atopy. J. Allergy Clin. Immunol. 128, murine microbiota, immune responses and allergic (2011). 948–955 (2011). asthma. Gut Microbes 4, 158–164 (2013). 14. Donaldson, G. P., Lee, S. M. & Mazmanian, S. K. Gut 37. Akay, H. K. et al. The relationship between 63. Hevia, A. et al. Allergic patients with long-term asthma biogeography of the bacterial microbiota. Nat. Rev. bifidobacteria and allergic asthma and/or allergic display low levels of Bifidobacterium adolescentis. Microbiol. 14, 20–32 (2016). dermatitis: a prospective study of 0–3 years-old PLoS ONE 11, e0147809 (2016). 15. Buffie, C. G. & Pamer, E. G. Microbiota-mediated children in Turkey. Anaerobe 28, 98–103 (2014). 64. Hua, X., Goedert, J. J., Pu, A., Yu, G. & Shi, J. Allergy colonization resistance against intestinal pathogens. 38. Marsland, B. J., Trompette, A. & Gollwitzer, E. S. associations with the adult fecal microbiota: analysis of Nat. Rev. Immunol. 13, 790–801 (2013). The gut–lung axis in respiratory disease. Ann. Am. the American Gut Project. EBioMedicine 3, 172–179 16. Zhernakova, A. et al. Population-based metagenomics Thorac. Soc. 12 (Suppl. 2), S150–S156 (2015). (2016). analysis reveals markers for gut microbiome 39. Trompette, A. et al. Gut microbiota metabolism of 65. He, F. et al. Stimulation of the secretion of pro- composition and diversity. Science 352, 565–569 dietary fibre influences allergic airway disease and inflammatory cytokines by Bifidobacterium strains. (2016). hematopoiesis. Nat. Med. 20, 159–166 (2014). Microbiol. Immunol. 46, 781–785 (2002).

PERSPECTIVES

66. Johnson, J. L., Jones, M. B. & Cobb, B. A. Bacterial increases C. albicans adhesion, growth, biofilm 110. Clarke, T. B. Early innate immunity to bacterial infection capsular polysaccharide prevents the onset of asthma formation, and EAP1, HWP1 and SAP2 gene in the lung is regulated systemically by the commensal through T cell activation. Glycobiology 25, 368–375 expression. BMC Microbiol. 14, 61 (2014). microbiota via Nod-like receptor ligands. Infect. Immun. (2015). 89. Hu, J., Wei, T., Sun, S., Zhao, A. & Xu, C. Effects of 82, 4596–4606 (2014). 67. Sehrawat, A., Sinha, S. & Saxena, A. Helicobacter cigarette smoke condensate on the production and 111. Oh, K. Z. et al. TLR5‑mediated sensing of gut microbiota

pylori neutrophil-activating protein: a potential Treg characterization of exopolysaccharides by is necessary for antibody responses to seasonal modulator suppressing allergic asthma. Front. Bifidobacterium. An. Acad. Bras. Cienc. 87, influenza vaccination. Immunity 41, 478–492 (2014). Microbiol. 6, 493 (2015). 997–1005 (2015). 112. Bernard, H. et al. Dietary pectin-derived acidic 68. Koch, K. N. et al. Helicobacter urease-induced activation 90. Reale, M. et al. Daily intake of Lactobacillus casei oligosaccharides improve the pulmonary bacterial of the TLR2/NLRP3/IL‑18 axis protects against asthma. Shirota increases natural killer cell activity in smokers. clearance of Pseudomonas aeruginosa lung infection J. Clin. Invest. 125, 3297–3302 (2015). Br. J. Nutr. 108, 308–314 (2012). in mice by modulating intestinal microbiota and 69. Engler, D. B. et al. Effective treatment of allergic 91. Verheijden, K. A. T. et al. Treatment with specific immunity. J. Infect. Dis. 211, 156–165 (2015). airway inflammation with Helicobacter pylori prebiotics or probiotics prevents the development 113. Kishino, E., Takemura, N., Masaki, H., Ito, T. immunomodulators requires BATF3‑dependent of lung emphysema in a mouse model of COPD. & Nakazawa, M. Dietary lactosucrose suppresses dendritic cells and IL‑10. Proc. Natl Acad. Sci. USA Eur. J. Pharmacol. 668, e12–e13 (2011). influenza A (H1N1) virus infection in mice. Biosci. 111, 11810–11815 (2014). 92. Mortaz, E. et al. Anti-inflammatory effects of Microbiota Food Health 34, 67–76 (2015). 70. Kim, Y. G. et al. Gut dysbiosis promotes M2 Lactobacillus rahmnosus and Bifidobacterium breve 114. Wang, J. et al. Respiratory influenza virus infection macrophage polarization and allergic airway on cigarette smoke activated human macrophages. induces intestinal immune injury via microbiota-

inflammation via fungi-induced PGE(2). Cell Host PLoS ONE 10, e0136455 (2015). mediated TH17 cell–dependent inflammation. Microbe 15, 95–102 (2014). 93. Tomoda, K. et al. Whey peptide-based enteral diet J. Exp. Med. 211, 2397–2410 (2014). 71. Zaiss, M. M. et al. The intestinal microbiota attenuated elastase-induced emphysema with increase 115. Huang, Y. J. & LiPuma, J. J. The microbiome in cystic contributes to the ability of helminths to modulate in short chain fatty acids in mice. BMC Pulm. Med. 15, fibrosis. Clin. Chest Med. 37, 59–67 (2016). allergic inflammation. Immunity 45, 998–1010 64 (2015). 116. Strachan, D. P. Hay fever, hygiene, and household size. (2015). 94. Tomoda, K. et al. Cigarette smoke decreases organic BMJ 299, 1259–1260 (1989). 72. Furusawa, Y. et al. Commensal microbe-derived acids levels and population of Bifidobacterium in 117. Riedler, J. et al. Exposure to farming in early life and butyrate induces the differentiation of colonic caecum of rats. J. Toxicol. Sci. 36, 261–266 (2011). development of asthma and allergy: a cross-sectional regulatory T cells. Nature 504, 446–450 (2013). 95. Kish, L. et al. Environmental particulate matter induces survey. Lancet 358, 1129–1133 (2001). 73. Huang, F. et al. Early-life exposure to Clostridium murine intestinal inflammatory responses and alters 118. Ball, T. M. et al. Siblings, day-care attendance, and the leptum causes pulmonary immunosuppression. the gut microbiome. PLoS ONE 8, e62220 (2013). risk of asthma and wheezing during childhood. N. Engl. PLoS ONE 10, e0141717 (2015). 96. Zuo, L. et al. Cigarette smoking is associated with J. Med. 343, 538–543 (2000). 74. Li, Y. N. et al. Effect of oral feeding with Clostridium intestinal barrier dysfunction in the small intestine but 119. Rook, G. A., Martinelli, R. & Brunet, L. R. Innate leptum on regulatory T‑cell responses and allergic not in the large intestine of mice. J. Crohns Colitis 8, immune responses to mycobacteria and the airway inflammation in mice. Ann. Allergy 1710–1722 (2014). downregulation of atopic responses. Curr. Opin. Allergy Asthma Immunol. 109, 201–207 (2012). 97. Huvenne, W. et al. Exacerbation of cigarette smoke- Clin. Immunol. 3, 337–342 (2003). 75. Segal, L. N. et al. Enrichment of lung microbiome induced pulmonary inflammation by Staphylococcus 120. Bieber, T. Atopic dermatitis. N. Engl. J. Med. 358, with supraglottic taxa is associated with increased aureus enterotoxin in mice. Respir. Res. 12, 69 (2011). 1483–1494 (2008). pulmonary inflammation. Microbiome 1, 19 98. Brass, D. M. et al. Chronic LPS inhalation causes 121. Gale, E. A. The rise of childhood type 1 diabetes in (2013). emphysema-like changes in mouse lung that are the 20th century. Diabetes 51, 3353–3361 (2002). 76. Pragman, A. A., Kim, H. B., Reilly, C. S., Wendt, C. associated with apoptosis. Am. J. Respir. Cell. Mol. 122. Alonso, A. & Hernan, M. A. Temporal trends in the & Isaacson, R. E. The lung microbiome in moderate Biol. 39, 584–590 (2008). incidence of multiple sclerosis: a systematic review. and severe chronic obstructive pulmonary disease. 99. Kelly, C. J. et al. Crosstalk between microbiota-derived Neurology 71, 129–135 (2008). PLoS ONE 7, e47305 (2012). short-chain fatty acids and intestinal epithelial HIF 123. Human Microbiome Project Consortium. Structure, 77. Sze, M. A. et al. The lung tissue microbiome in chronic augments tissue barrier function. Cell Host Microbe function and diversity of the healthy human obstructive pulmonary disease. Am. J. Respir. Crit. 17, 662–671 (2015). microbiome. Nature 486, 207–214 (2012). Care Med. 185, 1073–1080 (2012). 100. Suzuki, T., Yoshida, S. & Hara, H. Physiological 124. Ottman, N., Smidt, H., de Vos, W. M. & Belzer, C. The 78. Benjamin, J. L. et al. Smokers with active Crohn’s concentrations of short-chain fatty acids immediately function of our microbiota: who is out there and what do disease have a clinically relevant dysbiosis of the suppress colonic epithelial permeability. Br. J. Nutr. they do? Front. Cell. Infect. Microbiol. 2, 104 (2012). gastrointestinal microbiota. Inflamm. Bowel Dis. 18, 100, 297–305 (2008). 125. Okada, H., Kuhn, C., Feillet, H. & Bach, J. F. The 1092–1100 (2012). 101. Chen, L. W., Chen, P. H. & Hsu, C. M. Commensal ‘hygiene hypothesis’ for autoimmune and allergic 79. Biedermann, L. et al. Smoking cessation alters microflora contribute to host defense against diseases: an update. Clin. Exp. Immunol. 160, 1–9 intestinal microbiota: insights from quantitative Escherichia coli pneumonia through toll-like receptors. (2010). investigations on human fecal samples. Inflamm. Shock 36, 67–75 (2011). 126. Maizels, R. M., McSorley, H. J. & Smyth, D. J. Bowel Dis. 20, 1496–1501 (2014). 102. Ichinohe, T. et al. Microbiota regulates immune Helminths in the hygiene hypothesis: sooner or later? 80. Kabeerdoss, J., Jayakanthan, P., Pugazhendhi, S. defence against respiratory tract influenza A virus Clin. Exp. Immunol. 177, 38–46 (2014). & Ramakrishna, B. S. Alterations of mucosal infection. Proc. Natl Acad. Sci. USA 108, 5354–5359 127. Sun, X., Fiala, J. L. & Lowery, D. Patent watch: microbiota in the colon of patients with (2011). modulating the human microbiome with live inflammatory bowel disease revealed by real time 103. Wu, S. et al. Microbiota regulates the TLR7 signaling biotherapeutic products: intellectual property polymerase chain reaction amplification of 16S pathway against respiratory tract influenza A virus landscape. Nat. Rev. Drug Discov. 15, 224–225 ribosomal ribonucleic acid. Indian J. Med. Res. 142, infection. Curr. Microbiol. 67, 414–422 (2013). (2016). 23–32 (2015). 104. Vieira, A. T. et al. Control of Klebsiella pneumoniae 128. Brown, A. J. et al. Pharmacological properties of acid 81. Schwab, C. et al. Longitudinal study of murine pulmonary infection and immunomodulaation by oral N‑thiazolylamide FFA2 agonists. Pharmacol. Res. microbiota activity and interactions with the host treatment with commensal probiotic Bifidobacterium Perspect. 3, e00141 (2015). during acute inflammation and recovery. ISME J. 8, longum 51A. Microbes Infect. 18, 180–189 (2016). 129. Hudson, B. D. et al. Defining the molecular basis for 1101–1114 (2014). 105. Kawahara, T. et al. Consecutive oral administration of the first potent and selective orthosteric agonists of the 82. Khonsari, S. et al. A comparative study of Bifidobacterium longum MM‑2 improves the defense FFA2 free fatty acid receptor. J. Biol. Chem. 288, bifidobacteria in human babies and adults. system against influenza virus infection by enhancing 17296–17312 (2013). Biosci. Microbiota Food Health 35, 97–103 (2016). natural killer cell activity in a murine model. Microbiol. 130. Schmidt, J. et al. Selective orthosteric free fatty acid 83. Verschuere, S. et al. Cigarette smoking alters Immunol. 59, 1–12 (2015). receptor 2 (FFA2) agonists: identification of the epithelial apoptosis and immune composition in 106. Luoto, R. et al. Prebiotic and probiotic structural and chemical requirements for selective murine GALT. Lab. Invest. 91, 1056–1067 (2011). supplementation prevents rhinovirus infections in activation of FFA2 versus FFA3. J. Biol. Chem. 286, 84. Allais, L. et al. Chronic cigarette smoke exposure preterm infants: a randomized placebo-controlled trial. 10628–10640 (2011). induces microbial and inflammatory shifts and mucin J. Allergy Clin. Immunol. 133, 405–413 (2014). 131. Maslowski, K. M. et al. Regulation of inflammatory changes in the murine gut. Environ. Microbiol. 18, 107. Jespersen, L. et al. Effect of Lactobacillus paracasei responses by gut microbiota and chemoattractant 1352–1363 (2015). subsp. paracasei, L. casei 431 on immune response to receptor GPR43. Nature 461, 1282–1286 (2009). 85. Hammadi, M., Adi, M., John, R., Khoder, G. A. influenza vaccination and upper respiratory tract & Karam, S. M. Dysregulation of gastric H,K‑ATPase infections in healthy adult volunteers: a randomized, Acknowledgements by cigarette smoke extract. World J. Gastroenterol. double-blind, placebo-controlled, parallel-group study. The authors acknowledge the support of fellowships from the 15, 4016–4022 (2009). Am. J. Clin. Nutr. 101, 1188–1196 (2015). Australian National Health and Medical Research Council 86. Sapkota, A. R., Berger, S. & Vogel, T. M. Human 108. King, S., Glanville, J., Sanders, M. E., Fitzgerald, A. (NHMRC; to M.A.C. and P.M.H.), the Australian Research pathogens abundant in the bacterial metagenome of & Varley, D. Effectiveness of probiotics on the duration Council (ARC; to P.H.), the Brawn Foundation, the Faculty of cigarettes. Environ. Health Perspect. 118, 351–356 of illness in healthy children and adults who develop Health and Medicine at the University of Newcastle, Australia, (2010). common acute respiratory infectious conditions: a and grants from the NHMRC and the Rainbow Foundation (to 87. Kulkarni, R. et al. Cigarette smoke increases systematic review and meta-analysis. Br. J. Nutr. 112, P.M.H.). The authors thank F. Thomson and M. Thomson for Staphylococcus aureus biofilm formation via 41–54 (2014). their continued support. oxidative stress. Intect Immun. 80, 3804–3811 109. West, N. P. et al. Probiotic supplementation for (2012). respiratory and gastrointestinal illness symptoms in 88. Semlali, A., Killer, K., Alanazi, H., Chmielewski, W. healthy physically active individuals. Clin. Nutr. 33, Competing interests statement & Rouabhia, M. Cigarette smoke condensate 581–587 (2014). The authors declare no competing interests.