The airway mycobiome and interactions with immunity in health and chronic lung disease

Abstract The existence of commensal fungi that reside within the respiratory tract, termed the airway mycobiome, has only recently been discovered. Studies are beginning to characterize the spectrum of fungi that inhabit the human upper and lower respiratory tract but heterogeneous sampling and analysis techniques have limited the generalizability of findings to date. In this review, we discuss existing studies that have examined the respiratory mycobiota in healthy individuals and in those with inflammatory lung conditions such as asthma, chronic obstructive pulmonary disease and cystic fibrosis. Associations between specific fungi and features of disease pathogenesis are emerging but the precise functional consequences imparted by mycobiota upon the immune system remain poorly understood. It is imperative that further research is conducted in this important area as a more detailed understanding could facilitate the development of novel approaches to manipulating the mycobiome for therapeutic benefit.


Introduction to the mycobiome
The human microbiota comprises an extensive ecosystem of trillions of microorganisms that are known to colonize mucosal sites within the body including the skin, gut, and urogenital tract.Although the lower airways were traditionally believed to be a sterile environment, advances in molecular microbiological techniques and sequencing technologies have facilitated the characterization of the presence of microorganisms, which are now understood to be commensal inhabitants of respiratory mucosal surfaces [1].Termed airway microbiota, this community of commensal microorganisms is populated by complex networks of transient and colonizing bacteria, fungi, and viruses, which is unsurprising when one considers that the human respiratory tract represents the primary point of entry for countless microorganisms, especially through airborne particles.
The fungal components of this microbial ecosystem are termed mycobiota, or the mycobiome, where the former refers to the microorganisms themselves and the latter to their corresponding genomes [2].Body sites that are recognized to harbour a mycobiome include the oral cavity, gastrointestinal (GI) tract, respiratory tract, urogenital tract, and the skin [3].The composition of human mycobiota displays significant interindividual and intraindividual variability [4].The shaping of the human mycobiome begins immediately after birth and may be affected by a range of factors such as birth weight, gestational age/delivery and early feeding methods [5,6].In adulthood, the diversity of the mycobiota is further influenced by age, sex, diet, use of antibiotics or antifungals, and environmental factors [3,7].
To date, most studies of the pulmonary microbiome have focused on the presence and functional roles of bacterial communities.Commensal bacteria have been shown to actively contribute to airway homeostasis and determine the development and function of the local immune system through their capacity to influence immune cell maturation, produce antimicrobial molecules and balance pro-and anti-inflammatory responses [8].The bacteriome can be influenced by several factors, such as diet, age, environment, genetics, and antibiotic use.These external influences can induce perturbations leading to dysbiosis, an imbalance in the composition and function of the microbiome.Various chronic diseases have been linked to microbial dysbiosis with implications for the pathogenic progression of these illnesses [9,10].
More recently, it is becoming apparent that, similarly to bacterial communities, mycobiota play key roles in maintaining host homeostasis.However, our understanding of the key commensals and associated mechanisms remains limited.Although modest in number compared to other classes of resident commensals (representing approximately 0.1% of microorganisms in the body), the massive genomes of fungi could convey significant implications for the extent of their influence and role as keystone species in the airways [11].Indeed, certain fungal species are being implicated in chronic respiratory diseases (CRDs) where dysbiosis is likely to have an impact upon cardinal disease features such as chronic inflammation, defective mucociliary clearance and immunosuppression [12,13].
In addition to the apparent importance of the mycobiome in healthy homeostasis, opportunistic members of the mycobiome also have the potential to transcend from commensalism to pathogenicity, a phenomenon which is highly dependent upon altered host immunity, microbial dysbiosis and environmental factors [14,15].Importantly, fungal pathogens exhibit remarkable adaptability to the human lung, partly enabled by the abundance of biosynthetic gene clusters in these organisms, which produce bioactive secondary metabolites that include human toxins (e.g.aflatoxin) [16].Aspergillus fumigatus, the most frequently isolated colonizer in humans, represents a prime example with over 30 biosynthetic gene clusters [17,18].Studies are also beginning to highlight symbiotic and antagonistic transkingdom relationships between bacterial and fungal commensals.The presence of certain microbial taxa has implications for this switch to pathogenicity.It is therefore crucial to understand the complex biological synergy between fungi, bacteria and also viruses with the host immune response, to fully decipher the complex multifaceted interactions that seemingly govern airway homeostasis.
In healthy people, inhaled spores are expelled from the airways through the process of mucociliary clearance [19].However, a variety of factors including, but not limited to, a deficient immune response, presence or absence of CRD, and external factors such as smoking or pollution, can lead to the perturbation of mycobiota in the airways of patients, thus increasing the likelihood of pathogenic fungal colonization and infection.

Challenges in studying the mycobiome
Research around the respiratory mycobiome faces several challenges, both inherent to the microorganisms being interrogated and teething problems from the burgeoning nature of the field.Firstly, there are approximately three million species of fungi and fungus-like organisms, making them the second biggest group of eukaryotes based on global diversity [20].Furthermore, compared to other eukaryotes, fungi have simple cellular structures, often encompassing morphologically ambiguous structures, which poses challenges in accurate identification [20].Moreover, traditional culture-based classification techniques do not provide an authentic representation of respiratory mycobiota due to the large proportion of non-culturable species, biases towards faster growing species or even masking of rarer, morphologically similar species [21,22].With the advent of high throughput amplicon sequencing and shotgun metagenomics came greater power to more accurately discern the spectrum commensals that reside in the airways.However, technical obstacles pertaining to these techniques remain.For example, different DNA extraction methods have been suggested to influence bacterial and fungal community composition and contamination with environmental microorganisms in low biomass airway samples remains a major concern [22,23].The lack of suitable marker genes limits operational taxonomic unit (OTU) classification and accurate determination of relative abundance.Typically, methods target specific genes encoding internal transcribed spacer (ITS)1, ITS2 and 18S rRNA, however the complimentary primers for these sequences produce variable results and can amplify the genes of other eukaryotes [24].Moreover, intragenomic variation in the ITS sequences in DNA barcoding methodologies and a lack of sufficient and suitable reference databases for fungal identification also hinders comprehensive categorization of respiratory mycobiomes [25,26] with the existing databases (e.g.FungiDB, Mycobank, Ensembl Fungi) having limitations in terms of coverage of the full spectrum of fungal taxa.Finally, there is a lack of standardization between studies owing to heterogeneity in design, sampling, sequencing processes and the bioinformatic pipelines and analyses used.
Holistically, these limitations have led to inconsistencies between studies carried out by different groups.Most tend to agree at the phylum level; however, differences arise at the genus level, which along with the large interindividual mycobiome variation, results in a lack of uniform consensus regarding typical composition.Delving into the challenges of fungal species identification in great depth is beyond the scope of this review and we would point the reader towards recent publications by Tiew et al. [27] and Bharti et al. [28] for more extensive elaboration.Nevertheless, it is evident that a consensus on how to design and conduct studies for the characterization of the human airway mycobiome needs to be reached, to improve reproducibility and standardized comparison of results.

Airway mycobiome in health
Mycobiome composition in the respiratory tract is under constant flux with evolution from initial colonization at birth through childhood and adulthood.Several factors within the categories of vertical (from the mother) and horizontal (environmental) transmission modulate the spectrum of fungal microorganisms that inhabit the airways [29,30].These include: the delivery method, gestational age and later the feeding method, while diet continues to be a predominant factor throughout adulthood, along with weight and geographical location [30].The potential of transient mycobiota to establish longterm colonization in the airways can vary by species, microbial burden, and the pulmonary microenvironment of the host [31,32].Further variation between individual mycobiome profiles derives from the environmental niches found in the pulmonary ecosystem, arising from differences in mucus or surfactant secretion, gene expression, pH, and nutrient or oxygen availability [33].
Although ecosystem composition tends to overlap, the respiratory mycobiome can be split into two interconnected regionsthe upper and lower respiratory tract (URT and LRT).Previous studies on the respiratory mycobiome have been limited to focussing on potentially pathogenic fungi, which accounts for the fact that, until recently, fungal colonization of the respiratory tract was believed to be transient [3].More recent investigations have provided evidence that the respiratory tract is permanently populated by fungi, with specific genera and species preferentially colonizing the airways over the oral cavity [3,34].The majority of fungi identified in the human respiratory tract reside in the Basidiomycota and Ascomycota phyla, with the most commonly identified genera in lung tissue being Cladosporium, Eurotium, and Aspergillus [34,35].Importantly, the gut mycobiome has been suggested to influence the oropharyngeal and respiratory mycobiome via micro-aspiration [35], while other determinants such as geographic and climate variability, genetics, and environmental factors, have also been hypothesized to play a role in composition variability [11].The gut-lung axis is well established in the context of the bacteriome and roles for interactions between gut and lung mycobiota are also emerging [36].

Upper respiratory tract
The oropharyngeal mycobiome is important to consider as it represents a major pathway for transmission to the respiratory tract through micro-aspiration.The microbial communities of the oral cavity are amongst the most diverse in the human body and they exhibit huge interindividual variation [37].Composition constantly develops and evolves over the course of a lifetime, with colonization starting at birth and progressing with age [38].The prevailing consensus on the oral mycobiome is an ecosystem dominated by Candida, Cladosporium, Aureobasidium and Aspergillus, with a notable presence of Fusarium, Penicillium and Cryptococcus [39-41].In the adult oral mycobiome, most colonizers belong to the phyla Ascomycota, Basidiomycota, Glomeromycota and Mucoromycota [42,43].
One of the first studies to investigate the oral mycobiome was by Ghannoum et al. [40].This pioneering work examined 20 individuals and identified 85 species of fungi (74 culturable, 11 nonculturable) by multi-tag pyrosequencing, a predecessor to NGS, making it the first to use such technology to this end.Subjects were found to have a range of 5 to 39 fungal genera, and a core set comprising of Candida (75%); Cladosporium (60%); Aureobasidium (50%); Aspergillus (35%); Fusarium (30%), and Cryptococcus (20%) was identified.Further studies refined this 'normal' composition, notably Dupuy et al. who used an improved pyrosequencing approach to overcome process-induced sequencing errors and accurately assign fungal taxonomy, reporting the significant presence of Malassezia in saliva samples from all six participants [43].It is now recognized that two distinct oropharyngeal mycobiota profiles exist-one dominated by Candida and the other by Malassezia.Candida sp.thrive in the low pH environment of the oral cavity and is the only genus to have been shown to reach a significant biomass, being associated with a distinct oral ecology [44].Indeed, mechanistic animal studies have highlighted synergistic interactions with acidogenic bacteria, and the presence of Candida has been significantly correlated with caries including tooth loss, periodontitis and expansion of C. albicans in oral candidiasis [44][45][46].
Numerous ITS sequencing studies have suggested that Candida, Malassezia, Penicillium, Cladosporium, Pichia, Alternaria, Aspergillus, Cryptococcus, Trichosporon and Rhodotorula are high abundance genera which can be more easily isolated from oral samples, implying they are likely the main fungal colonizers [40].Furthermore, Pichia have been shown in vitro to inhibit the growth of potentially pathogenic Candida, Aspergillus and Fusarium [47].It is important to recognize that there are additional challenges associated with identification of fungal taxa in the oral mycobiome, as it represents the immediate region post ingestion of food or fungal spores from the environment.Thus, careful assessment of which species are colonizers and which are transient is required.Despite these limitations, ITS sequencing suggests that Candida, Pichia, and Fusarium species are common oral colonizers [33,40].

Lower respiratory tract
Compared to the oral mycobiome, the lower respiratory tract is even more poorly characterized with a very small number of studies to date focussing on pulmonary mycobiomes in subjects with chronic respiratory diseases.Figure 1 summarizes the different genera and their relative abundances identified from recent sequencing studies.
Up until 2015, the existence of a distinct, resident mycobiome in the lungs of immunocompetent individuals was not recognized [34].Indeed, fungal colonization of the lungs can be difficult to approach due to its potentially stochastic nature, giving the appearance of transience when colonization has in fact occurred [2,34].The precise diversity of fungal species that inhabit the lungs is likely governed by environmental exposure to filamentous fungi and yeasts, micro-aspiration from the oropharynx or by direct inhalation of spores [3].
One of the initial studies assessing the lower respiratory mycobiome was conducted by Van Woerden et al., who carried out 18s pyrosequencing of DNA extracted from sputum samples in 13 non-atopic healthy subjects [52].In total they identified 136 fungal taxa in the healthy lungs, with the species Eremothecium sinecaudum, Systenostrema alba, Cladosporium cladosporioides, Vanderwaltozyma polyspora and Entophlyctis helioformis being most prominent.These findings were further developed in a study by Cui et al.where 18s and ITS sequencing were performed on oral wash, induced sputum and bronchoalveolar lavage (BAL) collected from 56 individuals [34].The authors demonstrated clear patterns of shared and unique taxa between oral wash and BAL communities.Several fungal species were identified as predominant in the lower respiratory tract including Ceriporia lacerata, Saccharomyces cerevisiae, and Penicillium brevicompactum.
One of the major limitations with the current pool of literature on mycobiome composition in the lower respiratory tract is that our understanding is largely based on analysis of sputum samples.Prior studies examining the bacteriome indicate overlap between oropharyngeal and lower respiratory microbiome profiles in sputa [53].Sputum is thus less adequate for assessing fungi that may be present in more distal lung tissue.Therefore, a major future priority for the field is to build an anatomic catalogue or atlas of respiratory tract mycobiota through lower airways (e. g.BAL) sampling.Notably, studies that have employed lower airways sampling have shown greater prominence of Candida and Aspergillus [49,50].

Interactions of the mycobiota with the host immune response
We remain within the infancy of our understanding of how respiratory mycobiota may influence pulmonary host immunity, although inferred evidence of these potential interactions can be gained from studies of other mucosal sites such as the gastrointestinal tract.Structural ligands and metabolites from bacteria, viruses, and fungi of the lung microbiota have a range of functional effects upon innate and adaptive immunity, highlighting how these components can mechanistically influence the development and function of the immune system.While the interaction between bacterial pathogen-associated molecular patterns (PAMPs) and host pattern recognition receptors (PRRs) such as Toll-like receptors (TLRs) is well-established [54, 55], our understanding of fungal PAMPs and their interactions and downstream effects upon host immunity remains more limited.Nevertheless, studies are beginning to indicate that commensal fungi, similar to bacteria, provide basal stimulation to the immune system.This may contribute to protection against pathogens, as well as maintaining tolerance towards 'helpful' members of the microbiota [15].Jiang et al. provided evidence that susceptibility to colitis and influenza A infection in gut microbiota-depleted mice, could be reversed by monocolonization of the gut with C. albicans and S. cerevisiae with protective effects induced by stimulation by the fungal cell wall component mannans [56].
Moreover, as with commensal bacteria, metabolite synthesis by mycobiota may also be a major mechanism of immune modulation.Differential host sensing of fungal determinants from various species can significantly impact homeostatic immune modulation [56,57].Metabolites produced by Malassezia, including malassezin and indolo [3,2-b] carbazole, are known to act as potent ligands for the aryl hydrocarbon receptor (Ahr), leading to downstream signalling that is vital for epithelial tissue repair, barrier homeostasis and immune cell development [57,58].Furthermore, Malassezia produce lipases that catalyse the conversion of host triglycerides in the skin into short-chain fatty acids (SCFAs), metabolites with pleiotropic immunomodulatory effects that have been well-documented in numerous studies [59][60][61].Aspergillus fumigatus can also induce production of shortchain fatty acids through biodegradation of wheat straw lignin [61] but whether these processes occur in the human gut or lungs is unknown.Importantly, a recent study illustrated that the metabolites produced by microorganisms need not necessarily exert their effects exclusively on their local milieu.Specifically, the researchers showed that mice suffering from gut dysbiosis with an overgrowth of C. albicans were more likely to suffer from allergic airway inflammation [60].The authors determined that this was mediated by an increase in the secretion of prostaglandin E 2 (PGE 2 ) by C. albicans, which resulted in the polarization of alveolar macrophages towards the alternatively activated M2 inflammatory phenotype.This is a prime example of fungal metabolite translocation to the lungs being a key component of airway disease pathogenesis.The diversity of other similar metabolites produced by the mycobiota and the extent to which they influence the host immune response, both within and beyond the airways, remains to be investigated.
Factors such as disruption of the healthy microbiota, defects in host barrier functions and immunodeficiency allow fungi such as C. albicans and A. fumigatus to transition from commensalism to pathogenicity.These opportunistic fungal pathogens have been shown to modulate the human immune system, both to the benefit and detriment of the host.Pre-exposure of monocytes to C. albicans primed these cells towards a pro-inflammatory phenotype, leading to increased secretion of IL-6 and TNF-α upon stimulation with TLR ligands, an effect that could be recapitulated by inoculation with β-1,3-glucan alone, which is detected by the receptor dectin-1 [62].Whether or not this immune 'training' effect is replicated in vivo by resident C. albicans of the mycobiota, and how it is balanced with the requirement to dampen the immune response and remain tolerant against commensal members of the microbiota, is yet to be determined.
Control of the host immune system is where fungi seek to attenuate immunological responses.For example, C. albicans has been found to preferentially stimulate the TLR2 receptor, inducing a Th2 immune response, allowing it to persist in the absence of a potent pro-inflammatory Th1 response [63].It also synthesizes farnesol, which significantly reduces cytokine production by macrophages [64].Meanwhile conversion of tryptophan to kynurenine by A. fumigatus through host enzyme indolamine 2,3dioxygenase results in immunosuppression [65,66].
As discussed, it is important to note that many of the aforementioned findings derive from studies investigating the gut mycobiome.Therefore, future studies should be directed at examining whether these findings would be more specifically applicable to respiratory fungal commensals specifically residing within the pulmonary microenvironment.

Interactions between fungal and bacterial commensals
In addition to directly influencing the host immune response, members of the microbiota can also indirectly impact host im- Another form of communication between microorganisms that has garnered attention over the past few years is quorum sensing (QS).This phenomenon describes commensal fungi dictating gene expression within bacteria and vice versa [77].This dynamic form of communication relies on the accumulation of density-dependent small diffusible molecules that transmit a signal upon reaching a specific threshold, highlighting the significance of diversity and population numbers in the composition of the healthy microbiota [78,79].Although initially believed to be exclusive to bacteria, QS has been shown to be especially important in fungi as well, where it can regulate the transition from spherical to hyphae form, apoptosis and pathogenicity [78].Interestingly, the LasIR quorum-sensing system has been implicated in inhibiting A. fumigatus biofilms formation, while the QRregulated toxins pyrrolnitrin and pyocyanin have been shown to inhibit the growth of certain fungi.

Role of respiratory mycobiota in chronic respiratory diseases
Chronic respiratory diseases (CRDs) such as chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF) and asthma affect approximately 1 billion people worldwide, creating an immense financial burden for healthcare systems globally [78].These diseases are broadly characterized by chronic airway inflammation and defective mucociliary clearance which may predispose to fungal colonization and infection.This may be worsened by the frequent use of immunosuppressive (e.g.inhaled corticosteroid) and antimicrobial therapies.There is a growing body of evidence indicating that the respiratory mycobiome may play a key role in the etiopathogenesis of CRDs, mainly through inter-microbial and/or host interactions [27].Several studies have shown that differences in composition and diversity of lung mycobiota between healthy and diseased populations can influence CRD progression, morbidity and clinical outcomes [36]. A. fumigatus represents a prime example of this, since its identification by culture in airway samples from subjects with CRDs has been associated with increased clinical severity and higher risk of exacerbation [80,81].The following sections will highlight disease-specific alterations in respiratory mycobiota composition within a selection of the common lung diseases COPD, asthma, and CF, and discuss how these perturbations may impact upon progression, severity and management of each disease.

COPD
Chronic obstructive pulmonary disease (COPD) is a smokingrelated progressive respiratory condition marked by a persistent inflammatory response in the lower airways, resulting in irreversible airflow obstruction.Several mechanisms contribute to the development of COPD, encompassing an imbalance between proteolytic and anti-proteolytic activities, mitochondrial dysfunction, infiltration of inflammatory cells into the lungs, and oxidative stress [82].Key features of COPD include impaired mucociliary clearance, dysregulation of innate immune responses (e.g.type-I interferon) and chronic inhaled corticosteroid use, all of which could theoretically increase the likelihood of fungal dysbiosis [83-85].
To date, a few studies have attempted to characterize the differences between the lung mycobiome profiles of COPD patients and healthy individuals.Martinsen et al. profiled the oral and lung mycobiome in patients with COPD compared to healthy individuals but surprisingly found no significant differences between these groups with the most abundant genera being Candida, followed by Malassezia, Penicillium and Aspergillus [49].Conversely Tiew et al. reported increased fungal α-diversity in COPD compared to health [51].Specifically, COPD patients with expansion of Saccharomyces within the respiratory mycobiome exhibited increased respiratory symptoms.Subjects with a 'highrisk' mycobiome profile dominated by Aspergillus, Penicillium, and Curvularia dominance had more frequent exacerbations and greater mortality.A subset of these 'high-risk' patients also correlated with a sensitization response to these fungi, suggesting that mycobiota can be linked to a measurable host immune response.Importantly, as this study took participants from Singapore, Malaysia and Scotland, the authors were also able to determine, via linear discriminant analysis, that the airway mycobiome in COPD illustrates geographic variation.The differences between this study and the study by Martinsen et  Concomitant infections with A. fumigatus and P. aeruginosa have also been shown to be a key risk factor for exacerbation, again highlighting the importance of inter-microbial relations [95].Liu et al. reported simultaneous characterization of sputum bacterial and fungal microbiome in 84 stable COPD and 29 healthy subjects, identifying an inverse correlation between bacterial and fungal diversity [96].Perturbed bacterial-fungal interactions were associated with enhanced pro-inflammatory cytokine (IL-6, IL-8) expression and enrichment of fungal taxa with loss of bacterial commensals identified as a driver of exacerbation susceptibility.
Increased airway eosinophilic/type 2 inflammation is recognized to occur in around 10-30% of patients with COPD [97].Eosinophilic COPD is associated with an altered mycobiome composition compared to non-eosinophilic profiles with lower α-diversity and higher relative abundances of Aspergillus, Bjerkandera and Cladosporium [98].These taxa may play key roles in the pathogenesis and progression of COPD with Bjerkandera for example being shown to induce eosinophilic infiltration and type-2 inflammatory cytokine expression [99,100].Cladosporium spp.are known to cause allergic inflammation, airway hyperreactivity and remodelling in mice [101].Allergic sensitization to Aspergillus fumigatus is well recognized in a subset of patients with COPD and associated with a more severe clinical phenotype [102].
Such sensitization is not unique to Aspergillus fumigatus however as other fungi including Alternaria alternata, Schizophyllum commune, Aspergillus tamarii and Rhizopus spp.have been shown to be sensitizing agents in COPD [102].It is imperative therefore to consider that clinical symptoms could be attributable to fungi other than Aspergillus.
Pneumocystis is another key genus that has been identified as potentially important in COPD at stable-state and during exacerbation.In a small Colombian study, a colonization frequency of 32.3% was observed for P. jirovecii, with its presence significantly correlating with more severe disease (GOLD stage IV) [103].Morris et al. reported Pneumocystis colonization rates of 37% in severe COPD [104], which has also been linked to augmented Th1 inflammatory gene expression [105].At exacerbation, Pneumocystis spp, has also been identified in 11/58 patients (19%) with colonization being shown to be associated with higher Airway mycobiome in health and chronic lung disease | 5 serum IL-17 and CD26P levels [106].Using a sensitive LAMP assay, Xue et al. reported increased P. jirovecii colonization at exacerbation (67%) versus stable state (43%) [107].
In summary, perturbation of the mycobiome is a recognised feature of COPD with correlative evidence to indicate that these commensals may play key roles in pathogenesis, exacerbation susceptibility and disease progression.Although a wealth of descriptive data exists, the next step will be to utilize reductionist cellular and animal models to decipher functional roles played by fungal commensals.These models are currently lacking and will require future development and refinement.

Asthma
Asthma is a CRD characterized by airway inflammation, remodelling and airway hyperresponsiveness upon exposure to allergens that results in repeated episodes of heightened symptoms [108].Disease progression displays great heterogeneity by virtue of the complex interplay between genetic and environmental factors that define susceptibility to clinicopathological features [109].The role of fungi as a complicating factor in asthma has long been recognized with allergic bronchopulmonary aspergillosis (ABPA), where an allergic airway response develops to inhaled Aspergillus, estimated to affect around 11% of severe asthmatics [110].Detailed discussion about ABPA in the context of asthma is beyond the scope of this review and we point the reader to dedicated reviews on this topic [111].
The relatively small number of studies comparing the mycobiomes of healthy and asthmatic subjects have reached variable conclusions regarding the abundances of different fungi at the genus and/or species level, which could be attributed to differences in sample type, sampling method or sequencing techniques.Findings regarding the α-diversity (measurement of microbiome diversity applicable to a single sample) of the mycobiome in sputum samples from asthma patients have also been conflicting, where some studies suggest an increase and others a decrease of α-diversity in asthmatics compared to healthy subjects [50,112].
In terms of associations of specific fungi with disease severity and progression, sensitization with Aspergillus sp. has been linked, on multivariate analysis, to exacerbation frequency and greater corticosteroid requirement in severe asthma [113].These observations have been corroborated by mycobiome analysis studies, which found a higher fungal load among patients with asthma compared to healthy controls, with A. fumigatus complex accounting for the biggest part of this increase [114].Moreover, in a study using bronchoalveolar lavage fluid (BALF) samples, an increased abundance of Aspergillus along with Fusarium and Cladosporium was noted in patients with heightened type 2 immune responses [115].Conversely, in a study utilizing ITS2 sequencing in oropharyngeal swab samples, Aspergillus and Candida were reported to be more abundant in healthy subjects compared to asthma patients; the opposite was observed for Malassezia [116].Upper airway colonization with Malassezia has also been shown to be associated with lower risk of progression to severe exacerbation in asthma [117].Finally, Sharpe et al. identified an important association between increased asthma exacerbations and the presence of Cladosporium, Alternaria, Aspergillus, and Penicillium in samples collected at home from patient asthma, which hints at a link between fungal exposure and exacerbation susceptibility [118].
Fungal and bacterial dysbiosis in the human gut microbiome has also been implicated in the immunopathology of asthma.Early studies utilizing a mouse model of gut microbiota dysbiosis and intestinal overgrowth of C. albicans, reported the induction of a type 2-driven allergic airway response upon exposure to A. fumigatus, compared to control mice with an intact microbiota [119].A similar mouse model was used by Kim et al. to elucidate a mechanism of promotion of type 2 immunity through M2 polarization of alveolar macrophages via fungi-produced PGE 2 [60].Although detailed discussion about the gut mycobiome is outside the scope of this review, the importance of these studies lies in highlighting the principle of fungal commensals to modulate the host immune system.Similar mechanisms are likely to occur in the pulmonary environment, thus emphasizing the need for further research.
In summary, the mycobiome is likely to play an important role in asthma, particularly in relation to type 2 inflammatory pathways that characterize this disorder.More detailed functional insight will be required to determine the key fungi and mechanisms involved.

Cystic fibrosis
Cystic fibrosis (CF) is a monogenetic disorder affecting approximately 100 000 people worldwide, associated with decreased life expectancy and a huge treatment burden for patients [120].It is caused by autosomal recessive mutations in the CF transmembrane conductance regulator (CFTR) gene, which encodes an epithelial transmembrane protein that primarily functions as a chloride channel [120].In the pulmonary microenvironment, loss of CFTR function results in an electrolyte imbalance, leading to the secretion and accumulation of abnormally thickened mucus that compromises the airway lumen and impairs mucociliary clearance.This culminates in chronic microbial colonization of the lower airways [121].Recurrent and chronic respiratory infections trigger vigorous inflammatory responses that lead to pulmonary tissue destruction and progressive loss of function, representing the principal cause of loss of quality of life and decreased life expectancy [60].
Although colonizing bacterial pathogens such as P. aeruginosa, S. aureus and H. influenzae are well-recognized to be associated with pulmonary exacerbations, heightened inflammation and mortality in CF patients, far less is known about fungi.Delhaes et al. performed the first characterization of the respiratory mycobiome in CF patients, reporting lower α-diversity of both bacteria and fungi in patients with decreased lung function and poor clinical status, and highlighting C. albicans and A. fumigatus as particularly abundant colonizing species [122].Subsequent studies have shown a variety of results in terms of defining the prevalence of fungi in the respiratory environment of CF patients, with this variation being mainly attributed to geographical and environmental factors or differences in study design [123].Nevertheless, the majority of studies agree that Candida frequently dominates respiratory fungal communities within CF, with C. albicans, C. parapsilosis, and C. dubliniensis representing the most commonly encountered species.Malassezia and Aspergillus species have also been frequently identified as colonizers of the lower respiratory tract (LRT), likely due to micro-aspiration events [124][125][126][127].
Aspergillus has been the subject of much focus in CF and is detectable in around 10-50% subjects [128].A. fumigatus, has been reported to increase with patient age and to be associated with disease severity, as well as corticosteroid and chronic antibiotic use [128][129][130].As in asthma and other respiratory conditions, some of the major problems associated with chronic Aspergillus colonization include Aspergillus bronchitis, Aspergillus sensitization and ABPA.The latter in particular is an allergic inflammatory response process to fungal elements that results in airway remodelling and obstruction and is a frequent cause of pulmonary morbidity in CF [130].The prevalence of A. fumigatus has been reported to increase with patient age, and to be associated with disease severity and chronic antibiotic use [126].Accordingly, it has been speculated that the composition of the CF respiratory microbiota is subject to an array of dynamic selective pressures dependent on the disease stage, ranging from nutrient availability, pH, oxygen pressure, use of antibiotics, host immunity, and inter-microbial influences.For further information on Aspergillus in CF, we point the reader towards recent review articles dedicated to this topic [128].
Major fungal commensals that are expanded within the airway mycobiome of CRDs are summarized in Table 1.
Taken together the mycobiome appears to have roles in influencing the development and progression of CRDs.It is becoming increasingly apparent that the variety of geographical aspects, the genetic defects of the immune system, and the crosstalk between microorganisms of the wider microbiota, represent factors that could affect the clinical spectrum and the management of chronic respiratory diseases.The complex interactions between fungal and bacterial members of the respiratory microbiota and the host immune system are summarized in Fig. 2.

Conclusion and future perspectives
The discovery of the human respiratory microbiome has led to profound questions around how commensals may impact upon immunity and disease pathogenesis.Despite significant advancements being made in molecular and sequencing technologies, the characterization of the fungal component of this microbial community remains incomplete.We are beginning to recognize that the composition of the healthy respiratory mycobiota differs significantly to that of CRDs.Even more gaps remain in our understanding of the functional effects of the respiratory mycobiota, and its involvement in the development and progression of various respiratory diseases.Increased understanding could be driven by improved assessment of microbial function through integrated omics technologies, more standardized experimental design, and a greater emphasis on functional experiments in tractable experimental models.It is also vital to recognize that Figure 2. The complex interplay underlying the impact of the respiratory mycobiota on the development and progression of chronic respiratory diseases.Members of the respiratory mycobiota coinhabit the airways together with other commensal microorganisms.The inter-microbial relationships that exist between these groups can influence their growth and survival, and thus impact upon host immune homeostasis.Mycobiota can also directly manipulate host immunity via production of immunomodulatory metabolites and stimulation of host pattern recognition receptors.These interactions are particularly important in the context of chronic respiratory disease, where increasing evidence suggests that the composition of the respiratory mycobiota can impact the severity and frequency of disease exacerbations the information pool on the interactions of commensal fungi with other bacterial and viral communities is currently very limited.Future research should thus also focus upon trans-kingdom interactions in this context.Table 2 summarizes research areas that may be of future interest to the field.Overall, the mycobiome is anticipated to play a significant role in health and disease and development of our understanding in this important field has the potential to drive novel therapeutic approaches.
complex system in a dynamic relationship with the host.Wiley Interdiscip Rev Syst Biol Med 2019;11:e1438.Can alterations in the composition of the airway mycobiome be used as biomarkers for the early detection of various respiratory conditions?
� Characterization of airway mycobiome in health and disease contexts � Validation of biomarker use in patient cohorts What novel therapeutic strategies can be developed by manipulating the airway mycobiome in the context of various CRDs?
� Interventional experiments in CRD animal models � Clinical trials in human patients

Figure 1 .
Figure 1.A plethora of fungal genera inhabit the lungs but there is little in the way of composition consensus between recent studies.The different genera and their relative abundances identified from recent sequencing studies.These are mainly from analysis of airway samples from small groups of subjects with CRDs.Research articles referenced: Mac Aogain et al. [48], Martinsen et al. [49], Rick et al. [50], Tiew et al. [51] by interacting with each other[67][68][69].This is especially prominent in the gut and the lower airways, where fungi and bacteria of the microbiota coinhabit the same environmental niches in polymicrobial biofilms attached to mucosal surfaces[70, 71].These mixed biofilm environments can convey several benefits to the cohabitants, including metabolic cooperation and evasion of immune and anti-microbial agents[71].Indeed, the interplay between fungi and bacteria can be bidirectional, with mechanical and chemical interactions shown to selectively affect the growth and survival of different species, depending on the specific environmental conditions[72].For instance, several mutualistic and non-mutualistic competitive interactions between P. aeruginosa and A. fumigatus have been recorded in the context of co-infection in patients with cystic fibrosis.Mechanisms include inhibition of biofilm formation and growth suppression via nutrient sequestration or reactive oxygen species (ROS) production[72][73][74].It is also possible for fungal-bacterial interactions to enhance the virulence of pathogens, either directly, through the activity of metabolic products [75], or indirectly, through the incorporation of pathogens into biofilms, hampering the activity of the immune system and antimicrobials[76].
al. [49], conducted in Norway could thus partially be explained by differences in geographic location.This should be carefully considered when drawing conclusions from studies of this nature.Su et al. conducted a longitudinal study of six patients with acute exacerbations, showing that alterations in mycobiome composition occur during the episodes but consistent patterns could not be elucidated from this small study [86].Similarly, Enaud et al. reported that individuals with acute exacerbations exhibited lower α-diversity than stable-state, suggesting expansion of certain fungi towards a pulmonary mycobiome dominated by fewer taxa [87].Several studies support that Aspergillus spp.are detectable and potentially important during exacerbation.Bafadhel et al., observed that approximately 50% of stable patients with COPD at baseline had culturable filamentous fungi, 75% of which were A. fumigatus [88] with positive culture for A. fumigatus detected in 28% of exacerbations.In another study, 1.3-39% of hospitalized COPD patients developed invasive aspergillosis [89].Aspergillus colonization is strongly associated with corticosteroid use [90] and may promote airway inflammation, airways hyperresponsiveness and relapse of acute exacerbation [91-93].Moreover, Candida spp.isolation in the lower respiratory tract is associated with increased risk of recurrent exacerbations in COPD [94].

Table 1 .
Summary of fungal commensal taxa that have been shown to be increased within the airway mycobiome in COPD, asthma and cystic fibrosis

Table 2 .
Outstanding research areas and questions that need to be addressed in the future, in order to further advance our understanding of the role of the airway mycobiome in the development and progression of chronic respiratory diseases (CRDs)

questions that remain to be addressed Investigative approach
What are the key fungal species present in the airway mycobiome of healthy individuals?What is the airway mycobiome profile of various CRDs?� Comparative metagenomic analyses � High-throughput ITS sequencing � Standardized longitudinal cohort studies Does dysbiosis of the mycobiome contribute to development of invasive or chronic fungal disease?� Longitudinal studies in immunosuppressed cohorts Could airway mycobiome profiling be used as a diagnostic tool to predict CRD exacerbations in a clinical setting?� Characterization of airway mycobiome in health and disease context � Development of accurate techniques for identification of fungi in various airway sample types What are the functional roles played by different members of the airway mycobiome in maintaining immune homeostasis?� Metatranscriptomics and metabolomics studies � Comparative studies with CRD patients � Interventional experiments in CRD animal models What are the interactions of members of the airway mycobiome with bacterial species not typically associated with overt respiratory infections?� Longitudinal correlational studies in CRD patients How do environmental and lifestyle factors (e.g.pollution, allergens, smoking, diet) affect the composition and function of the airway mycobiome?� Epidemiological studies � Longitudinal correlational studies encompassing geographical and lifestyle variation How is the composition of the airway mycobiome influenced by the use of biologics therapies (e.g.monoclonal antibodies) for the treatment of type-2 asthma?
� Interventional experiments in CRD animal models � Characterization of airway mycobiome in human clinical trial patients