The respiratory tracts of healthy individuals have shown that they are not sterile as formerly thought but are composed of a complex microbial community termed the microbiome (Erb-Downward et al., 2011). The respiratory microbiome has been suggested to play an essential role in the pathogenesis of pulmonary diseases (Nguyen et al., 2015). The contribution of bacteria to the pathogenesis of pulmonary diseases has been extensively studied, whereas the roles of fungal species are much less understood (de Dios Caballero et al., 2022). In the last few years, the airway fungal microbiota (mycobiome) has drawn closer attention for the following reasons. The respiratory tract is the main portal of entry for various microorganisms, including fungal spores. Based on the microorganism characteristics, the human immune response will either clear them or allow them to colonize the respiratory tract (Pashley et al., 2012; Denning et al., 2014). As a continuous host–pathogen interaction environment, fungi in the respiratory tract are considered to be associated with chronic respiratory diseases (CRDs), including cystic fibrosis (CF), chronic obstructive pulmonary disease (COPD), asthma, and bronchiectasis (Nguyen et al., 2015). Growing evidence has shown a significant impact of the airway mycobiome on the clinical outcome of these diseases (Chotirmall et al., 2010; Fillaux et al., 2012). In addition to well-known fungal pathogens such as Aspergillus fumigatus and Candida albicans, a variety of other fungal species have been described to be involved in pulmonary diseases (Marguet et al., 2012). In addition, the widespread use of antifungal reagents facilitates the emergence of respiratory mycoses that are resistant to common antifungal drugs, posing great challenges to therapy strategies (Vermeulen et al., 2013).

The conventional approach to diagnose fungal infection is culture of respiratory samples. This method is useful for isolating and phenotyping microorganisms but has some limitations. A majority of fungal species are uncultivable on common culture media and many are unknown fungal pathogens (Paniz-Mondolfi et al., 2012). The next-generation sequencing (NGS) approach provided an indisputable method for identifying almost all the fungal species in a sample, including those that were undetected by classical culture methods. This approach could provide a comprehensive view of fungal ecology in various environments and enable the investigation of specific microbe markers in the pathogenesis and progression of diseases (McTaggart et al., 2019). Based on NGS technology, the mycobiomes of a number of pulmonary diseases have been investigated. A study of sputum samples from four CF patients suggested that the respiratory mycobiome was involved in the development of lung pathology and that Candida was the most commonly observed genus in CF (Delhaes et al., 2012). In the analysis of both bronchoalveolar lavage fluid (BALF) lower respiratory tract samples and oropharyngeal wash (OW) upper respiratory tract samples, the respiratory mycobiome showed a markedly different pattern between lung transplant patients and healthy subjects, indicating a possible fungal role in the transplant outcome (Charlson et al., 2012). A study comparing induced sputum samples of asthma patients with control subjects suggested significantly different fungal compositions between the two groups with representative species in each group (van Woerden et al., 2013). Despite these advances in mycobiome studies, the fungal microbiota and their roles in many kinds of pulmonary disease are still not well defined.

In this study, we analyzed the mycobiome in oropharynx swab (OS) samples of 65 patients with five kinds of pulmonary disease, including 20 patients with interstitial lung disease (ILD), 14 patients with bacterial pneumonia (BP), 11 patients with fungal pneumonia (FP), 8 patients with asthma (AS) and 12 patients with lung cancer (LC), as well as 10 healthy controls (HCs), based on high-throughput sequencing of the amplified fungal internal transcribed spacer (ITS) region. In addition to comparing the differences in the mycobiome between disease groups and HCs, the similarities and differences in fungal composition among the five disease groups were also investigated. These results deepen our understanding of the fungal community in the upper respiratory tract of different pulmonary diseases, thus facilitating further investigation of the association between the mycobiome and pathogenicity of pulmonary diseases.

The human respiratory tract is a complex system that is colonized with various microbes (Zhang et al., 2017). Changes in the diversity and abundance of the airway microbiota are frequently associated with chronic inflammatory pulmonary diseases (Zhang et al., 2017). The bacterial microbiota has been investigated in some of these pulmonary diseases (Fabbrizzi et al., 2019). For instance, in the BALF of patients with ILD, a significantly higher bacterial burden was observed than in the HCs (Molyneaux and Maher, 2013). The microbiome of patients with BP is characterized by low community diversity but increased bacterial burden (Dickson et al., 2014). Although the mycobiome is less diverse and less abundant than the bacteriome, fungal infection-related pulmonary diseases have become increasingly significant clinical problems (Tiew et al., 2020; Ali et al., 2021). Studies have shown that fungi directly contribute to airway inflammation and lung function decline (Amin et al., 2010; Chaudhary et al., 2012; Chotirmall et al., 2013; Liu et al., 2013). Thus, understanding the respiratory tract fungal composition of patients with different kinds of pulmonary disease would have diagnostic and therapeutic significance.

In this study, we investigated the upper respiratory tract mycobiome in patients with five kinds of pulmonary disease and compared it with that in HCs. Based on the analyses of fungi species and abundance, the five disease groups differentiated from the HC group, but among the disease groups, the differences were much smaller. These results suggested that the upper airway tract has different mycobiome compositions between patients with pulmonary diseases and healthy people, and great similarities were shared among the different kinds of pulmonary disease analyzed in our study.

In total, in the five disease groups, Basidiomycota was the most abundant phylum, with an average abundance of 76.55%. In contrast, Basidiomycota only accounted for 40.90% in the HCs. The differences in this phylum between the disease groups and HCs were mainly due to the significant difference in the abundance of the Malassezia genus, which was much more abundant in the disease groups than in the HCs.

Malassezia is a common part of the human and animal skin mycobiota, which is able to grow at an acidic pH (Soret et al., 2020). Recently, Malassezia has been recognized as an emerging infectious pathogen (Delhaes et al., 2012). This genus has been shown to be associated with several chronic inflammatory diseases, including AS, CF pulmonary exacerbation (CFPE) and inflammatory bowel disease (IBD) (van Woerden et al., 2013; Sokol et al., 2017; Soret et al., 2020; Bello et al., 2021). In our study, Malassezia was identified as the most abundant genus in all five disease groups and was much less abundant in the HC group. The common functions of Malassezia in different kinds of pulmonary disease and its specific roles in distinct disease groups need further investigation.

In addition, among the abundant fungal genera identified in our study, many were considered to colonize or act as pathogenic agents in various pulmonary diseases. For example, Alternaria, Cladosporium, Aspergillus, and Penicillium are the predominant aeroallergens that are commonly considered to cause asthma (Hughes et al., 2022). Among these genera, Alternaria and Cladosporium are unable to grow at body temperature, and their clinical effects are related to exposure levels (Rick et al., 2016). Furthermore, the spore sizes of Alternaria and Cladosporium are higher than 5 μm, and they are mainly deposited in the upper respiratory airways (Frohlich-Nowoisky et al., 2009; Gago et al., 2019). In our study, Alternaria and Cladosporium were both identified in all six groups and had similar abundances among the different groups. Thus, we speculated that these two genera identified in the OS samples may be due to daily exposure to the environment. In contrast, Aspergillus and Penicillium are thermotolerant genera that can colonize the respiratory tract (Rick et al., 2016). Thus, they could be persistent allergenic stimuli and act as opportunistic pathogens (Woolnough et al., 2015). Aspergillus is a ubiquitous fungus to which humans are constantly exposed, and Aspergillus colonization at the lower respiratory airways is common for patients with CF, COPD, AS and even in healthy people (Tong et al., 2018; Gago et al., 2019). Colonization is a prerequisite for subsequent fungal infection, but not all patients colonized with Aspergillus will develop disease, which is mainly dependent on the immune status of the host (Gago et al., 2019). Candida is also a thermotolerant genus and is frequently found in respiratory tract specimens (Meena and Kumar, 2022). However, the clinical relevance of Candida species remains controversial (Erami et al., 2022). In our study, both Aspergillus and Candida were identified in all six groups. Although no significant differences existed, these two genera were both more abundant in the HC group than in the disease group. The roles of these two commonly observed genera in the respiratory tract and their relations to pulmonary diseases should be studied further. In contrast, Penicillium was commonly identified in all five disease groups but was not identified in HCs, suggesting its specific pathogenic role in pulmonary diseases.

One interesting genus identified in our study is Knufia, which is a group of extreme stress-tolerant species from the environment, especially rock surfaces (Tesei et al., 2017). K. epidermidis was first described in 2008, and only several isolates have been reported (Martin-Gomez et al., 2019). Most of these isolates were from feet or nails, with little clinical information (Martin-Gomez et al., 2019). Recently, in 2019, this species was isolated from a pediatric dermatological sample (Martin-Gomez et al., 2019). In our study, Knufia was identified in all five disease groups with a relative abundance between 0.1 and 1% but was much less abundant in the HCs with a relative abundance of 0.00087%. Although their natural habitat, source and route of transmission remain unclear, our results provided some potential clues for their role in pulmonary diseases.

In addition to these pathogenic species, the genus Flammulina, and one of its species, F. velutipes, were only identified in HCs with a relative abundance of 1.6%. F. velutipes polysaccharides (FVPs) have been shown to improve gut health by affecting the gut microbiota (Hao et al., 2021). In addition, F. velutipes sterols (FVSs) showed potential for the treatment of liver cancer (Yi et al., 2013). However, this genus has not previously been identified in the respiratory tract or digestive tract to the best of our knowledge. This species may also be from a dietary source, and whether it was an accidentally detected species should be explored further.

In addition to the differences in abundance, some species were specific to either the disease group or HC group, further suggesting a different fungal community in the airway tract from pulmonary diseases to the healthy state. When comparing the five disease groups, a large number of taxa were shared by different disease groups, and some species considered potential pathogens or risk factors may be common in different kinds of pulmonary disease. In addition to the common characteristics shared in different pulmonary diseases, some species also showed differences in abundance among distinct disease groups, including M. restricta, M. globose, C. railenensis and C. santamariae. Furthermore, each disease group had specific species that were not identified in any other groups.

Based on the LefSe algorithm, we identified character species in each group, which were significantly different from all the other groups. Among the above discussed species, the species F. velutipes were characterized for HCs; the genus Malassezia was significantly abundant in the BP group; and genus Penicillium was significantly abundant in the LC group. Another species, Schizophyllum commune, which can cause respiratory system infections, was significantly enriched in the ILD group (Itoh et al., 2021). The genus Pyrenophora was the character taxa in the FP group. Pyrenophora is considered a barley pathogen, and its infection in humans is unclear (Dahanayaka et al., 2021). In our study, this genus was identified in the ILD group at a low abundance and in the FP group at a much higher abundance. Whether it plays a role in respiratory tract diseases should be investigated further.

Furthermore, our observations have potential clinical implications. Some specific species showed significant correlations with patient characteristics in each disease group. Aspergillus species were suggested to be positively associated with CFPE and increased inflammatory responses (Gangell et al., 2011; Soret et al., 2020). However, some studies showed inconsistent results in that Aspergillus colonization did not show any association with lung function or radiological abnormalities (Bargon et al., 1999; de Vrankrijker et al., 2011). Among the Aspergillus species, A. fumigatus was the most reported fungal agent colonizing and/or infecting the airways of CF patients and is normally considered to have clinical significance (Bouchara et al., 2009; Pihet et al., 2009; Amin et al., 2010; Sudfeld et al., 2010). Some other Aspergillus spp., including A. niger, A. flavus, A. tubingensis, A. nidulans and A. terreus, were also frequently detected in the airways of pulmonary disease patients (Rougeron et al., 2014; Shahi et al., 2015; Gautier et al., 2016; Rick et al., 2020; Zeng et al., 2021). In our study, A. fumigatus showed a slight yet significant negative correlation with the spirometry index only in BP patients but was positively related to spirometry in LC patients. Some other Aspergillus species that were negatively correlated with lung function included A. bisporus in ILD patients, A. flavus in BP patients and A. penicillioides in FP patients. Candida are normal components of the microflora of the oral cavity, skin, respiratory tract, gastrointestinal mucosa, and genitourinary tract, and are frequently isolated from airway samples in patients with pulmonary diseases, especially in immunocompromised patients (Kumamoto and Vinces, 2005; Meena and Kumar, 2022). Isolation of Candida from respiratory specimens is usually considered to indicate colonization, but the clinical relevance of Candida colonization remains controversial (Delhaes et al., 2012; Erami et al., 2022). Furthermore, different Candida species showed distinct relationships with clinical performance. Colonization of C. albicans has been reported to lead to reduced FEV1 values in CF patients (Chotirmall et al., 2010; Francis et al., 2021). In another study of adolescent CF patients, colonization with C. glabrata was shown to be significantly associated with lower lung function, while C. albicans colonization was associated with more preserved lung function (Hector et al., 2016). In our study, Candida did not show a similar tendency among the five disease groups. C. albicans did not correlate with lung function in LC patients, and in the other four disease groups, this species showed a slight positive correlation with the lung function index. Some other Candida species, including C. railenensis and C. santamariae in BP patients, C. zeylanoides in FP and LC patients and C. tropicalis in AS patients, showed a weak yet significant negative correlation with the lung function index. These inconsistent relations of Aspergillus and Candida species to lung function may reflect the difference in distinct species and their specific roles in different disease groups. In addition, fungal spp. may not be independent factors of clinical manifestation, and they may also interact with other species. For instance, C. albicans was reported to have a synergistic effect with Pseudomonas in patients with pneumonia (Roux et al., 2013). Increased virulence of Pseudomonas, E. coli and S. aureus was observed in the presence of Candida colonization (Meena and Kumar, 2022). Furthermore, C. albicans could also promote the biofilm formation of these mixed species, facilitating their antimicrobial resistance and persistence in the respiratory tract (Morales and Hogan, 2010). A. fumigatus biofilm formation and conidial germination were also influenced by P. aeruginosa, and an antagonistic effect has been proposed between these two species (Mowat et al., 2010).

Some other species also showed significant negative relations to the lung function index. P. solitum was significantly negatively related to lung function in both the FP and AS groups. This is consistent with a previous study in which Penicillium was suggested to be negatively associated with FEV1 in CF patients (Soret et al., 2020). Cladosporium sp. and Peniophora sp. showed a relatively strong and significant negative correlation with lung function in the AS group. The relations of these species to the lung function decrease deserve further investigation.

Few studies have focused on the fungal composition of airway samples of these five kinds of lung disease. When compared to these limited data, our study validated some of the previous results but also provided some inconsistent conclusions. For instance, a lung microbiota analysis of patients with central lung cancer showed a higher abundance of Candida in healthy controls and enrichment of Malassezia in patients, which is consistent with our results (Bello et al., 2021). In an analysis of BALF in idiopathic pulmonary fibrosis (IPF) patients, the mycobiome showed no significant difference in the overall mycobiome composition between the IPF group and control group, and both groups were dominated by Candida species (Fabbrizzi et al., 2021). In contrast, our study suggested that all the disease groups were dominated by the Malassezia genus. To investigate the fungal microbiota and the significant roles of specific species in these pulmonary diseases, many more studies are still needed.

In addition, mycobiome and some particular fungal species have significant roles in the tumourigenesis and the progression of cancer (Gamal et al., 2022). For instance, among subjects with inflammatory bowel disease (IBD), mycobiota dysbiosis is considered to be a trigger factor of colorectal carcinoma (CRC), through chronic inflammation and toxic metabolites secretion, which might cause DNA damage (Vallianou et al., 2021). For the pancreatic ductal adenocarcinoma (PDA) tumor, it showed a 3,000 fold higher abundance of fungi than the normal pancreatic tissue, and the composition of the mycobiome was also distinct in tumor and normal tissue (Aykut et al., 2019). Furthermore, Malassezia spp. was markedly enriched in PDA tumors and is a specific species in oncogenic progression, by activating the complement cascade through binding mannose-binding lectin (MBL) to the glycans of the fungal wall. This study suggested that particular fungal species are sufficient to promote the progression of PDA (Aykut et al., 2019). As suggested in our study, Malassezia was also the most abundant genus in LC accounting for 70.8% of the mycobiome, which was much more abundant than in the HC that only accounted for 23.5%. In addition, M. restricta was identified as a marker species in LC group that showed significant difference from all the other five groups. In accordance with our results, Malassezia was significantly enriched in bronchial biopsy samples from patients with central lung cancer compared with HCs (Bello et al., 2021). Few is known about the relations of fungal species and the LC development, however, the roles of Malassezia spp. in LC deserve further investigations. The mycobiome could be explored for the diagnosis and prognosis marker as well as therapeutic target of cancers (Narunsky-Haziza et al., 2022).

Limitations of our study exist. First, our study used noninvasive upper airway OS samples to illustrate the airway fungal composition, which is an alternative method for monitoring airway microbiota (Marsh et al., 2016). Previous studies have suggested that although imperfect, upper respiratory tract samples provide a reliable representation of the lung microbiota (Marsh et al., 2016; Al Bataineh et al., 2020). It was hypothesized that pulmonary microorganisms first colonize the oropharynx and then enter the airway tract (Zhang et al., 2020). In spite of this, a further investigation of the fungal microbiota in the low respiratory tract, such as BALF samples, would construct a whole view of the airway mycobiome and would better facilitate identifying fungal biomarkers of different kinds of pulmonary disease. Second, each disease group in our study included a small number of samples, which may not fully represent the features of each kind of disease. A larger cohort of populations should be included to further explore airway mycobiome characteristics and validate the results in this study.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://www.ncbi.nlm.nih.gov/, PRJNA907293.

The studies involving human participants were reviewed and approved by the Ethics Committee of Beijing Chaoyang Hospital in Capital Medical University and the Ethics Committee of the Institute of Pathogen Biology, Chinese Academy of Medical Sciences & Beijing Union Medical College. The patients/participants provided their written informed consent to participate in this study.

XX and XH performed the experiments and wrote the manuscript. FD collected the samples. FY, TZ, JD, and YX analyzed the data. TL, JW, and QJ designed the research and reviewed the manuscript. All authors read and approved the submitted version.

This work was supported by the CAMS Innovation Fund for Medical Sciences (CIFMS) (grant nos. 2021-I2M-1-038 and 2021-I2M-1-039), the National Science and Technology Major Project of China (grant nos. 2018ZX10712001-002-001 and 2018ZX10711001), the Open Project of the State Key Laboratory of Molecular Developmental Biology (grant no. 2022-MDB-KF-14), the National Science and Technology Infrastructure of China (Project No. National Pathogen Resource Center-NPRC-32), and the Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences (grant no. 2021-PT310-004).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2023.1117779/full#supplementary-material