Exposure of Aspergillus fumigatus to Klebsiella pneumoniae Culture Filtrate Inhibits Growth and Stimulates Gliotoxin Production

Aspergillus fumigatus is an opportunistic fungal pathogen capable of inducing chronic and acute infection in susceptible patients. A. fumigatus interacts with numerous bacteria that compose the microbiota of the lung, including Pseudomonas aeruginosa and Klebsiella pneumoniae, both of which are common isolates from cystic fibrosis sputum. Exposure of A. fumigatus to K. pneumoniae culture filtrate reduced fungal growth and increased gliotoxin production. Qualitative proteomic analysis of the K. pneumoniae culture filtrate identified proteins associated with metal sequestering, enzymatic degradation and redox activity, which may impact fungal growth and development. Quantitative proteomic analysis of A. fumigatus following exposure to K. pneumoniae culture filtrate (25% v/v) for 24 h revealed a reduced abundance of 1,3-beta-glucanosyltransferase (−3.97 fold), methyl sterol monooxygenase erg25B (−2.9 fold) and calcium/calmodulin-dependent protein kinase (−4.2 fold) involved in fungal development, and increased abundance of glutathione S-transferase GliG (+6.17 fold), non-ribosomal peptide synthase GliP (+3.67 fold), O-methyltransferase GliM (+3.5 fold), gamma-glutamyl acyltransferase GliK (+2.89 fold) and thioredoxin reductase GliT (+2.33 fold) involved in gliotoxin production. These results reveal that exposure of A. fumigatus to K. pneumoniae in vivo could exacerbate infection and negatively impact patient prognosis.


Introduction
A wide range of microbes can be present in the lung as commensals or pathogens, and there is significant microbial diversity present [1]. There is reduced microbial diversity in the sputum of patients with acute exacerbations of chronic obstructive pulmonary disease, indicating a poor prognosis, and the presence of Staphylococcus and absence of Veillonella in sputum is associated with a high one-year mortality risk in patients [2]. In addition, coinfection with Pseudomonas aeruginosa and Aspergillus fumigatus has been implicated in a worsened disease state in Cystic Fibrosis (CF) patients, with each species stimulating the growth and colonisation of the other [3]. The bacterium Klebsiella pneumoniae and the fungus A. fumigatus are commonly isolated from the lungs of CF patients [4], indicating a similar interaction could occur, impacting disease development in patients.
K. pneumoniae is a Gram-negative rod-shaped bacteria that is responsible for approximately one-third of all Gram-negative infections worldwide [5]. Pneumonia caused by K. pneumoniae is characterised by a strong inflammatory response which is due to the production of pro-inflammatory cytokines in addition to high neutrophil and macrophage counts [6]. Within the host, K. pneumoniae is found in the gastrointestinal tract and the upper respiratory tract, where it is most frequently acquired through nosocomial acquisition [7]. The first report of a CF patient becoming infected with a colistin-resistant strain of K. pneumoniae was in 2014 [8], and since then, the number of reports of critically ill patients becoming infected with such strains has increased.
The filamentous fungus A. fumigatus is widespread in the environment and can cause three types of infection in susceptible individuals. Allergic bronchopulmonary aspergillosis (ABPA) arises due to an allergic immune response to A. fumigatus stimulating T-helper cells to recruit immune cells, particularly eosinophils, resulting in inflammation of pulmonary tissue. Patients with asthma or CF are among the most susceptible to this form of aspergillosis, and early diagnosis of the disease is important for avoiding complications such as pulmonary fibrosis [9]. Chronic pulmonary aspergillosis (CPA) occurs when pre-damaged pulmonary tissue becomes colonised by A. fumigatus [10]. The colonisation results in the formation of an aspergilloma, or fungal ball, which can result in severe haemoptysis [11]. Invasive aspergillosis is the most serious form of aspergillosis, with a one-year survival rate for solid organ transplant patients of 59%, while the rate for stem cell transplant recipients can be as low as 25% [12]. Patients suffering from neutropenia were once considered to be the main target group for this type of aspergillosis; however, non-neutropenic patients such as transplant patients, AIDS patients and ICU patients are also susceptible to infection [13].
The interaction between A. fumigatus and P. aeruginosa has been characterised previously, indicating decreased fungal growth and increased gliotoxin production in the presence of bacterial cells. In contrast, exposure of A. fumigatus hyphae to P. aeruginosa culture filtrate led to increased growth and decreased gliotoxin production [14], and secreted products of A. fumigatus can promote P. aeruginosa growth in nutrient-poor conditions [15]. K. pneumoniae is often found in polymicrobial interactions with A. fumigatus within the lung, and these interactions may be antagonistic or synergistic [16]. For example, K. pneumoniae is capable of inhibiting spore germination and hyphal development in different Aspergillus spp., and this effect was dependent upon the direct physical interaction between the bacteria and the fungus [17]. A. fumigatus and K. pneumoniae can co-exist in the lungs of immunocompromised patients, but the interaction between these two pathogens is poorly characterised. The aim of the work presented was to characterise the effect of K. pneumoniae culture filtrate on the growth and proteomic response of A. fumigatus, as this might give an insight into an important bacterial-fungal interaction in the lungs of infected patients.

A. fumigatus Growth Conditions
Aspergillus fumigatus ATCC 26933 was grown on malt extract agar plates at 37 • C for a minimum of 96 h, and conidia were harvested by washing with 0.1% (v/v) Tween-20 and phosphate buffered saline (PBS-T). Czapek-Dox liquid medium (Duchefa Biochemie) (50 mls) was inoculated with conidia at a density of 1.2 × 10 6 /mL.

Preparation of the Bacterial Culture Filtrate
K. pneumoniae ATCC 13439 was grown on nutrient agar plates for 96 h. Czapek-Dox broth was inoculated with bacteria and incubated at 37 • C and 200 rpm for 96 h. Cells were harvested by centrifugation for 15 min at 1258× g at room temperature, and the culture filtrate was filter sterilised using 0.45 µm filtropur S filters. Culture filtrates were stored at −20 • C until required.

Assessment of Fungal Biomass
A. fumigatus was grown for 48 h in Czapek-Dox broth prior to the addition of K. pneumoniae culture filtrate or PBS (n = 3) at a concentration of 25% v/v. The cultures were incubated at 37 • C and 200 rpm. The mycelia were harvested after 24 h using Miracloth (Millipore). The wet weight of the mycelia was then measured using a weighing scale and expressed as weight in grams.

Gliotoxin Extraction and Quantification by RP-HPLC
K. pneumoniae culture filtrate was added to the 48 h old A. fumigatus cultures at a final concentration of 25% v/v and incubated at 37 • C and 200 rpm for a further 24 h. Culture filtrate (20 mL) was mixed continuously with an equal volume of chloroform for 2 h. The chloroform layer was removed and evaporated to dryness, and the extract was dissolved in 500 µL HPLC grade methanol. Gliotoxin was quantified using Reverse-Phase HPLC. The mobile phase was 34.9% (v/v) acetonitrile, 0.1% (v/v) trifluoroacetic acid (TFA) and 65% (v/v) HPLC-grade deionised water. Gliotoxin extract (20 µL) was injected into an Agilent ZORBAX SB-Aq 5 µm polar LC column. A standard curve of peak area vs. gliotoxin concentration was generated using gliotoxin standards (Merck) dissolved in HPLC-grade methanol.
2.5. Extraction of Protein from K. pneumoniae Culture Filtrate K. pneumoniae was grown for 96 h in Czapex Dox broth. Cells were harvested by centrifugation for 15 min at 1258× g at room temperature. The culture filtrate was passed through a 0.45 mm filtropur S filter. The culture filtrate was centrifuged at 14,500× g for 10 min to remove any remaining debris, and protein was acetone precipitated overnight at a ratio of 1:5 culture filtrate to acetone. Samples were processed in the same manner as the fungal mycelial samples for proteomic analysis.

Protein Extraction from A. fumigatus Exposed to Bacterial Culture Filtrate
A. fumigatus cultures (48 h growth) were supplemented with K. pneumoniae culture filtrate (25% v/v) for 24 h at 37 • C in Czapek-Dox media (n = 3 per group). Hyphae were harvested by filtration, snap-frozen in liquid nitrogen and ground to a fine dust in a mortar using a pestle. Lysis buffer [8 M urea, 2 M thiourea, and 0.1 M Tris-HCl (pH 8.0) dissolved in HPLC-grade dH2O], supplemented with protease inhibitors [aprotinin, leupeptin, pepstatin A, Tosyllysine Chloromethyl Ketone hydrochloride (TLCK) (10 µg/mL) and phenylmethylsulfonyl fluoride (PMSF) (1 mM/mL)] was added (4 mL/g of hyphae). The lysates were sonicated (Bandelin Senopuls) three times for 10 s at 50% power. The cell lysate was subjected to centrifugation (Eppendorf Centrifuge 5418) for 8 min at 14,500× g to cellular pellet debris. The protein concentration was quantified by the Bradford method, and samples (100 µg) were subjected to overnight acetone precipitation.

Label-Free Mass Spectrometry (LF/MS)
Following centrifugation for 10 min at 14,500× g, acetone was removed, and the protein pellet was resuspended in 25 µL sample resuspension buffer (8 M urea, 2 M thiourea, 0.1 M tris-HCL (pH 8.0) dissolved in HPLC grade dH2O). An aliquot of 5 µL was removed from each of the samples and quantified by the Qubit quantification system (Invitrogen). Ammonium bicarbonate (125 µL, 50 mM) was added to the remainder of the samples. Reduction of the sample was initiated by adding 1 µL 0.5 M dithiothreitol (DTT). The protein samples were then incubated for 20 min at 56 • C before alkylation with 2.7 µL 0.55 M iodoacetamide; this occurred at room temperature in the dark for 15 min. Protease max surfactant trypsin enhancer (Promega) (1 µL, 1% w/v) and sequence grade trypsin (ThermoFisher Scientific, Cork, Ireland) (0.5 µg/µL), were added to the proteins, respectively, and incubated for 18 h at 37 • C. TFA (1 µL, 100%) was added to each sample to end digestion. The samples were incubated for 5 min at room temperature and centrifuged for 10 min at 14,500× g. Samples were purified for mass spectrometry using C18 spin columns as per the manufacturer's instructions. A speedy vac concentrator was used to dry the peptides, and samples were resuspended in 2% v/v acetonitrile and 0.05% v/v TFA and sonicated for 5 min to help with this resuspension. The resulting final peptide concentration was (750 ng/2 µL).

Mass Spectrometry and the Parameters for A. fumigatus and K. pneumoniae Culture Filtrate Proteomic Data Procurement
The digested K. pneumoniae culture filtrate sample (500 ng) or A. fumigatus protein samples (750 ng) were each loaded onto a QExactive (ThermoFisher Scientific) high-resolution mass spectrometer which was connected to a Dionex Ultimate 3000 (RSlCnano) chromatography system. An increasing acetonitrile gradient was used to separate the peptides in the samples. This gradient was created on a 50 cm EASY-Spray PepMap C18 column with a 75 mm diameter using a 133 min reverse phase gradient at a flow rate of 300 nL/min. All of the data were obtained while the mass spectrometer was functioning in an automatic dependent switching mode. A quantitative analysis of the A. fumigatus proteome after exposure to bacterial culture filtrate was conducted using MaxQuant version 1.5.3.3 (http://www.maxquant.org accessed on 14 September 2022) using the settings outlined previously [14]. The search algorithm Andromeda included in the MaxQuant software was incorporated in the correspondence between MS/MS data and the Uniprot-SWISS-PROT database Neosartorya fumigata reference proteome obtained from a UniProt-SWISS-PROT database to identify proteins (9647 entries, downloaded July 2022). K. pneumoniae culture filtrate was analysed through proteome discoverer v 2.5, and proteins were searched against the UniProtKB database (Klebsiella Pneumoniae strain 342, 5738 entries, downloaded September 2022).

Data Analysis of A. fumigatus and K. pneumoniae Culture Filtrate Proteomes
Qualitative analysis of the proteome of the K. pneumoniae culture filtrate was investigated using Proteome Discoverer 2.5 and Sequest HT (SEQUEST HT algorithm; Thermo Scientific). Identified proteins were searched against the UniProtKB database (Klebsiella Pneumoniae strain 342, 5738 entries, accessed 16 September 2022). Search parameters applied for protein identification were as follows: (i) enzyme name-trypsin, (ii) an allowance of up to two missed cleavages, (iii) peptide mass tolerance set to 10 ppm, (iv) MS/MS mass tolerance set to 0.02 Da, (v) carbamidomethylation set as a fixed modification and (vi) methionine oxidation set as a variable modification. Peptide probability was set to high confidence (with an FDR ≤ 0.01% as determined by Percolator validation in Proteome Discoverer). Peptides identified by 2 or more unique peptides were retained for analysis.
Perseus v.1.6.15.0 was employed to analyse A. fumigatus proteomic data, as well as to process and visualise the data. Measurement of protein abundance was based on normalised LFQ intensity values. The data was initially filtered for the removal of contaminants before LFQ intensity values were Log2 transformed, and each sample was placed in its relative group. Proteins which were not found in three out of three replicates in at least one group were excluded from further analysis. Normal intensity values were also used for principle component analysis (PCA). Proteins which were distinctively expressed in one group compared to another or those which were completely absent in one group were noted and included for all further analysis. Gene Ontology (GO) mapping was also conducted using the UniProt gene ID to gain more knowledge about the biological processes and molecular processes of the identified proteins. To gain a better visualisation of the variances between the samples, pairwise student T-tests were conducted for all data using a cut-off of p < 0.05. The generation of a volcano plot was done by plotting the log2 fold change values on the x-axis and the log p-values on the y-axis, which allowed for pairwise comparison. The top 20 most increased and decreased in abundance proteins identified in the groups were included in these volcano plots. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [18] partner repository with the dataset identifier PXD037833.

Analysis of the Effects of K. pneumoniae Culture Filtrate on A. fumigatus
Exposure of 48 h old A. fumigatus to K. pneumoniae culture filtrate (25% v/v) for 24 h led to a decrease in fungal biomass (1.23 ± 0.20 g vs. 0.81± 0.12 g p = 0.03) ( Figure 1A) and a large increase in gliotoxin secretion with a 371.55% increase (p = 0.01) ( Figure 1B).

Characterisation of K. pneumoniae Culture Filtrate
Qualitative proteomic analysis of K. pneumoniae culture filtrate revealed the presence of 160 high-confidence proteins (Table S1), of which 35 were identified as having possible antifungal effects (Table 1). These could be subdivided into six categories; structurally bound (3 proteins), membrane-associated proteins (10 proteins), virulence-associated proteins (3 proteins), proteins with enzymatic activity (8 proteins), metal binding proteins (5 proteins) and proteins with detoxification activity (6 proteins).

Analysis of the Effects of K. pneumoniae Culture Filtrate on A. fumigatus
Exposure of 48 h old A. fumigatus to K. pneumoniae culture filtrate (25% v/v) for 24 h led to a decrease in fungal biomass (1.23 ± 0.20 g vs. 0.81± 0.12 g p = 0.03) ( Figure 1A) and a large increase in gliotoxin secretion with a 371.55% increase (p = 0.01) ( Figure 1B).

Characterisation of K. pneumoniae Culture Filtrate
Qualitative proteomic analysis of K. pneumoniae culture filtrate revealed the presence of 160 high-confidence proteins (Table S1), of which 35 were identified as having possible antifungal effects (Table 1). These could be subdivided into six categories; structurally bound (3 proteins), membrane-associated proteins (10 proteins), virulence-associated proteins (3 proteins), proteins with enzymatic activity (8 proteins), metal binding proteins (5 proteins) and proteins with detoxification activity (6 proteins).

Analysis of the Effect of K. pneumoniae Culture Filtrate on the A. fumigatus Proteome
Label-free mass spectrometry was employed to characterise the proteomic response of A. fumigatus following exposure to K. pneumoniae culture filtrate for 24 h. In total, 1960 proteins were identified, and 111 proteins were identified as being statistically significant (p < 0.05) differentially abundant (SSDA), having a fold change greater than ±1.5 (Table S2). A principle component analysis (PCA) was conducted on all significant proteins to identify distinct proteomic differences between each of the groups. The PCA indicates that the A. fumigatus exposed to the K. pneumoniae culture filtrate displays a distinct proteomic pattern when compared to control samples (Figure 2A). Hierarchal clustering carried out in Perseus highlights the clear differences in protein abundance between the control and the A. fumigatus culture that was exposed to the K. pneumoniae culture filtrate. These differences in protein abundance are highlighted in a heat map ( Figure 2B), with blue indicating proteins with increased abundance and orange indicating proteins with decreased abundance, respectively.
A volcano plot was generated by way of pairwise t-tests (p < 0.05) to determine the proteins which increased and decreased in abundance between control A. fumigatus samples and A. fumigatus exposed to K. pneumoniae culture filtrate (Figure 3). When analysing samples of A. fumigatus exposed to the K. pneumoniae culture filtrate, a significant increase in the relative abundance of proteins associated with secondary metabolism, in particular, proteins associated with gliotoxin biosynthesis was observed (Table 2), i.e., glutathione S-transferase GliG (+6.17 fold increase), non-ribosomal peptide synthase GliP (+3.67 fold), O-methyltransferase GliM (+3.5 fold), gamma-glutamyl acyltransferase GliK (+2.89 fold) and thioredoxin reductase GliT (+2.33 fold). In addition, ribonuclease mitogillin was increased in abundance by +4.9 fold. Fibrinogen C-terminal domain-containing protein and polysaccharide deacetylase family protein were increased by +63.3 fold and +6.9 fold, respectively, and both have been implemented in adherence to the lung epithelium. Decreased abundance of methyl sterol monooxygenase erg25B (−2.9 fold) and calcium/calmodulindependent protein kinase (−4.2 fold) indicates a decrease in fungal cell division (Table 3). Deoxyribose-phosphate aldolase was decreased in abundance by −2.3 fold and played a role in gluconeogenesis and lipid biogenesis. Protein synthesis was also affected by a decrease in the abundance of the 60 S ribosomal protein L22 and aspartyl aminopeptidase (−2.9 fold).  A volcano plot was generated by way of pairwise t-tests (p < 0.05) to determine the proteins which increased and decreased in abundance between control A. fumigatus samples and A. fumigatus exposed to K. pneumoniae culture filtrate (Figure 3). When analysing samples of A. fumigatus exposed to the K. pneumoniae culture filtrate, a significant increase in the relative abundance of proteins associated with secondary metabolism, in particular, proteins associated with gliotoxin biosynthesis was observed (Table 2)

Discussion
The aim of the work presented here was to characterise the response of A. fumigatus to K. pneumoniae culture filtrate, as this might give an indication of the in vivo interaction between the fungus and bacteria. The culture conditions were designed to represent conditions within the immunocompromised lung by exposing A. fumigatus to the K. pneumoniae culture filtrate as opposed to bacterial cells. Czapek-Dox was chosen for this investigation as this medium provides both a low-nutrient and high-nitrogen environment, which are characteristics of immunocompromised lungs [19,20].
The results demonstrated that exposure to the K. pneumoniae culture filtrate inhibited the growth of A. fumigatus and induced increased secretion of gliotoxin. Previously, an examination of the interaction between K. pneumoniae cells and A. fumigatus demonstrated a reduction in spore germination and hyphal development which was dependent upon direct physical interaction between bacterial and fungal cells [17]. Exposure of Penicillium verrucosum to actinobacter-species cell-free supernatant also resulted in stimulation of ochratoxin A production and inhibition of fungal growth [21]. Aspergillus flavus exposed to culture filtrate of Streptomyces spp. also showed reduced growth and elevated secretion of aflatoxin B1 [22].
Analysis of the K. pneumoniae culture filtrate identified 160 high-confidence proteins (Table S1), of which 35 were identified to have possible effects on fungal development ( Table 1) and three of which (putative lipoprotein, outer membrane protein A and chaperone protein DnaK) were also bound to the fungal mycelia. Outer membrane protein A is essential for A. baumannii cell attachment to Candida albicans filaments and A549 human alveolar epithelial cells [23] and has been demonstrated to induce leaky barriers in murine lungs and human A549 cells without affecting cell viability [24]. Elongation factor Tu was identified in the K. pneumoniae culture filtrate is involved in bacterial virulence and is associated with adhesion to host extracellular matrix components. Secretion of EF-Tu increased after Heliobacter pylori infection, suggesting that H. pylori secrete EF-Tu to facilitate attachment to host cells [25].
The ability of the K. pneumoniae culture filtrate to inhibit fungal growth could have also arisen as a result of enzymatic activity. DegP-like periplasmic serine endoprotease is a highly conserved periplasmic protease found in most Gram-negative bacteria. This protease is involved in the degradation of denatured or aggregated protein within the cell envelope in E coli [26,27]. Protease VII (OmpT), an aspartic protease, was identified in the culture filtrate and is associated with the inhibition of coagulation and antimicrobial peptide production [28].
Redox-active components and enzymes in the K. pneumoniae culture filtrate could also affect fungal growth and mycotoxin production. Thiol:disulfide interchange protein DsbA plays a role in oxidising the formation of disulfide bonds [29]. Thiol:disulfide interchange protein DsbA may be implemented in the oxidation of gliotoxin's disulfide bond, increasing its biological accumulation and activity [30]. Alkyl hydroperoxide reductase C was identified in the K. pneumoniae culture filtrate and catalyses the reduction of hydrogen peroxide, organic hydroperoxides and thioredoxin in HeLa cells [31].
The non-siderophore iron uptake component EfeO was identified in the K. pneumoniae culture filtrate, and this is a component in the main ferrous iron transport system that is present in both pathogenic and non-pathogenic microbes [32]. In addition, the highaffinity zinc uptake system protein ZnuA in K. pneumoniae culture filtrate is essential in acquiring zinc at the interface between bacteria and mammalian cells and counteracting host depletion mechanisms [33]. It has been previously demonstrated that zinc limitation results in increased gliotoxin production and the growth-limiting effects of exogenous gliotoxin are relieved by the presence of zinc in media [34]. In addition, certain genes of the gliotoxin biosynthetic cluster, including gliZ, are regulated by ZafA, which is the zinc-responsive transcription factor that controls the adaptive response to zinc starvation in A. fumigatus [35]. Analysis of P. aeruginosa culture filtrate, which also affected fungal growth and secondary metabolism, identified a similar protein profile to that found in this work [14]. P. aeruginosa culture filtrate contained chaperone protein DnaK and components involved in the uptake of nutrients, including ferric iron-binding periplasmic protein HitA. Detoxification proteins such as thiol:disulphide interchange protein DsbA and thioredoxin reductase were found in both culture filtrates. Despite these similarities, K. pneumoniae culture filtrate promoted gliotoxin biosynthesis, while P. aeruginosa inhibited it.
A. fumigatus exposed to the K. pneumoniae culture filtrate demonstrated a reduction in growth and a shift towards secondary metabolism, particularly through the increased secretion of gliotoxin. Gliotoxin biosynthesis is mediated by a gene cluster consisting of 13 genes. Proteomic analysis revealed that five proteins involved in the biosynthesis of gliotoxin were increased in abundance; glutathione S-transferase (+6.17 fold), nonribosomal peptide synthase (+3.67 fold), O-methyltransferase (+3.5 fold), gamma-glutamyl acyltransferase (+2.89 fold) and thioredoxin reductase (+2.33 fold). The increase in gliotoxin production may be attributed to the physical binding of outer membrane protein A, due to its ability to induce leakage in epithelial cells [24] in a similar manner to fungistatic concentrations of amphotericin B which increases A. fumigatus permeability and stimulates de novo gliotoxin biosynthesis [36].
There was also an increase (+63.3 fold) in the abundance of fibrinogen C-terminal domain-containing protein, and this could indicate increased virulence and adhesion to lung epithelium. Asthmatic lungs have damaged bronchioloalveolar epithelium, and fibrinogen deposits form at the surface of wounded epithelia, which facilitate microbial attachment [37]. A. fumigatus also demonstrated increased binding affinity to fibrinogen compared with less pathogenic Aspergillus spp., suggesting that adhesion to the extracellular matrix may be important in disease pathogenesis [38]. Polysaccharide deacetylase family protein was increased in abundance (+6.9 fold) and is associated with adherence as proteins with similar domains, such as Agd3, are part of a group of metal-dependent polysaccharide deacetylases, which have been shown to remove Nor O-linked acetate groups from chitin, peptidoglycan, acetylxylan, and poly-β-1,6-N-acetylglucosamine, deletion of agd3 was associated with a reduced adherence and virulence in murine models of invasive aspergillosis, indicating reduced fungal burden [39].
The proteome of A. fumigatus exposed to K. pneumoniae culture filtrate also showed a reduced abundance of several proteins; 1,3 beta glucanosyltransferase was reduced in abundance by −3.9 fold and plays a role in cell wall biogenesis [40]. The membrane protein methyl sterol monooxygenase erg25B, essential for ergosterol biosynthesis [41], was reduced by −2.9 fold, calcium/calmodulin-dependent protein kinase was reduced by −4.2 fold and is essential for fungal nuclear cell division and hyphal development in Aspergillus nidulans [42]. Deoxyribose-phosphate aldolase was reduced in abundance by −2.3 old and had a role in gluconeogenesis and the glyoxylate cycle [43]. Decreased abundance of methyltransferase LaeA-like putative protein −2.16 fold could explain the downregulation of proteins associated with specific secondary metabolites [44], including helvolic acid and fumagillin. Proteins associated with helvolic acid biosynthesis, including cytochrome P450 monooxygenase helB1 (−2.8 fold), short chain dehydrogenase helC (−3.8 fold), protostadienol synthase helA (−2.8 fold) and 3-ketosteroid 1-dehydrogenase helE (−2.4 fold) were decreased in abundance and the fumagillin associated protein polyketide transferase af380 (−3.8 fold) was also decreased in abundance [45,46].
This study provides evidence that exposure of A. fumigatus to K. pneumoniae culture filtrate results in a reduction in growth but an increase in gliotoxin production and in the abundance of associated proteins. The clinical implications of these alterations could inhibit the ability of the patient to mount an effective immune response to A. fumigatus infection resulting in more severe symptoms and an increase in mortality. This work provides novel insights into an important bacterial-fungal interaction that occurs in the lungs of immunocompromised patients. Understanding the extent and importance of such microbial interactions can help to identify better therapeutic strategies for the control of pulmonary infections in susceptible patients.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof9020222/s1, Table S1: All high-confidence proteins identified in K. pneumoniae culture filtrate. Table S2: All statistically significant differentially abundant proteins identified following exposure to Klebsiella pneumoniae cell-free culture filtrate at a concentration of 25% v/v for 24 h.
Author Contributions: Conceptualisation, K.K. and A.C.; methodology, K.K. and A.C.; formal analysis, A.C. and M.R. data curation, A.C. and M.R.; writing-original draft preparation, A.C. and M.R.; writing-review and editing K.K. and A.C. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement:
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [18] partner repository with the dataset identifier PXD037833.

Conflicts of Interest:
The authors declare no conflict of interest.