Biofilm Formation in Aspergillus fumigatus: A Comparative Study of Strains from Different Origins

Abstract

One of the most notable aspects of Aspergillus fumigatus, and related to its dynamic adaptation, is its ability to form biofilm and produce a wide variety of secondary metabolites. The aim of this study is to advance the characterization of biofilms generated by different A. fumigatus strains across their developmental stages and analytically evaluate their structure and composition and their relationship with secondary metabolism activation. An in vitro biofilm model was standardized to investigate structural and analytical differences among strains isolated from distinct clinical settings and associated with different pathologies. We found that all tested strains could form biofilms; however, the characteristics of these structures—including total biomass, cellular viability and overall structure—varied markedly among strains under the evaluated conditions. Strains isolated from cystic fibrosis patients exhibited distinct behaviors in most conducted assays compared to other strains. These findings provide new insights into the variability of biofilm composition and may contribute to a better understanding of the role of biofilms in fungal pathogenesis, persistence and treatment resistance.

1. Introduction

In recent years, an increase in the incidence of invasive fungal infections (IFIs) has been observed, primarily associated with the surge in the number of susceptible patients [1]. Invasive pulmonary aspergillosis (IPA), caused primarily by species of the Aspergillus genus, is one of the most severe and life-threatening IFI, particularly affecting immunocompromised patients such as those with neutropenia, organ transplants or undergoing immunosuppressive therapy [2]. It occurs following the germination of inhaled conidia, leading to invasion of the lung parenchyma with a potential hematogenous dissemination of the fungus. The mortality rate can reach 50–90% if not diagnosed and treated promptly.

Aspergillus fumigatus, the most frequent agent of IPA is a filamentous opportunistic fungus with a well-documented ability to form biofilms (multinucleated and branched filamentous cells or hyphae), both in vitro and in vivo [3]. This capacity is considered a key virulence trait, particularly in compromised pulmonary environments such as those present in patients with cystic fibrosis, chronic obstructive pulmonary disease or immunosuppression.

Biofilms are complex mono- or multispecies communities embedded in a biopolymeric matrix (extracellular matrix, ECM) that confer reduced susceptibility to antifungal agents and host immune responses due to limited drug penetration and profound physiological adaptations, including metabolic dormancy, hypoxic microenvironments, efflux pump overexpression, and persisterpersistent cell formation [4,5,6,7]. Although the term extracellular polymeric substance (EPS) is frequently used in microbiology to refer to these structures, the term ECM is more appropriate in the context of fungal biofilms, as it reflects the complex and heterogeneous nature of this structure, which comprises not only polysaccharides but also proteins, extracellular DNA, polyols, and lipids.

The formation of A. fumigatus biofilms follows a dynamic, structured and sequential process [3,4] that reflects the dynamic adaptation and growth process of A. fumigatus which might be important for understanding its persistence and resistance mechanisms in clinical settings [3,4]. Moreover, A. fumigatus mature biofilms display unique characteristics in the composition of their extracellular matrix (ECM). This structure is believed to play a crucial role in adaptation, persistence and antifungal resistance by acting as a physical barrier that prevents host cells or drug action and hinders its diffusion into the fungal cell interior [8,9,10]. This ECM is a hydrophobic structure that coats the fungal hyphae, and its disruption in combination with antimicrobial treatment may weaken biofilms and facilitate their eradication [11].

Another notable aspect of A. fumigatus, and related to its dynamic of adaptation, is its ability to produce a wide variety of secondary metabolites (more than 226 described). These are non-essential organic compounds for fungal growth, reproduction and development but play a significant role in its defense, survival and virulence [12,13]. The structure of the biofilm is shaped by environmental conditions and affects the kinetics of secondary metabolite production since these substances could function as cell communication signals even with other microorganisms during coexistence, contributing to the formation and maturation of the biofilm structure [14]. Among this large number of secondary metabolites, one of the most studied is gliotoxin (GT), a prototype mycotoxin of the epipolythiodioxopiperazine group (ETP) [15,16]. Several studies have shown that GT plays a crucial role in the virulence of A. fumigatus as it inhibits the host’s immune response and promotes tissue invasion, potentially leading to cellular apoptosis [15,17,18]. When GT is methylated at its two thiol groups, the derivative bismethylgliotoxin (bmGT) is generated. This compound does not exhibit toxicity but is involved in modulating GT synthesis, eliminating its toxic effects through a mechanism of negative regulation [19,20].

In essence, biofilm formation is a fungal adaptation strategy for survival. Due to their high complexity, it is important to describe and understand biofilms by adopting a dynamic biological approach through qualitative and quantitative analysis of a large number of chemically diverse and multifunctional molecules. Secondary metabolites are generated because of cellular interactions and the development of distinct biofilm phases. Studying these two processes simultaneously may offer clarifying results and establish theories related to the virulence of these strains. Our group has recently reported that A. fumigatus strains can differentially activate mechanisms related to secondary metabolism during the establishment of an in vitro biofilm, although we were unable to determine whether the phenotypic characteristics of the strains might influence biofilm development and the activation of secondary metabolism [21].

The aim of this study is to advance the characterization of biofilms generated by different A. fumigatus strains across their developmental stages, analytically evaluating their structure and composition, and their relationship with secondary metabolism activation. To achieve this, a combination of analytical methods and microscopy techniques was applied to investigate the structural and compositional features associated with biofilm formation and maturation. This approach enabled a preliminary comparative analysis of strains isolated from diverse clinical and environmental origins. Understanding the molecular and physiological mechanisms underlying A. fumigatus biofilm development may contribute to the design of more effective therapeutic strategies against persistent and drug-resistant infections.

2. Materials and Methods

2.1. Strain Description

Ten different strains of A. fumigatus were used: the reference strain ATCC 204305 (American Type Culture Collection), referred to in this work as “Af2580”; the reference strain “Af293” isolated from a patient with invasive aspergillosis; 3 clinical isolates [“Af9160” (CM9160), “Af119” (CM10236), “Af135” (CM10240)] from patients with pulmonary aspergillosis; 4 clinical isolates from patients with cystic fibrosis; CF [“AfP2-2”, “AfP2-8”, “AfP5-1”, and “AfP5-13”] and 1 environmental isolate [“Af126” (CM10247)] from the collection of Mycology Reference and Research Laboratory (MRRL). All strains were subcultured on Potato Dextrose Agar (PDA, OXOID, Madrid, Spain) at 30 °C.

2.2. Preparation of Spore Suspensions

From the cultures grown on PDA agar, a spore suspension was prepared in 2 mL of 0.1% Tween 20 (Sigma-Aldrich^®, Madrid, Spain), gently scraping the culture surface with a sterile swab. The concentration of these suspensions was adjusted by spore counting in a disposable hematocytometer (Cellometer^® Revvity, Madrid, Spain) to approximately 1–5 × 10⁷ CFU/mL. The adjusted inocula were kept at 4 °C until their use.

2.3. Growth Kinetics

The growth capacity of the study strains was evaluated through a kinetic assay that allows us to monitor the changes in optical density (O.D.) over time and establish differential growth patterns among strains. For this purpose, strains were grown in a 96-well flat-bottom polystyrene microdilution plate (Falcon^®, Madrid, Spain). This plate was incubated for 72 h at 37 °C in the Infinite^® MNano Reader (Tecan Diagnostics, Männedorf, Switzerland), which measured the O.D. (wavelength 535 nm) every 30 min. The assay was performed in triplicate. The data generated were exported to GraphPad Prism software v 9.0.2, appropriately tabulated, plotted, and visually inspected.

2.4. In Vitro Biofilm Generation

Biofilms were developed in 24-well flat-bottom polystyrene microdilution plates (Falcon^®, Madrid, Spain). Each well was filled with 500 μL of RPMI medium supplemented with 2% glucose (w/v) and buffered with MOPS (Sigma-Aldrich^®, Spain), along with 500 μL of a pre-prepared spore suspension from each of the 10 strains included in the study. Two inoculum sizes, 10⁴ CFU/mL and 10⁶ CFU/mL, were tested to assess the effect of inoculum on biofilm formation. Plates were incubated at 37 °C for the designated time points of the study (24, 48, and 72 h). RPMI-2%G was consistently prepared and controlled across all assays.

2.5. Biofilm Characterization

2.5.1. Measurement of Biofilm Cell Viability with XTT

For biofilm characterization, a technique based on the indirect measurement of metabolic activity was employed [22]. This method, used to determine cell viability, is based on the ability of the salt 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-tetrazolium hydroxide (XTT) to convert into a water-soluble formazan derivative in the presence of mitochondrial activity. This formazan derivative can absorb light in the visible spectrum (430–490 nm). Thus, the absorbance measured at each time point will be proportional to the amount of live or metabolically active cells present. After incubating the plates as described in Section 2.4, each well’s supernatant was carefully removed. The residual well content (biomass) was then washed twice with PBS using 500 μL/well; then, the same volume of this buffer was added. Finally, 250 μL of the XTT reagent (Cell Proliferation Kit II, Roche Diagnostics^®, Madrid, Spain, prepared following the manufacturer’s instructions by mixing 125 μL XTT labeling reagent and 2.5 μL Electron coupling reagent) was added to each well. Distilled water was used as negative control, following the same procedure.

The plates were kept in the dark at room temperature for 4 h; afterwards, the optical density (O.D.) was measured at 490 nm using an Infinite^® MNano reader (Tecan Diagnostics, Männedorf, Switzerland). The XTT reagent final concentration and incubation time were selected after a prior standardization process following instructions, as they provided an adequate signal to discriminate cell viability.

2.5.2. Quantification of Biomass with Crystal Violet (CV)

Crystal Violet (Sigma-Aldrich^®, Spain) is a non-specific fluorochrome that binds to negatively charged surface molecules and polysaccharides of the ECM, allowing the quantification of biofilm biomass. For this purpose, the method described by Domenech et al. was employed, with modifications [23].

After incubating the plates, the supernatant from the wells was removed and collected in eppendorf tubes for subsequent analysis. The plates were air-dried for 45 min and then 1000 μL of a 0.2% CV solution in distilled water was added, followed by incubation at room temperature for 15 min. After the incubation period, the CV solution was carefully removed, and the wells were washed three times with 1000 μL of distilled water. The biofilm was decolorized with 1000 μL of 96% ethanol and incubated at room temperature for 10 min. Finally, the 1000 μL of ethanol was transferred to a new plate to measure the O.D. (620 nm) using the Infinite^® MNano spectrophotometer (Tecan Diagnostics, Männedorf, Switzerland). Distilled water was used as negative control, following the same procedure.

2.5.3. Characterization of Secondary Metabolism Activation in Different Biofilm Phases: Analysis of Secondary Metabolites

The ability of the strains to activate secondary metabolism and generate compounds that promote adaptation during the biofilm development stages was evaluated. For this purpose, samples of supernatants collected at different times of biofilm growth (24, 48 and 72 h) and stored at −20 °C, were processed following the liquid–liquid extraction protocol with chloroform described below.

To 200 μL of each supernatant, 400 μL of acetonitrile (AcN), the precipitating agent, was added. The mixture was vortexed and centrifuged at 12,000 rpm for 10 min. The supernatant was transferred to a new tube and 400 μL of chloroform was added, vigorously mixed using a vortex mixer. Subsequently, a new centrifugation (12,000 rpm, 5 min) was performed, allowing the incorporation of the target metabolites, GT and bmGT, into the organic chloroform phase. This way, three distinct phases were obtained and the clarified chloroform phase was transferred to a new glass tube. This extraction was repeated two more times, gathering all the organic extracts in the same tube. Using the Concentrator plus evaporator equipment, Vacufuge^® plus (Eppendorf, Madrid, Spain), and the D-HV program at 60 °C for 70 min, the organic extracts were evaporated to obtain a dry residue. Subsequently, this residue was re-suspended in 100 μL of an AcN/H₂O mixture (50:50). The samples were vortexed to integrate the metabolites into the mixture and filtered. The content was transferred to UPLC vials for analysis.

These prepared suspensions were analyzed using a previously developed reversed-phase chromatographic method in the laboratory. For this purpose, the ACQUITY UPLC^® HSS, C₁₈, 1.8 μm, 2.1 × 75 mm column (Waters Chromatography^®, Barcelona, Spain) was used. The separation of different components was carried out using a gradient method in successive steps with a constant flow rate of 0.1 mL/minute, with the mobile phase being a mixture of AcN and Mili-Q water. The proportion of AcN varied within a range of 65% to 40%, with a higher proportion in the initial steps.

Characterization of GT and bmGT was achieved using a matrix photodiode detectorACQUITY PDA detector (Waters Chromatography^®, Barcelona, Spain) which allows establishing specific UV profiles and facilitates the identification of each analyte separately, as well as quantifies low concentrations of analytes in samples that are difficult to determine using conventional UV detection methods. Amount of GT and bmGT, measured by chromatographic peak area, was previously standardized by evaluating the detector signal (peak area over an analytical concentration range from 0.125 to 8 mg/L, prepared in an AcN/H₂O solution (53:47 v/v) from stock solutions in DMSO).

2.5.4. Characterization of the Extracellular Matrix

In 6 cm diameter Petri plates (Deltalab, Madrid, Spain), 5 mL of spore suspension containing 10⁴ CFU/mL and 10⁶ CFU/mL of the previously mentioned strains was inoculated in 5 mL of RPMI medium with 2% glucose and buffered with MOPS (10 mL/well in total). The plates were then incubated for 72 h at 37 °C. Once the mature biofilm was formed, the supernatant was extracted and subsequently mixed with five times the volume of ethanol cooled to 4 °C in a 15 mL Falcon^® tube. The mixture was kept at 4 °C for 5 days to promote the precipitation of the ECM [24]. Next, the different tubes were centrifuged at 10,000 g for 10 min at 4 °C, the supernatant was removed and the resulting precipitate was washed with sterile deionized water. The precipitate was kept frozen at −20 °C for subsequent lyophilization at −50 °C overnight (LyoQuest, Telstar^®, Madrid, Spain). The obtained lyophilized material was weighed and properly processed for further analysis.

Analysis of protein content of the ECM: For the estimation of the protein content of the ECM, the lyophilized matrices were dispersed in Mili-Q water (1 mg/mL). Subsequently, the amount of proteins present in the ECMs was quantified using the Bradford method, with bovine serum albumin (BSA) as a calibration standard. This method is based on the color change in Coomassie G-250 dye depending on the protein concentration. This dye binds to proteins in an acidic medium, causing the color to change from brown to blue [25].

For this purpose, 200 μL of the diluted Bradford reagent (1:4) along with 100 μL of the dispersed matrices were added to each well of a 96-well plate. The mixture was kept in the dark at room temperature for 5 min. Then, the absorbance at 595 nm was measured using the Infinite^® MNano Reader spectrophotometer (Tecan Diagnostics, Männedorf, Switzerland).

Analysis of presence of galactomannan in the ECM: Similarly, the presence of galactomannan was determined using the PlateliaTM Aspergillus Ag kit (Bio-Rad Laboratories, Madrid, Spain), a sandwich-type enzyme-linked immunosorbent assay (ELISA) that enables the detection of Aspergillus galactomannan antigen, following the manufacturer’s instructions.

2.5.5. Microscopic Study of Biofilm Structure

Fluorescence microscopy (Leica Thunder Imager, Leica Microsystems^®, Wetzlar, Germany) was used to perform a differential analysis of biofilm structure at various developmental stages. Following a prior standardization process of experimental conditions and staining protocols, specific strains and fluorescent dyes yielding the most consistent and informative results were selected, as detailed below.

A biofilm was developed in a 96-well microdilution plate (Falcon^®, Spain) containing RPMI 2% glucose buffered with MOPS (Sigma-Aldrich^®, Spain) and an inoculum of 10⁴ CFU/mL of the strains: Af293, Af2580, Af9160, Af126, AfP2-2, AfP2-8, AfP5-1 and AfP5-13. Three biofilm development times were analyzed for each strain (8, 12 and 72 h at 37 °C in an incubator). After the incubation time, the biofilm growth was halted and, following the removal of the medium and three successive washes (100 μL per well of PBS), the resulting structures were fixed with 200 μL of 4% para-formaldehyde (PFA) for 30 min in darkness at room temperature. After fixation, the supernatant was removed and all wells were washed with 200 μL of PBS.

FUN™ 1 Cell Stain (Invitrogen, Thermo Fisher Scientific, Madrid, Spain), a two-color fluorescent viability dye for yeast and filamentous fungi, was selected because it reported high-quality microscopic images. This stain diffuses into the cytoplasm, staining it with green fluorescence. Later, after the stain is taken up by live cells, the formation of vacuolar structures occurs, which exhibit red fluorescence, reducing the green fluorescence in the cytoplasm. In this case, after incubating the biofilms at the corresponding times, the different wells were not fixed, as this staining requires live cells.

A total of 100 μL of a working solution of FUN1 (5 μM) was added, mixing carefully with the pipette. The plate was incubated at 37 °C in darkness for 30 min and then washed once with PBS.

Finally, biofilms were visualized using the Leica Thunder Imager fluorescence microscope (Leica Microsystems^®, Wetzlar, Germany)and photographs were taken with the 10× objective. The images were processed using Fiji/ImageJ softwarev1.53t.

2.6. Statistical Analysis

The numerical data generated in the study were subjected to statistical analysis using the GraphPad Prism software (9.0.2. version). Biomass, cell viability, protein and galactomannan data were compared using one-way analysis of variance (ANOVA) for independent data. The level of statistical significance was set at p < 0.05.

3. Results

3.1. Differential Phenotypic Characterization of Developed Biofilms

After a preliminary analysis of the growth kinetics, all strains followed a sigmoidal growth pattern and three distinct phases were identified, regardless of the inoculum used: a latency phase, exponential growth phase and stationary phase. Generally, differences in strain growth were observed based on the tested inoculum size (Figure 1A, and in supplementary material (File S1: Excel file with data and graphs) for further details Supplementary File S1). With the lower inoculum (10⁴ CFU/mL), strains exhibited a longer latency phase (up to 10–12 h, calculated as the time during which O.D.₅₃₅ does not increase) and an extended exponential (active growth) phase. Conversely, the kinetic behavior of strains with a higher inoculum (10⁶ CFU/mL) showed a shorter latency phase (5–6 h), reduced exponential growth and a more pronounced stationary phase (Figure 1B,C).

Figure 1.

Representative kinetic profile of selected strains. (A) A. fumigatus 293 and two inocula tested (10⁴ and10⁶ CFU/mL). (B) A. fumigatus strains showing a sigmoidal growth pattern with the lower inoculum (10⁴ CFU/mL). (C) A. fumigatus strains from cystic fibrosis patients (AfP2-2, AfP2-8, AfP5-1, and AfP5-13). The graphs show the mean optical density (O.D.) values based on experiments performed in triplicate.

Regarding the individual growth of the strains, several specific behaviors should be highlighted. Strains isolated from CF patients (AfP2-2, AfP2-8, AfP5-1 and AfP5-13) exhibited a shorter latency phase compared with the remaining strains, regardless of the inoculum used.

3.1.1. Biofilm Cell Viability with XTT

To assess differences in biofilm-associated cell viability among strains, a comparative analysis was conducted at three incubation time points (24, 48 and 72 h). All strains demonstrated a time-dependent increase in viability under both inoculum conditions (10⁴ and 10⁶ CFU/mL; Figure 2A,B). The longer the incubation time, the greater the biofilm development and cell viability observed for all the strains (72 h >> 24 h). Comparatively, strains AfP2-2, AfP2-8, AfP5-1 and AfP5-13 (isolated from CF patients) consistently exhibited significantly higher viability levels when a higher inoculum size was used (Figure 2; p < 0.001, one-way ANOVA). It is important to note that strain Af293 showed the lowest viability at the final time point (72 h) under both inoculum conditions and compared to the other strains.

Figure 2.

Quantification of cell viability using the XTT assay (A,B) and biomass using CV (C,D) for all strains at different stages of biofilm development (24, 48 and 72 h). ns: non-statistically significant differences, * (p < 0.05).

3.1.2. Quantification of Biomass with Crystal Violet (CV)

Total biofilm biomass was quantified using CV staining. Measurements were taken at 24, 48 and 72 h of incubation for all strains included in the study. Data showed a general increase in biomass over time (Figure 2C,D), consistent with the trend observed in the cell viability assay. Biofilm biomass was compared using one-way ANOVA, which revealed no statistically significant differences among strains (p = 0.997). All strains, regardless of their origin, developed denser biofilms (with higher biomass) when the lowest inoculum (10⁴ CFU/mL, Figure 2) was used. No significant differences in biomass were observed between strains AfP2-2, AfP2-8, AfP5-1 and AfP5-13 and the remaining strains, although, as previously described, strain Af293 consistently exhibited the lowest biomass values, particularly at T = 72 h.

3.1.3. Quantification of Secondary Metabolites

Figure 3 shows the amount of GT and bmGT (chromatographic peak area) detected in the biofilm culture medium for each strain, inoculum size and development time. The production of these secondary metabolites exhibited time, inoculum and strain-dependent variation. GT was detected at earlier time points, whereas bmGT was only measurable after 48 and 72 h of biofilm development for most of the strains (except Af9160) and the lower inoculum. We also observed that a higher inoculum size was associated with biofilms showing increased production of secondary metabolites (10⁶ CFU/mL, higher GT and bmGT levels except Af293 and Af135).

Figure 3.

(A,B) Gliotoxin, GT and (C,D) bismethylgliotoxin bmGT content detected in biofilm supernatants generated with inoculum of 10⁴ CFU/mL and 10⁶ CFU/mL, incubated for 24, 48 and 72 h in 6 cm diameter Petri dishes. Each value represents the mean chromatographic peak areas of three independent experiments. Error bars show experimental variability.

Significant differences in GT production were observed among some strains (p < 0.01, one-way ANOVA). Notably, the GT production profiles of strains Af293 and Af9160 differed markedly from those of AfP2-2, AfP2-8, AfP5-1 and AfP5-13, which were isolated from CF patients. For these strains, GT production was significantly reduced, except for AfP2-8 and AfP5-1, which showed minimal GT production with the highest inoculum.

A similar trend was observed for bmGT: peak areas were substantially lower for the CF isolates and no bmGT was detected with the lower inoculum. Af9160 strain exhibited the highest bmGT peak at all time points and inoculum size. In contrast, bmGT was detectable as early as 24 h when the higher inoculum was used, with peak areas increasing proportionally over time and exceeding those measured under the 10⁴ CFU/mL condition.

3.1.4. Characterization of the Extracellular Matrix (ECM)

The amount of ECM extracted was highly variable. Overall, and with few exceptions, biofilms formed from the higher inoculum (10⁶ CFU/mL) produced a greater amount of ECM (4.5 to 38.1 mg and 5.8 mg to 32 mg) under the experimental conditions described.

The quantification of galactomannan content in the ECM (Figure 4A) revealed that all strains exhibited comparable levels of this polysaccharide, with index values ranging between 14.6 and 16.8. These indices were calculated by dividing the optical density (O.D.) of each sample by the mean O.D. value of the cutoff control provided by the commercial detection kit, indicating a consistent galactomannan contribution to the ECM across strains and inoculum sizes.

Figure 4.

Quantification of extracellular matrix components in (A) galactomannan content and (B) protein concentration (μg/mL), comparing biofilms generated with the two inoculum sizes (10⁴ and 10⁶ CFU/mL) across the different A. fumigatus strains. Protein levels were determined using the Bradford assay. Bars represent the mean ± standard deviation of three independent experiments. (* p < 0.05).

The differential analysis of the relative protein content in the ECM revealed differences among the studied A. fumigatus strains. Overall, biofilms formed using the lower inoculum (Figure 4B, 10⁴ CFU/mL) exhibited a higher protein concentration in the ECM compared to those formed with the higher inoculum (10⁶ CFU/mL).

Similarly, the environmental strain Af126, and the clinical isolates Af 119 and Af 135 exhibited the highest protein content among all evaluated strains (Figure 4B, p < 0.01; one-way ANOVA), indicating a potentially enhanced protein-rich matrix phenotype under the tested conditions.

3.1.5. Microscopic Characterization of Biofilm Structure

The microscopic biofilm structure formed by the different A. fumigatus strains, corresponding to distinct developmental stages (8, 12 and 72 h), is shown in Figure 5. The images were acquired using a 10× objective, allowing detailed visualization of the structural evolution of the biofilm over time for each strain.

Figure 5.

Biofilm structure incubated at 3 different development times (8, 12 and 72 h) of selected strains (Af293, Af2580, Af9160, Af126, AfP2-2, AfP2-8, AfP5-1 and AfP5-13) using fluorescence microscopy (10× objective) following staining with the viability dye FUN1. In all strains, the different stages of biofilm development were shown. SVfs with red staining were mainly noticeable after 72 h of incubation.

Figure 5 displays the biofilms of reference strains Af293 and Af2580, the clinical isolate Af9160, the environmental strain Af126, and four isolates from CF, stained with the fluorescent viability dye FUN1. Progressive changes in biofilm density and hyphal complexity can be observed over time, with marked differences in terms of biomass accumulation and three-dimensional organization among strains. In all of them, the three stages of biofilm development were clearly observed: at 8 h, the initiation of spore germination was noted; the transition to immature biofilm was achieved at 12 h, where hyphae expanded to form networks; finally, mature biofilm was obtained at 72 h, with an extensive network system spreading between different layers.

Although the spore germination pattern observed at 8 h of incubation appear similar in most strains, a notably higher number of germinated conidia was observed for the strains isolated from CF patients (specially Af P2-8 and Af P5-13); in the immature biofilm phases (12 h), variable mycelial density was found (Af9160 showed lower development). For all strains, small vacuolar formations (SVfs) with red staining were noticeable at 72 h of incubation (mature biofilm). This indicates that the live cells present in the biofilm could process the green stain present in the cytoplasm.

4. Discussion

Mycelial development in the form of biofilm, and the activation of secondary metabolism in A. fumigatus, are interconnected mechanisms that play a crucial role in its virulence and pathogenicity. Therefore, their combined analysis, as proposed in this work, provides a valuable framework for understanding the specific dynamics underlying fungal infection.

The analysis of biofilm-forming capacity across different laboratory conditions and among various A. fumigatus strains associated with distinct pathologies further reinforces the structural complexity and heterogeneity of this developmental mode. Previous studies have shown that biofilm structure and composition can influence the nature and severity of the resulting pathology [11,26]. In agreement with these observations, our results confirm that all tested strains could form biofilms; however, the characteristics of these structures—including total biomass, cellular viability and overall structure—varied markedly among strains under the evaluated conditions. These strain-dependent differences underscore the biological diversity of A. fumigatus and highlight the importance of considering both environmental and genetic factors when interpreting biofilm-associated pathogenicity. Some parameters that have been studied here include two different inoculum sizes (10⁴ CFU/mL and 10⁶ CFU/mL) and different incubation time points. Several authors have considered these parameters critical for improving assay reproducibility [27].

Biofilms generated from a lower spore population (lower inoculum, 10⁴ CFU/mL) exhibited higher cellular viability, as assessed by the XTT assay, as well as increased biomass according to the CV staining assay. These findings suggest that lower inoculum sizes appear to facilitate the development of thicker and more metabolically active biofilms, consistent with previous studies [27,28]. In contrast, and under similar nutrient conditions, a higher inoculum may limit metabolic activity and structural development, as reflected by the characteristics of the biofilms formed using the 10⁶ CFU/mL inoculum. Similarly, the kinetic behavior displayed by the strains when a higher number of cells was used per well (10⁶ CFU/mL) supports this hypothesis, showing shorter latency phase, reduced exponential (active growth) phase and a more pronounced stationary (slowed growth) phase (Figure 1B,C).

This phenomenon could be influenced by the release of active signaling molecules (quorum sensing), which might play a key role in biofilm regulation and development [29]. Keown et al. reported that the Pseudomonas aeruginosa quorum sensing (QS) system modulates its interaction with A. fumigatus. These signals regulate the production of virulence factors, siderophores and other toxic compounds that influence the survival, growth and biofilm formation of the latter in coinfection contexts. In this context of orchestrated signaling, the activation of secondary metabolism and the subsequent release of secondary metabolites may contribute to biofilm development and structural adaptation. In this study, we focused on GT, a known virulence factor of A. fumigatus, and its inactive derivative, bmGT. We found that inoculum size shapes the development and final structure of the biofilm, while also playing a role in the activation of secondary metabolism. Notably, under conditions of higher inoculum and reduced metabolic activity, both GT and bmGT were detected at earlier time points, although GT remained the predominant compound. These findings support the notion that GT production may play a key role in biofilm establishment and adaptation under stress or nutrient-limiting conditions. However, the results revealed strain-dependent differences in the kinetics of secondary metabolite production, which may affect their distinct pathogenic profiles, including variations in invasiveness and virulence. GT production occurred earlier and in greater amounts than bmGT in most of the strains analyzed, highlighting its relevance in early stages of biofilm formation.

Of note, strains isolated from CF patients exhibited significantly reduced GT and bmGT production compared to the rest, highlighting a potential role of GT in the biofilm development of these strains. Some previous studies have reported distinctive features in A. fumigatus strains isolated from CF patients, including differences in biofilm formation and antifungal susceptibility, high genetic diversity and complex interactions with co-colonizing bacteria (particularly Pseudomonas aeruginosa) [30,31,32]. However, there is no clear evidence that CF strains are universally “less virulent,” as one might infer from the lower GT production observed in our study. It has been speculated that A. fumigatus strains from CF patients may possess specific adaptations to the unique microenvironment of the CF airways (mucus, hypoxia, microbial interactions), although it is not yet fully understood how these adaptations affect their growth.

The observation of differences in growth patterns (shorter lag phase and longer exponential phase), biofilm structure and GT/bmGT production in a controlled in vitro model among the strains evaluated in this study highlights the unique characteristics of these isolates and suggests that the previously reported adaptations may be stable and play an important role in their virulence or persistence.

On the other hand, the differences in behavior among the Af2580, Af293 and Af9160 strains have already been reported in previous studies by our group. This study confirms the “high-producer” (HP) GT pattern for the Af9160 and Af293 strains compared with the others [21,33]. These phenotypes correspond to different patterns of activation of the associated molecular machinery, so further studies in this direction are needed to confirm the significance of these findings. Future in vivo screening assays using the Galleria mellonella model will allow us to characterize the differential virulence capacity among strains.

Considering the critical role of the ECM in A. fumigatus biofilms, numerous studies have focused on characterizing its biochemical composition. In the present study, we specifically investigated two key ECM components (proteins and galactomannan) to better understand their contribution to biofilm structure and variability among strains.

Higher protein content was obtained in ECMs isolated from biofilms formed with the lower inoculum (10⁴ CFU/mL), further supporting the hypothesis that larger inocula lead to reduced biofilm development due to limitations in nutrient availability and space. Published studies related to the ECM content of A. fumigatus show a protein content close to 40% and 43% of total polysaccharides (galactomannan, galactosaminogalactan and α-1,3 glucans). However, it has been observed that this proportion of polysaccharides might be variable [10]. The variability in the amount of extracted ECM and in its composition highlights the complexity involved in recovering this structure from each of the biofilms generated.

The ECM is difficult to completely solubilize, and during the extraction processes, it may undergo degradations or alterations that affect the estimations of its real composition. The variability in the amount of ECM obtained and in the determination of its components within experiments highlights the challenges of this characterization. Ethanol precipitation followed by lyophilization has shown low reproducibility and, in this context, the role of individual ECM components in the characteristics of A. fumigatus biofilms needs to be better defined.

The results show a similar galactomannan composition among the strains evaluated, with subtle differences that do not follow a defined pattern in the protein content of the ECM (one environmental strain and two clinical strains). Currently, there is no conclusive evidence indicating that environmental A. fumigatus strains possess overall adaptive advantages over clinical strains in terms of general pathogenicity. However, environmental strains may exhibit specific adaptations, such as azole resistance acquired in environmental settings or increased genetic diversity, which enable them to survive and proliferate under diverse selective pressures. These traits may contribute or facilitate their ability to cause infection. Considering the variability in the ECM extraction process and in the methodologies used to specifically characterize these compounds, further studies will be conducted to confirm these results, which we consider preliminary but reveal an interesting trend worth investigating.

Various methods have been used to characterize the ECM of A. fumigatus, allowing the understanding of its composition and three-dimensional structure (architecture), including colorimetric, microscopy-based techniques, and spectrophotometric methods [11]. These methods have used either mild detergent (or none), combined with gentle sonication to separate the cells from the matrix material. The use of more specific techniques such as liquid chromatography with post-column derivatization or mass spectrometry is anticipated, allowing the characterization of the protein and polysaccharide composition and the specific composition of the ECM in different strains. Several authors have used staining with Concanavalin A Alexa-Fluor 488 conjugate (CAAF) which, by binding to α-mannopyranosyl and α-glucopyranosyl residues of polysaccharides, allows for semiquantitative measurement of ECM fluorescence in a plate assay [34,35]. Other studies have employed different stains such as FUN1, Calcofluor White, Nile Red, fluorescein isothiocyanate (FITC) or SYTO 9 and confocal laser microscopy [36]. Other commonly used techniques include scanning electron microscopy (SEM), high-performance liquid chromatography (HPLC) and nuclear magnetic resonance (NMR) [10,37].

After microscopic characterization of biofilm structure with FUN-1 staining, the three development phases were clearly observed, with an increase in mycelium density and viability over time, consistent with the different phases previously described [3]. This is proof of metabolic activity, confirming that these are living structures, as observed in other studies [4]. However, strains isolated from CF patients appear to exhibit a higher number of germinated spores at 8 h compared to the rest of the isolates, despite using the same inoculum size. This suggests that biofilm formation in these strains begins at an earlier stage. Preliminary analysis of the biofilm using a 10× objective and a specific stain such as FUN-1 allows for an initial assessment of its structure and for identifying and/or demonstrating intra-strain differences; however, this should be complemented with more precise and detailed observations using advanced microscopy techniques. In the present work, we limited our approach to a qualitative description of the images without quantitative measurements. The use of specific stains and fluorescence microscopy highlights the differential structure of the biofilm formed at the various stages evaluated, and although it does not allow for a reliable quantitative assay, it can provide strain-specific details that lay the groundwork for more in-depth studies of the structure and cellular-level composition of the developed biofilm.

This study has several limitations. It did not include a detailed investigation of the biofilm’s structure or the genetic basis of its formation. In addition, the contribution of other factors influencing biofilm development was not explored. Despite these limitations, the findings of this work provide evidence to guide future investigations into biofilm development in A. fumigatus. The in vitro model developed and the characterization techniques described allow for the detection of differences between phenotypically distinct strains. These preliminary observations warrant further investigation of strains displaying distinctive features, including three-dimensional biofilm structural analysis using specific staining, confocal laser scanning microscopy, and image analysis, as well as characterization of the ECM protein composition through proteomic approaches. In addition, differential gene expression analyses focusing on the GT biosynthetic gene cluster would provide further insight into the underlying molecular mechanisms.

5. Conclusions

The methods employed for the phenotypic characterization of an A. fumigatus biofilm enable a comparative evaluation of growth, cell viability, biomass, and preliminary aspects of ECM composition and biofilm structure. All strains tested (reference, clinical and environmental) showed distinct biofilm-related behavior, which were notably influenced by inoculum size, incubation time, and isolate origin. The biofilm formation trigger A. fumigatus to activate its secondary metabolism, releasing compounds like GT, following a time- and strain-dependent pattern. GT production occurs in early stages and differs by strains. In particular, strains isolated from cystic fibrosis patients exhibited distinctive responses across most assays when compared with other isolates. Taken together, these observations underscore the need for further studies to better define the biological relevance and potential clinical implications of these phenotypic differences.

Abbreviations

The following abbreviations are used in this manuscript:

AcN	Acetonitrile
bmGT	Bismethylgliotoxin
CFU	Colony forming unit
CF	Cystic fibrosis
CV	Crystal Violet
ECM	Extracellular matrix
ETP	Epipolythiodioxopiperazines
GT	Gliotoxin
IFI	Invasive fungal infections
IPA	Invasive pulmonary aspergillosis
O.D.	Optical density
PFA	Paraformaldehyde
QS	Quorum sensing
SD	Standard deviation

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/microorganisms14020272/s1, File S1: Representative growth curves (OD535 nm) of selected strains.

Author Contributions

Conceptualization, A.G.-L.; methodology, A.G.-L., M.C.-P., J.d.D.C.P. and E.G.G.d.l.P.; validation, A.G.-L.; formal analysis, A.G.-L. and M.C.-P.; investigation, M.C.-P.; resources, A.G.-L. and M.C.-P.; data curation, A.G.-L. and M.C.-P.; writing—original draft preparation, M.C.-P. and A.G.-L.; writing—review and editing, A.G.-L., J.d.D.C.P., E.G.G.d.l.P. and M.C.-P.; visualization, M.C.-P.; supervision, A.G.-L.; project administration, A.G.-L.; funding acquisition, A.G.-L. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

We used microorganisms’ strains (A. fumigatus) that were retrospectively selected from an anonymized strain collection preserved in our laboratory and no patient-related details were reported for the study. Approval from the corresponding ethics committee was not required, as no new samples or data from human subjects were collected or analyzed specifically for this research.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This work was supported by Instituto de Salud Carlos III (Research Project AESi PI21CIII/00012, MPY MPY 435/21; Government of Spain). AGL belongs to Center for Biomedical Research in Network in Infectious Diseases (CIBERINFEC), CB21/13/00105, ISCIII, Madrid, Spain.