1. Introduction
Fungi are one of the most abundant and ecologically diverse microbial groups in the atmosphere [1–3]. The atmosphere harbors a diverse array of airborne fungal genera, including Aspergillus, Aureobasidium, Cladosporium, Penicillium [4, 5]. Fungi are not only critical contributors to ecosystem functioning but also pose risks to animal and plant health. Airborne fungi can trigger human respiratory and immune system issues including asthma and allergies and also cause crop diseases, considerably compromising yields [1, 6, 7]. Airborne fungal spores are the predominant fungal material in the atmosphere, with typical spore sizes ranging from 2 to 30 μm diameter, although variations exist depending on species and environmental conditions [2, 6, 8]. Their concentrations generally range from 103 to 105 spores per cubic meter, depending on season and weather conditions [2, 6].
To study airborne microorganisms, a variety of aerosol collection methods, including impaction, filtration, and liquid impingement, have been developed [2, 9, 10]. Among these, filtration is one of the most commonly used approaches for its simplicity and efficacy in capturing aerosols from air [11]. Porous filters made from various materials trap particles above a defined size threshold, which can then be analyzed using culture-based, molecular, or microscopic methods [9–11]. Filtration offers efficient collection of diverse bioaerosols, including bacteria, fungi, and microscopic eukaryotes, and enables the recovery of both culturable and non-culturable airborne propagules while preserving viability. The success of filtration-based methods depends on effective recovery of microorganisms from the filters [12]. Despite their advantages, filtration-based methods face challenges in bioaerosol sampling. One important challenge is the physical entrapment of microorganisms within the filter matrix, which can hinder microbial recovery and downstream analyses [11]. Additionally, dehydration and other mechanical stresses during filtration, influenced by the flow rate and duration, can lead to loss of viability [10, 13, 14].
Understanding and addressing the factors that influence recovery and viability are essential for refining filter-based methods to ensure accurate analysis of microbial communities. Microbial recovery is affected by multiple factors, including filter material, propagule properties, filtration conditions, and recovery methods [9, 15, 16]. For example, Wang, Reponen [13] found that the viability of Penicillium melinii and Aspergillus versicolor spores was affected by filtration time. Fungal spores, bacterial endospores, and vegetative cells differed in their susceptibility to filtration stress. Griffin, Gonzalez [14] observed a decreased recovery of viable bacterial and fungal cells with prolonged filtration times. Jeong and Kim [17] found that DNA recovery efficiency from bacterial consortia and bioaerosols depended on the filter material used. Additionally, propagule size can affect deposition and retention within the filters, thereby influencing recovery efficiency [18].
Given the considerable size variability of airborne microorganisms, propagule size can be a critical factor impacting the recovery efficiency of filtration-based methods. Understanding how propagule size impacts filtration results is essential for optimizing sampling and recovery processes. In this study, we aimed to evaluate the influence of propagule size on airborne microorganism recovery efficiency from different membrane filters to better understand the challenges associated with sampling airborne microorganisms. We obtained 25 airborne fungal isolates from outdoor air and selected five isolates with varying spore sizes. We examined the recovery and viability of the five fungal species using four different filters. We also investigated the impact of air filtration on spore recovery and viability.
2. Materials and Methods
2.1. Bioaerosol Sampling
Airborne fungi were isolated from outdoor air by vacuum-filtering aerosols through a polyethersulfone (PES) membrane filter (pore size: 0.2 μm; diameter: 47 mm) using a bioaerosol sampling device, as previously described [17]. The filtration setup consisted of a filter cup with a screw filter holder designed for 47-mm filters and a 250-mL glass funnel (DWK Life Sciences GmbH, Mainz, Germany). The system was connected to a 400 W rotary vane vacuum pump (W2V10 model, WSA Co., Ltd., Gunpo, Korea) that maintained a fixed flow rate of 100 L/min. A 1-mm nylon mesh was positioned over the filter cup to prevent the collection of macroorganisms. The device was positioned horizontally at a height of approximately 15 m above ground level outside the Biology Building on the main campus of Pusan National University, Busan, Korea (35°23′N 129°08′E). Aerosol sampling was conducted for 180 minutes daily over three consecutive days starting on June 26, 2020.
2.2. Isolation of Airborne Fungi
After aerosol collection, a membrane filter was transferred to a 50-mL conical tube containing 30 mL of sterile 0.9% NaCl solution and agitated at 200 rpm for 30 min. The resulting suspension was serially diluted with sterile 0.9% NaCl solution and plated onto potato dextrose agar (PDA; MB Cell, Kisan Bio Co., Ltd., Seoul, Korea) plates. The plates were incubated at 30°C for 4 d to facilitate fungal growth. Pure fungal colonies were obtained by transferring individual colonies onto fresh PDA plates at least twice. Genomic DNA was extracted from pure colonies using a HiGene Genomic DNA Prep Kit (Biofact Co., Ltd., Daejeon, Korea) according to the manufacturer’s protocol. The internal transcribed spacer (ITS) region was amplified using the primer pair ITS1 (5′-TCCGTAGGTGAACCT GCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′). Partial ITS sequences were analyzed and compared with reference sequences using the Basic Local Alignment Search Tool (http://blast.ncbi.nlm.nih.gov). A total of 25 fungal isolates were successfully identified and deposited in the Korean Collection for Type Cultures (KCTC) (http://kctc.kribb.re.kr).
2.3. Screening Fungal Isolates with Varying Spore Sizes
To obtain fungal spores of varying sizes, we measured spores from the 25 fungal isolates identified in this study. Each fungus was cultured on PDA at 30 °C for up to 1 week. Spores were harvested by scraping mycelia and spores from the PDA plates and transferring them to 500-mL flasks containing 200 mL sterile 0.9% NaCl solution. These suspensions were agitated at 200 rpm for 1 h and subsequently filtered through cheesecloth to remove mycelial debris. A 20 μL aliquot of each spore suspension was placed on a microscope slide and observed at 1000× magnification using a light microscope (Axio Observer 3, Carl Zeiss Co., Ltd., Oberkochen, Germany). Spore length and diameter were measured using ZEN 3.2 (Blue Edition) software (Carl Zeiss). Five fungal isolates with distinct spore sizes were selected: Penicillium sp. AB03, Aspergillus sp. AB25, Fusarium sp. AB21, Aureobasidium sp. AB12, and Fusarium sp. AB20 (Table 1).
Spore volumes were calculated based on measured spore sizes and shapes. For AB03 and AB25, which are globose, volumes were calculated using the formula for a sphere: 4/3•π•r3, where r is the radius. For AB21 and AB20, which are cylindrical, volumes were calculated using the formula for a cylinder: π•r2•h, where r is the radius and h is the height. For AB12, which is ellipsoid, the volume was determined using the formula for an ellipsoid: 4/3•π•a•b2, where a is the semi-major axis and b is the semi-minor axis. The relatively large standard deviations observed for some species (e.g., AB20) likely reflect both natural variation in spore morphology and measurement-related variability inherent to microscopic quantification.
2.4. Preparation of Fungal Spores
The five isolates were cultured on PDA plates at 30°C for 5 d. To promote sporulation, the plates were incubated for an additional week at 30°C under a 12-h light/dark cycle. Fungal spores were harvested by scraping mycelia and spores from the PDA plates and transferring them to 500-mL flasks containing 200 mL of sterile 0.9% NaCl solution. These mixtures were agitated at 200 rpm for 1 h and filtered through cheesecloth to remove mycelial debris, yielding spore suspensions. Spore concentrations were determined using a hemocytometer (Paul Marienfeld GmbH & Co. KG, Lauda-Königshofen, Germany). Each suspension was adjusted to a final concentration of 1 × 108 spores/mL.
2.5. Effects of Fungal Spore Size, Filter Material, and Air Filtration on Spore Recovery
Four hydrophilic membrane filter materials (PES, polyamide (PA), mixed cellulose ester (MCE), and polyvinylidene fluoride (PVDF) (Hyundai Micro Co., Ltd., Seoul, Korea)) were used in this study. All filters had a diameter of 47 mm and a pore size of 0.2 μm. Before use, filters were dried at 105°C for 24 h. Identical aerosol sampling devices were assembled, as described above. A 1 mL aliquot of fungal spore suspension was applied to a filter within the sampling device and briefly vacuumed for a few seconds. Subsequently, a 30-min vacuuming procedure was conducted inside a clean bench maintained at 25°C, ensuring that the fungal spores were embedded within the membrane matrix. No visible clogging or noticeable pressure drop was observed during this procedure across all filter types, confirming stable airflow conditions.
After filtration, the filter was removed from the device and transferred to a 50-mL conical tube containing 30 mL of sterile 0.9% NaCl solution. The tube was agitated at 200 rpm for 30 min to dislodge fungal spores while minimizing physical stress, as one objective was to assess their viability. Spore quantification was performed using two methods: direct microscopic counting using a hemocytometer as previously described, and dilution plating on PDA. For dilution plating, suspensions were serially diluted with sterile 0.9% NaCl solution and spread onto PDA plates. Plates were then incubated at 30°C for 2 d, after which fungal colonies were counted. Colony counts were restricted to plates containing 30–100 colonies to ensure accuracy. There were duplicates for each treatment.
2.6. Data Analysis
Recovery ratios were calculated to evaluate the efficiency of spore release from the filters, represented by the proportion of spores recovered either before or after air filtration relative to the total number of spores initially applied (i.e., inocula). Spore viability was calculated as the ratio of viable spores to total spore count. For example, inoculum viability was determined as the ratio of viable spores (via plate counting) to total spores (via microscopic counting) in the initial spore suspension. Viability ratios were calculated as pre- or post-filtration viability relative to inoculum viability. Finally, detachment efficiency, which reflects the proportion of spores successfully released from the filter after filtration, was determined using the following equation: detachment efficiency = 1 – ((pre-filtration direct count – post-filtration direct count) / pre-filtration direct count).
Statistical analyses were conducted using R version 4.3.0 (R Core Team, 2023). Three-factor analysis of variance (ANOVA) was used to evaluate the effects of spore size, filter material, and air filtration on spore recovery and viability, with all three factors treated as fixed effects. These analyses were conducted using the aov function. Additionally, the relationship between spore size and recovery ratio was assessed by fitting the data to an exponential-rise-to-maximum model (Y = Y0 + a•(1 – bx)).
3. Results
3.1. Fungal Spore Physical Characteristics
Spore dimensions were measured by light microscopy to characterize spore sizes and volumes (Fig. 1). Spores of AB03 and AB25 (globose in shape) had diameters of 3.71 ± 0.54 and 4.07 ± 0.13 μm, respectively. AB21 and AB20 produced cylindrical spores, with diameters of 2.79 ± 0.18 μm and 3.96 ± 0.69 μm, and corresponding lengths of 9.73 ± 2.82 μm and 24.76 ± 4.81 μm, respectively. AB12 spores (ellipsoid) were measured with a minor axis length of 6.08 ± 1.26 μm and a major axis length of 8.84 ± 1.98 μm.
Spore volumes were calculated based on shape and dimensions (n = 5). The smallest spores were from AB03, with a volume of 28.1 ± 12.8 μm3, followed by AB25 at 35.3 ± 3.4 μm3 (Table 1). Spores from AB21 and AB12 had volumes of 60.3 ± 21.9 μm3 and 188.0 ± 95.4 μm3, respectively. The largest spores were from AB20, with a volume of 325.2 ± 140.2 μm3.
3.2. Effects of Spore Size, Air Filtration, and Filter Materials on Spore Recovery
ANOVA indicated that spore size and air filtration affected spore recovery (P < 0.05), whereas filter material did not influence recovery (P > 0.05) (Fig. 2). For the PES filter, the relative ratio was 0.98 ± 0.07 (ranging from 0.89 to 1.06 across the spore size gradient) before filtration and 0.61 ± 0.30 (0.23 – 0.95) after filtration. The PA filter showed recovery ratios of 0.98 ± 0.13 (0.85 – 1.15) before filtration and 0.54 ± 0.36 (0.04 – 0.88) after filtration. The recovery ratio of MCE filter was 0.98 ± 0.08 (0.90 – 1.10) before filtration and 0.57 ± 0.37 (0.11 – 0.89) after filtration. For the PVDF filter, the recovery ratio was 0.91 ± 0.10 (0.77 – 1.05) before filtration and 0.62 ± 0.26 (0.27 – 0.87) after filtration.
The relationship between spore size and recovery ratio was analyzed for each filter type (Fig. 2). Before air filtration, no significant exponential-rise-to-maximum relationship was observed for any of the filters (P > 0.05). In contrast, after filtration, a clear exponential-rise-to-maximum relationship was evident for all filters (P < 0.05) (PES: Y = −1.37 + 2.24•(1 – 0.96x), R2 = 0.78, PA: Y = −7.78 + 8.57•(1 – 0.91x), R2 = 0.91, MCE: Y = −4.21 + 5.05•(1 – 0.94x), R2 = 0.77, and PVDF: Y = −1.11 + 1.90•(1 – 0.95x), R2 = 0.58)
3.3. Effects of Spore Size, Air Filtration, and Filter Materials on Spore Viability
ANOVA revealed that air filtration affected spore viability (P < 0.05), whereas spore size and filter material had no impact (P > 0.05) (Fig. 3). For the PES filter, the viability ratio was 0.98 ± 0.11 (0.85 – 1.14) before filtration and 0.88 ± 0.28 (0.51 – 1.14) after filtration. The PA filter showed viability ratios of 0.90 ± 0.22 (0.65 – 1.25) before filtration and 0.81 ± 0.73 (0.17 – 1.85) after filtration. For the MCE filter, the viability ratio was 0.95 ± 0.15 (0.69 – 1.05) before filtration and 0.48 ± 0.24 (0.25 – 0.88) after filtration. Finally, the viability ratio of PVDF filter was 0.81 ± 0.24 (0.53 – 1.17) before filtration and 0.62 ± 0.13 (0.48 – 0.79) after filtration. The relationship between spore size and viability ratio was analyzed for each filter type (Fig. 3). No significant exponential-rise-to-maximum relationship was observed for any filter before or after filtration (P > 0.05).
3.4. Overall Effects of Air Filtration on Spore Recovery and Viability
Given the strong negative effects of air filtration on both recovery and viability, the data were analyzed collectively, irrespective of filter material or spore size. This comprehensive analysis revealed that air filtration significantly reduced both recovery and viability (t-test, P < 0.05) (Fig. 4). The recovery ratio decreased by 39.6%, from 0.96 ± 0.14 before filtration to 0.58 ± 0.32 after filtration (P < 0.05; Fig. 4a). Similarly, the viability ratio declined by 23.1%, from 0.91 ± 0.25 to 0.70 ± 0.54 (P < 0.05; Fig. 4b).
3.5. Overall Trends of Spore Recovery, Viability, and Detachment with Increasing Spore Size
Owing to the lack of significant impact of filter material, data were pooled to evaluate the overall influence of spore size on recovery and viability before and after air filtration. The relationship between spore size and recovery ratio was analyzed collectively before and after air filtration, irrespective of filter material (Figs. 5a and b). Before air filtration, no significant exponential-rise-to-maximum relationship was observed between spore size and recovery ratio (P > 0.05). However, after filtration, a clear relationship emerged, modeled as Y = −3.01 + 3.83•(1 – 0.94x) (R2 = 0.74, P < 0.05).
Spore viability showed no positive association with spore size before or after air filtration, regardless of filter material (data not shown). Before filtration, viability was unaffected by spore size. However, after filtration, a very slight reduction in viability was observed with increasing spore size (Y = −0.002•x + 0.88, R2 = 0.10, P < 0.05).
ANOVA indicated that spore size significantly influenced detachment efficiency (P < 0.05), whereas filter material exerts no significant effect (Fig. 5c). Detachment efficiency also followed an exponential-rise-to-maximum relationship with spore size, irrespective of filter material (Y = −3.27 + 4.10•(1 – 0.94x), R2 = 0.65, P < 0.05).
4. Discussion
This study was performed to evaluate the recovery of five airborne fungal isolates, with spore sizes ranging from 28.1 to 325.2 μm3, through four different membrane filters in bioaerosol filtration. Our findings revealed an exponential-rise-to-maximum relationship between propagule size and recovery across all filters, emphasizing the challenges in recovering small propagules.
Airborne fungal spores typically ranged from 2 to 30 μm in diameter [2, 6, 8]. Although our study employed only five fungal isolates (3.7–24.8 μm), they were intentionally selected to cover a representative range of spore sizes commonly encountered in outdoor air. By spanning this biologically relevant size spectrum, the study was able to evaluate recovery efficiency across particle sizes most pertinent to airborne fungi. While we acknowledge that a larger set of isolates could provide additional resolution, the consistency of the observed size-dependent recovery trends supports the generalizability of our findings to broader airborne fungal communities. Air filtration (i.e., vacuuming air through a membrane filter) and spore size significantly affected recovery (P < 0.05), whereas the filter material had no significant impact (Fig. 2). The relationship between spore size and recovery was analyzed across four different filters before and after air filtration. Air filtration produced a clearly distinguishable relationship between spore size and recovery. Before air filtration, no exponential-rise-to-maximum relationship between spore size and recovery was observed for any filter (P > 0.05). By contrast, after air filtration, a clear exponential-rise-to-maximum relationship was found for all filters (R2 ≥ 0.58, P < 0.05). Irrespective of spore size and filter material, air filtration reduced the overall recovery of fungal spores by 39.6% with the recovery ratio decreasing from 0.96 to 0.58 (P < 0.05) (Fig. 4a). These results indicate that air filtration caused some microbial propagules to become firmly embedded within the filter matrix during bioaerosol sampling. The pronounced exponential-rise-to-maximum relationship observed after air filtration suggests that smaller spores were less effectively recovered, likely because of their tendency to become deeply embedded and tightly trapped within the filter matrix during filtration. Consistent with our observations, Zierdt [19] noted that smaller particles, owing to their size and greater surface-to-volume ratio, experience stronger electrostatic interactions with filter materials, increasing their adhesion to the matrix. Wakeman [18] reported that the specific resistance of filter cakes increased exponentially as particle size decreases, indicating that smaller spores are more likely to form tightly packed layers within the filter. Such layers exhibit high resistance to detachment, thereby reducing recovery efficiency. In addition, Wakeman [18] found that propagule shape also influenced specific resistance, with spherical particles exhibiting values nearly twice as high as those of cylindrical particles. In our study, the consistently lower recovery efficiency of species with smaller, globose spores (e.g., AB03 and AB25) may partly reflect this effect. Thus, propagule shape, through its influence on volume and surface area, can represent a factor governing both filtration and recovery.
Spore viability was significantly affected by air filtration (P < 0.05) but not by spore size or filter material (Fig. 3). Air filtration reduced the overall spore viability by approximately 23% from 0.91 to 0.70 (Fig. 4b). This reduction in viability is likely attributable to filtration-induced stresses, including dehydration, entanglement within the filter matrix, and clogging, which can compromise cellular integrity [13, 14]. Similarly, Wang, Reponen [13] reported a rapid decline in the viability of Penicillium and Aspergillus spores within the first 30 min of filtration. Stewart, Grinshpun [20] found that the increased impaction stress caused by higher filtration flow rates reduced the recovery rate of bacterial cells and increased their susceptibility to metabolic and structural injuries. These findings suggest that prolonged filtration may negatively impact the viability of microbial propagules. In addition, no exponential-rise-to-maximum trend was observed between spore size and viability for any filter, either before or after air filtration (Fig. 3). Although spore size did not affect viability prior to air filtration, a slight reduction in viability ratio with increasing spore size was observed after filtration (P < 0.05) (data not shown). Larger fungal spores (with greater surface areas) may be more vulnerable to filtration-induced stress than smaller spores. Thus, larger spores were more efficiently recovered in bioaerosol filtration, but did not exhibit greater viability.
As the filter material exerted no significant effect on recovery, the data were pooled to evaluate the overall influence of spore size. Spore recovery exhibited significant differences among spore sizes, both before and after air filtration (Fig. 5). The recovery ratio showed varying ranges, with a narrower difference between the lowest and highest values before air filtration (0.85 ± 0.11 to 1.08 ± 0.09) compared to a broader range after air filtration (0.20 ± 0.20 to 0.83 ± 0.15). With air filtration, an exponential-rise-to-maximum relationship between spore size and recovery became pronounced (R2 = 0.74). Similarly, the detachment ratio followed the same trend with increasing spore size (R2 = 0.65). Both recovery and detachment ratios consistently reached their maxima at a spore volume of 60.3 μm3. These findings suggest that spores larger than 60 μm3 are less likely to become deeply embedded within the membrane matrix and are more readily retained on the filter surface. This size-dependent recovery behavior highlights the need to adapt filtration and recovery methods to propagule size. While larger spores may be recovered effectively with simpler methods, finer propagules, such as those smaller than 60 μm3, may require additional processing to enhance detachment and extraction efficiency. Techniques such as ultrasonic agitation, bead beating, or optimized washing protocols can mitigate the forces that embed smaller spores in the filter matrix, thereby improving the recovery efficiency [17, 21, 22]. Underrepresentation of small-spored fungi can cause surveillance systems to underestimate exposure levels, leading to gaps in assessing risks for asthma, allergies, and other respiratory conditions.
Spore recovery, viability, and detachment did not differ significantly among the filter materials tested in this study. Nevertheless, filter materials can substantially affect the recovery and viability of microbial propagules. For example, polytetrafluorethylene (PTFE) filter exhibited significantly lower spore recovery and viability than the MCE, PES, PA, and PVDF filters (relative recovery: 0.1 vs. 0.6; relative viability: 0.1 vs. 0.7 for PTFE vs. other filters). Consequently, the PTFE filter was excluded from this study as it was considered unsuitable for airborne fungal spore sampling. Previous studies have highlighted the importance of filter material in microbial recovery. Jeong and Kim [17] found that PES filters outperformed MCE, PA, PTFE, and PVDF filters by recovering denser and more diverse bacterial and fungal populations during aerosol sampling. Similarly, Zierdt [19] reported that cellulose ester filters retained more bacteria and blood cells than PA filters. In our study, recovery variability among filter materials was particularly apparent for smaller spores (e.g., AB03 and AB25). Spore size emerged as the dominant factor influencing recovery, potentially masking differences attributable to filter material. Furthermore, the recovery protocol employed here (without chemical or physical lysis) may not have provided sufficient sensitivity to detect statistically significant differences in detachment efficiency among filter types. Overall, these findings suggest that filter characteristics, including membrane structure, porosity, pore size, thickness, surface roughness, and adsorptive capacity, play critical roles in determining the efficiency of microbial recovery from filters. Understanding these characteristics is essential for optimizing filter selection for bioaerosol sampling.
5. Conclusions
This study investigated the recovery and viability of airborne fungal spores of varying sizes in aerosol filtration using four different membrane filters. Air filtration significantly reduced recovery by 39.6% and viability by 23.1%, and spore size emerged as a significant factor affecting recovery efficiency. However, filter material exerted no significant impact on recovery or viability. Recovery showed an exponential-rise-to-maximum relationship with spore size, indicating that larger spores (> 60 μm3) were more effectively recovered, whereas smaller spores were more prone to entrapment within the filter matrix. In contrast, spore viability was not significantly influenced by spore size and only a slight decline was observed after air filtration. These findings underscore the importance of tailoring bioaerosol recovery protocols to propagule size, as unadjusted methods are likely to underestimate exposure to small-spored fungi.
PDF Links
PubReader
Full text via DOI
Download Citation 
Print 
