Isolation of Culturable Yeasts and Molds from Soils to Investigate Fungal Population Structure

Summary

This protocol is an effective, speedy method of culturing yeasts and the mold Aspergillus fumigatus from large sets of soil samples in as little as 7 days. The methods can be easily modified to accommodate a range of incubation media and temperatures as needed for experiments.

Abstract

Soil is host to an incredible amount of microbial life, with each gram containing up to billions of bacterial, archaeal, and fungal cells. Multicellular fungi such as molds and unicellular fungi, broadly defined as yeasts, fulfill essential roles in soil ecosystems as decomposers of organic material and as food sources for other soil dwellers. Fungal species diversity in soil is dependent on a multitude of climatic factors such as rainfall and temperature, as well as soil properties including organic matter, pH, and moisture. Lack of adequate environmental sampling, especially in regions of Asia, Africa, South America, and Central America, hinders the characterization of soil fungal communities and the discovery of novel species.

We characterized soil fungal communities in nine countries across six continents using ~4,000 soil samples and a protocol developed in the laboratory for the isolation of yeasts and molds. This protocol begins with separate selective enrichment for yeasts and the medically relevant mold Aspergillus fumigatus, in liquid media while inhibiting bacterial growth. Resulting colonies are then transferred to solid media and further processed to obtain pure cultures, followed by downstream genetic characterization. Yeast species identity is established via sequencing of their internal transcribed spacer (ITS) region of the nuclear ribosomal RNA gene cluster, while global population structure of A. fumigatus is explored via microsatellite marker analysis.

The protocol was successfully applied to isolate and characterize soil yeast and A. fumigatus populations in Cameroon, Canada, China, Costa Rica, Iceland, Peru, New Zealand, and Saudi Arabia. These findings revealed much-needed insights on global patterns in soil yeast diversity, as well as global population structure and antifungal resistance profiles of A. fumigatus. This paper presents the method of isolating both yeasts and A. fumigatus from international soil samples.

Introduction

Fungi in soil ecosystems play essential roles in organic matter decomposition, nutrient cycling, and soil fertilization¹. Both culture-independent (i.e., high-throughput sequencing) and culture-dependent approaches are widely used in the study of soil fungi^2,3. While the large amount of data generated by high-throughput metabarcode sequencing is useful for elucidating broad-scale patterns in community structure and diversity, the culture-dependent approach can provide highly complementary information on the taxonomic and functional structures of fungal communities, as well as more specific profiles of individual organisms through downstream diversity and functional analyses due to the availability of pure fungal cultures.

Despite rarely exceeding thousands of cells per gram of soil, yeasts, broadly defined as unicellular fungi, are essential decomposers and food sources for other soil dwellers^4,5. In fact, yeasts may be the predominant soil fungi in cold biospheres such as continental Antarctica^6,7. Soil is also a primary reservoir of medically relevant yeasts that cause serious opportunistic infections in humans and other mammals⁸. Despite morphological similarities, yeast species are phylogenetically diverse and occur among filamentous fungi in two major phyla, Ascomycota and Basidiomycota, within the fungal kingdom⁹. Yeasts lack a defining DNA signature at the fungal barcoding gene, the internal transcribed spacer (ITS) region of the nuclear ribosomal RNA gene cluster¹⁰, making them indistinguishable from other fungi in metagenomics investigations and thus necessitating the use of culture-dependent methods to isolate yeast species.

The protocol below was implemented to characterize soil yeast communities of nine countries and identify global trends and patterns in soil yeast diversity^9,11,12. Metagenomics approaches are of limited use when studying targeted groups of organisms such as yeasts^2,3. Due to their phylogenetic diversity, yeasts cannot be distinguished from other fungi based on DNA sequence alone. Thus, studying yeast populations requires the continued use of culture-dependent isolation. However, culturing is often significantly more time-consuming and requires more personnel to perform the experiments. Therefore, the protocol has been optimized and streamlined for faster processing with limited personnel. The main advantage of culturing is that the yeast species identified are living yeasts and not dead ones, and thus are more likely to be true soil dwellers rather than transient cells present in the soils. It has been estimated that approximately 40% of fungal DNA in soil are either contaminants from other environments, extracellular, or come from cells that are no longer intact, causing high-throughput sequencing approaches to overestimate fungal richness by as much as 55%¹³. Culture-dependent isolation can readily confirm yeast species identity with the added benefit of securing pure culture to be used in downstream analyses. Indeed, pure cultures of 44 putative new yeast species were identified using this soil isolation protocol that allowed the use of a range of methods to study their taxonomic and functional properties in detail¹⁴.

The protocol below can also be used to isolate molds present within soil, such as A. fumigatus. Aspergillus fumigatus is a thermophilic and saprophytic mold with a wide, global distribution in soil¹⁵. It has been isolated from numerous clinical and non-clinical environments. Non-clinical sampling commonly includes air, organic debris (compost, saw dust, tulip bulb waste), and soil (agricultural, garden, and natural soils)^16,17,18,19. Aspergillus fumigatus is a human opportunistic pathogen causing a range of infections collectively termed aspergillosis, affecting over 8 million people worldwide^16,20. Approximately 300,000 people around the globe suffer from invasive aspergillosis, which is the most severe form of aspergillosis¹⁶. Depending on factors such as the patient population, site of infection, and efficacy of antifungal therapy, mortality rate can be as high as 90%. Over the past several decades, resistance to antifungal therapies has increased, requiring global surveillance efforts in both clinical and environmental populations to track these resistance genotypes^21,22,23. Given its ability to grow at temperatures upward of 50 °C, this temperature can be exploited to select for A. fumigatus isolates from soil using culture-dependent methods. Aspergillus fumigatus isolates are commonly genotyped at nine highly polymorphic short tandem repeat (STR) loci, shown to have high discriminatory power between strains²⁴. These STR genotypes can be compared to other previously surveyed populations to track the spread of A. fumigatus genotypes, including drug-resistance genes, around the world.

Below we describe a protocol for the speedy isolation of yeasts and A. fumigatus from soil samples in a culture-dependent manner. Depending on the amount of soil obtained per sample, the soil samples can be shared between the two protocols. In comparison to similar methods that isolate yeast and A. fumigatus from soil, this protocol uses 10x less soil per isolate obtained. Studies attempting to isolate A. fumigatus from soil require between 1 and 2 g of soil per isolate, whereas this protocol requires only 0.1-0.2 g of soil^18,19,25. This protocol utilizes smaller plastics and containers that facilitate its high-throughput design. Therefore, a larger number of samples can be processed using less space for equipment such as incubators and roller drums. Soil samples can be fully processed to obtain isolates in as little as 7 days. This protocol has been optimized to allow processing of up to 150-200 samples per day per person.

Protocol

NOTE: Any steps utilizing international soil samples and/or A. fumigatus spores and mycelia require working within a biosafety cabinet for level 2 organisms (BSCII).

1. Isolation of yeast from soil

1. Preparation of antibacterial and antifungal solutions
   1. Suspend chloramphenicol powder in 70% ethanol to prepare a 50 g/L stock solution. Sterilize by syringe filtration and store at 4 °C.
      ​NOTE: This antibiotic will prevent the growth of most bacteria during soil yeast isolation. As chloramphenicol-resistant bacteria may still grow, colony morphology must be carefully taken into consideration when distinguishing yeasts from bacteria. Additional antibiotics may be added to media when working with soil suspected of containing antibiotic-resistant bacterial strains. Bacterial contamination in chloramphenicol-supplemented media was not an issue encountered when isolating yeasts from environmental soils.
   2. Suspend benomyl powder in DMSO to prepare a 5 g/L stock solution. Sterilize by syringe filtration and store at 4 °C.
      NOTE: This selective antifungal drug prevents the growth of most filamentous fungi without affecting yeast growth during soil yeast isolation^26,27.
2. Preparation of culture media and sterile equipment
   1. To prepare YEPD (Yeast Extract-Peptone-Dextrose) broth, add 10 g of Yeast Extract, 20 g of peptone, and 20 g of dextrose to 1 L of double-distilled water. Stir until well-mixed and autoclave for 40 min at 121 °C. Store at room temperature until use.
   2. YEPD solid agar medium
      1. Mix 10 g of yeast extract, 20 g of peptone, 20 g of dextrose, and 20 g of agar in 1 L of water. Stir well to mix and autoclave for 40 min at 121 °C.
      2. Once sufficiently cooled, add 1 mL of chloramphenicol and benomyl from stock solutions to bring the final concentrations of the two antimicrobials to 50 mg/L and 5 mg/L, respectively.
      3. Mix well by stirring and pour into 10 cm diameter Petri dishes. Leave at room temperature overnight to set and store at 4 °C until use.
         NOTE: From 1 L of YEPD, approximately 40 plates can be poured.
   3. Sterilize wooden plain-tipped applicator sticks and reusable cell spreaders by autoclaving and store at room temperature.
3. Incubation of soil in liquid broth
   1. Prepare a set of sterile 13 mL culture tubes by labeling them with soil sample ID.
   2. Using a serological pipette, add 5 mL of YEPD broth supplemented with chloramphenicol and benomyl into each tube.
   3. Working in a BSCII, transfer ~0.1 g of soil into the appropriate culture tube using a sterile, wooden plain-tipped applicator.
      NOTE: Use a fresh applicator for each soil sample and discard immediately upon use to avoid cross-contamination between samples.
   4. Cap the tube securely to the first stop to prevent spillage but still allow air exchange during incubation. Incubate the culture tubes in a roller drum for 24 h at a temperature deemed optimum for maximizing yeast growth.
      NOTE: The incubation temperature should be decided based on the mean annual temperature of the soil samples' country of origin. For example, when isolating soil yeasts from Iceland, the culture tubes were incubated at 14 °C, whereas soils from Saudi Arabia were incubated at 30 °C. Due to slower yeast growth at lower temperatures, incubation time might have to be extended for up to 72 h.
4. Transfer the supernatant to the solid medium.
   1. Using the set of YEPD + chloramphenicol + benomyl agar plates prepared in Step 1.2, label them with the soil sample ID.
   2. Remove the culture tubes prepared in Step 1.3 from the roller drum. Working in a BSCII, briefly vortex the tube to draw soil particles and cells that may have settled at the bottom back into suspension.
   3. Using a micropipette, transfer 100 µL of the supernatant onto a plate. Use a sterile, reusable cell spreader to spread the liquid thoroughly and evenly over the agar surface.
      NOTE: Working in sets of 10 samples can significantly speed up this process. Pipette the supernatant onto 10 plates first and then perform spreading. Use a fresh spreader for each sample to avoid cross-contamination between samples.
   4. Stack the plates in plastic bags, seal, and incubate upside down for 2-3 days at the same temperature used previously for liquid broth incubation until microbial growth is visible.
5. Detection of yeasts and streaking for single colonies
   1. After allowing for sufficient incubation time (typically 2-3 days, but may take longer at lower temperatures), inspect the plates in a BSCII for any yeast growth. Look for creamy, round, matte-like yeasts that are easily distinguished from bacterial and mold colonies.
      NOTE: Some yeasts produce colored pigments and may appear black/brown, yellow, or red, but their overall colony texture and shape would be similar to non-pigmented yeasts.
   2. Select one yeast-like colony from each plate for further processing.
      NOTE: If more than one type of morphology is observed on a single plate, select one representative colony for each morphological type.
   3. Using sterile, wooden plain-tipped applicator sticks, transfer each selected colony onto a fresh YEPD + chloramphenicol + benomyl plate and streak for single colonies. Perform three separate streaks.
      1. Streak back and forth over a third of the plate using an applicator stick. Begin the second and third streak by streaking the applicator across the preceding streak once. Use a new applicator stick for every streak.
         NOTE: Use one plate per isolate. Discard applicators immediately upon use to avoid cross-contamination between samples.
   4. Stack the plates in plastic bags, seal, and incubate upside down for 2-3 days at the same temperature used previously until single colonies become visible.
6. Identification of yeast species via ITS sequencing
   1. Select a well-separated, single colony per soil isolate and subculture onto a fresh YEPD + chloramphenicol + benomyl plate to obtain more cells. Incubate for 2-3 days at the same temperature used previously.
   2. Harvest the freshly grown cells and suspend them in -80 °C freezer tubes containing 1 mL of sterilized 30% glycerol in double-distilled water to create cell suspensions. Maintain these suspensions at -80 °C as stock solutions.
   3. Use fresh cells to perform colony PCR (polymerase chain reaction) with primers ITS1 (5' TCCGTAGGTGAACCTGCGG 3') and ITS4 (5' TCCTCCGCTTATTGATATGC 3') to amplify the fungal barcoding gene, ITS⁹. Use the following thermocycling conditions: an initial denaturation step at 95 °C for 10 min followed by 35 cycles of (i) 95 °C for 30 s, (ii) 55 °C for 30 s, and (iii) 72 °C for 1 min.
      NOTE: Colony PCR has a high success rate and a faster turnaround time than DNA extraction followed by PCR.
   4. If colony PCR repeatedly fails for a strain, extract DNA using the protocol of choice and perform ITS PCR using extracted genomic DNA as template (use same thermocycling conditions as above).
      NOTE: A relatively inexpensive chloroform-based DNA extraction is recommended²⁸.
   5. Perform Sanger sequencing to determine the DNA sequence of the amplified ITS region for each strain.
   6. Compare the obtained ITS sequence of the yeast strains to sequences deposited in public databases such as NCBI GenBank and UNITE to establish species identity.

2. Isolation of Aspergillus fumigatus from soil

1. Prepare 1 mL of sterile Sabouraud dextrose broth (SDB) supplemented with the antibiotic chloramphenicol per soil sample.
   1. Add 10 g of peptone and 20 g of dextrose to 1 L of distilled water. Autoclave at 121 °C for 40 min.
   2. Allow the SDB to cool to ~50 °C, add 1 mL of 50 g/L chloramphenicol to bring the concentration to 50 mg/L.
      NOTE: Chloramphenicol is prepared the same as described above for yeast isolation. Aside from inhibiting bacterial growth, chloramphenicol also prevents the production of gas from bacteria that will cause the tubes to open during the incubation step detailed in Step 2.2.
   3. Aseptically aliquot 1 mL of SDB into a 1.5 mL microcentrifuge tube per soil sample using a mechanical pipette.
2. Add soil to 1.5 mL microcentrifuge tubes.
   1. Lay bench coat or absorbent padding inside a BSCII to aid in the disposal of spilt soil.
   2. Using autoclaved applicator sticks, transfer approximately 0.1 g of soil into a 1.5 mL microcentrifuge tube containing 1 mL of SBD. Close the tube and vortex. Incubate the suspended soil at 50 °C for 3 days.
      NOTE: Shaking during incubation is not required.
3. Mycelial harvest of soil-inoculated broth
   1. Prepare malt extract agar (MEA) plates.
      1. Per liter of MEA: add 20 g of malt extract, 20 g of dextrose, 6 g of peptone, and 15 g of agar to 1 L of distilled water. Autoclave at 121 °C for 40 min.
      2. Allow the MEA to cool to ~50 °C, then add 1 mL of 50 g/L chloramphenicol to bring to a final concentration of 50 mg/L.
   2. Transfer the mycelia from the soil-inoculated broth onto MEA plates.
      1. Identify soil inoculums that have visible mycelial growth at the SDB to air boundary.
      2. Use sterilized wooden plain-tipped applicator sticks to transfer the mycelia to the center of a MEA plate. Incubate the MEA plates at 37 °C for 3 days.
4. Selection of mycelia with A. fumigatus morphological properties.
   1. Identify mold colonies that have characteristic A. fumigatus morphological properties (green suede like growth).
   2. Working in a BSCII, use sterilized wooden plain-tipped applicator sticks or an inoculation loop to harvest conidia/mycelia by scraping the surface once. Transfer the spores/mycelia to the center of a MEA plate by streaking onto the agar for single colonies. Incubate at 37 °C for 2 days.
      NOTE: As multiple A. fumigatus strains and/or other fungi may be present on the plate, it is important to streak for single colonies. The single-colony streaking protocol in Step 1.5.3 in the yeast isolation protocol can be used.
   3. Using a sterile applicator stick or inoculation loop, subculture a single colony generated in Step 2.4.1.2 onto MEA by streaking the colony once. Spread the harvested spores into the center of the plate. Incubate at 37 °C for 2 days.
5. Harvesting of A. fumigatus spores/mycelia for culture storage
   1. Prepare a sterile 30% glycerol solution (for a 100 mL solution, add 30 mL of 100% glycerol mixed with 70 mL of double-distilled water, sterilized at 121 °C for 40 min).
   2. Working in a BSCII, use a p1000 pipette to aspirate 1 mL of the 30% glycerol solution. Dispense the 1 mL of glycerol solution onto the A. fumigatus colony to harvest spores/mycelia.
      1. Due to the hydrophobicity of A. fumigatus spores/mycelia, use the pipette tip to scratch a densely sporulated region of the plate.
         NOTE: When glycerol is dispensed in the scratch, the glycerol would adhere to the scratch area rather than rolling over the agar.
      2. Slowly dispense the glycerol onto the scratch area to dislodge the spores and suspend them in the glycerol solution.
      3. Once fully dispensed, lightly tip the plate and aspirate the glycerol spore/mycelia suspension.
         NOTE: Approximately 750 to 800 µL will be aspirated.
      4. Transfer the aspirate to a sterile freezer tube and store at -80 °C. If required, create working stock by repeating steps 2.5.2.1 to 2.5.2.4.
6. Phenotypic identification of A. fumigatus strains
   1. Using the spore stock created from step 2.5.2, create a 100x dilution in water.
      1. Aspirate 10 µL of the mycelial and spore stocks and dispense in 990 µL of water. Vortex the suspension.
   2. Dispense 10 µL of the diluted spore suspension onto a standard microscope slide.
   3. OPTIONAL: Stain the mycelial and spore suspension with methylene blue.
      1. To stain with methylene blue, fix the conidia and conidiophores to the slide by passing the slide over a Bunsen burner until dry.
      2. Apply methylene blue for 1-2 min and wash off with water.
      3. Dry the slide with blotting paper.
   4. Using a compound microscope at 400x magnification, view the suspension and locate conidiophores. Compare the observed conidiophore morphology with A. fumigatus conidiophore morphology.
7. Molecular identification of A. fumigatus strains
   1. Extract DNA from each isolate following common fungal DNA extraction protocols.
   2. Using primers specific to the Aspergillus β-tubulin genes (β-tub1 and β-tub4), run PCR and obtain the sequence for the amplified products, following protocols described by Alcazar-Fuoli et al.¹⁷.
      1. Compare the obtained sequences to sequences deposited in public databases such as NCBI GenBank using BLAST.
      2. Confirm the strain sequences are a top match to A. fumigatus sequences in the database.
   3. Alternative to step 2.7.2, run a multiplex PCR reaction targeting the A. fumigatus mating types MAT1-1 and MAT1-2²⁹.
      1. Use the following three primer sequences in the multiplex PCR reaction: AFM1: 5'-CCTTGACGCGATGGGGTGG-3'; AFM2: 5′-CGCTCCTCATCAGAACAACTCG-3′; AFM3: 5′-CGGAAATCTGATGTCGCCACG-3′.
      2. Use the following thermocycler parameters: 5 min at 95 °C, 35 cycles of 30 s at 95 °C, 30 s at 60 °C, and 1 min at 72 °C before a final 5 min at 72 °C.
      3. Run gel electrophoresis to identify the products; look for 834 bp MAT-1 or 438 bp MAT1-2. Use A. fumigatus strains that have confirmed mating type amplification as positive and negative controls.
8. Microsatellite genotyping of A. fumigatus strains through fragment analysis
   NOTE: Although the steps listed below broadly cover genotyping A. fumigatus at nine microsatellite (STR) loci, only a few important considerations have been highlighted. For details on A. fumigatus STR genotyping, refer to De Valk et al.^24,30.
   1. Prepare three PCR multiplex master mixes using the STRAf primers previously described by De Valk et al.²⁴.
      1. Fluorescently label forward primers to determine fragment size through capillary electrophoresis. Ensure that the concentration of the forward primers is half (0.5 μM) that of the reverse primers (1 μM) within the master mix.
         NOTE: For the best results, use fluorescent labels that have absorbance wavelengths that match those of the chosen dye standard used during capillary electrophoresis
      2. Use hot start polymerase for best results.
      3. Use a DNA concentration of 0.1 ng per reaction.
   2. Run the multiplex PCR for each strain using the following PCR program: 95 °C for 10 min, 40 cycles of 95 °C for 30 s, 60 °C for 30 s, and 72 °C for 60 s, followed by 72 °C for 10 min and a hold at 4 °C.
   3. Check for amplified products through gel electrophoresis.
   4. Dilute the products to the desired level (typically ~50x) as recommended by fragment analyses specialists and run capillary electrophoresis. Perform three runs for each strain, with each run covering three multiplexed reactions with three different fluorescent probes.
   5. To determine the correct fragment size for each of the nine STR loci, use software capable of fragment analysis.
      1. Retrieve the raw data obtained from capillary electrophoresis. Score the fragment sizes based on the largest peak using the fragment analysis software (e.g., Osiris).
      2. Convert the fragment sizes to repeat numbers for each of the nine loci. Use the fragment sizes of the repeat numbers of the reference strain Af293 as previously described by De Valk et al.²⁴.
         NOTE: Slight variations in fragment sizes may occur between different capillary electrophoresis platforms. Thus, it is important to include a common reference strain (and an internal ladder) with known fragment size for each of the nine loci for genotyping strains.

Representative Results

Yeast isolation from soil
The above yeast isolation protocol was implemented to culture yeasts from soil samples originating from 53 locations in nine countries^9,12. In total, 1,473 yeast strains were isolated from 3,826 soil samples. Given the different climatic conditions of the nine originating countries, the best incubation temperature for each country was determined based on its mean annual temperature (Table 1). Given the slower yeast growth at 14 °C, soil samples from Iceland were incubated on a roller drum for an additional 48 h (96-120 h total). Following 2-3 days of incubation on solid medium, microbial growth was visible on plates for all samples (Figure 1A). A randomly selected colony for each yeast morphology present on a plate was streaked for single colonies on fresh plates (Figure 1B).

The rate of successful yeast isolation differed between countries (Table 1). For example, 97 yeast strains were cultured from 562 Saudi Arabian soil samples (17.3%), whereas 261 yeasts were cultured from 300 Canadian soil samples (87%). A rarefaction analysis should be performed to determine if sufficient soil sampling has been conducted to obtain accurate estimates of the true yeast diversity in sampled locations. An example rarefaction analysis for each country was preformed where the Shannon diversity index was used as a measure of soil yeast diversity (Figure 2). The resulting rarefaction curves approached the saturation asymptote indicating that additional sampling was not likely to have yielded more yeast diversity.

Sequencing of the ITS region revealed that the 1,473 yeast isolates can be categorized into 134 distinct species. This included 90 known species and 44 potentially novel species. We applied a mixed-effects model to identify predictors of global culturable soil yeast diversity, as quantified by the Shannon diversity index³¹. In this model, mean annual precipitation, mean annual temperature, elevation, and distance to equator of the sampling locations were fitted as fixed effects, while sampling country was set as a random effect⁹. This model identified mean annual precipitation to be significantly correlated with the Shannon diversity index (p = 0.012), while no significant correlations were found between other predictor variables and the Shannon diversity index⁹.

To compare this culture-based protocol to culture-independent methods, we compared these findings to a previous study by Tedersoo and colleagues that investigated global diversity of soil fungi using metagenomics². Tedersoo and colleagues performed high-throughput sequencing of the ITS region on DNA directly extracted from soil samples of 39 countries, four of which, namely Cameroon, Canada, China, and New Zealand, were also sampled in this study. We performed BLAST searches to identify ITS sequences present in both datasets³², and found 26% of the ITS sequences had a significant match (>98.41% nucleotide identity) in the metagenomics dataset. However, <3% matched a fungal sequence from the same country, while the remaining 23% matched a fungal sequence found in a different country⁹. We were more successful than the metagenomics study in annotating the ITS sequences with yeast species identity. Tedersoo and colleagues reported a total of 50,589 fungal operational taxonomic units (OTUs), with the number of yeast OTUs not known. Of these fungal OTUs, 33% were merely annotated as environmental_sequence (724, 1.4%), uncultured_soil_fungus (2,405, 4.8%), uncultured_ectomycorrhizal_fungus (1,407, 2.8%), or uncultured_fungus (11,898, 23.5%).

Aspergillus fumigatus isolation from soil
After 3 days of soil incubation at 50 °C in 1 mL of SDB in a microcentrifuge tube, mycelia may be found growing within the SDB, on the SDB to air boundary, and/or on the inside walls. Aspergillus fumigatus mycelia are typically found growing on the SDB to air boundary but can also be harvested from just below the SDB surface. After this step, the mycelia are transferred to MEA and grown for 3 days at 37 °C. Mycelial growth on plates may only form the green suede morphology typical of A. fumigatus (Figure 3A). More commonly, other thermotolerant molds may be present within the same soil and grow with A. fumigatus on the same plate or by themselves (Figure3B-D). Aspergillus fumigatus isolation rates may vary between geographic locations, similar to what has been found for soil yeasts. For example, within Vancouver, Canada, 251 isolates were obtained through this method from 540 soil samples (46.5%) (unpublished results). By contrast, within Cameroon, 51 A. fumigatus isolates were harvested from 495 soil samples (10.3%)³³.

A. fumigatus conidiophore structure can be viewed and identified using light microscopy. Conidiophores have a ball on stick morphology (Figure 4). Additionally, A. fumigatus conidiophores are uniseriate, where the phialides, attached to the conidia chains, are attached directly to the spherical vesicle. In other Aspergillus species, conidiophores are biseriate, where the phialides are connected to metulae attached to the vesicle.

Several software programs are available to perform fragment analysis of the raw data from capillary electrophoresis, converting the capillary electrophoresis spectrum to fragment sizes. Figure 5 is a chromatogram generated by the program Osiris. The three channels visualizing the fragment lengths of A. fumigatus dinucleotide (2A, 2B, and 2C), trinucleotide (3A, 3B, and 3C), and tetranucleotide (4A, 4B, and 4C) STR loci are shown. In addition, PCR artifacts that occur if the sample DNA concentration during PCR is too high are also shown. The highest peaks of each color represent the fragment sizes of the three STR loci in this plot.

The genetic variation present between A. fumigatus STR genotypes can be visualized using the R package poppr. The R script bruvo.msn creates a pairwise Bruvo's genetic distances matrix between each pair of genotypes. This matrix is then used to generate a minimum spanning network (MSN). A high-quality MSN can then be generated using the R script plot_poppr_msn or imsn (Figure 6). A discriminatory analysis of principal components (DAPC) is another method to visualize the genetic relationships among strains (Figure 7). The R script dapc in the R package adegenet is used to perform DAPC and can be used with known or unknown group priors. With unknown group priors, it uses K-means clustering to identify the likely number of groups of individuals.

Figure 1: Yeast isolation from soil samples. (A) Microbial growth is visible on solid agar following 2-3 days of incubation. Yeast colonies can be seen interspersed among other fungal/bacterial colonies. One representative yeast colony for each morphological type on a plate is selected to streak for single colonies. (B) Three streaks are performed to obtain single colonies of yeast isolates. One plate was used to obtain each soil isolate. Please click here to view a larger version of this figure.

Figure 2: Rarefaction analysis of soil sampling. In each of the nine sampled countries, soil yeast diversity, as quantified by the Shannon diversity index, approaches saturation as the number of soil sample increases. This figure was adapted from ⁹. Please click here to view a larger version of this figure.

Figure 3: Aspergillus fumigatus morphology on Sabouraud dextroseagar. (A) Green suede-like A. fumigatus growth morphology. (B-D) A. fumigatus growth with other thermophilic molds present within the soil sample. A. fumigatus conidiation is visibly reduced in B and D. Please click here to view a larger version of this figure.

Figure 4: Photo of an Aspergillus fumigatus conidiophore under a light microscope. Scale bar = 10 µm. Please click here to view a larger version of this figure.

Figure 5: Osiris output of nine microsatellite loci from an Aspergillus fumigatus strain. The outputs correspond to the (A) Dinucleotide (2A, 2B, and 2C), (B) trinucleotide (3A, 3B, and 3C), and (C) tetranucleotide (4A, 4B, and 4C) STR loci. Forward primer 5` fluorescent labels are: A = 6-FAM; B = HEX; C = ATTO550. Multiplex reaction products were diluted 30x prior to capillary electrophoresis. The LIZ600 dye standard was used during capillary electrophoresis. Raw data was analyzed using the peak analysis software Osiris identifying potential peaks between 60 and 400 bp. Several PCR artifacts are off-scale peaks that cause fluorescence bleeding, stutter peaks, and N-1 peaks (C). Abbreviations: 6-FAM = 6-carboxyfluorescein; HEX = hexachlorofluorescein; STR = short tandem repeats; RFU = relative fluorescence units; BPS = base pairs. Please click here to view a larger version of this figure.

Figure 6: Minimum-spanning network showing the genetic relationship between MLGs of A. fumigatus from Iceland and Northwest Territories in Canada. Each node represents one MLG, where node size corresponds to the number of strains for each MLG. Nodes that are more genetically similar have darker and thicker edges, whereas nodes genetically distant have lighter and thinner edges. This figure was adapted from ³⁴. Abbreviations: ISL = Iceland; NWT = Northwest Territories in Canada; MLG = multilocus genotype. Please click here to view a larger version of this figure.

Figure 7: Genetic clustering using Discriminant Analysis of Principal Components (DAPC) of Iceland, NWT, Eurasian, North American, and Oceanian A. fumigatus populations. Isolates were genotyped at nine microsatellite loci and clone-corrected, totaling 1,703 unique multilocus genotypes. Genotypes were colored according to geographic origins. This figure was adapted from ³⁴. Abbreviations: DAPC = discriminant analysis of principal components; NWT = Northwest Territories; ISL = Iceland; CMR = Cameroon; CAN = Hamilton, Ontario, Canada; BEL = Belgium; FRA = France; DEU = Germany; IND = India; NLD = Netherlands; NOR = Norway; NZL = New Zealand; ESP = Spain; CHE = Switzerland; USA = United States. Please click here to view a larger version of this figure.

Country	Incubation temperature (°C)	Soil samples	Yeast isolates	Known species/ Novel species
Cameroon	30	493	110	10/9
Canada	23	300	261	34/12
China	23	340	230	23/5
Costa Rica	30	388	95	20/2
France	23	327	175	12/2
Iceland	14	316	211	11/0
New Zealand	23	610	155	14/4
Peru	23	490	139	30/9
Saudi Arabia	30	562	97	8/1
Total		3826	1473	90/44

Table 1: Soil yeast isolation from nine countries in six continents. Incubation temperature for soil samples from each country was determined based on its mean annual temperature. Results presented here are adapted from ^{9, 12}.

Discussion

The protocol developed for isolating yeasts and A. fumigatus from soil is a fast and efficient method for high-throughput soil processing and fungal isolation. The protocol only requires a small amount of soil (0.1-0.2 g) per sample, allowing for more sites to be sampled with similar effort. The quick turnaround time ensures that results can be obtained within a short timeframe and allows time for troubleshooting and repeating experiments if necessary. This protocol can be easily replicated across many laboratory settings using standard microbiology and cell culturing equipment. However, when isolating yeast or molds from international soil, extra precautionary measures are required. All non-disposable plastic equipment, such as plastic trays and cell spreaders, should be sterilized by submersion in 10% bleach, followed by autoclaving. The used bench coat should be sealed in a plastic bag and sterilized in an autoclave prior to discarding.

To note, the number and identity of yeasts isolated using the protocol above can be limited by the incubating conditions and media used. For example, 24 h incubation in the roller drum can favor the isolation of fast-growing aerobic yeasts over slow-growing species. In addition, the use of nutrient-rich YEPD medium for yeast isolation may have excluded yeast species with different nutrient requirements and preferences. Furthermore, most locations sampled here experience seasonal variations in temperature and other climatic conditions throughout the year. If time and resources permit, the same soil samples can be processed under a range of different temperatures to allow for isolation of a broader range of yeasts that inhabit these soils.

While culture-independent methods are crucial for elucidating large-scale patterns in environmental fungal diversity, they fail to be sufficiently informative for targeted groups of fungi such as yeasts. The use of culture-dependent isolation allowed for a comprehensive investigation of global soil yeast diversity and its predictors⁹. While the metagenomics study conducted by Tedersoo and colleagues identified global predictors and patterns in soil fungal diversity, the extent to which these findings apply specifically to yeast diversity could not be elucidated². Tedersoo and colleagues identified mean annual precipitation as a major climatic driver of soil fungal diversity worldwide². This study found a similar correlation between mean annual precipitation and culturable yeast diversity in global soils, highlighting that culture-dependent methods can complement and expand on findings of high-throughput sequencing approaches⁹.

The addition of chloramphenicol is an important step in the preparation of both solid and liquid media to prevent fungal growth interference by bacteria. The addition of chloramphenicol to liquid media is crucial as the lack of antibiotics will allow the bacteria present in the soil to ferment. This will lead to the production of gases that will forcibly open microcentrifuge tubes or 13 mL culture tubes used during soil incubation in the protocol. If bacterial contamination in chloramphenicol containing agar/broth is an issue, chloramphenicol may be substituted or supplemented with stronger antibiotics. When isolating A. fumigatus from soil, SD broth may be substituted with a Tween 20 solution (0.2 M NaCl with 1% Tween 20) to help suspend the hydrophobic A. fumigatus conidia^35,36. Following suspension, 100 µL can be plated onto SD agar and incubated at 50 °C.

Aspergillus fumigatus grows quickly and sporulates abundantly when incubated alone on both SD agar and MEA media between 22 °C and 50 °C. However, depending on what other thermotolerant species are present in the soil sample, A. fumigatus growth and sporulation may be hindered (Figure 3A,D). Under such a situation, one to several subculturing steps may be required to obtain a pure colony for subsequent harvest and characterization.

The fragment analysis of A. fumigatus strains in Figure 5 was conducted using the program Osiris. Several PCR artifacts have been highlighted in Figure 5C and are discussed in detail by De Valk et al.³⁰. B-1 stutter peaks that have shorter repeat values are caused by strand slippage during the synthesis of the antisense strand by the DNA polymerase. N-1 peaks occur if the DNA concentration is too high during PCR. The recommended DNA concentration is 0.1 ng as stated in the protocol above. Another artifact is dye bleeding that can be remediated using fluorescent labels with non-overlapping absorbance wavelengths. For example, the fluorescent labels 6-FAM, HEX, and ATTO550, as well as the LIZ600 dye standard, generate distinct fragment peaks (Figure 5). Lastly, to prevent off-scale peaks, dilute each PCR reaction (in this case, it was ~50x dilution) prior to capillary electrophoresis. To further reduce fluorescence of the unused forward primers in the reaction mixture, halve the forward primer concentrations relative to the reverse primers when creating the master mix.

The high throughput and conservative nature of this protocol allows for a relatively quick and easy acquisition of many yeasts and molds from soils. However, there are two main limitations present with the sampling methodology. First, the morphological characteristics of yeast colonies grown from soil were used to select for unique species. These were then subcultured and used for species identification via ITS sequencing. This was done to maximize the representation of yeast species present. However, yeast species that shared similar or identical morphologies to the selected colonies may fail to be subcultured. Second, during A. fumigatus isolation, as only one individual colony is selected per soil sample, the presence of multiple individuals within the soil sample will be missed. This may lead to an underrepresentation of the true genetic diversity present within the sample population, as unique genotypes present within the same soil sample will not be collected. To mitigate this issue, during the streak for single colony step following the first culturing on solid media, several single colonies can be collected to obtain additional genotypes. The small volume of soil used per sample helps minimize this sampling limitation when compared to methods that used larger quantities of soil per sample.

The use of this high-throughput and labor-efficient protocol for the isolation of yeast and mold will increase the number of individuals within soil population while using less effort per sample. The increase in statistical power will provide a better picture of culturable yeast communities within the soil and help characterize soil populations of A. fumigatus.

Materials

Name	Company	Catalog Number	Comments
1.5 mL microcentrifuge tube	Sarstedt Inc	72.690.001
Benomyl powder	Toronto Research Chemicals	B161380
Chloramphenicol powder	Sigma-Aldrich	SKU: C0378-5G
Dextrose	Sigma-Aldrich	SKU: D9434-500G
Fragment Analysis Software	NCBI's Osiris		https://www.ncbi.nlm.nih.gov/osiris/
ITS sequence database	NCBI GenBank		https://www.ncbi.nlm.nih.gov/genbank/
ITS sequence database	UNITE		https://unite.ut.ee/
Peptone	Sigma-Aldrich	SKU: P5905-500G
Reusable cell spreaders	Fisher Scientific	08-100-12
Sterile 10 cm diameter Petri dishes	Sarstedt Inc	83.3902
Sterile 13 mL culture tubes	Sarstedt Inc	62.515.006
Wooden plain-tipped applicator sticks	Fisher Scientific	23-400-112
Yeast extract	Sigma-Aldrich	SKU: Y1625-250G