Search tips
Search criteria 


Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. 2017 July 1; 83(13): e00600-17.
Published online 2017 June 16. Prepublished online 2017 April 21. doi:  10.1128/AEM.00600-17
PMCID: PMC5478987

Mushroom Emergence Detected by Combining Spore Trapping with Molecular Techniques

Emma R. Master, Editor
Emma R. Master, University of Toronto;


Obtaining reliable and representative mushroom production data requires time-consuming sampling schemes. In this paper, we assessed a simple methodology to detect mushroom emergence by trapping the fungal spores of the fruiting body community in plots where mushroom production was determined weekly. We compared the performance of filter paper traps with that of funnel traps and combined these spore trapping methods with species-specific quantitative real-time PCR and Illumina MiSeq to determine the spore abundance. Significantly more MiSeq proportional reads were generated for both ectomycorrhizal and saprotrophic fungal species using filter traps than were obtained using funnel traps. The spores of 37 fungal species that produced fruiting bodies in the study plots were identified. Spore community composition changed considerably over time due to the emergence of ephemeral fruiting bodies and rapid spore deposition (lasting from 1 to 2 weeks), which occurred in the absence of rainfall events. For many species, the emergence of epigeous fruiting bodies was followed by a peak in the relative abundance of their airborne spores. There were significant positive relationships between fruiting body yields and spore abundance in time for five of seven fungal species. There was no relationship between fruiting body yields and their spore abundance at plot level, indicating that some of the spores captured in each plot were arriving from the surrounding areas. Differences in fungal detection capacity by spore trapping may indicate different dispersal ability between fungal species. Further research can help to identify the spore rain patterns for most common fungal species.

IMPORTANCE Mushroom monitoring represents a serious challenge in economic and logistical terms because sampling approaches demand extensive field work at both the spatial and temporal scales. In addition, the identification of fungal taxa depends on the expertise of experienced fungal taxonomists. Similarly, the study of fungal dispersal has been constrained by technological limitations, especially because the morphological identification of spores is a challenging and time-consuming task. Here, we demonstrate that spores from ectomycorrhizal and saprotrophic fungal species can be identified using simple spore traps together with either MiSeq fungus-specific amplicon sequencing or species-specific quantitative real-time PCR. In addition, the proposed methodology can be used to characterize the airborne fungal community and to detect mushroom emergence in forest ecosystems.

KEYWORDS: propagules, ectomycorrhizal, saprotroph, DNA barcoding, molecular identification, dispersion, fungi, Lactarius


Wild edible fungi are highly important nonwood forest products and are increasingly in demand at food markets worldwide (1). Up to 268 fungal taxa have been authorized to be commercialized in Europe (2), of which the most important marketed mushrooms are Boletus edulis, Cantharellus cibarius, Lactarius deliciosus, Morchella esculenta, and Agaricus campestris (2, 3). Spain, the Netherlands, France, and Poland are the largest mushroom producers in Europe (2). Despite the increasing interest and importance of mushrooms as nonwood forest products, mushroom production has only recently been included as a target in forest management and planning alongside the traditional goal of wood production (4). The inclusion of mushroom production into forest management planning requires the ability to evaluate potential mushroom yields in quantifiable terms. However, the variability in mushroom biomass across years, the ephemeral cycle of several fruiting bodies that are only observable for a few days, and the time needed to sample representative mushroom populations (5) represent serious challenges for monitoring purposes (6). In this context, new techniques that do not require the collection of mushrooms can improve the estimation of mushroom yields. Except for a few cases (7), no correlations have been found between the belowground mycelial biomass of specific species and their mushroom production (8,10), even though positive correlations have been found between root mycorrhizal colonization and soil mycelia (11). In this regard, new monitoring approaches using fungal spores, such as spore trapping, may represent a new way of addressing this challenge. Spore detection could be used to evaluate the fruiting bodies present without the need for regular or repeated samplings over time.

Traditionally, spore trapping has been used to detect and monitor airborne spores, particularly those of fungal pathogens (12). However, recently, spore trap methods have been combined with molecular techniques such as quantitative real-time PCR (qPCR) (13, 14). Spore traps need to be simple, robust devices for effective use in forest stands; they also need to be easy to handle and have low maintenance costs (14, 15). Active traps (e.g., volumetric or cyclonic traps) trap fungal spores more effectively than passive traps (12). However, they have some disadvantages when used under forest conditions; for example, they require electric power. In contrast, passive traps, such as filter or funnel traps, can be placed anywhere (13). Previous studies have shown that passive traps can capture ectomycorrhizal fungal propagules. These passive traps consisted of simple funnels attached to jars (16, 17). However, even simpler spore traps have been used to capture pathogenic fungi, such as filter traps (13). Although these traps collect spores passively, funnel-based traps collect spores in a jar, whereas filter traps retain spores within the filter, which may mean that the spore collection performance of these two traps is different. Despite the potential of these traps to capture fungal spores, to date, no quantitative relationship between fruiting bodies and airborne spores has been established.

Besides the type of trap, the dispersal capacity of the spores also affects spore detection. Unlike highly sporulating fungi, such as puffballs or several mold or yeast species that dominate the airborne community (18), ectomycorrhizal basidiomycetes seem to disperse spores less abundantly (19,21). Problems detecting rare species can be circumvented by using molecular techniques such as qPCR (22), which can detect even small amounts of fungal DNA, although specific primers need to be designed for the target species (13). High-throughput DNA sequencing technologies enable the relative proportions of each of the identified operational taxonomic units (OTUs) to be determined in a given sample (23). Besides monitoring mushroom emergence, this approach can also be used to understand fungal dispersal. To sum up, spore traps combined with qPCR can be used to detect airborne fungal spores of specific species, while spore traps in combination with high-throughput sequencing approaches can be used as a generic tool to detect and identify airborne spores of different taxa.

We hypothesized that (i) funnel and filter traps in combination with qPCR or MiSeq would differ in their capacity to detect spores, and (ii) there is a relationship between fruiting body yields and spore abundance over time (temporal relationship) and over plots (spatial relationship). To test these hypotheses, we identified the relative abundance of the spores by trapping spores during an 8-week period in the fall of 2014 in a Pinus pinaster forest in northeastern Spain. In parallel, we also determined the taxonomic identity and the yields of the fruiting body community.

First, we studied Lactarius vinosus as a model species to compare the two molecular techniques (qPCR versus MiSeq). We chose this species because of its commercial interest. It belongs to the Lactarius group deliciosus, which has well-established markets across Europe, Asia, and North Africa (1). Furthermore, this species is highly appreciated both by mushroom pickers and by consumers in Spain (24, 25). The commercialization of Lactarius group deliciosus has led to the sale of almost 500,000 kg of L. vinosus per year at the three most important Spanish markets; the estimated market value is €5.3 million year−1 (25).

Second, we studied the relationship between the spore abundance of a species and the yield of fruiting bodies from temporal (week) and spatial (plot) perspectives. For this, we compared the abundances of the fruiting bodies of the 12 most abundant fungal species (6 ectomycorrhizal and 6 saprotrophic fungal species) found in the fruiting body community.


Fungal species producing fruiting bodies in the studied plots.

Among the sampled fruiting bodies, we found 71 fungal species belonging to the ectomycorrhizal or saprotrophic fungal guilds. Among these ectomycorrhizal and saprotrophic species, we identified 37 species that were also present in the spore community. Of those species, 18 species produced fruiting bodies in at least two plots for a 2-week period, and the 12 most abundant species (6 ectomycorrhizal and 6 saprotrophic fungal species) were selected for analysis (Table 1). Thirteen species producing fruiting bodies did not have an internal transcribed spacer (ITS) reference in the UNITE (unified system for the DNA-based fungal species linked to classification) database or in the International Nucleotide Sequence Database (INSD), and, therefore, these species were not searched for within the spore community.

Selected most abundant speciesa

Spore abundance estimated by trapping techniques and molecular methods.

MiSeq data revealed that the relative abundance of L. vinosus was significantly higher in filter traps than in funnel traps (F[1,15] = 4.98; P < 0.001) (Table 2), representing on average 5.55 and 0.48 per thousand of the total number of reads, respectively (i.e., nearly 10 times more abundant in filter traps than in funnel traps) (Table 2). However, when using qPCR, the numbers of L. vinosus spores trapped by the filter and funnel traps did not seem to be significantly different (F[1,15] = 0.06; P = 0.951) (Table 2). When using funnel trap data, the spore abundance of L. vinosus estimated by qPCR correlated with that estimated by MiSeq (F[1,59] = 27.53; P < 0.001; R2 = 0.3) (Fig. 1a). However, there was no relationship between the spore abundance of L. vinosus detected in the filter traps and the abundance detected in the funnel traps (F[1,14] = 0.92; P = 0.352) (Fig. 1b).

Differences in the relative proportions of ectomycorrhizal and saprotrophic species detected by filter traps versus funnel trapsa
Relationship between the L. vinosus abundance measurements determined by qPCR (y axis) and by MiSeq data (x axis) (a) and relationship between the L. vinosus abundance measurements obtained at the funnel traps (y axis) and obtained at the filter traps ...

The higher relative abundance of L. vinosus spores detected by MiSeq when using filter traps rather than funnel traps was also observed for most of the saprotrophic and ectomycorrhizal fungal species studied (F[1,15] = 5.71; P < 0.001) (Table 2). When comparing guilds, we found that selected yeast species were better represented in funnel traps than in filter traps (F[1,15] = 4.70; P < 0.001) (Table 2), but no differences between spore trap types were observed for mold species.

Weekly relationship between fungal spores and the production of fruiting bodies.

The spore community composition varied significantly across weeks (F[7,59] = 3.72; R2 = 0.27; P < 0.001) and plots (F[7,59] = 3.45; R2 = 0.25; P < 0.001). The temporal effect (across weeks) was as high as the spatial effect (across plots) and together accounted for more than 50% of the total variation. The abundance of spores of different species found in the traps fluctuated over the sampling period, especially those of ectomycorrhizal fungal species (Fig. 2a and andbb).

Temporal fluctuation of mushroom yields (red line) produced by ectomycorrhizal (EcM) species (a) and saprotrophic species (b), the relative abundance of their spores (black line), and the total precipitation recorded each week (blue line). The sampling ...

Three saprotrophic (Marasmius androsaceus, Clitocybe phaeophthalma, and Hypholoma fasciculare) and two ectomycorrhizal fungal species (Russula torulosa and Russula chloroides) showed signs of having highly ephemeral fruiting bodies. Spore production by most of these species peaked 1 or 2 weeks after fruiting body production but then ended abruptly (Fig. 3a and andb).b). Species with at least three spore production peaks during the study period showed a significant relationship between fruiting body yields and their spore abundance, for example, Mycena pura (F[1,51] = 36.97; P < 0.001; R2 = 0.96), Lactarius deliciosus (F[1,51] = 11.56; P = 0.001; R2 = 0.68), Lactarius vinosus (F[1,51] = 57.39; P < 0.001; R2 = 0.77), Leucopaxillus gentianeus (F[1,51] = 12.58; P < 0.001; R2 = 0.59), and Tricholoma terreum (F[1,51] = 20.89; P < 0.001; R2 = 0.58). This relationship was also confirmed for L. vinosus using qPCR data (F[1,48] = 29.24; P < 0.001; R2 = 0.79). L. vinosus spore abundance patterns detected using either Illumina MiSeq or qPCR data were very similar, and both techniques showed that the production of L. vinosus spores peaked between 20 October and 29 October. This peak was estimated to represent an average of 1.5 × 106 L. vinosus spores · trap–1 (Fig. 4).

(a) Temporal changes in the number of fruiting bodies produced by saprotrophic species and their spores across the sampled weeks in all of the plots considered in this study. The solid black line represents mushroom production (number of fruiting bodies ...
Relationship between mushroom biomass (red line) and the relative abundance of L. vinosus spores detected by the spore traps, quantified by qPCR (black line) and by determining the relative proportions of L. vinosus using MiSeq (blue line). Average values ...

In contrast to the species already mentioned, Lycoperdon perlatum (F[1,51] = 0.01; P = 0.917) and Inocybe glabripes (F[1,51] = 0.89; P = 0.349) did not display a significant relationship between fruiting body yield and spore abundance. For example, the spore abundance of L. perlatum increased with time and did not correspond to variable fruiting body production (Fig. 3a), and the spore abundance of I. glabripes peaked at the start of the season, which was not reflected at the fruiting body level (Fig. 3b).

Relationship between spores and fruiting body production at plot level.

At plot level, only spores of saprotrophic fungal species correlated with fruiting body production (F[1,48] = 4.11; P = 0.048) (Fig. 5a). These significant differences were mainly attributed to the effect of one plot, in which large quantities of both spores and fruiting bodies were recorded. Nevertheless, no relationship was observed between spores of ectomycorrhizal fungi and their fruiting body yields (F[1,48] = 0.41; P = 0.527) (Fig. 5b). Only five species were found in more than two plots: Inocybe glabripes, Lactarius deliciosus, Lactarius vinosus, Lycoperdon perlatum, and Mycena pura. None of these species showed a significant relationship at plot level between spores and their fruiting bodies (P > 0.05). Thus, in most cases, fruiting bodies were found in only a few plots, but their spores were collected in a greater number of plots (e.g., Mycena pura) (Fig. 5c). In other cases, the plots with high levels of fruiting body production also had the highest spore abundance, but no quantitative relationships were found (e.g., Lactarius vinosus) (Fig. 5d). Among the saprotrophic fungal species, there was a clear altitudinal pattern (F[1,6] = 2.48; P = 0.047) (Fig. 6), with both fruiting bodies and spores decreasing toward higher altitudes, whereas no clear altitudinal pattern was observed among ectomycorrhizal fungal spores (F[1,6] = −0.28; P = 0.785) (Fig. 6).

Spatial relationships between fruiting body yields (number of fruiting bodies) and the relative abundance of spores evaluated by MiSeq for saprotrophic species (a), ectomycorrhizal species (b), Mycena pura (c), and Lactarius vinosus (d). Blue or green ...
Relationship between fruiting body yields (solid line) and their propagules (broken line) for ectomycorrhizal (a) and saprotrophic (b) species along an elevation gradient represented by all of the sampled plots considered in this study.


In this study, we showed that spore trapping coupled with qPCR or MiSeq can detect the emergence of epigeous mushrooms. This study demonstrates that qPCR and MiSeq can be used to detect fungal emergence. In general, a 1-week delay between mushroom emergence and airborne propagule detection was found using spore traps. This delay was most likely caused by a combination of (i) the time needed for mushrooms to mature and produce spores, and (ii) the time needed for spore ejection and deposition in the spore traps. However, an alternative explanation for the observed delay may be that our fruiting body sampling scheme favors the sampling of immature fruiting bodies before they initiate sporulation. It is generally accepted that spore deposition is driven by meteorological conditions. In theory, the initial phase of spore dispersal of some basidiomycete mushroom-forming species involves the discharge of basidiospores from the gill by a mechanism promoted by a droplet, also known as a Buller's drop (26), which is normally stimulated by the secretion of mannitol and other sugars (27). Then, convective airflows promoted by the pileus enhance the vertical movement of these spores, which may then reach dispersive winds (28). Finally, these spores may fall by gravity or rainfall events. However, in our study, we observed that spore deposition occurred regardless of rainfall events, indicating that spore deposition is not necessarily driven by meteorological conditions such as precipitation events (29). Future studies should focus on gaining a better understanding of the potential delay in spore detection and the factors driving spore deposition, such as the effect of winds on the movement of these spores.

In this study, we observed an interspecific short temporal variation in the abundance of spores, especially those produced by ectomycorrhizal fungal species. This finding agrees with the literature because spores can remain airborne for between a day and up to a few weeks, depending on their size and aerodynamic properties (30). The temporal variation in spore abundance has been reported previously (16, 20), and this variation has been attributed to different abiotic parameters, such as wind direction, rainfall, and species-specific life history traits (29, 31, 32). Based on our results, spore detection was related to mushroom production peaks; thus, it seems likely that seasonal differences in the composition of the airborne fungal community can also be partly driven by mushroom phenology (20, 33).

Despite the observed relationship between mushroom yields and spore abundance across the sampling weeks, there was no relationship at the spatial scale where mushroom production was assessed, i.e., plots of 10 by 10 m. Spores were captured in plots without fruiting bodies of the corresponding species, indicating that these spores were produced outside the sampled plots. The spores of ectomycorrhizal fungal species are usually dispersed approximately 10 to 1,000 m (16, 19, 21), depending on the species (17). The large amount of spores produced per fruiting body, e.g., 1.1 × 108 to 1 × 1010 spores (34, 35), makes it possible for spores of these species to be dispersed long distances, particularly if carried away by turbulences. The fine mapping of spore captures in future studies could be used to indirectly estimate dispersal curves for several species and, therefore, understand better how these fungal species disperse.

MiSeq analysis detected significantly greater relative proportions of our 12 selected fungal species belonging to the saprotrophic or ectomycorrhizal guilds in filter traps than in funnel traps. Funnel traps trapped a greater number of species from other fungal guilds, such as yeasts, than the filter traps, which translated into a relative decrease in the proportion of saprotrophic and ectomycorrhizal MiSeq reads. Differences between samples can arise because of the way in which samples were handled. Spores captured in the funnel traps were first filtered in situ and subsequently collected inside a jar with water (17), which may favor the growth of some opportunistic fungi such as yeasts. Both fungus-specific amplicon sequencing and species-specific quantitative real-time PCR successfully detected mushroom production peaks. Both methods revealed an almost identical pattern of spore abundance across the weeks for L. vinosus. The use of MiSeq avoids the need to design specific primers for each target species. However, the use of high-throughput sequencing data as a semiquantitative measure has been the subject of discussion, mostly because of the variation in ribosomal gene copy numbers among different fungal species (36), the read length biases caused by current high-throughput sequencing platforms (37), the DNA extraction efficiency (38), or biases due to interspecific primer binding differences. Despite these limitations, read abundance may still be used as a semiquantitative measure of abundance (relative proportions) (23).

We showed that a peak in ephemeral fruiting bodies was followed by a peak in the numbers of spores that were trapped. For most of the species, the spore peak mostly disappeared 2 weeks after mushroom emergence reached its highest level. Nevertheless, a quantitative relationship between fruiting body yields and their fungal spores was not observed for Inocybe glabripes and Lycoperdon perlatum. Although the spore abundance and fruiting body yields observed for I. glabripes showed a similar pattern, the lack of relationship between these two measures was most likely caused by a peak in spore abundance that was not reflected at the fruiting body level. Interestingly, the spore abundance pattern for L. perlatum suggests a progressive and slow spore deposition over time; the lack of relationship between spore abundance and time may be because these spores are rarely falling unless a rainfall event occurs, such as the event that occurred during the week of 4 December, which most likely washed away all of the spores.

Some species produced fruiting bodies, but their spores were not found among the spore community. Possible explanations for this include the following: (i) these species had a very short range of dispersal, or (ii) their sporulation was very small compared with the rest of the species. Alternatively, these species may have a very long ITS2 sequence and, thus, be underrepresented using MiSeq sequencing. We investigated whether this was a plausible explanation by studying the ITS2 region length of these species using ITS sequences from UNITE, specifically if the region theoretically amplified by the primers was longer and if the DNA sequence where each primers binds was different for the fungal species not detected at the spore traps. However, the lengths of the ITS2 regions of these species were similar to those of other species that were successfully captured by the spore traps. Primer sequence mismatch was also discarded as an alternative explanation because these species did not show any sequence mismatch where the primer binds. Thus, discarding these last possibilities, and given that fruiting bodies were collected <10 m from spore traps, we hypothesize that low sporulation may be the reason for the lack of detection of these species among the spore community.

Filter traps are more suitable for high-throughput sequencing approaches targeting saprotrophic or ectomycorrhizal fungal species. Our method represents a potential improvement on mushroom sampling approaches, as current sampling schemes are highly time-consuming and require expert knowledge to identify the fungal taxa (5). Here, we showed that mushroom production and fungal diversity of fungi with epigeous fruiting body structures can be monitored accurately by spore traps. Thus, spore trapping could represent a less time-consuming alternative for monitoring mushrooms, which is fundamental for effective management planning for these important nonwood forest products.


Study area.

The study was carried out at a long-term experimental setup located in the natural area of Poblet (northeastern Spain, 41°21′6.4728′'N latitude and 1°2′25.7496′'E longitude), which is characterized by a Mediterranean climate with an average annual temperature of 11.8°C and annual rainfall of 665.5 mm. The long-term experimental site comprises 28 fenced plots, of which we selected 8 plots (10 by 10 m), where fruiting bodies have been continuously monitored every fall since 2008 (39). Even-aged (60 years old) Pinus pinaster and Quercus ilex are the codominant species growing in the plots. Additional information about the study area and the plots can be found in Table 3.

Main sampling plot characteristicsa

Fruiting body sampling.

All plots were sampled for mushroom production weekly during the fall (October to December 2014) following previously described methods (5). In short, all of the fruiting bodies within the plot and with a minimum cap size of 1 cm were sampled on Thursdays to minimize errors resulting from recreational weekend collectors picking mushrooms. The plots were chosen according to the productivity gradient for Lactarius vinosus (from the highest to the lowest production), which was based on continuous production data that have been recorded since 2008 (39). All of the fruiting bodies from all of the species found within the plots were brought to the laboratory on the same day, identified, cleaned, and counted. Fruiting bodies were dried in an air-vented oven at 35°C to 40°C and weighed to the nearest 0.01 g. Data in this study were expressed as the number of fruiting bodies (n) per plot in kilograms · hectare–1 to show the temporal variation of L. vinosus fruiting bodies over time.

Spore trapping.

Two types of spore traps were installed: filter traps (13) and funnel traps (16). The two traps were placed 1 m apart in the center of each plot and captured spores 30 cm above ground level.

Funnel traps consisted of 15-cm-diameter funnels attached to 1-liter dark jars, with a 50-μm-pore-size nylon mesh fixed at the bottom of the funnel. These traps were erected 1 week after the first mushroom of L. vinosus was observed (14 October 2014) and remained in place until 11 December 2014, when no L. vinosus fruiting bodies were observed. Each week, the funnels were rinsed with ultrapure water (Milli-Q) to collect any spores, and then the jars were immediately removed and replaced with sterile jars. Liquids from the traps were filtered using sterile filter papers (90-mm diameter; Whatman no. 1) within 48 h of collection. Sample filtering was conducted in a flow chamber to prevent potential contamination. Filters were stored at −20°C until further analysis.

Filter traps consisted of a filter paper (Whatman no. 1; 90-mm-diameter filter paper) placed over a metal mesh, supported by a metal clamp, and attached to a vertical support 30 cm above ground level. The filter papers in the filter traps were sampled simultaneously to the funnel traps during the two most productive weeks when L. vinosus fruiting body production was greatest (from 14 October until 29 October 2014) and stored at −20°C after sampling.

Spore trap sample processing and DNA extraction.

Filters containing spores from the filter and the funnel traps were cut in half and placed in separate 50-ml Falcon tubes. A solution of 20 ml of sodium dodecyl sulfate (SDS buffer) was added to each tube before incubating at 65°C for 90 min. The tubes were vortexed three times before removing the filter from each tube. Twenty milliliters of 2-propanol was added to the resulting solution and then left overnight at room temperature. After centrifugation (700 × g for 10 min), the supernatant was carefully removed and 700 μl of SL2 lysis buffer (NucleoSpin NSP soil DNA extraction kit; Macherey-Nagel, Duren, Germany) was added. The resultant solution was vortexed and transferred to a 2-ml tube. After the addition of SX enhancer (NucleoSpin NSP soil DNA extraction kit), the spore solution was homogenized in a FastPrep-24 system (MP Biomedicals) at 5,000 × g for 30 s (twice) following the instructions provided by the manufacturer. The DNA obtained was eluted in a 50-μl elution buffer.

Quantification of Lactarius vinosus using real-time PCR.

For Lactarius vinosus spore quantification, we used a species-specific hydrolysis probe (40). In order to obtain comparable results between different plates, we prepared standard curves using known amounts of L. vinosus DNA, which was extracted with the NucleoSpin NSP soil kit (Macherey-Nagel, Duren, Germany). Threshold cycle (CT) values were converted to the number of L. vinosus spores in each reaction using serial DNA dilutions of known amounts of spores, starting with 15.10 × 106 spores and ending with 151.2 spores. Results are expressed as the number of spores · trap sample−1.

Real-time PCRs were prepared using 2× Premix Ex Taq (TaKaRa Bio Europe SAS, France) following the manufacturer's recommendations. The reaction mix was prepared as follows: 5 μl of DNA template, 400 nM each oligonucleotide, 200 nM TaqMan probe, 0.8 μl of ROX (rhodamine X), and a volume of water (high-pressure liquid chromatography [HPLC]) to adjust the final reaction volume to 20 μl. The cycling conditions in the StepOnePlus instrument (Applied Biosystems) were as follows: 30 s at 95°C followed by 40 cycles at 95°C for 5 s and at 60°C for 34 s. There were three replicates of each sample and the standards were included in the analysis, as well as a negative control using HPLC water instead of the template. Only amplification efficiencies ranging from 95% to 105% were accepted and considered for analysis. qPCRs, processing of data, and quantification of L. vinosus DNA were performed following the methods described in a previous study (11).

Spore trap sample sequencing using Illumina MiSeq.

Each spore trap sample was PCR amplified using the primers fITS7 and ITS4 (41), which amplify the ITS2 region of the ribosomal DNA (rDNA). Both primers were tagged using unique DNA sequences composed of eight bases. A PCR cycle test was performed prior to randomly selected samples in order to perform the minimum number of PCR cycles possible. PCR amplifications of samples and both negative controls from DNA extraction and PCR were conducted in a 2720 thermal cycler (Life Technologies) in 50-μl volumes. The final concentrations in the PCR mixture were as follows: 25 ng of template, 200 μM each nucleotide, 2.75 mM MgCl2, 200 nM each primer, and 0.025 U μl–1 polymerase (DreamTaq Green; Thermo Scientific, Waltham, MA) in 1× buffer PCR. The cycling conditions for PCR were as follows: 5 min at 95°C, followed by 24 to 30 cycles of 30 s at 95°C, 30 s at 56°C, and 30 s at 72°C, and a final extension step at 72°C for 7 min before storage at 4°C. Each sample was amplified in triplicate, purified using an AMPure kit (Beckman Coulter Inc., Brea, CA), and quantified using a Qubit fluorometer (Life Technologies, Carlsbad, CA). Equal amounts of DNA from each sample were pooled before sequencing. The final equimolar mix was purified using an EZNA Cycle Pure kit (Omega Bio-tek). Quality control of purified amplicons was carried out using a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA) 7500 DNA chip. Libraries were prepared from ~10 ng of fragmented DNA using the ThruPLEX-FD prep kit. The samples were sequenced using the Illumina MiSeq platform, with 300-bp paired-end read lengths, generating 13.4 million sequences.

Quality control and bioinformatic analysis.

Quality control, filtering, and clustering were assessed using the SCATA pipeline ( Sequences were filtered to remove data with a minimum allowed base quality score of <10 at any position, an average quality score of <20, and a minimum sequence length of 200 bp using the amplicon quality option. Sequences were also screened for primers (using 0.9 as a minimum proportional primer match for both primers) and sample tags. We used “usearch” as a search engine, considering a minimum match length of 85%. Homopolymers were collapsed to 3 bp before cluster analysis. Pairwise alignments were conducted using a mismatch penalty assigned a value of 1, a gap open penalty of 0, and a gap extension penalty with a value of 1. Sequences were clustered in OTUs with single linkage clustering, using 1.5% as a threshold distance with the closest neighbor. After quality control and clustering, all tags were identified and tag jumps were removed from the database (42). A tag jump has been defined as the generation of artifactual sequences in which amplicons carry tags different from those originally applied.

Identification of the fungal clusters.

From a total of 13,408,476 MiSeq DNA sequences, 1,992,529 passed the quality control. Of these, 95,104 (4.77%) sequences were discarded because they had two different tags. Finally, 1,897,425 reads were obtained, with an average of 19,361 reads per sample.

We searched among the most abundant 2,400 OTUs (representing OTUs with more than 20 read counts) for those potential fungal species representing the fruiting body community. We first identified the OTUs using the seriateBLAST option in the UNITE database (43). We included DNA sequences from both UNITE and INSD, but preference was given to the UNITE reference for identification. In case we could not find any given species, we verified that each species had a known reference ITS sequence by checking UNITE and INSD in order to differentiate those species not found among the spore community from those without a known ITS reference sequence. Taxonomic assignment of the OTUs was given using a 98.5% sequence similarity threshold, and all of the species found at or above this threshold were included in each OTU. Read count data were transformed to relative proportions, and data are shown as counts per thousand of the total read numbers per sample.


All statistical analyses were implemented in the R software environment (version 2.15.3; R Development Core Team 2013) using the “nlme” package for linear mixed models (LME [44]), the vegan package (45) for the multivariate analysis, and the “gstat” (46) and “maps” (47) packages for spatial kriging.

Comparison of spore traps and of the two molecular methods used to quantify the fungal spores.

LME models were used to test differences between spore traps (filter spore traps and funnel traps) using both qPCR and MiSeq data for L. vinosus and only MiSeq data for the other species. In these models, the temporal dependency was considered by defining plots nested with “week” as a random factor, whereas the square root spore abundance of 12 species was defined as a fixed factor. In these analyses, we included the 12 species considered in this study (6 ectomycorrhizal and 6 saprotrophic fungal species) in separate models and randomly selected yeast (i.e., Cryptococcus spp. and Rhodotorula spp.) and mold species (Mortierella elongata and Trichoderma sp.) to test whether the hypothetical effect of the spore trap type was due to the increase of some of these fungal guilds.

Temporal relationship between fungal spores and mushroom yields.

Temporal (week) and spatial (plot) effects on community data were tested using permutational multivariate analysis of variance (PERMANOVA) based on distance matrices (using the function adonis and Bray-Curtis distances). Plot and week were included as factors, and community data were previously Hellinger transformed. For this analysis, the fruiting body community data represented species that produced fruiting bodies in at least two plots for at least 2 weeks.

We analyzed the temporal (week) linear relations between spore abundance and the mushroom yields using LME models, using square root transformed qPCR and MiSeq data. To test for lagged relationships, different LME models were tested, with and without accounting for 1-week temporal autocorrelation among observations (corAR1[form~week]). The best model fit was selected using the Akaike information criterion (AIC) value. We confirmed this lagged relationship with LME models using data in which spores were 1 week manually lagged. Only species that produced fruiting bodies for at least 2 weeks were considered for analysis. As mushroom fresh weight, dry weight, and the number of mushrooms ha−1 were highly correlated (P < 0.001; R2 = 0.9), we only used fruiting body counts for analysis. Data were modeled with plots defined as random to deal with the intraplot variation and either qPCR data (L. vinosus) or MiSeq data (for each species) were used as explanatory variables.

Relationship between fungal spores and mushroom yields at plot level.

The distribution of fungal spores and mushroom counts at the spatial scale was obtained by ordinary kriging interpolating each value using the inverse distance weighting (IDW) function. Interpolated values were graphically represented using Universal Transverse Mercator (UTM) coordinates (European Datum 1950 [ED50]). The relationship between spores and fruiting bodies over the plots was tested with LME models, using only species found in more than two plots. In this case, data were modeled with week defined as random to deal with the temporal variation.

Accession number(s).

Raw sequence reads have been deposited in the NCBI Sequence Read Archive (SRA) under accession number PRJNA352156.


We are very grateful to the PNIN of Poblet for supporting the process of installing and maintaining the experimental plots.

We also thank Katarina Ihrmark and Johanna Boberg for providing the protocol and for advice on how to process and analyze the spore trap samples. We thank Josep Miró and Francesc Bolaño for their assistance with collecting the spore trap and fruiting body samples. We thank Caroline Woods for linguistic revision and for helping to clarify the manuscript.

This work was supported by an STSM grant from COST Action FP1203 and by the Spanish Ministry of Economy and Competitiveness (MINECO) through the projects MICOGEST AGL2012-40035-C03 and MYCOSYSTEMS AGL2015-66001-C3. Carles Castaño received the support of the Secretaria d'Universitats i Recerca del Departament d'Economia i Coneixement de la Generalitat de Catalunya through the program of Doctorats Industrials. José Antonio Bonet is supported by the Serra-Hunter fellowship, and Josu G. Alday is supported by the Juan de la Cierva fellowship (IJCI-2014-21393).

The funders had no role in the study design, data collection, and interpretation, or in the decision to submit the work for publication.


1. Boa E. 2004. Wild edible fungi: a global overview of their use and importance to people. Food and Agriculture Organization of the United Nations, Rome, Italy.
2. Peintner U, Schwarz S, Mešić A, Moreau PA, Moreno G, Saviuc P 2013. Mycophilic or mycophobic? Legislation and guidelines on wild mushroom commerce reveal different consumption behaviour in European countries. PLoS One 8(5):e63926. [PMC free article] [PubMed]
3. De Roman M, Boa E 2004. Collection, marketing and cultivation of edible fungi in Spain. Micol Apl Int 16:25–33.
4. Palahí M, Pukkala T, Bonet JA, Colinas C, Fischer CR, Martínez de Aragón J 2009. Effect of inclusion of mushroom values on the optimal management of even-aged pine stands of Catalonia. Forest Sci 55:503–511.
5. Martínez de Aragón J, Bonet JA, Fischer CR, Colinas C 2007. Productivity of ectomycorrhizal and selected edible saprotrophic fungi in pine forests of the pre-Pyrenees mountains, Spain: predictive equations for forest management of mycological resources. For Ecol Manage 252:239–256. doi:.10.1016/j.foreco.2007.06.040 [Cross Ref]
6. Vogt KA, Blomfield J, Ammirati JF, Ammirati SR 1992. Sporocarp production by basidiomycetes with emphasis on forest ecosystems, p 563–581. In Dighton J, White JF, Oudemans P, Carroll GC, Wicklow DT, (), The fungal community, its organization and role in the ecosystem, 2nd ed. Taylor & Francis, Boca Raton, FL.
7. Liu B, Bonet JA, Fischer CR, Martínez de Aragón J, Bassie L, Colinas C 2016. Lactarius deliciosus Fr. soil extraradical mycelium correlates with stand fruitbody productivity and is increased by forest thinning. For Ecol Manage 380:196–201. doi:.10.1016/j.foreco.2016.08.053 [Cross Ref]
8. van der Linde S, Alexander IJ, Anderson IC 2009. Spatial distribution of sporocarps of stipitate hydnoid fungi and their belowground mycelium. FEMS Microbiol Ecol 69:344–352. doi:.10.1111/j.1574-6941.2009.00716.x [PubMed] [Cross Ref]
9. De la Varga H, Agueda B, Martínez-Peña F, Parladé J, Pera J 2012. Quantification of extraradical soil mycelium and ectomycorrhizas of Boletus edulis in a Scots pine forest with variable sporocarp productivity. Mycorrhiza 22:59–68. doi:.10.1007/s00572-011-0382-2 [PubMed] [Cross Ref]
10. De la Varga H, Águeda B, Ágreda T, Martínez-Peña F, Parladé J, Pera J 2013. Seasonal dynamics of Boletus edulis and Lactarius deliciosus extraradical mycelium in pine forests of central Spain. Mycorrhiza 23:391–402. doi:.10.1007/s00572-013-0481-3 [PubMed] [Cross Ref]
11. Parladé J, Hortal S, Pera J, Galipienso L 2007. Quantitative detection of Lactarius deliciosus extraradical soil mycelium by real-time PCR and its application in the study of fungal persistence and interspecific competition. J Biotechnol 128:14–23. doi:.10.1016/j.jbiotec.2006.09.010 [PubMed] [Cross Ref]
12. Jackson SL, Bayliss KL 2011. Spore traps need improvement to fulfil plant biosecurity requirements. Plant Pathol 60:801–810. doi:.10.1111/j.1365-3059.2011.02445.x [Cross Ref]
13. Schweigkofler W, O'Donnell K, Garbelotto M 2004. Detection and quantification of airborne conidia of Fusarium circinatum, the causal agent of pine pitch canker, from two California sites by using a real-time PCR approach combined with a simple spore trapping method. Appl Environ Microbiol 70:3512–3520. doi:.10.1128/AEM.70.6.3512-3520.2004 [PMC free article] [PubMed] [Cross Ref]
14. Chandelier A, Helson M, Dvorak M, Gischer F 2014. Detection and quantification of airborne inoculum of Hymenoscyphus pseudoalbidus using real-time PCR assays. Plant Pathol 63:1296–1305. doi:.10.1111/ppa.12218 [Cross Ref]
15. Savage D, Barbetti MJ, MacLeod WJ, Salam MU, Renton M 2012. Mobile traps are better than stationary traps for surveillance of airborne fungal spores. Crop Prot 36:23–30. doi:.10.1016/j.cropro.2012.01.015 [Cross Ref]
16. Peay KG, Bruns TD 2014. Spore dispersal of basidiomycete fungi at the landscape scale is driven by stochastic and deterministic processes and generates variability in plant-fungal interactions. New Phytol 204:180–191. doi:.10.1111/nph.12906 [PubMed] [Cross Ref]
17. Peay KG, Schubert MG, Nguyen NH, Bruns TD 2012. Measuring ectomycorrhizal fungal dispersal: macroecological patterns driven by microscopic propagules. Mol Ecol 21:4122–4136. doi:.10.1111/j.1365-294X.2012.05666.x [PubMed] [Cross Ref]
18. Adams RI, Miletto M, Taylor JW, Bruns TD 2013. Dispersal in microbes: fungi in indoor air are dominated by outdoor air and show dispersal limitation at short distances. ISME J 7:1262–1273. doi:.10.1038/ismej.2013.28 [PMC free article] [PubMed] [Cross Ref]
19. Galante TE, Horton TE, Swaney DP 2011. 95% of basidiospores fall within 1 m of the cap: a field- and modeling-based study. Mycologia 103:1175–1183. doi:.10.3852/10-388 [PubMed] [Cross Ref]
20. Kivlin SN, Winston GC, Goulden ML, Treseder KK 2014. Environmental filtering affects soil fungal community composition more than dispersal limitation at regional scales. Fungal Ecol 12:14–25. doi:.10.1016/j.funeco.2014.04.004 [Cross Ref]
21. Li D-W. 2005. Release and dispersal of basidiospores from Amanita muscaria var. alba and their infiltration into a residence. Mycol Res 109:1235–1242. doi:.10.1017/S0953756205003953 [PubMed] [Cross Ref]
22. West JS, Atkins SD, Emberlin J, Fitt BDL 2008. PCR to predict risk of airborne disease. Trends Microbiol 16:380–387. doi:.10.1016/j.tim.2008.05.004 [PubMed] [Cross Ref]
23. Amend AS, Seifert KA, Bruns TD 2010. Quantifying microbial communities with 454 pyrosequencing: does read abundance count? Mol Ecol 19:5555–5565. doi:.10.1111/j.1365-294X.2010.04898.x [PubMed] [Cross Ref]
24. Górriz-Mifsud E, Domínguez-Torres G, Prokofieva I 2015. Understanding forest owners' preferences for policy interventions addressing mushroom picking in Catalonia (north-east Spain). Eur J For Res 134:585–598. doi:.10.1007/s10342-015-0874-2 [Cross Ref]
25. Voces R, Diaz-Balteiro L, Alfranca Ó 2012. Demand for wild edible mushrooms. The case of Lactarius deliciosus in Barcelona (Spain). J For Econ 18:47–60.
26. Stolze-Rybczynski JL, Cui Y, Stevens MHH, Davis DJ, Fischer MWF, Money NP 2009. Adaptation of the spore discharge mechanism in the Basidiomycota. PLoS One 4(1):e4163. doi:.10.1371/journal.pone.0004163 [PMC free article] [PubMed] [Cross Ref]
27. Webster J, Davey RA, Smirnoff N, Fricke W, Hinde P, Tomos D, Turner JCR 1995. Mannitol and hexoses are components of Buller's drop. Mycol Res 99:833–838. doi:.10.1016/S0953-7562(09)80737-5 [Cross Ref]
28. Dressaire E, Yamada L, Song B, Roper M 2016. Mushrooms use convectively created airflows to disperse their spores. Proc Natl Acad Sci U S A 113:2833–2838. doi:.10.1073/pnas.1509612113 [PubMed] [Cross Ref]
29. Oliveira M, Ribeiro H, Delgado JL, Abreu I 2009. The effects of meteorological factors on airborne fungal spore concentration in two areas differing in urbanisation level. Int J Biometeorol 53:61–73. doi:.10.1007/s00484-008-0191-2 [PubMed] [Cross Ref]
30. Després VR, Huffman JA, Burrows SM, Hoose C, Safatov AS, Buryak G, Fröhlich-Nowoisky J, Elbert W, Andreae MO, Pöschl U, Jaenicke R 2012. Primary biological aerosol particles in the atmosphere: a review. Tellus B 64:15598. doi:.10.3402/tellusb.v64i0.15598 [Cross Ref]
31. Burch M, Levetin E 2002. Effects of meteorological conditions on spore plumes. Int J Biometeorol 46:107–117. doi:.10.1007/s00484-002-0127-1 [PubMed] [Cross Ref]
32. di Giorgio C, Krempff A, Guiraud H, Binder P, Tiret C, Dumenil G 1996. Atmospheric pollution by airborne microorganisms in the city of Marseilles. Atmos Environ 30:155–160. doi:.10.1016/1352-2310(95)00143-M [Cross Ref]
33. Levetin E. 1990. Studies on airborne basidiospores. Aerobiologia 6:177–180. doi:.10.1007/BF02539111 [Cross Ref]
34. Dahlberg A, Stenlid J 1994. Size, distribution and biomass of genets in populations of Suillus bovinus (L.: Fr.) Roussel revealed by somatic incompatibility. New Phytol 128:225–234. doi:.10.1111/j.1469-8137.1994.tb04006.x [Cross Ref]
35. Kadowaki K, Leschen RA, Beggs JR 2010. Periodicity of spore release from individual Ganoderma fruiting bodies in a natural forest. Australas Mycol 29:17–23.
36. Baldrian P, Větrovský T, Cajthaml T, Dobiášová P, Petránková M, Šnajdr J, Eichlerová I 2013. Estimation of fungal biomass in forest litter and soil. Fungal Ecol 6:1–11. doi:.10.1016/j.funeco.2012.10.002 [Cross Ref]
37. Wommack KE, Bhavsar J, Ravel J 2008. Metagenomics: read length matters. Appl Environ Microbiol 74:1453–1463. doi:.10.1128/AEM.02181-07 [PMC free article] [PubMed] [Cross Ref]
38. Feinstein LM, Sul WJ, Blackwood CB 2009. Assessment of bias associated with incomplete extraction of microbial DNA from soil. Appl Environ Microbiol 75:5428–5433. doi:.10.1128/AEM.00120-09 [PMC free article] [PubMed] [Cross Ref]
39. Bonet JA, De Miguel S, Martínez de Aragón J, Pukkala T, Palahí M 2012. Immediate effect of thinning on the yield of Lactarius group deliciosus in Pinus pinaster forests in Northeastern Spain. For Ecol Manage 265:211–217. doi:.10.1016/j.foreco.2011.10.039 [Cross Ref]
40. Castaño C, Parladé J, Pera J, Martínez de Aragón J, Alday JG, Bonet JA 2016. Soil drying procedure affects the DNA quantification of Lactarius vinosus but does not change the fungal community composition. Mycorrhiza 26:799–808. doi:.10.1007/s00572-016-0714-3 [PubMed] [Cross Ref]
41. Ihrmark K, Bödeker ITM, Cruz-Martinez K, Friberg H, Kubartova A, Schenck J, Strid Y, Stenlid J, Brandström-Durling M, Clemmensen KE, Lindahl BD 2012. New primers to amplify the fungal ITS2 region–evaluation by 454-sequencing of artificial and natural communities. FEMS Microbiol Ecol 82:666–677. doi:.10.1111/j.1574-6941.2012.01437.x [PubMed] [Cross Ref]
42. Schnell IB, Bohmann K, Gilbert MTP 2015. Tag jumps illuminated–reducing sequence-to-sample misidentifications in metabarcoding studies. Mol Ecol Resour 15:1289–1303. doi:.10.1111/1755-0998.12402 [PubMed] [Cross Ref]
43. Kõljalg U, Larsson K-H, Abarenkov K, Nilsson RH, Alexander IJ, Eberhardt U, Erland S, Høiland K, Kjøller R, Larsson E, Pennanen T, Sen R, Taylor AFS, Tedersoo L, Vrålstad T, Ursing BM 2005. UNITE: a database providing web-based methods for the molecular identification of ectomycorrhizal fungi. New Phytol 166:1063–1068. doi:.10.1111/j.1469-8137.2005.01376.x [PubMed] [Cross Ref]
44. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team 2016. nlme: linear and nonlinear mixed effects models. R package version 3.1-128. The R Foundation, Vienna, Austria.
45. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O'Hara RB, Simpson GL, Solymos P, Stevens MHM, Wagner H 2015. VEGAN v2.1-10: community ecology package v 2.2-1. The R Foundation, Vienna, Austria.
46. Pebesma EJ. 2004. Multivariable geostatistics in S: the gstat package. Comput Geosci 30:683–691. doi:.10.1016/j.cageo.2004.03.012 [Cross Ref]
47. Becker R, Wilks A, Brownrigg R 2016. Package ‘maps’. The R Foundation, Vienna, Austria.

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)