No biogeographical pattern for a root-associated fungal species complex

Global Ecology and Biogeography, (Global Ecol. Biogeogr.) (2011) 20, 160–169

RESEARCH PA P E R

No biogeographical pattern for a root-associated fungal species complex

geb_589

160..169

Valentin Queloz1, Thomas N. Sieber1, Ottmar Holdenrieder1, Bruce A. McDonald2 and Christoph R. Grünig1*

1

Forest Pathology and Dendrology, Institute of Integrative Biology (IBZ), ETH Zurich, 8092 Zurich, Switzerland, 2Plant Pathology, Institute of Integrative Biology (IBZ), ETH Zurich, 8092 Zurich, Switzerland

A B S T R AC T

Aim The biogeography of microbes is poorly understood and there is an open debate regarding if and how microbial biodiversity is structured. At the beginning of the 20th century, Baas Becking laid the foundations for the biogeography of microbes by stating that ‘Everything is everywhere, but the environment selects’ (the EisE hypothesis). This hypothesis remained dogma for almost a century. However, the recognition that microbial ‘species’ are often assemblages of reproductively isolated lineages challenged the EisE hypothesis, leading to the now common assumption that microbial communities possess cryptic biogeographic structures. We tested the presence of a cryptic biogeographical structure for a well-characterized fungal species complex (the Phialocephala fortinii s.l.–Acephala applanata species complex, PAC) using precise molecular species resolution. In addition, we analysed factors that could govern PAC community assembling. Locations Forty-four study sites in temperate and boreal forests across the Northern Hemisphere were included. Methods (1) The distance–decay relationship among PAC communities was calculated and a resampling procedure was applied to analyse the effect of sampling intensity and geographic distances among PAC communities. (2) Factors shaping PAC communities (e.g. climatic factors and tree species composition) were studied. (3) We tested PAC communities for random composition. Results We found that the similarity of species assemblages did not decrease with increasing geographical distance. Moreover, species diversity did not increase by expanding the area sampled. Instead, species diversity increased by increasing the sampling effort. Community composition correlated neither with tree species composition nor climate, and no association among species was observed. Main conclusions We could not discover any cryptic biogeographic structure even after applying refined species assignment but we demonstrate the importance of sampling effort for understanding the biogeography of microorganisms. Moreover, we show that primarily stochastic effects are responsible for the species composition of PAC communities.

*Correspondence: Christoph. R. Grünig, Forest Pathology and Dendrology, Institute of Integrative Biology (IBZ), ETH Zurich, 8092 Zurich, Switzerland. E-mail: [email protected]

Keywords Cryptic species, distance–decay relationship, endophyte, everything is everywhere, microbial biodiversity, microorganisms, sampling intensity, species–area relationship.

I N TR O D U C TI O N The biogeography of microbes is poorly understood and there is an open debate regarding if and how microbial biodiversity is structured (Green et al., 2004; Bell et al., 2005; Martiny et al., 2006; Yang et al., 2010). Key factors that shape biodiversity in

160

microorganisms in space and time are the ability to disseminate, adaptive radiation, habitat differentiation and competition. Microorganisms have been regarded as cosmopolitan for many years because they have short generation times, huge populations sizes and disperse over long distances (Fenchel & Finlay, 2004), leading to the hypothesis of Baas Becking that ‘everything DOI: 10.1111/j.1466-8238.2010.00589.x © 2010 Blackwell Publishing Ltd www.blackwellpublishing.com/geb

Biogeography of a fungal species complex is everywhere, but the environment selects’ (the EisE hypothesis) (de Wit & Bouvier, 2006; O’Malley, 2007). However, the EisE hypothesis was challenged following the advent of molecular genetic markers that illustrated that species of microorganisms defined by morphological characters and/or conserved molecular markers are often assemblages of reproductively isolated lineages, i.e. cryptic species (CSP) (Taylor et al., 2000). This observation led researchers to hypothesize that a hidden biogeography of microorganisms may exist, and the influence of species definition became a controversial issue in studying the biogeography of microbes (Fenchel, 2005; Taylor et al., 2006; Peay et al., 2008). The precision of species assignment has a profound effect on the interpretation of the effects of isolation, competition and adaptive radiation in microbial assemblages. Indeed, hidden biogeographical structures were reported for several fungal microbes when resolution of species assignments was high (James et al., 1999; Jacobson et al., 2006; Geml et al., 2008) and in only a few cases has no biogeographic structure been found, especially for free-living fungal microbes (Pringle et al., 2005). The ascomycete Phialocephala fortinii s.l. (Pezizomycotina, Leotiomycetes, Helotiales) is an example of a morphologically defined microbial taxon composed of several CSP, seven of which were recently formally described (Grünig et al., 2008b). A closely related species was described as Acephala applanata (Grünig & Sieber, 2005). This assemblage of species is now known as the Phialocephala fortinii s.l.–Acephala applanata species complex (PAC). Although sequencing of the internal transcribed spacer (ITS) regions of the rDNA is often regarded as a ‘gold standard’ in species diagnosis in fungi (Crous & Groenewald, 2005; Peay et al., 2007), the resolution of ITS sequences was not sufficient to differentiate species in this complex (Grünig et al., 2004). Instead, several classes of molecular markers were developed for PAC species assignment including polymerase chain reaction (PCR) fingerprinting, single-copy restriction fragment length polymorphisms (RFLP), multilocus sequence typing and microsatellites. Each of these molecular markers supported the delineation of multiple species in this complex, with concordant cryptic species defined by all markers (Grünig et al., 2007; Queloz et al., 2010). Members of the PAC dominate the endophytic mycobiota in roots of conifers and members of the Ericaceae in heathlands, forests and alpine ecosystems (Addy et al., 2000; Grünig et al., 2006) and can be found on all parts of the root system, e.g. from the mycorrhizal root tips up to the root collar (Menkis, 2005; Grünig et al., 2008a). Among the strains sampled from a single study site, PAC species form communities of up to 10 sympatrically occurring species that were shown to remain stable for several years (Queloz et al., 2005). Species abundance distributions within these communities follow a hyperbolic distribution with a few abundant species and many ‘rare’ species (Grünig et al., 2006), consistent with the community structures observed in many other biological systems (McGill et al., 2007). Host specialization among PAC species is known to be low because many species have been isolated from a broad range of woody plant species. Only A. applanata shows a clear preference for

hosts belonging to the Pinaceae in forest ecosystems with ground vegetation formed by ericaceous plants (Grünig et al., 2006). Natural long-distance gene or genotype flow for soil-borne PAC species is assumed to be restricted (Grünig et al., 2008a) because: (1) A. applanata was never observed to sporulate and other PAC species rarely sporulate, and the conidia do not germinate in vitro; (2) PAC species have never been detected in arable soils (Ahlich-Schlegel, 1997; Brenn et al., 2008); and (3) no PAC species were detected in spore traps, although other species of the genus Phialocephala were recorded (Kauserud et al., 2005). But regional trade in colonized nursery plants may serve as an alternative source for the dissemination of PAC (Brenn et al., 2008) and could influence PAC community composition in plantations. Because of the limited ability of PAC species to disseminate and the high precision available to differentiate among PAC species, we expected to find evidence for a strong biogeographical pattern when comparing 5236 PAC isolates sampled from 44 study sites distributed across the Northern Hemisphere. We also sought to determine whether tree-species composition or climate (temperature, precipitation) influences PAC community assemblages. Surprisingly, we found no evidence for a biogeographic structure in this common, globally distributed species complex. MATER IAL S AND METHO DS Study sites included in the analysis We included 5236 isolates from 44 study sites sampled from across the Northern Hemisphere in our analysis. Europe was sampled most intensively (28 study sites) followed by North America (11 study sites) and Asia (5 study sites). The majority of sites were located in undisturbed or naturally regenerated forest to limit anthropogenic effects on PAC biogeography. Details on the study sites are given in Table 1. Two different sampling designs were applied. Most European study sites used 14 m ¥ 14 m plots (74 sampling points, five to seven root segments per sampling point) (Grünig et al., 2004). Study sites outside Europe were sampled using 50 m long transects (11 sampling points, 10 root segments per sampling point). Root endophytes were isolated from root segments as described previously (Grünig et al., 2002) and single-hyphal-tip cultures were prepared from darkly pigmented mycelia that emerged from the root segments. Species assignments Species definition in PAC follows a population genetic and phylogenetic approach as previously described by Grünig et al. (2007). Several classes of molecular genetic markers were combined to differentiate species within the PAC, including PCR fingerprinting (Grünig et al., 2002), single-copy RFLP (Grünig et al., 2004), multilocus sequence typing (Grünig et al., 2007) and microsatellites (Queloz et al., 2010). Each of these molecular markers supported the delineation of multiple species in this

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

161

162

Canada (Ontario) Canada (Québec) Canada (Alberta) USA (California) USA (California) USA (Maine) USA (Oregon) USA (Oregon) USA (Oregon) USA (Oregon) Austria Finland France France France France Italy Italy Lithuania Lithuania Lithuania Lithuania Poland Poland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Switzerland Ukraine Japan Kirgistan Kirgistan Nepal Russia

N America

Mansfield Val Cartier Redwater Yuba Sierra Nevada Marsh Island Catherine Creek Grass Creek Noonday Sheep Creek Rothwald Kevo Alpes Maritimes Le Kertoff La Fage Pyrénées Campolino Pollino Kelme Varena Varena plantation Veisejai Bialowieza Bialowieza transect Boedmeren Creux du Van Creux du Van transect Derborence Etang de la Gruère La Chaux La Chaux transect Runcaglia Scatlé Sellenbueren Uetliberg Walder Steinenberg Zinalgletscher Zuerichberg Tschornohora Fuji Ala Archa Chychkan Jiri Lake Baikal

Study site 44.215 N / 80.049 W 46.946 N / 71.496 W 53.926 N / 112.962 W 39.283 N / 121.283 W 39.509 N / 120.247 W 44.916 N / 68.657 W 45.181 N / 117.673 W 43.605 N / 122.57 W 43.593 N / 122.609 W 44.411 N / 122.198 W 47.776 N / 15.092 E 69.758 N / 26.975 E 44.088 N / 7.341 E 48.107 N / 6.852 E 45.063 N / 3.262 E 42.85 N / 1.063 E 44.119 N / 10.671 E 39.926 N / 16.211 E 55.65 N / 22.633 E 54.967 N / 24.5 E 54.333 N / 24.233 E 54.1 N / 23.733 E 52.77 N / 23.861 E 52.77 N / 23.861 E 46.983 N / 8.823 E 46.933 N / 6.733 E 46.933 N / 6.733 E 46.275 N / 7.215 E 47.239 N / 7.053 E 47.221 N / 7.04 E 47.228 N / 7.045 E 46.784 N / 9.394 E 46.791 N / 9.049 E 47.336 N / 8.432 E 47.367 N / 8.477 E 47.282 N / 8.353 E 46.097 N / 7.64 E 47.393 N / 8.565 E 48.136 N / 24.482 E 35.367 N / 138.786 E 42.553 N / 74.489 E 27.95 N / 86.667 E 42.081 N / 72.812 E 53.185 N / 107.343 E

Coordinates Plantation Plantation Natural regeneration Undisturbed Plantation Natural regeneration Undisturbed Undisturbed Natural regeneration Natural regeneration Undisturbed Undisturbed Undisturbed Undisturbed Undisturbed Undisturbed Undisturbed Undisturbed Nursery Nursery Plantation Nursery Undisturbed Undisturbed Undisturbed Undisturbed Plantation Undisturbed Undisturbed Undisturbed Undisturbed Natural regeneration Undisturbed Plantation Plantation Nursery Undisturbed Plantation Undisturbed Plantation Undisturbed Undisturbed Undisturbed Natural regeneration

Forest status Transect Transect Transect Transect Transect Transect Transect Transect Transect Transect Grid net Grid net Grid net Grid net Grid net Grid net Grid net Grid net Transect Transect Transect Transect Grid net Transect Grid net Grid net Transect Grid net Grid net Grid net Transect Grid net Grid net Grid net Grid net Grid net Transect Grid net Grid net Transect Transect Transect Transect Transect

Sampling scheme 110 110 110 110 110 110 110 110 110 110 515 508 456 487 372 462 508 387 100 150 100 100 447 80 494 468 66 537 656 1492 100 430 518 200 240 475 50 370 550 110 55 55 60 110

RF 84 65 37 26 59 58 50 37 57 44 248 270 256 208 185 26 152 40 24 71 42 12 161 47 290 253 27 244 290 419 39 222 308 75 227 57 11 271 115 28 18 19 11 53

PAC strains

76.4 59.1 33.6 23.6 53.6 52.7 45.5 33.6 51.8 40.0 48.2 53.1 56.1 42.7 49.7 5.6 29.9 10.3 24.0 47.3 42.0 12.0 36.0 58.8 58.7 54.1 40.9 45.4 44.2 28.1 39.0 51.6 59.5 37.5 94.6 12.0 22.0 73.2 20.9 25.5 32.7 34.5 18.3 48.2

COL

Sampling scheme: grid net, 14 m ¥ 14 m (196 m2), 74 grid points, five to seven root segments/grid point; transect, 50 m, 11 grid points, 10 root segments/grid point. Forest status: undisturbed, not managed; natural regeneration, managed but no seedlings planted; plantation, managed and seedlings planted; nursery, roots from seedlings in nursery beds. RF, total number of root fragments analysed; PAC strains, number of Phialocephala fortinii s.l.–Acephala applanata species complex (PAC) strains isolated and classified based on molecular markers; COL, percentage of root fragments colonized by PAC.

Asia

Europe

Country (State)

Region

Table 1 Description of study sites included in the present study.

V. Queloz et al.

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

Biogeography of a fungal species complex complex, with concordant species defined by all markers (Grünig et al., 2007; Queloz et al., 2010). Markers used to determine species in the study sites are given in Table S1 in Supporting Information. The following procedure was used to assign individual strains to species. In a first step, cluster analyses were performed for each marker type and study site independently using the freeware Populations 1.2.30. Concordant grouping of strains using different marker types was considered as evidence for the presence of CSP and helped to improve species assignment of strains (Grünig et al., 2004). Next, strains genotyped using microsatellite markers, i.e. strains from 40 of the 44 study sites, were assigned to species using GeneClass2 (Piry et al., 2004). This software was written to assign individuals to populations on the basis of multilocus genotypes. We used GeneClass2 to assign strains to species because closely related species can be regarded as populations separated by higher measures of population differentiation (Grünig et al., 2007). GeneClass2 was shown to be very effective in correctly assigning strains to PAC species, e.g. in the study of Queloz et al. (2010) 405 strains were assigned correctly to five PAC species. Strain assignment was performed by using the Bayesian criterion of Rannala & Mountain (1997). The dataset of Queloz et al. (2010) which includes 16 PAC species was used as the reference dataset. Special attention was given to ensure that newly recognized CSP were accurately defined. Congruence of at least two different classes of molecular markers was required to define new species (see Tables S1 & S2). Species diagnosis of the strains from the remaining four sites was based on a combination of single-copy RFLP, intersequence-specific (ISSR)-PCR and DNA sequence markers (Brenn et al., 2008). Assessment of sampling designs Rarefaction curves were used to determine whether PAC communities in each study site were sampled properly. Rarefaction curves are expected to reach a plateau if sampling has been exhaustive (Colwell & Coddington, 1994). The assignment of the strains to PAC species was recorded for each sampling point of a study site. Then rarefaction curves were calculated for each study site in Estimate S (Colwell, 2006) using the number of sampling points as a measure of sampling intensity. Analysing the biogeography of PAC species Two well-known community ecological approaches were applied to analyse the biogeography of PAC species. First, the distance–decay relationship of PAC communities was calculated (Green & Bohannan, 2007). Second, resampling was performed to test the effects of sampling intensity (number of study sites) and distance separating study sites on the number of species detected. The distance–decay relationship is one of the most commonly used analyses for assessing the cosmopolitanism of communities of microorganisms (Green & Bohannan, 2007). The distance– decay relationship assumes that community similarities will decrease with increasing geographical distance. But for truly

cosmopolitan microorganisms, the similarity of communities should be independent of the geographical distance separating them, i.e. communities separated by a few hundred metres or by thousands of kilometres should have the same probability of being very similar (or very different). We used the Morisita– Horn index to calculate dissimilarities among PAC communities sampled from different study sites. The Morisita–Horn index is less sensitive to missed species occurring at low frequencies than the Sørensen index that relies on presence/absence data only (Wolda, 1981; Zak & Willig, 2004). Community dissimilarities of PAC assemblages (Morisita–Horn) were calculated in R (R Development Core Team, 2006) and compared with pairwise distances among study sites using the function ‘mantel’ in the vegan package in R (http://r-forge.r-project.org/projects/ vegan/). In addition, the distance–decay relationship was recalculated as described above for a reduced dataset (31 sites) that did not include plantations to exclude effects due to humanmediated dissemination of PAC. In the resampling approach, groups of 2–14 study sites were randomly drawn from the 44 study sites, with 500 jackknife replicates per group. For each jackknife sample, the total number of species and the maximal distance (km) separating the communities were recorded. The maximum distance was used instead of the area sampled due to the large number of study sites and the difficulty in calculating precise areas for groups of study sites included in each jackknife sample. A general linear model was fitted to the artificial dataset using the number of species as the response variable and the number of study sites sampled as well as the maximum distance among study sites as the explanatory variables. This approach represents an alternative to the classical modelling of species–area relationships that is commonly used to analyse the biogeography of species (Rosenzweig, 1995; Green & Bohannan, 2007). The assumption is that more species will be detected as the sampling area increases in size (Tjørve, 2003). The effect of climate and tree-species composition on PAC assemblages Non-significant distance–decay or species–area relationships do not necessarily imply that PAC species are distributed randomly. Environmental factors such as similarities in climate or treespecies composition could mask possible distance–decay relationships (Martiny et al., 2006). Therefore, we tested the influence of climate (temperature and precipitation) and tree-species composition on the distribution of PAC species. First, climate data were obtained from the nearest meteorological station using either the climate explorer (http:// climexp.knmi.nl/start.cgi?someone@somewhere) or the Swiss Federal Office of Meteorology and Climatology (MeteoSwiss). For each study site, the average temperature in January and August for the past 20 years was calculated. Similarly, the mean precipitation for January and August was determined. Euclidean distances among study sites were calculated using the vegan package. The climate–distance matrix was then compared with community dissimilarities of PAC assemblages (Morisita–Horn)

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

163

V. Queloz et al. using the function ‘mantel’ in vegan. Pairwise distances between tree-species mixtures for each study site were calculated using the Sørensen index because frequency data were not available for tree-species composition. A Mantel test was performed using tree-species community distances and community dissimilarities of PAC assemblages as described above.

Testing PAC community assemblages for random composition We adapted the index of association (IA) to test whether species composition in PAC communities displays non-random composition patterns. The index of association was originally developed to detect multilocus linkage disequilibrium among loci of individuals in populations. In our analysis, we used the test to find associations among species in PAC communities. If the index of association deviates significantly from zero, this indicates that there is non-random association among loci, or in our case among species, meaning that some species combinations occur more frequently than expected by chance. For each community the presence or absence of a species was recorded result-

ing in a 0/1 matrix. Then, IA and its variance were calculated as described by Smith et al. (1993). R ESULTS The 5236 strains isolated from 12,808 root segments collected in 44 communities across the Northern Hemisphere were characterized and shown to belong to 21 PAC species (see Table S1). Five previously unknown species were discovered in the course of this analysis (CSP15, CSP17–20). These species occurred at low frequencies ranging from 0.08% to 0.78% of all strains. Assessment of sampling designs Appropriate sampling of communities is crucial to calculate similarities among communities and to analyse the distance– decay relationship. The number of recovered species varied from 1–10 at study sites independently of the sampling design (plots, transects) applied (Kruskal–Wallis test, a-level = 0.05, P = 0.059). Rarefaction curves showed that a plateau was reached for all study sites independently of whether transects or plots were sampled, suggesting that the PAC diversity in each study site was adequately sampled (Fig. 1).

Figure 1 Rarefaction analysis for selected study sites sampled using plots (a) or transects (b). 䉬, Creux du Van (number of strains, n, = 253); , Bialowieza (n = 161); 䉱, Rothwald (n = 248); 䉫, Catherine Creek (n = 50); 䊐, Marsh Island (n = 58); 䉭, Sierra Nevada (n = 59). Bars = standard deviation of the mean.

164

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

0.6 0.4 0.0

0.2

Morisita-Horn similarity

0.8

1.0

Biogeography of a fungal species complex • • • • ••• •• •• • •• •• • • • • • •• • ••••• • •• • • • ••• • • • • • • • •• • • • • •••• • •• • • • • ••••• ••• •••• •• •••• • • • • • • • •• • •• • •• • • • • • • • •• •• • • • ••••• •••• • ••• • • • • • • • • • • ••• • • •• • • •••••• • •• • • • • • • • ••• • ••• • • ••• • • •• • •• • • • • • •••••• •••• • • • • • • • • • •• • •• • •• • • •• • • • • •••• •• • •••• • • • • •• • •• • • •••••••• •• •••• • ••• • • • • •• •• •• • • • • •• • • •• • • • • ••• • •• •• •• •• ••••• ••••• • •• • • • • •••• •• • ••• •••• •• ••••• • • •• • • • • • • •• ••••• • • • •• •• • • •••••••••••••••••• •• • • •••••••••••• ••••• • • • • • • • • • • • • • • • • •• • ••••••••••••••••• ••••••••• ••• • •• • • •• • ••••• • • •• ••• • • • • •••••••••• ••• • •• • • • • • • • • • • •• • • ••• • ••••••••••• •• • • • • •• • ••••• ••• ••••• •• ••• • •••••••••• • •• ••••••••••••••••••••••••••••••••••••••• ••• • ••••••••• • ••••••••••••••• •••••••••• ••• •• •••••••••••••••••••••••••••••• ••• ••• 0

2,000

4,000

6,000

8,000

10,000

a-level = 0.01, climate P = 0.326, tree-species composition P = 0.588).

• ••

Testing PAC community assemblages for random composition

• • • • • • 12,000

Distance [km]

Figure 2 Distance–decay relationship for 44 Phialocephala fortinii s.l.–Acephala applanata communities. Pairwise community similarities (n = 946) were calculated using the Morisita–Horn index and plotted against the distance among study sites.

Distance–decay relationship and resampling analysis Forty-four PAC communities separated by distances ranging from 0.01 to 12,000 km (i.e. seven orders of magnitude) were included to examine the distance–decay relationship. No correlation was found between the Morisita–Horn similarity and geographical distance of communities (Mantel test, a-level = 0.01, P = 0.246) (Fig. 2). Similarly, no correlation was detected using the reduced dataset that included only undisturbed or naturally regenerated study sites (Mantel test, a-level = 0.01, P = 0.209). A resampling of the 44 PAC communities was performed to further test these findings. Whereas the maximal geographical distance among study sites did not correlate with the number of species detected (a-level = 0.01, P = 0.866), correlation between the number of study sites and the number of species was significantly positive (a-level = 0.01, coefficient (slope) = 0.78, P < 2.0 ¥10–16), indicating that the number of study sites is the main predictor of the number of species found in a defined geographical area. Mapping the geographical distributions of the most abundant species (Phialocephala subalpina) and one of the rarest species (CSP12) showed that both species were widespread (Fig. 3). CSP12 represented only 0.82% of all genotyped strains (43 strains) but was found in nine study sites on all studied continents (Europe, North America and Asia). Analyses of 14 study sites situated within a comparatively small area of c. 40,000 km2 in Switzerland led to the detection of 19 of the 21 species collected world-wide. Effect of climate and forest tree composition on PAC assemblages Similarities among PAC communities did not correlate with either tree-species composition or climate (Mantel tests,

Among the 44 PAC communities, species composition was unique in 41 communities. Species composition was identical in only three pairs of communities (Bialowieza–Valcartier; Bialowieza_transect–Tschornohora; Grass Creek–Noonday). No associations among PAC species in communities were found using the index of association measure (IA = 0.436, Var(Ve) = 0.558, P = 0.280) for the 44 study sites. DISC USSIO N We found that the similarity of microbial communities did not decrease with increasing geographical distance based on a dataset of fungal root endophytes comprising more than 5000 isolates of 21 PAC species sampled from across the Northern Hemisphere. Instead, species diversity increased with sampling effort. In addition, PAC community assemblages did not correlate with tree species composition or climate but seemed to be randomly assembled. Assessment of sampling designs An adequate sample must be drawn from each community to calculate similarities among communities and allow a robust analysis of the distance–decay relationship. In the rarefaction analysis, the PAC community of each study site reached a plateau, indicating that both sampling strategies allowed robust estimation of PAC species number and abundance. In contrast to many previous studies (e.g. Menkis, 2005; Tedersoo et al., 2006), we used the sampling point instead of the fungal strain as a measure of sampling intensity. Reaching a plateau in our case meant that we had sampled each species from at least two grid points independently. Earlier studies on below-ground fungal communities such as ectomycorrhizae rarely meet this standard (Peay et al., 2007). Despite the outcome of the rarefaction analysis, we cannot be certain that our sampling was sufficient to detect very rare species. Thus, we used the Morisita–Horn index to calculate dissimilarities among study sites because this measure is less sensitive to missed species with low frequencies than the Sørensen index (Zak & Willig, 2004). Biogeography of PAC species The EisE hypothesis could not be rejected for different microbial communities using morphological species concepts or highly conserved DNA sequence markers such as portions of the 18S or 28S rDNA (Finlay, 2002; Fenchel & Finlay, 2004; Martiny et al., 2006). But the EisE hypothesis has become controversial with the argument that classical morphological taxonomy and/or the use of conserved rDNA sequences are not suitable to distinguish all species (Taylor et al., 2006). Meanwhile, an increasing

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

165

V. Queloz et al.

Figure 3 Map representing the geographical distribution of the most abundant species (Phialocephala subalpina, black) and a rarely isolated species (CSP12, grey) through North America, Europe and Asia. Phialocephala subalpina and CSP12 represent 22.69% (1188 strains) and 0.82% (43 strains) of all genotyped strains, respectively. Nevertheless, both species were found on three continents.

number of CSP have been identified using highly polymorphic molecular genetic markers (Taylor et al., 2000; O’Donnell et al., 2004). As a consequence, fungi (Taylor et al., 2006), bacteria (Papke et al., 2003; Bell et al., 2005) and protists (Katz et al., 2005; Slapeta et al., 2005) previously believed to be cosmopolitan were redefined as assemblages of species with restricted biogeographical distributions and only few free-living fungal microbes are assumed to be truly cosmopolitan (Pringle et al., 2005). The PAC species used in this analysis were defined as reproductively isolated populations based on several classes of highly polymorphic molecular genetic markers (Grünig et al., 2004, 2007; Queloz et al., 2010). Despite the resulting high precision of resolution of PAC species, we found no biogeographical pattern for PAC species and we can confidently reject our hypothesis of a hidden biogeographic structure in PAC. The only predictor for species diversity was the sampling effort, determined by the number of exhaustively sampled study sites within a defined geographical area. The fact that PAC species are symbionts of non-motile hosts contradicts the idea that only free-living species are expected to be cosmopolitan (Taylor et al., 2006). Microorganisms are assumed to be cosmopolitan due to: (1) their large population sizes, and (2) their high rates of dispersal (Fenchel & Finlay, 2004). Precise measurements of population sizes for PAC species are not available. However, Godbold et al. (2003) reported that 1 m2 of spruce (Picea abies) forest soil typically harbours 1.7–2.2 km of fine roots with diameters of < 2 mm, indicating the enormous size of the preferred habitat of PAC species. Since the probability of isolating PAC from a randomly selected fine root segment ⱕ 5 mm long can be > 90% 166

(Table 1), PAC communities are expected to be extremely large, and large population sizes can also be assumed for less frequent PAC species. High dispersal rates for microorganisms were assumed for a long time, but recent studies have questioned this generalization (Martiny et al., 2006; Telford et al., 2006). Similarly, naturally occurring long-distance gene or genotype flow for soil-borne PAC species was found to be restricted (Brenn et al., 2008; Grünig et al., 2008a). Human-mediated genotype flow resulting from planting colonized nursery plants may have contributed to the dissemination of PAC species. Although previous studies showed that millions of trees originating from nurseries were planted (Bürgi & Schuler, 2003) and that nursery trees often are colonized by PAC (Kernaghan et al., 2003; Brenn et al., 2008), it is not known whether PAC on colonized nursery plants interferes with resident PAC communities (Brenn et al., 2008). Excluding plantations from our dataset did not change the finding of no significant distance–decay relationship. Which factors govern the assembly of PAC communities? The assembly and dynamics of communities of macroorganisms have received much attention (e.g. Hubbell, 2001; McGill et al., 2007). Models to describe and/or predict community structures of microorganisms have received much less attention, probably due to the difficulty of delineating species and individuals in microorganisms (Taylor et al., 2006; Grünig et al., 2007; Peay et al., 2008) as well as the difficulties associated with adequate sampling of microbial communities (Woodcock et al., 2006). Rules governing the assembly of PAC communities are not

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

Biogeography of a fungal species complex known, but our results shed at least some light on this question. The finding that tree-species composition does not correlate with the community structures found in PAC fits well with data showing that root symbionts in general (Jumpponen & Trappe, 1998), and PAC in particular, can be isolated from a broad range of plant species (Grünig et al., 2008a). For example, P. subalpina was isolated from 19 plant species belonging to six plant families including Poaceae (e.g. Deschampsia flexuosa). An exception to this rule is A. applanata that shows a preference for hosts belonging to the Pinaceae in forest ecosystems having ground vegetation formed by ericaceous plants (Grünig et al., 2006). Similarly, precipitation and temperature did not correlate with PAC community composition and can be ruled out as factors governing PAC community structure. The most intriguing result from our analysis was that PAC assemblages did not show significant associations among species, indicating that primarily stochastic effects are responsible for the composition of PAC communities. However, whether PAC communities really follow the principle of a neutral guild must be addressed in future studies. CO N C L U SI O N S In contrast to other recent analyses of microbial biogeography based on highly polymorphic molecular genetic markers (Taylor et al., 2006; Geml et al., 2008), our findings lead to the rejection of our hypothesis of a cryptic biogeography for PAC. Although we did not isolate all PAC species in all study sites, the absence of a biogeographic structure for PAC species is consistent with a broad geographical distribution of PAC species supporting the idea of Baas Becking that ‘everything is everywhere’. In addition, our study illustrates that an exhaustive and extensive sampling effort is critical to understand the biogeography of microorganisms. A C K N O W L ED G E M E N T S We gratefully acknowledge J. Dubé, R. S. Currah, M. E. Smith, P. Weisberg, W. H. Livingston, R. Danchok, J. Hill, J. Zaerr, A. Menkis, T. Nojima, M. Rehnus, G. Aas and L. Bont for making some of the collections used in this study. The Genetic Diversity Center (GDC, ETHZ) provided facilities for collecting much of the molecular genetic data. This study was funded by ETH-grant no. 0-20334-06 to V.Q. REF ER EN C ES Addy, H.D., Hambleton, S. & Currah, R.S. (2000) Distribution and molecular characterization of the root endophyte Phialocephala fortinii along an environmental gradient in the boreal forest of Alberta. Mycological Research, 104, 1213–1221. Ahlich-Schlegel, K. (1997) Vorkommen und Charakterisierung von dunklen, septierten Hyphomyceten (DSH) in Gehölwurzeln. PhD Thesis, Swiss Federal Institute of Technology, Zurich. Bell, T., Ager, D., Song, J.I., Newman, J.A., Thompson, I.P., Lilley, A.K. & van der Gast, C.J. (2005) Larger islands house more bacterial taxa. Science, 308, 1884.

Brenn, N., Menkis, A., Grünig, C.R., Sieber, T.N. & Holdenrieder, O. (2008) Community structure of Phialocephala fortinii s. lat. in European tree nurseries, and assessment of the potential of the seedlings as dissemination vehicles. Mycological Research, 112, 650–662. Bürgi, M. & Schuler, A. (2003) Driving forces of forest management – an analysis of regeneration practices in the forests of the Swiss Central Plateau during the 19th and 20th century. Forest Ecology and Management, 176, 173–183. Colwell, R.K. (2006) EstimateS: statistical estimation of species richness and shared species from samples. Available at: http:// viceroy.eeb.uconn.edu/estimates (accessed 15 August 2009). Colwell, R.K. & Coddington, J.A. (1994) Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society B: Biological Sciences, 345, 101–118. Crous, P.W. & Groenewald, J.Z. (2005) Hosts, species and genotypes: opinions versus data. Australasian Plant Pathology, 34, 463–470. Fenchel, T. (2005) Cosmopolitan microbes and their ‘cryptic’ species. Aquatic Microbial Ecology, 41, 49–54. Fenchel, T. & Finlay, B.J. (2004) The ubiquity of small species: patterns of local and global diversity. Bioscience, 54, 777–784. Finlay, B.J. (2002) Global dispersal of free-living microbial eukaryote species. Science, 296, 1061–1063. Geml, J., Tulloss, R.E., Laursen, G.A., Sazanova, N.A. & Taylor, D.L. (2008) Evidence for strong inter- and intracontinental phylogeographic structure in Amanita muscaria, a winddispersed ectomycorrhizal basidiomycete. Molecular Phylogenetics and Evolution, 48, 694–701. Godbold, D.L., Fritz, H.W., Jentschke, G., Meesenburg, H. & Rademacher, P. (2003) Root turnover and root necromass accumulation of Norway spruce (Picea abies) are affected by soil acidity. Tree Physiology, 23, 915–921. Green, J.L. & Bohannan, B.J.M. (2007) Biodiversity scaling relationships: are microorganisms fundamentally different? Scaling biodiversity (ed. by D. Storch, P.A. Marquet and J.H. Brown), pp. 129–149. Cambridge University Press, Cambridge. Green, J.L., Holmes, A.J., Westoby, M., Oliver, I., Briscoe, D., Dangerfield, M., Gillings, M. & Beattie, A. (2004) Spatial scaling of microbial eukaryote diversity. Nature, 432, 747– 750. Grünig, C.R. & Sieber, T.N. (2005) Molecular and phenotypic description of the widespread root symbiont Acephala applanata gen. et sp nov., formerly known as dark-septate endophyte Type 1. Mycologia, 97, 628–640. Grünig, C.R., Sieber, T.N., Rogers, S.O. & Holdenrieder, O. (2002) Spatial distribution of dark septate endophytes in a confined forest plot. Mycological Research, 106, 832–840. Grünig, C.R., McDonald, B.A., Sieber, T.N., Rogers, S.O. & Holdenrieder, O. (2004) Evidence for subdivision of the rootendophyte Phialocephala fortinii into cryptic species and recombination within species. Fungal Genetics and Biology, 41, 676–687. Grünig, C.R., Duò, A. & Sieber, T.N. (2006) Population genetic analysis of Phialocephala fortinii s.l. and Acephala applanata in

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

167

V. Queloz et al. two undisturbed forests in Switzerland and evidence for new cryptic species. Fungal Genetics and Biology, 43, 410– 421. Grünig, C.R., Brunner, P.C., Duò, A. & Sieber, T.N. (2007) Suitability of methods for species recognition in the Phialocephala fortinii–Acephala applanata species complex using DNA analysis. Fungal Genetics and Biology, 44, 773–788. Grünig, C.R., Queloz, V., Sieber, T.N. & Holdenrieder, O. (2008a) Dark septate endophytes (DSE) of the Phialocephala fortinii s.l.–Acephala applanata species complex in tree roots – classification, population biology and ecology. Botany, 86, 1355–1369. Grünig, C.R., Duò, A., Sieber, T.N. & Holdenrieder, O. (2008b) Assignment of species rank to six reproductively isolated cryptic species of the Phialocephala fortinii s.l.–Acephala applanata species complex. Mycologia, 100, 47–67. Hubbell, S.P. (2001) The unified neutral theory of biodiversity and biogeography. Princeton University Press, Princeton. Jacobson, D.J., Dettman, J.R., Adams, R.I., Boesl, C., Sultana, S., Roenneberg, T., Merrow, M., Duarte, M., Marques, I., Ushakova, A., Carneiro, P., Videira, A., Navarro-Sampedro, L., Olmedo, M., Corrochano, L.M. & Taylor, J.W. (2006) New findings of Neurospora in Europe and comparisons of diversity in temperate climates on continental scales. Mycologia, 98, 550–559. James, T.Y., Porter, D., Hamrick, J.L. & Vilgalys, R. (1999) Evidence for limited intercontinental gene flow in the cosmopolitan mushroom, Schizophyllum commune. Evolution, 53, 1665–1677. Jumpponen, A. & Trappe, J.M. (1998) Dark septate endophytes: a review of facultative biotrophic root-colonizing fungi. New Phytologist, 140, 295–310. Katz, L.A., McManus, G.B., Snoeyenbos-West, O.L.O., Griffin, A., Pirog, K., Costas, B. & Foissner, W. (2005) Reframing the ‘everything is everywhere’ debate: evidence for high gene flow and diversity in ciliate morphospecies. Aquatic Microbial Ecology, 41, 55–65. Kauserud, H., Lie, M., Stensrud, O. & Ohlson, M. (2005) Molecular characterization of airborne fungal spores in boreal forests of contrasting human disturbance. Mycologia, 97, 1215–1224. Kernaghan, G., Sigler, L. & Khasa, D. (2003) Mycorrhizal and root endophytic fungi of containerized Picea glauca seedlings assessed by rDNA sequence analysis. Microbial Ecology, 45, 128–136. McGill, B.J., Etienne, R.S., Gray, J.S., Alonso, D., Anderson, M.J., Benecha, H.K., Dornelas, M., Enquist, B.J., Green, J.L., He, F.L., Hurlbert, A.H., Magurran, A.E., Marquet, P.A., Maurer, B.A., Ostling, A., Soykan, C.U., Ugland, K.I. & White, E.P. (2007) Species abundance distributions: moving beyond single prediction theories to integration within an ecological framework. Ecology Letters, 10, 995–1015. Martiny, J.B.H., Bohannan, B.J.M., Brown, J.H., Colwell, R.K., Fuhrman, J.A., Green, J.L., Horner-Devine, M.C., Kane, M., Krumins, J.A., Kuske, C.R., Morin, P.J., Naeem, S., Øvreås, L., Reysenbach, A.L., Smith, V.H. & Staley, J.T. (2006) Microbial 168

biogeography: putting microorganisms on the map. Nature Reviews Microbiology, 4, 102–112. Menkis, A. (2005) Root associated fungi of conifer seedlings and their role in afforestation of agricultural land. PhD Thesis, Swedish University of Agricultural Sciences, Uppsala. O’Donnell, K., Ward, T.J., Geiser, D.M., Kistler, H.C. & Aoki, T. (2004) Genealogical concordance between the mating type locus and seven other nuclear genes supports formal recognition of nine phylogenetically distinct species within the Fusarium graminearum clade. Fungal Genetics and Biology, 41, 600–623. O’Malley, M.A. (2007) The nineteenth century roots of ‘everything is everywhere’. Nature Reviews Microbiology, 5, 647–651. Papke, R.T., Ramsing, N.B., Bateson, M.M. & Ward, D.M. (2003) Geographical isolation in hot spring cyanobacteria. Environmental Microbiology, 5, 650–659. Peay, K.G., Bruns, T.D., Kennedy, P.G., Bergemann, S.E. & Garbelotto, M. (2007) A strong species–area relationship for eukaryotic soil microbes: island size matters for ectomycorrhizal fungi. Ecology Letters, 10, 470–480. Peay, K.G., Kennedy, P.G. & Bruns, T.D. (2008) Fungal community ecology: a hybrid beast with a molecular master. Bioscience, 58, 799–810. Piry, S., Alapetite, A., Cornuet, J.M., Paetkau, D., Baudouin, L. & Estoup, A. (2004) GeneClass2: a software for genetic assignment and first-generation migrant detection. Journal of Heredity, 95, 536–539. Pringle, A., Baker, D.M., Platt, J.L., Wares, J.P., Latge, J.P. & Taylor, J.W. (2005) Cryptic speciation in the cosmopolitan and clonal human pathogenic fungus Aspergillus fumigatus. Evolution, 59, 1886–1899. Queloz, V., Grünig, C.R., Sieber, T.N. & Holdenrieder, O. (2005) Monitoring the spatial and temporal dynamics of a community of the tree-root endophyte Phialocephala fortinii s.l. New Phytologist, 168, 651–660. Queloz, V., Duò, A., Sieber, T.N. & Grünig, C.R. (2010) Microsatellite size homoplasies and null alleles do not affect species diagnosis and population genetic analysis in a fungal species complex. Molecular Ecology Resources, 10, 348–367. Rannala, B. & Mountain, J.L. (1997) Detecting immigration by using multilocus genotypes. Proceedings of the National Academy of Sciences USA, 94, 9197–9201. R Development Core Team (2006) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Rosenzweig, M.L. (1995) Species diversity in space and time. Cambridge University Press, Cambridge. Slapeta, J., Moreira, D. & Lopez-Garcia, P. (2005) The extent of protist diversity: insights from molecular ecology of freshwater eukaryotes. Proceedings of the Royal Society B: Biological Sciences, 272, 2073–2081. Smith, J.M., Smith, N.H., Orourke, M. & Spratt, B.G. (1993) How clonal are bacteria? Proceedings of the National Academy of Sciences USA, 90, 4384–4388. Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S. & Fisher, M.C. (2000) Phylogenetic species

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

Biogeography of a fungal species complex recognition and species concepts in fungi. Fungal Genetics and Biology, 31, 21–32. Taylor, J.W., Turner, E., Townsend, J.P., Dettman, J.R. & Jacobson, D. (2006) Eukaryotic microbes, species recognition and the geographic limits of species: examples from the kingdom Fungi. Philosophical Transactions of the Royal Society B: Biological Sciences, 361, 1947–1963. Tedersoo, L., Suvi, T., Larsson, E. & Köljalg, U. (2006) Diversity and community structure of ectomycorrhizal fungi in a wooded meadow. Mycological Research, 110, 734–748. Telford, R.J., Vandvik, V. & Birks, H.J.B. (2006) Dispersal limitations matter for microbial morphospecies. Science, 312, 1015. Tjørve, E. (2003) Shapes and functions of species–area curves: a review of possible models. Journal of Biogeography, 30, 827– 835. de Wit, R. & Bouvier, T. (2006) ‘Everything is everywhere, but, the environment selects’; what did Baas Becking and Beijerinck really say? Environmental Microbiology, 8, 755–758. Wolda, H. (1981) Similarity indices, sample size and diversity. Oecologia, 50, 296–302. Woodcock, S., Curtis, T.P., Head, I.M., Lunn, M. & Sloan, W.T. (2006) Taxa–area relationships for microbes: the unsampled and the unseen. Ecology Letters, 9, 805–812. Yang, J., Smith, H.G., Sherratt, T.N. & Wilkinson, D.M. (2010) Is there a size limit for cosmopolitan distribution in free-living microorganisms? A biogeographical analysis of testate amoebae from polar areas. Microbial Ecology, 59, 635–645. Zak, J.C. & Willig, M.R. (2004) Fungal biodiversity patterns. Biodiversity in fungi: inventory and monitoring methods (ed. by G.M. Mueller, G.F. Bills and M.S. Foster), pp. 59–75. Elsevier, Amsterdam.

SUPPO RTING INF O R MATIO N Additional Supporting Information may be found in the online version of this article: Table S1 Abundance of Phialocephala fortinii s.l.–Acephala applanata species complex (PAC) species at each of 44 study sites. Table S2 GenBank accession numbers for genotyped strains. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be reorganized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. B IO SK ETC H Our research team focuses on the ecology and evolution of fungal species and their hosts. V.Q. made most of the collections in Europe, genotyped samples with microsatellite markers, performed the statistical analysis, and prepared the manuscript; C.R.G. managed the sampling of the North American and Asian collections, genotyped a subset of the samples using single-copy RFLP, ISSR-PCR and sequence markers, and prepared the manuscript; T.N.S., B.A.M and O.H. were involved in the statistical analyses and the preparation of the manuscript. Editor: Arndt Hampe

Global Ecology and Biogeography, 20, 160–169, © 2010 Blackwell Publishing Ltd

169