Arbuscular mycorrhizal fungi associated with a single agronomic plant host across the landscape: The structure of an assemblage

Soil Biology & Biochemistry 64 (2013) 181e190

Contents lists available at SciVerse ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Arbuscular mycorrhizal fungi associated with a single agronomic plant host across the landscape: The structure of an assemblage Daniel J. Moebius-Clune, Zoe U. Anderson, Teresa E. Pawlowska* Department of Plant Pathology & Plant-Microbe Biology, Cornell University, 334 Plant Sciences Building, Ithaca, NY 14853-5904, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 June 2012 Received in revised form 23 September 2012 Accepted 25 October 2012 Available online 21 November 2012

Arbuscular mycorrhizal (AM) fungi (phylum Glomeromycota) are important components of natural and managed ecosystems. We explored the AM fungal assemblage in a selection of maize fields across a landscape in eastern New York State and characterized their diversity, dominance, and species abundance distribution. In this managed agroecosystem, we could investigate environment-influenced composition and diversity patterns unencumbered by immediate host species effects. We found that AM fungal taxon abundances were distributed lognormally, which suggests that the fungal community structure is shaped in a complex manner by many interacting niche-related factors rather than by only a single factor of disturbance associated with agricultural management. In addition to species abundance distribution, the focal assemblage shared with natural AM fungal communities a pattern of very strong dominance of certain taxa. To quantify this pattern, we developed two new indices “overdominance” and “inequitability”. Contrary to expectations based on observations of natural AM fungal communities, we found that most of the individual field communities were dominated by taxa from within a narrow phylogenetic range. At the landscape scale, we did not find an inverse relationship between the levels of taxonomic richness and phylogenetic relatedness expected in complex communities shaped by competitive exclusion. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Arbuscular mycorrhizal fungi Competitive exclusion Glomeromycota Idiosyncrasy Inequitability Overdominance Species abundance distribution

1. Introduction The arbuscular mycorrhizal (AM) fungi (phylum Glomeromycota) are important components of both natural and managed ecosystems. Symbiotic association with these fungi alters nutrient acquisition strategies and carbon budgets of plants (Smith and Read, 2008). While AM fungi make nutrients more available to plants in nutrient limiting conditions, thus buffering the impact of fluctuations in nutrient pool size, plants may also preferentially acquire nutrients through the fungal pathway even when no increase in nutrient status or absorption is observed (Smith et al., 2009). In turn, the fungi acquire substantial fractions of the plants’ photosynthate pool, up to and including amounts that may limit plant growth (Johnson et al., 1997). Association with AM fungi impacts plantepathogen interactions, reducing the severity of several root and foliar diseases, but exacerbating the effects of others (Borowicz, 2001; Hol and Cook, 2005). These fungi may mediate planteplant interactions as well, from moderating the effects of competition (Scheublin et al., 2007; van der Heijden,

* Corresponding author. Tel.: þ1 607 254 8726; fax: þ1 607 255 4471. E-mail address: [email protected] (T.E. Pawlowska). 0038-0717/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.soilbio.2012.10.043

2004) to facilitating the invasion of exotics (Carey et al., 2004). The nature and outcome of these many interactions with their host communities differs depending on the identities of the fungi involved. The AM fungi also impact the abiotic components of ecosystems, contributing to both soil aggregate stability and soil organic matter (Jastrow and Miller, 1998), and the ability of soils to provide key ecosystem services (Gianinazzi et al., 2010). While understanding about the role of AM fungi in shaping plant community composition is beginning to emerge (O’Connor et al., 2002; van der Heijden et al., 1998; van der Heijden et al., 2003), less is known of the forces that affect the diversity of AM fungal communities. In macroorganisms, insight into processes that influence the composition and diversity of an assemblage can be derived from species abundance distributions, SADs (Magurran, 2004). An SAD summarizes information on species richness (number of species) and abundance (number of individuals observed of each species) in a community (McGill et al., 2007). A collection of models is available, which describe and explain relative abundances of species in an assemblage in terms of niche occupancy. The general form of the SAD for natural AM fungal communities is uncertain, and may fit either the lognormal or broken stick model (Dumbrell et al., 2010a; Unterseher et al., 2011). The lognormal distribution of species abundance is

182

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

commonly found in large, equilibrium assemblages of macroorganisms (Magurran, 2004). A mathematical interpretation of this statistical model suggests that numerous ecological processes are responsible for determining the number of individuals per species in an assemblage (Magurran, 2004). The lognormal model is interpreted biologically to imply hierarchical partitioning, along several axes, of niche space (Sugihara, 1980) or of habitat (Kolasa and Strayer, 1988). In contrast, the broken stick distribution of species abundance suggests that niche space is partitioned by uniformly dividing a single resource-related axis (MacArthur, 1957; Sugihara, 1980). Even though the form of SAD for natural AM fungal communities is uncertain, all these communities appear to be very strongly dominated by taxa that are not together from any discernible, frequently dominant clade (Dumbrell et al., 2010a). While stochastic neutral processes, like dispersal limitation, have some role in shaping the structure of natural AM fungal communities, the primary mechanism regulating these communities is niche partitioning (Dumbrell et al., 2010b). However, in the case of AM fungi, niche space is remarkably complex. AM fungi are obligate biotrophs that depend on their plant hosts for energy. Consequently, their niche space is affected by the attributes of the plant hosts as well as by the properties of the physical environment (Dumbrell et al., 2010b). In particular, the identities of the plant species have an impact on the reproductive success of AM fungi (Bever et al., 1996), even though these fungi have historically been considered as generalists in the ability to colonize receptive hosts regardless of species (Smith and Read, 2008). One possible approach to explain relative abundances of AM fungal species in an assemblage in terms of niche apportionment is by isolating the effects of selective pressures generated by the host plant community from other pressures. Managed agroecosystems with their simplified pool of AM host plants offer a study system where this separation can be readily accomplished. Indeed, research characterizing the AM fungal communities in agricultural settings has concluded that they are very low in diversity (Daniell et al., 2001; Helgason et al., 1998), which is consistent with low complexity of niche space. However, most of these investigations were focused on AM fungal diversity in individual agronomic fields, in which niche space may be overly simplified and thus lead to underestimation of AM fungal diversity. The question of AM fungal diversity in natural and managed systems has been typically addressed by using taxon-based diversity approaches that do not consider the phylogenetic relatedness of the taxa involved. However, the recent observation that the extent of phylogenetic clustering in the AM fungal community affects species richness (Maherali and Klironomos, 2007) underscores the significance of considering phylogeny in studies of diversity. Communities that are overdispersed phylogenetically are expected to be more species-rich than communities that are phylogenetically clustered. The mechanism likely responsible for this pattern is competitive exclusion, which prevents closely related and functionally similar species from cooccurring. The goal of the present study was to elucidate the type of mechanisms, other than host species identity and host community structure, that influence the diversity of AM fungal assemblage. We addressed this goal by assessing AM fungal taxon abundance distributions in a landscape-wide assemblage of AM fungi in an agronomic system with a single plant host species (maize, Zea mays). We also examined patterns regarding: (i) low taxonomic diversity of AM fungi in agronomic systems, (ii) overdominance and idiosyncrasy in AM fungal communities, and (iii) the relationships between taxonomic and phylogenetic diversity measures.

2. Materials & methods 2.1. Sampling To assess sampling effort needed to characterize AM fungal diversity in New York State maize fields, we collected soil samples from an experimental maize field at the Willsboro Research Farm in Willsboro, NY, on a Kingsbury clay loam soil, 44 220 N; 73 230 W (Field A) in October 2008. This site is described further in Moebius et al. (2007). Twenty samples were taken at regularly spaced intervals along four transects. In the spring of 2009, additional samples were taken from seven active, conventionally managed fields, which were in maize cultivation for multiple seasons. No differentiation was made between cultivars of maize. Detailed site histories, including crops several seasons prior, or historical tillage intensity were not available. Fields were located in Flying Point, NY, on a Haven Loam, 40 530 56.000 N, 72 21034.500 W (Field B), and a Bridgehampton Silt Loam, 40 530 47.500 N, 72 21028.700 W (Field C); in Jamesport, NY, on a Riverhead Sandy Loam, 40 560 44.400 N, 72 350 47.900 W (Field D); in Queens County, NY, on a Riverhead Sandy Loam, 40 440 55.000 N, 73 430 22.700 W (Field E); in Dutchess County, NY, near Wappinger’s Falls, on Hoosic Gravelly Loam, 73 480 54.600 W (Field F), and 41380 2.700 N, 41380 7.200 N,  0 00 73 48 49.1 W (Field G); and in Colombia County, NY, near Red Hook, on a Blasdell Channery Loam, 41590 44.800 N, 73 370 8.100 W (Field H). The samples were collected using a modification of the methods used by the Cornell Soil Health Team (Gugino et al., 2009), keeping individual samples from within a field separate rather than compositing them. Ten samples from each field were taken as follows: five pin flags were placed at w20 m intervals in a “relaxed W” arrangement across the field, starting 20 m from a corner of the field, and with the angle described by the line between flags decreasing by 30 at each turn starting from a 120 angle. Two samples were taken near each pin flag, 2 m apart, along a line perpendicular to the line leading to the next pin flag. In this way, for each sample, one other sample was taken 2 m distant, and a series of distances were represented between samples in a field. Samples were collected by removing the top 2 cm of soil and excavating w4 L of maize rhizosphere soil, which was placed immediately in a zip-lock freezer bag, and cooled under ice until placed in a 4  C cooler for storage. 2.2. Trap culturing To capture the diversity of AM fungi present in maize fields, we established trap cultures for each individual field sampling point, using maize plants of cultivar Mandan Red (Seeds of Change, Rancho Dominguez, CA) to bait AM fungi and support their sporulation under shared conditions of a climate controlled greenhouse. In arable agronomic systems characterized by regular disturbance, trap culturing allows the use of spores as an integrative measure of AM fungal community composition (Oehl et al., 2003). Spores formed in trap cultures are expected to more accurately represent the AM fungal community than spores directly extracted from field samples (Johnson, 1993; Oehl et al., 2004) because the latter are often degraded or damaged (Douds and Millner, 1999) and thus unsuitable for diversity estimation. Results from trap culturing depend on the greenhouse host species used (Jansa et al., 2002). Therefore, the use of maize as a trap culture host to recover fungi that associate with maize plants in the field was expected to enable faithful reconstruction of the field AM fungal communities. The greenhouse pot cultures were established by mixing 0.5 L of field soil with 0.75 L of washed, autoclaved, pool filter sand (crushed sandstone), placing the mix in a standard 5½ inch black plastic greenhouse pot, and covering the mix surface with 0.25 L of

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

sand, for a total sand:soil ratio of 2:1. Trap cultures were planted with 3 maize seeds per pot. Prior to planting, seeds were surface disinfected with 30% H2O2 and rinsed with water. Pots were watered regularly to keep the surface moist during germination, watered intermittently approximately twice per week for 3 weeks, and then irrigated twice per day for 10 min at a rate of 2 L h1 pot1 using an automatic drip irrigation system. Plants were fertilized with a standard complete greenhouse nutrient solution as needed. After w100 days of growth, when plants showed visible signs of senescence, watering was ceased and trap cultures were left to dry on the greenhouse bench for 2 weeks. 2.3. Spore extraction from trap cultures Using a modification of standard methods (Douds and Millner, 1999; Gerdemann and Nicolson, 1963) spores were extracted from each trap culture by wet sieving and decanting of 100 mL of potting medium followed by sucrose centrifugation and collecting spores on a 47 mm diameter cellulose nitrate filter (0.45 mm). Spores were then transferred to the surface of 2% water agarose contained in a 60 mm Petri dish, by inverting the filter disk onto the agarose surface. A square piece of fiberglass window screen providing a grid with 5 units per cm was embedded in the agarose to guide spore sampling. Spores were collected under magnification by following a transect defined by a line of the mesh, and taking the spore closest to each intersection. Based on preliminary results from Field A, ten spores were sampled from each Petri dish, representing the diversity of AM fungi in each trap culture, which in turn corresponded to each individual field sample. Each spore was crushed in 2 mL of PCR H2O using fine forceps, and immediately frozen on dry ice. 2.4. Identification of AM fungi Spore genomic DNA was globally amplified using Illustra GenomiPhi v2 DNA Amplification Kit (GE Healthcare, Piscataway, NJ). The 50 end of the large subunit (LSU) rRNA gene was PCR amplified using primers LR1 and NDL22 (van Tuinen et al., 1998a; van Tuinen et al., 1998b). JumpStart REDTaq ReadyMix (Sigmae Aldrich) was used in 50 mL PCR reactions with 1 mL of GenomiPhi product as template and 10 pmol of each primer. PCR cycling conditions were: 1 min at 95  C followed by 30 cycles of 94  C for 45 s, 58  C for 45 s, and 72  C for 1 min, with a final extension at 72  C for 5 min. Amplicons were visualized by agarose gel electrophoresis, purified using QIAquick PCR Purification Kit (Qiagen, Valencia, CA), and sequenced in the forward and reverse direction using PCR primers with BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Carlsbad, CA) and the Applied Biosystems Automated 3730 DNA Analyzer. Sequences were base-called from raw data using CodonCode Aligner (CodonCode Corp., Dedham, MA) with the integrated Phred and Phrap workstation utilities. Forward and reverse reads were assembled using Geneious v5.1 (BioMatters Ltd, Auckland, NZ) after trimming off primer sequences. In the case of disagreements between reads, the base call with higher quality score was kept, and where the sum of quality scores for a particular position in the consensus sequence was less than 10, an N was called. Non-AM fungal sequences were identified by BLAST (Altschul et al., 1990) searches of GenBank (Benson et al., 2010). AM fungal rRNA gene sequences generated in this study were deposited at GenBank under accession numbers JN937121eJN937574. AM fungal operational taxonomic units (OTUs) were defined based on aligned LSU rRNA gene sequence similarity. The sequences were aligned using Muscle v3.7 (Edgar, 2004). A 95% sequence similarity level was used to cluster individual sequences

183

into OTUs in mothur (Schloss et al., 2009). This conservative cutoff has been used previously to group LSU rRNA gene sequences of AM fungi into OTUs (Li et al., 2010). To assess taxonomic affiliations of maize AM fungal OTUs, a representative sequence from each OTU, i.e. a sequence that was least different on average from all sequences comprising this OTU, was identified using mothur. These representative sequences were aligned with a set of reference taxa encompassing taxonomic diversity of Glomeromycota obtained from GenBank. Phylogenies were reconstructed with MrBayes v3.1.2 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003) and PhyML (Guindon and Gascuel, 2003) using the general time reversible (GTR) nucleotide substitution model (Tavaré, 1986) with invariable sites (I) and gamma (G) rate variation. Bayesian analyses were run for 10,000,000 generations. Statistical support for the maximum likelihood phylogeny was obtained from 1000 bootstrap replicates. Provisional taxon names were assigned to OTUs that clustered with morphologically defined reference taxa with posterior probability of 0.95 or greater, and with bootstrap support of 70% or greater. 2.5. AM fungal diversity We calculated several standard diversity indices for each field individually and for the landscape using OTUs as proxies for species. We expressed diversity as Shannon entropy H 0 (Shannon, 1948) using the natural log, and focused primarily on its expo0 nential form eH to represent the effective number of species (MacArthur, 1965), or true first order diversity (Jost, 2006), which indicates the number of species in a community with the same diversity and total abundance, but complete evenness in the distribution of abundance. To express how similar species are in their abundance, we used Heip’s index of evenness E (Heip, 1974), 0 which relates eH to the observed number of species S. We also computed Simpson’s concentration of dominance l (Simpson, 1949), which estimates the likelihood that two randomly drawn individuals are of the same species. We used EstimateS v8.2 (Colwell, 2009) for rarefaction analyses, and to calculate several estimates of true species richness, including estimation of the asymptote of the collector’s curve (Raaijmakers, 1987), along with computation of bootstrap (Smith and van Belle, 1984), Chao (Chao, 1984), and jackknife (Heltshe and Forrester, 1983) richness estimates. Rarefaction analysis of observed richness S, entropy H 0, and concentration l was used to assess sampling sufficiency for estimation of AM fungal diversity and dominance. To understand the relative contribution of within-field (alpha) and between-field (beta) diversity to the landscape-level (gamma) diversity, we partitioned gamma diversity into its alpha and beta components using the framework clarified and formalized by Jost (2006, 2007). 2.6. Taxon abundance distribution To determine the form of the taxon abundance distribution in the AM fungal landscape assemblage, the fit of a selection of taxon abundance distribution models to observed data was compared using Akaike’s information criterion (AIC) and the Bayesian information criterion (BIC), calculated with the vegan package in R (Oksanen et al., 2011; R Development Core Team, 2010). Goodnessof-fit was tested for the log series model using mothur (Schloss et al., 2009), and manually for the lognormal following Magurran (2004). We also conducted SHE analyses (Buzas and Hayek, 1996; Magurran, 2004) at the landscape and the individual field level using rarefaction estimates of observed richness S and Shannon entropy H0 generated in EstimateS, and the equitability index EBG of Buzas and Gibson (1969) and Sheldon (1969) calculated from S and H 0 .

184

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

2.7. Overdominance and inequitability To describe “overdominance”, which relates to the degree to which the most abundant taxon dominates the distribution, we developed a new measure to more consistently represent this feature. We defined “overdominance” O as the proportional abundance in excess of evenness of the top-ranked (most abundant) taxon. We calculated proportional abundance in excess of evenness, for each taxon i, as APEE ¼ ðpi  pe Þ=pe , where pi is the relative abundance of taxon i (i.e. the fraction of the total abundance across all taxa represented by the abundance of taxon i), and pe is the relative abundance of all taxa if the total abundance were distributed evenly among taxa. We also defined “inequitability” I as the proportion of taxa for which APEE is negative, i.e. where pi < pe. To compare observed O and I values with those expected for the lognormal SAD, 95% confidence intervals (CI) for expected abundance counts were obtained from Crow and Gardner (1959), treating each expected count as the mean of a Poisson variate, as in Bulmer (1974), and calculating O for the upper and lower CI bound values. An expected range was calculated for I, taking as the lower and upper ends of the range the values obtained for I where all counts are at their upper and lower CI bounds, respectively. For comparison to O, we calculated the “1:2 ratio” used previously (Dumbrell et al., 2010a; Poulin et al., 2008) to describe taxon overdominance. 2.8. Phylogenetic diversity measures In addition to taxon-based diversity measures, we computed two phylogeny-based indices, phylogenetic diversity PD (Faith, 1992) and the net relatedness index NRI (Webb, 2000). We used a maximum likelihood phylogeny of OTUs estimated by RAxML (Stamatakis et al., 2008) with GTR þ G nucleotide substitution model to calculate these indices. PD expresses the proportion of total phylogenetic tree branch length found in the individual field (Faith, 1992). We scaled this value such that the total branch length of the tree representing all sequences from the landscape was 1. As a consequence, PD for each field represents the proportion of the landscape phylogenetic breadth found in that field. NRI expresses the standardized effect size of clustering of the members of one field on a phylogenetic tree, compared to the variance in clustering expected from random community reassembly from the pooled landscape phylogeny (Webb, 2000). While PD describes the extent of phylogenetic divergence among the community taxa relative to the landscape-wide assemblage, independent of abundance, NRI reflects the overall relatedness, or degree of phylogenetic clustering, of community taxa relative to the landscape-wide assemblage. PD (Faith, 1992) and NRI (Webb, 2000) were calculated using Phylocom v4.2 (Webb et al., 2008). 3. Results 3.1. AM fungal OTU identification, taxon richness, and sampling sufficiency A total of 438 AM fungal individuals (spores) from eight maize fields across the eastern New York State (NYS) landscape, could be identified based on 28S rRNA gene sequences. These isolates grouped at a 95% sequence similarity cutoff into 33 operational taxonomic units, OTUs (Table 1). For OTUs that showed significant phylogenetic clustering with morphologically defined reference taxa, we applied the names of these morphospecies (Fig. S1, Table 1). In several cases, multiple OTUs were assigned the same morphospecies name, which is not unexpected given the phylogenetic breadth of some of the morphologically defined AM fungal taxa.

Table 1 Relative abundance and taxonomic affiliation of AM fungal OTUs in eight New York State maize fields. Taxonomic affiliation was assigned based on phylogenetic clustering with named reference taxa (Fig. S1). OTU

Rank

Abundance

Relative abundance (%)

Taxonomic affiliation Claroideoglomus etunicatum Claroideoglomus etunicatum/ claroideum/luteum lineage Funneliformis mosseae Paraglomus occultum Claroideoglomus etunicatum/ claroideum/luteum lineage Glomus trimurales Rhizophagus diaphanus Rhizophagus sp. Archaeospora schenckii Ambispora sp. Rhizophagus irregulare Rhizophagus cf. clarus Glomus viscosum Glomeromycota Funneliformis sp. Paraglomeraceae Paraglomus brasilianum Diversisporaceae Rhizophagus clarus Rhizophagus sp. Rhizophagus irregulare Rhizophagus cf. clarus Rhizophagus diaphanus Rhizophagus irregulare Rhizophagus cf. clarus Rhizophagus cf. clarus Glomus viscosum Glomeromycota Archaeospora schenckii Paraglomus occultum Paraglomus occultum Glomus microaggregatum Glomus candidum

OTU31 OTU33

1 2

156 57

35.6 13.0

OTU17 OTU25 OTU30

3 4 5

34 31 27

7.8 7.1 6.2

OTU32 OTU3 OTU9 OTU21 OTU26 OTU7 OTU11 OTU14 OTU27 OTU16 OTU19 OTU20 OTU1 OTU4 OTU5 OTU8 OTU12 OTU2 OTU6 OTU10 OTU13 OTU15 OTU18 OTU22 OTU23 OTU24 OTU28 OTU29

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

24 21 16 12 9 7 5 5 4 3 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1

5.5 4.8 3.7 2.7 2.1 1.6 1.1 1.1 0.9 0.7 0.7 0.7 0.5 0.5 0.5 0.5 0.5 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2

We used a combination of approaches to estimate true AM fungal taxon richness in our study system, and to assess the sufficiency of sampling for estimation of diversity and dominance in these communities. Based on: (i) a collector’s curve of cumulative number of OTUs over number of samples collected (Fig. S2), (ii) the MichaeliseMenten estimate (Raaijmakers, 1987) of the asymptote of a rarefaction curve of estimated OTU richness over sample size with resampling (Fig. 1A), and (iii) bootstrap estimation (Smith and van Belle, 1984), we determined that richness of the AM fungal assemblage across the landscape was likely about 38 OTUs, i.e. about 15% higher than the 33 OTUs that we observed (Text S1). We established that considering taxon abundance instead of frequency is appropriate for characterization of the AM fungal communities in these maize field sites (Text S1, Fig. S3). Furthermore, we verified that our sampling was sufficient to characterize the diversity and dominance of this assemblage (Text S1, Fig. 1B). Finally, we confirmed that none of fields appeared to exert a stronger influence than all others on the diversity related measures that we report in this paper (Text S1). 3.2. Richness and diversity of field communities and the landscape assemblage AM fungal communities of individual fields varied in richness from 3 OTUs in Field H to 12 OTUs in Field B, with a mean of 8.6 0 (Table 2). The diversity eH of the landscape level assemblage (gamma diversity) was 10.85. To understand the relative contributions of within-field (alpha) and between-field (beta) diversity to the pooled landscape (gamma) diversity, we partitioned gamma

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

Number of OTUs

A

B

3.3. Taxon abundance distribution

200 150 100 50 0

38.4

0

20

40

60

80

35 30

Index value

185

Sobs H’ e H’ λ

25 20 15 10 5 0 0

20

40

60

80

Number of samples

3.4. Overdominance and inequitability

Fig. 1. Rarefaction of taxon richness estimators and comparison with diversity indices. A. The MichaeliseMenten estimate of true richness, which stabilizes at just over 38 OTUs. B. A comparison of rarefaction curves by resampling for observed richness and 0 the key diversity indices: Shannon’s entropy H 0 , Shannon’s diversity eH , and Simpson’s concentration of dominance l.

diversity into its alpha and beta components following the frame0 work of Jost (2007). While field community diversity eH ranged 0 from 1.33 (Field H) to 8.86 (field B), landscape alpha diversity eH was 4.14, indicating that the average field contained the amount of diversity expected in a community of just over 4 evenly abun0 dant species. Landscape beta diversity eH was thus 2.62, representing the effective number of distinct communities in the region, and indicating that between-field homogeneity was M ¼ (1/beta) ¼ 0.38 (Jost, 2007; MacArthur, 1965). Simpson’s concentration of dominance l ranged from 0.14 (Field B) to 0.88 (Field H) and averaged 0.37; landscape level l was 0.17.

Table 2 Diversity indices computed for AM fungal communities of individual maize fields and across landscape. AM fungal community

S

Field A Field B Field C Field D Field E Field F Field G Field H Landscape

7 12 10 6 11 9 8 3 33

eH

0

E

l

PD

NRI

We combined several approaches to determine the form of the AM fungal taxon abundance distribution, as it is difficult to determine accurately with only one approach. We used AIC and BIC values to distinguish between the fit of several theoretical taxon abundance distribution models to the landscape-level empirical data. We found that the lognormal model fits the observed data better (AIC ¼ 131, BIC ¼ 134) than the broken stick model (AIC ¼ BIC ¼ 345), the geometric model (AIC ¼ 200, BIC ¼ 200), and the Zipf (AIC ¼ 143, BIC ¼ 146) and Zipf-Mandelbrot (AIC ¼ 136, BIC ¼ 141) models (Fig. S4). Goodness-of-fit tests rejected the log series model (P < 0.05), and confirmed the fit of the lognormal model (P >> 0.05). We also carried out SHE analysis (Buzas and Hayek, 1996) of the AM fungal assemblage across the landscape using rarefaction estimates of observed richness S and Shannon entropy H0, as well as the equitability index EBG calculated from S and H 0 . SHE analysis showed lnðSÞ and lnðH 0 Þ increasing from early to late in the range of sample size increase, while lnðEBG Þ declined and the ratio lnðEBG Þ=lnðSÞ remained flat (Fig. 2). This particular pattern is characteristic of a lognormal species abundance distribution (Magurran, 2004). The same pattern was evident in each individual field analyzed separately, with the exception of Field H, where H 0 increased only marginally, and lnðEBG Þ=lnðSÞ decreased (Fig. S5). Based on the collective results of these analyses, we conclude that the AM fungal abundance distribution in maize fields across the landscape is lognormal.

O

To quantify the strength of taxon dominance in the communities, we calculated the value ðpi  pe Þ=pe , where pi is the proportion or relative abundance of taxon i, and pe is the relative abundance of all taxa in a perfectly even community of the same richness. The value obtained for the top-ranked taxon we denoted “overdominance” O, whereas the fraction of taxa in a community for which this value is negative we denoted “inequitability” I in the sense that these taxa are “impoverished” in abundance relative to their value in a hypothetical evenly distributed community (Fig. 3). Across the landscape, the most abundant taxon Claroideoglomus etunicatum was nearly 11-fold more abundant (O ¼ 10.75) than it would be in an even community with the same richness and number of individuals, and 76% of the taxa were less abundant than

4

ln(S)

3

H’

2 1 10

I

20

30

40

50

60

70

80

0 4.23 8.86 6.67 3.56 2.89 3.31 5.53 1.33 10.85 0

0.54 0.71 0.63 0.51 0.19 0.29 0.65 0.16 0.31

0.29 0.14 0.19 0.33 0.59 0.50 0.22 0.88 0.17

0.56 0.41 0.40 0.28 0.46 0.34 0.32 0.25

2.64*** 1.96* 1.11 NS 8.19*** 3.74*** 2.02* 0.26 NS 6.14***

1.68 1.67 2.38 1.38 7.42 5.23 1.29 1.81 10.75

0.57 0.58 0.70 0.67 0.91 0.89 0.63 0.67 0.76

S, observed richness; eH , Shannon diversity; E, Heip’s evenness; l, Simpson’s dominance; PD, phylogenetic diversity; NRI, net relatedness index; O, overdominance; I, inequitability; NS, not significant; *P < 0.05; ***P < 0.001.

ln(EBG )/ln(S)

-1

ln(EBG )

-2

Number of samples Fig. 2. SHE analysis of AM fungal taxon abundance distribution in maize fields across the landscape. Resampling estimates of Shannon’s entropy H0 and log richness lnðSÞ increase across accumulation of number of samples, while log evenness lnðEBG Þ decreases, and the ratio lnðEBG Þ=lnðSÞ remains fairly stable. This is a characteristic pattern of lognormal SADs.

186

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

12 10

overdominance O

APEE

8 6 4

inequitability I

2 0 -2

5

10

15

20

25

30

35

OTU rank Fig. 3. Overdominance and inequitability in AM fungal assemblage in maize fields at the landscape scale. Proportional abundance in excess of evenness, APEE ¼ ðpi  pe Þ=pe , is plotted across OTU rank. The y value for the top-ranked OTU corresponds to overdominance O; the proportion of points with negative values corresponds to inequitability I.

(Faith, 1992), and net relatedness index NRI (Webb, 2000). The first of these, PD, was measured as the branch length found in each individual field community, expressed as a proportion of the total branch length of the pooled assemblage phylogeny. PD calculated for each field ranged from 0.25 (Field H) to 0.56 (Field A) (Table 2). PD did not rank fields in the same manner as the Shannon diversity 0 eH (Spearman r ¼ 0.33, P ¼ 0.42) or evenness E (r ¼ 0.286, P ¼ 0.49). Instead, PD correlated marginally well with taxon richness S (r ¼ 0.619, P ¼ 0.10). The second measure of phylogenetic diversity, NRI, expresses the average phylogenetic proximity of members of a community in comparison with a community of the same size randomly drawn from the pooled landscape assemblage phylogeny (Webb, 2000). NRI indicated that AM fungal communities of two fields were significantly overdispersed, i.e. members of each of these communities were less related than expected of a random sample (Field A, NRI ¼ 2.64, P < 0.001; Field B, NRI ¼ 1.96, P ¼ 0.017), whereas four fields (Fields D, E, F, and H) showed significant underdispersion (clustering) (Table 2). Furthermore, net relatedness NRI and richness S were not significantly correlated across the landscape (Spearman r ¼ 0.45, P ¼ 0.260). 4. Discussion

in such a community (I ¼ 0.76). In individual fields, the overdominance values ranged from O ¼ 7.42 in Field E, where the relative abundance of the most abundant taxon C. etunicatum was 0.77 and richness was S ¼ 11, to O ¼ 1.29 in Field G, where the relative abundance of the most abundant taxon Paraglomus occultum was 0.29 and richness was S ¼ 8 (Table 2, Fig. S6). Overall, the mean within-field overdominance value was 2.86, whereas the mean within-field inequitability was 0.70. The observed landscapelevel O ¼ 10.75 was approximately 40% greater than the O value of 7.7 that would be expected in a community with abundances distributed lognormally with the same mean and variance in abundance and the same richness, and outside of a 95% confidence interval (5.96e9.68) for such community. In contrast, the observed I ¼ 0.76 was approximately 7% greater than the expected value for a similar lognormal community (0.71), and well within the expected range (0.48e0.84).

4.1. Richness and diversity of field communities and the landscape assemblage

3.5. Identity of dominant taxa

4.2. Taxon abundance distribution

The most abundant and most frequent taxon OTU31, identified as C. etunicatum, accounted for 35.6% of the sampled individuals. The next three taxa in both abundance and frequency were OTU33 (C. etunicatum/claroideum/luteum lineage), Funneliformis mosseae, and P. occultum at 13%, 8%, and 7%, respectively (Table 1). A separate OTU found only in Field D, OTU30, was the fifth most abundant, at just over 6%, but only the ninth most frequent, occurring in only one field (Table S1). C. etunicatum was dominant in four of the five fields in which it was found (Table S1). OTU33, the second ranked taxon, was only dominant in two fields, though also found in five other fields. F. mosseae, while found in six fields, was dominant in none of them, and in five of the six fields, represented less than 10% of the field abundance. The four fields in which C. etunicatum was dominant were the four least even fields in distribution, and the three fields in which C. etunicatum was not found had higher diversity and lower concentration of dominance than the others.

Based on the information criteria, SHE analysis, and goodnessof-fit tests, we conclude that the AM fungal abundance distribution in maize fields across the landscape is lognormal, rejecting a number of alternative SAD models, including the geometric and log series models, the broken stick model, and the Zipf and Zipfe Mandelbrot models. The geometric series (niche preemption) model (Whittaker, 1965) describes a distribution that theoretically would arise if all taxa competed for a particular niche-related resource and could “preempt” a given fixed fraction of the unused amount of that resource available to them. The log series model (Fisher et al., 1943) is very similar to the geometric series model in interpretation, and both suggest that very few ecological factors influence the abundance distribution (Magurran, 2004; Zak and Willig, 2004). In the broken stick model (MacArthur, 1957), species abundances arise by competition for a simple set of resources as species divide these resources randomly (Wilson, 1991). The Zipf and ZipfeMandelbrot (Frontier, 1985, 1987) are distributional curves explained by the sequential or successional occupation of niche space by species with increasingly complex and costly requirements. Thus, like the lognormal, these two distributions arise from the effects of multiple niche-related factors, but acting sequentially, rather than simultaneously (Wilson, 1991).

3.6. Phylogenetic diversity measures We assessed two aspects of the phylogenetic diversity of the maize field AM fungal communities, the phylogenetic diversity PD

AM fungal richness and diversity in NYS maize fields were similar to those reported for a range of farming systems, including intensively managed continuously monocropped maize fields, extensive grasslands, and a vineyard in the Upper Rhine Valley in Europe (Oehl et al., 2005), and conventional and low-input maize and soybean cultivation in Pennsylvania (Franke-Snyder et al., 2001). Moreover, the NYS fields on average were higher in diversity than the European fields that led Hijri et al. (2006) to a general conclusion that AM fungal communities in agricultural fields were not necessarily low in diversity. Finally, landscape-level AM fungal richness in NYS maize fields was similar to that found recently in a grasslandewoodland transition with a wide pH range, with similar total abundance (Dumbrell et al., 2010a).

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

Our finding that the AM fungal taxon abundance distribution is lognormal is consistent with previous observations (Dumbrell et al., 2010a) that both the lognormal and broken stick distributions fit the AM fungal community across a natural grasslandewoodland ecotone in the U.K. However, since our approach permitted further discrimination between these two models, we were able to reject the broken stick model in favor of the lognormal. It is interesting to note that the patterns observed in AM fungal communities associated with natural and agronomic systems are similar despite differences between the methodologies used to study them. Dumbrell et al. (2010a) estimated abundances based on counts of cloned 18S rRNA gene sequences PCR-amplified from pooled root samples and a 97% sequence similarity level to discern taxa, whereas we counted fungal individuals directly and used a 95% 28S rRNA gene sequence similarity level to identify taxa. Agricultural management is expected to influence AM fungal community composition (Helgason et al., 1998). Indeed, the communities that we observed show a large proportional abundance of Claroideoglomus and Funneliformis, whereas other genera such as Paraglomus and Scutellospora would be more likely to be found in greater abundance in systems with a less intense disturbance regime (Douds et al., 1995; Jansa et al., 2003). Nevertheless, our observations challenge a widespread assumption that phosphorus sufficient agricultural soils (as is the case generally in NYS) exposed to repeated strong disturbance in the form of agricultural management have depauperate, simple communities of AM fungi (Daniell et al., 2001; Helgason et al., 1998). Depauperate and highly disturbance-simplified communities would be expected to exhibit a geometric abundance distribution pattern rather than the lognormal (Whittaker, 1965; Wilson, 1991). In contrast with the various rejected models, a lognormal species abundance distribution is often associated with large, stable, equilibrium communities (May, 1975; Zak and Willig, 2004) characterized by high diversity (Hughes, 1986; Whittaker, 1965) and productivity (McGill et al., 2007). However, there are substantial disagreements about the implications (Hughes, 1986), particularly in that the lognormal model is a statistical rather than biological model (Wilson, 1991), and that it may underestimate the number of low abundance species (Gray et al., 2006; Hubbell, 2001). Nevertheless, it is widely agreed that this pattern is observed where numerous niche-related environmental parameters combine to shape the form of the distribution (Magurran, 2004). Our focal agricultural community on the landscape scale (and based on the SHE analysis also on the field scale) shares the fundamental characteristic of taxon abundance distribution with the communities found across a diversity of non-agricultural sites rather than standing as a counterexample. Consequently, we conclude that the drivers of AM fungal community composition in maize fields are multifold and cannot be reduced solely to “agricultural management” as a single force. While beyond the scope of the present theoretical assessment, further work combining soil physical, chemical, and biological data from multiple research groups’ assessment of these sites identified several of these factors, including soil texture and available water capacity as important for AM fungal community structure (Moebius-Clune et al., 2012). 4.3. Overdominance and inequitability To quantify the very strong dominance of certain taxa apparent in AM fungal communities of maize fields, we developed two new indices, overdominance, O, and inequitability, I. Overdominance O was not significantly correlated with the widely used Simpson’s measure of dominance l (P ¼ 0.732), indicating that while both indices relate to the dominance structure of a community, they reflect substantially different features of it. Furthermore, Simpson’s

187

l did not seem to adequately represent the degree to which the

most abundant AM fungal species dominated in the NYS fields. Simpson’s l appeared quite sensitive to the evenness of the nondominant members of these communities. For example, in Fields B and C, the O value was roughly the same, and the number of species with lower than even abundances was the same, but l was twice as high in the field with lower evenness (Table 2). While l represents the probability that any two randomly chosen individuals are of the same species, the overdominance O more directly represents the degree to which the most abundant species, in particular, dominates the community abundance. On the landscape level, l was lower than the lowest individual field’s l value. This arises from the substantially higher richness at the landscape level, and the presence of numerous low abundance taxa. Nonetheless, the most abundant taxon was remarkably more overdominant at the landscape level than in any individual field community, alone representing more than a third of the abundance, giving an O value of 10.75, while l was 0.17. Taxon overdominance has been noted by Dumbrell et al. (2010a) as a peculiarity of AM fungal communities. However, these authors quantified it using the simple dominance index of Poulin et al. (2008), i.e. the ratio of abundances of the most abundant to the second-most abundant taxon, “1:2 ratio”. Our observations in NYS maize fields are consistent with these previous observations of overdominance in other AM fungal communities (Dumbrell et al., 2010a). In the NYS agroecosystem assemblage, the most abundant taxon was about 2.4 times as abundant as the second most abundant taxon, and within individual fields, the 1:2 ratio ranged from 1 to 29. The average 1:2 ratio of 7.6 was about twice that previously reported (Dumbrell et al., 2010a) from a set of communities whose most abundant taxa were on average 40% of the total abundance (mean of 3.5). Dumbrell et al. (2010a) compared the values of 1:2 ratios from AM fungal communities with other microbial communities and concluded that they were uncommonly high. However, the 1:2 ratio cannot relate the strength of dominance of the top taxon in relation to the whole community. For example the 1:2 ratio achieves its lowest possible value of 1 when two codominant taxa are present, even if together they are many times as abundant as the remainder of taxa. 4.4. Identity of dominant taxa The pattern of identities of the dominant taxon in individual NYS maize fields suggests an aspect of dominance in AM fungal communities that is inconsistent with the idiosyncrasy feature discussed by Dumbrell et al. (2010a). The meta-analysis conducted by these authors suggested that across many environmental types, it is not always the same or related species of AM fungi that are dominant. In other words, taxon dominance does not depend on its identity. However, in our study system, it appears that, while the landscape scale community is characterized by the overdominance structure, the degree to which different species can dominate within these highly similar environments may indeed be dependent on the identity of the dominant. OTU31 (C. etunicatum) was dominant in most of the fields in which we found it, and the evenness and diversity of these fields was lower than the others. Where OTU31 was dominant but evenness was not markedly low (Field D), it was the closely related OTU30 that together with it made up nearly 80% of the field abundance. The most diverse and most even fields were those that had no C. etunicatum present. Consequently, it appears that the degree of overdominance in our study system does depend on which species is dominant and may be due to differences in competitive ability, with C. etunicatum being particularly competitive in the maize fields. Perhaps within a number of sample sets from many dissimilar environments no

188

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

single group of species can be seen most frequently dominant but within particular environment types some species are better adapted and perhaps more competitive than others. 4.5. Phylogenetic diversity measures Of the two phylogenetic diversity measures that we computed, the net relatedness index NRI (Webb, 2000) proved more informative for assessment of phylogenetic community structure than phylogenetic diversity PD (Faith, 1992), as NRI is more sensitive to evenness of abundance, in addition to phylogenetic breadth or richness. Consequently, we used NRI to explore the interactions between taxon richness and phylogenetic dispersion of maize field AM fungal communities. Maherali and Klironomos (2007) argued that due to trait conservation within lineages and competitive exclusion, higher species richness should be expected in communities characterized by phylogenetic overdispersion. Several observations about our field communities are inconsistent with these predictions. Foremost, there is no significant relationship between NRI and taxon richness as a whole. Moreover, the field with the highest NRI (Field D) was not the least rich, but rather had twice the richness of the least rich field (Field H). Likewise, the field with the lowest relatedness (Field A) was not the richest, but rather the sixth richest. Additionally, the second richest community (Field E) was actually significantly underdispersed. Overall, in the assemblage of maize field communities, there is not a clear pattern of the more taxonomically rich communities being those with lower net phylogenetic relatedness. This contrast between our observations and those of Maherali and Klironomos (2007) may indicate that competitive exclusion is not a driving force shaping the phylogenetic diversity of the maize field AM fungal communities. Competitive exclusion likely is a strong shaping force in natural AM fungal communities (Maherali and Klironomos, 2007), leading to divergence of AM fungal functional trait complexes between forms or species. Our observations of a lack of correlation between NRI and S, and the presence of taxon-rich and phylogenetically underdispersed communities in individual fields suggest that compositional patterns in these fields may instead be a product of habitat filtering, whereby constraints imposed by environmental factors limit the possible phylogenetic range of the community, and thus allow closely related taxa with key adaptive functional traits to coexist. 4.6. Methodological considerations In this study, we used trap culturing and spore isolation methods to characterize field communities of AM fungi. Trap culturing induces the formation of mycorrhizal association and subsequently the generation of spores in conditions that approximate those encountered in the field. This may lead to biases in the assessment of field community composition as some species of fungi may be more favored by trap culture conditions than others are. We have attempted to minimize these biases by establishing trap cultures with the same host species as in the field, and retaining a large proportion of field sampled whole inoculum in the trap culture pots. Despite the potentially homogenizing effect of trap culturing, we nevertheless observed substantial differentiation between individual field communities. We further expect the differences between the field communities and the trap culture communities to be smaller than those that could be anticipated if we sampled undisturbed systems (Oehl et al., 2003). We described the assemblage of AM fungi by sequence-based identification and enumeration of their spores. Identifying and counting spores is the method usually used to characterize AM fungal assemblages in field situations (Smith and Read, 2008). By

focusing on spores rather than on hyphae in soil or intraradical structures, we were able to assess AM fungal diversity based on counts of discrete and differentiable individuals (Franke-Snyder et al., 2001; Rosendahl, 2008). However, sampling spores rather than intraradical structures will unavoidably omit AM fungal species that do not produce spores. Inconsistencies between spore populations and root-extracted rRNA gene sequence populations have been reported (Hijri et al., 2006; Schreiner and Mihara, 2009), with spores identified in soil samples that do not appear to be, at that time, colonizing roots taken from the same location, and the reverse. Nevertheless, the spore bank represents both an accumulated history and the potential inoculum pool of fungal partners likely to associate with plants introduced to the field, and overall is more informative than a survey of the fungi colonizing a particular plant at a particular moment in time (Hijri et al., 2006; Oehl et al., 2004). Furthermore, identifying intraradical hyphae in field sampled roots by massively parallel sequencing or cloning and sequencing of PCR-amplified rRNA gene fragments would introduce a set of biases associated with DNA extraction, amplification, and cloning (e.g. Jumpponen and Johnson, 2005; Schreiner and Mihara, 2009). By identifying spores based on rRNA gene sequences instead of morphology, we were able to approach diversity at a common level of sequence divergence, rather than using species delineations that are based only on visible aspects of morphology. Morphologybased identification is believed to substantially underestimate the actual diversity of AM fungal communities (Redecker and Raab, 2006). With sequence-based AM fungal identification we were also able to consider phylogenetic diversity of the AM fungal communities beyond taxon-based diversity measures, which do not account for the phylogenetic relatedness of the taxa involved. 5. Conclusion Overall, we found that the abundance of AM fungal taxa associated with a single agronomic host species across the landscape was distributed lognormally, which suggests that the fungal community structure was shaped in a complex manner by numerous environmental factors rather than only by a single factor of disturbance associated with agricultural management. The focal assemblage was more diverse at the landscape level than at any individual field level and resembled natural AM fungal communities in species abundance distribution and dominance structure. Observed taxon overdominance was of the order reported previously for AM fungi in natural communities. We suggested a refinement of the indices used to represent this pattern. Contrary to expectations from natural systems, we did not find evidence that each field was dominated by a different AM fungal species. Furthermore, the expectation of an inverse relationship between taxonomic richness and phylogenetic relatedness was not borne out. Instead, some of individual field communities that were taxonomically diverse were also phylogenetically clustered, which is consistent with habitat acting as a filter that limits the phylogenetic range of the community. While the community that we describe exhibits characteristics expected of natural AM fungal communities, such as lognormal species abundance distribution and an overdominance structure, it differs from natural systems in the pattern of identities of dominant taxa and in lacking features expected from the influence of competitive exclusion. Acknowledgments We thank B. Moebius-Clune for help with identifying sampling sites, J. Morton and D. van Tuinen for permission to use their unpublished LSU rRNA gene sequences deposited at GenBank, T.

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

Horton for discussions, S. Mondo and K. Toomer for comments on the manuscript. This research was supported by the Cornell University Agricultural Experiment Station federal formula funds, Project NYC153441, received from the National Institutes for Food and Agriculture, U.S. Department of Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2012.10.043. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic Local Alignment Search Tool. Journal of Molecular Biology 215, 403e410. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., Sayers, E.W., 2010. GenBank. Nucleic Acids Research 38, D46eD51. Bever, J.D., Morton, J.B., Antonovics, J., Schultz, P.A., 1996. Host-dependent sporulation and species diversity of arbuscular mycorrhizal fungi in a mown grassland. Journal of Ecology 84, 71e82. Borowicz, V.A., 2001. Do arbuscular mycorrhizal fungi alter plant-pathogen relations? Ecology 82, 3057e3068. Bulmer, M.G., 1974. On fitting the Poisson lognormal distribution to speciesabundance data. Biometrics 30, 101e110. Buzas, M., Hayek, L., 1996. Biodiversity resolution: an integrated approach. Biodiversity Letters 3, 40e43. Buzas, M.A., Gibson, T.G., 1969. Species diversity e benthonic foraminifera in Western North Atlantic. Science 163, 72e75. Carey, E.V., Marler, M.J., Callaway, R.M., 2004. Mycorrhizae transfer carbon from a native grass to an invasive weed: evidence from stable isotopes and physiology. Plant Ecology 172, 133e141. Chao, A., 1984. Nonparametric estimation of the number of classes in a population. Scandinavian Journal of Statistics 11, 265e270. Colwell, R.K., 2009. EstimateS: Statistical Estimation of Species Richness and Shared Species from Samples. Version 8.2. http://viceroy.eeb.uconn.edu/estimates. Crow, E.L., Gardner, R.S., 1959. Confidence intervals for the expectation of a Poisson variable. Biometrika 46, 441e453. Daniell, T.J., Husband, R., Fitter, A.H., Young, J.P.W., 2001. Molecular diversity of arbuscular mycorrhizal fungi colonising arable crops. FEMS Microbiology Ecology 36, 203e209. Douds, D.D., Galvez, L., Janke, R.R., Wagoner, P., 1995. Effect of tillage and farming system upon populations and distribution of vesicular-arbuscular mycorrhizal fungi. Agriculture Ecosystems & Environment 52, 111e118. Douds, D.D., Millner, P., 1999. Biodiversity of arbuscular mycorrhizal fungi in agroecosystems. Agriculture Ecosystems & Environment 74, 77e93. Dumbrell, A.J., Nelson, M., Helgason, T., Dytham, C., Fitter, A.H., 2010a. Idiosyncrasy and overdominance in the structure of natural communities of arbuscular mycorrhizal fungi: is there a role for stochastic processes? Journal of Ecology 98, 419e428. Dumbrell, A.J., Nelson, M., Helgason, T., Dytham, C., Fitter, A.H., 2010b. Relative roles of niche and neutral processes in structuring a soil microbial community. ISME Journal 4, 337e345. Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32, 1792e1797. Faith, D.P., 1992. Conservation evaluation and phylogenetic diversity. Biological Conservation 61, 1e10. Fisher, R.A., Corbet, A.S., Williams, C.B., 1943. The relation between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology 12, 42e58. Franke-Snyder, M., Douds, D.D., Galvez, L., Phillips, J.G., Wagoner, P., Drinkwater, L., Morton, J.B., 2001. Diversity of communities of arbuscular mycorrhizal (AM) fungi present in conventional versus low-input agricultural sites in eastern Pennsylvania, USA. Applied Soil Ecology 16, 35e48. Frontier, S., 1985. Diversity and structure in aquatic ecosystems. Oceanography and Marine Biology e An Annual Review 23, 253e312. Frontier, S., 1987. Applications of fractal theory to ecology. In: Legendre, P. (Ed.), Developments in Numerical Ecology. Springer-Verlag, Berlin, pp. 357e378. Gerdemann, J.W., Nicolson, T.H., 1963. Spores of mycorrhizal Endogone species extracted from soil by wet sieving and decanting. Transactions of the British Mycological Society 46, 235e244. Gianinazzi, S., Gollotte, A., Binet, M.N., van Tuinen, D., Redecker, D., Wipf, D., 2010. Agroecology: the key role of arbuscular mycorrhizas in ecosystem services. Mycorrhiza 20, 519e530. Gray, J.S., Bjorgesaeter, A., Ugland, K.I., 2006. On plotting species abundance distributions. Journal of Animal Ecology 75, 752e756.

189

Gugino, B.K., Idowu, O.J., Schindelbeck, R.R., van Es, H.M., Wolfe, D.W., MoebiusClune, B.N., Thies, J.E., Abawi, G.S., 2009. Cornell Soil Health Assessment Training Manual, second ed. Cornell University, Geneva, NY. Guindon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52, 696e704. Heip, C., 1974. A new index measuring evenness. Journal of the Marine Biological Association of the United Kingdom 54, 555e557. Helgason, T., Daniell, T.J., Husband, R., Fitter, A.H., Young, J.P.W., 1998. Ploughing up the wood-wide web? Nature 394, 431. Heltshe, J.F., Forrester, N.E., 1983. Estimating species richness using the jackknife procedure. Biometrics 39, 1e11. Hijri, I., Sykorova, Z., Oehl, F., Ineichen, K., Mader, P., Wiemken, A., Redecker, D., 2006. Communities of arbuscular mycorrhizal fungi in arable soils are not necessarily low in diversity. Molecular Ecology 15, 2277e2289. Hol, W.H.G., Cook, R., 2005. An overview of arbuscular mycorrhizal fungi-nematode interactions. Basic and Applied Ecology 6, 489e503. Hubbell, S.P., 2001. The Unified Theory of Biodiversity and Biogeography. Princeton University Press, Princeton, NJ. Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754e755. Hughes, R.G., 1986. Theories and models of species abundance. The American Naturalist 128, 879e899. Jansa, J., Mozafar, A., Anken, T., Ruh, R., Sanders, I.R., Frossard, E., 2002. Diversity and structure of AMF communities as affected by tillage in a temperate soil. Mycorrhiza 12, 225e234. Jansa, J., Mozafar, A., Kuhn, G., Anken, T., Ruh, R., Sanders, I.R., Frossard, E., 2003. Soil tillage affects the community structure of mycorrhizal fungi in maize roots. Ecological Applications 13, 1164e1176. Jastrow, J.D., Miller, R.M., 1998. Soil aggregate stabilization and carbon sequestration: feedbacks through organomineral associations. In: Lal, R., Kimble, J.M., Follett, R.F., Stewart, B.A. (Eds.), Soil Processes and the Carbon Cycle. CRC Press LLC, Boca Raton, FL, pp. 207e223. Johnson, N.C., 1993. Can fertilization of soil select less mutualistic mycorrhizae? Ecological Applications 3, 749e757. Johnson, N.C., Graham, J.H., Smith, F.A., 1997. Functioning of mycorrhizal associations along the mutualism-parasitism continuum. New Phytologist 135, 575e 586. Jost, L., 2006. Entropy and diversity. Oikos 113, 363e375. Jost, L., 2007. Partitioning diversity into independent alpha and beta components. Ecology 88, 2427e2439. Jumpponen, A., Johnson, L.C., 2005. Can rDNA analyses of diverse fungal communities in soil and roots detect effects of environmental manipulations - a case study from tallgrass prairie. Mycologia 97, 1177e1194. Kolasa, J., Strayer, D., 1988. Patterns of the abundance of species: a comparison of two hierarchical models. Oikos 53, 235e241. Li, L.F., Li, T., Zhang, Y., Zhao, Z.W., 2010. Molecular diversity of arbuscular mycorrhizal fungi and their distribution patterns related to host-plants and habitats in a hot and arid ecosystem, southwest China. FEMS Microbiology Ecology 71, 418e427. MacArthur, R.H., 1957. On the relative abundance of bird species. Proceedings of the National Academy of Sciences of the United States of America 43, 293e295. MacArthur, R.H., 1965. Patterns of species diversity. Biological Reviews 40, 510e533. Magurran, A., 2004. Measuring Biological Diversity. Blackwell Science Ltd, Malden, MA. Maherali, H., Klironomos, J.N., 2007. Influence of phylogeny on fungal community assembly and ecosystem functioning. Science 316, 1746e1748. May, R.M., 1975. Patterns of species abundances and diversity. In: Cody, M.L., Diamond, J.M. (Eds.), Ecology and Evolution of Communities. Harvard University Press, Cambridge, MA, pp. 81e120. 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, 995e1015. Moebius, B.N., van Es, H.M., Schindelbeck, R.R., Idowu, O.J., Clune, D.J., Thies, J.E., 2007. Evaluation of laboratory-measured soil properties as indicators of soil physical quality. Soil Science 172, 895e912. Moebius-Clune, D.J., Moebius, B.N., Van Es, H.M., Pawlowska, T.E., 2012. Arbuscular mycorrhizal fungi associated with a single agronomic plant host across the landscape: community differentiation along a soil textural gradient. in review. O’Connor, P.J., Smith, S.E., Smith, F.A., 2002. Arbuscular mycorrhizas influence plant diversity and community structure in a semiarid herbland. New Phytologist 154, 209e218. Oehl, F., Sieverding, E., Ineichen, K., Mäder, P., Boller, T., Wiemken, A., 2003. Impact of land use intensity on the species diversity of arbuscular mycorrhizal fungi in agroecosystems of Central Europe. Applied and Environmental Microbiology 69, 2816e2824. Oehl, F., Sieverding, E., Ineichen, K., Ris, E.A., Boller, T., Wiemken, A., 2005. Community structure of arbuscular mycorrhizal fungi at different soil depths in extensively and intensively managed agroecosystems. New Phytologist 165, 273e283. Oehl, F., Sieverding, E., Mader, P., Dubois, D., Ineichen, K., Boller, T., Wiemken, A., 2004. Impact of long-term conventional and organic farming on the diversity of arbuscular mycorrhizal fungi. Oecologia 138, 574e583.

190

D.J. Moebius-Clune et al. / Soil Biology & Biochemistry 64 (2013) 181e190

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., O’Hara, R.B., Simpson, G.L., Solymos, P., Henry, M., Stevens, H., Wagner, H., 2011. vegan: Community Ecology Package. R package version 1.17-6. http://CRAN.R-project.org/package¼vegan. Poulin, R., Luque, J.L., Guilhaumon, F., Mouillot, D., 2008. Species abundance distributions and numerical dominance in gastrointestinal helminth communities of fish hosts. Journal of Helminthology 82, 193e202. R Development Core Team, 2010. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Raaijmakers, J.G.W., 1987. Statistical analysis of the Michaelis-Menten equation. Biometrics 43, 793e803. Redecker, D., Raab, P., 2006. Phylogeny of the Glomeromycota (arbuscular mycorrhizal fungi): recent developments and new gene markers. Mycologia 98, 885e 895. Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572e1574. Rosendahl, S., 2008. Communities, populations and individuals of arbuscular mycorrhizal fungi. New Phytologist 178, 253e266. Scheublin, T.R., Van Logtestijn, R.S.P., Van der Heijden, M.G.A., 2007. Presence and identity of arbuscular mycorrhizal fungi influence competitive interactions between plant species. Journal of Ecology 95, 631e638. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., Weber, C.F., 2009. Introducing mothur: opensource, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75, 7537e7541. Schreiner, R.P., Mihara, K.L., 2009. The diversity of arbuscular mycorrhizal fungi amplified from grapevine roots (Vitis vinifera L.) in Oregon vineyards is seasonally stable and influenced by soil and vine age. Mycologia 101, 599e611. Shannon, C.E., 1948. A mathematical theory of communication. Bell System Technical Journal 27, 379e423. 623e656. Sheldon, A.L., 1969. Equitability indices e dependence on species count. Ecology 50, 466e467. Simpson, E.H., 1949. Measurement of diversity. Nature 163, 688. Smith, E.P., van Belle, G., 1984. Nonparametric estimation of species richness. Biometrics 40, 119e129. Smith, F.A., Grace, E.J., Smith, S.E., 2009. More than a carbon economy: nutrient trade and ecological sustainability in facultative arbuscular mycorrhizal symbioses. New Phytologist 182, 347e358. Smith, S.E., Read, D.J., 2008. Mycorrhizal Symbiosis. Academic Press.

Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for the RAxML web servers. Systematic Biology 57, 758e771. Sugihara, G., 1980. Minimal community structure e an explanation of species abundance patterns. American Naturalist 116, 770e787. Tavaré, S., 1986. Some Probabilistic and Statistical Problems in the Analysis of DNA Sequences, American Mathematical Society: Lectures on Mathematics in the Life Sciences. American Mathematical Society, pp. 57e86. Unterseher, M., Jumpponen, A., Opik, M., Tedersoo, L., Moora, M., Dormann, C.F., Schnittler, M., 2011. Species abundance distributions and richness estimations in fungal metagenomics e lessons learned from community ecology. Molecular Ecology 20, 275e285. van der Heijden, M.G.A., 2004. Arbuscular mycorrhizal fungi as support systems for seedling establishment in grassland. Ecology Letters 7, 293e303. van der Heijden, M.G.A., Klironomos, J.N., Ursic, M., Moutoglis, P., StreitwolfEngel, R., Boller, T., Wiemken, A., Sanders, I.R., 1998. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 69e72. van der Heijden, M.G.A., Wiemken, A., Sanders, I.R., 2003. Different arbuscular mycorrhizal fungi alter coexistence and resource distribution between cooccurring plant. New Phytologist 157, 569e578. van Tuinen, D., Jacquot, E., Zhao, B., Gollotte, A., Gianinazzi-Pearson, V., 1998a. Characterization of root colonization profiles by a microcosm community of arbuscular mycorrhizal fungi using 25S rDNA-targeted nested PCR. Molecular Ecology 7, 879e887. van Tuinen, D., Zhao, B., Gianinazzi-Pearson, V., 1998b. PCR in studies of AM fungi: from primers to application. In: Varma, A.K. (Ed.), Mycorrhiza Manual. SpringerVerlag, Heidelberg, pp. 387e399. Webb, C.O., 2000. Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. American Naturalist 156, 145e155. Webb, C.O., Ackerly, D.D., Kembel, S.W., 2008. Phylocom: software for the analysis of phylogenetic community structure and trait evolution. Bioinformatics 24, 2098e2100. Whittaker, R.H., 1965. Dominance and diversity in land plant communities e numerical relations of species express importance of competition in community function and evolution. Science 147, 250e260. Wilson, J.B., 1991. Methods for fitting dominance diversity curves. Journal of Vegetation Science 2, 35e46. Zak, J.C., Willig, M.R., 2004. Fungal biodiversity patterns. In: Mueller, G.M., Bills, G.F., Foster, M.S. (Eds.), Biodiversity of Fungi: Inventory and Monitoring Methods. Elsevier Academic Press, Boston, MA, pp. 59e75.