Fungal community composition and diversity vary with soil depth and landscape position in a no-till wheat-based cropping system

Fungal community composition and diversity vary with soil depth and landscape position in a... ABSTRACT Soil edaphic characteristics are major drivers of fungal communities, but there is a lack of information on how communities vary with soil depth and landscape position in no-till cropping systems. Eastern Washington is dominated by dryland wheat grown on a highly variable landscape with steep, rolling hills. High-throughput sequencing of fungal ITS1 amplicons was used to characterize fungal communities across soil depth profiles (0 to 100 cm from the soil surface) among distinct landscape positions and aspects across a no-till wheat field. Fungal communities were highly stratified with soil depth, where deeper depths harbored distinct fungal taxa and more variable, less diverse fungal communities. Fungal communities from deep soils harbored a greater portion of taxa inferred to have pathotrophic or symbiotrophic in addition to saprotrophic lifestyles. Co-occurrence networks of fungal taxa became smaller and denser as soil depth increased. In contrast, differences between fungal communities from north-facing and south-facing slopes were relatively minor, suggesting that plant host, tillage, and fertilizer may be stronger drivers of fungal communities than landscape position. fungal community, soil depth, aspect, slope, landscape position, no-till, wheat INTRODUCTION Fungi are critical components of agricultural systems where they play roles in carbon cycling, nutrient cycling, decomposition, and plant health (Raaijmakerset al. 2009). Fungal communities are highly diverse assemblages with an estimated 1.5 million species worldwide and hundreds of species within a single gram of soil (Hawksworth 2001; Taylor et al.2014; Peay, Kennedy and Talbot 2016). At the global scale, climatic conditions, soil characteristics and spatial patterns are believed to be the primary drivers of soil fungal communities (Tedersoo et al. 2014). The immense diversity of fungal communities across the landscape, however, is due in part to their extensive variability at small spatial scales, where soil edaphic parameters (eg. water, temperature, C, N), plant species identity, and plant diversity (LeBlanc, Kinkel and Kistler 2015, 2017; Peay, Kennedy and Talbot 2016; Zhanget al. 2017) drive variation in fungal community composition. Yet, with some exceptions (eg. Degrune et al. 2016; Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010), the vast majority of work examining spatial variation in soil fungal communities has focused only on communities found within the top 10–20 cm of soil. Thus, we have only scratched the surface of our understanding of how fungal communities vary within the soil profile. Fungi residing in deeper layers of the soil profile contribute to key processes, such as soil formation, carbon and nutrient cycling, and xenobiotic degradation, and form intimate associations with plant roots where they can enhance water and nutrient uptake, provide protection from pathogens, or be pathogens themselves (Swift, Heal and Anderson 1979; Schulz et al. 2013). Soil profiles are characterized by vertical gradients in carbon, nitrogen, pH and oxygen that frequently correspond with declines in microbial biomass and diversity (Akob and Kusel 2011; Ekelund, Rønn and Christensen 2001; Fierer, Schimel and Holden 2003). Further, deeper soil layers experience less extreme variation in moisture and temperature than surface soils (Kristensen 1959). Although it is recognized that fungal communities are likely to vary throughout the soil profile (Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010; Baldrian et al.2012; Peay, Kennedy and Talbot 2016; Toju et al. 2016; Lamit et al.2017), we have a poor understanding of the composition of deep-soil communities in agricultural systems and their diversity across the landscape. The Palouse region of eastern Washington and northern Idaho is characterized by steeply sloped, rolling hills, and although originally dominated by forest and prairie ecosystems, dryland wheat cropping has predominated in this area for the last century (Schillinger and Papendick 2008). Soil water is often limiting to crop production, since most precipitation occurs over the winter months (November to May). Crucially, the highly dynamic landscape topography creates a mosaic of microclimates across agricultural fields, especially where water, temperature, and plant productivity, vary considerably with the slope and aspect (Mahler and Bezdicek 1978). Specifically, north-facing slopes typically accumulate more snow over the winter, retain more water over the summer, and can be more productive than south-facing slopes. As such, it is likely that fungal communities are considerably impacted by the aspect of the landscape and plant productivity. Further, as a result of the landscape characteristics in the Palouse region, soil erosion due to wind, water, and mechanical tillage of soil is a major problem in this area, with an estimated average soil lost to erosion of 1 to 50 Mg ha–1 (Nagle and Ritchie 2004; Sharratt, Wendling and Feng 2012). To combat erosion, conserve soil water, increase soil organic matter, and save on fuel costs, growers are increasingly adopting no-till or direct-seed cropping practices (Huggins and Reganold 2008). In no-till, as the name implies, tillage is reduced or eliminated and crops are seeded directly into the previous crops’ stubble. Notably, there is little disturbance of the soil and crop residue is left on the soil surface rather than incorporated during tillage. Substantial differences in soil fungal communities exist between conventional tillage and no-till systems (Sharma-Poudyal et al.2017), yet little is known about how the spatial distribution of fungal communities is influenced by aspect, or how communities are stratified with soil depth in no-till fields across this highly-variable landscape. Additionally, unique characteristics of Palouse soils, which consist primarily of wind-deposited loess formed over 10000 years, may contribute to vertical stratification of fungal communities. These soils are very deep, with topsoil depths of 1.5–1.8 m or more (Horner, McCall and Bell 1948; Hipple 2011). Wheat roots can grow down to 3 m and could be a major source of carbon at these depths, but 50% of the roots are in the top 16 cm (Fanet al. 2016). Fungal interactions with soil carbon and wheat roots at different soil depths are likely to favor fungal taxa with different life-history strategies. For example, an abundance of carbon and plant roots in upper soil layers may generate a diverse community of saprotrophs and symbiotrophs, whereas deep soils, which are relatively depauperate in carbon, may select for a distinct community of pathotrophs and symbiotrophs. Furthermore, differences in fungal community composition and diversity among soil depths is likely to be reflected in unique species interactions and patterns in fungal networks. For example, AMF communities were shown to vary with soil depth (Sosa-Hernández et al. 2018). Powell et al. (2015) observed that fungi in the upper soil layer exhibit stochastic processes, possibly due to strong species interactions. Here, we use high-throughput DNA sequencing to explore the composition and spatial distribution of fungal communities at different depths (0 to 100 cm) across the landscape in a no-till system. Specifically, we hypothesize that (1) fungal community composition and diversity will differ across the soil depth profile; (2) soils on north-facing aspects will harbor distinct fungal communities from those of south-facing aspects; (3) fungi with saprotrophic lifestyles will be most abundant near the soil surface, while symbiotic and pathogenic modes will predominate at deeper depths, reflecting their nutritional preferences and (4) fungal co-occurrence patterns will be more complex near the soil surface than at deeper soil depths, reflecting a greater importance of species interactions. MATERIALS AND METHODS Site description and soil sampling Soil samples were collected from the Washington State University Cook Agronomy Farm, a USDA Long-Term Agroecosystem Research (LTAR) site in eastern Washington state, USA, located ∼7 miles northeast of Pullman, WA. Soils were sampled in April, 2016 from a transect spanning a range of slopes and aspects within a winter wheat-rotation (Fig. 1). This strip has been in a continuous cereal rotation for the 9 years previous to the sampling. This rotation was spring barley-spring wheat-winter wheat. Winter wheat was the last crop planted prior to sampling. On April 7, 2016, two soil cores were obtained from each of eight points using a tractor-mounted hydraulic Giddings soil probe (Windsor, CO). Briefly, the Giddings probe was fitted with plastic liners (PET-G, polyethylene terephthalate glycol-modified, Giddings Machine Co. Inc.) and 4.4 cm inside diameter cores were taken at each point to a depth of at least 100 cm. Cores were immediately capped, transported to the lab, and frozen at −20 C until processing as described below. The Cook Agronomy Farm has been studied extensively, and soil edaphic, environmental, and crop yield data exist for each sampling point (Table 1; Huggins and Uberuaga 2010). Figure 1. View largeDownload slide Landscape positions of sampling points at the Cook Agronomy Farm. Points 8B, 15K, and 16M are south-facing. Points 10D, 12G, and 18Q are north-facing. Points 9C and 12H are top-slope (summit) and bottom-slope (toe slope) cores, respectively. Contour interval = 2 m. Figure 1. View largeDownload slide Landscape positions of sampling points at the Cook Agronomy Farm. Points 8B, 15K, and 16M are south-facing. Points 10D, 12G, and 18Q are north-facing. Points 9C and 12H are top-slope (summit) and bottom-slope (toe slope) cores, respectively. Contour interval = 2 m. Table 1. General characteristics of sampling points. Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 a Yields were normalized to average yields (Blackmore et al.2003). View Large Table 1. General characteristics of sampling points. Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 a Yields were normalized to average yields (Blackmore et al.2003). View Large DNA extraction, PCR and Illumina sequencing Holes (∼2 cm diameter) were bored into the plastic sheaths of each soil frozen soil core at 10 cm, 25 cm, 50 cm, 75 cm and 100 cm from the soil surface with a drill and sealed with packing tape. Information on the soil horizons at each depth is provided in Supplemental Table S1. This includes the metadata such as EC, C, N, pH and bulk density, referenced to the points in Fig. 1. To obtain soil samples, cores were thawed individually on the benchtop for ∼30 min and ∼1 g of soil was taken from each bore hole and the surface (0 cm depth) of each core with a sterile spatula after scraping away the outer layer of soil. DNA was extracted from each sample (∼0.25 g) using the PowerSoil DNA extraction kit (MoBio/Qiagen) according to the manufacturers’ instructions using the ‘soil’ program on a FastPrep 5G homogenizer (MP Biomedicals) for the bead-beating step. A soil-free DNA extraction preparation was included as a negative control. DNA was quantified and checked for quality on a NanoDrop spectrophotometer (FischerThermo). The fungal ITS1 region was amplified with ITS1-F and ITS2 primers adapted to include linkers, barcoding tags, and additional bases to introduce a phase offset for Illumina sequencing (Supplemental Table S2). Primers were combined in equimolar ratio prior to PCR amplification. PCR reactions (20 µl total volume) consisted of 10 µl FastStart Master Mix (Roche), 0.7 µl of primers (10 µM), 8.3 µl PCR-grade H2O and 1 µl of template DNA (1/10 diluted). Reactions were performed in triplicate with three separate annealing temperatures (50°C, 53°C or 55°C). Thermocycling conditions consisted of an initial denaturation at 95°C for 4 min, 30 cycles of 95°C for 30 s, annealing at 50, 53 or 55°C for 30 s, and extension at 72°C for 45 s, with a final extension at 72°C for 7 min. PCR products were pooled and checked for amplification on a 1.5% gel. Illumina adapters and barcodes (provided by the University of Idaho) were added in an additional PCR step consisting of 10 µl 2x FastStart Master Mix, 0.75 µl barcoding primers, 8.25 µl H2O and 1 µl product from initial PCRs. Thermocycling for the barcoding PCR consisted of an initial denaturation at 95°C for 4 min, 10 cycles of 95°C for 30 s, 60°C for 30 s, 72°C for 1 min and a final extension at 72°C for 7 min. Successful addition of barcodes was confirmed by visualization of an expected size shift on a 1.5% gel. Pooled amplicons were size selected, bead purified and sequenced (paired-end 2x 300 Illumina MiSeq) at the University of Idaho IBEST Genomics Resources Core. Sequences are available in Genbank (SRP132277). Sequence processing Primer and barcode sequences were removed and reads were paired using PEAR (v0.9.6; (Zhang et al. 2014)). Sequences were clustered into operational taxonomic units (OTUs) using the UPARSE pipeline (Edgar 2013). Briefly, reads were quality filtered to include only sequences longer than 200 bp, no ambiguous bases, and a maximum expected error rate of 1, singletons were removed, and reads were dereplicated prior to clustering with the cluster_otus command in USEARCH. Raw reads were mapped to OTUs using vsearch (Rognes et al. 2016) to generate an OTU table. For fungal sequences, taxonomy was assigned to OTU representative sequences (centroids) with BLAST+ to the UNITE database (31.1.2016 general release). OTUs were filtered to include only those with a match length >75%, a percent identity >80% to UNITE representatives, and a total read count of five or more. Despite the lack of a noticeable band on the agarose gel or quantifiable PCR product, a single OTU with relatively high sequence counts (>2000) was found in the sample-free DNA extraction/negative control. This OTU had a high sequence similarity to Ascochyta skagwayensis (99.4% BLAST identity to accession GU054291| SH215821.07FU), a leaf blight pathogen, and was removed from all samples prior to rarefaction to a depth of 10000 sequences/sample. Raw (unrarefied) OTU tables were retained for differential abundance and network analyses with DESeq2 (Love, Huber and Anders 2014) and SPIEC-EASI (Kurtz et al. 2015) as described below. Data analyses Similarity in fungal communities among soil depths and aspects was evaluated using non-metric multidimensional scaling (NMDS) of Bray–Curtis distances using the metaMDS function in the vegan package of R (Oksanen et al.2016). Further, PERMANOVA was used to test for effects of soil depth and landscape position (aspect) using the ‘adonis’ function in vegan. Because there appeared to be differences in dispersion of fungal communities among soil depths, ANOVA was performed on community dispersions from centroids using the ‘betadisper’ function of vegan. Further, ANOVA was used on Log10(1+x)-transformed sequence counts of abundant phyla and families to identify fungal clades that differed significantly among soil depths or aspects, using an Benjamini–Hochberg correction to adjust p-values for false discovery. To identify OTUs that differed significantly among soil depths or between north- and south-facing aspects, unrarefied OTU tables were filtered to remove OTUs with <10 counts and DESeq2 (Love, Huber and Anders 2014) was used to contrast each successive soil depth (0 versus 10 cm, 10 versus 25 cm, 25 versus 50 cm, 50 versus 75 cm and 75 versus 100 cm), and north- and south- facing slopes. OTUs with an FDR-adjusted p-value <0.001 and a base mean >100 were considered differentially abundant. Observed richness (number of OTUs) and diversity (Shannon and inverse Simpson index) were calculated for each sample in vegan, and differences in richness and diversity among soil depths and aspects were evaluated using ANOVA. Further, correlations between richness and diversity with soil and environmental characteristics were evaluated using Spearman’s rank correlations. To assign putative functional characteristics to fungal OTUs, the Funguild script (Nguyen et al. 2016) was used. Among fungi that could be classified to a functional guild with ‘probable’ or ‘highly probable’ confidence, Kruskal–Wallis tests were performed to identify abundant groups that differed among soil depths. Finally, co-association networks among taxa at each depth or landscape position (aspect) were constructed using the sparCC algorithm implemented in the SPEIC-EASI package of R (Kurtz et al. 2015). Briefly, unrarefied OTU tables were filtered to include only samples belonging to each depth (n = 16 samples per depth) or aspect (n = 36 samples per aspect), then only OTUs found in at least 20% of samples with a total sequence count of >20 were retained for network inference. Co-associations with an R-value >0.5 and a p-value <0.01 after 100 bootstraps were considered significant. Network and node characteristics (graph density, clustering coefficient, node hub-scores) were evaluated with the igraph package in R (Csardi and Nepusz 2006), where the hub-score of a taxon reflects its structural importance in a network (Kleinberg 1999). Spearman correlations among taxon hub scores were used to compare the importance of taxa found in multiple networks among soil depths and aspects. RESULTS Illumina sequencing After data processing 2770 078 sequences were obtained among 96 samples with an average of 28855 +/− 9357 sequences per sample. After rarefaction (discarding three samples with <10000 sequences), sequences clustered into 678 OTUs belonging to seven identified phyla of which Ascomycota (472 OTUs or 87 +/- 14 % of sequences), Basidiomycota (105 OTUs or 7.8 +/- 13% of sequences) and Zygomycotya (29 OTUs, or 4.1 +/- 5.7 % of sequences) accounted for the vast majority of taxa. Minor groups included Cercozoa (1 OTU), Chytridiomycota (22 OTUs), Glomeromycota (7 OTUs) and Rozellomycota (3 OTUs). Fungal community composition varies with soil depth and landscape slope/aspect Fungal communities varied primarily with soil depth (Adonis r2 = 0.25, P = 0.001; Fig. 2), though members of the phylum Ascomycota were dominant among all depths (Fig. 2). In general, the prevalence of fungal phyla was similar across the depth profile, though, there was a distinct increase in the relative abundance of Basidiomycota at 10 cm below the soil surface (where the surface is 0 cm depth) versus at other strata (Fig. 3; ANOVA F = 8.44, P<0.0001) that corresponded with a decrease in the prevalence of Ascomycota (ANOVA F = 3.25, P = 0.01). At the family-level, 22 groups varied significantly with soil depth (FDR-adjusted p-values<0.05; Fig. 3), where members of some families tended to be more abundant at the soil surface (0 cm; eg. Glomerellaceae, Teratosphaeriaceae, Mycosphaerellaceae, Pleosporaceae, Pezizaceae, Lasiosphaeriaceae), some in shallow soil layers (10 and 25 cm; eg. Cephalothecaceae, Clavicipitaceae, Nectriaceae, Trichocomaceae, Chaetomiaceae, Hypicreaceae and Mycotrichaceae), and others in the deepest strata (50, 75 and 100 cm; eg. Bionectriaceae). Among OTUs with relative abundances that differed significantly with soil depth, the soil surface was dominated by OTUs identified as Sordariales, Fusarium, Mycosphaerella, Glarea and Humicola species, upper soil layers (10 and 25 cm) were dominated by Ascomycota, Nectriaceae, Hypocreales, Oidiodendron, Trichoderma and Penicillium species and the deepest soils were dominated by OTUs related to Bionectria, Ijuhya, Hypocreales, Mortierella and Ascomycota (Fig. 4). Thus, fungal community composition was highly stratified across soil depths, where some clades tended to dominate in shallow, intermediate or deep soil layers. Intriguingly, the variability in fungal community composition increased with depth (Fig. 2; ANOVA F = 4.34, P = 0.001); communities deeper in the soil were more different from one another than those nearer to the soil surface. Figure 2. View largeDownload slide NMDS plot (A) of fungal communities from different soil depths and landscape positions (stress=0.15) and the percentage of sequences belonging to fungal phyla among soil depths (B) and landscape positions (C). Distance to centroid at each depth is shown in D. Error bars represent standard deviations in D. Figure 2. View largeDownload slide NMDS plot (A) of fungal communities from different soil depths and landscape positions (stress=0.15) and the percentage of sequences belonging to fungal phyla among soil depths (B) and landscape positions (C). Distance to centroid at each depth is shown in D. Error bars represent standard deviations in D. Figure 3. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant fungal families that vary significantly (FDR p-value<0.05) among soil depths (cm) (triangle) or landscape position (circles). Dendogram on the top represents hierarchical clustering of fungal family abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 3. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant fungal families that vary significantly (FDR p-value<0.05) among soil depths (cm) (triangle) or landscape position (circles). Dendogram on the top represents hierarchical clustering of fungal family abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 4. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant OTUs (base mean among all samples >100) that differ significantly (FDR p-value<0.001) among soil depth (cm). Dendogram on the left represents hierarchical clustering of OUT abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 4. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant OTUs (base mean among all samples >100) that differ significantly (FDR p-value<0.001) among soil depth (cm). Dendogram on the left represents hierarchical clustering of OUT abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Landscape positio, based on slope and aspectn (north-facing slopes, south-facing slopes, toe-slope (bottom), and summit (top) also explained a significant portion of variation in fungal communities (Adonis r2 = 0.05, P = 0.001), however, only unidentified phyla differed significantly between aspects and were more abundant in south-facing versus north-facing slopes (F = 5.53, P = 0.002). At the family level, only three clades differed significantly with aspect (FDR p-value<0.05). Members of Chaetomiaceae tended to be more abundant in soils from north-facing slopes and toe-slopes, whereas members of Coniochaetaceae tended to be more abundant in soils from toe-slopes (Fig. 3). The relative abundance of taxa within Rustromiaceae was greatest in toe-slopes in the surface layers, yet was low in deeper soil depths. North facing positions had a higher relative abundance of OTUs related to Sordariales, Humicola, Lasiosphaeraceae, Cryptococcus and others (Fig. 5). In contrast, south facing positions had greater relative abundances of OTUs related to Phialophora, Lasiosphaeraceae, Sordariales, Mycosphaerella tassiana and unidentified Ascomycota (Fig. 5). Figure 5. View largeDownload slide Differentially abundant taxa between north-facing and south-facing slopes. Points to the right of 0 are more abundant on north-facing slopes, whereas those to the left of 0 are more abundant on south-facing slopes. The size of points is scaled by the mean abundance (baseMean) of that OTU and points are colored by the phylum to which they belong. Labels on the left indicate the nearest blast hit, blast % identity, and the OTU identifier. Figure 5. View largeDownload slide Differentially abundant taxa between north-facing and south-facing slopes. Points to the right of 0 are more abundant on north-facing slopes, whereas those to the left of 0 are more abundant on south-facing slopes. The size of points is scaled by the mean abundance (baseMean) of that OTU and points are colored by the phylum to which they belong. Labels on the left indicate the nearest blast hit, blast % identity, and the OTU identifier. Fungal diversity across soil depths and landscape slope/aspect Fungal richness and diversity declined significantly with increasing soil depth (ANOVA F = 108.7, 10.9 and 4.1; P<0.002 for observed richness, Shannon and inverse Simpsons indices; Fig. 6), though these indices did not differ significantly with aspect after excluding summit and toe slopes due to low replication (data not shown). Thus, fungal richness and diversity varied in concert with soil C (R = 0.81, 0.57, 0.38; P<0.001, for richness, Shannon, and Simpsons diversity, respectively), N (R = 0.80, 0.56, 0.37; P≤0.001, for richness, Shannon and Simpsons diversity, respectively) and pH (R = −0.85, −0.61, −0.39, P < 0.001, for richness, Shannon and Simpsons diversity, respectively). Moreover, although richness did not differ significantly between north- and south-facing slopes, fungal richness at the 10 cm depth was negatively related to global irradiance (R = −0.51, P = 0.04; Fig. 7). This pattern was marginally significant at the soil surface and 25 cm depth (R = −0.40, P = 0.13 and R = −0.44, P = 0.087 for 0 cm and 25 cm, respectively) though this pattern was not present at the deeper soil layers (50 cm: R = 0.06, P = 0.82; 75 cm: R = −0.04, P = 0.87; 100 cm: R = 0.20, P = 0.50). Thus, more intense solar irradiation may reduce fungal diversity within a landscape, though only near the soil surface. Figure 6. View largeDownload slide Relationship between soil depth and fungal richness, diversity and soil edaphic factors. Figure 6. View largeDownload slide Relationship between soil depth and fungal richness, diversity and soil edaphic factors. Figure 7. View largeDownload slide Relationships between fungal richness and global irradiance among soil depths (cm). Global irradiance expressed as MegaWatt-hours m-1 yr-1. Figure 7. View largeDownload slide Relationships between fungal richness and global irradiance among soil depths (cm). Global irradiance expressed as MegaWatt-hours m-1 yr-1. Variation in fungal life-history strategies among soil depths The relative abundance of fungi belonging to different common trophic modes (mean relative abundance >1%) differed among soil depths. Specifically, those taxa inferred as having pathotrophic-saprotrophic (Kruskal-Wallis test: chi-squared=20.5, P = 0.001), pathotrophic-symbiotrophic (Kruskal-Wallis test: chi-squared=11.0, P = 0.04) and symbiotrophic (Kruskal-Wallis test: chi-squared=22.7, P = 0.0004) lifestyles varied significantly among soil depths (Fig. 8). Taxa with pathotroph-symbiotroph or symbiotrophic trophic modes tended to be relatively more abundant at 10 cm below the soil surface, and those with pathotrophic-saprotrophic or pathotrophic-symbiotrophic abundance of different trophic modes did not vary significantly with landscape position (aspect), with the exception of saprotrophs, which tended to be more abundant in toe- and summit soils (Kruskal-Wallis test: chi-squared=8.9, P = 0.03). However, the difference in the frequency of saprotrophs was not significant between only north- and south- facing aspects (Kruskal-Wallis test: chi-squared=0.56, P = 0.45). Figure 8. View largeDownload slide Proportion of fungi classified to different trophic modes across soil depths (cm). Figure 8. View largeDownload slide Proportion of fungi classified to different trophic modes across soil depths (cm). Fungal co-occurrence patterns across soil depths Co-occurrence networks among fungal OTUs were strongly impacted by soil depth (Fig. 9; Supplemental Fig. S1). Specifically, there were fewer significant co-associations among a smaller number of OTUs as soil depth increased (Table 2), likely reflecting the lower fungal diversity in deeper soil layers. For example, there were significant correlations between the average taxon richness of communities at each depth and the number of network nodes (R = 0.99, P<0.001), positive edges (R = 0.99, P<0.001 ), negative edges (R = 0.99, P<0.001 ) and network density (R = −0.86, P = 0.03), but not with the positive:negative edge ratio or the clustering coefficient (R = −0.28, P = 0.59 and R = 0.05, P = 0.92, respectively). Similarly, there were few commonalities in the co-association patterns among taxa in networks from different depths, indicating vertical stratification of ecological dynamics and species interactions. However, the density of co-associations among taxa tended to increase with depth, suggesting that, despite lower diversity, fungal populations in deep soil strata are more likely to be driven by similar factors (eg. carbon availability) than those near the surface. In contrast, co-occurrence networks from north- and south- facing slopes were similar in size (number of nodes and edges) and had similar co-association patterns, though the co-association network among taxa from south-facing slopes was denser (Fig. 10, Table 2, Supplemental Fig. S2). Figure 9. View largeDownload slide Networks of positive co-occurrences among fungi from different sampling depths. Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from different depths (n = 16). Figure 9. View largeDownload slide Networks of positive co-occurrences among fungi from different sampling depths. Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from different depths (n = 16). Figure 10. View largeDownload slide Networks of positive co-occurrences among fungi from different aspects (landscape positions). Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from all depths for soil cores from north- or south-facing slopes (n = 36). Figure 10. View largeDownload slide Networks of positive co-occurrences among fungi from different aspects (landscape positions). Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from all depths for soil cores from north- or south-facing slopes (n = 36). Table 2. Characteristics of fungal co-association networks at different soil depths and aspects. Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 View Large Table 2. Characteristics of fungal co-association networks at different soil depths and aspects. Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 View Large The taxa that occupied central positions in the networks differed among networks of taxa from different depths, though hub-scores of taxa were significantly correlated among those from 10 cm versus 25 cm (R = 0.295, P = 0.01), 25 cm and 50 cm (R = 0.53, P = 0.002), suggesting that fungi share similar network positions at these depths. At the surface layer (0 cm) , taxa most central to the co-occurrence network included Verrucariles, Cryptococcus, Knufia and Penicillium species (Supplemental Fig. S3), those at 10 cm included Ascomycota, Chaetomiaceae, Sordariomycetes and Lasiosphaeraceae species, those at 25 cm included Bionectria, Mortierella, Chaetomiaceae, Sordariomycetes and Humicola species, those at 50 cm included Hypocreales, Mortierella, Nectriaceae and Ascomycota species, those at 75 cm included Basidiomycota, Sordariales, Ascomycota, Nectriaceae and Microdochium speices and at 100 cm only Hypocreales, Bionectria and Mortierella had more than one significant co-association. In contrast, taxa among networks from north- and south- facing slopes has similar positions within each network (R = 0.80, P<0.001), where Humicola, Bionectria, Mortierella and unidentified Ascomycota species were most central within each network (Supplemental Fig. S4). DISCUSSION Soil depth was a strong driver of fungal community composition and diversity in no-till in this no-till wheat cropping system, consistent with studies of forest (Fierer, Schimel and Holden 2003; Baldrian et al. 2012), prairie (Jumpponen, Jones and Blair 2010) and other agricultural soils (Derpsch et al. 2010; Tojuet al. 2016). For example, Sipila et al., (Sipila, Yrjala and Alakukku 2012) used T-RFLP and PLFA to characterize microbial communities and found that moldboard plowing homogenizes communities in the upper layer (0–20 cm), whereas they were more stratified in no-till soils, though other studies of fallow fields have suggested that there is little difference in fungal community composition or diversity with soil depth (Ko et al.2017). Variation in fungal communities across depths often tracks stratification in C, N, pH and other soil characteristics (Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010; Peay, Kennedy and Talbot 2016; Lamit et al. 2017), which is especially strong near the surface in agricultural fields under no-till versus those under conventional tillage (Zhao et al. 2015) It has been long understood that there is a succession of fungi on decomposing residue (Sadasivan 1939), since plant residue serves as a key resource base for fungi, stratification of fungal communities is likely to reflect different successional stages of residue decomposition. For example, the soil surface may have carbon derived from the standing crop residue and litter layer, and would have the greatest amounts of labile organic material. In parallel with studies of succession during decomposition, surface soil was dominated by many cellulolytic Ascomycetous fungi including those belonging to Sordariales, Coniochaeta (Lecythophora), Fusarium, Mycosphaerella, Humicola and Pleosporales, which are rapidly able to colonize wheat straw and are often found to be most abundant during early stages of decomposition (Harper and Lynch 1985; Maet al. 2013). The initial cellulolytic colonists of wheat residue, however, are estimated to only utilize a small portion of the available carbon (Harper and Lynch 1985). In later stages of decomposition of polymers (eg. cellulose), other taxa, such as Trichocladium (Chaetomiaceae), which tended to predominate at 10 cm depth, are expected to be more competitive (Poll et al. 2010). Although some residue may be incorporated by higher disturbance hoe-type no-till seeders, the roots of wheat and other rotation crops may also be significant sources of carbon at this depth. As soil depth increases and root density decreases, those taxa that dominated deeper soil layers (50 and 100 cm), related to Hypocreales, Mortierella, Ijuhya, Bionectria and unidentified Ascomycota species, are likely to be those able to degrade older, recalcitrant carbon sources (eg. lignin). Moreover, anecic earthworms (eg. Lumbricus terrestris), which are commonly present in no-till systems and can burrow to 3 m depth (Chan 2001, Aeschlimann, personal communication), may drag fresh surface residue into deep burrows, providing a significant source of C for microbial communities. Alternatively, fungi in deeper soil layers may act as pathogens or symbiotrophs in addition to saprotrophs as a means of obtaining nutrients, reflecting a shift in food web structures in deeper soil layers (Lindahl et al. 2007; Peay, Kennedy and Talbot 2016; Lamit et al. 2017). This is supported by the increase in the relative abundance of taxa assigned to trophic modes with pathogenic or symbiotrophic stages at deeper soil depths. In addition to shifts in fungal community composition with increasing soil depth, variation in soil characteristics (C, N, pH, moisture, temperature) are expected to impact the outcome of fungal interactions (Moore and Six 2015; Hiscox et al. 2016). Indeed, co-association patterns among fungal taxa, which are frequently used to infer species interaction or habitat preference (Poudelet al. 2016), differed considerably with soil depth. Although similar between networks from 0 and 10 cm, network hub taxa were often specific to a particular soil depth, suggesting that a given fungal taxa may play unique roles within communities at different soil depths. Moreover, the smaller, denser networks with increasing soil depth having distinct associations among taxa may jointly reflect fungal density and diversity, niche partitioning and the frequency of competitive or cooperative interactions (Poudel et al. 2016; Toju et al.2016). For example, the larger, more distributed networks at the upper 0 and 10 cm depths may result from the combination of greater microbial biomass and activity in the upper soil (Fierer, Schimel and Holden 2003), resulting in a more significant role of direct and indirect species interactions for fungal fitness. In contrast, in deep soil layers, where there is low resource availability and low microbial biomass, competitive species interactions among fungi may have little impact on biological fitness. Alternatively, the fungi detected in deep soil layers may exist primarily as inactive spores or represent ‘relic’ DNA (Cariniet al. 2016). Similar patterns in fungal co-occurrence networks, however, were described recently in forest systems by Toju et al. (2016) where fungal networks from different soil horizons differed in complexity. Together, the substantial variation in fungal co-occurrence networks suggests that variation in biotic and abiotic characteristics across the soil profile generate distinct ecological dynamics among soil fungi among soil layers. Fungal richness and diversity declined with soil depth in concert with decreasing C, N, and with increasing pH. However, there were still on average ∼24 OTUs found at 75 cm and 100 cm depths, consistent with previous observations of considerable fungal diversity in deep soils (Jumpponen, Jones and Blair 2010; Toju et al. 2016). In the upper soil layers, the small but significant negative associations between irradiance and fungal richness suggest that, despite relatively close proximity (<1 km), greater amounts of annual solar irradiance within a field can result in reduced fungal diversity in the upper 25 cm of soil. This may also reflect soil water levels, since south facing slopes with greater solar irradiance have more evapotranspiration. Finally, a unique finding of this study was that variation in fungal community composition increased with soil depth. The higher variance in fungal community composition with increasing soil depth may mirror a greater patchiness in soil C or other resources as depth increases, perhaps due to a greater reliance on resources provided by deep roots or earthworms (Shuster, Subler and McCoy 2001). Alternatively, more frequent dispersal of fungal propagules in the upper soil layers, perhaps facilitated by growing plant roots or earthworm/mesofaunal activity, may contribute to more homogeneous fungal communities near the soil surface. Contrary to our expectations that fungal communities would differ considerably between north- and south- facing aspects, only a small number of taxa was associated with one or the other aspects. Moreover, fungal co-occurrence networks from north- and south- facing slopes were strikingly similar, suggesting similar community dynamics at these aspects. Only small differences between north- and south- facing slopes may be due to the homogeneity of the previous crop species (wheat), since plant species identity can have a strong impact on fungal communities in soil (Peay, Baraloto and Fine 2013; LeBlanc, Kinkel and Kistler 2017). Moreover, consistent management practices, such as fertilizer treatments or mechanical disturbance may play a role in homogenizing the fungal community in upper soil layers (Weber, Vilgalys and Kuske 2013), whereas lower soil depths may not be strongly impacted by differences in temperature or water due to landscape position. Finally, samples were taken early in the season (April) when soils were likely to be more uniformly moist, and differences in communities due to aspect may only be significant for a small time period later on during the growing season when there are strong gradients in soil water or temperature. However, even though difference in community composition between north- and south-facing slopes were minor, there may be differences in the biomass or density of fungi that we were unable to detect with relative abundance data. In total, this work provides significant insight into variation in depth profiles of fungal communities across the landscape in a no-till wheat cropping system. Although dead roots and earthworm activity may be a significant source of nutrients for fungi in deep soils, differences in fungal communities among soil depths likely reflects the composition and quantity of available carbon sources, where fungal communities follow a pattern of succession during residue degradation. Moreover, stratification of fungal communities across the soil depth profile is likely to result in distinct ecological dynamics at each stratum, as exemplified by differences in co-association networks. However, there were only small differences in fungal communities between aspects, perhaps due to homogenization of communities by agricultural management practices. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. ACKNOWLEDGEMENTS Author Contributions: All authors discussed the results and edited and commented on the manuscript. TP and DCS designed the project. TP, DCS and KK collected the soil samples. DCS performed all the molecular sequencing and data analysis. BC provided metadata. DRH provided the study site, including long-term field work, maintenance, metadata, and LTAR funding and direction. DCS and TP wrote the manuscript. FUNDING Funding from USDA-ARS Postdoctoral Research Associate Award to the first author and REACCH (Regional Approaches to Climate Change- Pacific Northwest Agriculture) award #2011-68002-30191 from USDA National Institute for Food and Agriculture. Conflict of interest. None declared. REFERENCES Akob DM , Kusel K . Where microorganisms meet rocks in the Earth’s Critical Zone . Biogeosciences . 2011 ; 8 : 3531 – 43 . Google Scholar CrossRef Search ADS Baldrian P , Kolařík M , Štursová M , et al. Active and total microbial communities in forest soil are largely different and highly stratified during decomposition . ISME J . 2012 ; 6 : 248 – 58 . Google Scholar CrossRef Search ADS PubMed Blackmore S , Godwin RJ , Fountas S , et al. The analysis of spatial and temporal trends in yield map data over six years . Biosystems Engineering . 2003 ; 84 : 455 – 66 . Google Scholar CrossRef Search ADS Carini P , Marsden PJ , Leff JW et al. Relic DNA is abundant in soil and obscures estimates of soil microbial diversity . Nat Microbiol . 2016 ; 2 : 16242 . Google Scholar CrossRef Search ADS PubMed Chan K . An overview of some tillage impacts on earthworm population abundance and diversity — implications for functioning in soils . Soil Tillage Res . 2001 ; 57 : 179 – 91 . Google Scholar CrossRef Search ADS Csárdi G , Nepusz T . The igraph software package for complex network research . InterJournal . 2006 ; 1695 . Degrune F , Theodorakopoulos N , Dufrêne M , et al. No favorable effect of reduced tillage on microbial community diversity in a silty loam soil (Belgium) . Agric Ecosyst Environ . 2016 ; 224 : 12 – 21 . Google Scholar CrossRef Search ADS Derpsch R , Friedrich T , Kassam A , et al. Current status of adoption of no-till farming in the world and some its main benefits . Int J Agric Biol Eng . 2010 ; 3 , DOI: https://doi.org/10.3965/j.issn.1934-6344.2010.01.0-0 . Edgar RC . UPARSE: highly accurate OTU sequences from microbial amplicon reads . Nat Methods . 2013 ; 10 : 996 – 8 . Google Scholar CrossRef Search ADS PubMed Ekelund F , Rønn R , Christensen S . Distribution with depth of protozoa, bacteria and fungi in soil profiles from three Danish forest sites . Soil Biol Biochem . 2001 ; 33 : 475 – 81 . Google Scholar CrossRef Search ADS Fan JL , McConkey B , Wang H , et al. Root distribution by depth for temperate agriculture crops . Field Crops Research . 2016 ; 189 : 68 – 74 . Google Scholar CrossRef Search ADS Fierer N , Schimel JP , Holden PA . Variations in microbial community composition through two soil depth profiles . Soil Biol Biochem . 2003 ; 35 : 167 – 76 . Google Scholar CrossRef Search ADS Harper SHT , Lynch JM . Colonization and decomposition of straw by fungi . Trans Br Mycol Soc . 1985 ; 85 : 655 – 61 . Google Scholar CrossRef Search ADS Hawksworth DL . The magnitude of fungal diversity: the 1.5 million species estimate revisited . Mycol Res . 2001 ; 105 : 1422 – 32 . Google Scholar CrossRef Search ADS Hipple KW . Washington Soil Atlas , Washington, DC , USDA-NRCS Publication , 2011 . Hiscox J , Clarkson G , Savoury M , et al. Effects of pre-colonisation and temperature on interspecific fungal interactions in wood . Fungal Ecol . 2016 ; 21 : 32 – 42 . Google Scholar CrossRef Search ADS Horner GM , McCall AG , Bell FG . Investigations in erosion control and reclamation of eroded land at the Palouse Conservation Experiment Station 1931–1942 , U.S. Department of Agriculture, Soil Conservation Service, Technical Bulletin 860 . 1948 . Huggins DR , Reganold JP . No-till: the quiet revolution . Sci Am . 2008 ; 299 : 70 – 7 . Google Scholar CrossRef Search ADS PubMed Huggins DR , Uberuaga DP . Field heterogeneity of soil organic carbon and relationships to soil properties and terrain attributes . Climate friendly farming: Center for Sustaining Agriculture and Natural Resources Research Report 2010–001 . 2010 . Available at http://csanr.wsu.edu/pages/Climate_Friendly_Farming_Final_Report/. Jumpponen A , Jones KL , Blair J . Vertical distribution of fungal communities in tallgrass prairie soil . Mycologia . 2010 ; 102 : 1027 – 41 . Google Scholar CrossRef Search ADS PubMed Kleinberg JM . Authoritative sources in a hyperlinked environment . J ACM . 1999 ; 46 : 604 – 32 . Google Scholar CrossRef Search ADS Ko D , Yoo G , Yun S-T , et al. Bacterial and fungal community composition across the soil depth profiles in a fallow field . J Ecol Environ . 2017 ; 41 : 34 . Google Scholar CrossRef Search ADS Kristensen KJ . Temperature and heat balance of soil . Oikos . 1959 ; 10 : 103 . Google Scholar CrossRef Search ADS Kurtz ZD , Müller CL , Miraldi ER , et al. Sparse and compositionally robust inference of microbial ecological networks . PLOS Comput Biol . 2015 ; 11 : e1004226 . Google Scholar CrossRef Search ADS PubMed Lamit LJ , Romanowicz KJ , Potvin LR , et al. Patterns and drivers of fungal community depth stratification in Sphagnum peat . FEMS Microbiol Ecol . 2017 ; 93 , DOI: https://doi.org/10.1093/femsec/fix082 . LeBlanc N , Kinkel L , Kistler HC . Plant diversity and plant identity influence Fusarium communities in soil . Mycologia . 2017 ; 109 : 128 – 39 . Google Scholar CrossRef Search ADS PubMed LeBlanc N , Kinkel LL , Kistler HC . Soil fungal communities respond to grassland plant community richness and soil edaphics . Microb Ecol . 2015 ; 70 : 188 – 95 . Google Scholar CrossRef Search ADS PubMed Lindahl BD , Ihrmark K , Boberg J , et al. Spatial separation of litter decomposition and mycorrhizal nitrogen uptake in a boreal forest . New Phytol . 2007 ; 173 : 611 – 20 . Google Scholar CrossRef Search ADS PubMed Love MI , Huber W , Anders S . Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome Biol . 2014 ; 15 : 550 . Google Scholar CrossRef Search ADS PubMed Ma A , Zhuang X , Wu J , et al. Ascomycota members dominate fungal communities during straw residue decomposition in arable soil . PLoS ONE . 2013 ; 8 : e66146 . Google Scholar CrossRef Search ADS PubMed Mahler RL , Bezdicek DF . Diversity of Rhizobium legumosarium in the Palouse of eastern Washington . Applied and Environmental Microbiology . 1978 ; 36 : 780 – 2 . Google Scholar PubMed Moore ML , Six DL . Effects of temperature on growth, sporulation, and competition of mountain pine beetle fungal symbionts . Microb Ecol . 2015 ; 70 : 336 – 47 . Google Scholar CrossRef Search ADS PubMed Nagle GN , Ritchie JC . Wheat field erosion rates and channel bottom sediment sources in an intensively cropped northeastern Oregon drainage basin . Land Degradation and Development . 2004 ; 15 : 15 – 26 . Google Scholar CrossRef Search ADS Nguyen NH , Song Z , Bates ST , et al. FUNGuild: An open annotation tool for parsing fungal community datasets by ecological guild . Fungal Ecol . 2016 ; 20 : 241 – 8 . Google Scholar CrossRef Search ADS Oksanen J , Blanchette FG , Friendly M et al. Vegan: Community Ecology Package . 2016 . https://cran.r-project.org Peay KG , Baraloto C , Fine PV . Strong coupling of plant and fungal community structure across western Amazonian rainforests . ISME J . 2013 ; 7 : 1852 – 61 . Google Scholar CrossRef Search ADS PubMed Peay KG , Kennedy PG , Talbot JM . Dimensions of biodiversity in the Earth mycobiome . Nat Rev Microbiol . 2016 ; 14 : 434 – 47 . Google Scholar CrossRef Search ADS PubMed Poll C , Brune T , Begerow D , et al. Small-scale diversity and succession of fungi in the detritusphere of rye residues . Microb Ecol . 2010 ; 59 : 130 – 40 . Google Scholar CrossRef Search ADS PubMed Poudel R , Jumpponen A , Schlatter DC , et al. Microbiome networks: A systems framework for identifying candidate microbial assemblages for disease management . Phytopathology . 2016 ; 106 : 1083 – 96 . Google Scholar CrossRef Search ADS PubMed Powell JR , Karunaratne S , Campbell CD , et al. Deterministic processes vary during community assembly for ecologically dissimilar taxa . Nature Communications . 2015 ; 6 : 8444 . Google Scholar CrossRef Search ADS PubMed Raaijmakers JM , Paulitz TC , Steinberg C , et al. The rhizosphere: a playground and battlefield for soilborne pathogens and beneficial microorganisms . Plant Soil . 2009 ; 321 : 341 – 61 . Google Scholar CrossRef Search ADS Rognes T , Flouri T , Nichols B , et al. . VSEARCH: a versatile open source tool for metagenomics . PeerJ . 2016 ; 4 : e2584 . Google Scholar CrossRef Search ADS PubMed Sadasivan TS . Succession of fungi decomposing wheat straw in different soils, with special reference to Fusarium culmorum . Ann Appl Biol . 1939 ; 26 : 497 – 508 . Google Scholar CrossRef Search ADS Schillinger WF , Papendick RI . Then and Now: 125 years of dryland wheat farming in the inland Pacific Northwest . Agron J . 2008 ; 100 : S – 166 . Google Scholar CrossRef Search ADS Schulz S , Brankatschk R , Dümig A , et al. The role of microorganisms at different stages of ecosystem development for soil formation . Biogeosciences . 2013 ; 10 : 3983 – 96 . Google Scholar CrossRef Search ADS Sharma-Poudyal D , Schlatter D , Yin C , et al. Long-term no-till: A major driver of fungal communities in dryland wheat cropping systems . PLOS ONE . 2017 ; 12 : e0184611 . Google Scholar CrossRef Search ADS PubMed Sharratt BS , Wendling L , Feng G . Surface characteristics of a wind blown soil altered by tillage intensity during summer fallow . Aeol Res . 2012 ; 5 : 1 – 7 . Google Scholar CrossRef Search ADS Shuster W , Subler S , McCoy E . Deep-burrowing earthworm additions changed the distribution of soil organic carbon in a chisel-tilled soil . Soil Biol Biochem . 2001 ; 33 : 983 – 96 . Google Scholar CrossRef Search ADS Sipila TP , Yrjala K , Alakukku L , et al. Cross-site soil microbial communities under tillage regimes: Fungistasis and microbial biomarkers . Appl Environ Microbiol . 2012 ; 78 : 8191 – 8201 . Google Scholar CrossRef Search ADS PubMed Sosa-Hernández MA , Roy J , Hempel S , et al. Subsoil arbuscular mycorrhizal fungal communities in arable soil differ from those in topsoil . Soil Biology and Biochemistry . 2018 ; 117 : 83 – 86 . Google Scholar CrossRef Search ADS Swift MJ , Heal OW , Anderson JM . Decomposition in Terrestrial Ecosystems . Berkeley : University of California Press , 1979 . Taylor DL , Hollingsworth TN , McFarland JW , et al. A first comprehensive census of fungi in soil reveals both hyperdiversity and fine-scale niche partitioning . Ecol Monogr . 2014 ; 84 : 3 – 20 . Google Scholar CrossRef Search ADS Tedersoo L , Bahram M , Polme S , et al. Global diversity and geography of soil fungi . Science . 2014 ; 346 : 1256688 . Google Scholar CrossRef Search ADS PubMed Toju H , Kishida O , Katayama N , et al. Networks depicting the fine-scale co-occurrences of fungi in soil horizons . PLOS ONE . 2016 ; 11 : e0165987 . Google Scholar CrossRef Search ADS PubMed Weber CF , Vilgalys R , Kuske CR . Changes in fungal community composition in response to elevated atmospheric CO2 and nitrogen fertilization varies with soil horizon . Front Microbiol . 2013 ; 4 , DOI: https://doi.org/10.3389/fmicb.2013.00078 . Zhang J , Kobert K , Flouri T , et al. PEAR: a fast and accurate Illumina Paired-End reAd mergeR . Bioinformatics . 2014 ; 30 : 614 – 20 . Google Scholar CrossRef Search ADS PubMed Zhang Y , Dong S , Gao Q , et al. Soil bacterial and fungal diversity differently correlated with soil biochemistry in alpine grassland ecosystems in response to environmental changes . Sci Rep . 2017 ; 7 : 43077 . Google Scholar CrossRef Search ADS PubMed Zhao X , Xue J-F , Zhang X-Q , et al. Stratification and storage of soil organic carbon and nitrogen as affected by tillage practices in the north China plain . PLOS ONE . 2015 ; 10 : e0128873 . Google Scholar CrossRef Search ADS PubMed Published by Oxford University Press on behalf of FEMS 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Ecology Oxford University Press

Fungal community composition and diversity vary with soil depth and landscape position in a no-till wheat-based cropping system

Loading next page...
 
/lp/ou_press/fungal-community-composition-and-diversity-vary-with-soil-depth-and-MUhw3CYWNO
Publisher
Oxford University Press
Copyright
Published by Oxford University Press on behalf of FEMS 2018.
ISSN
0168-6496
eISSN
1574-6941
D.O.I.
10.1093/femsec/fiy098
Publisher site
See Article on Publisher Site

Abstract

ABSTRACT Soil edaphic characteristics are major drivers of fungal communities, but there is a lack of information on how communities vary with soil depth and landscape position in no-till cropping systems. Eastern Washington is dominated by dryland wheat grown on a highly variable landscape with steep, rolling hills. High-throughput sequencing of fungal ITS1 amplicons was used to characterize fungal communities across soil depth profiles (0 to 100 cm from the soil surface) among distinct landscape positions and aspects across a no-till wheat field. Fungal communities were highly stratified with soil depth, where deeper depths harbored distinct fungal taxa and more variable, less diverse fungal communities. Fungal communities from deep soils harbored a greater portion of taxa inferred to have pathotrophic or symbiotrophic in addition to saprotrophic lifestyles. Co-occurrence networks of fungal taxa became smaller and denser as soil depth increased. In contrast, differences between fungal communities from north-facing and south-facing slopes were relatively minor, suggesting that plant host, tillage, and fertilizer may be stronger drivers of fungal communities than landscape position. fungal community, soil depth, aspect, slope, landscape position, no-till, wheat INTRODUCTION Fungi are critical components of agricultural systems where they play roles in carbon cycling, nutrient cycling, decomposition, and plant health (Raaijmakerset al. 2009). Fungal communities are highly diverse assemblages with an estimated 1.5 million species worldwide and hundreds of species within a single gram of soil (Hawksworth 2001; Taylor et al.2014; Peay, Kennedy and Talbot 2016). At the global scale, climatic conditions, soil characteristics and spatial patterns are believed to be the primary drivers of soil fungal communities (Tedersoo et al. 2014). The immense diversity of fungal communities across the landscape, however, is due in part to their extensive variability at small spatial scales, where soil edaphic parameters (eg. water, temperature, C, N), plant species identity, and plant diversity (LeBlanc, Kinkel and Kistler 2015, 2017; Peay, Kennedy and Talbot 2016; Zhanget al. 2017) drive variation in fungal community composition. Yet, with some exceptions (eg. Degrune et al. 2016; Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010), the vast majority of work examining spatial variation in soil fungal communities has focused only on communities found within the top 10–20 cm of soil. Thus, we have only scratched the surface of our understanding of how fungal communities vary within the soil profile. Fungi residing in deeper layers of the soil profile contribute to key processes, such as soil formation, carbon and nutrient cycling, and xenobiotic degradation, and form intimate associations with plant roots where they can enhance water and nutrient uptake, provide protection from pathogens, or be pathogens themselves (Swift, Heal and Anderson 1979; Schulz et al. 2013). Soil profiles are characterized by vertical gradients in carbon, nitrogen, pH and oxygen that frequently correspond with declines in microbial biomass and diversity (Akob and Kusel 2011; Ekelund, Rønn and Christensen 2001; Fierer, Schimel and Holden 2003). Further, deeper soil layers experience less extreme variation in moisture and temperature than surface soils (Kristensen 1959). Although it is recognized that fungal communities are likely to vary throughout the soil profile (Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010; Baldrian et al.2012; Peay, Kennedy and Talbot 2016; Toju et al. 2016; Lamit et al.2017), we have a poor understanding of the composition of deep-soil communities in agricultural systems and their diversity across the landscape. The Palouse region of eastern Washington and northern Idaho is characterized by steeply sloped, rolling hills, and although originally dominated by forest and prairie ecosystems, dryland wheat cropping has predominated in this area for the last century (Schillinger and Papendick 2008). Soil water is often limiting to crop production, since most precipitation occurs over the winter months (November to May). Crucially, the highly dynamic landscape topography creates a mosaic of microclimates across agricultural fields, especially where water, temperature, and plant productivity, vary considerably with the slope and aspect (Mahler and Bezdicek 1978). Specifically, north-facing slopes typically accumulate more snow over the winter, retain more water over the summer, and can be more productive than south-facing slopes. As such, it is likely that fungal communities are considerably impacted by the aspect of the landscape and plant productivity. Further, as a result of the landscape characteristics in the Palouse region, soil erosion due to wind, water, and mechanical tillage of soil is a major problem in this area, with an estimated average soil lost to erosion of 1 to 50 Mg ha–1 (Nagle and Ritchie 2004; Sharratt, Wendling and Feng 2012). To combat erosion, conserve soil water, increase soil organic matter, and save on fuel costs, growers are increasingly adopting no-till or direct-seed cropping practices (Huggins and Reganold 2008). In no-till, as the name implies, tillage is reduced or eliminated and crops are seeded directly into the previous crops’ stubble. Notably, there is little disturbance of the soil and crop residue is left on the soil surface rather than incorporated during tillage. Substantial differences in soil fungal communities exist between conventional tillage and no-till systems (Sharma-Poudyal et al.2017), yet little is known about how the spatial distribution of fungal communities is influenced by aspect, or how communities are stratified with soil depth in no-till fields across this highly-variable landscape. Additionally, unique characteristics of Palouse soils, which consist primarily of wind-deposited loess formed over 10000 years, may contribute to vertical stratification of fungal communities. These soils are very deep, with topsoil depths of 1.5–1.8 m or more (Horner, McCall and Bell 1948; Hipple 2011). Wheat roots can grow down to 3 m and could be a major source of carbon at these depths, but 50% of the roots are in the top 16 cm (Fanet al. 2016). Fungal interactions with soil carbon and wheat roots at different soil depths are likely to favor fungal taxa with different life-history strategies. For example, an abundance of carbon and plant roots in upper soil layers may generate a diverse community of saprotrophs and symbiotrophs, whereas deep soils, which are relatively depauperate in carbon, may select for a distinct community of pathotrophs and symbiotrophs. Furthermore, differences in fungal community composition and diversity among soil depths is likely to be reflected in unique species interactions and patterns in fungal networks. For example, AMF communities were shown to vary with soil depth (Sosa-Hernández et al. 2018). Powell et al. (2015) observed that fungi in the upper soil layer exhibit stochastic processes, possibly due to strong species interactions. Here, we use high-throughput DNA sequencing to explore the composition and spatial distribution of fungal communities at different depths (0 to 100 cm) across the landscape in a no-till system. Specifically, we hypothesize that (1) fungal community composition and diversity will differ across the soil depth profile; (2) soils on north-facing aspects will harbor distinct fungal communities from those of south-facing aspects; (3) fungi with saprotrophic lifestyles will be most abundant near the soil surface, while symbiotic and pathogenic modes will predominate at deeper depths, reflecting their nutritional preferences and (4) fungal co-occurrence patterns will be more complex near the soil surface than at deeper soil depths, reflecting a greater importance of species interactions. MATERIALS AND METHODS Site description and soil sampling Soil samples were collected from the Washington State University Cook Agronomy Farm, a USDA Long-Term Agroecosystem Research (LTAR) site in eastern Washington state, USA, located ∼7 miles northeast of Pullman, WA. Soils were sampled in April, 2016 from a transect spanning a range of slopes and aspects within a winter wheat-rotation (Fig. 1). This strip has been in a continuous cereal rotation for the 9 years previous to the sampling. This rotation was spring barley-spring wheat-winter wheat. Winter wheat was the last crop planted prior to sampling. On April 7, 2016, two soil cores were obtained from each of eight points using a tractor-mounted hydraulic Giddings soil probe (Windsor, CO). Briefly, the Giddings probe was fitted with plastic liners (PET-G, polyethylene terephthalate glycol-modified, Giddings Machine Co. Inc.) and 4.4 cm inside diameter cores were taken at each point to a depth of at least 100 cm. Cores were immediately capped, transported to the lab, and frozen at −20 C until processing as described below. The Cook Agronomy Farm has been studied extensively, and soil edaphic, environmental, and crop yield data exist for each sampling point (Table 1; Huggins and Uberuaga 2010). Figure 1. View largeDownload slide Landscape positions of sampling points at the Cook Agronomy Farm. Points 8B, 15K, and 16M are south-facing. Points 10D, 12G, and 18Q are north-facing. Points 9C and 12H are top-slope (summit) and bottom-slope (toe slope) cores, respectively. Contour interval = 2 m. Figure 1. View largeDownload slide Landscape positions of sampling points at the Cook Agronomy Farm. Points 8B, 15K, and 16M are south-facing. Points 10D, 12G, and 18Q are north-facing. Points 9C and 12H are top-slope (summit) and bottom-slope (toe slope) cores, respectively. Contour interval = 2 m. Table 1. General characteristics of sampling points. Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 a Yields were normalized to average yields (Blackmore et al.2003). View Large Table 1. General characteristics of sampling points. Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 Grid point Soil series Global irradiance (MegaWatt-hours m−1 yr−1.) Elevation (m) Slope position Relative yield (2015) Relative yield (2016)a 10D Thatuna 1568109 790.6 N Shoulder 1.27 1.55 12G Thatuna 1546946 784.1 N Footslope 0.57 0.77 12H Thatuna 1605529 782.3 Toeslope 0.75 0.91 15K Palouse 1723038 790.4 S Backslope 1.20 1.36 16M Palouse 1652716 796.5 S Shoulder 1.10 0.53 18Q Thatuna 1431788 790.7 N Backslope 0.96 0.61 8B Naff 1739815 789.6 S Shoulder 1.41 0.90 9C Palouse 1695119 791.0 Summit 1.28 0.67 a Yields were normalized to average yields (Blackmore et al.2003). View Large DNA extraction, PCR and Illumina sequencing Holes (∼2 cm diameter) were bored into the plastic sheaths of each soil frozen soil core at 10 cm, 25 cm, 50 cm, 75 cm and 100 cm from the soil surface with a drill and sealed with packing tape. Information on the soil horizons at each depth is provided in Supplemental Table S1. This includes the metadata such as EC, C, N, pH and bulk density, referenced to the points in Fig. 1. To obtain soil samples, cores were thawed individually on the benchtop for ∼30 min and ∼1 g of soil was taken from each bore hole and the surface (0 cm depth) of each core with a sterile spatula after scraping away the outer layer of soil. DNA was extracted from each sample (∼0.25 g) using the PowerSoil DNA extraction kit (MoBio/Qiagen) according to the manufacturers’ instructions using the ‘soil’ program on a FastPrep 5G homogenizer (MP Biomedicals) for the bead-beating step. A soil-free DNA extraction preparation was included as a negative control. DNA was quantified and checked for quality on a NanoDrop spectrophotometer (FischerThermo). The fungal ITS1 region was amplified with ITS1-F and ITS2 primers adapted to include linkers, barcoding tags, and additional bases to introduce a phase offset for Illumina sequencing (Supplemental Table S2). Primers were combined in equimolar ratio prior to PCR amplification. PCR reactions (20 µl total volume) consisted of 10 µl FastStart Master Mix (Roche), 0.7 µl of primers (10 µM), 8.3 µl PCR-grade H2O and 1 µl of template DNA (1/10 diluted). Reactions were performed in triplicate with three separate annealing temperatures (50°C, 53°C or 55°C). Thermocycling conditions consisted of an initial denaturation at 95°C for 4 min, 30 cycles of 95°C for 30 s, annealing at 50, 53 or 55°C for 30 s, and extension at 72°C for 45 s, with a final extension at 72°C for 7 min. PCR products were pooled and checked for amplification on a 1.5% gel. Illumina adapters and barcodes (provided by the University of Idaho) were added in an additional PCR step consisting of 10 µl 2x FastStart Master Mix, 0.75 µl barcoding primers, 8.25 µl H2O and 1 µl product from initial PCRs. Thermocycling for the barcoding PCR consisted of an initial denaturation at 95°C for 4 min, 10 cycles of 95°C for 30 s, 60°C for 30 s, 72°C for 1 min and a final extension at 72°C for 7 min. Successful addition of barcodes was confirmed by visualization of an expected size shift on a 1.5% gel. Pooled amplicons were size selected, bead purified and sequenced (paired-end 2x 300 Illumina MiSeq) at the University of Idaho IBEST Genomics Resources Core. Sequences are available in Genbank (SRP132277). Sequence processing Primer and barcode sequences were removed and reads were paired using PEAR (v0.9.6; (Zhang et al. 2014)). Sequences were clustered into operational taxonomic units (OTUs) using the UPARSE pipeline (Edgar 2013). Briefly, reads were quality filtered to include only sequences longer than 200 bp, no ambiguous bases, and a maximum expected error rate of 1, singletons were removed, and reads were dereplicated prior to clustering with the cluster_otus command in USEARCH. Raw reads were mapped to OTUs using vsearch (Rognes et al. 2016) to generate an OTU table. For fungal sequences, taxonomy was assigned to OTU representative sequences (centroids) with BLAST+ to the UNITE database (31.1.2016 general release). OTUs were filtered to include only those with a match length >75%, a percent identity >80% to UNITE representatives, and a total read count of five or more. Despite the lack of a noticeable band on the agarose gel or quantifiable PCR product, a single OTU with relatively high sequence counts (>2000) was found in the sample-free DNA extraction/negative control. This OTU had a high sequence similarity to Ascochyta skagwayensis (99.4% BLAST identity to accession GU054291| SH215821.07FU), a leaf blight pathogen, and was removed from all samples prior to rarefaction to a depth of 10000 sequences/sample. Raw (unrarefied) OTU tables were retained for differential abundance and network analyses with DESeq2 (Love, Huber and Anders 2014) and SPIEC-EASI (Kurtz et al. 2015) as described below. Data analyses Similarity in fungal communities among soil depths and aspects was evaluated using non-metric multidimensional scaling (NMDS) of Bray–Curtis distances using the metaMDS function in the vegan package of R (Oksanen et al.2016). Further, PERMANOVA was used to test for effects of soil depth and landscape position (aspect) using the ‘adonis’ function in vegan. Because there appeared to be differences in dispersion of fungal communities among soil depths, ANOVA was performed on community dispersions from centroids using the ‘betadisper’ function of vegan. Further, ANOVA was used on Log10(1+x)-transformed sequence counts of abundant phyla and families to identify fungal clades that differed significantly among soil depths or aspects, using an Benjamini–Hochberg correction to adjust p-values for false discovery. To identify OTUs that differed significantly among soil depths or between north- and south-facing aspects, unrarefied OTU tables were filtered to remove OTUs with <10 counts and DESeq2 (Love, Huber and Anders 2014) was used to contrast each successive soil depth (0 versus 10 cm, 10 versus 25 cm, 25 versus 50 cm, 50 versus 75 cm and 75 versus 100 cm), and north- and south- facing slopes. OTUs with an FDR-adjusted p-value <0.001 and a base mean >100 were considered differentially abundant. Observed richness (number of OTUs) and diversity (Shannon and inverse Simpson index) were calculated for each sample in vegan, and differences in richness and diversity among soil depths and aspects were evaluated using ANOVA. Further, correlations between richness and diversity with soil and environmental characteristics were evaluated using Spearman’s rank correlations. To assign putative functional characteristics to fungal OTUs, the Funguild script (Nguyen et al. 2016) was used. Among fungi that could be classified to a functional guild with ‘probable’ or ‘highly probable’ confidence, Kruskal–Wallis tests were performed to identify abundant groups that differed among soil depths. Finally, co-association networks among taxa at each depth or landscape position (aspect) were constructed using the sparCC algorithm implemented in the SPEIC-EASI package of R (Kurtz et al. 2015). Briefly, unrarefied OTU tables were filtered to include only samples belonging to each depth (n = 16 samples per depth) or aspect (n = 36 samples per aspect), then only OTUs found in at least 20% of samples with a total sequence count of >20 were retained for network inference. Co-associations with an R-value >0.5 and a p-value <0.01 after 100 bootstraps were considered significant. Network and node characteristics (graph density, clustering coefficient, node hub-scores) were evaluated with the igraph package in R (Csardi and Nepusz 2006), where the hub-score of a taxon reflects its structural importance in a network (Kleinberg 1999). Spearman correlations among taxon hub scores were used to compare the importance of taxa found in multiple networks among soil depths and aspects. RESULTS Illumina sequencing After data processing 2770 078 sequences were obtained among 96 samples with an average of 28855 +/− 9357 sequences per sample. After rarefaction (discarding three samples with <10000 sequences), sequences clustered into 678 OTUs belonging to seven identified phyla of which Ascomycota (472 OTUs or 87 +/- 14 % of sequences), Basidiomycota (105 OTUs or 7.8 +/- 13% of sequences) and Zygomycotya (29 OTUs, or 4.1 +/- 5.7 % of sequences) accounted for the vast majority of taxa. Minor groups included Cercozoa (1 OTU), Chytridiomycota (22 OTUs), Glomeromycota (7 OTUs) and Rozellomycota (3 OTUs). Fungal community composition varies with soil depth and landscape slope/aspect Fungal communities varied primarily with soil depth (Adonis r2 = 0.25, P = 0.001; Fig. 2), though members of the phylum Ascomycota were dominant among all depths (Fig. 2). In general, the prevalence of fungal phyla was similar across the depth profile, though, there was a distinct increase in the relative abundance of Basidiomycota at 10 cm below the soil surface (where the surface is 0 cm depth) versus at other strata (Fig. 3; ANOVA F = 8.44, P<0.0001) that corresponded with a decrease in the prevalence of Ascomycota (ANOVA F = 3.25, P = 0.01). At the family-level, 22 groups varied significantly with soil depth (FDR-adjusted p-values<0.05; Fig. 3), where members of some families tended to be more abundant at the soil surface (0 cm; eg. Glomerellaceae, Teratosphaeriaceae, Mycosphaerellaceae, Pleosporaceae, Pezizaceae, Lasiosphaeriaceae), some in shallow soil layers (10 and 25 cm; eg. Cephalothecaceae, Clavicipitaceae, Nectriaceae, Trichocomaceae, Chaetomiaceae, Hypicreaceae and Mycotrichaceae), and others in the deepest strata (50, 75 and 100 cm; eg. Bionectriaceae). Among OTUs with relative abundances that differed significantly with soil depth, the soil surface was dominated by OTUs identified as Sordariales, Fusarium, Mycosphaerella, Glarea and Humicola species, upper soil layers (10 and 25 cm) were dominated by Ascomycota, Nectriaceae, Hypocreales, Oidiodendron, Trichoderma and Penicillium species and the deepest soils were dominated by OTUs related to Bionectria, Ijuhya, Hypocreales, Mortierella and Ascomycota (Fig. 4). Thus, fungal community composition was highly stratified across soil depths, where some clades tended to dominate in shallow, intermediate or deep soil layers. Intriguingly, the variability in fungal community composition increased with depth (Fig. 2; ANOVA F = 4.34, P = 0.001); communities deeper in the soil were more different from one another than those nearer to the soil surface. Figure 2. View largeDownload slide NMDS plot (A) of fungal communities from different soil depths and landscape positions (stress=0.15) and the percentage of sequences belonging to fungal phyla among soil depths (B) and landscape positions (C). Distance to centroid at each depth is shown in D. Error bars represent standard deviations in D. Figure 2. View largeDownload slide NMDS plot (A) of fungal communities from different soil depths and landscape positions (stress=0.15) and the percentage of sequences belonging to fungal phyla among soil depths (B) and landscape positions (C). Distance to centroid at each depth is shown in D. Error bars represent standard deviations in D. Figure 3. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant fungal families that vary significantly (FDR p-value<0.05) among soil depths (cm) (triangle) or landscape position (circles). Dendogram on the top represents hierarchical clustering of fungal family abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 3. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant fungal families that vary significantly (FDR p-value<0.05) among soil depths (cm) (triangle) or landscape position (circles). Dendogram on the top represents hierarchical clustering of fungal family abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 4. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant OTUs (base mean among all samples >100) that differ significantly (FDR p-value<0.001) among soil depth (cm). Dendogram on the left represents hierarchical clustering of OUT abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Figure 4. View largeDownload slide Heatmap of log2(1+X)-transformed sequence counts of abundant OTUs (base mean among all samples >100) that differ significantly (FDR p-value<0.001) among soil depth (cm). Dendogram on the left represents hierarchical clustering of OUT abundance. Color scale represents range of log2(1+X)-transformed sequence counts. Landscape positio, based on slope and aspectn (north-facing slopes, south-facing slopes, toe-slope (bottom), and summit (top) also explained a significant portion of variation in fungal communities (Adonis r2 = 0.05, P = 0.001), however, only unidentified phyla differed significantly between aspects and were more abundant in south-facing versus north-facing slopes (F = 5.53, P = 0.002). At the family level, only three clades differed significantly with aspect (FDR p-value<0.05). Members of Chaetomiaceae tended to be more abundant in soils from north-facing slopes and toe-slopes, whereas members of Coniochaetaceae tended to be more abundant in soils from toe-slopes (Fig. 3). The relative abundance of taxa within Rustromiaceae was greatest in toe-slopes in the surface layers, yet was low in deeper soil depths. North facing positions had a higher relative abundance of OTUs related to Sordariales, Humicola, Lasiosphaeraceae, Cryptococcus and others (Fig. 5). In contrast, south facing positions had greater relative abundances of OTUs related to Phialophora, Lasiosphaeraceae, Sordariales, Mycosphaerella tassiana and unidentified Ascomycota (Fig. 5). Figure 5. View largeDownload slide Differentially abundant taxa between north-facing and south-facing slopes. Points to the right of 0 are more abundant on north-facing slopes, whereas those to the left of 0 are more abundant on south-facing slopes. The size of points is scaled by the mean abundance (baseMean) of that OTU and points are colored by the phylum to which they belong. Labels on the left indicate the nearest blast hit, blast % identity, and the OTU identifier. Figure 5. View largeDownload slide Differentially abundant taxa between north-facing and south-facing slopes. Points to the right of 0 are more abundant on north-facing slopes, whereas those to the left of 0 are more abundant on south-facing slopes. The size of points is scaled by the mean abundance (baseMean) of that OTU and points are colored by the phylum to which they belong. Labels on the left indicate the nearest blast hit, blast % identity, and the OTU identifier. Fungal diversity across soil depths and landscape slope/aspect Fungal richness and diversity declined significantly with increasing soil depth (ANOVA F = 108.7, 10.9 and 4.1; P<0.002 for observed richness, Shannon and inverse Simpsons indices; Fig. 6), though these indices did not differ significantly with aspect after excluding summit and toe slopes due to low replication (data not shown). Thus, fungal richness and diversity varied in concert with soil C (R = 0.81, 0.57, 0.38; P<0.001, for richness, Shannon, and Simpsons diversity, respectively), N (R = 0.80, 0.56, 0.37; P≤0.001, for richness, Shannon and Simpsons diversity, respectively) and pH (R = −0.85, −0.61, −0.39, P < 0.001, for richness, Shannon and Simpsons diversity, respectively). Moreover, although richness did not differ significantly between north- and south-facing slopes, fungal richness at the 10 cm depth was negatively related to global irradiance (R = −0.51, P = 0.04; Fig. 7). This pattern was marginally significant at the soil surface and 25 cm depth (R = −0.40, P = 0.13 and R = −0.44, P = 0.087 for 0 cm and 25 cm, respectively) though this pattern was not present at the deeper soil layers (50 cm: R = 0.06, P = 0.82; 75 cm: R = −0.04, P = 0.87; 100 cm: R = 0.20, P = 0.50). Thus, more intense solar irradiation may reduce fungal diversity within a landscape, though only near the soil surface. Figure 6. View largeDownload slide Relationship between soil depth and fungal richness, diversity and soil edaphic factors. Figure 6. View largeDownload slide Relationship between soil depth and fungal richness, diversity and soil edaphic factors. Figure 7. View largeDownload slide Relationships between fungal richness and global irradiance among soil depths (cm). Global irradiance expressed as MegaWatt-hours m-1 yr-1. Figure 7. View largeDownload slide Relationships between fungal richness and global irradiance among soil depths (cm). Global irradiance expressed as MegaWatt-hours m-1 yr-1. Variation in fungal life-history strategies among soil depths The relative abundance of fungi belonging to different common trophic modes (mean relative abundance >1%) differed among soil depths. Specifically, those taxa inferred as having pathotrophic-saprotrophic (Kruskal-Wallis test: chi-squared=20.5, P = 0.001), pathotrophic-symbiotrophic (Kruskal-Wallis test: chi-squared=11.0, P = 0.04) and symbiotrophic (Kruskal-Wallis test: chi-squared=22.7, P = 0.0004) lifestyles varied significantly among soil depths (Fig. 8). Taxa with pathotroph-symbiotroph or symbiotrophic trophic modes tended to be relatively more abundant at 10 cm below the soil surface, and those with pathotrophic-saprotrophic or pathotrophic-symbiotrophic abundance of different trophic modes did not vary significantly with landscape position (aspect), with the exception of saprotrophs, which tended to be more abundant in toe- and summit soils (Kruskal-Wallis test: chi-squared=8.9, P = 0.03). However, the difference in the frequency of saprotrophs was not significant between only north- and south- facing aspects (Kruskal-Wallis test: chi-squared=0.56, P = 0.45). Figure 8. View largeDownload slide Proportion of fungi classified to different trophic modes across soil depths (cm). Figure 8. View largeDownload slide Proportion of fungi classified to different trophic modes across soil depths (cm). Fungal co-occurrence patterns across soil depths Co-occurrence networks among fungal OTUs were strongly impacted by soil depth (Fig. 9; Supplemental Fig. S1). Specifically, there were fewer significant co-associations among a smaller number of OTUs as soil depth increased (Table 2), likely reflecting the lower fungal diversity in deeper soil layers. For example, there were significant correlations between the average taxon richness of communities at each depth and the number of network nodes (R = 0.99, P<0.001), positive edges (R = 0.99, P<0.001 ), negative edges (R = 0.99, P<0.001 ) and network density (R = −0.86, P = 0.03), but not with the positive:negative edge ratio or the clustering coefficient (R = −0.28, P = 0.59 and R = 0.05, P = 0.92, respectively). Similarly, there were few commonalities in the co-association patterns among taxa in networks from different depths, indicating vertical stratification of ecological dynamics and species interactions. However, the density of co-associations among taxa tended to increase with depth, suggesting that, despite lower diversity, fungal populations in deep soil strata are more likely to be driven by similar factors (eg. carbon availability) than those near the surface. In contrast, co-occurrence networks from north- and south- facing slopes were similar in size (number of nodes and edges) and had similar co-association patterns, though the co-association network among taxa from south-facing slopes was denser (Fig. 10, Table 2, Supplemental Fig. S2). Figure 9. View largeDownload slide Networks of positive co-occurrences among fungi from different sampling depths. Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from different depths (n = 16). Figure 9. View largeDownload slide Networks of positive co-occurrences among fungi from different sampling depths. Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from different depths (n = 16). Figure 10. View largeDownload slide Networks of positive co-occurrences among fungi from different aspects (landscape positions). Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from all depths for soil cores from north- or south-facing slopes (n = 36). Figure 10. View largeDownload slide Networks of positive co-occurrences among fungi from different aspects (landscape positions). Each point (nodes) represents a fungal taxon and points connected by lines (edges) are significantly positively associated. Nodes are colored by the phylum to which they belong. Networks were constructed using samples from all depths for soil cores from north- or south-facing slopes (n = 36). Table 2. Characteristics of fungal co-association networks at different soil depths and aspects. Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 View Large Table 2. Characteristics of fungal co-association networks at different soil depths and aspects. Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 Depth (cm) #Nodes (OTUs) #Positive edges #Negative edges Positive:negative ratio Graph density Clustering coefficient 0 215 587 349 1.68 0.041 0.29 10 134 245 203 1.21 0.05 0.301 25 109 204 132 1.55 0.057 0.361 50 39 33 13 2.54 0.062 0.158 75 22 14 7 2 0.091 0.632 100 15 5 4 1.25 0.086 0 Aspect North 81 151 59 2.56 0.065 0.409 South 73 168 53 3.17 0.084 0.557 View Large The taxa that occupied central positions in the networks differed among networks of taxa from different depths, though hub-scores of taxa were significantly correlated among those from 10 cm versus 25 cm (R = 0.295, P = 0.01), 25 cm and 50 cm (R = 0.53, P = 0.002), suggesting that fungi share similar network positions at these depths. At the surface layer (0 cm) , taxa most central to the co-occurrence network included Verrucariles, Cryptococcus, Knufia and Penicillium species (Supplemental Fig. S3), those at 10 cm included Ascomycota, Chaetomiaceae, Sordariomycetes and Lasiosphaeraceae species, those at 25 cm included Bionectria, Mortierella, Chaetomiaceae, Sordariomycetes and Humicola species, those at 50 cm included Hypocreales, Mortierella, Nectriaceae and Ascomycota species, those at 75 cm included Basidiomycota, Sordariales, Ascomycota, Nectriaceae and Microdochium speices and at 100 cm only Hypocreales, Bionectria and Mortierella had more than one significant co-association. In contrast, taxa among networks from north- and south- facing slopes has similar positions within each network (R = 0.80, P<0.001), where Humicola, Bionectria, Mortierella and unidentified Ascomycota species were most central within each network (Supplemental Fig. S4). DISCUSSION Soil depth was a strong driver of fungal community composition and diversity in no-till in this no-till wheat cropping system, consistent with studies of forest (Fierer, Schimel and Holden 2003; Baldrian et al. 2012), prairie (Jumpponen, Jones and Blair 2010) and other agricultural soils (Derpsch et al. 2010; Tojuet al. 2016). For example, Sipila et al., (Sipila, Yrjala and Alakukku 2012) used T-RFLP and PLFA to characterize microbial communities and found that moldboard plowing homogenizes communities in the upper layer (0–20 cm), whereas they were more stratified in no-till soils, though other studies of fallow fields have suggested that there is little difference in fungal community composition or diversity with soil depth (Ko et al.2017). Variation in fungal communities across depths often tracks stratification in C, N, pH and other soil characteristics (Fierer, Schimel and Holden 2003; Jumpponen, Jones and Blair 2010; Peay, Kennedy and Talbot 2016; Lamit et al. 2017), which is especially strong near the surface in agricultural fields under no-till versus those under conventional tillage (Zhao et al. 2015) It has been long understood that there is a succession of fungi on decomposing residue (Sadasivan 1939), since plant residue serves as a key resource base for fungi, stratification of fungal communities is likely to reflect different successional stages of residue decomposition. For example, the soil surface may have carbon derived from the standing crop residue and litter layer, and would have the greatest amounts of labile organic material. In parallel with studies of succession during decomposition, surface soil was dominated by many cellulolytic Ascomycetous fungi including those belonging to Sordariales, Coniochaeta (Lecythophora), Fusarium, Mycosphaerella, Humicola and Pleosporales, which are rapidly able to colonize wheat straw and are often found to be most abundant during early stages of decomposition (Harper and Lynch 1985; Maet al. 2013). The initial cellulolytic colonists of wheat residue, however, are estimated to only utilize a small portion of the available carbon (Harper and Lynch 1985). In later stages of decomposition of polymers (eg. cellulose), other taxa, such as Trichocladium (Chaetomiaceae), which tended to predominate at 10 cm depth, are expected to be more competitive (Poll et al. 2010). Although some residue may be incorporated by higher disturbance hoe-type no-till seeders, the roots of wheat and other rotation crops may also be significant sources of carbon at this depth. As soil depth increases and root density decreases, those taxa that dominated deeper soil layers (50 and 100 cm), related to Hypocreales, Mortierella, Ijuhya, Bionectria and unidentified Ascomycota species, are likely to be those able to degrade older, recalcitrant carbon sources (eg. lignin). Moreover, anecic earthworms (eg. Lumbricus terrestris), which are commonly present in no-till systems and can burrow to 3 m depth (Chan 2001, Aeschlimann, personal communication), may drag fresh surface residue into deep burrows, providing a significant source of C for microbial communities. Alternatively, fungi in deeper soil layers may act as pathogens or symbiotrophs in addition to saprotrophs as a means of obtaining nutrients, reflecting a shift in food web structures in deeper soil layers (Lindahl et al. 2007; Peay, Kennedy and Talbot 2016; Lamit et al. 2017). This is supported by the increase in the relative abundance of taxa assigned to trophic modes with pathogenic or symbiotrophic stages at deeper soil depths. In addition to shifts in fungal community composition with increasing soil depth, variation in soil characteristics (C, N, pH, moisture, temperature) are expected to impact the outcome of fungal interactions (Moore and Six 2015; Hiscox et al. 2016). Indeed, co-association patterns among fungal taxa, which are frequently used to infer species interaction or habitat preference (Poudelet al. 2016), differed considerably with soil depth. Although similar between networks from 0 and 10 cm, network hub taxa were often specific to a particular soil depth, suggesting that a given fungal taxa may play unique roles within communities at different soil depths. Moreover, the smaller, denser networks with increasing soil depth having distinct associations among taxa may jointly reflect fungal density and diversity, niche partitioning and the frequency of competitive or cooperative interactions (Poudel et al. 2016; Toju et al.2016). For example, the larger, more distributed networks at the upper 0 and 10 cm depths may result from the combination of greater microbial biomass and activity in the upper soil (Fierer, Schimel and Holden 2003), resulting in a more significant role of direct and indirect species interactions for fungal fitness. In contrast, in deep soil layers, where there is low resource availability and low microbial biomass, competitive species interactions among fungi may have little impact on biological fitness. Alternatively, the fungi detected in deep soil layers may exist primarily as inactive spores or represent ‘relic’ DNA (Cariniet al. 2016). Similar patterns in fungal co-occurrence networks, however, were described recently in forest systems by Toju et al. (2016) where fungal networks from different soil horizons differed in complexity. Together, the substantial variation in fungal co-occurrence networks suggests that variation in biotic and abiotic characteristics across the soil profile generate distinct ecological dynamics among soil fungi among soil layers. Fungal richness and diversity declined with soil depth in concert with decreasing C, N, and with increasing pH. However, there were still on average ∼24 OTUs found at 75 cm and 100 cm depths, consistent with previous observations of considerable fungal diversity in deep soils (Jumpponen, Jones and Blair 2010; Toju et al. 2016). In the upper soil layers, the small but significant negative associations between irradiance and fungal richness suggest that, despite relatively close proximity (<1 km), greater amounts of annual solar irradiance within a field can result in reduced fungal diversity in the upper 25 cm of soil. This may also reflect soil water levels, since south facing slopes with greater solar irradiance have more evapotranspiration. Finally, a unique finding of this study was that variation in fungal community composition increased with soil depth. The higher variance in fungal community composition with increasing soil depth may mirror a greater patchiness in soil C or other resources as depth increases, perhaps due to a greater reliance on resources provided by deep roots or earthworms (Shuster, Subler and McCoy 2001). Alternatively, more frequent dispersal of fungal propagules in the upper soil layers, perhaps facilitated by growing plant roots or earthworm/mesofaunal activity, may contribute to more homogeneous fungal communities near the soil surface. Contrary to our expectations that fungal communities would differ considerably between north- and south- facing aspects, only a small number of taxa was associated with one or the other aspects. Moreover, fungal co-occurrence networks from north- and south- facing slopes were strikingly similar, suggesting similar community dynamics at these aspects. Only small differences between north- and south- facing slopes may be due to the homogeneity of the previous crop species (wheat), since plant species identity can have a strong impact on fungal communities in soil (Peay, Baraloto and Fine 2013; LeBlanc, Kinkel and Kistler 2017). Moreover, consistent management practices, such as fertilizer treatments or mechanical disturbance may play a role in homogenizing the fungal community in upper soil layers (Weber, Vilgalys and Kuske 2013), whereas lower soil depths may not be strongly impacted by differences in temperature or water due to landscape position. Finally, samples were taken early in the season (April) when soils were likely to be more uniformly moist, and differences in communities due to aspect may only be significant for a small time period later on during the growing season when there are strong gradients in soil water or temperature. However, even though difference in community composition between north- and south-facing slopes were minor, there may be differences in the biomass or density of fungi that we were unable to detect with relative abundance data. In total, this work provides significant insight into variation in depth profiles of fungal communities across the landscape in a no-till wheat cropping system. Although dead roots and earthworm activity may be a significant source of nutrients for fungi in deep soils, differences in fungal communities among soil depths likely reflects the composition and quantity of available carbon sources, where fungal communities follow a pattern of succession during residue degradation. Moreover, stratification of fungal communities across the soil depth profile is likely to result in distinct ecological dynamics at each stratum, as exemplified by differences in co-association networks. However, there were only small differences in fungal communities between aspects, perhaps due to homogenization of communities by agricultural management practices. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. ACKNOWLEDGEMENTS Author Contributions: All authors discussed the results and edited and commented on the manuscript. TP and DCS designed the project. TP, DCS and KK collected the soil samples. DCS performed all the molecular sequencing and data analysis. BC provided metadata. DRH provided the study site, including long-term field work, maintenance, metadata, and LTAR funding and direction. DCS and TP wrote the manuscript. FUNDING Funding from USDA-ARS Postdoctoral Research Associate Award to the first author and REACCH (Regional Approaches to Climate Change- Pacific Northwest Agriculture) award #2011-68002-30191 from USDA National Institute for Food and Agriculture. Conflict of interest. None declared. REFERENCES Akob DM , Kusel K . Where microorganisms meet rocks in the Earth’s Critical Zone . Biogeosciences . 2011 ; 8 : 3531 – 43 . Google Scholar CrossRef Search ADS Baldrian P , Kolařík M , Štursová M , et al. Active and total microbial communities in forest soil are largely different and highly stratified during decomposition . ISME J . 2012 ; 6 : 248 – 58 . Google Scholar CrossRef Search ADS PubMed Blackmore S , Godwin RJ , Fountas S , et al. The analysis of spatial and temporal trends in yield map data over six years . Biosystems Engineering . 2003 ; 84 : 455 – 66 . Google Scholar CrossRef Search ADS Carini P , Marsden PJ , Leff JW et al. Relic DNA is abundant in soil and obscures estimates of soil microbial diversity . Nat Microbiol . 2016 ; 2 : 16242 . Google Scholar CrossRef Search ADS PubMed Chan K . An overview of some tillage impacts on earthworm population abundance and diversity — implications for functioning in soils . Soil Tillage Res . 2001 ; 57 : 179 – 91 . Google Scholar CrossRef Search ADS Csárdi G , Nepusz T . The igraph software package for complex network research . InterJournal . 2006 ; 1695 . Degrune F , Theodorakopoulos N , Dufrêne M , et al. No favorable effect of reduced tillage on microbial community diversity in a silty loam soil (Belgium) . Agric Ecosyst Environ . 2016 ; 224 : 12 – 21 . Google Scholar CrossRef Search ADS Derpsch R , Friedrich T , Kassam A , et al. Current status of adoption of no-till farming in the world and some its main benefits . Int J Agric Biol Eng . 2010 ; 3 , DOI: https://doi.org/10.3965/j.issn.1934-6344.2010.01.0-0 . Edgar RC . UPARSE: highly accurate OTU sequences from microbial amplicon reads . Nat Methods . 2013 ; 10 : 996 – 8 . Google Scholar CrossRef Search ADS PubMed Ekelund F , Rønn R , Christensen S . Distribution with depth of protozoa, bacteria and fungi in soil profiles from three Danish forest sites . Soil Biol Biochem . 2001 ; 33 : 475 – 81 . Google Scholar CrossRef Search ADS Fan JL , McConkey B , Wang H , et al. Root distribution by depth for temperate agriculture crops . Field Crops Research . 2016 ; 189 : 68 – 74 . Google Scholar CrossRef Search ADS Fierer N , Schimel JP , Holden PA . Variations in microbial community composition through two soil depth profiles . Soil Biol Biochem . 2003 ; 35 : 167 – 76 . Google Scholar CrossRef Search ADS Harper SHT , Lynch JM . Colonization and decomposition of straw by fungi . Trans Br Mycol Soc . 1985 ; 85 : 655 – 61 . Google Scholar CrossRef Search ADS Hawksworth DL . The magnitude of fungal diversity: the 1.5 million species estimate revisited . Mycol Res . 2001 ; 105 : 1422 – 32 . Google Scholar CrossRef Search ADS Hipple KW . Washington Soil Atlas , Washington, DC , USDA-NRCS Publication , 2011 . Hiscox J , Clarkson G , Savoury M , et al. Effects of pre-colonisation and temperature on interspecific fungal interactions in wood . Fungal Ecol . 2016 ; 21 : 32 – 42 . Google Scholar CrossRef Search ADS Horner GM , McCall AG , Bell FG . Investigations in erosion control and reclamation of eroded land at the Palouse Conservation Experiment Station 1931–1942 , U.S. Department of Agriculture, Soil Conservation Service, Technical Bulletin 860 . 1948 . Huggins DR , Reganold JP . No-till: the quiet revolution . Sci Am . 2008 ; 299 : 70 – 7 . Google Scholar CrossRef Search ADS PubMed Huggins DR , Uberuaga DP . Field heterogeneity of soil organic carbon and relationships to soil properties and terrain attributes . Climate friendly farming: Center for Sustaining Agriculture and Natural Resources Research Report 2010–001 . 2010 . Available at http://csanr.wsu.edu/pages/Climate_Friendly_Farming_Final_Report/. Jumpponen A , Jones KL , Blair J . Vertical distribution of fungal communities in tallgrass prairie soil . Mycologia . 2010 ; 102 : 1027 – 41 . Google Scholar CrossRef Search ADS PubMed Kleinberg JM . Authoritative sources in a hyperlinked environment . J ACM . 1999 ; 46 : 604 – 32 . Google Scholar CrossRef Search ADS Ko D , Yoo G , Yun S-T , et al. Bacterial and fungal community composition across the soil depth profiles in a fallow field . J Ecol Environ . 2017 ; 41 : 34 . Google Scholar CrossRef Search ADS Kristensen KJ . Temperature and heat balance of soil . Oikos . 1959 ; 10 : 103 . Google Scholar CrossRef Search ADS Kurtz ZD , Müller CL , Miraldi ER , et al. Sparse and compositionally robust inference of microbial ecological networks . PLOS Comput Biol . 2015 ; 11 : e1004226 . Google Scholar CrossRef Search ADS PubMed Lamit LJ , Romanowicz KJ , Potvin LR , et al. Patterns and drivers of fungal community depth stratification in Sphagnum peat . FEMS Microbiol Ecol . 2017 ; 93 , DOI: https://doi.org/10.1093/femsec/fix082 . LeBlanc N , Kinkel L , Kistler HC . Plant diversity and plant identity influence Fusarium communities in soil . Mycologia . 2017 ; 109 : 128 – 39 . Google Scholar CrossRef Search ADS PubMed LeBlanc N , Kinkel LL , Kistler HC . Soil fungal communities respond to grassland plant community richness and soil edaphics . Microb Ecol . 2015 ; 70 : 188 – 95 . Google Scholar CrossRef Search ADS PubMed Lindahl BD , Ihrmark K , Boberg J , et al. Spatial separation of litter decomposition and mycorrhizal nitrogen uptake in a boreal forest . New Phytol . 2007 ; 173 : 611 – 20 . Google Scholar CrossRef Search ADS PubMed Love MI , Huber W , Anders S . Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome Biol . 2014 ; 15 : 550 . Google Scholar CrossRef Search ADS PubMed Ma A , Zhuang X , Wu J , et al. Ascomycota members dominate fungal communities during straw residue decomposition in arable soil . PLoS ONE . 2013 ; 8 : e66146 . Google Scholar CrossRef Search ADS PubMed Mahler RL , Bezdicek DF . Diversity of Rhizobium legumosarium in the Palouse of eastern Washington . Applied and Environmental Microbiology . 1978 ; 36 : 780 – 2 . Google Scholar PubMed Moore ML , Six DL . Effects of temperature on growth, sporulation, and competition of mountain pine beetle fungal symbionts . Microb Ecol . 2015 ; 70 : 336 – 47 . Google Scholar CrossRef Search ADS PubMed Nagle GN , Ritchie JC . Wheat field erosion rates and channel bottom sediment sources in an intensively cropped northeastern Oregon drainage basin . Land Degradation and Development . 2004 ; 15 : 15 – 26 . Google Scholar CrossRef Search ADS Nguyen NH , Song Z , Bates ST , et al. FUNGuild: An open annotation tool for parsing fungal community datasets by ecological guild . Fungal Ecol . 2016 ; 20 : 241 – 8 . Google Scholar CrossRef Search ADS Oksanen J , Blanchette FG , Friendly M et al. Vegan: Community Ecology Package . 2016 . https://cran.r-project.org Peay KG , Baraloto C , Fine PV . Strong coupling of plant and fungal community structure across western Amazonian rainforests . ISME J . 2013 ; 7 : 1852 – 61 . Google Scholar CrossRef Search ADS PubMed Peay KG , Kennedy PG , Talbot JM . Dimensions of biodiversity in the Earth mycobiome . Nat Rev Microbiol . 2016 ; 14 : 434 – 47 . Google Scholar CrossRef Search ADS PubMed Poll C , Brune T , Begerow D , et al. Small-scale diversity and succession of fungi in the detritusphere of rye residues . Microb Ecol . 2010 ; 59 : 130 – 40 . Google Scholar CrossRef Search ADS PubMed Poudel R , Jumpponen A , Schlatter DC , et al. Microbiome networks: A systems framework for identifying candidate microbial assemblages for disease management . Phytopathology . 2016 ; 106 : 1083 – 96 . Google Scholar CrossRef Search ADS PubMed Powell JR , Karunaratne S , Campbell CD , et al. Deterministic processes vary during community assembly for ecologically dissimilar taxa . Nature Communications . 2015 ; 6 : 8444 . Google Scholar CrossRef Search ADS PubMed Raaijmakers JM , Paulitz TC , Steinberg C , et al. The rhizosphere: a playground and battlefield for soilborne pathogens and beneficial microorganisms . Plant Soil . 2009 ; 321 : 341 – 61 . Google Scholar CrossRef Search ADS Rognes T , Flouri T , Nichols B , et al. . VSEARCH: a versatile open source tool for metagenomics . PeerJ . 2016 ; 4 : e2584 . Google Scholar CrossRef Search ADS PubMed Sadasivan TS . Succession of fungi decomposing wheat straw in different soils, with special reference to Fusarium culmorum . Ann Appl Biol . 1939 ; 26 : 497 – 508 . Google Scholar CrossRef Search ADS Schillinger WF , Papendick RI . Then and Now: 125 years of dryland wheat farming in the inland Pacific Northwest . Agron J . 2008 ; 100 : S – 166 . Google Scholar CrossRef Search ADS Schulz S , Brankatschk R , Dümig A , et al. The role of microorganisms at different stages of ecosystem development for soil formation . Biogeosciences . 2013 ; 10 : 3983 – 96 . Google Scholar CrossRef Search ADS Sharma-Poudyal D , Schlatter D , Yin C , et al. Long-term no-till: A major driver of fungal communities in dryland wheat cropping systems . PLOS ONE . 2017 ; 12 : e0184611 . Google Scholar CrossRef Search ADS PubMed Sharratt BS , Wendling L , Feng G . Surface characteristics of a wind blown soil altered by tillage intensity during summer fallow . Aeol Res . 2012 ; 5 : 1 – 7 . Google Scholar CrossRef Search ADS Shuster W , Subler S , McCoy E . Deep-burrowing earthworm additions changed the distribution of soil organic carbon in a chisel-tilled soil . Soil Biol Biochem . 2001 ; 33 : 983 – 96 . Google Scholar CrossRef Search ADS Sipila TP , Yrjala K , Alakukku L , et al. Cross-site soil microbial communities under tillage regimes: Fungistasis and microbial biomarkers . Appl Environ Microbiol . 2012 ; 78 : 8191 – 8201 . Google Scholar CrossRef Search ADS PubMed Sosa-Hernández MA , Roy J , Hempel S , et al. Subsoil arbuscular mycorrhizal fungal communities in arable soil differ from those in topsoil . Soil Biology and Biochemistry . 2018 ; 117 : 83 – 86 . Google Scholar CrossRef Search ADS Swift MJ , Heal OW , Anderson JM . Decomposition in Terrestrial Ecosystems . Berkeley : University of California Press , 1979 . Taylor DL , Hollingsworth TN , McFarland JW , et al. A first comprehensive census of fungi in soil reveals both hyperdiversity and fine-scale niche partitioning . Ecol Monogr . 2014 ; 84 : 3 – 20 . Google Scholar CrossRef Search ADS Tedersoo L , Bahram M , Polme S , et al. Global diversity and geography of soil fungi . Science . 2014 ; 346 : 1256688 . Google Scholar CrossRef Search ADS PubMed Toju H , Kishida O , Katayama N , et al. Networks depicting the fine-scale co-occurrences of fungi in soil horizons . PLOS ONE . 2016 ; 11 : e0165987 . Google Scholar CrossRef Search ADS PubMed Weber CF , Vilgalys R , Kuske CR . Changes in fungal community composition in response to elevated atmospheric CO2 and nitrogen fertilization varies with soil horizon . Front Microbiol . 2013 ; 4 , DOI: https://doi.org/10.3389/fmicb.2013.00078 . Zhang J , Kobert K , Flouri T , et al. PEAR: a fast and accurate Illumina Paired-End reAd mergeR . Bioinformatics . 2014 ; 30 : 614 – 20 . Google Scholar CrossRef Search ADS PubMed Zhang Y , Dong S , Gao Q , et al. Soil bacterial and fungal diversity differently correlated with soil biochemistry in alpine grassland ecosystems in response to environmental changes . Sci Rep . 2017 ; 7 : 43077 . Google Scholar CrossRef Search ADS PubMed Zhao X , Xue J-F , Zhang X-Q , et al. Stratification and storage of soil organic carbon and nitrogen as affected by tillage practices in the north China plain . PLOS ONE . 2015 ; 10 : e0128873 . Google Scholar CrossRef Search ADS PubMed Published by Oxford University Press on behalf of FEMS 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US.

Journal

FEMS Microbiology EcologyOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off