Dryland forest management alters fungal community composition and decouples assembly of root- and soil-associated fungal communities

Soil Biology & Biochemistry 109 (2017) 14e22

Contents lists available at ScienceDirect

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

Dryland forest management alters fungal community composition and decouples assembly of root- and soil-associated fungal communities Yong Zheng a, b, Hang-Wei Hu a, Liang-Dong Guo b, Ian C. Anderson a, Jeff R. Powell a, * a b

Hawkesbury Institute for the Environment, Western Sydney University, Locked Bag 1797, Penrith, NSW, 2751, Australia State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, Beijing, 100101, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 June 2016 Received in revised form 10 January 2017 Accepted 31 January 2017

Land management practices considerably influence ecosystem processes and functioning, particularly in dryland ecosystems where nutrient and water limitations have direct (via effects on productivity) and indirect (via effects on soil biota) effects on ecosystem properties. Fungi in soils and associated with roots play critical roles in soil nutrient cycling and plant nutrient acquisition, but their responses to land management practices in dryland ecosystems remain equivocal. Here we evaluate the responses of fungal communities in roots and soils associated with a Eucalyptus saligna plantation after six years of forest management practices (irrigation and fertilisation) and in two different microenvironments within treated plots, in the presence and absence of understorey grasses. We observed that the richness and evenness of fungal communities were higher in soil than in root samples, but these two parameters did not vary among any of the management treatments. Effects of fertilisation and irrigation on fungal community composition were observed and appeared to be related to variation in soil pH, moisture, and nitrogen availability. Both fertilisation and irrigation decreased the ratios of ectomycorrhizal fungi to total fungi and increased the frequencies of saprotrophic and/or plant pathogenic fungi. We observed that some OTUs were shared between soil and root-associated fungal communities but that fertilisation was associated with lower frequencies of shared OTUs, suggesting a decoupling of these communities. In the absence of grasses, where only tree roots were present, we observed fewer tight relationships between fungal occurrence in root and soil samples. Our findings highlight the importance of forest management practices for fungal community assembly processes in dryland ecosystems, which may have consequences for the predictability of fungal community dynamics and nutrient cycling. © 2017 Published by Elsevier Ltd.

Keywords: 454 pyrosequencing Alpha diversity Beta diversity Community assembly Dryland Fertilisation Irrigation

1. Introduction Drylands, including arid and semi-arid ecosystems, are widely distributed around the world and cover c. 41% of Earth's terrestrial surface (Schimel, 2010). These ecosystems are characterized by low levels of soil moisture and nutrients owing to low precipitation and high evaporation (Delgado-Baquerizo et al., 2013), and are considered to be extremely vulnerable to anthropogenic perturbations and climate change (Singh et al., 2010; Martins et al., 2015). Microbial communities, particularly those associated with plant roots and rhizospheric soil, play a key role in ecosystem functions including biogeochemical cycling and affect plant growth and tolerance to biotic and abiotic stresses (Philippot et al., 2013; Singh

* Corresponding author. E-mail address: [email protected] (J.R. Powell). http://dx.doi.org/10.1016/j.soilbio.2017.01.024 0038-0717/© 2017 Published by Elsevier Ltd.

et al., 2014). Given the critical importance of microbes to ecosystem services and the continuing expansion of drylands in many regions, it is important to generate an understanding of how microbes in dryland ecosystems respond to environmental shifts (Johnson et al., 2012; Hu et al., 2015). Fungi fill important roles in carbon (C) cycling and plant nutrition in forest soils through their functioning as decomposers and mutualists (Tedersoo et al., 2014). Plant parasitic fungi can cause a range of plant diseases and negatively affect forest production, while mycorrhizal fungi facilitate water and mineral nutrient uptake by plants, and thus enhance fitness of plants experiencing environmental stress (Smith and Read, 2008). Therefore, understanding the responses of belowground fungal diversity and community composition in forest ecosystems facing environmental manipulation may aid sustainable forest management. This is particularly true if the disturbance influences taxa that are involved in regulating functional aspects of these ecosystems. Soil fungal

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

diversity and community composition have been observed to respond to environmental variation associated with the plant community present (Peay et al., 2013) and fertilisation (He et al., 2008), while a global-scale study reported that soil fungal diversity could be largely explained by mean annual precipitation and soil calcium (Ca) concentrations across various ecosystems (Tedersoo et al., 2014). Therefore, fungi in managed dryland forests are expected to respond to practices associated with nutrient and/ or water inputs, and understanding how these community responses are related to altered soil properties is essential for the prediction of outcomes related to plant nutrition, plant-soil-water interactions, and soil carbon cycling in other systems in other systems. In addition, how fungal communities respond to management may depend on where they are observed given that soil- and rootassociated fungal communities may be governed by differing assembly processes (Beck et al., 2015). Fungal communities associated with roots may be made up of members recruited from the surrounding soil fungal community (Danielsen et al., 2012), but there are conflicting observations in that root-associated fungal communities may be less diverse (Goldmann et al., 2016) or considerably more species-rich (Saks et al., 2014; Beck et al., 2015) than the surrounding soil fungal community. While the reasons behind these differences in fungal community assembly in soil versus roots still needs to be determined, they may have to do with the direct roles that root-associated fungi play in plant nutrition and fitness maintenance while soil-associated fungi influence plant nutrition and fitness indirectly via their effects on organic matter turnover and nutrient cycling. Fertilisers are extensively applied in pursuit of higher crop yield, forage production, and wood output in agricultural, grassland, and forest ecosystems, respectively. Increased soil fertility alters the diversity, composition, and productivity of the aboveground plant community (Magnani et al., 2007; Fornara and Tilman, 2012) as well as some aspects of the belowground fungal community (Bradley et al., 2006; Veresoglou et al., 2012; Weber et al., 2013; Li et al., 2015; Nielsen et al., 2015; Zheng et al., 2016). These impacts on fungi may be due to direct effects on edaphic properties (Zheng et al., 2014) or indirect effects via changes in the abundance and/or composition of associated plant communities (Liu et al., 2012). Irrigation is also a common practice, especially in dryland systems, to alleviate water stress and promote productivity, with effects on edaphic properties and plant communities (Hawkes et al., 2011; Zhou et al., 2013; Li et al., 2015). Soil fungal communities are expected to respond to changes in water availability. Nutrient diffusion is enhanced under increased water availability, which may reduce the dependence of host plants on mycorrhizal fungal mycelial uptake of these nutrients, and even
, 2001). Water addition may also indirectly influence of water (Auge fungi via changes in plant community abundance and composition (Hawkes et al., 2011; Cregger et al., 2012; Li et al., 2015). Shifts in the aboveground plant community, together with changes in belowground resource availability due to altered plant litter, exudates, and soil characteristics may impact both mycorrhizal and saprotrophic fungi (Husband et al., 2002; Waldrop et al., 2006; Hawkes et al., 2011; Li et al., 2015). However, relatively little is known about the impacts of altered precipitation regimes on belowground dryland fungal communities compared to our current knowledge of plant responses and other microbes (Cregger et al., 2012; Zhang et al., 2013; Nielsen and Ball, 2015). The main objective of this study was to explore the direct and indirect responses of fungal communities in dryland forests to long-term (six years) forestry management practices, namely fertilisation and irrigation. We sampled fungal communities from soil and roots in a dryland Eucalyptus saligna Sm. plantation. To evaluate

15

the direct (via edaphic properties) and indirect (via plant responses) effects, we sampled soil and roots from patches that contained understorey grasses or were bare and only influenced by the tree roots and the management treatments. We hypothesized that management strategies would have strong effects on rootassociated fungal communities, especially in patches with understorey grasses, because of the potential for both direct effects of edaphic properties and indirect effects associated with plant responses. Alternatively, root-associated fungal communities may be less affected by the treatments because the root environment already presents a strong environmental filter and altered edaphic properties contribute little to subsequent variation in rootassociated fungal communities, leading to a decoupling of soiland root-associated fungal communities (Beck et al., 2015). We also speculated that irrigation and fertilisation would have strong interactive effects in this ecosystem due to both water (Colombo et al., 2016) and soil fertility (especially phosphorus [P]; Nielsen et al., 2015) being limiting in local soils. 2. Materials and methods 2.1. Site and experimental design The experimental field is located at the Hawkesbury Forest Experiment site (33
360 4000 S, 150
440 26.500 E) in Richmond, NSW, Australia. The experiment consists of a series of plots planted with Eucalyptus saligna at a density of 1000 trees ha1. The plots also contain an understorey of grasses, with the space between tree rows dominated by Eragrostis curvula (Schrad.) Nees, Microlaena stipoides (Labill.) R. Br., and Elymus repens (L.) Gould, as well as smaller quantities of Digitaria sanguinalis (L.) Scop., Setaria incrassata (Hochst.) Hack., Chloris truncata R. Br., and Dactylis glomerata L. (Frew et al., 2013). Grasses are largely absent within tree rows as a result of a relatively dense layer of leaf litter and bark. Roundup was sprayed within tree rows, but not between rows, occasionally during the first two years of the study to facilitate tree establishment. The soil is a sandy loam and is characterized by low water holding capacity, low organic matter content (0.7%), low nitrogen ([N] < 1 mg kg1) and P (8 mg kg1) concentrations (Barton et al., 2010). Mean annual temperature at this site is 17
C, and the mean annual precipitation is 801 mm. The ratio of precipitation to evapotranspiration at the site is 0.6; therefore, the site is classified as a dry sub-humid environment under UNEP classification (Millennium Ecosystem Assessment, 2005). The field experiment consisted of four management treatments (fertilisation and ambient rainfall [F], fertilisation plus irrigation [IF], irrigation without additional fertiliser [I], and a control with ambient rainfall and without fertilisation [Control]), applied to experimental plots (38.5  41.6 m) in a randomized complete block design with four replicates for each treatment (resulting in 16 independent plots). Each plot was planted with 160 Eucalyptus saligna in 10 rows in April 2007, and experimental treatments were applied to the whole plot. At planting, 50 g of diammonium phosphate starter blend (N 15.3%, P 8.0%, potassium [K] 16.0%, sulphur [S] 7.7%, and Ca 0.3%) was applied to each tree to promote establishment. The Control treatment received no additional fertiliser or water. The first fertilisation in the F and IF treatments was undertaken with a solid N fertilizer (N 20.6%, P 3.0%, K 7.5%, S 3.8%, and Ca 4.4%) at a rate of 25 kg N ha1 year1 in January 2008. Solid N fertiliser (N 21.6%, P 8.1%, K 12.0%, and S 0.6%), at a rate of 150 kg N ha1 year1, was applied uniformly to the F treatment beginning in October 2008. The IF treatment received liquid fertiliser at a rate of 150 kg N ha1 year1 (Nutrifeed19 and Liquid N, Amgrow Fertilisers, Lidcombe, NSW, Australia), beginning in October 2008. In both the I and IF treatments, grey water (pH 8.8, total N 0.6 mg L1,

16

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

and total P 3.0 mg L1) was supplied at a rate of 15 mm every 4 days from September to April, and 7e10 mm every 4 days from April to September. Therefore, the irrigated plots received c. 1000 mm year1 plus the ambient rainfall. Management practices were chosen based on how treatments would be applied under actual field conditions, hence the use of solid and liquid forms of fertiliser in the two irrigation treatments. The detailed information on treatments was described by Frew et al. (2013). 2.2. Root and soil sampling and soil physicochemical analysis Plant root and soil samples were collected from the 16 plots in April 2012. Three trees were selected from each plot along a diagonal line through the centre of the plot. Around each tree, one soil core (2.5 cm in diameter, 15 cm in depth) was collected 1 m on each side of the tree trunk within the tree row where grasses were absent; the two samples adjacent to each tree were combined into a composite sample (three composite samples per plot; Fig. S1). Two more samples were collected 2 m on either side of each selected tree, extending into the space between tree rows where grasses were present and combined into a composite sample (a further three composite samples per plot), resulting in a total of 96 composite samples across the 16 plots (Fig. S1). At the time of sampling, the canopy was closed and extended over all soil coring points, regardless of whether they were taken from within or between tree rows. Samples were transported on ice to the laboratory and stored at 4
C until processed within five days. Fine roots (<1 mm diameter) were separated from soil by gentle brushing, while soil was sieved through a 2 mm mesh. Both plant root and soil samples were stored at 80
C prior to DNA extraction. Soil physicochemical property data were adopted from the same experimental field as measured by Hu et al. (2015), in which the soil samples were collected in November 2012. Briefly, soil pH was determined using a soil-to-water ratio of 1:2.5 with a Delta pHmeter (Mettler-Toledo Instruments, Columbus, OH, USA). Soil moisture was measured by oven-drying the samples at 105
C for  24 h. Soil ammonium (NHþ 4 eN) and nitrate (NO3 eN) were extracted with 2 M KCl solution and determined by a SEALAQ2 Analyzer (SEAL Analytical, Maquon, WI, USA). Soil total C and total N were measured on a LECO macro-CN analyzer (LECO, St Joseph, MI, USA). 2.3. DNA extraction, PCR and 454-pyrosequencing Soil DNA was extracted from 0.5 g of frozen soil samples using MoBio Ultra Clean® Soil DNA Isolation Kit (San Diego, CA, USA) according to the manufacturer's protocol. After carefully removing visible soil particles and rinsing with sterile water, the roots were cut into small pieces (<1 mm) using sterile scissors. Genomic DNA was extracted from 50 mg of frozen roots with a DNeasy Plant Mini Kit (Qiagen, Crawley, UK) and the DNA concentration was fluorometrically quantified using the Qubit dsDNA HS Assay Kit (Invitrogen, Carlsbad, CA, USA) according to manufacturer's instructions. Equal volumes of DNA were pooled at the plot level for each sample type-origin combination (i.e. root e grasses present, root e grasses absent, soil e grasses present, and soil e grasses absent), resulting in a total of 64 composite DNA samples (Fig. S1). DNA was diluted by five fold in sterile water prior to PCR amplification. The fungal large subunit (LSU; 28S rRNA gene) region was amplified using the primers of LR0R (ACCCGCTGAACTTAAGC, Cubeta et al., 1991) and LR3 (CCGTGTTTCAAGACGGG, R. Vilgalys Website: http://sites.biology.duke.edu/fungi/mycolab/primers.htm) linked with adaptors A (50 -CCATCTCATCCCTGCGTGTCTCCGAC-30 ) and B (50 -CCTATCCCCTGTGTGCCTTGGCAGTC-30 ), respectively. A 10-base

barcode (MID) sequence was inserted between adaptor A and the LR0R primer (Table S1). The 25 mL PCR cocktail consisted of 2.5 mL of 10  buffer, 1.5 mL of Mg2þ (50 mM), 4.0 mL of dNTP mixture (each 2.5 mM), 1.0 mL of each primer (10 mM), 2.5 U BioTaq DNA polymerase (Bioline, London, UK), and 1.0 mL (c. 5 ng) of template DNA. Thermal cycling conditions were as follows: 95
C for 6 min, 34 cycles of 95
C for 40 s, 55
C for 30 s and 72
C for 1 min, followed by 72
C for 10 min. The PCR products were purified using the Wizard SV Gel and PCR Clean-Up System (Promega, San Luis Obispo, CA, USA), and quantified using the Qubit dsDNA HS Assay Kit (Invitrogen, Carlsbad, CA, USA). A total of 50 ng of DNA from each sample was pooled and adjusted to 10 ng mL1 as required for pyrosequencing on a Roche Junior Genome Sequencer System (454 Life Sciences, Branford, CT, USA) using Titanium 454 chemistry by the Western Sydney University Next Generation Sequencing Facility. 2.4. Bioinformatics analysis Raw sequences generated from pyrosequencing have been deposited in the NCBI Sequence Read Archive under BioSample Accessions SAMN06204527-SAMN06204590. Sequences were assigned to individual samples in Mothur 1.32.2 (Schloss et al., 2009) based on the sample-specific barcodes, and trimmed (parameters: minlength ¼ 300; maxambig ¼ 0; maxhomop ¼ 10; qwindowaverage ¼ 20; pdiffs ¼ 1) to exclude short and low-quality reads. Denoising of the sequences was performed using Acacia (Bragg et al., 2012). The output file was further checked for potential chimeras using the chimera.uchime command in Mothur. The resultant dataset was aligned against the template of silva.eukarya.fasta (version 2014-03-12_id161395), and the sequences classified as fungi were extracted using the classify.seqs command. The number of sequences per sample was normalized to the smallest sample size (abundance-based rarefaction) using the sub.sample command. We also performed coverage-based rarefaction (Chao and Jost, 2012), which produced qualitatively similar results to those obtained from abundance-based rarefaction (Table S2); the outcome of analyses following abundance-based normalization are reported below. The sequences were clustered into operational taxonomic units (OTUs) at a 97% similarity level using the hcluster command in Mothur using the furthest neighbour algorithm. The UPARSE pipeline (Edgar, 2013) was also used to generate an OTU-sample matrix, which led to qualitatively similar results (Table S3) to those obtained after sequence processing using Mothur, suggesting that our inferences are not strongly sensitive to the method used for DNA sequence processing; analyses using OTUs generated using Mothur are reported below. Representative sequences from OTUs were picked through the command of get.oturep, and was identified by a basic local alignment search tool (BLAST; Altschul et al., 1990) search against the SILVA database (Pruesse et al., 2007) and the UNITE (unite.ut.ee) database. All sequences classified as fungi and the corresponding OTU table were exported for downstream analyses. In addition, the OTUs identified to the genus level were further assigned to trophic modes using FUNGuild (Nguyen et al., 2016) when a positive match to the database was observed. Briefly, (1) symbiotroph represent the fungi receiving nutrients by exchanging resources with host, and specify only ectomycorrhizal (EcM) fungi in this study; (2) saprotrophic and pathotrophic fungi which receive nutrients from dead or alive host cells; and (3) unassigned fungus portion (Tedersoo et al., 2014; Nguyen et al., 2016). 2.5. Data analysis All the analyses of fungal diversity and community composition

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

were conducted based on the relative abundance of OTUs. Sample OTU richness was calculated from the rarefied index of richness (Chao1) using the ‘estimateR’ function of the vegan package (Oksanen et al., 2013) in R version 3.1.3 (R Core Team, 2015). The alpha diversity indices including Shannon and Simpson's, as well as Pielou's evenness indices were calculated using the ‘diversity’ function in vegan. A four-way analysis of variance (ANOVA) was used to examine the effects of sample type (root, soil), fertilisation, irrigation, and sample origin [grass present (þ) or absent ()] on fungal alpha diversity. To evaluate the effects of sample type (i.e. root and soil), fertilisation, irrigation, sample origin and their interactions on fungal community composition, four-way permutational multivariate analysis of variance (PerMANOVA) was carried out based on Bray-Curtis dissimilarity matrices using the ‘adonis’ function in vegan with 999 permutations. The relative abundances of each OTU in the OTU-sample matrix were treated with the Hellinger transformation, and principal coordinate analysis (PCoA) was performed with the BrayeCurtis dissimilarity index using the ‘wcmdscale’ function in vegan for visualizing the differences in fungal community composition among samples. The function ‘indval’ in the labdsv package (Roberts, 2010) was run to determine fungal indicator ‘species’ (i.e. OTUs, P < 0.05); the significant OTUs with Indval values  0.3 were considered ‘critical indicator OTUs’ (Logares et al., 2013) and the remaining significant indicator OTUs were considered ‘common indicator OTUs’. The function ‘ggiNEXT’ in the iNEXT package (Hsieh et al., 2016) was used to construct rarefaction curves for root and soil samples. The sample type was found to be the most significantly important driver. Therefore, subsequent analyses were done using threeway ANOVA for alpha diversity and PerMANOVA and PCoA for community compositions of root and soil data separately. We also estimated the correspondence between fungal communities derived from root and soil samples to estimate the degree of interchange between root and soil environments and the effects of the management treatments on this correspondence. To estimate this correspondence, we calculated (separately for grass present and grass absent samples) the frequency of OTUs that were observed in both root and soil samples within each plot relative to those OTUs that were only observed for one sample type in the plot. To estimate the effect of management treatments and the presence of grasses on this correspondence, we performed logistic regression by fitting a generalised linear mixed effects model using the lme4 package (Bates et al., 2015) for R. In this analysis, we modelled a binomial response consisting of the number of OTUs observed in

17

both root and soil samples under each treatment combination compared with the number of OTUs found in only one sample type; sample origin (grass present/absent) within each plot was considered the unit of replication, and non-independence of samples was accounted for by including ‘block’ and ‘plot nested within block’ as random effects. 3. Results 3.1. 454 pyrosequencing analysis and fungal community diversity In total, 122,994 high-quality reads were retained from 32 root samples and 32 soil samples after the denoising step. Subsequently 1563 potential chimeras were excluded from the dataset, resulting in 121,431 non-chimeric reads. These sequences were further normalized to 404 reads for each sample, resulting in a dataset containing 1597 fungal OTUs (25,856 reads). Among these 1597 OTUs, 34 non-fungal OTUs identified as Alveolata, Amoebozoa, and Animalia were removed from the dataset, retaining 1563 fungal OTUs for subsequent analyses. We conducted all analyses either including or excluding singletons (549) and observed highly similar results, and therefore only the results of analyses including singletons are presented here. All fungal community diversity properties were significantly affected by sample type (Table 1), indicating greater diversity in soil samples than in root samples (Table 2, Fig. S2). However, management practices (fertilisation and irrigation) and the presence/absence of grass (sample origin) were not observed to affect any of the estimated fungal alpha diversity patterns (Table 1). 3.2. Changes in fungal community composition The PerMANOVA results indicated significant effects of sample type (R2 ¼ 0.131, P ¼ 0.001), fertilisation (R2 ¼ 0.041, P ¼ 0.001), and irrigation (R2 ¼ 0.056, P ¼ 0.001) on fungal community compositions (Table 3). Significant interactions between fertilisation and irrigation (R2 ¼ 0.025, P ¼ 0.001) and between sample type and irrigation (R2 ¼ 0.027, P ¼ 0.005) were also observed. Sample origin also had an effect, but the effect size was of marginal significance (R2 ¼ 0.019, P ¼ 0.042) and no interactions with other factors were observed. To address whether the magnitude of responses to management differed among sample types, analyses were performed independently for each of the two subsets based on sample type. Significant

Table 1 Summary of the four-way ANOVA on the comparison of fungal diversity indices, including estimated Richness (Chao1), Shannon-Wiener index, Simpson's index, and Pielou's evenness index. Effect

Sample type (ST) Fertilisation (F) Irrigation (I) Sample origin (SO) ST  F ST  I IF ST  SO F  SO I  SO ST  F  I ST  F  SO ST  I  SO I  F  SO ST  F  I  SO

Df

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Richness

Shannon

Simpson

Pielou

F

P

F

P

F

P

F

P

73.9 3.22 0.96 1.70 0.45 2.15 0.08 0.00 0.30 0.29 1.92 2.44 1.88 0.22 0.04

<0.001 0.079 0.334 0.199 0.506 0.150 0.784 0.962 0.589 0.595 0.172 0.125 0.177 0.640 0.846

46.8 0.19 0.94 0.41 2.87 0.00 1.15 0.97 0.26 0.06 1.24 2.11 0.04 0.27 0.08

<0.001 0.667 0.337 0.523 0.097 0.995 0.289 0.329 0.610 0.805 0.271 0.153 0.850 0.603 0.775

10.2 0.59 0.89 0.47 1.51 0.50 0.27 2.97 0.99 0.32 0.35 1.55 0.40 0.01 0.01

0.003 0.448 0.350 0.498 0.225 0.483 0.607 0.091 0.325 0.575 0.558 0.220 0.530 0.946 0.917

20.1 0.01 1.09 0.17 2.61 0.27 0.92 2.08 0.37 0.09 0.97 1.94 0.10 0.05 0.01

<0.001 0.908 0.302 0.680 0.113 0.607 0.342 0.156 0.545 0.765 0.330 0.170 0.756 0.829 0.921

Note: The P values < 0.05 were indicated in bold.

18

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

Table 2 Estimated diversity properties for root- and soil-associated fungal communities. Mean values plus/minus standard errors are presented. Sample type

Richness (Chao1)

Shannon

Simpson

Pielou

Root Soil P value (T-test)

122 ± 9b 219 ± 8a <0.001

3.03 ± 0.12b 3.99 ± 0.07a <0.001

0.871 ± 0.027b 0.959 ± 0.004a 0.002

0.723 ± 0.023b 0.836 ± 0.010a <0.001

Note: The P values < 0.05 were indicated in bold.

Table 3 Detailed PerMANOVA outcomes using all data (n ¼ 64) and re-analysed individually for each sample type: root- (n ¼ 32) and soil- (n ¼ 32) associated fungal communities. Predictors included fertilisation and irrigation treatments, as well as whether samples were collected from between tree rows where understorey grasses were present or within rows where grasses did not grow (sample origin). Effect

Sample type (ST) Fertilisation (F) Irrigation (I) Sample origin (SO) ST  F ST  I IF ST  SO F  SO I  SO ST  F  I ST  F  SO ST  I  SO F  I  SO ST  F  I  SO

Df

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

All data

Root data

Soil data

F value

R2

P value

F value

R2

P value

F value

R2

P value

10.7 3.39 4.58 1.57 1.24 2.21 2.08 1.10 1.28 1.05 0.92 0.93 0.84 1.02 1.13

0.131 0.041 0.056 0.019 0.015 0.027 0.025 0.013 0.016 0.013 0.011 0.011 0.010 0.012 0.014

0.001 0.001 0.001 0.042 0.137 0.005 0.001 0.259 0.142 0.321 0.507 0.491 0.658 0.381 0.224

n.a. 2.14 2.98 1.42 n.a. n.a. 1.17 n.a. 1.12 0.99 n.a. n.a. n.a. 1.19 n.a.

n.a. 0.061 0.085 0.041 n.a. n.a. 0.033 n.a. 0.032 0.028 n.a. n.a. n.a. 0.034 n.a.

n.a. 0.001 0.001 0.083 n.a. n.a. 0.223 n.a. 0.289 0.481 n.a. n.a. n.a. 0.201 n.a.

n.a. 2.56 3.98 1.21 n.a. n.a. 1.95 n.a. 1.09 0.89 n.a. n.a. n.a. 0.91 n.a.

n.a. 0.070 0.109 0.033 n.a. n.a. 0.053 n.a. 0.030 0.024 n.a. n.a. n.a. 0.025 n.a.

n.a. 0.001 0.001 0.154 n.a. n.a. 0.004 n.a. 0.300 0.694 n.a. n.a. n.a. 0.643 n.a.

Note: The P values < 0.05 were indicated in bold and ‘n.a.’ represents the data and relevant analysis is not available.

main effects were observed in root- and soil-associated fungal communities for both fertilisation (P ¼ 0.001) and irrigation (P ¼ 0.001), but the strength of these effects was greater for soilassociated fungal communities (Table 3), particularly for irrigation (R2 ¼ 0.085 and 0.109 for root and soil subsets, respectively). These patterns were further corroborated by the ordination results, with some separation associated with the irrigation treatment along the PCoA1 axis for roots (Fig. 1a) and a clear separation associated with irrigation for soil (Fig. 1b). Analysis of these subsets did not reveal a significant effect of sample origin for either root- or soil-associated fungi (P  0.05) further demonstrating the relatively weak effect of grass presence/absence. In total, 325 fungal OTUs out of 1563 OTUs were shared between root and soil communities. The frequencies of these 325 common OTUs in root and soil samples were 3.13e65.6% (i.e. occurred in 1e21 out of 32 root samples) and 3.13e100% (1e32 soil samples), respectively. The proportion of OTUs observed to co-occur in both roots and soil varied between 6 and 11% of OTUs across all treatments (Fig. 2) and tended (marginally statistically non-significant) to increase in samples where grasses were present when compared to those where grasses were absent (P ¼ 0.054) and decrease under fertilisation (P ¼ 0.065). The main effect of irrigation and effects of all interactions on the frequency of OTU co-occurrence were statistically non-significant (P > 0.10). 3.3. Indicator OTUs analysis A total of 22, 115, and 227 significant (P < 0.05) indicator OTUs (phylotypes) were screened from three analyses including: (1) sample origin, (2) forest management practices, and (3) sample type within each of the irrigation treatments (Table 4). Of these indicator OTUs, 3 and 19, 46 and 69, and 108 and 119 were found to be critical (Indval values  0.3) and common (<0.3) indicator OTUs, respectively (Table 4). All indicator OTUs were further analyzed in

terms of their potential functional classification (i.e. mycorrhizal, saprotrophic/pathotrophic, and unassigned fungi). Fewer indicator OTUs were observed from grass absent (8 OTUs) than grass present (14 OTUs) samples, and most (18 out of 22) of them were identified as saprotrophic or pathotrophic (i.e. plant or animal pathogen) fungi, regardless of grass presence or not (Table 4). Indicator OTUs were more likely to be identified as EcM fungal in the control treatment (100% and 66.7% of critical and common indicator OTUs, respectively) than in the treatments including fertilisation and/or irrigation, although the maximum 17 EcM fungal indicator OTUs were observed in the irrigation treatment (Table 4). More saprotrophic/pathotrophic than EcM fungal indicator OTUs (regardless of whether total, critical or common) were found in the analyses of sample type within each of the irrigation treatments. Irrigation tended to increase the percentage of EcM OTUs in root samples (from 20.0 to 36.4%), but to decrease the percentage of EcM OTUs in soil samples (from 21.1 to 15.8%). Under irrigation, higher EcM fungal critical indicator OTUs occurrence (36.4%) was observed in root samples when compared with soil samples (15.8%), whereas similar occurrences of common indicator OTUs (25.9% vs. 20.0%) were observed in root and soil samples without irrigation (Table 4). A total of 157 indicator OTUs where classified as critical (i.e. Indval values  0.3). The relative abundance of OTUs was used to examine the correlation between indicator OTUs and available soil parameters. Of these critical indicator OTUs, 14 OTUs had strongly significant (P  0.001) correlations with at least one of the measured soil variables including soil pH, moisture, NO 3 eN, and/or NHþ 4 eN (Table 5). For these 14 critical indicator OTUs, the OTU00018, OTU00031, and OTU00130 were classified into EcM fungal lineages: ‘/ramaria-gautieria’ (positively correlated with NHþ 4 -N), ‘/cortinarius’ (positively correlated with pH), and ‘/tomentella-thelephora’ (positively correlated with soil moisture), respectively (Table 5).

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

19

Fig. 2. The effect of management treatment on the proportion (incidence) of OTUs detected in both root and soil samples. Filled circles represent the mean proportion; horizontal lines associated with boxes represent the median and interquartile range; whiskers represent the range, except where potential outliers (small circles) were observed (then represent 1.5 times the interquartile range). Samples were collected within tree rows where grasses were absent [Grass ()] and between tree rows where a grassy understorey was present [Grass (þ)]. Ctrl, Control; F, Fertilisation; I, Irrigation; IF, Fertilisation plus Irrigation.

Fig. 1. Outcomes of principal coordinates analyses showing effects of fertilisation and irrigation of a Eucalyptus saligna Sm. plantation on the composition of fungal communities associated with root (a) and soil (b) samples. Samples were collected from two microenvironments in each treatment: between rows where a grassy understorey was present [Grass (þ)] and within rows where litter largely prevented growth of grasses [Grass ()].

4. Discussion 4.1. Forest management practices altered fungal community composition, but not diversity Previous studies have reported that fungal richness was not only related to plant diversity and functional diversity (Gao et al., 2013; Peay et al., 2013), but also could be shaped by climatic and/or abiotic factors, such as temperature, soil moisture, pH and N availability (Allison et al., 2007; Bi et al., 2012; Tedersoo et al., 2014; Wang et al., 2015). In this study, however, we did not observe any relationship between fungal diversity and either fertilisation or irrigation. In line with our results, some other studies demonstrated that the diversity of soil fungi was poorly correlated with environmental manipulations, such as long-term N deposition in forests (Entwistle et al., 2013) and altered precipitation frequency in grasslands (Jumpponen and Jones, 2014). We speculate that the responses of fungal diversity to environmental changes and/or anthropogenic disturbances might be dependent on the scales and categories of the ecosystem, because microbial communities including fungi were found to be resilient or adapted to simulated environmental or climatic condition changes in some ecosystems (Zhang et al., 2013; Jumpponen and Jones, 2014).

However, we found that fertilisation and irrigation strongly influenced fungal community compositions in both root and soil samples. Fungal community structure responses to fertilisation have been extensively studied, with varying results and interpretations. For example, N fertiliser addition can significantly change fungal community composition in different ecosystems (Allison et al., 2007; Bi et al., 2012; Liu et al., 2013; Kim et al., 2015). The fertilisation may influence fungal community composition through at least three mechanisms. First, some fungi such as mycorrhizal and saprotrophic fungi play vital roles in belowground N and C mobilisation, respectively (e.g. Lindahl et al., 2007; Entwistle et al., 2013), and fertilisation influences the relative availability of these resources in soil. Second, fertilisation influences the aboveground plant community, with consequences for resource quantity and quality for symbionts and detritivores, which can impact root and/or soil fungal communities (Antoninka et al., 2011; Liu et al., 2012; Kim et al., 2015). Finally, altered soil physicochemical properties (e.g. pH) under fertilisation would also greatly influence the fungal community composition (e.g. Weber et al., 2013; Zheng et al., 2014). Understanding how fungal communities are influenced by precipitation is essential for better prediction of soil functional responses to future climate change (Hawkes et al., 2011; Curlevski et al., 2014). There has been evidence that the impacts of water availability on soil fungal communities can vary across different ecosystems and precipitation regimes. For instance, water addition had no effect on soil fungal PLFA composition in a semiarid steppe (Bi et al., 2012; Zhang et al., 2013) and in a desert steppe (Huang et al., 2015). However, after a c. 20% water addition treatment, the soil fungal community became less abundant, less diverse, and more variable compared to those under drought conditions in a grassland ecosystem in northern California (Hawkes et al., 2011). Different response of fungal community composition to water addition might be attributed to the relative levels of water

20

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

Table 4 Comparisons of the numbers of OTUs associated with fungal trophic groups and identifies and indicators of treatments associated with the presence [Grass (þ)] and absence [Grass ()] of understorey grasses and specific forest management practices. The estimated proportions of indicator OTUs identified as ectomycorrhizal fungi are also shown for each scenario. Effect

Properties

Sum

Critical (common) indicator OTUs

Symbiotroph/ EcM fungi

Saprotroph/ Pathotroph

Unassigned fungi

% of EcM fungi

Sample origin

Grass (þ) Grass () Total

14 8 22

1 (13)a 2 (6) 3 (19)

0 (2) 1 (1) 1 (3)

1 (11) 1 (5) 2 (16)

0 (0) 0 (0) 0 (0)

0 (15.4) 50.0 (16.7) 33.3 (15.8)

Forest management practices

Control Fertilisation (F) Irrigation (I) IF Total

17 41 37 20 115

5 (12) 16 (25) 19 (18) 6 (14) 46 (69)

5 (8) 2 (8) 9 (8) 3 (4) 19 (28)

0 (3) 13 (17) 9 (9) 2 (8) 24 (37)

0 1 1 1 3

(1) (0) (1) (2) (4)

100 (66.7) 12.5 (32.0) 47.4 (44.4) 50.0 (28.6) 41.3 (40.6)

Irrigation (I) vs. sample origin

noI_Root noI_Soil I_Root I_Soil Total

37 68 42 80 227

10 (27) 38 (30) 22 (20) 38 (42) 108 (119)

2 (7) 8 (6) 8 (5) 6 (10) 24 (28)

8 (20) 29 (20) 13 (12) 30 (24) 80 (76)

0 1 1 2 4

(0) (4) (3) (8) (15)

20.0 21.1 36.4 15.8 22.2

a

(25.9) (20.0) (25.0) (23.8) (23.5)

Data of indicator OTUs are shown separately as critical indicator OTUs (Indval values  0.3), followed by common indicator OTUs (<0.3) in parentheses.

Table 5 Pearson correlation coefficients (r) and P-values associated with 14 OTUs identified as critical indicators of treatments and observed to be correlated with measured soil properties. Moisture (%)

NHþ 4 -N

NO 3 -N

Name

Reflect treatment

Identification of trophic status

pH r

P

r

P

r

P

r

P

r

P

r

P

OTU00018

Fertilisation

0.324

0.009

0.215

0.088

0.407

<0.001

0.281

0.024

0.049

0.703

0.078

0.541

OTU00017 OTU00170 OTU00136 OTU00093 OTU00152 OTU00193 OTU00031

Fertilisation Fertilisation Fertilisation Fertilisation Fertilisation Fertilisation Irrigation (I)

0.454 0.297 0.269 0.300 0.491 0.280 0.452

<0.001 0.017 0.032 0.016 <0.001 0.025 <0.001

0.331 0.216 0.198 0.199 0.423 0.207 0.220

0.008 0.086 0.117 0.115 <0.001 0.100 0.081

0.526 0.508 0.560 0.442 0.330 0.430 0.193

<0.001 <0.001 <0.001 <0.001 0.008 <0.001 0.127

0.286 0.254 0.210 0.394 0.075 0.282 0.151

0.022 0.043 0.095 0.001 0.557 0.024 0.233

0.013 0.118 0.139 0.078 0.000 0.017 0.067

0.919 0.351 0.272 0.539 0.998 0.897 0.599

0.067 0.043 0.038 0.028 0.213 0.013 0.092

0.599 0.736 0.765 0.828 0.092 0.920 0.468

OTU00026 OTU00203 OTU00403 OTU00188 OTU00167 OTU00130

Irrigation (I) noI_Soil noI_Soil I_Root I_Root I_Root

Symbiotroph (/ramaria-gautieria) Saprotroph Saprotroph Pathotroph Saprotroph Pathotroph Saprotroph Symbiotroph (/cortinarius) Saprotroph Saprotroph Saprotroph Saprotroph Saprotroph Symbiotroph (/tomentella-thelephora)

0.379 0.293 0.313 0.501 0.268 0.208

0.002 0.019 0.012 <0.001 0.032 0.100

0.410 0.213 0.228 0.391 0.411 0.439

<0.001 0.091 0.070 0.001 <0.001 <0.001

0.297 0.439 0.478 0.309 0.052 0.192

0.017 <0.001 <0.001 0.013 0.685 0.128

0.006 0.349 0.313 0.108 0.022 0.052

0.963 0.005 0.012 0.395 0.861 0.681

0.157 0.023 0.009 0.137 0.124 0.321

0.215 0.855 0.941 0.282 0.328 0.010

0.334 0.026 0.018 0.256 0.174 0.387

0.007 0.835 0.886 0.041 0.169 0.002

TC

TN

Note: noI_Soil, soil fungal data without irrigation; I_Root, root fungal data under irrigation condition. The lineage names of three ectomycorrhizal fungal OTUs (indicated in bold) are shown in parentheses. TC, soil total carbon; TN, soil total nitrogen. The P values which are 0.001 were indicated in bold.

availability for microbial metabolic activity (Raich and Potter, 1995), the amount of precipitation and the water application time (Nielsen and Ball, 2015), and the distinctly accumulative effects of precipitation regimes (Williams and Rice, 2007). Indeed, we observed strong impacts of a six-year irrigation on root and soil fungal community composition (Fig. 1), thereby supporting the recent findings that water addition exerted crucial effects on other functional microbial activities such as ammonia oxidation, greenhouse gas emissions, and potential enzyme activities in the same experimental system (Hu et al., 2015; Martins et al., 2015; Colombo et al., 2016). 4.2. Distinct differences between root- and soil-associated fungal communities In this study, fungal diversity indices were substantially greater in soil samples, indicating that soil harbored more diverse fungal communities than plant roots. Our results were consistent with the findings in which fungal species and family richness were lower in

root than those in soil as reported by Danielsen et al. (2012). We also observed that, while a minority of fungal OTUs were detected both in root and soil samples, the degree to which these soil- and root-associated fungal communities overlap may depend on local environmental conditions (i.e. the presence of understorey grasses and the manipulation of soil fertility). This expands on the hypothesis proposed by Beck et al. (2015), based on a survey of fungal communities associated with an ericacious host plant, that rootassociated fungal communities do not exhibit niche-based assembly and instead are more governed by neutral dynamics independent of local environmental conditions. This difference could be due to the relatively steep environmental gradient generated in the current study compared with that of Beck et al. (2015), where environmental selection in soil may be less important than dispersal in determining the local abundance of fungi available to colonise roots. While we cannot determine exactly why the degree of co-occurrence varied in our study, we can speculate that reduced abundance and/or diversity of host plants, or the process of colonising roots of a specific tree host (here E. saligna), as well as

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

increasing the pool of labile nutrients available to fungi, lead to a decoupling of community assembly processes within the root environment from that of the soil environment. Lessening the importance of environmental selection in soil on the assembly of root-associated fungal communities may play a role in generating a continuum of somewhat predictable to highly unpredictable community outcomes (Powell and Bennett, 2016). This has consequences for understanding the dynamics of root-associated mutualists and pathogens, but also for our theoretical understanding of complex dynamics occurring during the interchange of entire fungal communities (i.e. community coalescence; Rillig et al., 2015, 2016). However, more work is required to evaluate these possible consequences. Fertilisation, irrigation, and their combination were each associated with unique indicator fungal OTUs. All of these treatments exhibited indicators that were identified as putative EcM taxa, but EcM fungal OTUs made up a greater proportion of indicators in the control treatment. Mycorrhizal fungi, forming mutualistic associations with most land plant species and obtaining C resource from their plant partners in exchange for mineral nutrients (Smith and Read, 2008), are influenced by different types of soil organic C (Verbruggen et al., 2010; Kiers et al., 2011) and N or P fertilisation (i.e. Liu et al., 2012). Under fertilisation and water addition conditions, host plants would be less dependent on mycorrhizal symbionts for seeking more efficient nutrient and water uptake. In this case, C allocation to mycorrhizal fungi may decrease and thus result in shifted mycorrhizal fungal community composition. However, reduced proportions of indicator OTUs identified as EcM fungi was largely due to a greater frequency of saprotrophic and pathotrophic indicator OTUs in each of the treatments including fertilisation, irrigation, or both. Thus, in general, other components of the fungal community may be more responsive to these changes. For example, soil water dynamics (i.e. potential or availability) can directly determine microbial physiological responses and metabolic activity through, for instance, cell dehydration (Stark and Firestone, 1995), and also indirectly affects microbial growth and metabolism via reduced substrate availability due to the diffusive limitations (Hawkes et al., 2011; Hu et al., 2015). Different microbial groups have distinct tolerance to soil water availability, and there is strong variability in water-stress threshold even among soil fungi (Manzoni et al., 2012). Irrigation will also promote degradation of litter and increase nutrient mineralization by other components of the detritivore food web, with subsequent consequences for saprotrophic fungi that form the base of a substantial portion of this web. Although we did not directly measure any ecosystem function that may be linked to compositional shifts in fungal communities, others have demonstrated that these shifts can lead to differences in community functioning (e.g. Amend et al., 2015). The detection of several OTUs belonging to different functional guilds and responsive to management practices suggests the potential for functional consequences of fungal community shifts. In addition, the effects of management practices in combination with characteristics of the local microenvironment highlights an understudied aspect of research in plantation systems (the presence of absence of a grassy understorey) with consequences for the predictability of fungal communities associated with the roots of host plants. Our findings thus necessitate more research into dynamical aspects of fungal community assembly and functioning, particularly in dryland forest ecosystems and their responses to environmental change. Competing interests The authors declare no conflict of interest.

21

Acknowledgments We thank Emma Greenlaw for help during soil sampling and processing and Burhan Amiji and Craig Barton for site access and overall management of the Hawkesbury Forest Experiment. We appreciate Niuniu Ji, Liang Chen, and Cheng Gao for comments on fungal molecular identification and Jessica Rigg and Jennifer Walker for comments on an earlier version of the manuscript. We also thank two anonymous reviewers for very helpful suggestions. This study was financially supported by the Australian Research Council (DP130102501) and by the National Natural Science Foundation of China (31570499). Y.Z. received a fellowship as a visiting scholar from the China Scholarship Council (2011491255). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.soilbio.2017.01.024. References Allison, S.D., Hanson, C.A., Treseder, K.K., 2007. Nitrogen fertilization reduces diversity and alters community structure of active fungi in boreal ecosystems. Soil Biology and Biochemistry 39, 1878e1887. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. Journal of Molecular Biology 215, 403e410. Amend, A.S., Matulich, K.L., Martiny, J.B.H., 2015. Nitrogen addition, not initial phylogenetic diversity, increases litter decomposition by fungal communities. Frontiers in Microbiology 6, 109. Antoninka, A., Reich, P.B., Johnson, N.C., 2011. Seven years of carbon dioxide enrichment, nitrogen fertilization and plant diversity influence arbuscular mycorrhizal fungi in a grassland ecosystem. New Phytologist 192, 200e214.
, R.M., 2001. Water relations, drought and vesicular-arbuscular mycorrhizal Auge symbiosis. Mycorrhiza 11, 3e42. Barton, C.V.M., Ellsworth, D.S., Medlyn, B.E., Duursma, R.A., Tissue, D.T., Adams, M.A., Eamus, D., Conroy, J.P., McMurtrie, R.E., Parsby, J., Linder, S., 2010. Whole-tree chambers for elevated atmospheric CO2 experimentation and tree scale flux measurements in south-eastern Australia: the Hawkesbury Forest Experiment. Agricultural and Forest Meteorology 150, 941e951. Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67, 1e48. Beck, S., Powell, J.R., Drigo, B., Cairney, J.W.G., Anderson, I.C., 2015. The role of stochasticity differs in the assembly of soil- and root-associated fungal communities. Soil Biology and Biochemistry 80, 18e25. Bi, J., Zhang, N.L., Liang, Y., Yang, H.J., Ma, K.P., 2012. Interactive effects of water and nitrogen addition on soil microbial communities in a semiarid steppe. Journal of Plant Ecology 5, 320e329. Bradley, K., Drijber, R.A., Knops, J., 2006. Increased N availability in grassland soils modifies their microbial communities and decreases the abundance of arbuscular mycorrhizal fungi. Soil Biology and Biochemistry 38, 1583e1595. Bragg, L., Stone, G., Imelfort, M., Hugenholtz, P., Tyson, G.W., 2012. Fast, accurate error-correction of amplicon pyrosequences using Acacia. Nature Methods 9, 425e426. Chao, A., Jost, L., 2012. Coverage-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology 93, 2533e2547. Colombo, F., Macdonald, C.A., Jeffries, T.C., Powell, J.R., Singh, B.K., 2016. Impact of forest management practices on soil bacterial diversity and consequences for soil processes. Soil Biology and Biochemistry 94, 200e210. Cregger, M.A., Schadt, C.W., McDowell, N.G., Pockman, W.T., Classen, A.T., 2012. Response of the soil microbial community to changes in precipitation in a semiarid ecosystem. Applied and Environmental Microbiology 78, 8587e8594. Cubeta, M.A., Echandi, E., Abernethy, T., Vilgalys, R., 1991. Characterization of anastomosis groups of binucleate Rhizoctonia species using restriction analysis of an amplified ribosomal-RNA gene. Phytopathology 81, 1395e1400. Curlevski, N.J.A., Drigo, B., Cairney, J.W.G., Anderson, I.C., 2014. Influence of elevated atmospheric CO2 and water availability on soil fungal communities under Eucalyptus saligna. Soil Biology and Biochemistry 70, 263e271. Danielsen, L., Thurmer, A., Meinicke, P., Buee, M., Morin, E., Martin, F., Pilate, G., Daniel, R., Polle, A., Reich, M., 2012. Fungal soil communities in a young transgenic poplar plantation form a rich reservoir for fungal root communities. Ecology and Evolution 2, 1935e1948. Delgado-Baquerizo, M., Maestre, F.T., Gallardol, A., et al., 2013. Decoupling of soil nutrient cycles as a function of aridity in global drylands. Nature 502, 672e676. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature Methods 10, 996e998. Entwistle, E.M., Zak, D.R., Edwards, I.P., 2013. Long-term experimental nitrogen deposition alters the composition of the active fungal community in the forest floor. Soil Science Society of America Journal 77, 1648e1658.

22

Y. Zheng et al. / Soil Biology & Biochemistry 109 (2017) 14e22

Fornara, D.A., Tilman, D., 2012. Soil carbon sequestration in prairie grasslands increased by chronic nitrogen addition. Ecology 93, 2030e2036. Frew, A., Nielsen, U.N., Riegler, M., Johnson, S.N., 2013. Do eucalypt plantation management practices create understory reservoirs of scarab beetle pests in the soil? Forest Ecology and Management 306, 275e280. Gao, C., Shi, N.N., Liu, Y.X., Peay, K.G., Zheng, Y., Ding, Q., Mi, X.C., Ma, K.P., Wubet, T., Buscot, F., Guo, L.D., 2013. Host plant genus-level diversity is the best predictor of ectomycorrhizal fungal diversity in a Chinese subtropical forest. Molecular Ecology 22, 3403e3414. € ter, K., Pena, R., Scho €ning, I., Schrumpf, M., Buscot, F., Polle, A., Goldmann, K., Schro Wubet, T., 2016. Divergent habitat filtering of root and soil fungal communities in temperate beech forests. Scientific Reports 6, 31439. Hawkes, C.V., Kivlin, S.N., Rocca, J.D., Huguet, V., Thomsen, M.A., Suttle, K.B., 2011. Fungal community responses to precipitation. Global Change Biology 17, 1637e1645. He, J.Z., Zheng, Y., Chen, C.R., He, Y.Q., Zhang, L.M., 2008. Microbial composition and diversity of an upland red soil under long-term fertilisation treatments as revealed by culture-dependent and culture-independent approaches. Journal of Soils and Sediments 8, 349e358. Hsieh, T.C., Ma, K.H., Chao, Anne, 2016. iNEXT: iNterpolation and EXTrapolation for Species Diversity. R package version 2.0.8. https://cran.r-project.org/web/ packages/iNEXT/index.html. Hu, H.W., Macdonald, C.A., Trivedi, P., Holmes, B., Bodrossy, L., He, J.Z., Singh, B.K., 2015. Water addition regulates the metabolic activity of ammonia oxidizers responding to environmental perturbations in dry subhumid ecosystems. Environmental Microbiology 17, 444e461. Huang, G., Li, Y., Su, Y.G., 2015. Divergent responses of soil microbial communities to water and nitrogen addition in a temperate desert. Geoderma 251, 55e64. Husband, R., Herre, E.A., Turner, S.L., Gallery, R., Young, J.P.W., 2002. Molecular diversity of arbuscular mycorrhizal fungi and patterns of host association over time and space in a tropical forest. Molecular Ecology 11, 2669e2678. Johnson, S.L., Kuske, C.R., Carney, T.D., Housman, D.C., Gallegos-Graves, L., Belnap, J., 2012. Increased temperature and altered summer precipitation have differential effects on biological soil crusts in a dryland ecosystem. Global Change Biology 18, 2583e2593. Jumpponen, A., Jones, K.L., 2014. Tallgrass prairie soil fungal communities are resilient to climate change. Fungal Ecology 10, 44e57. Kiers, E.T., Duhamel, M., Beesetty, Y., Mensah, J.A., Franken, O., Verbruggen, E., Fellbaum, C.R., Kowalchuk, G.A., Hart, M.M., Bago, A., Palmer, T.M., West, S.A., Vandenkoornhuyse, P., Jansa, J., Bücking, H., 2011. Reciprocal rewards stabilize cooperation in the mycorrhizal symbiosis. Science 333, 880e882. Kim, Y.C., Gao, C., Zheng, Y., He, X.H., Yang, W., Chen, L., Wan, S.Q., Guo, L.D., 2015. Arbuscular mycorrhizal fungal community response to warming and nitrogen addition in a semiarid steppe ecosystem. Mycorrhiza 25, 267e276. Li, X., Zhu, T., Peng, F., Chen, Q., Lin, S., Christie, P., Zhang, J., 2015. Inner Mongolian steppe arbuscular mycorrhizal fungal communities respond more strongly to water availability than to nitrogen fertilization. Environmental Microbiology 17, 3051e3068. Lindahl, B.D., Ihrmark, K., Boberg, J., Trumbore, S.E., Hogberg, P., Stenlid, J., Finlay, R.D., 2007. Spatial separation of litter decomposition and mycorrhizal nitrogen uptake in a boreal forest. New Phytologist 173, 611e620. Liu, L., Zhang, T., Gilliam, F.S., Gundersen, P., Zhang, W., Chen, H., Mo, J.M., 2013. Interactive effects of nitrogen and phosphorus on soil microbial communities in a tropical forest. PLoS One 8, e61188. Liu, Y.J., Shi, G.X., Mao, L., Cheng, G., Jiang, S.J., Ma, X.J., An, L.Z., Du, G.Z., Johnson, N.C., Feng, H.Y., 2012. Direct and indirect influences of 8 yr of nitrogen and phosphorus fertilization on Glomeromycota in an alpine meadow ecosystem. New Phytologist 194, 523e535. Logares, R., Lindstrom, E.S., Langenheder, S., Logue, J.B., Paterson, H., LaybournParry, J., Rengefors, K., Tranvik, L., Bertilsson, S., 2013. Biogeography of bacterial communities exposed to progressive long-term environmental change. ISME Journal 7, 937e948. Magnani, F., Mencuccini, M., Borghetti, M., Berbigier, P., Berninger, F., Delzon, S., Grelle, A., Hari, P., Jarvis, P.G., Kolari, P., Kowalski, A.S., Lankreijer, H., Law, B.E., Lindroth, A., Loustau, D., Manca, G., Moncrieff, J.B., Rayment, M., Tedeschi, V., Valentini, R., Grace, J., 2007. The human footprint in the carbon cycle of temperate and boreal forests. Nature 447, 848e850. Manzoni, S., Schimel, J.P., Porporato, A., 2012. Responses of soil microbial communities to water stress: results from a meta-analysis. Ecology 93, 930e938. Martins, C.S.C., Nazaries, L., Macdonald, C.A., Anderson, I.C., Singh, B.K., 2015. Water availability and abundance of microbial groups are key determinants of greenhouse gas fluxes in a dryland forest ecosystem. Soil Biology and Biochemistry 86, 5e16. Millennium Ecosystem Assessment, 2005. Drylands systems. In: Hassan, R., Scholes, R., Ash, N. (Eds.), Ecosystems and Human Wellbeing: Current State and Trends. Island Press, Washington, DC, USA, pp. 625e655. Nguyen, N.H., Song, Z.W., Bates, S.T., Branco, S., Tedersoo, L., Menke, J., Schilling, J.S., Kennedy, P.G., 2016. FUNGuild: an open annotation tool for parsing fungal community datasets by ecological guild. Fungal Ecology 20, 241e248. Nielsen, U.N., Ball, B.A., 2015. Impacts of altered precipitation regimes on soil communities and biogeochemistry in arid and semi-arid ecosystems. Global Change Biology 21, 1407e1421. Nielsen, U.N., Prior, S., Delroy, B., Walker, J.K.M., Ellsworth, D.S., Powell, J.R., 2015. Response of belowground communities to short-term phosphorus addition in a phosphorus-limited woodland. Plant and Soil 391, 321e331.

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2013. Vegan: Community Ecology Package. R package version 2.0-10. http://CRAN.R-project.org/ package¼vegan. Peay, K.G., Baraloto, C., Fine, P.V.A., 2013. Strong coupling of plant and fungal community structure across western Amazonian rainforests. ISME Journal 7, 1852e1861. Philippot, L., Raaijmakers, J.M., Lemanceau, P., van der Putten, W.H., 2013. Going back to the roots: the microbial ecology of the rhizosphere. Nature Reviews Microbiology 11, 789e799. Powell, J.R., Bennett, A.E., 2016. Unpredictable assembly of arbuscular mycorrhizal fungal communities. Pedobiologia 59, 11e15. Pruesse, E., Quast, C., Knittel, K., Fuchs, B.M., Ludwig, W.G., Peplies, J., Glockner, F.O., 2007. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Research 35, 7188e7196. R Core Team, 2015. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project. org/. Raich, J.W., Potter, C.S., 1995. Global patterns of carbon-dioxide emissions from soils. Global Biogeochemical Cycles 9, 23e36. Rillig, M.C., Antonovics, J., Caruso, T., Lehmann, A., Powell, J.R., Veresoglou, S.D., Verbruggen, E., 2015. Interchange of entire communities: microbial community coalescence. Trends in Ecology & Evolution 30, 470e476. Rillig, M.C., Lehmann, A., Aguilar-Trigueros, C.A., Antonovics, J., Caruso, T., Hempel, S., Lehmann, J., Valyi, K., Verbruggen, E., Veresoglou, S.D., Powell, J.R., 2016. Soil microbes and community coalescence. Pedobiologia 59, 37e40. Roberts, D.W., 2010. LABDSV: Ordination and Multivariate Analysis for Ecology. http://cran.r-project.org/web/packages/labdsv/labdsv.pdf. € Saks, Ü., Davison, J., Opik, M., Vasar, M., Moora, M., Zobel, M., 2014. Root-colonizing and soil-borne communities of arbuscular mycorrhizal fungi in a temperate forest understorey. Botany 92, 277e285. Schimel, D.S., 2010. Drylands in the earth system. Science 327, 418e419. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., Weber, C.F., 2009. Introducing mothur: opensource, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75, 7537e7541. Singh, B.K., Bardgett, R.D., Smith, P., Reay, D.S., 2010. Microorganisms and climate change: terrestrial feedbacks and mitigation options. Nature Reviews Microbiology 8, 779e790. Singh, B.K., Quince, C., Macdonald, C.A., Khachane, A., Thomas, N., Abu Al-Soud, W., Sørensen, S.J., He, Z.L., White, D., Sinclair, A., Crooks, B., Zhou, J.Z., Campbell, C.D., 2014. Loss of microbial diversity in soils is coincident with reductions in some specialized functions. Environmental Microbiology 16, 2408e2420. Smith, S.E., Read, D.J., 2008. Mycorrhizal Symbiosis, third ed. Academic Press, London, UK. Stark, J.M., Firestone, M.K., 1995. Mechanisms for soil-moisture effects on activity of nitrifying bacteria. Applied and Environmental Microbiology 61, 218e221. ~lme, S., et al., 2014. Global diversity and geography of Tedersoo, L., Bahram, M., Po soil fungi. Science 346, 1256688. Verbruggen, E., Roling, W.F.M., Gamper, H.A., Kowalchuk, G.A., Verhoef, H.A., van der Heijden, M.G.A., 2010. Positive effects of organic farming on below-ground mutualists: large-scale comparison of mycorrhizal fungal communities in agricultural soils. New Phytologist 186, 968e979. Veresoglou, S.D., Chen, B.D., Rillig, M.C., 2012. Arbuscular mycorrhiza and soil nitrogen cycling. Soil Biology and Biochemistry 46, 53e62. Waldrop, M.P., Zak, D.R., Blackwood, C.B., Curtis, C.D., Tilman, D., 2006. Resource availability controls fungal diversity across a plant diversity gradient. Ecology Letters 9, 1127e1135. Wang, J.T., Zheng, Y.M., Hu, H.W., Zhang, L.M., Li, J., He, J.Z., 2015. Soil pH determines the alpha diversity but not beta diversity of soil fungal community along altitude in a typical Tibetan forest ecosystem. Journal of Soils and Sediments 15, 1224e1232. Weber, C.F., Vilgalys, R., Kuske, C.R., 2013. Changes in fungal community composition in response to elevated atmospheric CO2 and nitrogen fertilization varies with soil horizon. Frontiers in Microbiology 4, 78. Williams, M.A., Rice, C.W., 2007. Seven years of enhanced water availability influences the physiological, structural, and functional attributes of a soil microbial community. Applied Soil Ecology 35, 535e545. Zhang, N.L., Liu, W.X., Yang, H.J., Yu, X.J., Gutknecht, J.L.M., Zhang, Z., Wan, S.Q., Ma, K.P., 2013. Soil microbial responses to warming and increased precipitation and their implications for ecosystem C cycling. Oecologia 173, 1125e1142. Zheng, Y., Chen, L., Luo, C.Y., Zhang, Z.H., Wang, S.P., Guo, L.D., 2016. Plant identity exerts stronger effect than fertilization on soil arbuscular mycorrhizal fungi in a sown pasture. Microbial Ecology 72, 647e658. Zheng, Y., Kim, Y.C., Tian, X.F., Chen, L., Yang, W., Gao, C., Song, M.H., Xu, X.L., Guo, L.D., 2014. Differential responses of arbuscular mycorrhizal fungi to nitrogen addition in a near pristine Tibetan alpine meadow. FEMS Microbiology Ecology 89, 594e605. Zhou, X.Q., Chen, C.R., Wang, Y.F., Xu, Z.H., Duan, J.C., Hao, Y.B., Smaill, S., 2013. Soil extractable carbon and nitrogen, microbial biomass and microbial metabolic activity in response to warming and increased precipitation in a semiarid Inner Mongolian grassland. Geoderma 206, 24e31.