Introduction
Over the past few decades, human activities have significantly intensified nitrogen deposition, far exceeding the safe thresholds for environmental sustainability1. This poses a serious threat to ecosystems, impacting diversity and function through soil acidification, reduced plant variety, and disrupted soil nutrient balance2. The alterations in plant community characteristics (diversity, composition, and productivity) and soil characteristics (pH, nitrogen, phosphorus, and heavy metals) due to nitrogen deposition also impact the community structure of soil microbiomes3,4. Arbuscular mycorrhizal fungi (AMF) form symbiotic associations with over two-thirds of terrestrial plant species through root colonization, playing pivotal roles in mediating carbon, nitrogen, and phosphorus cycling within ecosystems5, 6, 7–8. The functions of the AMF-associated microbiome have garnered significant attention for their ecological importance in recent years9, 10–11. The coordination between mycorrhizal fungi and soil microbiome is crucial for ecosystem stability and functionality, as it regulates soil nutrient dynamics and helps alleviate environmental pressures12,13. However, the response of AMF and AMF-associated microbiome to changes in soil chemistry induced by nitrogen addition remains uncertain. The responses of AMF communities to nitrogen addition are often manifested in taxonomic diversity (alpha diversity), community composition, and beta diversity, yet no consensus has been reached thus far (Supplementary Table S1). For example, divergent studies have reported that AMF alpha diversity may exhibit increases, decreases, or no significant changes under nitrogen addition14, 15, 16–17.
Besides, AMF community beta diversity may increase, decrease, or remain unchanged in response to nitrogen addition18,19. For example, nitrogen addition in the Qinghai-Tibet Plateau alpine meadow ecosystem has been shown to reduce AMF alpha diversity while increasing AMF beta diversity, a trend evident when beta diversity was quantified using the Sørensen index but not the Bray–Curtis index18; whereas nitrogen addition in Swedish farmland reduced AMF alpha diversity and putatively decreased AMF beta diversity, though statistical evidence for beta dispersion remained lacking20. The inconsistent findings might be attributable to variations in environmental contexts across different scenarios. When nitrogen addition alleviates nitrogen limitation and without causing phosphorus limitation21,22, we suggest that alpha diversity of the AMF community decreases while beta diversity increases when plants with lower mycorrhizal dependency allocate a smaller proportion of photosynthate and exert weaker selection on the AMF community. However, when nitrogen addition alleviates nitrogen limitation and induces phosphorus limitation23, both AMF alpha and beta diversity may increase, decrease, or remain unchanged, depending on the level of nitrogen and phosphorus limitation20,24,25. Besides, when nitrogen addition causes soil acidification, heavy metal release, and plant diversity loss26, 27–28, both AMF alpha and beta diversities might decrease. However, despite the different mechanisms involving nitrogen availability, phosphorus availability, soil pH, soil heavy metal content, and plant richness caused by nitrogen addition, the response of AMF alpha and beta diversity to nitrogen addition has not been well resolved (Supplementary Table S1). Here, in a semiarid grassland ecosystem where nitrogen addition alleviates nitrogen limitation and is not likely to cause phosphorus limitation21, we simply hypothesize (hypothesis-1) that nitrogen addition causes a negative association between beta diversity and alpha diversity.
The functionality of AMF is at least partially dependent on its associated microbiome, which might be inferred from co-occurrence network analyses identifying specific microbial community subsets co-occurring with AMF taxa. Several prior studies have emphasized the critical role of the mycorrhiza-associated microbiome in fundamental ecological processes, including phosphorus mineralization, stress/toxin tolerance, and plant growth promotion6,7,29, 30, 31, 32–33. A major challenge in investigating the mycorrhiza-associated microbiome is distinguishing mycorrhizal fungi-interacting microbial members from free-living microbial members, for which sophisticated compartmented microcosms have been developed29,34,35. In natural ecosystems, the composition and function of the mycorrhiza-associated microbiome might be informed by network analysis on the co-occurrences of AMF taxa with microbial taxa and genes, yet this area, to our knowledge, remains largely untouched36. When nitrogen addition relieves nitrogen limitation without causing phosphorus limitation21,22, we propose that nitrogen addition does not necessarily enrich the phosphorus mineralization function of the mycorrhiza-associated microbiome. When nitrogen addition alleviates nitrogen limitation and results in phosphorus limitation23, we suggest that the function of phosphorus mineralization would be enriched for the mycorrhiza-associated microbiome. Additionally, when nitrogen addition results in soil acidification and heavy metal release26, 27–28, we suggest enrichment in the functions of signal transduction and motility, as the benign microhabitat of the hyphosphere can be viewed as a refuge for stress-non-tolerant microbes. In this semiarid grassland ecosystem of Inner Mongolia, nitrogen addition relieves nitrogen and phosphorus limitations but imposes stress on mycorrhizal fungi through soil acidification, heavy metal toxicity, and ammonia toxicity15,21,27. We hypothesize (hypothesis-2) that nitrogen addition did not enrich the phosphorus mineralization function of the mycorrhiza-associated microbiome.
The system we used to test these two hypotheses consists of a semi-arid grassland field with seven nitrogen addition levels (Nlevel), and five replicate plots per treatment. Soil samples were collected in 2017 and 2019, from which DNA was extracted to characterize the AMF community via 18S rRNA gene sequencing, the bacterial community via 16S rRNA gene sequencing, and conduct functional analysis via shotgun metagenomics (Supplementary Fig. S1). Our analyses provided support for hypothesis-1 by demonstrating that nitrogen addition induced a negative correlation between beta diversity and alpha diversity. We also supported hypothesis-2 by finding no evidence for the enrichment of phosphorus mineralization functions in bacterial genes associated with the AMF community, despite the enrichment of functions related to transporters, amino acid synthesis and metabolism, and replication repair.
Results and discussion
We characterized the AMF community structure via 18S rRNA gene metabarcoding for a total of 70 samples. These samples were collected from 35 plots representing seven nitrogen addition levels across two years (Supplementary Fig. S1). Our analysis detected a total of 116 AMF OTUs, with community composition dominated by members of the genera Glomus, Claroideoglomus, Paraglomus, and Scutellospora (Supplementary Fig. S2). Principal Coordinate Analysis (PCoA) based on Jaccard, Bray–Curtis, and Sørensen distances indicated that the composition of the AMF community was significantly altered by varying levels of nitrogen addition (Fig. 1). A Mantel test revealed that AMF community composition was significantly correlated with soil available phosphorus, soil available nitrogen, and pH (Supplementary Fig. S3). Threshold indicator analysis (TITAN) showed that nitrogen addition decreased the relative abundance of Ambispora (OTU16), Glomus (OTU4, OTU7, OTU10, OTU13, OTU14, OTU18, OTU21, OTU23, OTU37, OTU57), Paraglomus (OTU17, OTU122), and Scutellospora (OTU51) (Supplementary Fig. S4).
Fig. 1 A negative association between beta diversity and alpha diversity of arbuscular mycorrhizal fungi (AMF) community along a nitrogen addition gradient. [Images not available. See PDF.]
a The occurrences of 116 AMF operational taxonomic units (OTUs) across 70 samples (n = 70) showed changes of AMF community structure under nitrogen addition. Pink-filled squares denote the presence (1) of an AMF OTU in a sampling plot, while white squares indicate the absence (0). b Nitrogen addition significantly decreased AMF α-diversity as measured by OTU richness (n = 70). Note AMF Shannon’s diversity also decreased with nitrogen addition as showed in the Supplementary Fig. S5. c–h Nitrogen addition significantly increased AMF β-diversity (n = 70). The c Jaccard dissimilarity-, d Bray–Curtis dissimilarity-, and e Sørensen dissimilarity-based principal coordinates (PCo) analysis followed by permutational analysis of variance (adonis2) and beta dispersion showed significant effects of nitrogen addition on AMF community composition, and on AMF community beta diversity as measured by an increase in the distance to centroid among replications of a nitrogen level. Fitting curve showing the significant increase of AMF β-diversity (distance-to-centroid) measured based on f Jaccard, g Bray–Curtis, and h Sørensen dissimilarity with increasing level of nitrogen addition. AMF α-diversity correlates negatively with β-diversity (distance-to-centroid) measured based on i Jaccard, j Bray–Curtis, and k Sørensen dissimilarity along a nitrogen addition gradient (n = 70).
To test hypothesis-1 that nitrogen addition causes a negative association between beta diversity and alpha diversity, we first computed the alpha and beta diversity of the AMF community. On one hand, AMF alpha diversity as measured by operational taxonomic unit (OTU) richness and Shannon’s index showed significant decreases by nitrogen addition (Fig. 1b, Supplementary Fig. S5). On the other hand, AMF beta diversity as measured by Jaccard, Bray–Curtis, and Sørensen dissimilarity indices showed significant increases by nitrogen addition (Fig. 1f–h). Taken together, for the increasing beta diversity and decreasing alpha diversity, we detected a negative correlation between them along the nitrogen addition gradient (Fig. 1i–k), thereby supporting our hypothesis-1.
Our finding of a negative correlation between decreasing alpha diversity and increasing beta diversity is consistent with the prediction made when nitrogen addition alleviates nitrogen limitation without causing phosphorus limitation. Our analysis of soil properties confirmed that nitrogen addition increased the concentration of soil available phosphorus by about eightfold (N0: 2.3 ± 0.25, N50: 17.5 ± 1.48 mg kg-1) in the semiarid grassland (Supplementary Fig. S6). As stated in the introduction, when nitrogen addition alleviates nitrogen limitation without inducing phosphorus limitation21,22, plants would reduce their dependence on mycorrhizal fungi for acquiring phosphorus and nitrogen via the mycorrhizal pathway37. On one hand, plants with reduced mycorrhizal dependence allocate a lower proportion of photosynthates to AMF, consequently leading to a shrinking niche that supports fewer AMF species. On the other hand, because plants with reduced mycorrhizal dependence impose weaker selective pressures on the AMF community, this decrease in host-mediated homogeneous selection forces thereby increases stochasticity in the AMF community (Supplementary Fig. S7). The generalizability of our findings is partially supported by Lu et al.18 who reported that nitrogen addition in a Tibetan meadow decreased AMF alpha diversity and increased AMF beta diversity as measured by the Sørensen index, but not that measured by the Bray–Cutis index. However, the association between AMF alpha diversity and beta diversity was not tested due to a short nitrogen addition gradient in Lu et al.’s study18, and a mechanistic explanation was not proposed due to the lack of information about phosphorus availability.
Alternatively, the response of the AMF community to nitrogen addition might be attributed to factors such as soil acidification, heavy metal contamination, or reduced plant diversity (see references in Supplementary Table S1). In our study, we also found that nitrogen addition reduced soil pH, increased the concentrations of Cu and Mn, and decreased plant species richness (Supplementary Fig. S8). However, under soil acidification and heavy metal release, the number of AMF species would be reduced due to the selection of stress tolerators, while AMF beta diversity would decline via homogeneous selection. The latter prediction conflicts with our finding that nitrogen addition increased AMF beta diversity (Fig. 1). For the reduction in plant diversity, AMF alpha diversity would decrease due to the loss of niches associated with different plant species, while AMF beta diversity would also decline as a result of homogeneous selection (see references in Supplementary Table S1). Again, the latter prediction conflicts with our finding that nitrogen addition increased AMF beta diversity (Fig. 1). Thus, in our study, the observed negative association between beta diversity and alpha diversity was unlikely due to soil acidification, heavy metal release, or plant diversity loss, factors which typically cause declines in both alpha or beta diversities15,19,38.
Since the responses of AMF alpha diversity, beta diversity, and community composition to nitrogen addition are consistent across 2017 and 2019 (Supplementary Fig. S9), we performed bacterial amplicon sequencing and metagenomic analysis on soil samples collected in 2019. This enabled us to further investigate how AMF and AMF-associated bacteria and functions respond to nitrogen addition. To explore the correlations between the AMF community, bacterial community, and bacterial functions, we performed a multiple co-inertia (MCo) analysis on three datasets: AMF community composition assessed via 18S rRNA gene amplicon sequencing, bacterial community composition assessed via 16S rRNA gene amplicon sequencing, and bacterial functional composition assessed via shotgun metagenomics. Our results revealed the coordinated response of AMF, bacterial community, and bacterial functions to nitrogen addition. In the MCo analysis, the first axis, explaining 44.56% of the total variance across datasets, was most strongly correlated with soil-available phosphorus (R² = 0.905, P < 0.001) (Fig. 2a–c). These results highlight the interconnected nature of soil nitrogen and phosphorus cycles and underscore the critical role of coordination between AMF and bacteria in mediating nitrogen- and phosphorus-driven ecological processes. We then conducted non-metric multidimensional scaling (NMDS) analysis to show that the AMF community, bacterial community, and bacterial functional composition were all significantly altered by nitrogen addition (Fig. 2d–f).
Fig. 2 Coordination of arbuscular mycorrhizal fungal (AMF) community structure with bacterial taxonomic and functional profiles association with soil available phosphorus. [Images not available. See PDF.]
a Multiple co-inertia (MCo) analysis of multi-omics datasets composed by bacterial functional composition (KOs, KEGG Orthologies) as detected from shotgun metagenome (open circles), bacterial community composition as measured by 16S rRNA gene amplicon sequencing (open triangles), and AMF community composition as measured by 18S rRNA gene amplicon sequencing (open squares), collected from 34 plots (n = 34) along a nitrogen addition gradient. Note that the first axis (MCo1) captured 44.56% of the total variation, followed by 18.5% of MCo2. b, c Strong association of MCo1 and soil available phosphorus. b The bar plot showing the strength (R2) of significant associations (P < 0.05) of MCo1 with plant and soil variables. Note that soil available phosphorus exhibited the strongest association with MCo1 (R2 = 0.91, P < 0.001). c Scatter plot with fitting curve showing the strong association of MCo1 and soil available phosphorus (n = 34). d–f The non-metric multidimensional scaling (NMDS) of AMF community, bacterial community, and bacterial KOs significantly changed by nitrogen addition levels (n = 34). g–i Co-occurrence networks showing significant associations between AMF operational taxonomic unit (OTUs) and KOs, between AMF OTUs and bacterial OTUs, and between subsets of bacterial KOs and subsets of OTUs, both of them significantly correlated with AMF OTUs.
Our previous studies had primarily focused on bacterial communities, demonstrating that nitrogen addition enriched functions related to translation and xenobiotics degradation in the overall microbial community4. Building upon these foundational findings, we employed a combined approach of co-occurrence network analysis and integrated metagenomic-amplicon sequencing to systematically characterize the responses of AMF and their associated microbiomes to nitrogen addition. We then constructed co-occurrence networks using the sparse inverse covariance estimation for ecological association inference (SPIEC-EASI) approach. The results showed significant correlations between 30 AMF OTUs and 1823 KOs, between 30 AMF OTUs and 823 bacterial OTUs, and between 823 bacterial OTUs and 1823 KOs (Fig. 2 g–i). This study, using the SPIEC-EASI algorithm consistent with previous studies to infer the cross-kingdom network of AMF and AMF-associated microbiomes36, identified the bacterial taxa and functional genes associated with AMF and further explored the synergistic roles of AMF-associated microbiomes in nutrient cycling and soil ecosystem dynamics.
To test our hypothesis-2, we explored the 11 KOs involved in phosphorus metabolism among the 1823 bacterial KOs correlated with 30 AMF OTUs (Fig. 3). We found that six genes exhibited increased abundances following nitrogen addition. Among them, one gene (olpA) is involved in organic phosphoester hydrolysis, one (ADE2) in purine metabolism, one (pckG) in pyruvate metabolism, one (rtpR) in pyrimidine metabolism, and two (phnS and phnV) in transporters. In contrast, five genes showed decreased abundances in response to nitrogen addition, with functions in organic phosphoester hydrolysis (phoN), phosphonate and phosphinate metabolism (phnH), purine metabolism (ppx and purD), and transport (glpTg) (Fig. 3a). Focusing on the phosphatase genes, we found that nitrogen addition increased the abundance of olpA and decreased that of phoN. Note that the abundance of the phoN gene was about tenfold higher than that of olpA gene (Fig. 3b). The phoN gene encodes phosphomonoesterase, a class A acid phosphatase that converts organic phosphorus into inorganic phosphates by breaking phosphoester bonds32. In this study, nitrogen addition increased soil phosphorus availability, enabling AMF and plants to directly access available phosphorus and reducing their reliance on bacteria for organic phosphorus mineralization via phosphomonoesterase, the enzyme encoded by the phoN gene. Extending our analysis to all phosphorus mineralization-related functional genes, seven were upregulated and only one downregulated under increasing nitrogen addition (Supplementary Fig. S10). Further supporting this, acid phosphatase activity remained unchanged with nitrogen addition (Fig. 3c), despite the net decrease in phosphatase gene abundance. Together, these findings indicate that nitrogen addition alleviates phosphorus limitation predominantly through soil acidification rather than enhanced microbial-mediated organic phosphoester hydrolysis. Crucially, our results provide no evidence for upregulated phosphorus mineralization functions in bacterial genes associated with the AMF community, thereby supporting our second hypothesis.
Fig. 3 No evidence of phosphorus mineralization function upregulation in bacterial genes correlated with arbuscular mycorrhizal fungal (AMF) community. [Images not available. See PDF.]
a The heatmap showing relative abundance of AMF correlated bacterial genes involving in phosphorus cycling in each plot (n = 34). Note total of six genes showed an increase in relative abundance upon nitrogen addition. Among them, one gene (olpA) is involved in organic phosphoester hydrolysis, one (ADE2) in purine metabolism, one (pckG) in pyruvate metabolism, one (rtpR) in pyrimidine metabolism, and two (phnS and phnV) in transporters. In contrast, five genes exhibited a decrease in relative abundance with nitrogen addition, which function in organic phosphoester hydrolysis (phoN), phosphonate and phosphinate metabolism (phnH), purine metabolism (ppx and purD), and transport (glpT). b, c Number of acid phosphatase genes and enzyme activity of acid phosphatase along a nitrogen addition gradient (n = 34). b Nitrogen addition increased the KO number of olpA and decreased that of phoN, two acid phosphatase genes. Note the number of phoN gene was about tenfold higher than olpA gene. c The measured activity of acid phosphatase showed no significant increase with nitrogen addition. d A schematic graph showing the up and down regulations of AMF correlated bacterial genes involving in phosphorus cycling in response to nitrogen addition. This graph displayed the chemical process catalyzed by the enzyme of the 11-phosphorus cycling related bacterial genes significantly correlated with AMF community as shown in Fig. 3a. Note we found no evidence of phosphorus mineralization function upregulation in bacterial genes associated with AMF community.
To explore the nitrogen response of AMF positively correlated microbial function, we firstly perform a differential analysis to detect KOs significantly upregulated or downregulated in N50 as compared to N0, followed by function enrichment analysis (Supplementary Fig. S11). We found that the KOs positively correlated with AMF and upregulated in N50 were enriched in multiple functions, including ABC transporters, histidine metabolism, valine, leucine and isoleucine biosynthesis, base excision repair, mismatch repair, pentose phosphate pathway, and nucleotide metabolism (Fig. 4a). The functional enrichment in amino acid synthesis and metabolism aligns with previous studies demonstrating that interactions between AMF and bacteria enhance organic nitrogen utilization7. On the other hand, the KOs negatively correlated with AMF and upregulated in N50 enriched functions related to polycyclic aromatic hydrocarbon degradation, naphthalene degradation, nitrotoluene degradation, bacterial secretion system, riboflavin metabolism, biofilm formation, phospholipase D signaling pathway, lipopolysaccharide biosynthesis, RNA polymerase, ethylbenzene degradation, nitrogen cycle, acarbose and validamycine biosynthesis (Supplementary Fig. S12). We also explored the nitrogen response of AMF-associated microbial taxonomic composition by performing differential analysis to detect bacterial OTUs up- or downregulated in N50 and other N treatments (N20, N15, N10, N5, N2) relative to N0 (Supplementary Fig. S13). We found that nitrogen addition increased the richness of the AMF-associated microbial taxa, while in the N50 treatment, there was a higher relative proportion of Actinobacteria members as compared to N0 (Fig. 4c, d, and Supplementary Fig. S14).
Fig. 4 Nitrogen response of positively AMF-associated bacterial genes enriched functions leading by ABC transport and amino acid metabolism. [Images not available. See PDF.]
a The 34 microbial KOs that are significantly higher in N50 (n = 5) than N0 (n = 4) enriched functions, including ABC transporters, histidine metabolism, valine leucine and isoleucine biosynthesis, base excision repair, mismatch repair, pentose phosphate pathway, and nucleotide metabolism. Note that we focused exclusively on the N50 level because we did not identify a sufficient number of differentially expressed genes for functional enrichment analysis in other nitrogen addition levels (Supplementary Fig. S11). b A heatmap illustrates the association between microbial genes underpinning these enriched functions and environmental variables. It highlights positive correlations with nitrogen addition level and soil available phosphorus, and negative correlations with soil pH and plant richness. c A bar plot demonstrates that there is a significant difference in the richness of AMF-associated bacteria between N50 (n = 5) and N0 (n = 4). d A stacked bar plot illustrates that in N50 (n = 5), the abundance of AMF-associated Actinobacteria was significantly higher compared to N0 (n = 4).
Our finding that nitrogen addition did not enrich the phosphorus mineralization function of the mycorrhiza-associated microbiome is consistent with the prediction made when nitrogen addition alleviates nitrogen limitation without causing phosphorus limitation (Supplementary Fig. S6). Because the ecosystem is not limited by phosphorus, it is not likely that biotic processes involving phosphorus mobilization and mineralization would be upregulated. Besides, the downregulation of the phosphorus mineralization function in the mycorrhizal-associated microbiome aligns with the condition where plants with lower mycorrhizal dependence allocate a smaller proportion of photosynthates to AMF and their associated microbiome.
Despite the obvious acidification of soil caused by nitrogen addition, we found no evidence for the enrichment of AMF-associated bacterial taxa or functions related to acid stress tolerance. For example, we found that the nitrogen-induced soil acidification favored AMF-associated bacterial taxa belonging to Actinobacteria, which are less tolerant to acidic conditions compared to Acidobacteria, while the relative abundance of acid-tolerant Acidobacteria remained unchanged. Meanwhile, we found that the nitrogen-driven soil acidification process favored AMF-associated functions of ABC transporters, histidine metabolism, valine, leucine and isoleucine biosynthesis, base excision repair, mismatch repair, pentose phosphate pathway, and nucleotide metabolism, none of which were likely involved in acid stress tolerance. When the soil is acidified by nitrogen addition, the hyphosphere of AMF might represent a benign microhabitat with a neutral pH, which might serve as a shelter for acid stress non-tolerant organisms without enriching genes underpinning acid stress tolerance.
Conclusion
In a semiarid grassland, we found that the nitrogen response of the AMF community was characterized by a negative association between beta diversity and alpha diversity. This pattern likely arises from niche contraction, reducing the number of coexisting species, coupled with relaxed resource limitation, enhancing stochastic processes within the AMF community. Multiple co-inertia analyses revealed a coordinated response of the AMF community, bacterial community, and bacterial functions to nitrogen addition, which as a whole was strongly predicted by soil-available phosphorus (R2 = 0.905, P < 0.001). Besides, we found no evidence for the enrichment of phosphorus mineralization functions in bacterial genes associated with the AMF community, despite the enrichment of functions related to transporters, amino acid synthesis and metabolism, and replication repair.
Materials and Methods
Site description and sampling
We conducted a field experiment in a temperate grassland near the Inner Mongolia Grassland Ecosystem Station (116°14’E, 43°13’N), located in the Inner Mongolia Autonomous Region, China. The study site receives an average annual precipitation of ~327.7 mm, with over 80% of rainfall occurring during the growing season from May to September. The average annual temperature at the study site is ~1.0 °C (1983–2018), with the coldest temperatures recorded in January (–21.4 °C) and the warmest temperatures in July (19.7 °C). Over the past 20 years, the average annual nitrogen deposition has remained below 2.0 g N m−2 yr−1, with no fertilizers were applied prior to the experiment. The soil at this site is classified as Calcic–Orthic Aridisol according to the US soil classification system. Dominant vegetation consists of Leymus chinensis and Stipa grandis, accompanied by common co-occurring species such as Artemisia frigida, Caragana microphylla, Potentilla bifurca, Carex korshinskyi, and Chenopodium glaucum. Since 1999, the experimental field site has been enclosed with a fence to prevent large animal grazing.
Each block comprises 36 plots (3.5 m × 8 m), representing various treatments in a full factorial design. These treatments include two mowing regimes (mown and unmown), two nitrogen addition frequencies (two or twelve times per year), and nine nitrogen addition levels (0, 1, 2, 3, 5, 10, 15, 20, and 50 g N m-2 yr-1), denoted as N0, N1, N2, N3, N5, N10, N15, N20, and N50, respectively. Our objective was to explore how the response of AMF to nitrogen addition is associated with the composition and function of the bacterial community. Therefore, we concentrated solely on different nitrogen addition levels, excluding the variables of mowing regime and nitrogen addition frequency. A previous study conducted at the same experimental site showed that low nitrogen addition levels (1, 2, 3 g N m-2 yr-1) had no significant effect on AMF diversity and composition15. Thus, in this study, we selected only seven nitrogen addition levels (0, 2, 5, 10, 15, 20, 50 g N m-2 yr-1). To reduce labor and costs, we chose five replicates out of the ten originally planned replicates. In September 2017 and 2019, we collected soil samples from plots subjected to seven distinct nitrogen addition treatments, with application rates of 0, 2, 5, 10, 15, 20, and 50 g N m-2 yr-1, respectively. For each treatment, five replicates were selected, yielding a total of 70 samples (Supplementary Fig. S1). We harvested the aboveground parts of all vascular plants within a 1 m × 1 m quadrat in each plot, oven-dried and weighed them in the laboratory to obtain the aboveground net primary productivity, while the number of species occurring in the 1 m × 1 m quadrat represented the plant richness of the plot. The soil cores from the same subplot were sieved through a 2 mm mesh, pooled together, and divided into three parts for analysis: one part stored at 4 °C for measuring soil acid phosphatase activity, soil water content, and inorganic nitrogen content; another part frozen at –80 °C for DNA extractions; and the third part air-dried for determining soil available phosphorus, pH and heavy metal ions (e.g., Cu, Mn).
Chemical analyses
Soil available phosphorus was determined using Olsen’s method: samples were extracted with a 0.5 M NaHCO3 solution and quantified by the molybdenum blue colorimetric method at 700 nm using a UV-visible spectrophotometer (Shimadzu UV-2550). Acid phosphatase activity was assessed through the fluorometric and colorimetric methods using a microplate reader. Soil slurries were prepared by blending 2.75 g of fresh soil with 91 ml of 50 mM acetate buffer at pH 5.0, and homogenizing for 1 min with a blender. Standard wells containing 4-methylumbelliferyl (MUB) or 7-amino-4-methylcoumarin (MUC) at concentrations ranging from 0 μM to 100 μM were prepared. Fluorescence levels were measured using a microplate fluorometer equipped with a 365 nm excitation and a 450 nm emission filter. Inorganic nitrogen in the soil was extracted using a 1 M KCl solution (w/v = 1:5), shaken at 180 rpm for 30 min, and filtered through a Whatman No. 1 filter paper (9 cm in diameter). The concentration of inorganic nitrogen was quantified using a flow injection analyzer (Autoanalyzer 3 SEAL, Bran and Luebbe, Norderstedt, Germany). Soil pH was measured using a pH meter (FE20-FiveEasy, Mettler Toledo, Columbus, Ohio, USA) at a soil-to-water ratio of 1:2.5. Heavy metals ions (Cu, Mn) in the soil were extracted using a 20 ml solution containing 0.005 M diethylenetriamine pentaacetate (DTPA), 0.01 M CaCl2, and 0.1 M triethanolamine (TEA) adjusted to pH 7.3, with the mixture shaken at 25 °C and180 rpm for 2 h, and filtered through Whatman No. 1 filter paper. The concentrations of heavy metal ions (Cu and Mn) were quantified using an inductively coupled plasma-optical emission spectrometer (ICP-OES; iCAP 6300, Thermo Scientific, USA).
DNA extraction and AMF 18S rRNA gene amplicon sequencing
Soil DNA was extracted from 0.5 g of previously frozen soil using the FastDNA® SPIN Kit (MP Biomedicals, Irvine, CA), according to the manufacturer’s protocol. The concentration and purity of DNA were assessed using a NanoDrop 2000 UV-vis spectrophotometer (Thermo Scientific, Wilmington, USA), while DNA integrity was evaluated using 1% agarose gels. For Illumina MiSeq sequencing, amplicons of the 18S rRNA gene were generated by a two-step PCR procedure, respectively, using GeoA2 and AML2 primers and NS31 and AMDGR primers39. PCR amplification was performed using the VeritiTM 96-Well Thermal Cycler (Thermo Fisher Scientific, Singapore) with cycling conditions including initial pre-denaturation at 98 °C for 10 s, 20 cycles of 98 °C for 10 s, 55 C for 30 s, and 72 °C for 40 s, and a final extension at 72 °C for 40 s. Each amplification was carried out in a 20 μl reaction mixture containing 2.5 μl template DNA (10 ng/μl), 2.5 μl forward and reverse primer (1 μM), 10 μl Phanta Flash Master Mix (2X), 2 μl BSA (2%), and 0.5 μl nuclease-free water. In the second round of PCR, both the forward and reverse primers were linked with 12-base barcode sequences for sample distinction. The second round of PCR amplification was performed with pre-denaturation at 98 °C for 30 s, followed by 30 cycles of 98 °C for 10 s, 55 °C for 30 s, and 72 °C for 40 s, and a final extension at 72 °C for 40 s. Each amplification was carried out in a 20 μl reaction mixture containing 2.5 μl product of the first round of PCR (diluted 50 times), 2.5 μl forward and reverse primer (1 μM), 10 μl buffer (2X), 2 μl BSA (2%), and 0.5 μl nuclease-free water. The concentration of PCR products was measured using the dsDNA HS Assay Kit for Qubit®. The quantified PCR products were pooled with the same molar amount (100 ng) from each sample. The mixed PCR products were purified using an E.Z.N.A Gel Extraction Kit (Omega Bio-Tek, Norcross, GA, USA). The purified products were sequenced on the Illumina MiSeq platform (Shanghai Personal Biotechnology Co. Ltd, Shanghai, China).
Forward and reverse reads were merged using the fastq_mergepairs command in usearch v11.040. Primers were removed using cutadapt v 4.441. The raw sequences were subjected to removal of low-quality sequences using fastx_filter command (-fastq_maxee 1.0 -fastq_qmax 42) in VSEARCH (v.2.221)42. The high-quality sequences were clustered into OTUs with a 97% sequence similarity threshold using the UPARSE pipeline in USEARCH v.11.0, which included dereplication and removal of singleton sequences before clustering. Representative sequences (the most abundant sequence from each OTU) were analyzed using the Basic Local Alignment Search Tool (BLAST) against both the National Center for Biotechnology Information (NCBI) nucleotide database and the MaarjAM 18S rRNA gene database. OTUs were considered AMF if their closest BLAST hit was annotated as ‘Glomeromycotina’ in both NCBI and MaarjAM with E-value < e-5043. It should be noted that measuring alpha diversity by OTUs has limitations. On the one hand, OTUs are grouped based on genetic sequence similarity, typically at 97% similarity, which means this approach may not precisely mirror true species diversity. On the other hand, sample biases in high-throughput sequencing technologies, like PCR amplification preferences, can cause the OTU data to be distorted, thus having an impact on the measurement of alpha diversity.
Bacterial 16S rRNA gene amplicon sequencing
The hypervariable region V3-V4 of the bacterial 16S rRNA gene was amplified using the primer pair 338F and 806R on a Gene Amplification PCR System (BioRad Laboratories Inc.), incorporating 8-base barcode sequences on both the forward and reverse primers for sample differentiation. PCR amplification was conducted in triplicate using the Gene Amplification PCR System (BioRad Laboratories Inc.), starting with an initial denaturation at 95 °C for 3 min, followed by 35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 10 min. Each amplification reaction comprised a 20 μl mixture containing 2 μl template DNA (10 ng), 0.8 μl forward and reverse primer (5 μM each), 2 μl of dNTPs (2.5 mM), 0.4 μl of TransStart FastPfu DNA Polymerase, 4 μl of 5 × TransStart FastPfu buffer, and 10 μl nuclease-free water. The AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) and Quantus™ Fluorometer (Promega, USA) were employed to both purify and quantify the PCR product following the manufacturer’s protocols. Purified amplicons were subsequently pooled at equimolar concentrations and subjected to paired-end sequencing on an Illumina MiSeq PE300 platform (Illumina, San Diego, USA) by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China), using standard protocols.
The raw 16S rRNA gene sequencing reads were demultiplexed, quality-filtered by the software of fastp v 0.20.044 and merged by Flash v 1.2.1145. OTUs with a 97% similarity cutoff were clustered using Uparse v 7.0.109035. Each bacterial OTU was taxonomically classified using RDP Classifier v 2.2 against the Silva database (Silva 128)46, with a confidence threshold set at 70%.
Metagenomic sequencing workflow and data processing
Thirty-five DNA samples from the 2019 sampling year were chosen for metagenomic analysis. The quality of DNA extracts was assessed on a 1% agarose gel, and their concentration and purity were determined using TBS-380 and NanoDrop2000. High-quality DNA samples were subsequently fragmented to an average size of ~400 bp using a Covaris M220 system (Gene Company Limited, China) in preparation for paired-end library construction. The NEXTflexTM Rapid DNA-Seq (Bioo Scientific, Austin, TX, USA) was employed for constructing the paired-end library. Adapters containing the full complement of sequencing primer hybridization sites were ligated to the blunt ends of the DNA fragments. Paired-end sequencing was conducted on an Illumina NovaSeq platform (Illumina Inc., San Diego, CA, USA) at Majorbio Bio-Pharm Technology Co., Ltd (Shanghai, China), using NovaSeq Reagent Kits.
Raw reads from shotgun metagenome sequencing were cleaned by removing adaptor sequences, trimming, and removing low-quality reads (reads with N bases, a minimum length threshold below 50 bp, and a minimum quality threshold less than 20) using fastp (v 0.20.0) on the Majorbio Cloud Platform. Each DNA sample yielded an average of ~10 GB of clean data. Subsequently, high-quality reads were assembled into contigs using Megahit v1.2.9 (parameters: k-mer = 21, 29, 39, 59, 79, 99, 119, 141)47. A non-redundant gene catalog was constructed using CD-hit v4.8.148 with criteria set at 95% similarity and 90% coverage. Gene abundance across samples was quantified using Salmon v1.5.149, and functional profiles or microbiomes were annotated using eggNOG databases v5.050.
Statistics and reproducibility
Amplicon analysis of AMF and soil/plant property measurements utilized 70 samples collected over two years (2017 and 2019) across seven nitrogen addition levels (0, 2, 5, 10, 15, 20, 50 g N m-2 yr-1; designated N0-N50) with five replicate blocks (70 samples = 2 years × 7 N addition levels × 5 replicates). Bacterial metagenome and amplicon analysis employed 35 samples collected from seven nitrogen addition levels (0, 2, 5, 10, 15, 20, 50 g N m-2 yr-1; designated N0-N50) with five replicate blocks in 2019 (35 samples = 7 N addition levels × 5 replicates). All statistical analyses were performed in R (v 4.2.2) (R Core Team 2025). The 16S rRNA gene amplicon data, the 18S rRNA gene amplicon data, and the shotgun metagenomics data were rarefied to the minimum read number, respectively, using the rrarefy command in the vegan package51, to account for the variation in read number among samples. Specifically, the bacterial community data was rarefied to 14,914 reads per sample, the AMF community data was rarefied to 239 reads per sample, and the shotgun metagenomics data was rarefied to 468,055 reads per sample. Note that shotgun metagenomic data offers an opportunity to validate the consistency of bacterial composition detected by 16S rRNA amplicon sequencing. Although shotgun metagenomics has gained significant popularity recently, amplicon sequencing remains the most accessible approach, and in our study, the results from 16S rRNA gene sequencing can be compared with those from previous studies.
Principal coordinate (PCo) analysis, followed by permutational analysis of variance (permANOVA), was performed to test and visualize the variation in AMF community composition and beta-dispersion caused by nitrogen addition, based on Jaccard dissimilarity, Bray–Curtis, or Sørensen distances in the Ape and vegan packages51. For AMF alpha diversity analysis, we calculated richness and Shannon’s diversity indices using the diversity function in the vegan package based on our 18S rRNA gene amplicon dataset. To assess beta diversity patterns, we employed the betadisper function in vegan package, which quantifies community dispersion through the distance of individual samples to their group centroid in principal coordinates analysis (PCoA) space. Relationships between AMF alpha diversity and nitrogen addition levels, between AMF beta diversity and nitrogen addition levels, and between AMF alpha diversity and AMF beta diversity were tested by Spearman’s correlation analysis, using corr.test function in the psych package52. We employed the iCAMP (Inferring Community Assembly Mechanisms by Phylogenetic-bin-based null model analysis) framework to quantitatively infer the assembly mechanisms of AMF communities53. The observed AMF taxa were first grouped into phylogenetic bins, followed by null model analysis using the beta net relatedness index (βNRI) and the modified Raup-Crick metric (RC) to determine the dominant assembly processes. Specifically, homogeneous selection (HoS) was inferred when βNRI < –1.96, while heterogeneous selection (HeS) was identified when βNRI > 1.96. For cases where |βNRI| ≤ 1.96, RC values further employed to classify as homogenizing dispersal (HD, RC < –0.95), dispersal limitation (DL, RC > 0.95), and drift (DR, |RC| ≤ 0.95). Finally, the relative importance of each ecological process at the whole-community level was estimated by aggregating the abundance-weighted proportions of each process across all bins.
Multiple co-inertia (MCo) analysis in the ade4 package54 was used to identify co-relationships of multi-omics datasets composed of AMF taxonomic, bacterial taxonomic, and functional composition. The first dimension of MCo (MCo1) and second dimension of MCo (MCo2) were extracted. Relationships between MCo1, MCo2, and plant and soil variables were tested by Spearman’s correlation analysis, using corr.test function in the psych package52. The NMDS analysis in the vegan package51 was used to visualize AMF taxonomic, bacterial taxonomic, and functional composition. Since the method of SPIEC-EASI often shows robustness in handling the bias of community composition55,56, we used this method to conduct network analysis among AMF OTUs, bacterial OTUs, and bacterial functional genes. The meta-matrix was produced using the R package SpiecEasi55, which applies Least Absolute Shrinkage and Selection Operator (LASSO) regularization and cross-validation to identify the most parsimonious network structure in high-dimensional microbial datasets. A lambda ratio of 0.01 was used, and network assessment was conducted over 50 lambda values across 100 cross-validation permutations, selecting the least variable links with the Stability Approach to Regularization Selection (StARS) criterion. The Meinshausen and Bühlmann graph estimation method was employed to estimate networks during each permutation. The networks were visualized using Cytoscape (v3.10.1)57. It is critical to acknowledge that the co-occurrence networks inferred via the SPIEC-EASI method (used to explore ecological associations in this study) may not fully capture real-world interactions, as correlations identified by this approach could be confounded by indirect ecological links or environmental factors.
A heatmap was constructed to depict the abundance of phosphorus functional genes over nitrogen addition levels using the pheatmap package58. The relationship between acid phosphatase genes and nitrogen addition levels was visualized using a scatterplot in ggplot259. We used the hypergeometric test to test the KO pathway difference or bacteria taxa difference between N50 and N0 using the stats package59. The enrichment of the KO pathway was visualized by scatterplot in the ggplot2 package60. Spearman’s correlation analysis in the psych package52 was further used to test the relationship between the relative abundance of functional genes and plant and soil variables. The richness of AMF-associated bacteria or the abundance of AMF-associated bacteria phyla between N50 and N0 were visualized by bar plot or stacked bar plot in the ggplot2 package60.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Acknowledgements
This work was financially supported by the National Natural Science Foundation of China (32301353, 32322053, 32301355), China Postdoctoral Science Foundation (2023M733731, GZC20232981), and Central Asian Drug Discovery and Development Centre of the Chinese Academy of Sciences (CAM202202). We are grateful to Hui Li, Ran Guo, Yan Shen, and Liangchao Jiang for their help during the course of the experiment. We thank the Inner Mongolia Grassland Ecosystem Research Station for logistical support. We thank all members of the Gao Lab for their discussion and communications. We thank Nancy Collins Johnson for the valuable feedback on the manuscript.
Author contributions
Jianxia Yang, Xingguo Han, and Cheng Gao conceived and designed the study. Jianxia Yang, Yang Peng, Junjie Yang, and Yunhai Zhang performed the experiments. Jianxia Yang, Yang Peng, Qiang Dong, and Qiushi Li analyzed the output data. Jianxia Yang, Yang Peng and Cheng Gao wrote the first draft of the manuscript.
Peer review
Peer review information
Communications Biology thanks Shengjing Jiang, Sulaimon Basiru and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: David Favero. A peer review file is available.
Data availability
The raw reads from the 16S and 18S amplicon sequencing and metagenomic data have been deposited in the NCBI Sequence Read Archive (SRA) under accession numbers PRJNA807483, PRJNA1249125, and PRJNA1084095. Analysis data supporting this study’s findings are openly available in Zenodo at https://zenodo.org/records/16247763 (https://doi.org/10.5281/zenodo.16247763) and in the Supplementary Data.
Code availability
Code is available from Zenodo: https://zenodo.org/records/16247763 (https://doi.org/10.5281/zenodo.16247763).
Competing interests
The authors declare no competing interests.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1038/s42003-025-08681-w.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1. Rockström, J. et al. A safe operating space for humanity. Nature461, 472–475 (2019).
2. Sala, OE et al. Global biodiversity scenarios for the year 2100. Science; 2000; 287, pp. 1770-1774.1:CAS:528:DC%2BD3cXhvVWltLk%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10710299]
3. Zhou, Z; Wang, C; Luo, Y. Meta-analysis of the impacts of global change factors on soil microbial diversity and functionality. Nat. Commun.; 2020; 11, 1:CAS:528:DC%2BB3cXht1Srs7jN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32555185][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7300008]3072.
4. Yang, JX et al. Gene horizontal transfers and functional diversity negatively correlated with bacterial taxonomic diversity along a nitrogen gradient. NPJ Biofilms Microbiomes; 2024; 10, 1:CAS:528:DC%2BB2cXisFWmtb7K [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39550371][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11569254]128.
5. Tisserant, E et al. Genome of an arbuscular mycorrhizal fungus provides insight into the oldest plant symbiosis. Proc. Natl Acad. Sci. USA; 2013; 110, pp. 20117-20122.1:CAS:528:DC%2BC3sXhvFKms7fF [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/24277808][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3864322]
6. Zhang, L; Zhou, J; George, TS; Limpens, E; Feng, G. Arbuscular mycorrhizal fungi conducting the hyphosphere bacterial orchestra. Trends Plant Sci.; 2022; 27, pp. 402-411.1:CAS:528:DC%2BB3MXisVOksr%2FI [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34782247]
7. Rozmos, M et al. Organic nitrogen utilisation by an arbuscular mycorrhizal fungus is mediated by specific soil bacteria and a protist. ISME J.; 2022; 16, pp. 676-685.1:CAS:528:DC%2BB38Xjt1ags7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34545172]
8. Kakouridis, A et al. Arbuscular mycorrhiza convey significant plant carbon to a diverse hyphosphere microbial food web and mineral-associated organic matter. New Phytol.; 2024; 242, pp. 1661-1675.1:CAS:528:DC%2BB2cXksFChur0%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38358052]
9. Bunn, RA; Simpson, DT; Bullington, LS; Lekberg, Y; Janos, DP. Revisiting the ‘direct mineral cycling’ hypothesis: arbuscular mycorrhizal fungi colonize leaf litter, but why?. ISME J.; 2019; 13, pp. 1891-1898. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30911130][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6775977]
10. Wang, F et al. Exploring the secrets of hyphosphere of arbuscular mycorrhizal fungi: processes and ecological functions. Plant Soil; 2022; 481, pp. 1-22.1:CAS:528:DC%2BB38XhvF2ls73E
11. Basiru, S; Ait Si Mhand, K; Hijri, M. Disentangling arbuscular mycorrhizal fungi and bacteria at the soil-root interface. Mycorrhiza; 2023; 33, pp. 119-137. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36961605]
12. Jansa, J; Bukovska, P; Gryndler, M. Mycorrhizal hyphae as ecological niche for highly specialized hypersymbionts - or just soil free-riders?. Front Plant Sci.; 2013; 4, 134. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23720665][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3655320]
13. Salvioli, A et al. Symbiosis with an endobacterium increases the fitness of a mycorrhizal fungus, raising its bioenergetic potential. ISME J.; 2016; 10, pp. 130-144.1:CAS:528:DC%2BC2MXitVymtb3E [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26046255]
14. Jiang, S et al. Dynamics of arbuscular mycorrhizal fungal community structure and functioning along a nitrogen enrichment gradient in an alpine meadow ecosystem. New Phytol.; 2018; 220, pp. 1222-1235. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29600518]
15. Wu, H et al. Identifying thresholds of nitrogen enrichment for substantial shifts in arbuscular mycorrhizal fungal community metrics in a temperate grassland of northern China. New Phytol.; 2023; 237, pp. 279-294.1:CAS:528:DC%2BB38XislGhs7jN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36177721]
16. Shi, G. et al. Nitrogen addition drives changes in arbuscular mycorrhizal fungal richness through changes in plant species richness in revegetated alpine grassland. Fungal Ecol.67https://doi.org/10.1016/j.funeco.2023.101303 (2024).
17. Han, Y; Feng, J; Han, M; Zhu, B. Responses of arbuscular mycorrhizal fungi to nitrogen addition: a meta-analysis. Glob. Change Biol.; 2020; 26, pp. 7229-7241.
18. Lu, Y., Liu, X. & Zhou, S. Nitrogen addition and arbuscular mycorrhizal fungi beta diversity: patterns and mechanisms. Front. Environ. Sci.9https://doi.org/10.3389/fenvs.2021.701653 (2021).
19. Cheng, Y et al. Long-term nitrogen fertilization alters phylogenetic structure of arbuscular mycorrhizal fungal community in plant roots across fine spatial scales. Plant Soil; 2022; 483, pp. 427-440.
20. Williams, A; Manoharan, L; Rosenstock, NP; Olsson, PA; Hedlund, K. Long-term agricultural fertilization alters arbuscular mycorrhizal fungal community composition and barley (Hordeum vulgare) mycorrhizal carbon and phosphorus exchange. New Phytol.; 2017; 213, pp. 874-885.1:CAS:528:DC%2BC28XitFGgtLrL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27643809]
21. Wang, R et al. Nitrogen enrichment buffers phosphorus limitation by mobilizing mineral-bound soil phosphorus in grasslands. Ecology; 2022; 103, e3616. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34923633]
22. Chen, J et al. Long-term nitrogen loading alleviates phosphorus limitation in terrestrial ecosystems. Glob. Change Biol.; 2020; 26, pp. 5077-5086.
23. Li, Y; Niu, S; Yu, G. Aggravated phosphorus limitation on biomass production under increasing nitrogen loading: a meta-analysis. Glob. Change Biol.; 2016; 22, pp. 934-943.
24. Cao, J. et al. Warming exerts a stronger effect than nitrogen addition on the soil arbuscular mycorrhizal fungal community in a young subtropical Cunninghamia lanceolata plantation. Geoderma367https://doi.org/10.1016/j.geoderma.2020.114273 (2020).
25. Egerton-Warburton, LM; Johnson, NC; Allen, EB. Mycorrhizal community dynamics following nitrogen fertilization: a cross-site test in five grasslands. Ecol. Monogr.; 2007; 77, pp. 527-544.
26. Wang, Z et al. Forms of nitrogen inputs regulate the intensity of soil acidification. Glob. Change Biol.; 2023; 29, pp. 4044-4055.1:CAS:528:DC%2BB3sXpsFyju7Y%3D
27. Zhang, Y et al. Rapid plant species loss at high rates and at low frequency of N addition in temperate steppe. Glob. Change Biol.; 2014; 20, pp. 3520-3529.
28. Bowman, WD et al. Limited ecosystem recovery from simulated chronic nitrogen deposition. Ecol. Appl.; 2018; 28, pp. 1762-1772. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30179279]
29. Hestrin, R; Hammer, EC; Mueller, CW; Lehmann, J. Synergies between mycorrhizal fungi and soil microbial communities increase plant nitrogen acquisition. Commun. Biol.; 2019; 2, 233. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31263777][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6588552]
30. Zhang, L et al. Carbon and phosphorus exchange may enable cooperation between an arbuscular mycorrhizal fungus and a phosphate-solubilizing bacterium. New Phytol.; 2016; 210, pp. 1022-1032.1:CAS:528:DC%2BC28XlvF2kt7k%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27074400]
31. Li, X et al. Mycorrhiza-mediated recruitment of complete denitrifying Pseudomonas reduces N(2)O emissions from soil. Microbiome; 2023; 11, 1:CAS:528:DC%2BB3sXlsFSgsbs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36890606][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9996866]45.
32. Wang, G; Jin, Z; George, TS; Feng, G; Zhang, L. Arbuscular mycorrhizal fungi enhance plant phosphorus uptake through stimulating hyphosphere soil microbiome functional profiles for phosphorus turnover. New Phytol.; 2023; 238, pp. 2578-2593.1:CAS:528:DC%2BB3sXjsFKms70%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36694293]
33. Deveau, A et al. Bacterial-fungal interactions: ecology, mechanisms and challenges. FEMS Microbiol. Rev.; 2018; 42, pp. 335-352.1:CAS:528:DC%2BC1MXltFKhsQ%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29471481]
34. Zhang, C et al. A tripartite bacterial-fungal-plant symbiosis in the mycorrhiza-shaped microbiome drives plant growth and mycorrhization. Microbiome; 2024; 12, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38243337][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10799531]13.
35. Wang, L; Zhang, L; George, TS; Feng, G. A core microbiome in the hyphosphere of arbuscular mycorrhizal fungi has functional significance in organic phosphorus mineralization. New Phytol.; 2023; 238, pp. 859-873.1:CAS:528:DC%2BB38XjtFeltLjE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36444521]
36. Qiu, T et al. Arbuscular mycorrhizal fungal interactions bridge the support of root-associated microbiota for slope multifunctionality in an erosion-prone ecosystem. Imeta; 2024; 3, 1:CAS:528:DC%2BB2MXpvFGms7Y%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38898982][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11183171]e187.
37. Johnson, NC. Resource stoichiometry elucidates the structure and function of arbuscular mycorrhizas across scales. New Phytol.; 2010; 185, pp. 631-647.1:CAS:528:DC%2BC3cXitFKnsrs%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19968797]
38. Ma, X et al. Global negative effects of nutrient enrichment on arbuscular mycorrhizal fungi, plant diversity and ecosystem multifunctionality. New Phytol.; 2021; 229, pp. 2957-2969.1:CAS:528:DC%2BB3MXnvVCku74%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33188641]
39. Babalola, BJ et al. Nitrogen fertilisation disrupts the temporal dynamics of arbuscular mycorrhizal fungal hyphae but not spore density and community composition in a wheat field. New Phytol.; 2022; 234, pp. 2057-2072. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35179789]
40. Edgar, RC. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods; 2013; 10, pp. 996-998.1:CAS:528:DC%2BC3sXht1ylsLzE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23955772]
41. Martin, M. CUTADAPT removes adapter sequences from high-throughput sequencing reads. EMBnet J.17, 10–12 (2011).
42. Rognes, T; Flouri, T; Nichols, B; Quince, C; Mahe, F. VSEARCH: a versatile open source tool for metagenomics. PeerJ.; 2016; 4, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27781170][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5075697]e2584.
43. Davison, J et al. Global assessment of arbuscular mycorrhizal fungus diversity reveals very low endemism. Science; 2015; 349, pp. 970-973.1:CAS:528:DC%2BC2MXhsVSlsr3O [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26315436]
44. Chen, S; Zhou, Y; Chen, Y; Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics; 2018; 34, pp. i884-i890. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30423086][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6129281]
45. Magoc, T; Salzberg, SL. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics; 2011; 27, pp. 2957-2963.1:CAS:528:DC%2BC3MXhtlGkur7M [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21903629][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3198573]
46. Quast, C et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res.; 2013; 41, pp. D590-D596.1:CAS:528:DC%2BC38XhvV2ksb%2FN [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23193283]
47. Li, D; Liu, C-M; Luo, R; Sadakane, K; Lam, T-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics; 2015; 31, pp. 1674-1676.1:CAS:528:DC%2BC28XhtFyltL3N [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25609793]
48. Fu, L; Niu, B; Zhu, Z; Wu, S; Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics; 2012; 28, pp. 3150-3152.1:CAS:528:DC%2BC38XhslKnsb%2FE [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23060610][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3516142]
49. Patro, R; Duggal, G; Love, MI; Irizarry, RA; Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods; 2017; 14, pp. 417-419.1:CAS:528:DC%2BC2sXltVWgtL8%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28263959][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5600148]
50. Huerta-Cepas, J et al. eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. Nucleic Acids Res.; 2019; 47, pp. D309-D314.1:CAS:528:DC%2BC1MXhs1Gqtr%2FL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30418610]
51. Dixon, P. VEGAN, a package of R functions for community ecology. J. Veg. Sci.; 2003; 14, pp. 927-930.
52. Revelle, W. psych: procedures for personality and psychological research, Northwestern University, Evanston, Illinois, USA, https://CRAN.R-project.org/package=psych Version = 2.2.9. (2022).
53. Ning, D et al. A quantitative framework reveals ecological drivers of grassland microbial community assembly in response to warming. Nat. Commun.; 2020; 11, 1:CAS:528:DC%2BB3cXhvFSis7bK [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32948774][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7501310]4717.
54. Dray, S; Dufour, AB. The ade4 package: Implementing the duality diagram for ecologists. J. Stat. Softw.; 2007; 22, pp. 1-20.
55. Kurtz, ZD et al. Sparse and compositionally robust inference of microbial ecological networks. PLoS Comput. Biol.; 2015; 11, e1004226. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25950956][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4423992]
56. Layeghifard, M; Hwang, DM; Guttman, DS. Disentangling interactions in the microbiome: a network perspective. Trends Microbiol.; 2017; 25, pp. 217-228.1:CAS:528:DC%2BC28XhvFCgt7zL [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27916383]
57. Shannon, P et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res.; 2003; 13, pp. 2498-2504.1:CAS:528:DC%2BD3sXovFWrtr4%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/14597658][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC403769]
58. R. Kolde, M. R. K. Package ‘pheatmap’. R package. (2015).
59. R Core Team, R. R: A language and environment for statistical computing. (2013).
60. Wickham, H. ggplot2. Wiley Interdiscip. Rev. Comput. Stat.; 2011; 3, pp. 180-185.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the "License"). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Mycorrhiza interplays with the microbiome in adaptation to environmental fluctuation, yet how arbuscular mycorrhizal fungi (AMF) and the associated microbiome respond to nitrogen addition remains poorly understood. Here, we addressed this gap by conducting amplicon sequencing of AMF 18S rRNA and bacterial 16S rRNA operons, along with shotgun metagenome sequencing, using soil samples collected from a semiarid grassland that has received nitrogen inputs for 11 years at different levels. We found that the nitrogen response of the AMF community was characterized by a negative association whereby increasing nitrogen addition leads to higher beta diversity and lower alpha diversity. Multiple co-inertia analyses revealed a coordinated response of the AMF community, bacterial community, and bacterial functions to nitrogen addition, which as a whole was strongly related to soil phosphorus availability. Besides, through network analysis of AMF with bacteria and bacterial functional genes, we found that nitrogen addition selected Actinobacteria and enriched functions of transporters, amino acid synthesis and metabolism, and replication repair, whereas there was no evidence for the enrichment of phosphorus mineralization functions.
A study on steppe grasslands found that nitrogen addition alters arbuscular mycorrhizal fungi and bacterial networks but does not promote phosphorus mineralization.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
; Dong, Qiang 1 ; Li, Qiu-Shi 4 ; Han, Xing-Guo 5
; Gao, Cheng 4
1 State Key Laboratory of Microbial Diversity and Innovative Utilization, Institute of Microbiology, Chinese Academy of Sciences, Beijing, China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309)
2 School of Grassland Science, Beijing Forestry University, Beijing, China (ROR: https://ror.org/04xv2pc41) (GRID: grid.66741.32) (ISNI: 0000 0001 1456 856X); State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Beijing, China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309)
3 State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, Chinese Academy of Sciences, Beijing, China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309)
4 State Key Laboratory of Microbial Diversity and Innovative Utilization, Institute of Microbiology, Chinese Academy of Sciences, Beijing, China (ROR: https://ror.org/034t30j35) (GRID: grid.9227.e) (ISNI: 0000000119573309); University of Chinese Academy of Sciences, Beijing, China (ROR: https://ror.org/05qbk4x57) (GRID: grid.410726.6) (ISNI: 0000 0004 1797 8419)
5 College of Life Sciences, Hebei University, Baoding, China (ROR: https://ror.org/01p884a79) (GRID: grid.256885.4) (ISNI: 0000 0004 1791 4722)