Article | Open

Consistent effects of nitrogen fertilization on soil bacterial communities in black soils for two crop seasons in China

  • Scientific Reports 7, Article number: 3267 (2017)
  • doi:10.1038/s41598-017-03539-6
  • Download Citation
Published online:


Long-term use of inorganic nitrogen (N) fertilization has greatly influenced the bacterial community in black soil of northeast China. It is unclear how N affects the bacterial community in two successive crop seasons in the same field for this soil type. We sampled soils from a long-term fertilizer experimental field in Harbin city with three N gradients. We applied sequencing and quantitative PCR targeting at the 16S rRNA gene to examine shifts in bacterial communities and test consistent shifts and driving-factors bacterial responses to elevated N additions. N addition decreased soil pH and bacterial 16S rDNA copy numbers, and increased soil N and crop yield. N addition consistently decreased bacterial diversity and altered bacterial community composition, by increasing the relative abundance of Proteobacteria, and decreasing that of Acidobacteria and Nitrospirae in both seasons. Consistent changes in the abundant classes and genera, and the structure of the bacterial communities across both seasons were observed. Our results suggest that increases in N inputs had consistent effects on the richness, diversity and composition of soil bacterial communities across the crop seasons in two continuous years, and the N addition and the subsequent edaphic changes were important factors in shaping bacterial community structures.


The global ecosystem is receiving elevated levels of nitrogen (N), often >100 kg N ha−1 y−1, in order to achieve high crop yields1, 2. The average global rate of N deposition to terrestrial ecosystems is likely to increase by a factor of 2.5 over the next century3. Previous studies have shown that elevated N additions to ecosystems lead to climate change, emission of greenhouse gases, species extinction and even human health threats4. A significant increase in carbon (C) storage in the form of plant biomass5 and productivity6, but a loss of plant community diversity7 is commonly observed across most experimental N gradients in ecosystems. Furthermore, increased N additions change microbial activities by altering microbial biomass8, 9 and bacterial10, 11 and fungal communities12, 13, and thereafter change their subsequent ecological function. In addition to N, crop rotations14, precipitation15, temperature16 and season17 strongly affect the composition, richness and diversity of soil microbial communities. Therefore, it is likely that research using single time-point sampling cannot differentiate the influences causing variation in microbial community structure. Understanding the bacterial community in response to N gradients and seasonal variations will help determine the ‘real’ effect of N and guide us in the appropriate development of agricultural ecological systems for food production.

Black soil, belonging to the pachic Haploborolls subtype of Haploborolls in the Boroll suborder, is one of the most fertile soils in China, and the agricultural area with this soil is very important for grain production and cultivation17. Very high inputs of inorganic fertilizers in this region have seriously degraded the soil physicochemical properties and environmental health since the 1950s18. These nutrient-based alterations of microbiota are reflected in significant shifts in the soil N-cycling and wider communities, including denitrifying bacteria18, 19, ammonia-oxidizing bacteria20 and the whole bacterial kingdom11. Ramirez et al.21 observed consistent effects of N fertilization on microbial respiration regardless of soil and N fertilizer type. However, in black soils, whether such effects of N occur over two continuous crop seasons remains unclear.

Long-term fertilization trials in Heilongjiang Province, China, allow investigations of the effects of repeated additions of N fertilizer on soil microorganisms in black soil. In this experimental field, the influence of higher levels of N and phosphorus (P) fertilizer treatments on bacterial communities was greater than that of lower levels after wheat harvest in 2013 in our previous study11. However, soybean was cultivated in 2014 and the weather, including mean annual temperature and precipitation, differed from that in 2013. Meanwhile, the soil concentration of nitrate (NO3) – shown to be an important factor in shaping bacterial communities after 34 years of N and P addition in our previous study11 significantly differed in the two crop seasons. In the present study, using experimental gradients of N, the specific questions addressed were as follows: (i) Could N additions have consistent effects on the abundance and composition of the entire soil bacterial kingdom and the dominant phyla/classes/genera in both crop seasons? (ii) Will shifts in specific bacterial taxa correspond with the fertilizer-induced changes in soil pH and/or NO3 concentration? Thus, we used long-term N experiments to assess the influence of N additions on bacterial community abundance, composition and diversity using real-time PCR coupled with pyrosequencing targeting the bacterial 16S rRNA gene.


Soil and crop responses

In the wheat season, the N fertilization significantly (P < 0.05) affected the soil properties and wheat yields. Soil dissolved inorganic N (NH4+ and NO3) was generally correlated with the N gradient, in the ranges of 33.8–40.7 and 7.7–24.5 mg kg−1 soil, respectively, accompanied by a corresponding decrease in soil pH from 6.5 to 4.6. The N addition also induced significant increases in soil Avail P from 9.3 to 15.3 mg/g soil and OM from 27.3 to 29.5 g kg−1 soil. From low to high fertilization, wheat yield significantly increased from 1548 to 2155 kg ha−1 (Table S1).

In the soybean season, the three soil properties (pH, Avail P and NO3) and soybean yield were also significantly correlated with the N fertilization regime (P < 0.05). For low and high fertilization, pH was 6.5 and 5.4, respectively, corresponding soil NO3 content was 2.4 and 5.6 mg g−1 soil, and soil Avail P was 1.8 and 64.9 mg g−1 soil. From low to high fertilization, soybean yield significantly increased from 1800 to 2761 kg ha−1 (Table S1).

Effects of long-term N inputs on bacterial 16S rDNA copy numbers

The long-term N inputs significantly decreased the number of bacterial 16S rDNA copies (Fig. 1). These effects were significant (P < 0.01) on gene copy numbers/g soil in both wheat (r = −0.949) and soybean seasons (r = −0.970). For low and high N inputs, the numbers of bacterial 16S rDNA copies g−1 soil were 4.28 × 109 and 2.17 × 109 in the wheat season and 1.79 × 1010 and 9.35 × 109 in the soybean season, respectively (Fig. 1).

Figure 1
Figure 1

Abundance of bacteria as indicated by the numbers of 16S rDNA copies measured using quantitative PCR. sCK, sN1 and sN2 indicate different N treatments in the soybean season; wCK, wN1 and wN2 indicate different N treatments in the wheat season. The ‘r’ indicates Pearson’s correlation coefficient between N added and 16S rDNA copies (**P < 0.01, *P < 0.05).

Differences in bacterial alpha diversity between different N fertilizer regimes

After OTU picking and chimera checking, a total of 2,078,821 high-quality sequences were assigned to 5148 non-singleton OTUs, resulting in the classification of 533 taxa at genus level. The numbers of OTUs in soil samples were in the range of 1624–2467 in the six fertilizer treatments (Table S2). The Good’s coverage values were in the range of 85.5–96.0% at 97% similarity cutoff, indicating sufficient sequence reads to capture the bacterial diversity in these soils (Table S2).

Several alpha diversity measures were calculated, including estimated OTUs, Shannon’s diversity index (Fig. 2A) and Chao1 (Fig. 2B). For community Shannon diversity comparison, lower N addition samples (N1) had significantly lower Shannon values than those without N addition (CK) in both crop seasons (P < 0.05, Fig. 2A). Moreover, in the wheat season, the higher N addition (N2) led to significantly lower Shannon values than N1 samples did (P < 0.01, Fig. 2A). In the soybean season, Shannon values in N2 samples were lower than those in N1, although there was no significant difference between them. Shannon values in the two crop seasons were significantly negatively correlated with N fertilization inputs (P < 0.01, for both cases; Table S2).

Figure 2
Figure 2

Differences in bacterial community diversity and richness between different N fertilizer regimes. (A) Community diversity between different N fertilizer regimes (both wheat and soybean seasons). (B) Community richness between different N fertilizer regimes (both wheat and soybean seasons). Asterisks show significant differences between samples (**P < 0.01, *P < 0.05, Tukey r-test). sCK, sN1 and sN2 indicate different N treatments in the soybean season; wCK, wN1 and wN2 indicate different N treatments in the wheat season.

There was no significant difference in community richness between N1 and CK samples, while N2 samples had significantly (P < 0.01) lower richness (Chao1) than N1 samples (both wheat and soybean seasons) (Fig. 2B).

Differences in bacterial composition

Analysis of the community composition under different N fertilizer regimes in the two crop seasons showed that the top abundant 20 phyla accounted for 99.7% of the reads (Fig. 3). Proteobacteria was the most abundant phylum, representing 28.5–44.5% of the total sequences (Fig. S1). The bacterial communities in N2 samples differed from those in N1 and CK samples by showing higher relative abundances of Proteobacteria in both wheat and soybean seasons (Figs 3 and S1). Classes Alphaproteobacteria, Gammaproteobacteria and Deltaproteobacteria occupied 47.9, 13.0 and 11.7% of the sequences of this phylum, respectively (data not shown). With increasing N, the relative abundance of Alphaproteobacteria and Gammaproteobacteria increased significantly, and were highly positively correlated with N addition in wheat and soybean seasons (r = 0.949, P < 0.01, for both cases; Fig. S2A and B). However, the relative abundance of Deltaproteobacteria decreased significantly with increasing N for wheat (r = −0.949, P < 0.01) and soybean seasons (r = −0.738, P < 0.01; Fig. S2E). In the Alphaproteobacteria class, genera Devosia and Sphingomonas were significantly positively correlated with N addition (r = 0.944 and 0.764, respectively; P < 0.05), whereas Balneimonas showed the opposite trend (r = −0.832, P < 0.01; Fig. 4).

Figure 3
Figure 3

Bacterial compositions of different N fertilizer regimes. Each bar represents the average relative abundance of each bacterial taxon within a group. The top 20 abundant phyla are shown (relative abundance >0.01%). sCK, sN1 and sN2 indicate different N treatments in the soybean season; wCK, wN1 and wN2 indicate different N treatments in the wheat season.

Figure 4
Figure 4

Changes in the relative abundances of bacterial genera across the N gradients in the wheat (A) and soybean (B) seasons. Only shown are those classes correlated with N added (Spearman’s r-values). Bars indicate one standard deviation. Asterisks show significant correlations (**P < 0.01, *P < 0.05). sCK, sN1 and sN2 indicate different N treatments in the soybean season; wCK, wN1 and wN2 indicate different N treatments in the wheat season.

Acidobacteria was the second most abundant phylum in the sampled soils, and constituted 10.5–15.6% of total sequences (Fig. 3). The N addition significantly (P < 0.01) decreased the relative abundance of this phylum in both seasons (Fig. S1). Class Solibacteres occupied 19.4% of sequences of this phylum and was positively correlated with N addition in wheat (r = 0.843, P < 0.01) and soybean seasons (r = 0.896, P < 0.01; Fig. S2D). Classes Chloracidobacteria and Acidobacteria occupied 33.8 and 17.5% of sequences of this phylum, respectively (data not shown) and were negatively correlated with N addition in both seasons (P < 0.01 for both; Fig. S2F and G). At genus level, N addition significantly (r = 0.920, P < 0.01) increased the relative abundance of Koribacter (Fig. 4).

Actinobacteria constituted only 9.5–14.8% of the total sequences (Tables S4 and S5) – N addition increased the abundance of this phylum only in the wheat season (r = 0.738, P < 0.05; Fig. 3B); and also increased class Thermoleophilia in wheat (r = 0.949, P < 0.01) in soybean season (r = 0.896, P < 0.01; Fig. S2C). Four genera, Salinibacterium, Actinoplanes, Nocardioides and Streptacidiphilus, of Actinobacteria showed distinct correlations with N addition in both seasons (Fig. 4B).

Verrucomicrobia was the fourth most abundant phylum with 5.4–9.4% of sequences in these soils. Compared to that in unfertilized samples, the relative abundance of Verrucomicrobia in N1 samples increased not significantly, whereas those in N2 samples decreased significantly in both seasons (Tables S4 and S5). There was a significant (P < 0.01) decline in abundance of Chloroflexi and Nitrospirae (Tables S4 and S5) from unfertilized to high N treatments in both seasons.

OTU-level bacterial β-diversity analysis and redundancy analysis

The PCoA based on the weighted UniFrac distance matrices (Fig. 5) clustered all three replicate soil samples within the same treatment plot together and they displayed stronger dissimilarities to those from other treatments. Across wheat and soybean seasons, the phylogenetic structure of the bacterial communities shifted in a similar manner. As N inputs increased, communities became progressively more distinct from those receiving no N fertilizer in both seasons. Furthermore, the overall weighted UniFrac distances between communities were highly correlated with N additions in both wheat (r = 0.932, P < 0.01) and soybean (r = 0.932, P < 0.01) seasons, as well as with edaphic characteristics (pH, NO3 and Avail P; Table 1).

Figure 5
Figure 5

Principal coordinate analysis of pyrosequencing reads obtained from soils subjected to different fertilization regimes based on the weighted Fast UniFrac metric. The first three axes are shown and the percent of variance explained by each axis is given.

Table 1: Mantel test of UniFrac distances with soil properties and N addition in two seasons.

According to the forward-selection option in CANOCO, soil pH (contribution of 35.8%, P = 0.002) NO3 concentration (contribution of 30.4%, P = 0.002), Avail P (contribution of 13.8%, P = 0.002) and TN (contribution of 9.6%, P = 0.006) were the four most important contributors to the variation in bacterial communities (Table S3). These soil properties together explained 72.8% of the variation in microbial communities among samples. Based on this model, a total of 58.61% of the total variance was explained by the first two constrained axes of the RDA: the first axis explained 34.32% and the second 24.29% (Fig. S3).


Variability in bacterial 16S rDNA copy numbers

In this study, N inputs significantly (P < 0.01) decreased 16S rDNA copy numbers in wheat (r = −0.949) and soybean seasons (r = −0.970), which partially addressed question concerning N addition consistently decreasing bacterial abundance. The result was in accordance with our previous research11, which concluded that bacterial 16S rDNA copy numbers declined remarkably across different doses of N and P fertilizer inputs. It was interesting that soil pH also showed the same trend with 16S rDNA copy numbers across N gradients in this experiment. Particularly, in N2 plots, after long-term urea-N fertilizer application, the soil was acidic and pH declined to 4.64 and 5.4 in the wheat and soybean seasons, respectively; the corresponding abundances of bacteria were 9.79 × 107 and 5.92 × 107 copies/g soil and declined by 49.2% and 47.8% compared to CK. The results indicated that bacterial strains had narrow pH optima and N fertilization induced a significant reduction in bacterial numbers22. A high correlation between the gradient of soil pH and the shift in 16S rDNA copy numbers was previously reported23, 24 and indicated that soil pH could have a major effect on population sizes of bacteria. As farmers use more inorganic N fertilizers, bacteria abundance will decrease with time – the fungi will grow faster than bacteria25 and this will lead to lower bacteria to fungi ratios in soil, which is an important parameter in assessing the balance of soil microbial communities26.

Variability in bacterial community diversity

Recent studies have shown that overuse of inorganic N can reduce bacterial biodiversity27. In the present study, in both wheat and soybean seasons, there was a consistent decline in bacterial α-diversity (Shannon index) (Fig. 2A) across N gradients, and these shifts were driven largely by N addition (Table S2) – this was in good agreement with previous reports27, 28. The decline in biodiversity caused by N accumulation was reported to be due to stimulating the expansion of nitrophilous species and competitive exclusion of others29. Although soil pH11, soil type30, moisture31 and crop rotation32 have all been shown to be important factors affecting bacterial diversity, the direct or indirect interactions among plant, soils and bacterial diversity in response to N deposition remain unclear. However, in this study, the high correlation between N addition and bacterial diversity in two different crop seasons indicates that bacterial diversity may be affected more by N enrichment than crops. The loss of microbial diversity can alter terrestrial ecosystem processes, and suggests that the importance of functional redundancy in soil microbial communities has been overstated33. In the present study, a lower bacterial α - diversity in high N inputs soils may lead to a less stable ecosystem34, and in this sense very high N addition is not suitable for sustainable microbial functions and processes.

ACE and Chao1, two estimators of richness, revealed higher bacterial richness in control and N1 plots than that in N2 plots. However, their shifts were not related to N addition, and the result was similar to the report of Zhang et al.35.

Variability in the composition of the dominant taxa

In both seasons, we observed consistent shifts in bacterial community composition (Fig. 3) and these shifts were driven largely by changes in the relative abundance of specific bacterial groups (Figs 3, S1 and S2), which addressed question concerning whether N additions consistently changed the bacterial community composition. Proteobacteria became more dominant in N2 treatments (34.71% and 44.47% in wheat and soybean seasons, respectively; Fig. 3) than in CK. Among them, the relative abundance of class Alphaproteobacteria increased by 33.6% and 77.6% compared to that in CK in wheat and soybean seasons, respectively (Fig. 3B). Alphaproteobacteria (e.g. family Acetobacteraceae) can use recalcitrant forms of C and supply important C intermediates to other microorganisms in acidic fermentation reactions36. There was obviously lower pH and higher Alphaproteobacteria abundance in N2 soils. Devosia species are N-fixing bacteria and have been shown to be positively correlated with potato yield37; and bacteria of genus Sphingomonas can metabolize C1 compounds38. An increase in these two genera in N2 soils may be an advantage of using urea fertilizers because these groups are also known to contribute to plant health by promoting plant growth and protecting against plant diseases37, 39.

Gammaproteobacteria abundance increased by 68.2% and 65.9% compared to the unfertilized controls in wheat and soybean seasons, respectively. Previous work also suggested that Gammaproteobacteria were more abundant in long-term N fertilized soils compared with unfertilized controls35, 36. Some Gammaproteobacteria species are considered to be organotrophic and can flourish in soils with large amounts of nutrients, and utilize reduced inorganic compounds such as ammonia, nitrite or NO3 as energy sources24. For this reason, higher relative abundance of Gammaproteobacteria was observed in higher N addition treatments with higher concentrations of nitrite, NO3 and OM. In contrast, Deltaproteobacteria abundance decreased across N gradients, and the decline patterns were consistent with response of Deltaproteobacteria to soils amended with inorganic N4.

Phylum Actinobacteria demonstrated a positive trend with increasing N addition in both seasons (Fig. S1), which was not in accordance with previous results showing that Actinobacteria decreased at low pH40. However, this shift was consistent with the prediction of Ramirez et al.41 that Actinobacteria were copiotrophic groups with fast growth rates, relying on more labile C sources and more likely to increase in abundance with increased nutrient inputs28, such as in high N amendments (N2) in the present study. In contrast, the relative abundances of Acidobacteria and Verrucomicrobia decreased across N gradients in both seasons, possibly because they are oligotrophic groups42 with slower growth rates, and in all likelihood, with the ability to metabolize nutrient-poor and recalcitrant C substrates. This shift is consistent with the microbial N mining hypothesis that soil microbes reduce decomposition of recalcitrant C and lead to a shift towards labile C decomposition under N-enriched conditions43. Thus, they have lower competitiveness than copiotrophic microbes in nutrient-rich soil. The presumed taxon-specific responses might be an indirect N effect arising from the increase in organic C availability associated with the increase in wheat and soybean productivity in soils amended with N. The bacterial taxon responses observed across the N gradients are consistent with characteristics of metagenomes44, suggesting that the communities became more copiotrophic as N inputs increased.

Factors driving the shifts in the bacterial community

Mantel tests showed that N addition was strongly related to the overall UniFrac distances between communities. This result highly corresponded to previous studies demonstrating that shifts in community composition were highly correlated with N additions in two long-term experimental N-gradient grasslands21. N fertilization decreased soil pH, and soil pH was highly correlated with UniFrac distance between bacterial communities. The RDA revealed that a large percentage (35.8%) of the variation within the pyrosequencing data was attributable to soil pH. The results validated question (ii) and previous work11, 45 also concluded that soil pH had a strong influence on soil bacterial community structure and diversity in both seasons. Meanwhile, concentration of NO3 in soil (contribution of 30.4%, r = 0.506) was also an important factor driving the shift of bacterial community according to the Mantel test (Table 1) and RDA (Table S3) results, which was consistent with our previous studies11 and addressed question (ii). Furthermore, soil Avail P (contribution of 13.8%, r = 0.688, P = 0.002) was also significantly correlated with the variation in bacterial communities. Similar results were previously reported, in which both the bacterial46 and AM fungal communities47 were significantly correlated with soil Avail P48. Our results showed that long-term N addition in the absence of P had a significant effect on the soil Avail P content, making it a limiting nutrient for some sensitive bacterial groups. For example, the relative abundances of order Legionellales (in class Gammaproteobacteria) and Sphingomonadales (in class Sphingobacteria) were highly positively correlated with the Avail P concentration in our previous study11.

Although soybean and wheat can produce different root exudates, and root exudates shape soil bacterial community structure49, we observed consistent shifts in bacterial abundance, diversity and community structure in both wheat and soybean seasons. These shifts were driven largely by changes in the dose of N addition. However, our observed trends were in direct contrast with predictions, suggesting that crop root exudates may be important in structuring the communities, but were not the dominant factor responsible for the pronounced shifts in bacterial community composition across these N gradients. The bacterial compositional changes caused by N fertilizer were more dramatic than those caused by crop root exudates in our experiment.


By simultaneously investigating bacterial community responses to urea -N across two contrasting crop seasons, we determined that N additions enhanced the concentration of soil N nutrients, but decreased bacterial 16S rDNA copy numbers. Our work suggests that N inputs elicited similar responses in the entire soil bacterial community structure and the dominant groups, although the temperature, precipitation and cultivated crops significantly differed in the two continuous crop seasons. Moreover, we observed a significant and distinct shift in the bacterial community resulting from N fertilizer regimes and their subsequent effect on edaphic characteristics, while crop season also led to a significant but much smaller variation in bacterial community structure. This work highlights that shifts of bacterial community may be important in understanding the corresponding consequences of urea-N addition and provides an important step towards identifying the likely community shifts associated with urea-N addition.


Field description and soil sampling

The experimental field is located in Harbin city, Heilongjiang Province, China (45°40′ N, 126°35′ E and altitude 151 m). This region has a temperate continental monsoon climate, with a frost-free period of 135 d. Wheat, soybean and maize were continuously grown on the field and the fertilization experiment was commenced in 1980, including three replicates (each plot was 9 m × 4 m) in a randomized complete block design. In the present study, we analyzed soil samples from three treatments. In early July annually since 1980, each plot in each row was randomly assigned to one of three fertilizer treatments: CK (without fertilizer), N1 (150 kg urea ha−1 y−1) and N2 (300 kg urea ha−1 y−1). The period of crop growing started in early April and ended in late September every year and we collected soil samples in September 2013 and 2014 after wheat and soybean harvests, respectively. The annual average soil temperature, 10 cm below the surface of soil was 7.4 and 5.9 °C, and annual precipitation was 2262 and 296 mm in 2013 and 2014, respectively ( Ten cores (7.5 cm in diameter) were randomly collected from the plow layer of soil (5–20 cm) in selected subplots within each treatment plot. The cores from each replicate plot were mixed together, and transported to the laboratory on ice. Samples were sieved (2 mm pore size) to remove roots, debris and stones, and divided into two parts: one was stored at −80 °C for molecular analysis, and the other was air-dried and used to determine soil properties.

Soil physicochemical properties and crop yield

Soil pH was measured using a glass electrode meter in a soil water ratio (mass: volume) of 1:1, after being shaken for 1 h50. Organic matter (OM) was determined using the potassium dichromate external heating method51. Total N (TN) was measured according to Strickland and Sollins52. Soil KCl-extractable NO3 and ammonium (NH4+) were quantified by extraction with 2 M KCl, steam distillation and titration53. Available P was extracted with sodium bicarbonate and then determined using the molybdenum blue method54. Available K was extracted with ammonium acetate and determined by flame photometry22. Wheat and soybean yield were determined by the soil testing laboratory in Harbin at the black soil experimental station of the Heilongjiang Academy of Agricultural Sciences.

Real-time PCR

Soil (0.25 g) was used for DNA extraction using the PowerSoil DNA extraction kit (MOBIO Laboratories Inc., Carlsbad, CA, USA) following the manufacturer’s method, modified with an additional incubation step at 65 °C for 10 min followed by 2 min of bead beating55. To minimize the DNA extraction bias, three successive extractions of microbial DNA from the same soil samples were combined and purified using a DNeasy Tissue kit (Qiagen, Valencia, CA, USA)2. Quantitative real-time PCR was performed to determine the relative 16S rRNA gene abundance. We used the 515f/806r primer sets56 to quantify the total bacterial populations. The standard templates were made from 10-fold dilutions of linearized plasmids containing the gene fragment of interest that was cloned from amplified pure 16S rRNA gene. The R2 of the standard curve was >0.99.

Sequencing of 16S rRNA gene

DNA was amplified using the 515f/806r primer set with the reverse primer containing a 6-bp error-correcting barcode unique to each sample. Three replicate PCR reactions for each sample were purified, and pyrosequencing was conducted on an Illumina Miseq 2 × 250 platform according to the protocols of Caporaso et al.56. Raw sequences were deposited in the NCBI Sequence Read Archive with accession number SRX1126010.


The quality filtering of raw reads and operational taxonomic unit (OTU) clustering was done with mothur v1.3257. The quality filter removed sequences outside the range of 200–300 bases long and those with primer mismatches. OTUs were picked using a de novo OTU picking protocol with a 97% similarity threshold. Rare OTUs (<4 reads each; <2% of total reads) were excluded from this analysis in order to reduce the datasets and thus provide more consistent alignments. High-quality sequences were classified using the Greengene database.

Alpha diversity analysis included Shannon index, Chao1 and number of OTUs within each sample was calculated in the QIIME pipeline, based on rarefaction of the OTU table. To compare between-sample differences in the total bacterial community composition, unweighted and weighted UniFrac distances58 were calculated with the QIIME pipeline55. Principal Coordinate Analysis (PCoA) was performed on the basis of the weighted UniFrac distance measured, and coordinates were used to draw 3D graphical outputs57.

Statistical analysis

The concentrations of OM, TN, NO3, NH4+ and available P and pH in soil samples were tested using a one-way analysis of variance (ANOVA). Correlations between soil properties, the relative abundances of each bacterial group, and estimated diversity levels with N inputs were tested for significance using Pearson’s correlations in the SPSS (v19.1) statistical package. To analyze changes in phylogenetic structure, pairwise weighted UniFrac distance matrices were compared against N addition and each of the measured soil properties using Mantel tests (999 permutations) in the R statistical language (x64 3.1.3) using the ‘vegan’ package following methods of Ramirez et al.36. Redundancy analysis (RDA) was applied to visualize the effect of edaphic factors on bacterial community structure, and was carried out with CANOCO software (version 5.0, Microcomputer Power Inc., Ithaca, NY, USA) followed by Monte Carlo permutation tests (999 permutations).

Additional Information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


  1. 1.

    Galloway, J. N. et al. Transformation of the nitrogen cycle: recent trends, questions, and potential solutions. Science 320, 889–892 (2008).

  2. 2.

    Zhou, J. et al. Thirty four years of nitrogen fertilization decreases fungal diversity and alters fungal community composition in black soil in northeast China. Soil Biology & Biochemistry 95, 135–143 (2016).

  3. 3.

    Lamarque, J. F. et al. Assessing future nitrogen deposition and carbon cycle feedback using a multimodel approach. Analysis of nitrogen deposition. 110, 2657–2677 (2005).

  4. 4.

    Simkin, S. M. et al. Conditional vulnerability of plant diversity to atmospheric nitrogen deposition across the United States. P Natl Acad Sci USA 113, 4086–4091 (2016).

  5. 5.

    Wang, N. et al. Variability in responses of bacterial communities and nitrogen oxide emission to urea fertilization among various flooded paddy soils. Fems Microbiology Ecology 91 (2015).

  6. 6.

    LeBauer, D. S. & Treseder, K. K. Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology 89, 371–379 (2008).

  7. 7.

    Cleland, E. E. & Harpole, W. S. Nitrogen enrichment and plant communities. Ann Ny Acad Sci 1195, 46–61 (2010).

  8. 8.

    Waldrop, M. P. & Firestone, M. K. Microbial community utilization of recalcitrant and simple carbon compounds: impact of oak-woodland plant communities. Oecologia 138, 275–284 (2004).

  9. 9.

    Zeglin, L. H., Stursova, M., Sinsabaugh, R. L. & Collins, S. L. Microbial responses to nitrogen addition in three contrasting grassland ecosystems. Oecologia 154, 349–359 (2007).

  10. 10.

    Fierer, N. & Lennon, J. T. The generation and maintenance of diversity in microbial communities. American journal of botany 98, 439–448 (2011).

  11. 11.

    Zhou, J. et al. Influence of 34-years of fertilization on bacterial communities in an intensively cultivated black soil in northeast China. Soil Biology & Biochemistry 90, 42–51 (2015).

  12. 12.

    Leff, J. W. et al. Consistent responses of soil microbial communities to elevated nutrient inputs in grasslands across the globe. Proceedings of the National Academy of Sciences 112, 10967–10972 (2015).

  13. 13.

    Paungfoolonhienne, C. et al. Nitrogen fertilizer dose alters fungal communities in sugarcane soil and rhizosphere. Scientific Reports 5, 8678 (2015).

  14. 14.

    Tahir, M., Mirza, M. S., Hameed, S., Dimitrov, M. R. & Smidt, H. Cultivation-Based and Molecular Assessment of Bacterial Diversity in the Rhizosheath of Wheat under Different Crop Rotations. PloS one 10, e0130030 (2015).

  15. 15.

    Hawkes, C. V. et al. Fungal community responses to precipitation. Global Change Biology 17, 1637–1645 (2011).

  16. 16.

    Ultra, V. U., Han, S.-H. & Kim, D.-H. Soil properties and microbial functional structure in the rhizosphere of Pinus densiflora (S. and Z.) exposed to elevated atmospheric temperature and carbon dioxide. Journal of Forest Research 18, 149–158 (2012).

  17. 17.

    Zhao, J. et al. Responses of bacterial communities in arable soils in a rice-wheat cropping system to different fertilizer regimes and sampling times. Plos One 9, e85301 (2014).

  18. 18.

    Yin, C. et al. Denitrification potential under different fertilization regimes is closely coupled with changes in the denitrifying community in a black soil. Applied Microbiology and Biotechnology 99, 5719–5729 (2015).

  19. 19.

    Tang, H. et al. Diversity analysis of nitrite reductase genes (nirS) in black soil under different long-term fertilization conditions. Ann Microbiol 60, 97–104 (2010).

  20. 20.

    Yu, R., Kampschreur, M. J., Loosdrecht, M. C. M. V. & Chandran, K. Mechanisms and specific directionality of autotrophic nitrous oxide and nitric oxide generation during transient anoxia. Environmental Science & Technology 44, 1313–1319 (2010).

  21. 21.

    Ramirez, K. S. & Fierer, N. Consistent effects of nitrogen fertilization on soil bacterial communities in contrasting systems. Ecology 91, 3463 (2010).

  22. 22.

    Shen, Z. et al. Induced soil microbial suppression of banana fusarium wilt disease using compost and biofertilizers to improve yield and quality. European Journal of Soil Biology 57, 1–8 (2013).

  23. 23.

    Wessén, E., Hallin, S. & Philippot, L. Differential responses of bacterial and archaeal groups at high taxonomical ranks to soil management. Soil Biology & Biochemistry 42, 1759–1765 (2010).

  24. 24.

    Ahn, J. H. et al. Characterization of the bacterial and archaeal communities in rice field soils subjected to long-term fertilization practices. Journal of Microbiology 50, 754 (2012).

  25. 25.

    Thiet, R. K., Frey, S. D. & Six, J. Do growth yield efficiencies differ between soil microbial communities differing in fungal:bacterial ratios? Reality check and methodological issues. Soil Biology & Biochemistry 38, 837–844 (2006).

  26. 26.

    de Vries, F. T., Hoffland, E., van Eekeren, N., Brussaard, L. & Bloem, J. Fungal/bacterial ratios in grasslands with contrasting nitrogen management. Soil Biology and Biochemistry 38, 2092–2103 (2006).

  27. 27.

    Wang, J. et al. Impact of inorganic nitrogen additions on microbes in biological soil crusts. Soil Biology and Biochemistry 88, 303–313 (2015).

  28. 28.

    Zeng, J. et al. Nitrogen fertilization directly affects soil bacterial diversity and indirectly affects bacterial community composition. Soil Biology & Biochemistry 92, 41–49 (2015).

  29. 29.

    Bobbink, R. et al. Global assessment of nitrogen deposition effects on terrestrial plant diversity: a synthesis. Ecological Applications A Publication of the Ecological Society of America 20, 30–59 (2010).

  30. 30.

    Griffiths, R. I. et al. The bacterial biogeography of British soils. Environmental Microbiology 13, 1642 (2011).

  31. 31.

    Cruzmartínez, K. et al. Despite strong seasonal responses, soil microbial consortia are more resilient to long-term changes in rainfall than overlying grassland. Isme Journal 3, 738–744 (2009).

  32. 32.

    Lupwayi, N. Z., Rice, W. A. & Clayton, G. W. Soil microbial diversity and community structure under wheat as influenced by tillage and crop rotation. Soil Biology & Biochemistry 30, 1733–1741 (1998).

  33. 33.

    Philippot, L. et al. Loss in microbial diversity affects nitrogen cycling in soil. The ISME journal 7, 1609–1619 (2013).

  34. 34.

    Chaer, G., Fernandes, M., Myrold, D. & Bottomley, P. Comparative resistance and resilience of soil microbial communities and enzyme activities in adjacent native forest and agricultural soils. Microb Ecol 58, 414–424 (2009).

  35. 35.

    Zhang, X., Wei, H., Chen, Q. & Han, X. The counteractive effects of nitrogen addition and watering on soil bacterial communities in a steppe ecosystem. Soil Biology & Biochemistry 72, 26–34 (2014).

  36. 36.

    Campbell, B. J., Polson, S. W., Hanson, T. E., Mack, M. C. & Schuur, E. A. The effect of nutrient deposition on bacterial communities in Arctic tundra soil. Environmental microbiology 12, 1842 (2010).

  37. 37.

    Barnett, B. A., Holm, D. G., Koym, J. W., Wilson, R. G. & Manter, D. K. Site and Clone Effects on the Potato Root-Associated Core Microbiome and its Relationship to Tuber Yield and Nutrients. American Journal of Potato Research 92, 1–9 (2015).

  38. 38.

    Sonoki, T. et al. Tetrahydrofolate-dependent vanillate and syringate O -demethylation links tightly to one-carbon metabolic pathway associated with amino acid synthesis and DNA methylation in the lignin metabolism of Sphingomonas paucimobilis SYK-6. Journal of Wood Science 48, 434–439 (2002).

  39. 39.

    Toyota, K. & Watanabe, T. Recent Trends in Microbial Inoculants in Agriculture. Microbes & Environments 28, 403–404 (2013).

  40. 40.

    Lauber, C. L., Hamady, M., Knight, R. & Fierer, N. Pyrosequencing-Based Assessment of Soil pH as a Predictor of Soil Bacterial Community Structure at the Continental Scale. Appl Environ Microb 75, 5111–5120 (2009).

  41. 41.

    Ramirez, K. S., Craine, J. M. & Noah, F. Consistent effects of nitrogen amendments on soil microbial communities and processes across biomes. Global Change Biol 18, 1918–1927 (2012).

  42. 42.

    Bergmann, G. T. et al. The under-recognized dominance of Verrucomicrobia in soil bacterial communities. Soil Biology & Biochemistry 43, 1450–1455 (2011).

  43. 43.

    Craine, J. M., Morrow, C. & Fierer, N. Microbial nitrogen limitation increases decomposition. Ecology 88, 2105–2113 (2007).

  44. 44.

    Fierer, N. et al. Comparative metagenomic, phylogenetic and physiological analyses of soil microbial communities across nitrogen gradients. The ISME journal 6, 1007–1017 (2012).

  45. 45.

    Stres, B., Sul, W. J., Murovec, B. & Tiedje, J. M. Correction: Recently Deglaciated High-Altitude Soils of the Himalaya: Diverse Environments, Heterogenous Bacterial Communities and Long-Range Dust Inputs from the Upper Troposphere. Plos One 8, 1521–1525 (2013).

  46. 46.

    Beauregard, M. S., Hamel, C., Atul-Nayyar & St-Arnaud, M. Long-Term Phosphorus Fertilization Impacts Soil Fungal and Bacterial Diversity but not AM Fungal Community in Alfalfa. Microb Ecol 59, 379–389 (2010).

  47. 47.

    Liu, Y. et al. Direct and indirect influences of 8 yr of nitrogen and phosphorus fertilization on Glomeromycota in an alpine meadow ecosystem. New Phytologist 194, 523 (2012).

  48. 48.

    Michael, D. et al. Tree girdling provides insight on the role of labile carbon in nitrogen partitioning between soil microorganisms and adult European beech. Soil Biology & Biochemistry 41, 1622–1631 (2009).

  49. 49.

    Haichar, F. Z. et al. Plant host habitat and root exudates shape soil bacterial community structure. The ISME journal 2, 1221–1230, doi:10.1038/ismej.2008.80 (2008).

  50. 50.

    Li, X. et al. Shifts of functional gene representation in wheat rhizosphere microbial communities under elevated ozone. Isme Journal Multidisciplinary Journal of Microbial Ecology 7, 660 (2012).

  51. 51.

    Ciavatta, C., Govi, M., Antisari, L. V. & Sequi, P. Determination of organic carbon in aqueous extracts of soils and fertilizers. Communications in Soil Science and Plant Analysis 22, 795–807 (1991).

  52. 52.

    Strickland, T. C. & Sollins, P. Improved Method for Separating Light- and Heavy-Fraction Organic Material from Soil. Soil Science Society of America Journal 51, 1390–1393 (1987).

  53. 53.

    Sparks, D. L. et al. Methods of Soil Analysis. Part III. Chemical Methods. (1996).

  54. 54.

    Olsen, S. R. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. Miscellaneous Paper Institute for Agricultural Research Samaru Pp (1954).

  55. 55.

    Fierer, N. et al. Cross-biome metagenomic analyses of soil microbial communities and their functional attributes. Proceedings of the National Academy of Sciences 109, 21390–21395 (2012).

  56. 56.

    Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335 (2010).

  57. 57.

    Schloss, P. D. et al. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 75, 7537–7541 (2009).

  58. 58.

    Lozupone, C., Lladser, M. E., Knights, D., Stombaugh, J. & Knight, R. UniFrac: an effective distance metric for microbial community comparison. Isme Journal 5, 169 (2011).

Download references


This work was funded by China National Foundation of Science (No. 41573066), National Basic Research Program of China (973 Program: 2015CB150506), National Plan for Science & Technology Support (2013BAD07B01), the China Agriculture Research System(CARS-004) and Natural Science Foundation of Shandong Province (ZR2015CL015).

Author information


  1. Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing, 100081, PR China

    • Jing Zhou
    • , Xin Jiang
    • , Mingchao Ma
    • , Fengming Cao
    • , Dawei Guan
    •  & Jun Li
  2. College of Biological Sciences, China Agricultural University, Beijing, 100094, PR China

    • Jing Zhou
    •  & Sanfeng Chen
  3. The Institute of Soil Fertility and Environmental Sources, Heilongjiang Academy of Agricultural Sciences, Harbin, 150086, PR China

    • Dan Wei
  4. Laboratory of Quality & Safety Risk Assessment for Microbial Products (Beijing), Ministry of Agriculture, Beijing, 100081, PR China

    • Xin Jiang
    • , Baisuo Zhao
    • , Mingchao Ma
    • , Fengming Cao
    • , Delong Shen
    •  & Jun Li


  1. Search for Jing Zhou in:

  2. Search for Xin Jiang in:

  3. Search for Dan Wei in:

  4. Search for Baisuo Zhao in:

  5. Search for Mingchao Ma in:

  6. Search for Sanfeng Chen in:

  7. Search for Fengming Cao in:

  8. Search for Delong Shen in:

  9. Search for Dawei Guan in:

  10. Search for Jun Li in:


Jun Li, Baisuo Zhao and Xin Jiang conceived the experiments. Jing Zhou and Dan Wei conceived and conducted the experiments. Jing Zhou, Mingchao Ma, Sanfeng Chen, Fengming Cao, Delong Shen, and Dawei Guan analyzed the results. And Jing Zhou wrote the manuscript. All authors reviewed the manuscript.

Competing Interests

The authors declare that they have no competing interests.

Corresponding authors

Correspondence to Xin Jiang or Jun Li.

Supplementary information


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BY

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit