Long-Term Nitrogen Fertilization Elevates the Activity and Abundance of Nitrifying and Denitrifying Microbial Communities in an Upland Soil: Implications for Nitrogen Loss From Intensive Agricultural Systems

Abstract

The continuous use of nitrogen (N) fertilizers to increase soil fertility and crop productivity often results in unexpected environmental effects and N losses through biological processes, such as nitrification and denitrification. In this study, multidisciplinary approaches were employed to assess the effects of N fertilization in a long-term (~20 years) field experiment in which a fertilizer gradient (0, 200, 400, and 600 kg N ha^-1 yr^-1) was applied in a winter wheat-summer maize rotation cropping system in the North China Plain, one of the most intensive agricultural regions in China. The potential nitrification/denitrification rates, bacterial community structure, and abundances of functional microbial communities involved in key processes of the N cycle were assessed during both the summer maize (SM) and winter wheat (WW) seasons. Long-term N fertilization resulted in a decrease in soil pH and an increase in soil organic matter (OM), total N and total carbon concentrations. Potential nitrification/denitrification and the abundances of corresponding functional N cycling genes were positively correlated with the fertilization intensity. High-throughput sequencing of the 16S rRNA gene revealed that the increased fertilization intensity caused a significant decrease of bacterial diversity in SM season, while changed the microbial community composition such as increasing the Bacteroidetes abundance and decreasing Acidobacteria abundance in both SM and WW seasons. The alteration of soil properties markedly correlated with the variation in microbial structure, as soil pH and OM were the most predominant factors affecting the microbial structure in the SM and WW seasons, respectively. Furthermore, consistently with the results of functional gene quantification, functional prediction of microbial communities based on 16S rRNA sequence data also revealed that the abundances of the key nitrificaiton/denitrification groups were elevated by long-term N inputs. Taken together, our results suggested that soil microbial community shifted consistently in both SM and WW seasons toward a higher proportion of N-cycle microbes and exhibited higher N turnover activities in response to long-term elevated N fertilizer. These findings provided new insights into the molecular mechanisms responsible for N loss in intensively N fertilized agricultural ecosystems.

Keywords: denitrification; high-throughput sequencing; microbial community structure; nitrification; nitrogen fertilization.

Figure 1

(A,B) The abundance of ammonia-oxidizing archaea (AOA) amoA gene, ammonia-oxidizing bacteria (AOB) amoA gene, nirK, nirS, and nosZ genes normalized to sample volume (copy numbers/g dry soil) in the SM (A) and WW (B) seasons. Error bars indicate the standard deviation of three replicates. Differing letters indicate significant differences of means of each gene in pairwise comparisons (Duncan test; P < 0.05) for each treatment. Means followed by the same footnote symbol(s) for each gene are not significantly different at P < 0.05.

Figure 2

(A) Shannon and Invsimpson of the bacterial communities in four fertilization gradients in the SM and WW seasons. (B) Principal coordinate analysis (PCoA) based on the Bray-Curtis distance showing the overall distribution pattern of bacterial communities in both SM and WW seasons.

Figure 3

Taxonomic compositions of the dominant bacterial phyla (A) (relative abundance > 1%), and classes (B) in the SM and WW seasons. Each bar represents the average value of three replicates.

Figure 4

Taxonomic distributions of the dominant bacterial genus which sigfinificantly changed across fertilization intensities (relative abundance > 1%) in the SM (A) and WW (B) seasons. The downward arrow indicates a significant decrease, the upward arrow indicates a significant increase. Un, Unclassified genus.

Figure 5

(A,B) Redundancy analysis (RDA) plots depict the correlation between bacterial communities and soil properties in the SM (A, P = 0.001) and WW (B, P = 0.039) seasons; (C,D) Multivariate regression tree (MRT) analysis of bacterial communities under different N treatments in the SM (C) and WW (D) seasons. Specific treatment and the number of samples included in the analysis are shown under bar plots. OM, soil organic matter; NO₃/${\text{NO}}_3^{-}$-N, nitrate concentration; NH4, ammonium; TC, total carbon; TN, total N; C/N: C/N ratio. Different shapes of symbols represent different treatments.