linkage between n2o emission and functional gene abundance in an intensively managed calcareous fluvo aquic soil

www.nature.com/scientificreports OPEN received: 13 December 2016 accepted: 19 January 2017 Published: 24 February 2017 Linkage between N2O emission and functional gene abundance in an intensively managed calcareous fluvo-aquic soil Liuqing Yang1, Xiaojun Zhang2 & Xiaotang Ju1 The linkage between N2O emissions and the abundance of nitrifier and denitrifier genes is unclear in the intensively managed calcareous fluvo-aquic soils of the North China Plain We investigated the abundance of bacterial amoA for nitrification and narG, nirS, nirK, and nosZ for denitrification by in situ soil sampling to determine how the abundance of these genes changes instantly during N fertilization events and is related to high N2O emission peaks We also investigated how long-term incorporated straw and/or manure affect(s) the abundance of these genes based on a seven-year field experiment The overall results demonstrate that the long-term application of urea-based fertilizer and/or manure significantly enhanced the number of bacterial amoA gene copies leading to high N2O emission peaks after N fertilizer applications These peaks contributed greatly to the annual N2O emissions in the crop rotation A significant correlation between annual N2O emissions and narG, nirS, and nirK gene numbers indicates that the abundance of these genes is related to N2O emission under conditions for denitrification, thus partly contributing to the annual N2O emissions These findings will help to draw up appropriate measures for mitigation of N2O emissions in this ‘hotspot’ region Nitrous oxide (N2O) is a powerful long-lived greenhouse gas and has a 300-times stronger warming effect than that of carbon dioxide in the troposphere on a 100-yr time horizon1 It also reacts with ozone in the stratosphere and became the dominant substance in ozone depletion in the 21st century2 Nitrous oxide is emitted from both natural and anthropogenic sources, and agricultural soils amended with chemical nitrogen (N) fertilizers and/or manure are mainly anthropogenic sources, which contribute up to 66% of the current global anthropogenic N2O emissions3 and approximately 75% of anthropogenic emission in China4 This makes agricultural soils an important target in the efforts to mitigate anthropogenic N2O emissions both regionally and globally5 A better understanding of the processes, mechanisms and factors controlling N2O production and emission is a prerequisite for managing agricultural N2O emissions Much progress has been made in the last five decades6 Although abiotic reactions are important N2O sources under certain circumstances7, biological processes play more important roles through at least four microbially mediated processes, i.e., nitrifier nitrification, nitrifier denitrification, denitrifier denitrification and coupled nitrification-denitrification8–12 which are enhanced by the application of N fertilizers and/or manure to most agricultural soils The occurrence of each N2O production process and its contribution to the total N2O emissions depend on the prevailing soil conditions such as pH, temperature, moisture, oxygen, and microbial available C and N13,14 The above factors could be classified into three categories, i.e., edaphic conditions, climatic conditions and agricultural management practices15 Due to the temporal and spatial variation in these factors, specifically measuring N2O fluxes and studying the processes among typical soil-climatic regions are important for the mitigation of agricultural N2O emissions locally and can greatly contribute to the global total N2O budget3 The North China Plain (NCP), an important agricultural region of China with an area of approximate 35 million hectares, is located in northeast China (32–41°N, 113–120°E) on the alluvial plain of the Yellow River and has a warm-temperate sub-humid climate with cold winters and hot summers16 The current agricultural College of Resources and Environmental Sciences, China Agricultural University, Yuanmingyuan West Road, Beijing 100193, China 2State Key Laboratory of Microbial Metabolism and School of Life Science and Biotechnology, Shanghai Jiaotong University, 800 Dongchuan Road, Shanghai 200240, China Correspondence and requests for materials should be addressed to X.Z (email: xjzhang68@sjtu.edu.cn) or X.J (email: juxt@cau.edu.cn) Scientific Reports | 7:43283 | DOI: 10.1038/srep43283 www.nature.com/scientificreports/ Treatment code Soil Organic C (g C kg−1) Total N (g N kg−1) C/N Ratio NH4-N (mg N kg−1) NO3-N (mg N kg−1) N0 7.78 ±​  0.92 ba 0.99 ±​  0.08 b 7.84 ±​  0.45 a 1.6 ±​  0.9 a Nopt 9.01 ±​  0.64 b 1.09 ±​  0.03 b 8.27 ±​  0.89 a 2.0 ±​  0.1 a CNopt 9.22 ±​  0.95 b 1.14 ±​  0.05 b 8.09 ±​  0.86 a CM 13.00 ±​  1.62 a 1.51 ±​  0.06 a 8.57 ±​  1.05 a Potential nitrification rate (mg NO3-N kg−1 d−1) Potential denitrification rate (mg N2O-N kg−1 d−1) 10.6 ±​  7.1 b 60.2 ±​  17.48 b 1.62 ±​  0.10 c 38.5 ±​  4.3 a 204.5 ±​  22.65 a 1.76 ±​  0.93 c 1.9 ±​  0.2 a 36.5 ±​  18.4 a 235.0 ±​  45.04 a 2.92 ±​  0.73 b 1.6 ±​  0.3 a 26.2 ±​  5.0 a 228.9 ±​  26.56 a 5.06 ±​  0.61 a Table 1.  Some soil chemical and biological properties (mean ± standard error, n = 3) at 0–20 cm soil depth sampling before the sowing of winter wheat in 2012 aWithin each parameter, different letters indicate significant differences (P ​  Nopt, but there were no significant differences among these treatments, possibly also due to the high spatial variation of measured N2O fluxes in the field plots (Fig. 1c) In our study the annual N2O emission factor was 0.20–0.40%, much lower than the IPCC default value of 1%28 but was in line with our previous studies18,20,23–25 Abundance of the 16S rRNA gene and nitrification and denitrification genes.  The order of the 16S rRNA gene copy number was CM >​  CNopt >​  Nopt >​  N0 on all three sampling dates (Fig. 3), a similar order to SOC, but only significantly higher in the CM treatment compared with the other three treatments The 16S rRNA gene copy numbers in the CM treatment were 1.6–1.9-, 1.7–2.1- and 1.3–2.2-times larger than in the CNopt, Nopt, and N0 treatments on the three sampling dates, indicating that soil with long-term incorporated manure harbored the largest 16S rRNA gene number Surprisingly, the 16S rRNA gene copies in all four treatments were higher on the 16th April soil sampling date than on 9th August or 14th August, possibly due to the different crops because the first sampling date was winter wheat and the last two sampling dates were summer maize (further explanation in the Discussion section) The soil 16S rRNA gene copy numbers in the CNopt treatment appeared to be 27.2, 16.7 and 27.7% higher than in the Nopt treatment on the three sampling dates, indicating that straw return increased the soil 16S rRNA gene copy number but this was not statistically significant The gene copy numbers of bacterial amoA in all three fertilization treatments were significantly higher than in the N0 control but were not significantly different among fertilization treatments on all three sampling dates (Fig. 4), reflecting the legacy of historical long-term application of urea-based fertilizer and/or manure as distal drivers to regulate nitrification These numbers were 1.4e +​ 7 to 1.9e +​ 7 and 1.4e +​ 7 to 1.6e +​ 7 on 16th April and 9th August, respectively On the third day after the 10th leaf fertilization, the amoA gene number of Nopt, CNopt and CM increased 14.7, 124.5 and 107.6% compared to that before fertilization (on 9th August), respectively, indicating that the 10th leaf fertilization of maize enhanced the amoA gene copy number as proximal factor and led to production of high N2O emission peaks in this short period (Fig. 1a) This result further confirms our previous conclusion that ammonia oxidation is an engine to generate nitrous oxide in this intensively managed calcareous Fluvo-aquic soil24 Scientific Reports | 7:43283 | DOI: 10.1038/srep43283 www.nature.com/scientificreports/ Figure 4.  Gene copy numbers of the bacterial ammonia monooxygenase gene (amoA) (AOB) of different treatments at 0–20 cm soil depth on the sampling dates in 2013 Different letters indicate significant differences (P ​  Nopt >​  N0, on 16th April and 9th August, but in most cases these genes were significantly higher only in the CM treatment compared with the other three treatments (Fig. 5a,b,c,d), which seems to follow the trend of SOC and TN The third sampling date on 14th August, after fertilization, likely disturbed this trend, and none of the genes were significantly different among fertilization treatments These results indicate that the abundance of denitrification genes was regulated by both distal drivers, such as SOC and TN, and proximal drivers, such as N fertilization The gene copy numbers of narG, nirS and nirK in different treatments on 16th April and 9th August corresponded well with the daily N2O emissions during these two zero fertilization periods and with the annual N2O emissions (Fig. 1a,b) The gene copy numbers of narG, nirS and nirK in different treatments on 14th August seem also to partly correspond with high N2O emission after fertilization Although the nosZ gene number was significantly higher in the CM treatment in most cases, the N2O emission factor in this treatment was still higher than that of the other treatments, likely due to the effect of the nosZ gene being partly offset by increased effects of narG, nirS and nirK genes Sequencing analysis.  The PCoA score plots based on the Bray-Curtis distance reveal that fertilization with straw and/or manure changed the structure of the soil microbiota (Fig. 6a) and there were statistically significant differences between the N0 control and the fertilization treatments (Fig. 6b), i.e., Nopt, CNopt and CM were separate Scientific Reports | 7:43283 | DOI: 10.1038/srep43283 www.nature.com/scientificreports/ Figure 6.  Alterations in the soil microbiota under different treatments at 0–20 cm soil depth on the sampling dates in 2013 (a) Principal coordinate analysis (PCoA) score plots based on the Bray-Curtis distance (b) Clustering of soil microbiota based on Mahalanobis distances calculated with a multivariate analysis of variance (MANOVA) Each point represents the mean principal coordinate (PC) score of three replicate soil samples from one treatment at one time point ***P