Evidence for the Cooccurrence of Nitrite-Dependent Anaerobic Ammonium and Methane Oxidation Processes in a Flooded Paddy Field

Abstract

Anaerobic ammonium oxidation (anammox) and nitrite-dependent anaerobic methane oxidation (n-damo) are two of the most recent discoveries in the microbial nitrogen cycle. In the present study, we provide direct evidence for the cooccurrence of the anammox and n-damo processes in a flooded paddy field in southeastern China. Stable isotope experiments showed that the potential anammox rates ranged from 5.6 to 22.7 nmol N₂ g⁻¹ (dry weight) day⁻¹ and the potential n-damo rates varied from 0.2 to 2.1 nmol CO₂ g⁻¹ (dry weight) day⁻¹ in different layers of soil cores. Quantitative PCR showed that the abundance of anammox bacteria ranged from 1.0 × 10⁵ to 2.0 × 10⁶ copies g⁻¹ (dry weight) in different layers of soil cores and the abundance of n-damo bacteria varied from 3.8 × 10⁵ to 6.1 × 10⁶ copies g⁻¹ (dry weight). Phylogenetic analyses of the recovered 16S rRNA gene sequences showed that anammox bacteria affiliated with “Candidatus Brocadia” and “Candidatus Kuenenia” and n-damo bacteria related to “Candidatus Methylomirabilis oxyfera” were present in the soil cores. It is estimated that a total loss of 50.7 g N m⁻² per year could be linked to the anammox process, which is at intermediate levels for the nitrogen flux ranges of aerobic ammonium oxidation and denitrification reported in wetland soils. In addition, it is estimated that a total of 0.14 g CH₄ m⁻² per year could be oxidized via the n-damo process, while this rate is at the lower end of the aerobic methane oxidation rates reported in wetland soils.

INTRODUCTION

Microbially mediated anaerobic ammonium oxidation (anammox), which was predicted by Broda (1) on the basis of thermodynamic calculations, was first confirmed in the 1990s in a denitrifying pilot plant (2). Thermodynamically, it was believed that microorganisms capable of using nitrite as an electron acceptor for anaerobic methane oxidation could also exist in nature (3). Nitrite-dependent anaerobic methane oxidation (n-damo) was first confirmed in 2006 in an enrichment culture (4).

Currently, five genera of anammox bacteria (“Candidatus Brocadia,” “Candidatus Kuenenia,” “Candidatus Scalindua,” “Candidatus Anammoxoglobus,” and “Candidatus Jettenia”), which form a monophyletic order of bacteria, the Brocadiales (5), have been enriched and described. At present, it is believed that the anammox process is responsible for 50% of dinitrogen gas (N₂) production in marine ecosystems (6–8). Although a limited number of recent studies has reported the presence of anammox bacteria and the occurrence of the anammox process in freshwater wetlands (9–11), the overall importance of this process in wetland systems is still unclear owing to a lack of data.

The n-damo process is catalyzed by “Candidatus Methylomirabilis oxyfera” (12), which is affiliated with the NC10 phylum. This process constitutes a unique association between the two major global nutrient cycles of carbon and nitrogen (4) and might serve as an important and overlooked sink of the greenhouse gas methane (13). Until now, however, the distribution of n-damo bacteria and the occurrence of the n-damo process in environments are not well-known. Two recent studies have reported the presence of n-damo bacteria in the sediments of two freshwater lake ecosystems, Lake Constance in Germany (14) and Lake Biwa in Japan (15), and the activity of the n-damo process was confirmed in Lake Constance using radiotracer experiments. Wang et al. (16) and Zhou et al. (17) provided molecular evidence for the presence of n-damo bacteria in paddy fields. Furthermore, Shen et al. (18, 19) reported the presence of n-damo bacteria in the sediments of the Qiantang River and the Jiaojiang Estuary in China. The distribution of n-damo bacteria was also confirmed in the sediments of the South China Sea (20). Recently, Hu et al. (21) and Shen et al. (22) reported the presence of n-damo bacteria and the occurrence of the n-damo process in wetlands.

Among the different types of wetlands (23), paddy fields are one of the most important nitrogen sinks, represent one of the most important sources of the greenhouse gas methane, and are responsible for 10 to 25% of global methane emissions (24). Furthermore, paddy fields are characterized by cultivation patterns, including water logging, which cause anoxic soil conditions (16). The anoxic soil conditions theoretically provide suitable habitats for both anammox bacteria and n-damo bacteria. In addition, the application of nitrogen-rich fertilizers further makes the paddy fields suitable habitats for these two groups of bacteria.

The primary objectives of the present study were to investigate the distribution, diversity, and significance of anammox bacteria and n-damo bacteria in a flooded paddy field (at a depth of 0 to 100 cm). Previous studies have indicated that the anammox bacteria were mainly present in the surface paddy soils, while the n-damo bacteria were mainly present in the deep paddy soils (16). To further ascertain the vertical distribution characteristics of these two groups of bacteria, two representative surface soil layers (0 to 10 and 20 to 30 cm) and two representative deep soil layers (50 to 60 and 90 to 100 cm) were analyzed in the current study. The distribution and diversity of anammox bacteria and n-damo bacteria were studied on the basis of 16S rRNA gene clone library analyses, and the abundance of these bacteria was quantified by quantitative PCR (qPCR). The potential rates of the anammox and n-damo processes were determined using ¹⁵N and ¹³C stable isotope labeling experiments, respectively.

MATERIALS AND METHODS

Site description and sampling.

The flooded paddy field selected for this study is located in Zhejiang Province, China, in a region that represents a typical agricultural region of subtropical southeastern China. The paddy field has been planted in a rice rotation and has a long history of fertilization. The total rate of nitrogen fertilization is approximately 300 to 350 g N m⁻² per year. The paddy field was used for investigation of n-damo bacteria in a previous study (21), but a different sampling site was selected in the current study. A total of five soil cores were collected from the paddy field in September 2012 using a stainless steel ring sampler (diameter, 5 cm; length, 100 cm). The cores were sliced at 10-cm intervals, and samples from each depth were mixed in the field to form one composite sample. The samples were immediately placed in sterile containers, sealed, and transported to the laboratory on ice within 12 h. The collected soil samples were subsequently divided into three parts. The first part was incubated immediately after arrival at the laboratory to determine the potential anammox activity and n-damo activity, the second part was stored anaerobically at 4°C for subsequent physicochemical analyses, and the third part was stored at −80°C for later molecular analyses.

Physicochemical analyses.

The pH and temperature of the intact soil were determined in situ using an IQ150 pH meter (IQ Scientific Instruments Inc., Carlsbad, CA, USA). Soil ammonium and nitrate were extracted using 2 M KCl, as previously described (25, 26). The amount of extracted NH₄⁺ was determined through the salicylic acid method (27). The amount of NO_x⁻ was determined by reduction of NO₃⁻ to NO₂⁻ via cadmium reduction and measured through the N-(1-naphthyl)ethylenediamine dihydrochloride method (28). NO₂⁻ and NO₃⁻ were not differentiated in the current study. Because the methods used for determination of the amounts of soil ammonium and nitrate did not exclude the possibility of interference from humic substances, the current methods may overestimate the lower end of the concentrations of soil NH₄⁺ and NO_x⁻. The soil organic carbon content was determined using the K₂Cr₂O₇ oxidation method (25), and the soil total nitrogen content was determined using a FOSS Kjeltec2300 analyzer (FOSS Group, Höganäs, Sweden). The water content of the soils was determined by oven drying overnight at a temperature of 110°C. Below-ground gas samples were gathered at 10-cm intervals through the use of soil gas samplers, as previously described (21). The methane was determined using an Agilent 6890N gas chromatograph (Agilent) as previously described (21). All the above-described analyses were performed on the soil samples or gas samples in triplicate.

Isotope tracer experiments.

Soil samples were transferred to He-flushed 75-ml glass vials together with He-purged deionized water. The soil slurries were preincubated under anaerobic conditions for at least 30 h to remove the residual NO_x⁻ and oxygen in the slurries. The slurries were subsequently divided into six groups that received one of the following treatments: (i) ¹⁵NH₄⁺ (¹⁵N at 99.6%), (ii) ¹⁵NH₄⁺ plus NO₂⁻, (iii) ¹⁵NO₂⁻ (¹⁵N at 99.6%), (iv) ¹³CH₄ (¹³C at 99.9%), (v) ¹³CH₄ plus NO₂⁻, and (vi) ¹³CH₄ plus SO₄²⁻. The final concentrations of NH₄⁺ or NO_x⁻ ranged from 67.8 to 150.0 μmol kg⁻¹ (dry weight) soil in treatments (i), (ii), (iii), and (v). Three independent experiments per sample in each treatment group were performed. Immediately after the preincubation step, 2 ml of headspace gas was removed and replaced with an equal volume of ¹³CH₄, resulting in a final concentration of 4.5 × 10³ μmol liter⁻¹ in the headspace of each vial in treatments (iv), (v), and (vi). The production of ²⁹N₂, ³⁰N₂, and ¹³CO₂ was measured directly from the headspace of each vial with a gas chromatograph-mass spectrometer (Agilent 7890/5975C inert MSD; Agilent, USA) as previously described (21, 29, 30). The potential anammox rates could be calculated by the linear regression of the concentration of ²⁹N₂ produced from slurries amended with ¹⁵NO₂⁻ or ¹⁵NH₄⁺ plus NO₂⁻. However, the soil samples (especially the surface soil samples) used in this study contained relatively high concentrations of NH₄⁺. As a result, the background NH₄⁺ in the slurries could not be exhausted under anoxic conditions after preincubation. Thus, the potential rates would be underestimated on the basis of the results for slurries amended with ¹⁵NH₄⁺ plus NO₂⁻ because the background NH₄⁺ could react with NO₂⁻ for the production of ²⁸N₂. On the other hand, the background NO₂⁻/NO₃⁻ in the slurries could be exhausted by denitrification and anammox under anoxic conditions. Actually, the potential anammox rates obtained from slurries amended with ¹⁵NH₄⁺ plus NO₂⁻ were approximately 85 to 97% of the rates obtained from slurries amended with only ¹⁵NO₂⁻ in our preexperiment. Therefore, the concentration of ²⁹N₂ produced from slurries amended with ¹⁵NO₂⁻ was finally used for determination of the potential anammox rates. The potential n-damo rates were calculated by linear regression of the concentration of ¹³CO₂ produced from slurries amended with ¹³CH₄ plus NO₂⁻ in the headspace of the vial over time. The coefficients of determination (R² values) for linear regression of the change in ²⁹N₂ and ¹³CO₂ concentrations over time were greater than 0.9 for most data sets.

DNA extraction and PCR amplification.

Soil DNA was extracted using a Power soil DNA kit (Mo Bio Laboratories, Carlsbad, CA, USA) according to the manufacturer's instructions. Approximately 0.3 g of homogenized soil was used for DNA isolation. The quality of the extracted DNA was evaluated on a 1% agarose gel, and the DNA concentration was measured with a NanoDrop spectrophotometer (ND-1000; Isogen Life Science, the Netherlands).

The 16S rRNA genes of anammox bacteria were amplified using a nested PCR protocol, as previously described (31). In the first round of PCR, the forward primer Pla46f (32) and the reverse primer 1545r (33) were used. In the second round, the PCR was conducted using the anammox bacterium-specific primers Amx368f (34) and Amx820r (35). A nested PCR protocol was also used to amplify the 16S rRNA genes of n-damo bacteria, as previously described (36). In the first round of PCR, the n-damo bacterium-specific forward primer 202f (30) and the general bacterial reverse primer 1545r (33) were used. In the second round, the PCR was performed using primers qP1f and qP2r (30), which are specific for n-damo bacteria. Detailed information on the primers used in this study is shown in Table 1.

TABLE 1.

PCR primers used in this study

^aND, not determined.

Cloning and sequencing.

The PCR products were cloned using the pMD19-T vector (TaKaRa, Bio Inc., Shiga, Japan) according to the manufacturer's instructions. Randomly selected positive clones for each sample were subjected to sequencing (Life Technology, Shanghai, China).

Phylogenetic analysis.

The recovered 16S rRNA gene sequences were aligned with the MUSCLE algorithm (37) and imported into MEGA (version 5) software (38), where the alignment was manually checked and trimmed. Phylogenetic analysis of the sequences was performed by MEGA (version 5) software using the neighbor-joining method (38), and a BLAST search was performed to search for related sequences in GenBank. The evolutionary distances were computed using the maximum composite likelihood method. The robustness of the tree topology was tested with a bootstrap analysis (1,000 replicates), and bootstrap values of >70 (700 replicates) are shown at the branches.

Quantitative PCR.

Hydrazine synthase (hzs) is a very important enzyme in anammox metabolism and is responsible for the synthesis of hydrazine from nitric oxide and ammonium (39, 40). In this study, the primer set hzsA_1597f-hzsA_1857r, targeting subunit α of the hzs genes of anammox bacteria, was used to determine the abundance of anammox bacteria, as previously described (41). The abundance of n-damo bacteria was estimated by quantifying their 16S rRNA genes using the primer set qP1f-qP1r, as previously described (30). The standard curves were constructed from a series of 10-fold dilutions of a known copy number of plasmid DNA containing the target genes. Negative-control reactions in which the DNA template was replaced by nuclease-free water were also performed. Triplicate qPCR analyses were performed for each sample. Single peaks were observed in the melting curves for both qPCR assays, and the amplification efficiency was greater than 90% for both qPCR assays. In addition, the specificity of the primer sets for the anammox bacteria and n-damo bacteria was further confirmed by sequencing the qPCR products from several soil samples. Phylogenetic analysis showed that the sequences recovered with primer set hzsA_1597f-hzsA_1857r were all very closely related to the sequences of the hzsA genes of anammox bacteria (see Fig. S1 in the supplemental material). Similarly, phylogenetic analysis showed that the sequences of the qPCR products obtained with primer set qP1f-qP1r were all closely related to the sequences of the 16S rRNA gene of “Candidatus Methylomirabilis oxyfera” (see Fig. S2 in the supplemental material).

Statistical analyses.

The operational taxonomic units (OTUs) for the determination of the 16S rRNA gene diversity of anammox bacteria and n-damo bacteria were defined using 3% differences in the nucleotide sequences and the furthest-neighbor algorithm in the DOTUR program (42). The Chao1 estimator and the Shannon index were also generated using the DOTUR program.

Nucleotide sequence accession numbers.

The 16S rRNA gene sequences reported in this study have been deposited in the GenBank database under accession numbers KF754815 to KF754836 (anammox 16S rRNA), KM403486 to KM403495 (anammox hzsA), and KF754837 to KF754861 (n-damo 16S rRNA).

RESULTS

Physicochemical analyses of the collected core samples.

The vertical distribution profiles of the soil NH₄⁺ content, NO_x⁻ content, CH₄ content, pH, temperature, total nitrogen content, and organic carbon content at 10-cm intervals are shown in Fig. 1. The NH₄⁺ content peaked in the layer at a depth of 10 to 20 cm and then decreased with depth from 2,270.7 ± 122.6 to 9.0 ± 0.4 μmol kg⁻¹ (dry weight) soil. The content of NO_x⁻ peaked in the layer at a depth of 0 to 10 cm and then decreased with depth from 745.1 ± 45.5 to 7.9 ± 0.4 μmol kg⁻¹ (dry weight) soil. The simultaneous decrease of both NH₄⁺ and NO_x⁻ with depth may indicate the occurrence of anammox and denitrification. As opposed to NO_x⁻, the CH₄ concentration in soil gas showed a trend toward an increase with depth from 13.1 × 10³ ± 0.7 × 10³ to 6.3 × 10³ ± 0.7 × 10³ μmol liter⁻¹. The coexistence of NO_x⁻ and methane may suggest that the paddy soil could provide a suitable habitat for n-damo bacteria. Soil samples collected from four representative layers (0 to 10 cm, 20 to 30 cm, 50 to 60 cm, and 90 to 100 cm) were selected for further molecular analyses and activity tests.

FIG 1.

Vertical distribution of soil NH₄⁺ (a), NO_x⁻ (b), CH₄ (c), pH (d), temperature (e), water content (f), total nitrogen (TN) (g), and organic carbon (OrgC) (h) in core samples collected from the paddy field.

Phylogenetic analyses of anammox bacteria and n-damo bacteria.

Phylogenetic analysis of the recovered 16S rRNA gene sequences of anammox bacteria showed that these sequences were grouped into three distinct clusters (Fig. 2). The sequences of the “Candidatus Brocadia” cluster, which were recovered from the layer at a depth from 90 to 100 cm, showed 95.0 to 96.2% identity to the sequence of the 16S rRNA gene of “Candidatus Brocadia anammoxidans.” The sequences in this cluster were the most closely related to the sequences obtained from the Baiyangdian Lake sediments (10), with 99% identity. The sequences of the “Candidatus Kuenenia” cluster, which were recovered from the layer at a depth of 0 to 10 cm, showed 94.8 to 97.5% identity to the sequence of the 16S rRNA gene of “Candidatus Kuenenia stuttgartiensis.” The closest relatives of this cluster were the sequences retrieved from Qiantang River sediments (43), with 98% identity. Furthermore, a new anammox cluster which was distantly related, with 92.6 to 95.0% identity, to the 16S rRNA gene of “Candidatus Kuenenia” formed in the phylogenetic tree (Fig. 2). This cluster was most closely related, with 99% identity, to the clones obtained from another paddy field also located in southeastern China (9).

FIG 2.

Neighbor-joining phylogenetic tree showing the phylogenetic affiliations of the anammox bacterial 16S rRNA gene sequences in core samples collected from the paddy field. The bootstrap values included 1,000 replicates, and the scale bar represents 2% sequence divergence. The identifiers PFG10, PFG30, PFG60, and PFG100 represent core samples collected from layers at depths of 0 to 10 cm, 20 to 30 cm, 50 to 60 cm, and 90 to 100 cm, respectively. The numbers in parentheses indicate the number of clones in each cluster/total number of clones sequenced, and the other designations in parentheses are GenBank accession numbers.

Phylogenetic analysis of the recovered 16S rRNA gene sequences of n-damo bacteria showed that the recovered sequences were grouped into three separate clusters (Fig. 3), which were assigned to two groups of n-damo bacteria, group A and group B, as described by Ettwig et al. (30). The sequences of cluster I, which were primarily recovered from the layers at depths of 50 to 60 cm and 90 to 100 cm, showed 95.8 to 96.9% identity to the sequence of the 16S rRNA gene of “Candidatus Methylomirabilis oxyfera.” The closest relatives of this cluster, with 98% identity, were the clones recovered from another paddy field (16). The sequences of clusters II and III, which were primarily recovered from the layer at depths of 0 to 10 cm and 20 to 30 cm, showed only 91.6 to 92.1% and 90.1 to 90.9% identities to the sequence of the 16S rRNA gene of “Candidatus Methylomirabilis oxyfera,” respectively. These two clusters were also the most closely related, with 98% identity, to clones recovered from another paddy field (16).

FIG 3.

Neighbor-joining phylogenetic tree showing the phylogenetic affiliations of the n-damo bacterial 16S rRNA gene sequences in core samples collected from the paddy field. The bootstrap values included 1,000 replicates, and the scale bar represents 2% sequence divergence. The numbers in parentheses indicate the number of clones in each cluster/total number of clones sequenced, and the other designations in parentheses are GenBank accession numbers.

Genetic diversity analyses of anammox bacteria and n-damo bacteria.

The diversity levels of the 16S rRNA genes of anammox bacteria and n-damo bacteria in each sample were determined on the basis of the number of OTUs, the Shannon index, and the Chao1 estimator (see Table S1 in the supplemental material). A total of 7 and 11 OTUs of the 16S rRNA genes of anammox bacteria and n-damo bacteria, respectively, were observed. A similar diversity of anammox bacterial 16S rRNA genes was observed between the different layers of the soil cores (see Table S1). The diversity of the 16S rRNA genes of n-damo bacteria was also very similar between the different layers of the soil cores (see Table S1). It could be observed that the diversity of the n-damo bacteria was higher than that of the anammox bacteria in each layer (see Table S1).

Quantitative analyses of anammox bacteria and n-damo bacteria.

The qPCR results further confirmed the coexistence of anammox bacteria and n-damo bacteria in the different layers of the soil cores. The abundance of anammox bacteria ranged from 1.0 × 10⁵ ± 0.04 × 10⁵ to 2.0 × 10⁶ ± 0.14 × 10⁶ copies g⁻¹ (dry weight), assuming that the anammox bacteria contain one copy of the hzsCBA gene cluster, as previously reported (39, 40). Different abundances of anammox bacteria were observed in the different layers of the soil cores, with the highest abundance being in the layer at a depth of 0 to 10 cm (Fig. 4, top). Different abundances of n-damo bacteria were also observed in different layers of the soil cores, with the highest abundance (6.1 × 10⁶ ± 0.25 × 10⁶ copies g⁻¹ [dry weight]) being in the layer at a depth of 90 to 100 cm and the lowest abundance (3.8 × 10⁵ ± 0.12 × 10⁵ copies g⁻¹ [dry weight]) being in the layer at a depth of 20 to 30 cm (Fig. 4, top).

FIG 4.

Copy numbers of anammox bacterial hzsA genes and n-damo bacterial 16S rRNA genes (top) and potential anammox rates and n-damo rates (bottom) in core samples collected from different layers of the paddy field.

Activity analyses of the anammox process and n-damo process.

Stable isotope experiments confirmed the cooccurrence of anammox and n-damo processes in the paddy field examined (Fig. 5). The results showed that the potential anammox rates ranged from 5.6 ± 0.6 to 22.7 ± 1.0 nmol N₂ g⁻¹ (dry weight) day⁻¹, which contributed 8.7 to 29.8% N₂ to soil N₂ production. Different potential anammox rates were observed in the different layers of the soil cores, with the higher potential anammox rates being observed in the layers at depths of 0 to 10 cm and 20 to 30 cm (Fig. 4, bottom). The cell-specific anammox rates ranged from 9.5 to 36.2 fmol N per cell per day. The potential n-damo rates ranged from 0.2 ± 0.01 to 2.1 ± 0.08 nmol CO₂ g⁻¹ (dry weight) day⁻¹ in the core samples examined. No n-damo activity could be detected in the layer at a depth of 0 to 10 cm, while obvious n-damo activities were observed in the layers at depths of 20 to 30 cm, 50 to 60 cm, and 90 to 100 cm (Fig. 4, bottom). The cell-specific n-damo rates ranged from 0.3 to 0.4 fmol CO₂ cell⁻¹ day⁻¹.

FIG 5.

Examples of concentrations of ²⁹N₂/³⁰N₂ produced from core samples (collected from the layer at a depth of 90 to 100 cm) amended with ¹⁵NH₄⁺ (a), ¹⁵NH₄⁺ plus NO₂⁻ (b), and ¹⁵NO₂⁻ (c) and examples of concentrations of ¹³CO₂ produced from core samples (collected from the layer at a depth of 90 to 100 cm) amended with ¹³CH₄ (d), ¹³CH₄ plus NO₂⁻ (e), and ¹³CH₄ plus SO₄⁻ (f).

DISCUSSION

Distribution and diversity of anammox bacteria and n-damo bacteria.

Multiple cooccurring anammox populations were found together, and a higher level of n-damo bacterial diversity was also observed in the current study. Soil is a highly heterogeneous environment, and the paddy field may provide diverse microenvironments for different species of anammox bacteria and n-damo bacteria. It was found that only “Candidatus Kuenenia” was detected in the soil layer at a depth of 0 to 10 cm, while only “Candidatus Brocadia” was detected in the soil layer at a depth of 90 to 100 cm (Fig. 2). All the sequences retrieved from the layers at depths of 20 to 30 cm and 50 to 60 cm were affiliated with the new cluster (Fig. 2). The vertical variation in the community structures of anammox bacteria was more or less similar to the results reported by Zhu et al. (9). For the n-damo bacteria, the group A members, which were reported to be the dominant bacteria responsible for the n-damo process (30, 36, 44–46), were detected only in the soil layers at depths of 50 to 60 cm and 90 to 100 cm (Fig. 3). The group A members were also primarily present in the deep layer of the reported wetland systems (below the layer at a depth of 40 to 50 cm) (16, 21, 22).

Abundance of anammox bacteria and n-damo bacteria.

The abundance of anammox bacteria (1.0 × 10⁵ ± 0.04 × 10⁵ to 2.0 × 10⁶ ± 0.14 × 10⁶ copies g⁻¹ [dry weight]) observed in this study was lower than the values reported in freshwater river sediments (10⁶ to 10⁷copies g⁻¹ sediment) (43), while it was within the range found in another paddy field (10⁵ to 10⁷ copies g⁻¹ soil) (9). The abundance of n-damo bacteria (3.8 × 10⁵ ± 0.12 × 10⁵ to 6.1 × 10⁶ ± 0.25 × 10⁶ copies g⁻¹ soil) in the paddy field examined was similar to the values reported for lake sediments (10⁵ to 10⁶ copies g⁻¹ sediment) (15), river sediments (10⁶ to 10⁷ copies g⁻¹ sediment) (18), and wetland systems (10⁶ to 10⁷ copies g⁻¹ soil) (21, 45). The abundance of anammox bacteria showed a trend toward a decrease from the layer at a depth of 0 to 10 cm to that at a depth of 90 to 100 cm (Fig. 4, top), as previously described (9). In contrast, the abundance of n-damo bacteria showed a trend toward an increase from the layer at a depth of 0 to 10 cm to the layer at a depth of 90 to 100 cm (Fig. 4, top). Previous studies also suggested that n-damo bacteria were most abundant in deep wetland soils (16, 17, 21, 45).

Activities and roles of the anammox process and n-damo process.

The potential anammox rates (5.6 ± 0.6 to 22.7 ± 1.0 nmol N₂ g⁻¹ [dry weight] day⁻¹) measured in the paddy field examined were in the same range as those reported in most marine and freshwater environments (11, 47, 48) but lower than the values reported in the land-freshwater interfaces of Baiyangdian Lake (84 to 240 nmol N₂ g⁻¹ day⁻¹) (10). The contribution (8.7 to 29.8%) of anammox to soil N₂ production in the paddy field examined was similar to the values reported for another paddy field (4 to 37%) (9). The potential n-damo rates (0.2 ± 0.01 to 2.1 ± 0.08 nmol CO₂ g⁻¹ [dry weight] day⁻¹) measured in this paddy field were in the same range as those reported in lake sediments (1.8 to 3.6 nmol CO₂ ml⁻¹ day⁻¹) (14) and wetland soils (0.2 to 14.5 nmol CO₂ g⁻¹ day⁻¹) (17, 21, 22).

The cooccurrence of the anammox and n-damo processes was confirmed in different layers of the paddy field examined by incubation experiments. It should be noted that a high concentration of in situ NO_x⁻ (228.6 to 745.1 μmol kg⁻¹ [dry weight] soil) was detected in the upper layer (0 to 30 cm) of the paddy field examined, but a relatively lower concentration of NO_x⁻ (7.9 to 101.7 μmol kg⁻¹ [dry weight] soil) was observed at depths below 50 cm (Fig. 1). The occurrence of the anammox and n-damo processes in the deep layers may be limited by the availability of NO_x⁻ in in situ environments. As a result, the incubation experiments could overestimate the in situ rates of the anammox and n-damo processes in the layers at depths of 50 to 60 cm and 90 to 100 cm because a concentration of 67.8 to 150.0 μmol NO₂⁻ kg⁻¹ (dry weight) soil was added in the slurries. To make a conservative estimate of the nitrogen flux from the anammox process occurring in in situ environments, only the potential anammox rates in the layers at depths of 0 to 10 cm and 20 to 30 cm were used. Therefore, it can be calculated that the nitrogen flux by the anammox process in the paddy field examined was approximately 50.7 g N m⁻² per year on the basis of the reported mean density of paddy soil (1.24 g cm⁻³) (49). The aerobic ammonium oxidation rates reported in wetland soils ranged from 3.7 to 784.8 g N m⁻² per year (50, 51), and the nitrogen loss by denitrification reported in wetland soils ranged from 1.1 to 372.3 g N m⁻² per year (50, 52). Thus, the nitrogen flux (50.7 g N m⁻² per year) by the anammox process is at intermediate levels for the nitrogen flux ranges of aerobic ammonium oxidation and denitrification reported in wetland soils. Similarly, only the potential n-damo rates in the layer at a depth of 20 to 30 cm were used to make a conservative estimate of the level of methane oxidation by the n-damo process in the paddy field examined, and it is estimated that approximately 0.14 g CH₄ m⁻² per year could be oxidized via the n-damo process. The methane oxidation rate by the n-damo process is at the lower end of aerobic methane oxidation rates reported in wetland soils (53).

Higher potential anammox rates were observed in the layers at depths of 0 to 10 cm and 20 to 30 cm, while higher potential n-damo rates were observed in the layers at depths of 50 to 60 cm and 90 to 100 cm (Fig. 4, bottom). In the surface soil layer, a higher NH₄⁺ concentration and a higher NO_x⁻ concentration were observed in the paddy field examined (Fig. 1), and these higher concentrations could stimulate the occurrence of the anammox process, as previously reported (9). Previous studies also indicated that anammox activities in surface soil/sediments were greater than those in deep soil/sediments (9, 10, 54, 55). It was found that members of group A of n-damo bacteria were primarily present in the deep layers (Fig. 3), where a higher abundance of n-damo bacteria was observed (Fig. 4, top). These findings can explain the higher potential n-damo rates measured in the deep layers. Generally, the microbial process using NO_x⁻ as an electron acceptor would be limited by its availability in the deep soil layer because a major part of NO_x⁻ could be consumed in the upper soil layer. However, a certain concentration of NO_x⁻ (7.9 to 101.7 μmol kg⁻¹ [dry weight] soil) was observed in the deep layer of the paddy field examined (Fig. 1). The presence of NO_x⁻ in the deep layer may be because of NO_x⁻ leaching from the upper layer (22). The paddy field examined was frequently irrigated because it has been planted in a rice rotation. In addition, the paddy field has a long history of fertilization, and the total rate of nitrogen fertilization is approximately 300 to 350 g N m⁻² per year. NO_x⁻ leaching has been shown to depend on rates of irrigation and nitrogen fertilization and increases with rates of irrigation and fertilization (56–58).