Ammonia- and Methane-Oxidizing Bacteria: The Abundance, Niches and Compositional Differences for Diverse Soil Layers in Three Flooded Paddy Fields

Ammonia oxidizing bacteria (AOB), Ammonia oxidizing archaea (AOA) and methane oxidizing bacteria (MOB) play cogent roles in oxidation and nitrification processes, and hence have important ecological functions in several ecosystems. However, their distribution and compositional differences in different long-term flooded paddy fields (FPFs) management at different soil depths remains under-investigated. Using qPCR and phylogenetic analysis, this study investigated the abundance, niches, and compositional differences of AOA, AOB, and MOB along with their potential nitrification and oxidation rate in three soil layers from three FPFs (ShaPingBa (SPB), HeChuan (HC), and JiDi (JD)) in Chongqing, China. In all the FPFs, CH4 oxidation occurred mainly in the surface (0–3 cm) and subsurface layers (3–5 cm). A significant difference in potential methane oxidation and nitrification rates was observed among the three FPFs, in which SPB had the highest. The higher amoA genes are the marker for abundance of AOA compared to AOB while pmoA genes, which is the marker for MOB abundance and diversity, indicated their significant role in the nitrification process across the three FPFs. The phylogenetic analysis revealed that AOA were mainly composed of Nitrososphaera, Nitrosospumilus, and Nitrosotalea, while the genus Nitrosomonas accounted for the greatest proportion of AOB in the three soil layers. MOB were mainly composed of Methylocaldum and Methylocystis genera. Overall, this finding pointed to niche differences as well as suitability of the surface and subsurface soil environments for the co-occurrence of ammonia oxidation and methane oxidation in FPFs.


Introduction
Nitrification, through which ammonia (NH 3 ) is converted into nitrite (NO 2 -) or nitrate (NO 3 -), is a key process in global nitrogen (N) cycling and is strongly linked with the release of greenhouse gases to the atmosphere, NO 2 leaching into ground water and N availability for plant use [1]. The biological oxidation of NH 3 into NO 3 is currently considered to be catalyzed by Ammonia oxidizing Bacteria (AOB) within two Proteobacteria subclasses (β and γ subclasses Proteobacteria) and Ammonia oxidizing Archaea (AOA) [2]. Conversely, the oxidation of the second most abundant greenhouse We applied qPCR techniques to identify AOA, AOB and MOB and compared their abundance and occurrence among three FPFs.

Soil Sampling and Site Characteristics
The soil samples used in the present study were selected from three different soil layers, including the surface (0-3 cm), the subsurface (3-5 cm) and the bottom (5-20 cm) layers at three FPFs following crop harvest. The three experimental fields are at the Purple Soil Ecology Experimental Station of Southwest University (JD) (30 • 18.1 • C and 18.2 • C, respectively. The management practices in the three FPFs included conventional fertilization treatments. At the JD sites, the application of 190 kg hm −2 human and animal feces or urine containing urea, 500 kg hm −2 calcium superphosphate, 75 kg hm −2 potassium chloride, 80 kg hm −2 urea topdressing of rice at the tillering stage and 75 kg hm −2 potassium chloride were common practices. At the HC and SPB sites, ammonium sulfate combined with human waste was applied as a fertilizer. In October 2016, three replicate soil samples were randomly collected at three layers in each of the paddy fields using a ZYA-QY Earth drill soil sampler (70 mm × 300 mm, Hangzhou Billion McNair Tech., Zhejiang, China). The soils were homogenized by using manually mixing and removing visible plant debris, rocks and soil micro-fauna, and were then stored at 4 • C. Soil samples for molecular analysis were stored at −20 • C. Other soil samples were air-dried and used for the determination of basic physical and chemical properties.

Determination of Soil Properties
Soil pH was measured in a 1:2.5 (soil:water) suspension using a pH meter (320-S, Mettler-Toledo Instruments Co., Ltd., Shanghai, China). Soil organic matter (SOM) concentration was measured using the dichromate oxidation method [22]. Total nitrogen (TN) was measured using the kjeldahl method [23] and bulk density using the ring knife method. Redox potential was calculated using the depolarization method (FJA-4 redox depolarization automatic analyzer) [24]. The results for the soil chemical property tests are presented in Table 1.

Potential Nitrification Rates (PNRs) and Methane Oxidation Potentials (MOP)
Potential nitrification rates (PNRs) were tested using the Chlorate (sodium chlorate) inhibition method [13]. Dry soil samples (5 g) were added into a 50 mL centrifuge tube that contained 20 mL of 1 mM (NH 4 ) 2 SO 4 and 1 mL of sodium chlorate solution. The tube was maintained at 25 • C for 24 h in the dark and later retrieved for further analysis. The oxidation rate was calculated as follows: S-Soil sample value (mg N); C-Control (mg N) The determination of soil CH 4 oxidation potential was carried out according to Lü et al. (2005) [25] with some modifications. Briefly, the CH 4 gas initially added to the pre-culture samples was released following the opening of the culture flask. As 1% CH 4 gas was added to the samples, 1% difluoromethane was also added to the control treatment as a CH 4 oxidation inhibitor. The culture suspensions were carefully shaken after sealing the pinhole. Each treatment had three replicates and was incubated at 25 • C in the dark. Sampling and soil CH 4 oxidation potential determination were carried out on the first, fourth, seventh and tenth days after the beginning of the culture. The CH 4 oxidation rate was calculated based on the slope when the concentration of CH 4 in the bottle decreased linearly.
The ratio of the difference of the CH 4 concentration to the sampling time interval measured based on two consecutive samples during the culture period represented the CH 4 oxidation rate during the period, and the oxidation rate was calculated as follows: where: Q: the oxidation rate of CH 4 (µg g −1 h −1 ); dc/dt: change of CH 4 concentration per unit time (ppm mol −1 h −1 ); V: the volume of gas in the culture flask (L); W: the dry soil weight (g); MW: the molecular weight (g) of CH 4 ; MV: the volume (L) of 1 mol of gas in a standard state.

Molecular Analyses
A FastDNA SPin Kit for Soil (MP Biomedicals, LLC, United States)was used to extract soil DNA. Amplification of the qPCR products for AOA, AOB and MOB was conducted on a CFX-96 Optical Real-Time PCR System (Bio-Rad Inc. Hercules, CA, US). The primer pairs amoA-1F/amoA-2R [26], Arch-amoAF/Arch-amoAR [27], and A189f/682r [28], A189f/mb661 [29] (Table 2) were adopted for the quantitative PCR (q-PCR) assays of AOA, AOB and MOB abundance, respectively. The total volume of the PCR reaction system was 20 µL, which was 10 µL of SYBR@ Premix EX TaqTM (TaKaRa), 0.3 µL of the upstream and downstream primers (20 pmol µL −1 ), 1.0 µL of the sample and 8.4 µL of sterilized double distilled water (ddH 2 O). The reaction conditions are listed in Table 3. The standard curve for quantitative PCR was used to clone an AOA and AOB amoA or an MOB pmoA functional gene into a vector plasmid (from Shanghai Meiji Sequencing Co., Ltd.).

Cloning, Sequencing, and Phylogenetic Analysis
The target gene and Escherichia coli were recombined in vitro and then introduced into competent cells for expression and amplification. The positive clones were sent to Shanghai Meiji Company for further sequencing using a pEASY-T3 Cloning Kit (TransGen Biotech, Beijing, China). The result obtained for the cloned sequences was subjected to sequence quality control using Chromas (Technelysium, Helensvale, Queensland, Australia), and the vector sequences were deleted from the text to obtain the target gene sequences. A Blast comparison was conducted by searching the National Center for Biotechnology Information (NCBI) Gen Bank database to obtain highly similar homologous gene sequences. MEGA v4.0 [30] was used for the analysis and the sequences of all gene clones were classified based on Operating Taxonomic Units (OTUs). The neighbor joining method was used to establish phylogenetic trees to determine the phylogenetic status of sequencing clones based on their positions and genetic distances in phylogenetic trees.

Potential Nitrification Rates
Potential nitrification rates were different among the FPFs and among the different soil layers. The nitrification activities ranged from 0.0150 µg N g −1 h −1 in HC soil to 0.1680 µg N g −1 h −1 in SPB soil ( Figure 1). In all the FPFs, nitrification activity was significantly higher in the subsurface layers (3-5 cm), although at different rates. Compared to the surface soil layer (0-3 cm), the potential nitrification rate in HC soil was significantly lower in the bottom layer (5-20 cm), although it was higher in SPB and JD soil ( Figure 1).

AOA, AOB, and MOB Abundance
The results of the q-PCR analysis of amoA and pmoA were used to determine the AOA, AOB and MOB abundance in the SPB, HC and JD soils. AOA abundance ranged from 5.28 to 1.09 × 10 7 /g dry soil, 5.36 to 1.59 × 10 7 /g dry soil and 3.64 to 1.10 × 10 7 /g dry soil archaeal amoA gene copies in the SPB, HC and JD soils, respectively (Figure 3a). The highest AOA amoA abundance was observed in the bottom (5-20 cm) layer of the SPB and JD soils. AOB abundance ranged from 3.81 to 7.50 × 10 6 /g dry soil, 1.10 to 4.27 × 10 6 / g dry soil, and 1.43 to 3.14 × 10 6 /g dry soil bacterial amoA gene in the SPB, HC and JD soil samples, respectively (Figure 3b). The AOB amoA gene copies in the subsurface (3-5 cm) layer of the SPB, HC and JD soils were significantly higher than in the bottom (5-20 cm) layer. MOB abundance ranged from 2.56 to 1.05 × 10 6 / g dry soil, 7.46 to 1.02 × 10 6 /g dry soil, and 2.92 to 1.04 × 10 6 / g dry soil pmoA gene copies in the SPB, HC and JD soils, respectively (Figure 3c). The MOB pmoA gene copies at the SPB site decreased with an increase in soil layer depth; however, no consistent trend in MOB pmoA abundance was observed in the HC and JD sites among the soil layers. The AOA/AOB Sustainability 2020, 12, 953 8 of 24 ratio at 0-3 cm and 5-20 cm was greater than 10 across the three sites (Table 4) while the AOB/MOB ratio in all the three sites and at the different layers was much lower than 10.  layer of the SPB, HC and JD soils were significantly higher than in the bottom (5-20 cm) layer. MOB abundance ranged from 2.56 to 1.05 × 10 6 / g dry soil, 7.46 to 1.02 × 10 6 /g dry soil, and 2.92 to 1.04 × 10 6 / g dry soil pmoA gene copies in the SPB, HC and JD soils, respectively (Figure 3c). The MOB pmoA gene copies at the SPB site decreased with an increase in soil layer depth; however, no consistent trend in MOB pmoA abundance was observed in the HC and JD sites among the soil layers. The AOA/AOB ratio at 0-3 cm and 5-20 cm was greater than 10 across the three sites (Table 4) while the AOB/MOB ratio in all the three sites and at the different layers was much lower than 10.

AOA, AOB, and MOB Community Composition
A total of 90 positive sequences were used for AOA, AOB amoA and MOB pmoA functional gene sequencing and the construction of a clone library (Figure 4). Phylogenetic analysis revealed that the AOA amoA in the surface layer in HC belonged to Group 1.1b, while 97% of AOA amoA in the subsurface layer belonged to the subordinates Group 1.1b and 3% to Group 1.1a associated. Sixty-four percent of AOA amoA in the bottom layer belonged to the subordinates Group 1.1b and 36% to Group 1.1a associated (Figure 4a). In addition, the phylogenetic analysis revealed that 92% of AOA amoA in the surface layer of SPB belonged to Group 1.1b, while 8% belonged to Group 1.1a. Fifty-six percent of AOA amoA in the subsurface layer of SPB belonged to subordinates Group 1.1b and 44% to Group 1.1a associated; while in the bottom layer, 12% of the AOA amoA belonged to Group 1.1b, and 88% of the subordinates belonged to Group 1.1a associated (Figure 4b).
At the JD site, 84% of AOA amoA belonged to Group 1.1b and 16% to Group 1.1a, while 36% belonged to Group 1.1b and 64% belonged to Group 1.1a in the subsurface and the bottom layers (Figure 4c). Nitrosospira clusters were observed in the surface and bottom layers, accounting for 17% and 14%, respectively. In addition, the community diversity of AOA exceeded that of AOB considerably, and AOB amoA was affiliated only with Nitrosomonas in the HC and SPB sites ( Figure 5).
The phylogenetic analysis of MOB pmoA assigned 69% of the sequences to Type Ia and the residual 31% of the sequences to type II MOB in the HC surface layer. In the HC subsurface layer, MOB pmoA consisted of subordinate Type Ia (18%), Type Ib (37%) and Type II (45%). In the HC bottom layer, 24% of the sequences belonged to Type Ib and 76% belonged to Type II. In the surface layer at the SPB site, only 11% of the sequences were assigned to type II methanotrophs while the rest belonged to Type Ia (30%) and Type Ib (59%). In the SPB subsurface layer, 44% of the sequences were associated with Type Ib, and 56% were associated with Type I methanotrophs. Only 8% of MOB pmoA sequences were classified into Type Ia methanotrophs and the rest were associated with Type II. In the surface layer of the JD site, 17% of the sequences were assigned to Type Ia and the rest of the sequences were type Ib MOBs. The subsurface MOB pmoA belonged to Type Ib bacteria (38%) and Type II bacteria (62%). For the bottom layer soil at the JD site, the sequences belonged to Type Ia (25%), Type Ib (21%) and Type II (54%) (Figure 6).     The phylogenetic analysis of MOB pmoA assigned 69% of the sequences to Type Ia and the residual 31% of the sequences to type II MOB in the HC surface layer. In the HC subsurface layer,  The number in brackets after the gene or clone name represents the sequence number of the OTU. The scale represents the genetic distance, and the number represents five differences in every 100 bases. H-1 represents a 0-3 cm soil layer, H-2 represents a 3-5 cm soil layer and H-3 represents a 5-20 cm soil layer in Figure 5a; Q-1 represents a 0-3 cm soil layer, Q-2 represents a 3-5 cm soil layer, and Q-3 represents a 5-20 cm soil layer in Figure 5b;J-1 represents a 0-3 cm soil layer, J-2 represents a 3-5 cm soil layer, and J-3 represents a 5-20 cm soil layer in Figure 5c.The scale indicates the number of nucleotide substitutions per site. associated with Type Ib, and 56% were associated with Type I methanotrophs. Only 8% of MOB pmoA The number in brackets after the gene or clone name represents the sequence number of the OTU. The scale represents the genetic distance, and the number represents five differences in every 100 bases. H-1 represents a 0-3 cm soil layer, H-2 represents a 3-5 cm soil layer and H-3 represents a 5-20 cm soil layer in Figure 6a; Q-1 represents a 0-3 cm soil layer, Q-2 represents a 3-5 cm soil layer and Q-3 represents a 5-20 cm soil layer in Figure 6b; J-1 represents a 0-3 cm soil layer, J-2 represents a 3-5 cm soil layer and J-3 represents a 5-20 cm soil layer in Figure 6c. The scale indicates the number of nucleotide substitutions per site. Numbers in parentheses against each abbreviation indicate the number of sequences recovered from each sample.

Spatial Occurrence and Abundance of AOA, AOB and MOB among the FPFs
The present study explored the occurrence and niche distribution of AOA, AOB and MOB among three FPFs. Although some studies have suggested that the oxidation of NH3 is mainly

Spatial Occurrence and Abundance of AOA, AOB and MOB among the FPFs
The present study explored the occurrence and niche distribution of AOA, AOB and MOB among three FPFs. Although some studies have suggested that the oxidation of NH 3 is mainly attributed to AOA [31,32], the presence of both AOA and AOB in all the three FPFs in the present study indicated that the two groups could jointly play a role in the transformation of NH 3 to NO 3 -. Consistent with the findings of previous studies across different ecosystems [16,33], we observed that AOA abundance has a more considerable role than AOB and MOB across the three FPFs, with the highest amoA observed in the bottom soil layers, particularly in the SPB and JD sites. Although the results of previous studies have suggested that AOA numerically outnumber AOB in acidic soil, in the present study we observed that despite the soil pH values across the FPFs moving away from being acidic to alkaline (Table 1), the AOA abundance was still significantly higher than AOB abundance in all FPFs, and were consistent with the higher potential nitrification rates (PNRs), particularly in the SPB and JD sites. The results indicated a reduced affinity of soil pH and AOA for NH 3 , which enabled them to utilize low concentrations of NH 3 in the acidic soils and thus influenced their abundance in the present study. However, the higher AOA observed in the present study could be attributed to their capacity to adapt to low O 2 concentration environments, which is a common phenomenon in paddy fields [34], or the high organic matter concentrations that create NH 3 assimilation pathways in all the three sites (Table 1 in reference [35]). In addition, although the AOA/AOB ratio was higher than 10 in all the three sites, the AOB/MOB ratio was much lower than ten, indicating that the AOA amoA functional gene copy numbers in the three sites were significantly higher than the AOB amoA and MOB pmoA functional gene copy numbers, by at least one order of magnitude. Since potential nitrifying activity is linked strongly to soil NO 2 content, the higher PNRs in the 3-5 cm soil layer indicated that the risk of NO 2 leaching could be higher in the surface soil. Therefore, in all the studied flooded paddy fields, the surface soil requires more attention with regard to NO 2 leaching.
Unlike AOA and AOB, approximately 50% to 90% of the CH 4 in the soil surface and the rhizosphere is oxidized by MOB prior to being emitted into the atmosphere, thus there is a need to pay further attention to their occurrence and distribution in FPFs [6]. Similar to AOA and AOB, MOB occurred in all the studied FPFs and there was no significant difference in their abundance among the sites. However, the potential CH 4 oxidation rates across the FPFs indicated that SPB paddy fields had the capacity to increase CH 4 oxidation capacity considerably compared to the other FPFs ( Figure 2). MOB abundance across the FPFs was low compared to AOA and AOB abundance, which suggests that although MOB might contribute significantly to the nitrification process in all the three FPFs, their specific nitrification rates could be much lower than the AOA and AOB nitrification rates [5].

Community Structure of AOA, AOB, and MOB at Different Soil Depth among the FPFs
The results of the phylogenetic analysis in the present study revealed that the members of three archaea genera, including Nitrososphaera, Nitrosopumilus, and Nitrosotalea were enriched in the three FPFs. Members of bacterial genus Nitrosomonas were the most dominant NH 3 oxidizers in all the soils. In addition, genus Nitrososphaera was among the most abundant NH 3 oxidizers in the HC and the SPB surface soil. Members of genus Nitrosotalea were highly enriched in the JD subsurface and bottom soil layers, while members of Nitrosopumilus were only dominant in the JD subsurface and bottom soil layers, and Nitrososphaera were still dominant in the surface layer. The results indicated that different NH 3 oxidizers occupied different environmental niches in the FPFs. Several factors, including substrate concentration, pH, temperature and O 2 have been reported to influence the structural composition of NH 3 oxidizers in different ecosystems [36,37]. However, the distinction in the community structure across the FPFs in the present study could not be attributed solely to the environmental drivers because there were no significant differences in either temperature or pH among the FPFs.
In the present study, MOB pmoA distribution in the FPFs varied considerably; however, community distribution increased with an increase in the soil layer depth. Type I MOB were enriched in the surface and subsurface layers, although Type II MOB were enriched in the bottom layer. The results indicated that type I MOB abundance was higher than type II MOB, which was consistent with the findings of Steenbergh et al. (2010) [38], who proposed different life strategies for type I and type II methanotroph. It has been reported that Type I MOB preferentially grow at high O 2 and low CH 4 concentration environments, while Type II MOB preferentially grow at low O 2 and high CH 4 concentration environments [39]. Therefore, Type I and II MOB occupy different niches [40]. The results also demonstrated the uniqueness of each soil environment and the effects of ecological factors on MOB at the species level. The MOB communities in the FPFs were relatively stable. The Type I MOB belong to the Methylococcaceae type I methanotroph genera (Type I a) and the genus Methylocaldum (Type I b), while Type II MOB are the genus Methylocystis. The results are consistent with the findings of a previous study that suggested that the major contributor to CH 4 oxidation in the atmosphere is type II MOB, particularly Methylocystis sp [41]. Plant cultivation, moisture content and temperature also influence MOB diversity [42]. In addition, soil moisture could influence soil CH 4 absorption [43]. In the present study, the application of fertilizer in paddy fields enhanced CH 4 oxidation. The inorganic N concentration in the soil influenced MOB activity. The MOB CH 4 oxidation activity is near zero when N ions in the soil are exhausted, with three potential mechanisms: (1) NH 4 + -or NO 2 --nitrogen increase the activity of CH 4 oxidizing microorganisms; (2) N participates in the synthesis of enzymes involved in CH 4 oxidation; and (3) an increase in nitrification bacteria diversity and activity could enhance CH 4 oxidation [44,45]. Nevertheless, the specific mechanism of action requires further study.

Conclusions
This study highlighted that AOA abundance was higher than AOB and MOB abundance in all the FPFs, which was consistent with the potential nitrification rates (PNRs), particularly in the SPB and JD sites. Three AOA genera (Nitrososphaera, Nitrosopumilus, Nitrosotalea) were observed in all the FPFs, and Nitrososphaera were dominant in the surface layers. In addition, Nitrososphaera were among the most abundant NH 3 oxidizers in the HC and in the SPB surface soil, while members of genus Nitrosotalea were considerably enriched in the JD subsurface and bottom layers. MOB community distribution increased with an increase in the depths of the soil layers, and Type I MOB were enriched in the surface and subsurface layers, while Type II MOB were enriched in the bottom layers. Overall, this study provides a basis for improving soil nitrogen use efficiency and mitigating soil greenhouse gas emissions in flooded paddy fields.