Calcium Superphosphate-Mediated Reshaping of Denitrifying Bacteria Community Contributed to N2O Mitigation in Pig Manure Windrow Composting

Composting is recognized as an effective strategy for the sustainable use of organic wastes, but also as an important emission source of nitrous oxide (N2O) contributing to global warming. The effects of calcium superphosphate (CaSSP) on N2O production during composting are reported to be controversial, and the intrinsic microbial mechanism remains unclear. Here, a pig manure windrow composting experiment lasting for ~60 days was performed to evaluate the effects of CaSSP amendment (5%, w/w) on N2O fluxes in situ, and to determine the denitrifiers’ response, and their driving factors. Results indicated that CaSSP amendment significantly reduced N2O emissions as compared to the control pile (maximum N2O emission rate reduced by 64.5% and total emission decreased by 49.8%). CaSSP amendment reduced the abundance of nirK gene encoding for nitrite reductase, while the abundance of nosZ gene (N2O reductase) was enriched. Finally, we built a schematic model and indicated that the abundance of nirK gene was likely to play a key role in mediating N2O production, which were correlated with NH4+-N and NO3−-N changing responsive to CaSSP. Our finding implicates that CaSSP application could be a potential strategy for N2O mitigation in manure windrow composting, and the revealed microbial mechanism is helpful for deepening the understanding of the interaction among N-cycle functional genes, physicochemical factors, and greenhouse gases (GHG) emissions.


Introduction
Composting is considered to be one of the effective ways for organic wastes disposal, which is presented as an environmentally friendly and cost-efficient process to manage and recycle organic solid wastes [1]. Nevertheless, gaseous nitrogen (N) losses constitute a significant challenge of manure composting, which are detrimental to the environment and can cause secondary pollution [2][3][4][5][6]). Nitrous oxide (N 2 O) is a potent greenhouse gas (GHG) that exhibits global warming potential of 298 times that of carbon dioxide (CO 2 ) on a 100-year time horizon [7]. The N 2 O emissions from manure composting may constitute 0.2-9.9% of initial total nitrogen (TN) loss of raw materials [8,9].
Several strategies have been developed to mitigate gaseous emissions during composting, including aeration control [3,[10][11][12], bulking agents [13], and amendment 2 of 11 addition [14][15][16][17], as well as microbial inoculation [18]. Calcium superphosphate (CaSSP), a commonly used and low-cost phosphorus-containing additive fertilizer, has been proposed as a potential strategy to mitigate N loss from manure composting systems [19,20]. Luo et al. [15] and Yuan et al. [21] reported that the addition of CaSSP decreased N 2 O emissions by 22.2-55.4% during composting. By contrast, Yang et al. [22] showed that the CaSSP amendment increased N 2 O emissions during kitchen waste composting. These contradictory effects of CaSSP addition on N 2 O emissions may relate to the variation of physicochemical properties, and particular the involved microbial processes during composting.
It is well documented that the formation of N 2 O during composting were contributed to different microbial nitrogen-transforming pathways, including nitrification and denitrification processes under various oxygen conditions [23,24]. Nitrification is a two-step oxidation of ammonium (NH 4 + ) to nitrate (NO 3 − ) via nitrite (NO 2 − ) by chemolithoautotrophic ammonia oxidizers and nitrite oxidizers, where ammonia monooxygenase encoding by amoA is known to an important enzyme in this process; while microbial denitrification represents the stepwise reduction of nitrate or nitrite to N 2 covered by four denitrified genes (narG, nirK/S, norB and nosZ) [25][26][27]. It has been highlighted that nitrifying and denitrifying gene abundance could predict N 2 O emissions as potential proxies [28]. Importantly, these two nitrogen-transforming processes were regulated by various abiotic factors [29]. Changes in crucial environmental factors during composting can shape the N 2 O-related functional microbes [24], which will then directly affect N 2 O emissions. Several studies have documented that CaSSP addition during composting can affect the abiotic characteristics of composts, such as decreasing pH and shifting the NH 4 + -N and NO 3 − -N content, which was considered to be the primary factors driving the increase/decrease in N 2 O emissions [30,31]. However, the correlation between these changes in physicochemical factors and microbial community is ignored, and it is not yet clear how CaSSP addition can affect the behavior of microbial functional groups during composting. Therefore, more attention is needed to survey the response of functional microbial populations to CaSSP addition during composting, as well as the intermediary physicochemical factors, so as to understand the underlying mechanism that CaSSP addition affects N 2 O emissions.
In this study, commercial windrow composting was carried out to measure N 2 O fluxes in situ. The abundance of functional genes involved in N 2 O emissions were quantified by real-time quantitative PCR. The main objectives of this study were to (1) examine the dynamics of N 2 O emissions with CaSSP addition during composting; (2) identify the CaSSP-mediated key microbial pathways that affect N 2 O emissions during composting; (3) illustrate the CaSSP-responsive physicochemical factors that shape the functional microbial groups. We successfully identified that CaSSP addition mediated NH 4 + -N/NO 3 − -N transformation in pig manure composting, consequently suppressed the abundance of nirK genes, and significantly decreased N 2 O emissions, which firstly explains the microbiology mechanism involved in CaSSP-mediated N 2 O mitigation in composting.

Windrow Composting Experiment
The windrow composting system for~60 days was conducted in a fertilizer company, located at Nanjing, Jiangsu province, China. Two treatments were carried out using pig manure mixed completely with vegetable (Broccoli) and wheat straw at a dry mass ratio of 6:1 as raw materials. In addition to the control treatment (CK), calcium superphosphate (CaSSP) was amended with a weight of 5% (w/w, dry weight (DW); according to Luo et al. [15] and Yuan et al. [21] of the raw composting materials in calcium superphosphate treatment (CaSSP treatment). Both treatments contained three windrow composts for replications. The volume of each composting pile was~20 m 3 (12 m in length, 1.5 m in width and 1 m in height). The composting process included two phases: phase I was the bio-oxidative phase with regularly turning by a composter turner every two days for~30 days; phase II was the post-maturation phase without turning. Each pile was located for gas sampling respectively. Composting material samples were collected near the positions of gas sampling and included three replicates.

Gas Sampling and Flux Measurement
N 2 O fluxes were measured using closed static chamber system (diameter 0.45 m, height 5 m) [13,32]. For this, PVC chamber bases (30 cm length × 30 cm width × 25 cm height) were inserted 25 cm into the pile at 10-12 h before gas sampling. Gas samples were collected once a week. The opaque chamber was placed on the peak of each windrow with rim of the chamber fitted into the groove of collar. As sampling, the groove in the top edge of the collar was filled with water to seal the rim of the chamber ( Figure S1). At 0, 5, 10, 20, and 30 min after the chamber closure, 60 mL of gas samples were extracted with an airtight syringe and immediately injected into 100 mL pre-evacuated gas sampling bags. Gas samples were taken mainly between 8:00 and 10:00 am on each sampling day and then transported to the laboratory for analysis by gas chromatograph (GC) within a few hours. Concentrations of N 2 O were measured by a Gas Chromatograph (Agilent 7890A, Agilent Technologies, Inc. Wilmington, USA) with a flame ionization detector (FID) and an electron capture detector (ECD) [32,33]. N 2 O fluxes were calculated by a linear regression as described by Hutchinson and Mosier [34]. Total N 2 O emissions were sequentially accumulated from the fluxes between every two adjacent measurements [32].

Physicochemical Properties Determination
During the aerobic composting process, windrow temperature at a depth of 30 cm was monitored by a mercury thermometer and recorded at each sampling day. Material samples were divided into three parts after passing through 2-mm sieve: one part was air-dried for total carbon (TC) and total nitrogen (TN) analyses. The other two parts were immediately stored at 4 • C and −80 • C for further analysis, respectively. Moisture content of fresh samples was determined as a weight loss upon drying at 105 • C in an oven for 24 h. Inorganic nitrogen, including ammonium (NH 4 + ) and nitrate (NO 3 − ), were extracted from material samples using 2 mol L −1 potassium chloride (KCl) solution and then analyzed using the segmented flow analyzer (Auto Analyzer 3, Seal, Germany). For pH measurement, the samples were diluted with deionized water at a volume ratio of 1:10 and analyzed by a pH electrode (PHS-3C mv/pH detector, Shanghai, China). The TC content, TN content and C/N ratio were determined using an element analyzer (Elementar Vario Macro cube, Hanau, Germany).

Real-Time Quantitative PCR (q-PCR) Assays of the Functional Genes
According to the manufacturer's instructions, DNA samples were extracted from composting materials (days 1, 10, 24, 31, 45, and 60) in triplicate using the UltraClean soil DNA isolation kit (Qiagen, Hilden, Germany), and then stored at −20 • C for q-PCR assays. The Nanodrop (ThermoFisher Scientific, Waltham, MA, USA) was used to determine the quantity and quality of DNA samples. Functional genes associated with nitrification and denitrification were quantified from the composting process to elucidate the role of nitrifying and denitrifying bacteria on N 2 O dynamics. Ammonium monooxygenase gene (amoA) was selected as nitrify marker gene, which include the ammonia oxidizing bacteria (AOB) and archaea (AOA), while the nitric oxide reductase gene (nirK/S) and nitrous oxide reductase gene (nosZ) were targeted as denitrify marker genes. The standard curves were made using stepwise 10-fold diluted pMD-18 vectors with correct functional gene fragments. Each DNA sample was diluted with sterile water to 10 ng DNA µL −1 before q-PCR assays. Amplification was performed using StepOne Plus Real-time PCR System (Thermo Fisher Scientific, Waltham, MA, USA). PCR mixtures contained 10 µL SYBR@ Premix Ex Taq TM (Takara, Dalian, China), 0.4 µL forward and reverse primers, 0.4 µL ROX reference dye II (50×), 2 µL template DNA, and 6.8 µL sterile water with a final volume of 20 µL. Primers sequence and PCR reaction conditions were listed in Table 1.

Statistical Analysis
All data were reported as means and standard errors. After checking the normal distribution of results and the homogeneity of variances using Kolmogorov-Smirnov (K-S) and Shapiro-Wilk (S-W) test by SPSS 22.0 (SPSS Inc., Chicago, IL, USA), a Tukey post hoc test was used to determine the significant difference between Control and CaSSP treatments. Pearson correlation analyses between N 2 O fluxes, functional genes abundance and physicochemical properties were carried out by using SPSS 22.0. Structural equation modeling (SEM) analysis conducted using AMOS 22.0 (Amos, Development Corporation, Meadville, PA, USA) was established to address the direct and indirect effects of calcium superphosphate on the soil properties, functional genes abundance, and N 2 O emissions.

N 2 O Fluxes
Over the 60-day period of manure composting, N 2 O emission patterns and temporal trends for the different treatments were similar (Figure 1a). At the beginning, N 2 O fluxes maintained a constantly low level, then increased sharply on day 52 and lasted until the end of the composting. However, the peak N 2 O emission rate from CaSSP treatment (58.32 mg m −2 h −1 ) was significantly lower than that from Control (164.46 mg m −2 h −1 ), resulting in a 64.5% reduction of the maximum N 2 O emission rate brought by CaSSP application. Consequently, CaSSP amendment significantly reduced total N 2 O emission over the total composting period by 49.8% (1.18 g m −2 vs. 2.35 g m −2 ; Supplementary Figure S2).

Physical and Chemical Characteristics Measurement
Composting temperature in Control and CaSSP treatments generally showed similar variation patterns with a scope of 42 • C to 71 • C (Figure 1b), which went through three typical degradation phases (mesophilic, thermophilic and curing phases). For both treatments, the temperature increased and entered the thermophilic phase on day 10 and peaked on day 31 (68-71 • C). In addition, similar trends of moisture of the two treatments over the whole manure composting were observed. They decreased constantly to day 38 and then remained a stable level, with a range from 12-14% (Figure 1b).

Physical and Chemical Characteristics Measurement
Composting temperature in Control and CaSSP treatments generally showed similar variation patterns with a scope of 42 °C to 71 °C (Figure 1b), which went through three typical degradation phases (mesophilic, thermophilic and curing phases). For both treatments, the temperature increased and entered the thermophilic phase on day 10 and peaked on day 31 (68-71 °C). In addition, similar trends of moisture of the two treatments over the whole manure composting were observed. They decreased constantly to day 38 and then remained a stable level, with a range from 12-14% (Figure 1b).
The pH values of the two treatments remained at 6.8 to 7.6 during the whole composting process (Figure 2a). An increase in pH was noted in the control treatment within the first 10 days, and the peak was achieved at 7.5 on day 17. On the contrary, adding CaSSP decreased the pH due to its acidic feature.
The dynamic changes of TN with time were shown in Figure 2b. Along with the reduction of carbonaceous materials (Figure 2c), the TN content of the both treatments increased continuously. During composting, the rate of organic nitrogen mineralization was lower than that of organic carbon, causing the C/N ratio of the two treatments to decrease ( Figure 2d).
As shown in Figure 3a,b, soil NO3 − -N and NH4 + -N contents ranged from 0.21 to 0.77 mg N kg −1 and 1.96 to 3.21 mg N kg −1 , respectively, which showed a trade-off between each other during the composting, as NH4 + -N contents generally revealed a decreased pattern along the time while that for NO3 − -N contents was opposite. As expected, NH4 + -N contents of CaSSP treatment were slightly higher than the Control, which could be at- The pH values of the two treatments remained at 6.8 to 7.6 during the whole composting process (Figure 2a). An increase in pH was noted in the control treatment within the first 10 days, and the peak was achieved at 7.5 on day 17. On the contrary, adding CaSSP decreased the pH due to its acidic feature.   The dynamic changes of TN with time were shown in Figure 2b. Along with the reduction of carbonaceous materials (Figure 2c), the TN content of the both treatments increased continuously. During composting, the rate of organic nitrogen mineralization was lower than that of organic carbon, causing the C/N ratio of the two treatments to decrease ( Figure 2d).
As shown in Figure 3a,b, soil NO 3 − -N and NH 4 + -N contents ranged from 0.21 to 0.77 mg N kg −1 and 1.96 to 3.21 mg N kg −1 , respectively, which showed a trade-off between each other during the composting, as NH 4 + -N contents generally revealed a decreased pattern along the time while that for NO 3 − -N contents was opposite. As expected, NH 4 + -N contents of CaSSP treatment were slightly higher than the Control, which could be attributed to that the lower pH caused by CaSSP addition hindered the conversion of NH 4 + -N to NH 3 . The NO 3 − -N concentration was distinctly lower than the NH 4 + -N concentration during the composting of CK and CaSSP. The NO 3 − -N concentration in CK and CaSSP piles decreased from 0.33 and 0.29 g kg −1 DW to 0.26 and 0.22 g kg −1 DW after the first 10 days composting, respectively (Figure 3b). The high temperature inhibited nitrifying bacteria activity and maintained the low NO 3 − -N concentration in the both treatments from day 10-30. Afterwards, the NO 3 − -N concentration in CK and CaSSP treatments slowly elevated until the end of the experiment. first 10 days composting, respectively (Figure 3b). The high temperature inhibited nitrifying bacteria activity and maintained the low NO3 − -N concentration in the both treatments from day 10-30. Afterwards, the NO3 − -N concentration in CK and CaSSP treatments slowly elevated until the end of the experiment.

Quantification of the Functional Genes Involved in N2O Emission
The nitrifying genes (AOA amoA and AOB amoA) and denitrifying genes (narG, nirK, nirS and nosZ) were quantified in the composting process ( Figure 4). The copy numbers of AOB amoA in CK and CaSSP piles increased during the first 10 days composting, and

Quantification of the Functional Genes Involved in N 2 O Emission
The nitrifying genes (AOA amoA and AOB amoA) and denitrifying genes (narG, nirK, nirS and nosZ) were quantified in the composting process ( Figure 4). The copy numbers of AOB amoA in CK and CaSSP piles increased during the first 10 days composting, and then gradually decreased. After 31 days, a slightly increase was observed and the highest value occurred on day 60 of composting ( Figure 4a); similar changing trend was observed in AOA amoA in both treatments (Figure 4b). The AOB amoA abundance in CaSSP treatment was higher than that in the Control pile, especially between 10 and 50 days; while the AOA amoA population was found to be similar in both treatments (Figure 4a,b). The abundance of nirS and nirK generally remained a consistently level at the beginning of the composting, and then showed a continuous increase and reached the peak finally (Figure 4c,d). Importantly, after 30 days, both the population of nirS and nirK in Control was observed to be significantly higher than that in CaSSP treatment (p < 0.05) (Figure 4c,d). On the contrary, the abundance of nosZ increased dramatically on day 10 and then showed a continuous decrease until the end of experiment; the Control pile revealed a lower nosZ population than CaSSP treatment before 30 days (Figure 4e). of the composting, and then showed a continuous increase and reached the peak finally (Figure 4c,d). Importantly, after 30 days, both the population of nirS and nirK in Control was observed to be significantly higher than that in CaSSP treatment (p < 0.05) (Figure  4c,d). On the contrary, the abundance of nosZ increased dramatically on day 10 and then showed a continuous decrease until the end of experiment; the Control pile revealed a lower nosZ population than CaSSP treatment before 30 days (Figure 4e).

Linking N2O Emissions to the Abundance of N-Cycling Functional Microbial Genes in CaSSP-Amended Compost
The N2O emissions were correlated with the abundance of N-cycling microbial functional genes and soil properties ( Table 2). The N2O cumulative emissions measured in the CaSSP treatment displayed significant positive correlations with the abundance of nirK (r = 0.79, p < 0.05). However, correlation analyses showed that the N2O cumulative emissions shared negative correlations with pile moisture, Total C, C/N ratio, and NH4 + -N.

Linking N 2 O Emissions to the Abundance of N-Cycling Functional Microbial Genes in CaSSP-Amended Compost
The N 2 O emissions were correlated with the abundance of N-cycling microbial functional genes and soil properties ( Table 2). The N 2 O cumulative emissions measured in the CaSSP treatment displayed significant positive correlations with the abundance of nirK (r = 0.79, p < 0.05). However, correlation analyses showed that the N 2 O cumulative emissions shared negative correlations with pile moisture, Total C, C/N ratio, and NH 4 + -N. Table 2. Correlations between the abundance of N 2 O-forming and consuming microbial functional genes and compost physicochemical properties with calcium superphosphate (CaSSP) addition during composting. To better understand the effect of CaSSP on N 2 O emissions linked to the nitrifier and denitrifier functional gene abundance, path analysis was used to examine the interactions among the variables ( Figure 5). From this, the addition of CaSSP had direct and significant influence on NH 4 + -N, whereas indirectly influenced nirK abundance via changing NO 3 − -N. Therefore, the abundance of the nirK gene showed a significant and direct effect on N 2 O emissions (λ = 0.97, p < 0.05).

Parameters
To better understand the effect of CaSSP on N2O emissions linked to the nitrifier and denitrifier functional gene abundance, path analysis was used to examine the interactions among the variables ( Figure 5). From this, the addition of CaSSP had direct and significant influence on NH4 + -N, whereas indirectly influenced nirK abundance via changing NO3 − -N. Therefore, the abundance of the nirK gene showed a significant and direct effect on N2O emissions (λ = 0.97, p < 0.05).

Discussion
The present study was performed to better understand the process by addressing the effects of CaSSP amendment on N2O mitigation during composting, as well as improve the mechanistic understanding of the role of microbial denitrification contributing to this mitigation.
N2O emissions in Control and CaSSP treatments observed in maturity stage during composting (Figure 1a), which may be attributed to accumulated NO3 − -N during the thermophilic phase. It has been reported that high temperature (above 40 °C) and high concentrations of free ammonia may inhibit nitrifiers [12,40]. Here, it was observed that the CaSSP application significantly decreased the total N2O emissions by 49.8% com-

Discussion
The present study was performed to better understand the process by addressing the effects of CaSSP amendment on N 2 O mitigation during composting, as well as improve the mechanistic understanding of the role of microbial denitrification contributing to this mitigation. N 2 O emissions in Control and CaSSP treatments observed in maturity stage during composting (Figure 1a), which may be attributed to accumulated NO 3 − -N during the thermophilic phase. It has been reported that high temperature (above 40 • C) and high concentrations of free ammonia may inhibit nitrifiers [12,40]. Here, it was observed that the CaSSP application significantly decreased the total N 2 O emissions by 49.8% compared to the Control pile (Figure 1a), as compared with the 22.2-55.4% N 2 O mitigation rates in response to CaSSP or biochar addition reported previously [15,17,21,24], the amendment of CaSSP can be recognized as an effective strategy for reducing GHG from composting.
It is known that N 2 O emissions during composting are a microbe-mediated process, which is a balance between production (NO 3 − to N 2 O as catalyzed by nitrite reductase encoding by nirS/nirK) and consumption (N 2 O to N 2 as driven by nitrite reductase encoding by nosZ) [28]. In this study, CaSSP addition repressed the nirK and nirS abundance while increased nosZ copy numbers, as compared with the Control composting pile, which can explain the N 2 O mitigation based on the microbial functional genes relevant to N-cycle. It has been addressed that biochar, another popular amendment for reducing N 2 O production, also decreased nirK population in manure composting [17,24], or nirS density under field condition [41,42], and suppressed nosZ abundance [17].
The following question is which compost physicochemical properties can mediate the denitrifier community after CaSSP application. Correlation analyses showed that the abundance of nirK shared negative relationship with NH 4 + -N, moisture, TOC and C/N ratio, and positive correlations with TN as well as NO 3 − N. In contrast, the copy number of nosZ revealed positive and negative relationship with NH 4 + -N and NO 3 − -N, respectively ( Table 2). Comparison of these characteristics between Control and CaSSPadded piles suggests that NH 4 + -N and NO 3 − -N (as well as TN) are the major factors that shaped the population of nirK and nosZ. In addition, the NH 4 + -N concentrations of CaSSP treatment were higher than that of CK in the composting process, indicating that CaSSP had a considerable effect on NH 4 + -N adsorption and retention. The contents of NO 3 − -N maintained a low level from initial stage to the 31th day after composting, but increased at the end of the experiment. This result coincided with Zeng et al. [43] who reported that nitrifying bacteria multiplied with the decrease of temperature and biodegradation rate would promote the transformation of NH 4 + -N to NO 3 − -N. Additionally, a growing number of studies have illustrated that changes in physicochemical parameters, such as TN, NH 4 + -N and NO 3 − -N, drove different time course patterns of N 2 O-related functional genes [44][45][46][47], which are consistent with our observation.
Additionally, previous studies had indicated that CaSSP could affect the composting conditions such as reducing pH, which led to an adverse effect on microbial growth [48][49][50]. Results in this study illustrated that the decreased pH brought by CaSSP application did not show significant negative effects on N-cycle functional gene abundance (Table 2). Interestingly, moisture content was found to be negatively correlated with AOA amoA, nirS, and nirK abundance.
Based on physicochemical and biological variables measurements during the composting and their correlation analyses, a schematic model was developed for explaining the dynamics of N 2 O fluxes associated with bacterial functional genes and physicochemical parameters during manure composting ( Figure 5). In the schematic model, NH 4 + -N and NO 3 − -N are correlated with each other and interacting to shape the dynamics of bacterial functional gene abundance. Given this, there is a strong correlation between nirK genes and N 2 O fluxes. The present data advised particularly that the abundance of nirK genes was likely to play a key role in controlling N 2 O production.
In conclusion, results of this study suggested that calcium superphosphate (CaSSP) addition mediated NH 4 + -N/NO 3 − -N transformation in pig manure composting, consequently suppressed the abundance of nirK genes and significantly decreased N 2 O emissions. Therefore, the use of CaSSP is an effective and economical way to reduce environmental risks and improve compost quality. Additionally, using high-throughput sequencing and a culturomics strategy to determine the functional microbial groups involved in the CaSSPmediated N 2 O mitigation process, as well as investigating their response to the changing pie physicochemical factors and detailed contribution to N 2 O emissions, will be of great importance for us to deeply understand the GHG mitigation mechanisms mediated by CaSSP, and should be performed in the future.
Supplementary Materials: The following are available online at https://www.mdpi.com/1660-460 1/18/1/171/s1. Figure S1: The pictures of the experimental static chamber system for gas sampling; Figure S2: Accumulative N 2 O emission over the total composting period in Control and calcium superphosphate-added (CaSSP) piles.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to conditional confidentiality of the requirements by the National environmental funding project.