E ﬀ ects of Zeolite and Biochar Addition on Ammonia-Oxidizing Bacteria and Ammonia-Oxidizing Archaea Communities during Agricultural Waste Composting

: The e ﬀ ects of zeolite and biochar addition on ammonia-oxidizing bacteria (AOB) and archaea (AOA) communities during agricultural waste composting were determined in this study. Four treatments were conducted as follows: Treatment A as the control with no additive, Treatment B with 5% of zeolite, Treatment C with 5% of biochar, and Treatment D with 5% of zeolite and 5% biochar, respectively. The AOB and AOA amoA gene abundance as well as the ammonia monooxygenase (AMO) activity were estimated by quantitative PCR and enzyme-linked immunosorbent assay, respectively. The relationship between gene abundance and AMO enzyme activity was determined by regression analysis. Results indicated that the AOB was more abundant than that of AOA throughout the composting process. Addition of biochar and its integrated application with zeolite promoted the AOB community abundance and AMO enzyme activity. Signiﬁcant positive relationships were obtained between AMO enzyme activity and AOB community abundance ( r 2 = 0.792; P < 0.01) and AOA community abundance ( r 2 = 0.772; P < 0.01), indicating that both bacteria and archaea played signiﬁcant roles in microbial ammonia oxidation during composting. Using biochar and zeolite might promote the nitriﬁcation activity by altering the sample properties during agricultural waste composting.


Introduction
Composting has been widely recognized as an effective way to convert agricultural waste into valuable organic products [1,2]. Gaseous emissions inevitably reduced the nutrient content, which is a major challenge for the composting process [3,4]. During composting, a large quantity of nitrogen (9.6-50%) in the raw materials was released in the form of gas, mainly including ammonia (NH 3 ) and nitrous oxide (N 2 O) [5,6]. CH 4 and N 2 O, which are reported by the international panel on climate change (IPCC), are 30-210 times more likely to contribute to global warming than carbon dioxide (CO 2 ) and are responsible for climate change or ozone depletion. Therefore, it is necessary to control the emission of gases to provide an efficient, economical and environmentally friendly process for the treatment of agricultural waste.
Nitrogen in composts could be fixed and the final product's quality can be improved by a variety of physical, chemical and biological methods. Microbial inoculation and chemical/mineral material additives improved the composting efficiency, reduced nitrogen losses and mitigated greenhouse gas from rice straw in a hypoxia environment (500 • C, 3 h) by using a tubular carbonization furnace (HeFeiJingKe, GDL−1500×, China). Composting piles (50 kg wet weight) were set up and packed loosely in open boxes with good thermal insulation performance. The initial C/N ratio and moisture content were about 30:1 and 60%, respectively. Composting piles were manually turned to avoid possible anaerobic conditions. The physicochemical properties of the raw materials and the different piles are shown in Table 1.

Samples Collection and Parameters Determination
The subsamples were collected on days 0, 7, 14, 21, 28, 35 and 42. Samples for DNA extraction for discerning the functional gene abundance and activity were stored at −20 • C before being used. Samples for determination of the physical-chemical properties were stored at 4 • C before being used. Pile temperature, pH, ammonium (NH 4 + -N) and nitrate (NO 3 − -N) were determined according to our previous research [23,24]. After sampling, sterile deionized water was periodically added to maintain the moisture content of each pile in the range of 50% to 60%.

AMO Enzyme Activity Measurement
The activity of the AMO enzyme activity depends on, in part, on the content of the AMO protein. An AMO ELISA Kit was used to determine the AMO enzyme activity. The color of the stop solution changed from blue to yellow, and the color intensity was determined at 450 nm by a spectrophotometer (SpectraMax iD5). The AMO ELISA Kit has a series of calibration standards to determine the concentration of the AMO enzyme in the samples. Both the calibration and the sample standards were determined simultaneously. Meanwhile, the standard curve of the optical density versus AMO concentration was produced by the operator. Data of the AMO enzyme activity was calculated by comparing the value of the sample optical density with the standard curve [25].

Quantitative PCR
Total DNA was extracted from freeze-dried samples by using the PowerSoil kit (MoBio Laboratories, USA). The DNA extracts for each sample were combined to reduce possible variability and stored at −20 • C. Primers amoA−1f/amoA−2r [26] and CrenamoA−23f/CrenamoA−616r [26] were chosen for the AOB and AOA amoA gene abundance quantification, respectively. Quantitative PCR was conducted on an iCycler IQ5 Thermocycler (Bio-Rad, USA) with a 10 µL volume involving 0.5 µL of DNA extract, 0.2 µL of each primer, 5 µL of a real-time PCR mixture (2 × SYBR, Bioteke, Beijing) and 4.1 µL of sterile water. The quantitative PCR reaction was as follows: 95 • C for 3 min, 40 cycles of 40 s at 95 • C, 40 s at 55 • C and 40 s at 72 • C. Data were retrieved at 72 • C. The standard curves for the quantitative PCR were prepared with linearized plasmids containing cloned amoA after a 10-fold serial dilution. Melting curves were used to verify the amplification specificity of the amoA gene.

Data Analysis
Three replicates were used for the parameter, amoA gene abundance and AMO enzyme activity determination. The original data of the gene abundances were log 10 -transformed before further analysis. Least-significant difference (LSD) tests were performed to compare the mean values of the amoA gene abundance and the AMO enzyme activity in the different treatments on each sampling occasion by using SPSS (version 11.5). A regression analysis was performed between the AMO enzyme activity with log 10 -trasnformed AOB and AOA abundance to obtain the possible relationships of AOB and AOA with microbial ammonia-oxidizing activity.

Physico-Chemical Parameters
The addition of zeolite and biochar changed the pile temperature during composting, with the maximum pile temperature for Treatments A, B, C and D being 51.0, 54.4, 62.4 and 56.5 • C, respectively ( Figure 1a). The addition of biochar might accelerate the decomposition of the easy-degradable organics. Compared with the control treatment (Run A), the high temperature period was shortened by adding biochar. analysis. Least-significant difference (LSD) tests were performed to compare the mean values of the amoA gene abundance and the AMO enzyme activity in the different treatments on each sampling occasion by using SPSS (version 11.5). A regression analysis was performed between the AMO enzyme activity with log10-trasnformed AOB and AOA abundance to obtain the possible relationships of AOB and AOA with microbial ammonia-oxidizing activity.

Physico-Chemical Parameters
The addition of zeolite and biochar changed the pile temperature during composting, with the maximum pile temperature for Treatments A, B, C and D being 51.0, 54.4, 62.4 and 56.5 °C, respectively ( Figure 1a). The addition of biochar might accelerate the decomposition of the easydegradable organics. Compared with the control treatment (Run A), the high temperature period was shortened by adding biochar. The pH increased significantly during the first two weeks and then decreased gradually afterwards (Figure 1b). Significant changes in the nitrogen-relevant substances were obtained, indicating that the biochar and zeolite showed nitrogen conservation potential during composting. Because of the pH increase and mineralization of the nitrogen compounds, the NH4 + -N rapidly accumulated during the thermophilic stage, but decreased afterwards due to NH3 volatilization ( Figure 2a). The NO3 − -N contents in Treatments C and D with biochar addition were significantly higher than that of Treatments A and B during the maturation stage ( Figure 2b  The pH increased significantly during the first two weeks and then decreased gradually afterwards ( Figure 1b). Significant changes in the nitrogen-relevant substances were obtained, indicating that the biochar and zeolite showed nitrogen conservation potential during composting. Because of the pH increase and mineralization of the nitrogen compounds, the NH 4 + -N rapidly accumulated during the thermophilic stage, but decreased afterwards due to NH 3 volatilization ( Figure 2a). The NO 3 − -N contents in Treatments C and D with biochar addition were significantly higher than that of Treatments A and B during the maturation stage ( Figure 2b

AOB and AOA amoA Gene Abundance
The bacterial ( Figure 3) and archaeal amoA genes ( Figure 4) widely existed during the whole composting process. The abundance of the AOB and AOA amoA gene ranged from 4.9 × 10 6 to 1.9 × 10 8 and 1.1 × 10 6 to 3.4 × 10 7 gene copies per g −1 DW compost sample. The addition of biochar or zeolite stimulated the AOB abundance. The AOB amoA gene was 1-2 orders of magnitude higher than its archaeal community counterparts, indicating that AOB rather than AOA might play more roles in microbial ammonia oxidization during composting with biochar and zeolite addition. The abundance of the AOB amoA genes was relatively higher during the second fermentation phase than that of the first fermentation. In the whole composting process, the addition of biochar rather than zeolite had a significant effect on the AOB and AOA communities. The bacterial amoA gene abundance for the different treatments. Different letters above the error bars indicate significant differences (P < 0.05) at each sampling occasion. Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). The error bar is the mean value ± standard error.

AOB and AOA amoA Gene Abundance
The bacterial ( Figure 3) and archaeal amoA genes ( Figure 4) widely existed during the whole composting process. The abundance of the AOB and AOA amoA gene ranged from 4.9 × 10 6 to 1.9 × 10 8 and 1.1 × 10 6 to 3.4 × 10 7 gene copies per g −1 DW compost sample. The addition of biochar or zeolite stimulated the AOB abundance. The AOB amoA gene was 1-2 orders of magnitude higher than its archaeal community counterparts, indicating that AOB rather than AOA might play more roles in microbial ammonia oxidization during composting with biochar and zeolite addition. The abundance of the AOB amoA genes was relatively higher during the second fermentation phase than that of the first fermentation. In the whole composting process, the addition of biochar rather than zeolite had a significant effect on the AOB and AOA communities.

AOB and AOA amoA Gene Abundance
The bacterial ( Figure 3) and archaeal amoA genes ( Figure 4) widely existed during the whole composting process. The abundance of the AOB and AOA amoA gene ranged from 4.9 × 10 6 to 1.9 × 10 8 and 1.1 × 10 6 to 3.4 × 10 7 gene copies per g −1 DW compost sample. The addition of biochar or zeolite stimulated the AOB abundance. The AOB amoA gene was 1-2 orders of magnitude higher than its archaeal community counterparts, indicating that AOB rather than AOA might play more roles in microbial ammonia oxidization during composting with biochar and zeolite addition. The abundance of the AOB amoA genes was relatively higher during the second fermentation phase than that of the first fermentation. In the whole composting process, the addition of biochar rather than zeolite had a significant effect on the AOB and AOA communities. compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). The error bar is the mean value ± standard error.  compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). The error bar is the mean value ± standard error.

AMO Enzyme Activity
The activities of the AMO enzyme in all treatments are presented in Figure 5. Treatment A without any addition had the lowest AMO enzyme activity. The AMO enzyme activity was significantly increased in Treatments C and D with biochar addition. The AMO activities in Treatments B and D were increased on Days 28 and 42. These results indicated that biochar rather than zeolite amendment promoted the microbial conversion of NH4 + -N and increased the activity of the AMO genes. Figure 5. Changes in ammonia monooxygenase (AMO) enzyme activity for the different treatments. Different letters above the error bars indicate significant differences (P < 0.05) at each sampling occasion. Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). AMO: ammonium monooxygenase activity. The error bar is the mean value ± standard error. compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). The error bar is the mean value ± standard error.

AMO Enzyme Activity
The activities of the AMO enzyme in all treatments are presented in Figure 5. Treatment A without any addition had the lowest AMO enzyme activity. The AMO enzyme activity was significantly increased in Treatments C and D with biochar addition. The AMO activities in Treatments B and D were increased on Days 28 and 42. These results indicated that biochar rather than zeolite amendment promoted the microbial conversion of NH 4 + -N and increased the activity of the AMO genes.

AMO Enzyme Activity
The activities of the AMO enzyme in all treatments are presented in Figure 5. Treatment A without any addition had the lowest AMO enzyme activity. The AMO enzyme activity was significantly increased in Treatments C and D with biochar addition. The AMO activities in Treatments B and D were increased on Days 28 and 42. These results indicated that biochar rather than zeolite amendment promoted the microbial conversion of NH4 + -N and increased the activity of the AMO genes. Different letters above the error bars indicate significant differences (P < 0.05) at each sampling occasion. Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). AMO: ammonium monooxygenase activity. The error bar is the mean value ± standard error. Different letters above the error bars indicate significant differences (P < 0.05) at each sampling occasion. Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%). AMO: ammonium monooxygenase activity. The error bar is the mean value ± standard error.

Relationship Between AMO Enzyme Activity and amoA Gene Abundance
Significant positive relationships were obtained between the AMO enzyme activity and bacterial amoA gene abundance ( Figure 6) as well as archaeal amoA gene abundance (Figure 7). The AMO enzyme activity increased as the bacterial and archaeal amoA community exponentially increased. The bacterial and archaeal amoA gene-coding communities can account for 79.2% (P < 0.01) and 77.2% (P < 0.01) of the variation in AMO enzyme activity. These results suggested that the AOB and AOA communities were both closely related to the AMO enzyme activity in the composting substrate.

Relationship Between AMO Enzyme Activity and amoA Gene Abundance
Significant positive relationships were obtained between the AMO enzyme activity and bacterial amoA gene abundance ( Figure 6) as well as archaeal amoA gene abundance (Figure 7). The AMO enzyme activity increased as the bacterial and archaeal amoA community exponentially increased. The bacterial and archaeal amoA gene-coding communities can account for 79.2% (P < 0.01) and 77.2% (P < 0.01) of the variation in AMO enzyme activity. These results suggested that the AOB and AOA communities were both closely related to the AMO enzyme activity in the composting substrate. Figure 6. Relationships between ammonia monooxygenase (AMO) enzyme activity and bacterial amoA gene abundance during agricultural waste composting process. Line indicates the fitted curve, where y is the AMO enzyme activity (mmol g −1 d −1 DW sample) and x is the Log10 transformed AOB amoA gene abundance (gene copies g −1 DW sample) (n = 28). Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%).

Discussion
Pile temperatures exceeding 50 °C lasted for more than 5 days, which effectively killed the pathogenic microorganisms and ensured the compost is harmless for all treatments [27]. The pile Figure 6. Relationships between ammonia monooxygenase (AMO) enzyme activity and bacterial amoA gene abundance during agricultural waste composting process. Line indicates the fitted curve, where y is the AMO enzyme activity (mmol g −1 d −1 DW sample) and x is the Log 10 transformed AOB amoA gene abundance (gene copies g −1 DW sample) (n = 28). Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%).

Relationship Between AMO Enzyme Activity and amoA Gene Abundance
Significant positive relationships were obtained between the AMO enzyme activity and bacterial amoA gene abundance ( Figure 6) as well as archaeal amoA gene abundance (Figure 7). The AMO enzyme activity increased as the bacterial and archaeal amoA community exponentially increased. The bacterial and archaeal amoA gene-coding communities can account for 79.2% (P < 0.01) and 77.2% (P < 0.01) of the variation in AMO enzyme activity. These results suggested that the AOB and AOA communities were both closely related to the AMO enzyme activity in the composting substrate. Figure 6. Relationships between ammonia monooxygenase (AMO) enzyme activity and bacterial amoA gene abundance during agricultural waste composting process. Line indicates the fitted curve, where y is the AMO enzyme activity (mmol g −1 d −1 DW sample) and x is the Log10 transformed AOB amoA gene abundance (gene copies g −1 DW sample) (n = 28). Run A: control; Run B: compost + zeolite (5%); Run C: compost + biochar (5%); Run D: compost + zeolite (5%) + biochar (5%).

Discussion
Pile temperatures exceeding 50 • C lasted for more than 5 days, which effectively killed the pathogenic microorganisms and ensured the compost is harmless for all treatments [27]. The pile temperature decreased gradually afterwards with the depletion in organic matter [19,28]. The bulking effect of the zeolite might enhance the heat radiation, thus lowering the pile temperature of the cooling stage and the maturation stage [29,30]. Adding biochar significantly increased the pH, which could be caused by the alkalinity characteristics of the biochar [31]. The combined addition of biochar and zeolite promoted nitrogen retention and nutrient transformation during composting [18]. Zeolite has selective adsorption and sieving properties, especially the high polarity molecules such as H 2 O, NH 3 , CO 2 , etc. [32]. Zeolite can reduce the NH 3 volatilization loss up to 44% [33]. Biochar is widely used as an additive to improve composting conditions and improve the quality of compost products because of its stable porous structure and good adsorption properties [34]. Biochar addition during composting significantly increased the NO 3 − -N content and pile temperature, and decreased the pH and NH 4 + -N [35]. Biochar and zeolite mixing decreased the water-filled pore space, causing better aeration, and might be a potential reason explaining the greenhouse gas emission reduction. Soil pH is one of the main factors of AOB community structure, including the direct effects on AOB and indirect effects on soil activity [36]. pH directly determined the presence of ammonia in the soil form: when the pH is lowered, the ammonia (NH 3 ) will be converted to ammonium (NH 4 + ), which reduces the amount of substrate NH 3 , and affects the activity, abundance and even species of AOB [37]. In most acidic soils, the number of AOA is higher than that of AOB, suggesting that AOA has stronger adaptability to low-pH habitats. Indeed, [38] also found a similar conclusion in Chinese tea-garden soils. In acidic tea-garden soil, the AOA/AOB ratio increased with the decrease in soil pH value, and the number of AOA showed a good positive correlation with soil ammonia-oxidizing activity. Amendment of biochar and zeolite resulted in increased input of carbon and would stimulate microbial community activity during composting. Biochar amendment induced high respiration rates and fast organic matter decomposition, indicating higher microbial activity [39]. Organic and inorganic compounds were filled into biochar pores after composting [40]. Biochar can serve as niches for microbial community cultivation, as the biochar will provide a nitrogen source and decrease the free NH 3 toxicity towards the microbial species [41]. Moreover, improved microbial activities caused higher nutrient consumption and lignocellulose biodegradation, thereby reducing the availability of the carbon and nitrogen substances. AOB and AOA both contain AMO genes that catalyze the first step of ammonium oxidation, responsible for converting ammonia to hydroxylamine. Some studies showed that the AOA amoA gene abundance was significantly higher than that of the AOB [26,42]. However, another study also indicated that AOB rather than its archaeal counterparts were distributed widely during composting of manure from field-scale facilities [43].
Enzyme activities reflect the biochemical reaction extent in environmental samples (e.g., composts, soils and waters), and also serve as a potential biological indicator for nutrition condition [44]. The AMO enzyme is encoded by the amoA, amoB and amoC genes [45]. Its activity influenced the rate-limiting step of the reaction process of NH 4 + -N to NH 2 OH [46,47]. Stimulation or inhibition of AMO was the combination results of multiple factors and some NO 3 − -N production did not require AMO participation [48]. Different responses of AOB, AOA and AMO activity can be used as important indicators of the potential changes in the physico-chemical/biological condition of the composting process. Previous research indicated no inhibitory effect was observed on the activities of a microbial community after clinoptilolite application [49]. Biochar and zeolite amendment will also affect the availability of toxins that determine the growth and activity of the microbial community. Zeolite can absorb the NH 4 + cations in water [16], soils [50] and composting piles [51]. At the thermophilic stage, addition of biochar consequently alleviated the initial low pH [35]. As an indirect substrate of AOB, the concentration of NH 4 + -N is directly related to the quantity and species of AOB: the increase in ammonium concentration can lead to the increase in AOB quantity. A series of microcosmic experiments and ecological studies also confirmed that AOA is more suitable for growth in low-substrate environments [52]. The growth of two AOB and AOA species were significantly different under different substrate concentrations [53]. AOB grows well under high concentration of ammonia nitrogen. AOA can grow under high, medium and low matrix concentrations, whose growth is actually inhibited under high matrix concentrations. In grassland soil with a high substrate concentration, the increase in ammonia-oxidizing activity was accompanied by the increase in the number of AOB, indicating that AOB were the main drivers of the soil ammonia-oxidizing reaction at high substrate concentrations [54]. It was also reported that a high NH 4 + -N content inhibited the CO 2 assimilation of thermophilic AOA, indicating that high levels of NH 4 + -N could affect the AOA community abundance and structure [55].
Which community was relatively more important seemed to be site-specific, depending on the raw material condition and different control strategies during agricultural waste composting. The AOB in this study was more sensitive than the AOA to changes in samples with biochar and zeolite addition. The different responses of the AOA and AOB due to the various changes in the physical and chemical composting conditions can be used as an important indicator of potential environmental condition change after amendment with biochar and zeolite. A previous study showed that the AOA community was relatively more abundant than that of the AOB during tropical agricultural waste composting [56]. In agricultural soil, nitrogen-rich grassland ecosystems and mangrove sediments, AOB has an obvious influence [57]. Higher organic matter and O 2 content during composting provided unfavorable conditions for the AOA communities.

Conclusions
The effects of zeolite and biochar addition on the AOB and AOA amoA gene abundance and AMO enzyme activity were determined during agricultural waste composting. Results showed that the addition of zeolite and biochar changed the abundance of the AOB and AOA communities. A higher community abundance was found in samples with zeolite and biochar addition. Both the bacterial and archaeal amoA communities were found to be significantly and positively related to AMO enzyme activity during composting. The combined additives of zeolite and biochar stimulated the AOA and AOB communities and promoted nitrogen preservation during agricultural waste composting.