Amending biochar affected enzyme activities and nitrogen turnover in Phaeozem and Luvisol

Soil nitrogen (N) is a vital source of nutrients for maintaining soil fertility and crop production. However, the effect of biochar application rate on the mechanism of organic N transformation and the contribution of enzyme mineralization is still unclear. Therefore, we conducted two 5‐year field experiments in contrasting soils (Phaeozem and Luvisol) with biochar application rate at 0 t hm−2 (CK, 0), 22.5 t hm−2 (D1, 1%), 67.5 t hm−2 (D2, 3%), and 112.5 t hm−2 (D3, 5%) to investigate the potential effects of biochar application rate on soil organic nitrogen (N) turnover and its linkage to enzymatic mineralization in contrasting soil. The results showed that soil organic carbon (SOC) and microbial biomass nitrogen (MBN) contents, microbial biomass carbon to nitrogen ratio (MBC:MBN) and protease activity are significantly influenced by biochar application rate whereas not by soil type. Ammonium nitrogen (NH4+‐N) and nitrate nitrogen (NO3−‐N) contents, and dehydrogenase activity are significantly changed by soil type whereas not by biochar application rate. Based on the redundancy analysis, we found that organic N fractions are associated with MBN, SOC, and protease in Phaeozem, but related to protease activity in Luvisol. Our findings indicate that organic N turnover is not only related to the bioavailability of N but also requires carbon substrates in Phaeozem, whereas the transformation of organic N in Luvisol is dominated by enzymatic mineralization as the relatively low level of bioavailable N.

. Moreover, the low molecular weight organic N can be directly taken up (Tian et al., 2017), whereas the high molecular weight organic N generally requires further mineralized by extracellular enzymes to be uptake in the form of mineral N (Nannipieri & Eldor, 2009). According to the method of Stevenson (1982), soil organic N is classified into hydrolyzable ammonium N (AM), amino acid N (AA), amino sugar N (AS), acid-insoluble N (AIN), and hydrolyzable unknown N (HUN), and its composition can affect the utilization of mineralized N (Prieto-Fernández & Carballas, 2000). Among them, the active-soil organic N such as AM, AA, and AS (Li et al., 2014), which is easily to be utilized by microorganisms and plants than stable-soil organic N (i.e., AIN and HUN) (Zhu et al., 2019). Therefore, organic N pools in agricultural production are significant for crop growth.
Soil organic N depolymerization and mineralization are mediated by microbial enzymatic processes (Mengel, 1996). Of these, the most important processes involved in protease catalyzing hydrolysis proteins, amidase releasing ammonia from amides, and N-acetyl-βd-glucosaminidase (NAG) degrading chitin (Kandeler et al., 2011;Mengel, 1996). Moreover, organic N mineralization requires the involvement of these enzymes, such as the depolymerization of protein by extracellular protease as a rate-limiting step of organic N mineralization (Schimel & Bennett, 2004;Kandeler et al., 2011). Wallenstein and Weintraub (2008) also indicated that the depolymerization processes in soils can be influenced either by extracellular enzyme activities (enzyme limitation) or by substrate availability (substrate limitation). Conversely, soil substrate availability also affects the catalytic process of soil extracellular enzymes.
Biochar is a recalcitrant carbonaceous product that can be derived from crop residues and other biomass pyrolysis under limited oxygen or hypoxia condition, with carbon-rich and well-developed pores, which can improve soil quality . Extensive attention has been attracted to the potential benefits of biochar as a soil amendment in modulating soil N cycle and providing a new way to improve soil fertility Sohi et al., 2010). Biochar can reduce nutrients leaching, improve the availability of soil N, and increase SOC content, but how much improvement will depend on soil characteristic (DeLuca et al., 2009;Van Zwieten et al., 2010). In addition, Kuppusamy et al. (2016) also reported that biochar can improve the retention capacity of soil fertilizer, and promote sustainable supply of soil nutrients. Soil N transformations are enzyme-mediated process, and many of which have been influenced by biochar application rate (Thies et al., 2015). Several studies have reported the effects of biochar application on the relationship between soil enzymes and N mineralization, such as NAG, protease, and amidase (Bailey et al., 2011;Oleszczuk et al., 2014). Moreover, biochar has property of adsorbing organic molecules on the surface, including adsorbing enzyme substrates to promote enzymatic reactions and masking the binding sites of enzymatic reactions to inhibit enzymatic reactions, respectively (Bailey et al., 2011;Elzobair et al., 2016). Although the average residual time of biochar applied to soil exceeds 19 years, biochar surface will be rapidly oxidized and aging phenomenon over time due to its properties would be influenced by biotic and abiotic factors, such as surface functional groups, ion exchange capacity, surface area, pore volume, and surface properties (Cheng et al., 2006;Hamer et al., 2004;Lehmann et al., 2003;Noyce et al., 2015).
In general, biochar application could modify soil N dynamic by improving N bioavailability for crops. However, the effects of biochar on soil organic N turnover are highly variable, and it remains uncertain whether this variability varied with the characteristics of different soil types. Therefore, the objectives of this study were (i) to assess the effect of biochar application rate on organic N fractions and N-mineralization enzyme activities in contrasting soils, and (ii) to elucidate the potential relationship between organic N fractions and enzyme activities in Phaeozem and Luvisol. We hypothesized that (1) soil C content and N-related enzyme activities would increase with the increase in biochar application rate; and (2) there were different mechanisms of N transformation in Phaeozem and Luvisol under biochar application, which might be influenced by the availability of soil C and N.

| Site description and experimental design
Two microplot experiments were performed with singleharvest spring maize (Zea mays L.) as a test crop at experimental sites of the Key Lab of Conservation Tillage and Ecological Agriculture, Liaoning Province and National Field Observation and Research Station of Shenyang Agroecosystems. They were located in Lower Liaohe River and Songnen Plains, respectively, which are major grainproducing areas of Northeast China. Soils are classified as Phaeozem and Luvisol (FAO, 1998), respectively. Significant differences in the climatic condition, soil development, organic matter, and nutrient contents were observed between two soils (Wu et al., 2021;Zhang et al., 2022). The experiments were rain-fed and conventional managed to ensure optimal nutrient conditions and avoid stresses from weeds, insects, and diseases. Detail experimental designs were shown by Wang et al. (2017) and Zhang et al. (2022), respectively. Briefly, two sets of 12 microplots (1.7 m × 1.5 m) were conducted using a randomly complete block design with three replicates in Phaeozem and Luvisol. The experiments started in April 2013, and the four biochar application rates (0, 22.5, 67.5, and 112.5 t hm −2 ) were mixed thoroughly the 0~20 cm soil, respectively. Biochar application rate is equivalent to 0, 1%, 3%, and 5% of soil weight in 0-20 cm plough layer, abbreviated as CK, D 1 , D 2 , and D 3 , respectively. The D 1 biochar application rate is equivalent to its production from maize straw in the area. It was inconvenient to apply the power biochar in agriculture so we explored to reduce the frequency of biochar application in the experiments. D 2 and D 3 rates are equivalent to three and five times of D 1 rate, which simulate once biochar application every 3 and 5 years, respectively. Fertilization was according to the common regional practice with 180 kg hm −2 of N, 75 kg hm −2 of P (P 2 O 5 ), and 75 kg hm −2 of K (K 2 O) in each microplot. The raw materials for biochar preparation in this study were maize straw, pyrolyzed at 300-500 C for 3 h (THL-IV). The basic properties of soil and biochar are shown in Table 1.

| Soil sampling
Sampling was collected at the end of the maize growing season in October 2018. Specifically, we used a fivepoint sampling method with an auger (3 cm in diameter) to collect at 0-20 cm depth from each plot. Soil samples were immediately transported to the laboratory, which were homogenized and passed through a 2-mm sieve for each plot after removing visible stones and plant residues. These already prepared soil samples were divided into two subsamples, one of them was stored at 4°C to analyze biological and biochemical characteristics, while another sample was air-dried for analyzing chemical characteristics and nitrogen fractionation.

| Biochar properties analysis
Biochar total carbon (TC) and total nitrogen (TN) were measured with an automatic elemental analyzer (Analyzer vario MICRO cube; Elementar). The ammonium nitrogen (NH 4 + -N) and nitrate nitrogen (NO 3 − -N) in biochar were extracted with deionized water for 16 h, filtered through a 0.45-μm membrane after centrifuged at 8000 r min −1 , and then the filtrate was decolorized with sulfuric acid, and the NH 4 + -N and NO 3 − -N were determined by distillation. Soil and biochar pH is measured by a pH meter (pH 700 Bench Meter; Eutech Instruments) at a ratio of 1:2.5 (w/v).

| Soil properties analysis
Soil organic carbon (SOC) was determined by the chemical oxidation method with K 2 Cr 2 O 7 -H 2 SO 4 , and then corrected for oxidation efficiency with a corrected constant of 1.08 (Nelson & Sommers, 1983). Soil TC and TN were measured with an automatic elemental analyzer (Analyzer vario MICRO cube; Elementar). Soil NH 4 + -N and NO 3 − -N were extracted with 2 M KCl solution and analyzed using an AutoAnalyser III continuous Flow Analyzer (Bran+Luebbe; AA3 AutoAnalyzer). The chloroform fumigation extraction method was used to determine microbial biomass carbon and nitrogen (MBC and MBN). Briefly, 10.00 g of fresh subsample was fumigated with ethanol-free chloroform at room temperate for 24 h and another 10.00 g of fresh subsample was unfumigated. Then the fumigated and unfumigated samples were extracted with 40 mL 0.5 M K 2 SO 4 solution, and the extracted liquid was determined using a TOC analyzer (Multi C/N 3000; Analytik Jena) (Joergensen, 1996). 2.3.3 | Soil organic N fractionation measurement Soil organic N (SON) fractions were measured according to the acid-hydrolysis method, which was used to separate SON forms (e.g., hydrolyzed ammonium N, AA, AS, AIN, and hydrolyzable unknown N) as described by Stevenson (1982). A brief description is as follows: soil subsample containing about 1.00 g soil N was hydrolyzed in an autoclave for 6 h at 15 lb./in 2 by mixing 20 mL 6 M HCl and a small amount of octanol. The different SON fractions were measured by steam distillation with different additives after hydrolysis. Specifically, total hydrolyzable N was determined by steam distillation with   Tian et al. (2017) and Wu et al. (2021). Moreover, the proportion of soil total N was used to calculate the N distributions (%) in various fractions.

| Soil enzyme assays
The N-related enzyme activities were described as follows: protease activity was assayed according to the method described by Ladd and Butler (1972), soil amidase activity was measured by the method of Frankenberger and Tabatabai (1981), soil NAG activity was determined by the method of Parham and Deng (2000), and the determination of dehydrogenase activity (DHA) according to the reduction of 2,3,5-triphenyl tetrazolium chloride as described by Tabatabai (1994).

| Statistical analysis
All values are based on the weight of oven-dried soil (105°C), and the soil data are the mean values of three replicates with their standard errors. An analysis of variance (ANOVA) was performed using R software. If interactions between biochar application rate and soil type were significant, differences among application rates and between two soil types were further compared with a one-way ANOVA (S-N-K method) and t-test, respectively. Main effect comparisons were conducted for those nonsignificant interaction effect indicators.
Pearson correlation was used to analyze the relationship between soil organic nitrogen fractions and enzyme activities. Redundancy analysis (RDA) was performed using R software to investigate the relationships between soil organic N fractions and soil chemical and biological properties. Soil organic N fractions were normalized by log transformation The statistical significance of factors of variance was tested using a Monte Carlo permutation test. All data analysis was performed using R software (v. 4.2.1).

| RESULTS
The results of two-way ANOVA (Figures 1 and 2; Tables 2 and 3) showed that both biochar application rate and soil type could influence soil chemical properties and biological characteristics. The interaction between biochar application rate and soil type was significantly affected by TC content and DHA, NAG, and protease activity. TC and TN contents, C:N ratio, and NAG and amidase activities were significantly changed by biochar application rate and soil type. SOC and MBN contents, MBC:MBN ratio, and protease activity were significantly influenced by biochar application rate whereas not by soil type. NH 4 + -N and NO 3 − -N contents, and DHA activity were significantly changed by soil type whereas not by biochar application rate.

| Effects of biochar application on soil chemical properties
Phaeozem showed higher TN and NH 4 + -N contents, whereas showed lower NO 3 − -N content and C:N ratio compared to Luvisol (Table 2). Biochar application rate caused significant changes in SOC, TN, and TC contents, and C:N ratio, and no significant change in NH 4 + -N and NO 3 − -N content, respectively (Table 2; Figure 1). D 2 and D 3 treatments significantly increased TC content in Phaeozem and Luvisol, and TC content at D 3 level was significantly higher in Phaeozem than that in Luvisol ( Figure 1). Moreover, SOC and TN contents were significantly increased at D 2 and D 3 levels, while C:N ratio was significantly increased at each biochar level (Table 2). F I G U R E 1 Effects of different treatments on soil total carbon. * and ** above the columns indicate significant differences between two soil types in the biochar treatment. CK, D 1 , D 2 , and D 3 represent the biochar application intensity equivalent to 0, 1%, 3%, and 5% of soil weight in 0-20 cm plough layer, respectively. Values are means and error bars represent standard errors (n = 3). Different uppercase and lowercase letters indicate significant differences (p < 0.05) among biochar application rates in Phaeozem and Luvisol, respectively.

| Effect of biochar application on soil microbial properties
Generally, microbial biomass, N cycling, and metabolic activity of living microorganisms in soil are usually characterized by MBC and MBN contents, and DHA activity, respectively. The effect of biochar application on MBC content was not significant, and the effect of soil type on MBC and MBN contents, and MBC:MBN ratio was not significant. MBN content significantly decreased with increasing biochar application rate, and only significantly at D 3 level. MBC:MBN ratio significantly increased at D 3 level (Table 3). DHA activity showed an increasing trend with increasing biochar application rate in Luvisol, and significantly increased at D 3 level, whereas a decreasing trend with increasing biochar application rate in Phaeozem. Moreover, DHA activity was significantly higher in Phaeozem than those in Luvisol under CK and D 1 treatments (Figure 2c).

| Effects of biochar application on soil N-hydrolyzing enzyme activities
Soil organic N-mineralization enzymes mainly include NAG, protease, and amidase. Compared with biochar application rate, NAG activity was significantly decreased at all treatments (D 1 , D 2 , and D 3 ) in Phaeozem, but showed a decreasing trend at all treatment and a F I G U R E 2 Effects of different treatments on soil N-acetyl-βd-glucosaminidase, protease and dehydrogenase activity. * Significant correlation at 0.05 level, ** significant correlation at 0.01 level. Different uppercase and lowercase letters indicate significant differences (p < 0.05) among biochar application rates in Phaeozem and Luvisol, respectively. NS, nonsignificant.  T A B L E 2 Main effect comparisons of the chemical properties among the biochar application intensity and between the soil types. significantly only at D 1 level in Luvisol (Figure 2a). Protease activity has increasing trends at all treatments in Phaeozem with a significantly increased in D 1 treatment, whereas protease activity was significantly increased at D 3 treatment in Luvisol (Figure 2b). Amidase activities were significantly increased in D 2 and D 3 treatments (Table 3). A few significant differences of N-mineralization enzyme activities were observed between two soil types. For instance, amidase activity in Phaeozem was significantly higher than those in Luvisol, NAG activity in Phaeozem was significantly higher than that in Luvisol at CK, and protease activity in Luvisol was significantly higher than that in Phaeozem at D 3 level (Table 3; Figure 2a).

| Effects of biochar application on soil organic N fractions
The content and distribution of organic N fractions under biochar application conditions are shown in Table 4 and Figure 3. Soil organic N fractions showed different proportions in different soil types after 5 years of biochar application. In Phaeozem, AM, AA, AS, and HUN contents did not significantly change, but AS content had an increasing trend with biochar application compared to CK. The content and proportion of AIN in Phaeozem tended to increase with biochar application, while the D 3 treatment showed a significantly increase. There were no significant differences in organic N fraction contents except for AIN,

| Relationship between soil organic N fractions and chemical and biological properties under biochar application
Correlation analysis showed that soil organic N fractions were affected by soil enzymes when biochar applied to different soils. Soil amidase activity was significantly positively correlated with AM and AIN (Figure 5a, p < 0.05) in Phaeozem.
There was a significantly negative correlation between AA and soil protease activity in Luvisol (Figure 5b, p < 0.05).
Redundancy analysis results are shown in Figure 6, the first two axes explained 45.01% and 36.59% of total environmental variation in Phaeozem and Luvisol, respectively (Figure 6a,b). Soil MBN, SOC, and protease explained significant proportion of the variation in organic N fractions and explained 34.29%, 10.72%, and 3.27% of overall variations in organic N fractions in Phaeozem, while protease was important factor that influenced organic N fractions and explained 20.25% of overall variations in organic nitrogen fractions in Luvisol (Table 6, p < 0.05).

| Effects of biochar application on the fractions and availability of soil N
In our study, organic N fractions showed a different distribution after 5 years of biochar application, which can also influence crop nutrient requirements by regulating the availability of soil N. NH 4 + -N can be directly absorbed and utilized by crops, and biochar can hold NH 4 + -N and NO 3 − -N, thereby reducing N leaching. Moreover, the pore structure of biochar could increase soil permeability and promote nitrification to transform NH 4 et al., 2006). Biochar can also affect the mineralization process of soil organic N with its unique properties (Sahrawat, 2010), which promotes organic N mineralization and increases NO 3 − -N content (Yoo & Kang, 2012). Zheng et al. (2013) reported that biochar application can increase TN content, Liang et al. (2014) also reported that biochar application can significantly increase TN content, which is related to biochar application rate. Comparing the contrasting soils, TN and NH 4 + -N contents were significantly higher, and NO 3 − -N content was significantly lower in Phaeozem than that in Luvisol, respectively F I G U R E 3 Distribution (%) of organic nitrogen fraction under the application of biochar. Different lowercase and uppercase letters indicate significant differences (p < 0.05) of each organic nitrogen fraction distribution among biochar application rates in Phaeozem and Luvisol, respectively. AA, amino acid N; AIN, acid insoluble N; AM, hydrolyzable ammonium N; AS, amino sugar N; HUN, hydrolyzable unknown N.
(Table 2), indicating that biochar could promote N accumulation and improve N availability and reduce N leaching with its inherent nutrients and structural characteristics in Phaeozem compared with Luvisol. However, the TN content in Phaeozem was reduced compared to that before the biochar application (Tables 1 and 2). This may be explained by the fact that biochar significantly increased the crop yield and aboveground biomass (Farhangi-Abriz et al., 2021), although biochar application also promotes the accumulation of soil N (Table 2), which was not sufficient to compensate for the loss of removed N.
To better evaluate the bioavailability of soil N, we divided soil organic N (SON) into soil active-SON (AM, AA, and AS) and soil stable-SON (HUN and AIN) (Wu et al., 2021). These are ascribed to AM, AA, and AS are more available for microorganisms and crops compared with HUN and AIN (Qiu et al., 2012). Based on this division, we found that D 3 treatment showed a significantly higher proportion of stable-SON and lower active-SON compared to D 1 treatment in Luvisol ( Figure 4). As shown in Table 5, our results also found that biochar application insignificantly increased active-and stable-SON contents in Phaeozem, whereas D 2 and D 3 treatments increased stable-SON contents and showed a significant increase compared to D 1 treatment.
AM serves as a reservoir of rapidly released and available pool of N for growth and metabolism of crops and microorganisms, which is derived from adsorption and immobilization of ammonium, deamination of amino acids, and amino sugars, and hydrolysis of amide compounds (Lü et al., 2013;Qiu et al., 2012;Stevenson, 1982). In our study, we found that AM content was insignificantly increased in biochar treatment in Phaeozem, while insignificantly increased at D 2 level in Luvisol (Table 4). The insignificant increases in biochar application on soil AM are probably due to the fact that biochar has the property of absorbing AM. Another possible reason is that the welldeveloped pores of biochar provide a living place for microorganisms, which indirectly increases transformation of organic N fractions  biochar application impeded the release of fixed NH 4 + , thereby AM content has not significantly changed. AA is a kind of easily oxidized organic N compound, and its content is related to residues of microorganisms, animals, and plants and the composition of secretions, which plays an important role in storage capacity for immobilized N (Bremner, 1965;Lu et al., 2018;Lü et al., 2013). Stevenson (1982) found that AA is similar to the hydrolyzates of microbial structural protein, and Schulten and Schnitzer (1997) also reported that the synthesis and dynamics of AA in the soil-plant system are closely linked to microbial metabolism. Our results found that the insignificant changes of biochar application on AA content partly due to the increase in aboveground biomass with biochar application, which might leads to a decrease in AA by plant uptake that is not related to inorganic N concentrations (Jones & Darrah, 1994;Wu et al., 2021). It has been shown that the variation of AA content is closely related to MBC, which is attribute to a majority of AA in soils seems to be present in soil microbes (Stevenson, 1982;Tian et al., 2017). Our result found that MBC was not significantly changed in biochar application, which is consistent with Castaldi et al. (2011). Moreover, the adsorption of microbial biomass in pores of biochar during chloroform fumigation (Liang et al., 2010) can also influence MBC content.
Amino sugar N, as most of them are macromolecular compounds, is mainly synthesized by soil microorganisms, which can closely bind soil colloid (Bremner, 1967). AS content in different soil was not significantly changed by biochar application, but the proportion of AS was increased in Phaeozem. Some studies indicated that AS is mostly released by fungal cell walls, and a few released by bacterial cell walls or other tissues (Nannipieri & Eldor, 2009;Olk, 2008). Furthermore, the MBC:MBN ratio, as an indicator of soil microbial structure, is often used to describe the relative contributions of bacterial and fungal cell populations to the soil microbial biomass (Tian et al., 2017). Compared to the CK treatment, biochar F I G U R E 5 Correlations between soil organic N and soil enzyme activities in Phaeozem (a) and Luvisol (b). * Significant correlation at 0.05 level (both sides), ** significant correlation at 0.01 level (both sides). AA, amino acid N; AIN, acid insoluble N; AM, hydrolyzable ammonium N; AS, amino sugar N; HUN, hydrolyzable unknown N; NAG, N-acetyl-βd-glucosaminidase.

F I G U R E 6
Redundancy analysis (RDA) between soil organic nitrogen fraction and soil environmental parameters under the addition of biochar. (a, b) For Phaeozem and Luvisol, respectively. AA, amino acid N; AIN, acid insoluble N; AM, hydrolyzable ammonium N; AS, amino sugar N; HUN, hydrolyzable unknown N; MBN, microbial biomass nitrogen; NAG, N-acetyl-βd-glucosaminidase; SOC, soil organic C; TN, soil total nitrogen. application rate increased the MBC:MBN ratio and significantly increased at D 3 level (Table 3), indicating that biochar application increases the relative abundance of fungi. There were opposite trends of AS and HUN between two soils in our study (Table 4). One possible reason is that biochar application stimulates the fungal population, which causes the fungi to predominant in Phaeozem compared to Luvisol. Another possible reason is related to soil MBN. For example, the lower MBN content in Phaeozem than that in Luvisol under biochar application, which required enhance transformation of stable-SON into active-SON, thereby improving the availability of soil N for microbial growth and metabolism.
AIN mainly exists in the form of heterocyclic or aromatic ring compounds, which is considered to be a structural component of humic substances, the most important sources of them come from the senescent materials of above-and belowground detritus (Horwath, 2007;Nannipieri & Eldor, 2009;Tian et al., 2017). Previous study reported that inorganic N and soluble organic N existence in biochar can be directly absorbed by crops, while insoluble organic N can increase soil TN content (Liu et al., 2019). AIN content is the largest among all components, and not easy to be mineralized (Rovira & Vallejo, 2002). Our result found that TN content was significantly increased under D 2 and D 3 treatments (Table 2), which might be not only related to the pore structure of biochar improving soil aeration and inhibiting N loss by denitrification , but also to the AIN content. In our study, the content and proportion of AIN were increased at all levels and significantly increased at D 3 level in Phaeozem compared to CK (Table 4; Figure 3). The reason is biochar contains a certain amount of non-hydrolyzable N, which is structurally stable and cannot be decomposed by soil microorganisms and enzyme for crop utilization in short term (Liu et al., 2019). However, AIN content was significantly reduced at D 2 treatment in Luvisol (Table 4). This may be ascribed to the fact that the lower availability N is not enough to sustain biological requirements in Luvisol, so a small amount of biochar application can be degraded and utilized stable N by enzymatic process.
Hydrolyzable unknown N refers to the unidentified Ncontaining substance whose morphology and properties need to be further explored (Stevenson, 1982). According to some studies, this component consists mainly of nonα-AA, n-phenoxy AA, and heterocyclic N such as pyrimidine and purine (Kelley & Stevenson, 1996). HUN content had an increasing trend with biochar application in Luvisol, whereas tended to decrease in Phaeozem. This result indicated that HUN can also participate in mineralization process of soil organic N. In addition, biochar application has different effects on HUN in different soil, which may be related to application rate and soil type.

| Effects of biochar application on soil N-hydrolyzing enzyme activities
Soil enzymes can reflect soil quality and participate in biochemical cycles, and their activities are influenced by soil type and biochar application rate. Protease, amidase, and NAG in soil are mainly secreted by soil microorganisms, and they are involved in the mineralization of organic N (Kandeler et al., 2011;Landi et al., 2011). Wu et al. (2021) reported that N-hydrolyzing enzyme activities are linked with microbial biomass, and they showed a positive correlation.
As we known, NAG was mainly expressed by a diverse group of fungi. The MBC:MBN ratio, as an indicator of soil microbial community structure, is generally applied to discribe the relative contribution of fungal cell population to microbial biomass (Tian et al., 2017). The higher MBC:MBN ratio may indicates a higher proportion of fungi in microbial biomass. The increased NAG activity under biochar application may be explained by stimulating the relative abundance of fungi (Table 3). However, we found that NAG activity was significantly reduced at D 1 , D 2 and D 3 levels in Phaeozem, whereas NAG activity T A B L E 6 Statistic results of redundancy analysis between soil organic nitrogen fractions and explanatory environmental parameters in different soils. showed a decreasing trend and significantly decreased at D 1 level in Luvisol (Figure 2a). These results might be explained by the N supply of soil itself is sufficient, although biochar application increased the relative abundance of fungi, it does not need enzyme to promote catalytic hydrolysis of N. Moreover, the large specific surface area and surface CEC of biochar increase the fixation of NH 4 + -N and adsorption of free NO 3 − -N, and inhibit denitrification of soil (Cayuela et al., 2010), thereby reducing the soil N leaching. Therefore, the effect of biochar on NAG activity is not only related to the relative abundance of fungi by biochar but also to the availability of contrasting soil N.

Nitrogen fractions in the Phaeozem Nitrogen fractions in the Luvisol
Previous study reported that protease belonging to hydrolytic enzymes that reflect the whole soil microbial activity and are mainly derived from soil bacteria and fungi . Our result also found that protease activity significantly increased at D 1 level in Phaeozem and D 3 level in Luvisol (Figure 2b). The results might be attributed to the adsorption properties of biochar, which adsorbs intermediates and increases the availability of inorganic N, thus promoting enzyme activity (Bailey et al., 2011). Biochar application could increase the availability of soil N (Chintala et al., 2014), and improve the activity of protease, which might be related to C:N ratio and its caused N immobilization (Lehmann et al., 2003). Furthermore, our results found that biochar application increased C:N ratio and protease activity (Table 2; Figure 2b). The high C input with biochar application led to N deficiency and caused relative abundance fungi, and to obtain available N through increasing protease activity. Moreover, it has been documented that N-hydrolyzing enzyme such as protease activities is significantly associated with SOC content (Wu et al., 2021). This is ascribed to the ability of SOC to provide abundance substrates, which induce the production and secretion of extracellular enzyme. Generally, although biochar is difficult to be utilized by microorganisms due to its stability and difficult in decomposition (Angst & Sohi, 2013;Zimmerman et al., 2011), the gradual decomposition of unstable carbon components increases substrates for microbial metabolism (Lehmann et al., 2011).
Amidases can hydrolyze amide compounds to produce ammonia and corresponding carboxylic acids (Kandeler et al., 2011). Some studies reported that amidase activity is related to microbial biomass (Nayak et al., 2007). In our study, amidase activity increased with biochar application rate, and significantly at D 2 and D 3 levels ( Table 3). The reasons for the increase in amidase activity are as follows: First, biochar is aged under natural conditions after applied to soil, and structure is broken so that the absorbed nutrients are released to provide nutrients for growth and metabolism of microorganisms and crops. Second, the functional groups on aging surface of biochar are changing, which weak competition with amidase substrates (Frankenberger & Tabatabai, 1981). Therefore, biochar application could influence enzymatic mineralization of organic N through promoting the availability of N.

| Soil organic nitrogen fractions in relation to soil properties and enzyme activities
More than 95% of soil organic N is not directly available for crops and needs to be further mineralized by soil microbial and enzymatic processes, and small molecules (such as amino acids and amino sugars) derived from organic N degradation are available for crop uptake (Knowles, 1977;Nannipieri & Eldor, 2009;Tian et al., 2017). In our study, we found that organic N fractions were related to N-mineralization enzyme under biochar applied to contrasting soil, indicating that biochar application could affect soil organic N turnover. Amidase activity was significantly and positively correlated with AM and AIN in Phaeozem (Figure 5a), which might be attributed to the fact that the changes in functional groups on aged surface of biochar weaken substrates competing with amidase, and that AM is end product of enzymatic mineralization of organic N (Wu et al., 2021). However, protease was significantly negatively associated with AA in Luvisol ( Figure 5b) probably because biochar provided the abundance substrates and adsorbs intermediates that increased the availability of inorganic N and decreased N immobilization. Therefore, the analysis results indicating that N-mineralization enzyme was involved in organic N fraction turnover in different soil types.
Furthermore, RDA results suggested that organic N turnover was different in Phaeozem and Luvisol. In Phaeozem, the variation in organic N fractions related to protease activity, and MBC and SOC contents (Table 6; Figure 6a), indicating that microbial regulation of organic N turnover was not only involved in enzymatic mineralization but also required abundant C source. SOC is considered to be an indicator of soil quality, which plays a vital role in maintaining soil physical, chemical, and biological properties (Karami et al., 2012). Previous study reported that the significant increase in SOC content is related to biochar application rate (Demisie et al., 2014). In our study, biochar application rate significantly influenced SOC content and it showed significantly increased at D 2 and D 3 levels ( Table 2). Our results also indicated that both biochar application rate and soil types could significantly affect TC content, since biochar is rich in nutrients to meet the requirement of microorganisms. Soil MBC is not only most active part of SOC but also is reservoir of soil available nutrients, which can drive nutrient turnover (McGill et al., 1986). Moreover, the excessive application rate resulted in the higher C:N ratio would stimulate microbial community activities induce a higher N requirement, thereby enhancing microbial immobilization of N in Phaeozem. The low molecular weight Ncontaining substrates (e.g., AA and AS) can be directly taken up by microbes as C sources with intracellular deamination, whereas microorganisms which utilize organic N molecules also need to provide C substrates (Geisseler et al., 2010;Nannipieri & Eldor, 2009;Tian et al., 2017).
In Luvisol, protease activity was the most important factors influenced by organic N fractions (Table 6b; Figure 6), indicating that protease can regulate organic N turnover. Our results found that biochar application rate could reduce MBN content and significantly decreased in D 3 treatment (Table 3). This is consistent with the result by Zavalloni et al. (2011) who found that biochar application rate could reduce MBN content. This may be the fact that biochar application can stimulate microorganisms to participate in mineralization of organic N (Dempster et al., 2012). Another reason may be the relative deficiency of bioavailable N due to the C-source input of biochar application. Protease plays a vital role in driving the degradation of amino acid production and releasing oligopeptides and free amino acids that are then taken up by microbes (Prommer et al., 2014). In our study, N availability (AM) was relatively low in Luvisol, where soil microorganisms were forced to decompose unstable organic N pools by secreting extracellular enzymes to meet their requirement (Wu et al., 2021), thereby enhancing enzymatic mineralization. Our result found that DHA activity was significantly increased at D 3 level in Luvisol (Figure 2c), which may be the fact that the well-developed pores of biochar can provide a good environment for microorganisms, thereby increasing enzyme activity. However, DHA exists in living cells and participates in soil oxidation-reduction reaction, which largely characterizes soil microbial activity (Bi et al., 2012), which is influenced by the disturbance of soil pH due to alkaline biochar application. Furthermore, Luvisol is a typical acidic soil, but biochar application could provide an optimal pH for microorganism growth and metabolism in Luvisol. Therefore, biochar application significantly increased DHA activity in Luvisol, indicating that the bioavailable C source and nutrients in biochar could provide substrates and mineral nutrients for maintaining microbial activity. Previous research has reported that soil amino sugars are another N source for microorganisms when soil N availability is low, but amino acids are preferred over amino sugars . In addition, we also found that protease was related to AA (Figure 5b), suggesting that protease was an important indicator in regulating organic N turnover in Luvisol. Therefore, our results suggested that the circulation of soil organic N is not only related to bioavailability of soil N but also related to C source (Geisseler et al., 2010;Yang et al., 2016).

| CONCLUSION
Overall, our result indicated that soil TC content, and DHA, NAG, and protease activities could be significantly influenced by biochar application rate, soil types, and their interaction. The conversion mechanism of N was different when biochar applied to different soil, that is, N significantly accumulated in Phaeozem and a little increase in Luvisol. Furthermore, we found that biochar application increased active-and stable-SON contents in Phaeozem, whereas the higher amount of application could increase stable-SON contents. Therefore, the application of biochar has the potential to improve soil N supply and maintain soil N storage in Phaeozem. Specially, organic N turnover in Phaeozem was not only related to bioavailable N but also need carbon substrates, whereas enzymatic mineralization was the dominant process involved in decomposing organic N as bioavailable N was relatively low in Luvisol.