Responses of Soil Enzyme Activity to Long-Term Nitrogen Enrichment and Water Addition in a Typical Steppe

: Enzyme activity plays an important role in soil biochemical processes and is a key factor driving nutrient cycling. Although a great number of studies examined the effects of nitrogen (N) enrichment and water (W) addition on soil enzyme activity, most of them focused on the effect of only one resource and are based on short-term investigations. The separate and interactive effects of long-term changes in nitrogen and water on soil enzyme activity remain largely unexplored. In this study, we demonstrated the responses of two types of soil enzyme, β -1,4-glucosidase (BG) and acid phosphatase (APA), to increased nitrogen and water based on a 16-year experiment conducted in a typical grassland in northern China. The results show that: (1) nitrogen addition inhibited BG and APA in 2019 and 2020; (2) water addition had no signiﬁcant effect on BG activity, but signiﬁcantly reduced APA activity in 2020; and (3) redundancy analysis (RDA) showed that nitrogen and water addition affected soil enzyme activity mainly by affecting soil microbial biomass carbon (MBC). The present research offers a comprehensive explanation of how atmospheric nitrogen deposition and precipitation patterns affect the characteristics of microorganisms and the cycling of nutrients in grassland ecosystems.


Introduction
Global increase in nitrogen deposition [1] and changes in precipitation pattern [2] fundamentally changed the physicochemical properties and biochemical cycling of soils [3][4][5]. Soil microbes are key drivers of nutrient cycling, and soil enzyme regulation is the most important process in microbial regulation of nutrient cycling [6]. To meet microbial growth under conditions of soil resource scarcity, soil microorganisms secrete enzymes [7]. Enzymes, as product of soil biological activity, are indicator of microbial nutrient requirements. Soil enzyme activity reflects the rate of soil microbial metabolism and biochemical cycling processes. The soil enzyme converts complex organic matter into plant-available nutrients and is a key factor in determining soil organic matter decomposition and nutrient cycling [8][9][10]. β-1,4-glucosidase (BG) is a soil enzyme secreted by cellulose, which often takes soil carbon as the matrix and participates in the carbon cycle. Phosphorus transformation-related acid phosphatase (APA) hydrolyzes the ester bonds in soil organic phosphorus to promote phosphorus conversion, determining the phosphorus availability for plants. Soil enzyme activity would change with environmental fluctuations. While there is a growing amount of research investigating the effects of changes in precipitation or nitrogen availability on soil enzyme activity, the majority of these studies focused

Soil Sampling and Chemical Property Analysis
From May to September in 2019, soil samples were collected every two weeks in each subplot. A total of five soil cores, measuring 3 cm in diameter and 10 cm in depth, were obtained. The percentage weight loss of fresh soil cores was determined by calculating the soil moisture as the difference in weight between fresh and dry soil cores after they were weighed and dried at a temperature of 105 °C for a duration of 48 h. At the conclusion of August in the years 2019 and 2020, seven soil samples were randomly chosen from every plot. The levels of inorganic nitrogen in soil samples were measured with a FIAstar 5000 Analyzer, a flow injection autoanalyzer manufactured by Foss Tecator in Denmark. We stirred dry soil in CO2-free deionized water (1:2.5 soil to water ratio) for one minute and allowed the soil to stabilize for an hour before measuring its pH using a pH meter (METTLER TOLEDO, FE28, Zurich, Switzerland).

Soil Enzyme Activities Analysis
The activities of BG and APA were analyzed according to Treseder [36]. For this test, a fresh soil sample weighing 1 g was combined with 4 milliliters of adjusted universal buffer (pHBG = 6.0, pHacid APA = 6.5) and 1 milliliter of substrate (pHAPS = 6.5). BG is represented by p-nitrophenol-β-D-glucopyranoside, while APA is represented by p-nitrophenyl-phosphate. After the reaction, 4 mL of Tris (hydroxymethyl) aminomethane was introduced with a concentration of 0.1 M and 1 mL of CaCl2 (THAM, pH = 12) at 37 °C. The reaction in the case of APA was halted by the addition of 1 mL of CaCl2 and 4 mL of NaOH. Following the filtration process, the solution was assessed at a wavelength of 450 nm using a UV-VIS spectrometer treasurer hectometer (MAPADA, P9, Shanghai, China). The activities of BG and APA were measured in milligrams of p-nitrophenol (PNP) per kilogram per hour.

Soil Sampling and Chemical Property Analysis
From May to September in 2019, soil samples were collected every two weeks in each subplot. A total of five soil cores, measuring 3 cm in diameter and 10 cm in depth, were obtained. The percentage weight loss of fresh soil cores was determined by calculating the soil moisture as the difference in weight between fresh and dry soil cores after they were weighed and dried at a temperature of 105 • C for a duration of 48 h. At the conclusion of August in the years 2019 and 2020, seven soil samples were randomly chosen from every plot. The levels of inorganic nitrogen in soil samples were measured with a FIAstar 5000 Analyzer, a flow injection autoanalyzer manufactured by Foss Tecator in Denmark. We stirred dry soil in CO 2 -free deionized water (1:2.5 soil to water ratio) for one minute and allowed the soil to stabilize for an hour before measuring its pH using a pH meter (METTLER TOLEDO, FE28, Zurich, Switzerland).

Soil Enzyme Activities Analysis
The activities of BG and APA were analyzed according to Treseder [36]. For this test, a fresh soil sample weighing 1 g was combined with 4 milliliters of adjusted universal buffer (pHBG = 6.0, pHacid APA = 6.5) and 1 milliliter of substrate (pHAPS = 6.5). BG is represented by p-nitrophenol-β-D-glucopyranoside, while APA is represented by p-nitrophenyl-phosphate. After the reaction, 4 mL of Tris (hydroxymethyl) aminomethane was introduced with a concentration of 0.1 M and 1 mL of CaCl 2 (THAM, pH = 12) at 37 • C. The reaction in the case of APA was halted by the addition of 1 mL of CaCl 2 and 4 mL of NaOH. Following the filtration process, the solution was assessed at a wavelength of 450 nm using a UV-VIS spectrometer treasurer hectometer (MAPADA, P9, Shanghai, China). The activities of BG and APA were measured in milligrams of p-nitrophenol (PNP) per kilogram per hour.

Statistical Analysis
ANOVAs with split plot designs were employed to examine the impact of water on soil enzymatic activity, specifically focusing on the interaction between nitrogen and water.
A study was performed to analyze the correlation between soil enzymatic activity and soil moisture (SM) using Pearson correlation analysis. The study also considered soil  TC:TN, TN:TP, soil microbial biomass carbon (MBC), soil microbial biomass nitrogen (MBN), and MBC:MBN. All statistical analyses were performed using the R software (V.3.3.2), and p < 0.05 was considered a statistically significant difference. Redundancy analysis (RDA) was undertaken using the Canoco software (Canoco for Windows 5.02). All bar graphs were drawn using Origin 2021 (https://www.originlab.com/, accessed on 1 May 2023).

Effects of Nitrogen and Water Addition on the Physical and Chemical Properties of Soils
Soil temperature experienced significant alterations as a result of both nitrogen enrichment and the addition of water (p < 0.001, Table 1, Figure 2A). The addition of water significantly affected soil moisture, while the addition of nitrogen had no effect on SM (Table 1). Water increased SM by 71.57% (p < 0.05, Figure 2B), the simultaneous addition of nitrogen and water increased SM by 82.81% (p < 0.05). In 2019, adding water increased SM by 84.32% (p < 0.05), while adding nitrogen and water simultaneously increased SM by 87.15% (p < 0.05), and adding nitrogen had no effect. In 2020, both nitrogen and water increased SM, nitrogen application increased SM by 20.03% (p < 0.05), and water addition increased SM by 56.01% (p < 0.05), the simultaneous addition of nitrogen and water increased SM by 77.53 (p < 0.05). Both nitrogen and water had an effect on pH (p < 0.001, Table 1), and the simultaneous addition of nitrogen and water had an interactive effect on pH (p <0.01, Table 1). Addition of nitrogen reduced pH by 14.57% (p < 0.05, Figure 2C), addition of water increased pH by 8.65% (p <0.05), and simultaneous addition of nitrogen and water reduced pH by 4.10% (p < 0.05). In 2019, addition of nitrogen decreased pH by 22.14% (p <0.05), addition of water increased pH by 3.72% (p < 0.05), and simultaneous addition of nitrogen and water decreased pH by 10.67% (p < 0.05). In 2020, addition of nitrogen reduced pH by 15.74% (p < 0.05), while simultaneous addition of nitrogen and water reduced pH by 7.22%, addition of water had no effect on pH.  Nitrogen and water addition had no impact on TC or TN (Table 1, Figure 3A, B). There was a significant impact on TP after nitrogen addition (p < 0.001, Table 1). Nitrogen and water had an interactive effect on TP (p < 0.05) and reduced TP ( Figure 3C). The addition of nitrogen and water did not have any impact on AP, as shown in Table 1 and Figure 3D. Nitrogen addition significantly affected IN (p < 0.05, Table 1), and adding nitrogen increases inorganic nitrogen ( Figure 3E). Nitrogen addition significantly affected Nitrogen and water addition had no impact on TC or TN (Table 1, Figure 3A, B). There was a significant impact on TP after nitrogen addition (p < 0.001, Table 1). Nitrogen and water had an interactive effect on TP (p < 0.05) and reduced TP ( Figure 3C). The addition of nitrogen and water did not have any impact on AP, as shown in Table 1 and Figure 3D. Nitrogen addition significantly affected IN (p < 0.05, Table 1), and adding nitrogen increases inorganic nitrogen ( Figure 3E). Nitrogen addition significantly affected NO 3 − −N (p < 0.05, Table 1), and the NO 3 − −N showed a significant increase due to the nitrogen addition treatments ( Figure 3F). Water addition significantly affected NH 4 + −N (p < 0.01, Table 1). The addition of water significantly increased the NH 4 + −N (p < 0.05, Figure 3G). In 2019, the addition of nitrogen and water had no effect on NH 4 + −N. In 2020, the addition of nitrogen had no effect on NH 4 + −N, Water addition increased NH 4 + −N (p < 0.05). Both nitrogen and water had no effect on TC:TN (Table 1, Figure 3H). Nitrogen addition significantly affected TN:TP (p < 0.001), and nitrogen and water had an interactive effect on TN:TP (p < 0.01). The addition of nitrogen increased TN:TP (p < 0.05, Figure 3I), while the addition Agronomy 2023, 13, 1920 6 of 13 of water also increased TN:TP (p < 0.05), and simultaneous addition of nitrogen and water increased TN:TP (p < 0.05). 0.05). Both nitrogen and water had no effect on TC:TN (Table 1, Figure 3H). Nitrogen addition significantly affected TN:TP (p < 0.001), and nitrogen and water had an interactive effect on TN:TP (p < 0.01). The addition of nitrogen increased TN:TP (p < 0.05, Figure  3I), while the addition of water also increased TN:TP (p < 0.05), and simultaneous addition of nitrogen and water increased TN:TP (p < 0.05).

Effect of Nitrogen Addition and Water Addition on Enzyme Activity and Microbial Biomass
Nitrogen addition had a tendency to inhibit MBC (Table 1, Figure 4A), with nitrogen addition decreasing MBC by 14.39% and water addition increasing MBC by 15.53%. Simultaneous addition of nitrogen and water had no effect on MBC. In 2019, nitrogen reduced MBC by 16.88%, while water addition increased MBC by 18.00%. In 2020, nitrogen reduced MBC by 9.61% and water addition increased MBC by 10.80%.
Nitrogen addition promoted MBN (Table 1, Figure 4B), nitrogen addition increased MBN by 51.44%, water addition had a tendency to promote MBN increases, water addition increased MBN by 36.06%, and simultaneous addition of nitrogen and water had no effect on MBN. In 2019, neither nitrogen nor water had an effect on MBN. But in 2020, nitrogen addition significantly promoted MBN by 100.92% (p < 0.05), while water addition showed a trend of promoting MBN by 84.77%. Both nitrogen and water additions had an effect on MBC:MBN (Table 1, Figure 4C). However, nitrogen addition significantly reduced MBC:MBN by 37.72% in 2019 (p < 0.05), and water addition significantly promoted MBC:MBN by5.37% (p < 0.05), the addition of nitrogen and water had no effect on MBC:MBN. In 2020, nitrogen addition significantly reduced MBC: MBN by 69.74% (p < 0.05), while water addition significantly reduced MBC:MBN by 43.67% (p < 0.05). The simultaneous addition of nitrogen and water to MBC:MBN had no effect (p > 0.05).
Nitrogen addition had a tendency to inhibit MBC (Table 1, Figure 4A), with nitrogen addition decreasing MBC by 14.39% and water addition increasing MBC by 15.53%. Simultaneous addition of nitrogen and water had no effect on MBC. In 2019, nitrogen reduced MBC by 16.88%, while water addition increased MBC by 18.00%. In 2020, nitrogen reduced MBC by 9.61% and water addition increased MBC by 10.80%. Nitrogen addition promoted MBN (Table 1, Figure 4B), nitrogen addition increased MBN by 51.44%, water addition had a tendency to promote MBN increases, water addition increased MBN by 36.06%, and simultaneous addition of nitrogen and water had no effect on MBN. In 2019, neither nitrogen nor water had an effect on MBN. But in 2020, nitrogen addition significantly promoted MBN by 100.92% (p < 0.05), while water addition showed a trend of promoting MBN by 84.77%. Both nitrogen and water additions had an effect on MBC:MBN (Table 1, Figure 4C). However, nitrogen addition signifi-  Although BG activity did not respond to nitrogen addition or water addition conditions (Tables 1 and 2, Figure 5A), nitrogen addition showed an inhibitory trend on BG compared to the control. The addition of nitrogen reduced BG by 7.90%, but the addition of water promoted BG by 14.93%, the addition of nitrogen and water simultaneously decreased BG by 1.21%. The trend of BG in 2019 and 2020 was similar. In 2019, the addition of nitrogen and water had no effect on BG, the addition of water increased BG by 15.95%, and the simultaneous addition of nitrogen and water decreased BG by 18.92%. In 2020, the effects of nitrogen and water were different, with nitrogen addition significantly inhibiting BG by 16.06% (p < 0.05). Water addition had a tendency to promote BG, with water addition increasing BG by 13.93%, while nitrogen and water addition simultaneously increased BG by 16.16%.
The effect of adding water on APA activity varied over time (Tables 1 and 2, Figure 5B), and nitrogen and water interacted to affect APA activity. Both nitrogen (p < 0.001) and water significantly inhibited APA (p < 0.01, Table 1). APA activity decreased by 9.86% with nitrogen addition (p < 0.05, Figure 4B), and decreased by 24.45% with water addition (p < 0.05, Figure 4B). In 2019, the addition of nitrogen inhibited APA by 24.07%, while the Agronomy 2023, 13,1920 8 of 13 addition of water had no effect on APA. The simultaneous addition of nitrogen and water inhibited APA, resulting in a decrease in APA by 37.52% (p < 0.05). In 2020, both nitrogen and water inhibited APA, and decreased by 75.27% and 80.76%, respectively (p < 0.05). The simultaneous addition of nitrogen and water reduced APA by 76.95%. Table 2. Results (F-values) of linear mixed-effects models with a repeated measures split-plot design on the effects of year (Y), water (W) addition, nitrogen (N) addition, and their interactions on BG and APA activity.  The effect of adding water on APA activity varied over time (Tables 1 and 2, Figure  5B), and nitrogen and water interacted to affect APA activity. Both nitrogen (p < 0.001) and water significantly inhibited APA (p < 0.01, Table 1). APA activity decreased by 9.86% with nitrogen addition (p < 0.05, Figure 4B), and decreased by 24.45% with water addition (p < 0.05, Figure 4B). In 2019, the addition of nitrogen inhibited APA by 24.07%, while the addition of water had no effect on APA. The simultaneous addition of nitrogen and water inhibited APA, resulting in a decrease in APA by 37.52% (p < 0.05). In 2020, both nitrogen and water inhibited APA, and decreased by 75.27% and 80.76%, respectively (p < 0.05). The simultaneous addition of nitrogen and water reduced APA by 76.95%.

Key Factors Driving Changes in Soil Enzyme Activity
To better understand the links of soil physicochemical properties to BG and APA, linear regressions were plotted. The results showed that BG was significantly correlated with TC, TN Figure S3). In Figure 6, we employed RDA to differentiate the influence of environmental factors on soil enzyme activity. BG and APA were used as biological variables, while soil physical Agronomy 2023, 13, 1920 9 of 13 and chemical property indicators were used as environmental variables. The results showed that the RDA1 and RDA2 of soil physicochemical properties explained 55.53% and 0.82% of soil enzyme activity varies, respectively. Therefore, the alterations in enzyme function were primarily influenced by the first axis. The results showed that MBC had the strongest effect on enzyme activity 24.2% (p = 0.002), TN:TP explained 16.7% (p = 0.002) and NH 4 + −N (p = 0.008) explained 7.6% of variations in soil enzyme activity (Table S1).  (Table S1).

Effects of Nitrogen and Water Addition on Soil Enzyme Activity
The nutrient needs of microorganisms [29] and their functional response to environmental changes [9] can be evaluated by measuring soil enzyme activity. BG and APA, as components of soil extracellular enzymes, are often used to elucidate changes in C and P cycling processes in response to environmental changes [12,13]. Previous studies found that nitrogen addition significantly affected enzyme activity associated with soil P-conversion [13,14,16]. Our results showed that nitrogen addition inhibited APA (Table  1, Figure 5A,B), which was line with some empirical studies that nitrogen addition reduced soil enzyme activity [35,36]. The main reason for these phenomena was that the addition of nitrogen in the external environment leads to an increase in inorganic nitrogen in the soil, which reacts chemically with soil organic matter to produce substances that are unavailable to microorganisms [10], thus reducing microbial activity. Another study found that microbial inhibition was more common in fertilizer treatments over a five-year experiment [37]. The addition treatments of nitrogen and water for 16 years demonstrated that long-term nitrogen addition resulted in soil acidification ( Figure 2C), which is harmful to the activity of enzymes in the soil, leading to a decrease in soil enzyme activity [38]. At the same time, long-term nitrogen application can damage soil aggregates, destroy the living environment of microorganisms, and inhibit microbial activity [39]. Our results showed that nitrogen addition had no effect on BG (Table 1, Figure  5A,B). The possible reason for this phenomenon is the long residence time required for nutrients in the soil and the untimely feedback to nitrogen [5,17].
The addition of water was determined to have no notable impact on BG activity, except for a significant decrease in APA activity in 2020. Most previous studies showed that water addition would not affect most soil enzyme activities [15]. As the feedback of soil enzyme activity to water is related to the pre-soil moisture content [40], and there is a threshold of microbial community response to moisture content [34]. Once the threshold is exceeded, increased soil moisture can lead to nutrient leaching and reduction in microbial resource use, and therefore soil microbial enzyme activity [5]. A study showed that β-1,4-glucosidase activity did not increase with increasing precipitation, and changes

Effects of Nitrogen and Water Addition on Soil Enzyme Activity
The nutrient needs of microorganisms [29] and their functional response to environmental changes [9] can be evaluated by measuring soil enzyme activity. BG and APA, as components of soil extracellular enzymes, are often used to elucidate changes in C and P cycling processes in response to environmental changes [12,13]. Previous studies found that nitrogen addition significantly affected enzyme activity associated with soil P-conversion [13,14,16]. Our results showed that nitrogen addition inhibited APA (Table 1, Figure 5A,B), which was line with some empirical studies that nitrogen addition reduced soil enzyme activity [35,36]. The main reason for these phenomena was that the addition of nitrogen in the external environment leads to an increase in inorganic nitrogen in the soil, which reacts chemically with soil organic matter to produce substances that are unavailable to microorganisms [10], thus reducing microbial activity. Another study found that microbial inhibition was more common in fertilizer treatments over a five-year experiment [37]. The addition treatments of nitrogen and water for 16 years demonstrated that long-term nitrogen addition resulted in soil acidification ( Figure 2C), which is harmful to the activity of enzymes in the soil, leading to a decrease in soil enzyme activity [38]. At the same time, long-term nitrogen application can damage soil aggregates, destroy the living environment of microorganisms, and inhibit microbial activity [39]. Our results showed that nitrogen addition had no effect on BG (Table 1, Figure 5A,B). The possible reason for this phenomenon is the long residence time required for nutrients in the soil and the untimely feedback to nitrogen [5,17].
The addition of water was determined to have no notable impact on BG activity, except for a significant decrease in APA activity in 2020. Most previous studies showed that water addition would not affect most soil enzyme activities [15]. As the feedback of soil enzyme activity to water is related to the pre-soil moisture content [40], and there is a threshold of microbial community response to moisture content [34]. Once the threshold is exceeded, increased soil moisture can lead to nutrient leaching and reduction in microbial resource use, and therefore soil microbial enzyme activity [5]. A study showed that β-1,4-glucosidase activity did not increase with increasing precipitation, and changes in soil water content also affected microbial intracellular and extracellular pressures [41], which in turn affected the release of soil enzymes to the environment [42]. Our results showed a significant reduction in APA activity for the water augmentation treatment in 2020 ( Figure 5B); this may be related to annual precipitation. Our study found that annual precipitation in 2020 was 410 mm, significantly higher than the historical average annual precipitation of 379 mm ( Figure S1). High precipitation can easily create a hypoxic environment for the soil, thereby inhibiting APA activity [12]. In addition, adding water can improve the physical and chemical properties of the soil, as well as increase the total and available phosphorus content of the soil, resulting in soil microorganisms being in a state of phosphorus nutrient 'enrichment' [43], which in turn reduces the secretion or activity of phosphorus nutrient-related enzymes.

Key Factors Driving Soil Enzyme Activity
The RDA results indicated that the main indicator factor affecting soil enzyme activity was MBC (Figure 6), which was positively correlated with BG and APA, indicating that MBC stimulated enzyme activity. It is reported that MBC plays the most important role in the soil carbon-phosphorus cycle [44]. The number of soil microorganisms can to some extent indicate the vigorous level of material metabolism in the soil, which is a comprehensive reflection of soil physical and chemical properties and an important indicator of soil fertility. Most soil enzymes are derived from microorganisms and other organic tissues (animals, plants, organisms, and their residues), so the number of soil microorganisms is closely related to soil enzyme activity. In addition, Allison et al. [45] discussed the importance of microbial physiology and biomass in the soil carbon and nitrogen cycle under climate change and found that microbial biomass and enzyme activity are affected by climate change in the same direction. With the increase in enzyme activity and microbial biomass, the soil carbon cycle was affected differently. However, when enzyme activity and microbial biomass adapt to climate change, the effect of climate change on the carbon cycle is reduced. In this study, nitrogen addition reduced MBC and enzyme activity, which was consistent with previous studies. Yang et al. [46] found that the presence of soil microbial biomass carbon has a significant impact on soil enzyme activity, with enzyme activity showing an increase as microbial biomass carbon levels rise. In summary, the direction of change in microbial biomass carbon and enzyme activity was consistent.
There was a negative correlation between NH 4 + −N and both BG and APA, indicating that NH 4 + −N inhibits the activity of the enzymes. The effect of NH 4 + −N on carbon and phosphorus cycling enzyme activity is enhanced by nitrogen application, and NH 4 + −N has a stronger effect on soil acidification than NO 3 − −N. Water addition was shown to increase NH 4 + −N in the soil and react with soil organic matter to produce indoles and pyrroles [47], which are difficult for microorganisms to utilize [48], thereby inhibiting microbial activity. In addition, NH 4 + −N often displaces salt-based ions (Ca 2+ , Mg 2+ , K + , and Na + ) from the surface of soil colloids, making them susceptible to leaching. When NH 4 + −N is taken up by plants, it releases H + into the soil solution, thus causing soil acidification [49], which is unsuitable for microbial survival. TN:TP is negatively correlated with APA, indicating that APA decreases as TN:TP increases, suggesting that soil stoichiometry may be an important driver of soil enzyme activity. Ecological stoichiometric ratios can indicate constraints to biogeochemical cycling as a predictor of carbon cycling [50,51]. Sinsabaugh et al. [48] found that soil enzyme activity is sensitive to soil stoichiometry (C:N:P) [52]. It was suggested that the effect of soil stoichiometry on enzyme activity is mainly through the maintenance of stoichiometric balance by microorganisms [43]. Soil stoichiometry can strongly influence the structure and activity of microbial communities and regulate the metabolism of microbial communities, leading to changes in soil enzyme activity [53][54][55].

Conclusions
A field experiment based on 16 years of nitrogen and precipitation manipulated in temperate grassland in northern China, our study showed that nitrogen addition decreased both BG and APA, and that water addition only inhibited APA in 2020. Nitrogen and water addition affected enzyme activity mainly by affecting MBC, NH 4 + −N, and TN:TP, with MBC having the most important effect. In general, the addition of nitrogen or water over a long-term of time did not affect BG, suggesting that BG is a resilient enzyme that remains unaffected by changes in the environment. In contrast, APA is sensitive to nitrogen and water addition and responds dramatically to interannual fluctuations in precipitation, suggesting that APA can be used as an indicator of changes in soil nutrient cycling due to climate change. However, our study used only BG and APA to explore the effects of environmental changes and their implications on carbon and phosphorus cycling. Since different enzymes could respond differently to environmental variations, the representativeness of only one enzyme on carbon or phosphorus cycling is not convincing enough, which warrants further investigations on more enzymes in future studies. Although based on the long-term experiment of nitrogen and water addition, our sampling on enzyme activity was only conducted for two years, and the interannual variation was rather great. Thus, long-term monitoring is needed in future studies to reveal the more reliable patterns and mechanisms of enzyme activity in response to global change in the semi-arid grassland.