N₂O Production and Reduction in Chinese Paddy Soils: Linking Microbial Functional Genes with Soil Chemical Properties

Abstract

Nitrous oxide (N₂O) emissions from paddy soils significantly contribute to global warming; however, the regulatory mechanisms of microbial denitrification remain poorly understood. This study investigated the biotic and abiotic drivers of N₂O production and reduction across seven paddy soils spanning China’s major rice-growing regions, using integrated qPCR, incubation experiments, and multivariate analyses. Results demonstrated niche partitioning among denitrifying microorganisms. Pearson correlation analysis demonstrated significant positive correlations between potential N₂O production rates and the abundances of denitrification genes (nirS, nirK, and fungal nirK), as well as between N₂O reduction rates and nosZ gene abundances (both clade I and II). Key soil chemical properties, including pH, total carbon (TC), and NH₄⁺-N content, showed significant relationships with both potential N₂O production rates and reduction rates. Furthermore, random forest analysis identified nirS, fungal nirK, TC, and pH as key predictors of N₂O production, while nosZ (clade I and II), TC, and pH governed N₂O reduction. Structural equation modeling revealed that nirS-type bacteria predominantly drove N₂O production, whereas nosZ II-encoded microorganisms primarily mediated N₂O reduction. Moreover, TC exhibited direct positive effects on both processes, while pH indirectly influenced N₂O production by regulating nirS abundance and affected reduction via nosZ Ⅱ modulation. These findings provide a mechanistic framework for mitigating agricultural denitrification-derived N₂O emissions through a targeted management of soil carbon and pH conditions to optimize complete denitrification.

1. Introduction

Nitrous oxide (N₂O) is a potent greenhouse gas with a global warming potential 298 times that of CO₂ over a 100-year timescale and contributes to stratospheric ozone depletion [1]. Agricultural soils, particularly paddy fields, account for approximately 60% of global anthropogenic N₂O emissions [2]. Therefore, understanding and controlling N₂O emissions in agricultural soil is crucial for mitigating climate change.

The denitrification pathway involves the stepwise reduction of nitrate (NO₃⁻) to dinitrogen (N₂) via nitrite (NO₂⁻), nitric oxide (NO), and N₂O as intermediates. This process is mediated by microorganisms carrying functionally distinct genes [3,4], including those encoding nitrite reductases (nirK and nirS) and nitrous oxide reductase (nosZ) [5]. The reduction of NO₂⁻ to NO is catalyzed by either a copper-dependent (nirK-encoded) or a heme-dependent (nirS-encoded) nitrite reductase [6,7]. Although these genes exhibit environmental preferences, current understanding remains contradictory [8,9]. While nirS-type denitrifiers typically dominate in neutral-alkaline soils (pH 6.5–8.0), nirK-types prevail in acidic conditions (pH < 6.0) [6]. However, exceptions frequently occur due to microbial adaptation and local soil heterogeneity [10,11,12]. Carbon and nitrogen availability differentially regulate these communities, with organic matter frequently favoring nirK-types [13], while long-term nitrogen fertilization tends to enrich nirS- and nosZ-harboring populations, altering N₂O emissions [14].

The terminal step of denitrification, the reduction of N₂O to N₂, represents the only known biological sink for N₂O in agricultural systems [15]. This critical process is catalyzed by nitrous oxide reductase, encoded by two phylogenetically distinct nosZ clades (nosZ I and nosZ II) [16]. Recent studies reveal that fungi and approximately 30% of denitrifiers lack nosZ genes entirely, while the two clades appear to occupy different ecological niches [17]. Substantial uncertainties remain regarding their environmental regulation, as soil management factors (e.g., pH adjustment, tillage practices, and fertilization) influence nosZ clade abundance and activity, yet reported effects are inconsistent across studies [18,19]. Notably, emerging evidence suggests that nosZ II-harboring microorganisms may play a more significant role in N₂O reduction than previously recognized, particularly in acidic soils, challenging the traditional paradigm of nosZ I dominance [18].

These knowledge gaps are particularly pronounced in paddy ecosystems, where the complex interplay between flooding regimes, soil properties, and microbial communities creates unique but poorly understood denitrification dynamics. To address these limitations, we conducted a comprehensive investigation across seven representative paddy soils spanning China’s major rice-growing regions (24–42° N). This study employs an integrated approach combining qPCR, incubation experiments, and multivariate analysis to (1) quantify variations in potential N₂O production rates and N₂O reduction rates; (2) identify key abiotic and biotic drivers of denitrification partitioning; and (3) evaluate the relative importance of the direct versus indirect effects of environmental factors on denitrification outcomes. The findings will provide a mechanistic framework for developing targeted N₂O mitigation strategies tailored to different paddy soil ecosystems.

2. Materials and Methods

2.1. Soil Sampling

During November and December 2023, we collected soil samples from typical paddy fields with a range of 50 × 50 m² at seven sites across China (24.84° N–41.83° N, 102.86° E–119.23° E, Figure 1). From each site, five subsamples were taken from the top 20 cm of soil, thoroughly mixed, and divided into three replicates. Soils were stored in aseptic plastic bags on ice and immediately transported to the laboratory, sieved through a 2-mm mesh, and stored at 4 °C for physicochemical analysis and incubation or at −80 °C for DNA extraction.

2.2. Soil Chemical Properties Analysis

Soil pH was measured in a 1:5 (w/v) soil-to-water suspension using a calibrated pH meter (Seven Compact S210, Mettler-Toledo, Greifensee, Switzerland). Ammonium (NH₄⁺-N) and nitrate (NO₃⁻-N) were extracted with 2 M KCl (1:10 soil-to-solution ratio) and quantified via flow injection analyzer (SAN++, Skalar, Breda, The Netherlands) [20]. Dissolved organic carbon (DOC) and nitrogen (DON) were extracted with 0.5 M K₂SO₄ and measured using a TOC analyzer (Multi N/C 3100, Analytik Jena, Jena, Germany) [21]. Total carbon (TC) and nitrogen (TN) were determined by high-temperature oxidation methodology using an elemental analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany) [22]. Available phosphorus (AP) was extracted with 0.5 M NaHCO₃ and analyzed via the molybdenum blue method [23].

2.3. DNA Extraction and Quantitative PCR

Total genomic DNA was extracted from 0.5 g of soil using the FastDNA SPIN Kit for Soil (MP Biomedicals, LLC, Solon, OH, USA) [22]. DNA purity and concentration were verified spectrophotometrically (NanoDrop 2000, Thermo Fisher Scientific, Waltham, MA, USA) [22], with A260/A280 ratios maintained between 1.8 and 2.0. Quantitative PCR (qPCR) was performed on a LightCycler^® 480II (Roche, Penzberg, Germany) to quantify bacterial (16s rRNA), fungal (ITS), and denitrification-related gene (nirS, nirK, nosZI, nosZII) abundance. Primers and cycling conditions are listed in

Table 1. Primer sets and thermal cycling conditions for qPCR.

Gene	Primer	Sequence 5′ to 3′	Thermal Cycling Conditions
16s rRNA	515F/806R	GTGYCAGCMGCCGCGGTAA/ GGACTACNVGGGTWTCTAAT	95 °C for 1 min, 1 cycle; 95 °C for 30 s, 55 °C for 30 s, 72 °C for 30 s, 40 cycles; 75 °C for 10 min. [24]
Fungal ITS	5.8S-F/ITS1f-R	CGCTGCGTTCTTCATCG/ TCCGTAGGTGAACCTGCGG	95 °C for 1 min, cycle; 98 °C for 10 s, 53 °C for 30 s, 72 °C for 30 s, 40 cycles; 75 °C for 10 min. [25]
nirS	nirS-cd3aF/ nirS-R3cd	GASTTCGGRTGSGTCTTGA/ ATCATGGTSCTGCCGCG	95 °C for 1 min, 1 cycle; 98 °C for 10 s, 53 °C for 30 s, 72 °C for 30 s, 40 cycles; 75 °C for 10 min. [26]
nirK	nirK-FlaCu/ nirK-R3Cu-GCb	ATCATGGTSCTGCCGCG/ GCCTCGATCAGRTTGTGGTT	94 °C for 3 min, 1 cycle; 94 °C for 30 s, 57 °C for 1 min, 73 °C for 1 min, 35 cycles; 75 °C for 10 min. [27]
Fungal nirK	nirKfF/ nirKfR	TACGGGCTCATGtaygtnsarcc/ AGGAATCCCACAscnccyttntc	95 °C for 5 min, 1 cycle; 95 °C for 30 s, 61.5 °C for 30 s, 72 °C for 1 min, 35 cycle; 72 °C for 10 min. [28]
nosZ I	nosZ I F/ nosZ I R	GGCAARCTVTCDCCVAC/ AVCGGTCYTTVGAGAAYTT	95 °C for 2 min, 1 cycle; 95 °C for 45 s, 53 °C for 45 s, 72 °C for 1 min, 35 cycles; 72 °C for 10 min. [29]
nosZ II	nosZ- II -F/ nosZ- II -R	CTIGGICCIYTKCAYAC/ GCIGARCARAAITCBGTRC	95 °C for 5 min, 1 cycle; 95 °C for 30 s, 54 °C for 60 s, 72 °C for 1 min, 35 cycles; 72 °C for 10 min. [30]

. All amplification efficiencies were 94–101%, and R² values of 0.993–0.999.

2.4. Measurement of Potential N₂O Production Rate and N₂O Reduction Rate

Prior to analysis, 30 g of air-dried, 2-mm sieved soil for each sample was pre-incubated in 120 mL serum bottles at 25 °C in the dark for 5 days at 50% water-filled pore space (WFPS) to stabilize microbial activity. Following this stabilization period, the potential N₂O production rate and denitrification rate were determined using a modified acetylene inhibition technique with some modifications [31]. The N₂O reduction rate is the difference between the potential N₂O production rate and the denitrification rate.

Six parallel bottles were prepared for each sample. All bottles were sealed and purged by evacuating the air and filling with Ar. Three of them were used to measure the potential N₂O production rate. Acetylene (a specific inhibitor of the N₂O reductase) was added to the other three bottles to reach 0.1 atm partial pressure to measure the potential denitrification rate. Subsequently, all six bottles for each sample were amended with substrates (0.2 mg N g⁻¹ soil as NaNO₃; 2 mg C g⁻¹ soil as glucose), and adjusted moisture to 100% (w/w) [31,32]. Then, bottles were purged with Ar after 20 min shaking (180 rpm, 25 °C).

For potential N₂O production rate measurements, three bottles were incubated without acetylene. For denitrification rate determination, parallel triplicate cultures received 10% C₂H₂ (v/v, 0.1 atm partial pressure) to inhibit N₂O reductase [33]. All bottles were incubated for 4 h at 25 °C at 180 rpm. Headspace gas (10 mL) was sampled at 0.5 h, 1 h, 2 h, and 4 h intervals using gas-tight syringes, with N₂O concentrations quantified by gas chromatograph (GC-2010 Plus, Shimadzu, Japan) equipped with an electron capture detector. The potential N₂O production rate and denitrification rate were calculated by Equation (1)

$$F=\rho V{\displaystyle \frac{dC}{dt}}{\displaystyle \frac{273}{273+T}}/W$$

(1)

where F is potential N₂O production rate or denitrification rate (mg kg⁻¹ h⁻¹), $\rho$ is the density of N₂O at standard conditions (mg mL⁻¹), V is the serum bottle volume (mL), W is the soil weight (kg), T is the air temperature (°C), dC/dt is the variation of N₂O (ppb h⁻¹).

2.5. Statistical Analysis

One-way analysis of variance (ANOVA) was performed in SPSS (IBM SPSS version 24, SPSS Inc., Chicago, IL, USA) [20] to test the difference in soil chemical properties, microbial abundances, potential N₂O production, and reduction rates among different soils by the least significant difference (LSD) test at a 0.05 p-level. Principal component analysis (PCA) visualized the dissimilarities of microbial gene abundances and soil chemical properties using Origin (Origin 2023, OL, USA) [23]. Pearson correlation analysis was performed to correlate soil chemical properties and microbial gene abundances by the “ggcoorplot” package in R (Version 4.0.2). Random forest was performed using the “RandomForest” package in R to identify key predictors of N₂O production and reduction among multiple parameters, including soil chemical properties and microbial gene abundances. Structural equation modeling (SEM) was conducted by SPSS Amos (SPSS Amos IBM, version 24.0, USA) [22] to evaluate the direct or indirect effect of factors on denitrification-derived N₂O dynamics.

3. Results

3.1. Soil Chemical Properties

Soil chemical properties exhibited significant variation (Figure 2). pH ranges from 4.4 (Nanchang, lowest) to 7.6 (Beijing, highest). Nitrogen showed distinct patterns; NO_3⁻-N content varied between 3.4 mg kg⁻¹ (Zhengzhou) and 8.3 mg kg⁻¹ (Nanchang); NH₄⁺-N levels spanned 3.3 mg kg⁻¹ (Kunming) to 6.7 mg kg⁻¹ (Beijing); Total nitrogen (TN) ranged from 0.09% to 0.16%, with Fuzhou significantly higher than other sites. Dissolved organic nitrogen (DON) varied between 2.6 mg kg⁻¹ (Zhengzhou) and 9.0 mg kg⁻¹ (Nanchang). Dissolved organic carbon (DOC) ranged from 4.9 mg kg⁻¹ (Fuzhou) to 23.8 mg kg⁻¹ (Shenyang). Total carbon (TC) peaked in Beijing (3.6%) and reached its minimum in Zhengzhou (1.1%). Available phosphorus (AP) showed limited spatial variation, maintaining consistent levels across regions (31.0–35.2 mg kg⁻¹).

3.2. Microbial Gene Abundance

Microbial gene abundance exhibited significant variations across soils (Figure 3). Bacterial abundance (16s rRNA gene copies, Figure 3) ranged from 8.9 × 10⁸ copies g⁻¹ (Beijing) to 1.4 × 10⁹ copies g⁻¹ (Shenyang). Fungal abundance (fungal ITS copies, Figure 3 varied greatly, with Nanchang exhibiting the highest (3.9 × 10⁶ copies g⁻¹) and Kunming the lowest (8.7 × 10⁵ copies g⁻¹). For nitrite reductase genes, Beijing dominated in nirK (1.9 × 10⁷ copies g⁻¹) and fungal nirK (4.6 × 10⁴ copies g⁻¹), while Fuzhou showed preference for nirS (5.7 × 10⁷ copies g⁻¹). Regarding nitrous oxide reductase genes, Fuzhou exhibited peak nosZ I abundance (1.3 × 10⁸ copies g⁻¹) while Nanchang had the lowest (3.0 × 10⁷ copies g⁻¹). Beijing showed the highest nosZ II abundance (2.6 × 10⁸ copies g⁻¹), contrasting with Quzhou, which had the lowest (4.4 × 10⁷ copies g⁻¹).

3.3. Potential N₂O Production Rate and N₂O Reduction Rate

Beijing demonstrated the highest potential N₂O production rate (470 mg kg⁻¹ h⁻¹) and N₂O reduction rate (447 mg kg⁻¹ h⁻¹) among the soils (Figure 4). In contrast, Zhengzhou, Nanchang, Quzhou, and Kunming exhibited relatively low potential N₂O production rate (57–90 mg kg⁻¹ h⁻¹) and reduction rate (1–7 mg kg⁻¹ h⁻¹). Notably, Fuzhou exhibited both a high potential N₂O production rate (191 mg kg⁻¹ h⁻¹) and reduction rate (107 mg kg⁻¹ h⁻¹).

3.4. Principal Component Analysis

Principal Component Analysis (PCA) revealed distinct patterns in microbial gene abundance (Figure 5a) and soil chemical properties (Figure 5b) across the studied soils. The first two principal components explained 55.3% (PC1) and 18.4% (PC2) of the total variance in microbial gene abundance data. Denitrification-related microbial groups, including nirS, nirK, fungal nirK, nosZ I, and nosZ II, exhibited strong loading along PC1, suggesting their dominant role in shaping microbial structure.

For soil chemical properties, PC1 (37.8% variance explained) was strongly associated with variables including TC, pH, and DOC, collectively highlighting gradients in soil nutrient availability and chemical characteristics.

3.5. Correlation and Random Forest Analysis

Pearson correlation analysis demonstrated significant (p < 0.001) positive correlations between potential N₂O production rates and the abundances of N₂O production-related genes, which include nirS, nirK, and fungal nirK (Figure 6a). N₂O reduction rates were significantly correlated with nosZ gene abundances (both clade I and II). Significant positive correlations were observed between potential N₂O production rates, N₂O reduction rates, and specific soil chemical properties, including TC, pH, and NH₄⁺-N. The abundance of N₂O production-related genes (nirS, nirK, and fungal nirK) was significantly correlated with TC and pH, whereas nosZ gene abundances were significantly correlated with pH, TC, and NH₄⁺-N.

Random forest analysis confirmed the differential controls on N₂O production and reduction (Figure 6b). For N₂O production, nirS, fungal nirK, TC, and pH were identified as the most critical predictors. N₂O reduction was primarily governed by the nitrous oxide reductase genes nosZ II and nosZ I, as well as TC and pH.

3.6. Structural Equation Modeling

Structural equation modeling (SEM) revealed distinct pathways through which abiotic and biotic factors regulate potential N₂O production rate and N₂O reduction rate (Figure 7). The model showed excellent fit indices for both potential N₂O production rate (χ²/df = 1.064, RMSE = 0.057, CFI = 0.999, GFI = 0.975) and N₂O reduction rate (χ²/df = 2.161, RMSE = 0.024, CFI = 0.989, GFI = 0.951).

For the potential N₂O production rate, both total carbon (TC, p = 0.005) and pH (p < 0.001) showed significant positive direct effects. Among biotic factors, nirS gene abundance directly promoted N₂O production (p = 0.013). pH additionally showed indirect effects on N₂O production through its positive influence on nirS abundance.

Total carbon demonstrated a significant positive direct effect on N₂O reduction (p < 0.001), while the nosZ II gene abundance also directly promoted N₂O reduction (p < 0.001). pH indirectly enhanced N₂O reduction by positively influencing nosZ II abundance (p = 0.003).

4. Discussion

4.1. Biotic and Abiotic Factors Influencing Denitrification-Derived N₂O Production

Our study reveals intricate interactions between soil chemical properties and microbial functional genes in regulating denitrification-derived N₂O production across Chinese paddy soils. Although nirS, nirK, and fungal nirK were all significantly correlated with potential N₂O production rates, both random forest analysis and structural equation modeling confirmed the dominant role of nirS in driving N₂O production. Notably, sites with higher N₂O production (e.g., Beijing and Fuzhou) also exhibit higher nirS abundance, while patterns of nirK were inconsistent. These findings align with previous research suggesting that nirS-type denitrifiers are the primary contributors to N₂O production in agricultural soils [34]. For instance, Lee et al. (2017) quantitatively analyzed the spatial distribution of nirK and nirS along environmental gradients and based on gene expression profiles, proposed that nirS-type enzymes generally have a lower K_m and stronger substrate affinity, which may lead to increased N₂O accumulation [35]. Furthermore, nirS abundance showed a significant correlation with soil pH, consistent with earlier studies. Structural equation modeling demonstrated both direct and indirect pH-mediated effects on potential N₂O production. While Song et al. (2020) [36] found that higher soil pH promotes nirS gene abundance and that nirK tends to dominate in acidic conditions, our study observed unexpectedly low nirK abundance in acidic soils from Nanchang and Quzhou (pH 4.4–5.0). This finding contrasts with the conventional view and mirrors observations from Northeast China’s black soil belt, where multiple environmental factors—beyond pH—shape the biogeographical distribution of nirS and nirK denitrifiers [37].

In addition to pH, total carbon (TC) emerged as a key abiotic driver of N₂O production. Our findings support the carbon modulation hypothesis, which posits that organic matter availability regulates denitrification enzyme synthesis [38]. This carbon effect operates through direct impacts on denitrification enzyme activity, as demonstrated by studies showing that increased TC content can enhance denitrifying enzyme activity and alter N₂O emissions [38,39].

Notably, both correlation and random forest analysis identified fungal nirK as having a significant impact on N₂O production, highlighting the underappreciated role of fungal denitrifiers in paddy soil emissions. This is in line with the recent findings by Wei et al. (2024) [40]. However, structural equation modeling revealed that nirS-harboring bacteria exerted the most direct influence on N₂O production. This apparent discrepancy suggests that while fungal denitrifiers contribute substantially to overall N₂O emissions, bacterial denitrification remains the dominant pathway in these systems.

4.2. Biotic and Abiotic Factors Influencing N₂O Reduction

Our study provides new insights into the regulation of N₂O reduction in paddy soils by integrating microbial functional gene data with key soil chemical properties. While earlier studies proposed functional redundancy between nosZ clades I and II in agricultural ecosystems [41,42], our analyses (Figure 6 and Figure 7) revealed that nosZ II plays a predominant role in N₂O reduction. This aligns with recent spatial-phylogenetic evidence indicating ecological niche partitioning between the two clades [43]. Interestingly, our findings contrast with earlier reports that emphasized stronger correlations between nosZ I and N₂O reduction [44], highlighting the importance of local soil conditions in shaping denitrification dynamics.

Among these conditions, pH was found to play a central role in regulating nosZ II. Our findings support observations by Shaaban et al. (2023) [45], who reported similar trends in acidic soils, and further extend these results across a broader pH range (4.4–7.6). Low pH generally inhibits nosZ gene expression and suppresses the activity of nosZ-harboring microbes [46]. Multiple studies have shown that increasing soil pH can reduce N₂O emissions by enhancing the nosZ gene abundance and expression [47]. For instance, dolomite application in rice-rapeseed rotation soils (pH 5.44) significantly increased pH and decreased N₂O emissions while boosting nosZ transcript levels [47]. Notably, nosZ I appears more sensitive to acidification than nosZ II, which shows greater pH adaptability [46]. Therefore, pH changes may shift the relative abundance of these two clades, thereby altering the balance of N₂O reduction pathways.

Total carbon (TC) was also identified as a critical driver of N₂O reduction. Structural equation modeling revealed a strong direct effect of TC, supporting the carbon limitation theory proposed by Domeignoz-Horta et al. (2020) [48]. Carbon availability is essential for providing the energy and electrons needed for complete denitrification. When carbon is limited, the process may be hindered, resulting in N₂O accumulation rather than complete reduction to N₂ [49]. These findings underscore TC’s central role in modulating N₂O reduction capacity in paddy soils.

5. Conclusions

This study elucidates the complex regulation of N₂O dynamics in Chinese paddy soils through an integrated analysis of microorganisms and soil chemical properties. Results demonstrate that niche partitioning governs denitrification processes, with nirS predominantly driving N₂O production while nosZ II mediates N₂O reduction. Total carbon content directly modulates both N₂O production and reduction, while pH influences N₂O production both directly and indirectly through nirS abundance regulation, and indirectly affects N₂O reduction via nosZ II modulation. These mechanistic insights enable targeted N₂O mitigation strategies through optimized carbon and pH management in rice cultivation systems.