Optimizing the ratio of conventional urea to polyurethane-coated urea via single basal application improves growth, nitrogen use efficiency and economic return of tillering mustard

1 Introduction

With the improvement of living standards, the demand for green and high-quality vegetables has been increasing continuously. Vegetables are the second most widely cultivated crop in China after cereals, with a cultivation area of 22.33 million hectares and a yield of approximately 769 million tonnes (1). Mustard (Brassica juncea), native to China and belonging to the Brassica genus of the Brassicaceae family, is an important leafy vegetable and oilseed crop in the country (2). Tillering mustard, a specific variety of leaf mustard, is highly valued for its unique fresh taste and nutritional benefits. It is also processed into preserved products like Meigan cai (dried preserved mustard greens) and cured vegetables, contributing significantly to economic returns (3). Tillering mustard has high nutrient demands during its growth, requiring sufficient nitrogen fertilization to ensure optimal quality and yield (4). In China, vegetable production accounts for about 10% of total nitrogen (TN) fertilizer consumption (5). However, vegetable cultivation in the country has long been dependent on conventional nitrogen fertilizers, which are often over-applied, resulting in low nitrogen use efficiency and significant nutrient losses. This not only heightens the risk of agricultural non-point source pollution but also threatens regional ecological and environmental security (6). Novel fertilizers are products that enhance the nutritional functions of traditional chemical fertilizers or incorporate new resources, theories, methods, and technologies (7). Given the dual challenges of achieving efficient mustard production and ensuring environmental safety, traditional nitrogen fertilizers can no longer meet the evolving needs of the mustard industry. The promotion and use of novel fertilizers have thus become critical to fostering the green and sustainable development of the mustard sector (8).

Controlled-release fertilizer, as an advanced nutrient management tool, physically encapsulates fertilizer granules within polymer or inorganic coatings, forming a barrier that regulates water infiltration and subsequent nutrient diffusion (9). Following soil application, moisture gradually permeates the coating, dissolving the encapsulated nutrients, which are then released outward through the coating layer in a diffusion-controlled pattern. The release kinetics are primarily governed by coating thickness, membrane permeability, and ambient temperature (10, 11). This mechanism enables a sustained nutrient supply that is better synchronized with crop demand, thereby enhancing fertilizer use efficiency, curtailing nutrient losses, and alleviating environmental pollution (12). Conventional urea, however, is highly susceptible to nitrogen loss, with 20–60% of applied N being lost via ammonia (NH₃) volatilization, nitrate leaching, and denitrification—resulting in a nitrogen use efficiency (NUE) of merely 30–40% for most crop production systems (13, 14). Beyond nitrogen dissipation, urea application also contributes considerably to greenhouse gas emissions. Notably, while coated urea products have been shown to effectively mitigate N₂O emissions and NH₃ volatilization, they may concurrently incur elevated CO₂ emissions attributable to the carbon footprint embedded in coating material production (15, 16). Consequently, controlled-release blended fertilizers (CRBF) have attracted growing interest as a promising strategy to align nitrogen supply with crop uptake while simultaneously curbing these environmental burdens (17). The mechanisms behind yield enhancement and efficiency improvement of CRBF have been well-documented in major field crops like maize and rice (18–20). Studies by Xu et al. (2021) demonstrated that CRBF can improve rice yield, quality, and nitrogen use efficiency. Guo et al. (2017) found that, compared to conventional urea, a single application of CRBF not only enhanced maize yield and quality but also fostered a better maize production system. However, unlike grain crops, leaf mustard has a short growth cycle, rapid biomass accumulation, and, with leaves as the harvested organ, is more sensitive to the spatial and temporal match of nitrogen supply (3, 4). Existing research on the application of controlled-release fertilizers in leafy vegetables primarily focuses on common varieties like spinach and lettuce. There is a research gap in understanding the comprehensive effects of a single application of CRBF on tillering mustard, a variety with significant processing and economic value. Thus, investigating the impact of CRBF on the growth, nutrient uptake, and soil microbial communities of tillering mustard not only expands the application scope of CRBF but also represents a breakthrough in achieving green and efficient mustard production.

In this context, the study hypothesizes that a single application of CRBF can promote the growth and nutrient uptake of tillering mustard, improving yield and economic returns by enhancing soil physicochemical properties, boosting soil enzyme activities, and optimizing the microbial community structure, thereby achieving synergistic regulation of the fertilizer-soil-crop system. The specific objectives of this study are: (1) to evaluate the effects of single CRBF application on agronomic traits, yield, nitrogen use efficiency, and economic benefits of tillering mustard; (2) to examine the effects of CRBF application on soil physicochemical properties, enzyme activities, and microbial community structure; (3) to analyze the interactions among crop growth, soil characteristics, and microbial communities, elucidating the mechanisms through which CRBF application increases the yield of tillering mustard.

2 Materials and methods

2.1 Experimental site

A field experiment was carried out from October 2024 to May 2025 at the experimental station of the Three Gorges Academy of Agricultural Sciences, Chongqing Sanxia University of Science and Technology, located at 108.395°E, 30.762°N, in Wanzhou District, Chongqing (altitude~175 m). The site has a humid subtropical monsoon climate with a frost-free period of ~340 days. During the trial, monthly mean temperatures ranged from 8.30 °C to 18.20 °C, monthly rainfall from 18.60 mm to 107.70 mm, with 6–14 rainy days per month (Supplementary Table 1).

2.2 Experimental materials

The mustard variety tested was Yujie 7 [tillering mustard (Brassica juncea var. multiceps Tsen et Lee), selected by the Three Gorges Academy of Agricultural Sciences, Chongqing Sanxia University of Science and Technology], hereafter referred to as mustard. The fertilizers tested included conventional urea (N ≥ 46.00%, produced by Sichuan Lutianhua Co., Ltd.), polyurethane-coated urea (N ≥ 43.00%, 3-month release period, supplied by the Rubber Research Institute, Chinese Academy of Tropical Agricultural Sciences), potassium sulfate (K₂O ≥ 52.00%, produced by SDIC Xinjiang Luobupo Potassium Salt Co., Ltd.), and superphosphate (P₂O₅ ≥ 12.00%, produced by Hubei Three Gorges Ecological Fertilizer Co., Ltd.). The soil tested was brown soil with a pH of 7.60, organic matter content of 9.84 g kg^-1, alkali-hydrolyzable nitrogen of 51.60 mg kg^-1, available phosphorus (AP) of 12.90 mg kg^-1, and available potassium (AK) of 119.90 mg kg^-1.

2.3 Experimental design

A randomized block design was employed, with six treatments: no nitrogen application (CK), 100% conventional urea (U), 70% conventional urea + 30% coated urea (CRU1), 50% conventional urea + 50% coated urea (CRU2), 30% conventional urea + 70% coated urea (CRU3), and 100% coated urea (CRU4). Each treatment had three replications, and each plot had an area of 8 m². The nitrogen, phosphorus, and potassium ratios were configured according to local fertilization recommendations. The fertilization rates for all treatments were N 156.17 kg ha^-1, P₂O₅ 75.60 kg ha^-1, and K₂O 121.94 kg ha^-1. The mustard planting density was 90, 000 plants ha^-1 with a spacing of 33 cm × 33 cm. Plots were separated by drainage ditches and guard rows. Mustard was raised in plug trays on September 25, 2024, and vigorous seedlings were transplanted on October 18. All phosphorus and potassium fertilizers were applied once as basal fertilizer before land preparation. For the U treatment, 50% of the nitrogen fertilizer was applied as basal fertilizer, while 30% and 20% were top dressed at the rosette initiation stage (November 15) and tillering stage (December 16), respectively. For the CRBF treatments, all nitrogen fertilizers were applied as a single basal fertilizer application before land preparation. Field management, including pest, disease, and weed control, was consistently maintained throughout the mustard growth period.

2.4 Index measurement

2.4.1 Mustard growth index and yield measurement

Ten mustard plants were selected from each plot as observation plants. Agronomic traits, including plant height, leaf length, leaf width, and tiller count, were regularly measured with a ruler at various stages: Rosette Stage (RS; 20 days after transplanting), Peak Rosette Stage (PRS; 35 days after transplanting), Tillering Stage (TS; 50 days after transplanting), Middle Tillering Stage (MTS; 69 days after transplanting), Peak Tillering Stage (PTS; 77 days after transplanting), and Harvest Period (93 days after transplanting). At harvest, all plants in each plot were harvested, weighed, and the yield per plot was determined.

2.4.2 Determination of mustard leaf SPAD value

The SPAD value of the 3rd fully expanded functional leaf (counting from the top) of mustard was determined with a SPAD-502 chlorophyll meter at RS, PRS, TS, MTS and PTS. (Konica Minolta, Tokyo, Japan). Ten plants per plot were selected for measurement.

2.4.3 Determination of mustard leaf photosynthetic parameters

At the PTS stage, the net photosynthetic rate (Pn), stomatal conductance (Gs), intercellular CO₂ concentration (Ci) and transpiration rate (Tr) were determined on the 3rd fully expanded functional leaf (counting from the top) of tillering mustard using a Li-6400XT portable photosynthesis system (LI-COR Inc., USA). Measurements were performed with an open gas flow and a fixed light intensity of 1500 lx.

2.4.4 Determination of nitrogen metabolism enzyme activities in mustard leaves

At maturity, functional leaves (the fully expanded third leaf from the top) were collected from five plants per plot, immediately frozen in liquid nitrogen, and stored at –80 °C for the analysis of nitrogen metabolism enzyme activities, including glutamine synthetase (GS), NADH-glutamate synthase (NADH-GOGAT), glutamate dehydrogenase (GDH), and NR. The analysis was conducted by Suzhou Comin Biotechnology Co., Ltd (21).

2.4.5 Determination of plant nitrogen content and nitrogen use efficiency

At maturity, ten plants per plot were collected. After oven-drying at 105 °C for 30 minutes to deactivate enzymes, the plants were further dried at 75 °C to a constant weight. The dried samples were ground, digested with concentrated sulfuric acid and hydrogen peroxide, Total nitrogen (TN) content was determined by a continuous flow analyzer. Nitrogen recovery efficiency (NRE) was calculated using the following formula (Equation 1):

(1) NRE=(N−N0)/FN

where N represents the aboveground TN uptake in the nitrogen application treatment, N0 represents the uptake in the no-nitrogen treatment, and FN is the nitrogen application rate (22).

2.4.6 Determination of soil physicochemical properties

At maturity, 0–20 cm soil samples were collected from each plot using the five-point sampling method. The collected soil samples were divided into three portions: one portion was passed through a 2 mm sieve to remove residual roots and gravel, then stored at –80 °C for soil microbial sequencing; another portion was stored at –20 °C. Soil nitrate nitrogen (NO₃^--N) and ammonium nitrogen (NH₄⁺-N) were extracted from fresh soil with 1 mol L^-1 KCl and determined using a continuous flow analyzer (AA3) (23). Soil organic matter (SOM) was measured by the potassium dichromate oxidation-volumetric method with oil bath heating using air-dried and sieved soil (24). Total nitrogen (TN) was determined by the Kjeldahl method after digesting the air-dried soil with concentrated sulfuric acid and a catalyst (25). Available phosphorus (AP) was extracted from air-dried soil with 1 mol L^-1 NaHCO₃ and measured with a spectrophotometer at 700 nm (26). Available potassium (AK) was extracted from air-dried soil with 1 mol L^-1 ammonium acetate and determined by flame photometry (27). Soil pH was measured by weighing 10 g of air-dried soil, adding 25 mL of deionized water, stirring thoroughly, allowing the suspension to stand for 30 min, and then using a glass electrode method (28).

The activities of urease (UE), sucrase (SC), nitrate reductase (NR), acid phosphatase (ACP), polyphenol oxidase (PPO), and catalase (CAT) were determined by weighing 1 g of air-dried soil, adding the corresponding substrate solutions, incubating for 24 h, and then using the indophenol blue colorimetry (29), 3, 5-dinitrosalicylic acid colorimetry (30), phenoldisulfonic acid colorimetry (31), disodium phenyl phosphate colorimetry (32), pyrogallol colorimetry (33), and potassium permanganate titration (34), respectively. The activities of β-1, 4-glucosidase (BG), β-1, 4-N-acetylglucosaminidase (NAG), and L-leucine aminopeptidase (LAP) were assayed using 1 g of fresh soil with p-nitrophenyl-β-D-glucoside (pNPG) (35), p-nitrophenyl-β-D-N-acetylglucosaminide (pNP-NAG) (36), and L-leucine-p-nitroanilide (L-Leu-pNA) (37) as substrates, respectively. After incubation for 1 h, the absorbance was measured with a microplate reader at 405 nm.

2.4.7 Soil microbial DNA extraction, PCR amplicon sequencing, and library construction

Using a Soil Microbial Genomic DNA Extraction Kit (Tiangen Biotech (Beijing) Co., Ltd.), microbial DNA was extracted from 18 soil samples collected from the mustard harvest stage. Each DNA extract was then quantified/measured in triplicate. The bacterial 16S rRNA gene (V3-V4 region) was amplified using specific primers 341F (CCTAYGGGRBGCASCAG) and 806R (GGACTACNNGGGTATCTAAT), and the fungal ITS region was amplified using primers ITS1-1F-F (CTTGGTCATTTAGAGGAAGTAA) and ITS1-1F-R (GCTGCGTTCTTCATCGATGC) (38). PCR amplification was performed as follows: all PCR mixtures contained 15 µL of Phusion High-Fidelity PCR Master Mix, 0.2 µM primers, and 10 ng genomic DNA template. The amplification protocol was: initial denaturation at 98 °C for 1 minute, followed by 30 cycles of denaturation at 98 °C (10 s), annealing at 50 °C (30 s), and extension at 72 °C (30 s), with a final extension at 72 °C for 5 minutes. The PCR products were purified using magnetic beads, pooled in equal amounts based on concentration, and the target bands were detected and recovered. After purification and quantification, the PCR products were subjected to paired-end sequencing on the Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Raw sequences were quality-filtered, denoised using the DADA2 method, and amplicon sequence variants (ASVs) were generated using the QIIME2 platform. Species annotation was performed based on the SILVA 138.1 (bacteria) and UNITE 9.0 (fungi) databases. Subsequent analyses were performed after randomly rarefying valid sequences to the same depth by Novogene Bioinformatics Technology Co., Ltd.

2.4.8 Economic benefit calculation

Economic benefit and its components were calculated as economic output value minus production cost. The economic output value was obtained by multiplying the yield of each treatment by the selling price and a loss coefficient of 0.95. The selling price of tuber mustard was ¥1.20 kg^-¹. Total production costs consisted of expenditures on seeds, fertilizers and labor.

The market prices of fertilizers were ¥ 2.30 kg^-¹ for conventional urea, ¥ 3.80 kg^-¹ for coated urea, ¥ 1.00 kg^-¹ for superphosphate, and ¥ 3.80 kg^-¹ for potassium sulfate. The cost of tuber mustard seeds was ¥ 100.00 ha^-¹. Labor costs were 1200.00 ha^-¹ for land preparation, ¥ 500.00 ha^-¹ for fertilization, ¥ 500.00 ha^-¹ for sowing, ¥ 500.00 ha^-¹ for pesticide application, and ¥ 500.00 ha^-¹ for harvesting. All prices corresponded to the local market rates in the experimental year.

2.5 Data analysis

Data organization and table creation were performed using Excel 2016 (Microsoft), and statistical analysis was conducted with SPSS 22.0 (SPSS Institute Inc., United States). Agronomic traits, leaf enzyme activities, photosynthetic parameters, soil physicochemical properties, and soil enzyme activities were analyzed using one-way analysis of variance, followed by Duncan’s multiple range test for significance (p < 0.05). Significant differences in mean values were indicated by different lowercase letters (e.g., a, b, c). Graphs were created using Origin 2021 (OriginLab Corporation, United States). Principal coordinate analysis (PCoA), Mantel test, and partial least squares path modeling (PLS-PM) were performed and visualized using R (ver. 4.5.1).

3 Results

3.1 Effects of different fertilizer treatments on agronomic traits of mustard

At TS, the CRU3 treatment exhibited superior growth parameters compared to the U treatment, whereas the CRU4 treatment showed reductions of 14.70%, 13.84%, 11.84%, and 12.22% in plant height, leaf width, leaf length, and tiller number, respectively (p < 0.05). At harvest, all agronomic traits followed a unimodal response pattern as the proportion of coated urea increased, peaking in CRU3 (Figure 1). Compared to the U treatment, the CRU3 treatment resulted in significant increases of 7.76%, 8.08%, 11.97%, and 9.09% in plant height, leaf length, leaf width, and tiller number, respectively (p < 0.05). These results indicate that the 30:70 blending ratio optimized vegetative growth, whereas CRU4 (100% coated urea) resulted in reduced growth parameters, possibly due to insufficient early-season nitrogen availability.

3.2 Effects of different fertilizer treatments on nitrogen metabolism enzyme activities in leaves

Leaf GS activity initially decreased, then increased with the increasing proportion of coated urea, with the lowest value recorded in the CRU2 treatment (Figure 2B). As presented in Figures 2C, D, leaf NADH-GOGAT and GDH activities both initially increased and then decreased as the proportion of coated urea rose from 30% to 100%, with the highest activities observed in the CRU3 treatment. Compared to the CRU1, CRU2, and CRU4 treatments, NADH-GOGAT activity in the CRU3 treatment increased by 4.88%, 39.59%, and 21.75%, respectively, while GDH activity increased by 3.93%, 15.06%, and 9.89%, respectively. The significantly higher NADH-GOGAT and GDH activities in CRU3 than in CRU1 and CRU4 (p < 0.05) indicate that CRU3 enhanced the ammonium assimilation capacity, which is critical for efficient nitrogen utilization in mustard (Figure 2).

3.3 Effects of different fertilizer treatments on soil physicochemical properties and enzyme activities

Figure 3A shows no significant differences in SOM, AK, and AP among the nitrogen-applied treatments (p > 0.05), indicating that the CRBF formulations did not substantially alter the baseline soil fertility parameters. The NO₃^--N content in the CRU1 treatment was the highest, with an increase of 82.21% to 1247.37% compared to other treatments (p < 0.05). The NH₄⁺-N content in the CRU1, CRU2, and CRU3 treatments was significantly higher than in the CK, U, and CRU4 treatments (p < 0.05), with the CRU3 treatment showing a 61.25% increase compared to the U treatment. Figure 3B shows that CAT activity was significantly higher in the CRU3 treatment compared to other treatments, with an increase of 20.16% relative to the U treatment (p < 0.05). UE activity in the CRU3 treatment was significantly higher than in the U, CK, and CRU2 treatments, with a 55.56% increase relative to the U treatment (p < 0.05). BG and NAG activities were significantly higher in the U and CRU1 treatments than in other treatments (p < 0.05). The ACP and SC activities in the CRU1 treatment were also significantly higher than those in other treatments (p < 0.05). However, no significant differences in PPO activity were observed among the treatments (p > 0.05). The elevated UE activity in CRU3 suggests enhanced nitrogen transformation potential, while the increased CAT activity reflects more active overall soil microbial metabolism under this treatment (Figure 3B).

3.4 Effects of different fertilizer treatments on soil microorganisms of mustard

3.4.1 Soil microbial alpha diversity

As shown in Table 1, compared to the U treatment, the observed features index of soil bacteria in the CRU3 and CRU4 treatments increased by 10.96% and 3.76%, respectively (p < 0.05). The bacterial Chao1 index in the CRU3 and CRU4 treatments increased by 11.79% and 4.11%, respectively, compared to the U treatment, but the differences among treatments were not significant (p > 0.05). Nitrogen application reduced the alpha diversity indices of soil fungi. The Chao1, observed features, Shannon, and Simpson indices in the CRU1 treatment were significantly lower by 53.34%, 62.62%, 41.93%, and 54.29%, respectively, compared to the CK treatment (p < 0.05). Notably, the fungal Chao1 index and Observed features in CRU3 increased by 22.79% and 19.06% compared with U, suggesting that this formulation (CRU3) supported a more diverse fungal community than conventional urea alone (Table 1). However, no significant differences in fungal indices were detected between any CRBF treatment and U (p > 0.05). The higher bacterial and fungal richness observed in CRU3 indicates that a balanced nitrogen supply from this formulation created a more favorable soil environment for microbial communities.

Table 1. The effect of different fertilization treatments on the alpha diversity of mustard soil microorganisms.

The data are from a randomized complete block design with six treatments: CK (no nitrogen), U (100% conventional urea, split-applied), CRU1 (70% conventional urea + 30% coated urea), CRU2 (50% conventional urea + 50% coated urea), CRU3 (30% conventional urea + 70% coated urea), CRU4 (100% coated urea). Soil samples were collected at harvest (0–20 cm depth). The values are expressed as mean ± SE (n = 3). Different lowercase letters within the same column indicate significant differences at the level of p < 0.05 according to one-way ANOVA followed by Duncan’s multiple range test. Chao1: an index used to estimate the total number of species (including undetected species) in a community, reflecting species richness; Observed features: the actual number of observed species (OTUs or ASVs) in a sample, directly reflecting species richness; Shannon index: a comprehensive index that considers both species richness and evenness, with higher values indicating greater species diversity; Simpson index: an index reflecting the probability that two individuals randomly selected from a community belong to the same species, with higher values indicating lower diversity. In this study, the Simpson index values approach 1, indicating high species evenness across all samples.

3.4.2 Soil microbial beta diversity

As presented in Figures 4A, B, PCoA explained 71.35% (PCo1: 52.66%, PCo2: 18.69%) of the variation in bacterial communities and 94.52% (PCo1: 75.64%, PCo2: 18.88%) of the variation in fungal communities, effectively capturing the differences in both soil bacterial (p = 0.0060) and fungal (p = 0.0100) communities. The bacterial communities in the CK, U, and CRU3 treatments overlapped considerably, indicating similar community structures with relatively minor differences. For fungi, the CRU3 and CK treatments showed partial separation along the PCo2 axis.

3.4.3 Soil microbial community composition

As shown in Figure 4C, Nitrogen application increased the relative abundance of Proteobacteria while decreasing that of Acidobacteriota compared to the CK treatment. No significant changes were observed in the relative abundances of the dominant bacterial phyla (Proteobacteria, Acidobacteriota, Actinobacteriota, Gemmatimonadota, Chloroflexi) in the CRBF treatments compared to the U treatment. However, the relative abundance of Crenarchaeota in the CRU1, CRU2, CRU3, and CRU4 treatments increased by 48.03%, 89.37%, 96.85%, and 24.01%, respectively. As shown in Figure 4D, the dominant fungal phyla were Ascomycota (32.50%–60.61%), Olpidiomycota (21.33%–42.58%), and Mortierellomycota (2.97%–11.73%). Compared to the U treatment, the abundance of Ascomycota in the CRU1 treatment decreased by 13.18%, whereas it increased by 14.93% in the CRU3 treatment. The abundance of Olpidiomycota in the CRU1 treatment increased by 21.34% compared to the U treatment, but decreased by 17.56% in the CRU3 treatment.

3.4.4 Environmental drivers of soil microbial community structure

As presented in Figure 5, soil ACP had a highly significant positive correlation with NO₃^--N and TN, while AK was strongly positively correlated with SOM. In contrast, pH exhibited a highly significant negative correlation with TN and ACP. The Mantel test indicated that NR, pH, and SC were key drivers of the bacterial community structure, while the fungal community structure was primarily influenced by NO₃^--N, BG, NAG, and ACP.

3.5 Effects of different fertilizer treatments on nitrogen use efficiency and economic benefits of mustard

The CRU4 treatment yielded the lowest output, significantly lower than all other nitrogen-applied treatments (p < 0.05). In comparison to the U treatment, the CRU1, CRU2, and CRU3 treatments all improved mustard nitrogen use efficiency. The CRU3 treatment achieved a nitrogen use efficiency of 67.91%, significantly higher by 13.96% compared to the U treatment (p < 0.05). In contrast, the nitrogen use efficiency in the CRU4 treatment was the lowest, significantly reduced by 20.89% compared to the U treatment (p < 0.05). Although production costs in CRBF treatments increased with coated urea proportion (from ¥ 7, 290 to ¥ 7, 730 ha^-¹), the economic output and net benefit were highest in CRU3, with significant increases of 6.66% and 8.11% compared to U (p< 0.05) (Table 2). The benefit-cost ratio (BCR) followed the same trend, with CRU3 achieving the highest value (10.90), 16.2% higher than U (9.38). These results establish that the CRU3 formulation achieves concurrent improvements in agronomic and economic performance.

Table 2. Effects of different fertilizer treatments on yield and economic benefit in mustard.

The data are from a randomized complete block design with six treatments: CK (no nitrogen), U (100% conventional urea, split-applied), CRU1 (70% conventional urea + 30% coated urea), CRU2 (50% conventional urea + 50% coated urea), CRU3 (30% conventional urea + 70% coated urea), CRU4 (100% coated urea). The values are expressed as mean ± SE (n = 3). Different lowercase letters within the same column indicate significant differences at the level of p< 0.05 according to one-way ANOVA followed by Duncan’s multiple range test. NRE: nitrogen recovery efficiency, calculated as (N - N₀)/FN × 100%, where N is aboveground N uptake in fertilized plots, N0 is N uptake in no-N control, and FN is N application rate. Economic Output Value = Yield × Price × 0.95 (loss coefficient). Production Cost includes seeds, fertilizers, and labor. Economic Benefit = Economic Output Value − Production Cost. BCR: Benefit-Cost Ratio, calculated as Economic Benefit/Production Cost.

3.6 Structural equation modeling analysis

To investigate the regulatory effects of CRBF on soil physicochemical properties, enzyme activities, soil microbial community structure, crop growth, and nitrogen uptake, a partial least squares path modeling (PLS-PM) was constructed. As shown in Figure 6, the model had a GOF of 0.4759, indicating a good fit. The structural equation model demonstrated that CRBF directly influenced soil nutrients (PC = 0.381), promoted mustard growth (PC = 1.624, p < 0.05), which in turn enhanced nitrogen uptake by mustard (PC = 0.792, p < 0.05), ultimately leading to an increase in mustard yield (PC = 0.531, p < 0.05).The path coefficients indicate that plant growth served as the primary direct driver of nitrogen uptake, while soil nutrients played a mediating role in the pathway from CRBF application to plant growth, nitrogen uptake, and yield (Figure 6).

4 Discussion

4.1 Effects of different fertilizer treatments on mustard growth, nitrogen use efficiency, and economic benefits

Crop growth status serves as a direct indicator of fertilization effectiveness (39). This study demonstrated that the CRU3 treatment (30% conventional urea + 70% coated urea) most significantly promoted mustard growth, with superior agronomic traits compared to other fertilization treatments (Figure 1). The observed growth enhancement is closely tied to optimal nitrogen nutrition. However, Guo et al. (2017) reported that a blending ratio of 40% conventional urea to 60% coated urea achieved the best yield increase in maize, which differs from our findings. This discrepancy may be attributed to the fact that leafy vegetables have a continuous nitrogen demand, whereas cereal crops exhibit a peak-demand pattern, suggesting that the crop-specific nitrogen demand pattern determines the optimal blending ratio of controlled-release blended fertilizer (CRBF).

Nitrogen is not only a vital component of chlorophyll, amino acids, and proteins but also a core nutrient for photosynthesis (40). As a constituent of chlorophyll molecules and key photosynthetic enzymes, nitrogen promotes the synthesis of chlorophyll and photosynthetic proteins (41). At PTS, mustard leaves under the CRU3 treatment exhibited the highest SPAD values (29.40; Supplementary Table 2) and Pn (23.66 μmol m^-² s^-¹; Supplementary Table 3). Compared with U, CRU3 also showed a concurrent increase in Gs (58.41%) and decrease in Ci, suggesting that the photosynthetic enhancement may be associated with improved stomatal conductance and CO₂ assimilation capacity under this treatment. This physiological foundation facilitated later-stage material accumulation in mustard.

Furthermore, the CRU3 treatment boosted the activities of key nitrogen metabolism enzymes in leaves (Figure 2). Notably, NADH-GOGAT and GDH activities in CRU3 were significantly higher than in CRU1 and CRU4 (p < 0.05), with increases of 21.75% and 9.89%, respectively (Figures 2C, D). These enzymes played a central role in assimilating NH₄⁺-N in plants (42), the primary form of nitrogen utilized by plants. Absorbed NO₃^--N must first be converted to NH₄⁺-N prior to assimilation (43). The elevated NADH-GOGAT and GDH activities under CRU3 indicated enhanced ammonium assimilation capacity, which likely contributed to improved nitrogen utilization efficiency (44). This inference was supported by the significantly higher NUE in CRU3 (67.91%) relative to U (53.95%) (Table 2).

Mixing controlled-release fertilizer with conventional urea is a common strategy to coordinate nitrogen supply throughout the crop growing season (45). The resulting temporal complementarity was supported by our SPAD and photosynthetic data across growth stages (Supplementary Tables 2, S3). This study found that mustard yield, nitrogen use efficiency, and economic benefits initially increased and then decreased with increasing proportions of controlled-release nitrogen. The yield in the CRU3 treatment was significantly higher by 6.66%, 7.10%, 4.99%, and 27.54% compared to the U, CRU1, CRU2, and CRU4 treatments, respectively. NUE and economic benefits in CRU3 increased by 13.96% and 8.11%, respectively, compared to the U. In contrast, the CRU4 (100% coated urea) treatment showed the lowest yield, NUE, and economic benefits, due to insufficient early-stage nitrogen supply as its release rate failed to match mustard’s initial demand (Supplementary Table 2), inhibiting both growth and nitrogen uptake. The BCR analysis further confirmed the economic advantage of CRU3, which achieved the highest ratio (10.90), 16.2% above that of U (9.38). This indicates that the additional investment in coated urea for CRU3 was more than offset by the yield gain, whereas CRU4 incurred the highest cost but delivered the lowest return among nitrogen-applied treatments.

The non-linear response across CRU1 to CRU4 treatments, where performance peaked at CRU3 and declined at both lower and higher coated urea proportions, reveals a critical “optimal synchrony window” for mustard. At the lower end of the spectrum, CRU1 (30% coated urea) failed to sustain late-season nitrogen supply, as evidenced by declining SPAD values at MTS and PTS. At the higher end, CRU4 (100% coated urea) could not meet early nitrogen demand, showing the lowest SPAD values at RS and PRS. These contrasting patterns highlight that the ideal formulation is not simply the one with the highest controlled-release proportion, but rather the one that best matches the crop’s developmental nitrogen demand curve.

In conclusion, the single application of CRBF (30% conventional urea + 70% coated urea) can promote mustard growth, enhance photosynthetic efficiency and leaf nitrogen metabolism, improve nitrogen use efficiency, increase yield and economic benefits, and reduce labor requirements.

4.2 Effects of different fertilizer treatments on soil physicochemical properties and enzyme activities

CRBF not only promote crop growth but also enhance soil conditions, thereby indirectly benefiting crop productivity. Soil physicochemical properties and enzyme activities are key indicators for assessing soil quality (46). Coated urea sustains a stable nitrogen supply in the soil over an extended period, effectively reducing nitrogen losses, preserving soil nitrogen pools, and increasing crop yields (47–49). This highlights the importance of synchronizing nitrogen release with crop demand (20).

The application of CRBF significantly improves various indicators related to soil nitrogen supply capacity. In this study, the CRU3 treatment increased NH₄⁺-N content by 61.25% compared to the U treatment (Figure 3A), which aligned with the higher activities of NADH-GOGAT and GDH in mustard leaves (Figures 2C, D), suggesting that plant enzyme activities are influenced by soil nutrient status (50). In contrast, CRU4 (100% coated urea) did not increase soil NH₄⁺-N, likely because its slow release rate failed to meet the crop’s early nitrogen demand (0–35 DAT), resulting in net NH₄⁺ depletion due to vigorous plant uptake (Supplementary Table 2). CRU3, by contrast, supplied sufficient available N for early growth while the 70% coated urea component sustained release through later stages, leaving residual NH₄⁺ at harvest. Soil enzyme activity serves as a sensitive indicator of microbial metabolic activity and nutrient cycling capacity (51). In CRU3, CAT and UE activities were significantly higher than in U (Figure 3B), indicating enhanced soil microbial activity and improved nitrogen transformation capacity (52). CAT activity increased by 20.16%, while UE activity increased by 55.56% compared to U (Figure 3B). The higher UE activity in CRU3 suggests accelerated urea hydrolysis and ammonium generation, which was consistent with the elevated NH₄⁺-N content observed in this treatment. This urease enhancement likely reflects a substrate-induced response, whereby the gradual release of urea from coated granules provided a sustained, moderate substrate supply that avoided the transient urease inhibition triggered by the high NH₃ concentrations from conventional urea application. Although enhanced urease activity theoretically promotes urea hydrolysis and could potentially increase local NH₃ concentration, the overall risk of ammonia volatilization in CRU3 remains lower than that in the U treatment. This is because only 30% of the total N in CRU3 was supplied as conventional urea, whereas the remaining 70% was coated urea with a slow release pattern. Consequently, the total amount of urea substrate available for rapid hydrolysis at any given time was substantially less than that in the U treatment (100% conventional urea). Moreover, the gradual release from coated granules prevented the formation of excessively high localized NH₃ concentrations, which are a primary driver of volatilization losses in conventional urea applications (53). Therefore, despite the higher urease activity, the overall ammonia volatilization potential under CRU3 is expected to be lower, contributing to the improved nitrogen recovery efficiency observed in this treatment. Meanwhile, the elevated CAT activity reflects enhanced soil antioxidant capacity, which alleviates oxidative stress on microorganisms and supports more active overall soil metabolism.

The Mantel test (Figure 5) revealed a significant negative correlation between pH and ACP activity as well as TN, suggesting that increased pH may inhibit ACP activity, thereby reducing phosphorus activation in plant-fungal interactions and potentially increasing ammonia volatilization (54). The structural equation model (Figure 6) further confirmed that soil nutrients were the most direct factor influencing mustard growth (with the highest path coefficient). CRBF directly impacted soil nutrients (PC = 0.381), promoted mustard growth (PC = 1.624, p < 0.05), which in turn enhanced nitrogen uptake (PC = 0.792, p < 0.05), and ultimately boosted mustard yield (PC = 0.531, p < 0.05), consistent with the findings of Cai et al. (2019) (55).

4.3 Effects of different fertilizer treatments on soil microbial community structure

Soil microorganisms play a critical role in linking fertilizer input with crop response (56). This study demonstrated that the soil microbial community in the CRU3 was more diverse and stable. The Observed features and Chao1 indices for bacteria were increased by 10.96% and 11.79%, respectively, compared to the U treatment (Table 2), while the corresponding indices for fungi increased by 22.79% and 19.06%. In contrast, microbial diversity declined in some treatments (e.g., CRU1, CRU2), likely due to the rapid hydrolysis of conventional urea upon application, resulting in localized high concentrations of NH₃ and ammonium toxicity (57). However, CRU3, with its balanced combination of readily available and controlled-release nitrogen, may have provided a more moderate nitrogen supply, which likely avoided the extreme NH₃ peaks associated with conventional urea and contributed to the higher microbial richness observed in this treatment. The weaker microbial response in CRU4 (100% coated urea) versus CRU3 can be attributed to its slower N release, which delayed early root growth and microbial colonization and failed to match crop peak demand. In contrast, CRU3’s optimal blend of immediate and slow-release N provided both early stimulation and sustained supply, supporting greater microbial diversity and plant-microbe synergies. Similar findings were reported by Chen et al. (2024) (58) using lignin-based controlled-release urea, which enhanced soil N availability, enzyme activities, and microbial diversity. Together, these results indicate that a balanced coated/uncoated ratio, rather than full coating, is more effective in synchronizing N supply with both microbial and crop needs.

PCoA analysis (Figures 4A, B) further revealed that the microbial community structure in the CRU3 closely resembled that of CK, indicating minimal disturbance to the original microbial community (59). The bacterial communities in CK, U, and CRU3 overlapped considerably (Figure 4A), suggesting that the CRU3 did not drastically alter the bacterial community composition compared to the unfertilized control. Compared to U, the CRBF treatments did not significantly alter the relative abundances of dominant soil bacterial phyla (Proteobacteria, Acidobacteriota, Actinobacteriota, Gemmatimonadota, Chloroflexi; Figure 4C) but notably increased the relative abundance of Crenarchaeota, with the CRU3 treatment showing a 96.85% increase (Figure 4C). Crenarchaeota, a core group of soil archaea, plays a key role in soil ammonia oxidation (60). In our study, under the CRU3 treatment, the abundance of Crenarchaeota and soil NH₄⁺ content exhibited a synchronous increasing trend. The synchronous increase can be explained by two non-mutually exclusive mechanisms. First, the sustained release of urea from the 70% coated fraction maintained a continuous NH₄⁺ supply, while the timing of soil sampling (at harvest, 93 DAT) coincided with a period when crop N uptake had ceased. This created a net positive balance, where the rate of NH₄⁺ generation exceeded its consumption by nitrification and immobilization. Second, Crenarchaeota proliferation may reflect a long-term community adaptation to the stable, low-ammonium environment created by CRU3, rather than an immediate increase in per-capita nitrification activity. In fact, their dominance under such conditions may have limited the conversion of NH₄⁺ to NO₃^-, thereby contributing to the maintenance of a larger soil NH₄⁺ pool. This interpretation is consistent with the high substrate affinity of archaeal ammonia monooxygenase (AMO), which enables AOA to thrive under low NH₃ conditions without causing a rapid net decline in the NH₄⁺ pool when supply is continuously replenished (61). However, given that Crenarchaeota is a phylum-level taxonomic group encompassing a large number of non-ammonia-oxidizing lineages, an increase in its abundance cannot be directly interpreted as an enhancement of soil ammonia oxidation capacity (62, 63). To further elucidate the underlying mechanisms, future studies should employ targeted amplification of the amoA functional gene to separately quantify ammonia-oxidizing archaea (AOA), ammonia-oxidizing bacteria (AOB), and comammox bacteria abundances, alongside measurements of potential nitrification rate (PNR) and high-frequency temporal sampling across the growing season. In contrast, the slower release from CRU4 failed to establish a similarly favorable niche for Crenarchaeota, as the prolonged N limitation during early growth likely constrained root development and reduced rhizodeposition, thereby limiting the substrate and energy supply for microbial proliferation.

Additionally, compared to the U treatment, the relative abundance of Olpidiomycota in the CRU3 treatment was significantly reduced by 17.56% (Figure 4D). Olpidiomycota, a group of pathogenic fungi that infect Brassicaceae roots (64), can cause root necrosis and lodging (65). The suppression of this potential pathogen under CRU3 likely contributed to improved root health, as reflected by enhanced aboveground biomass and yield (Figure 1; Table 2). This finding suggests that, beyond direct N nutrition, modulation of soil microbial communities represents an additional mechanism through which optimized fertilizer management improves crop performance, consistent with previous reports that controlled-release fertilizers can indirectly suppress soil-borne pathogens by reshaping rhizosphere microbial community structure and enhancing plant vigor (66).

Although the CRBF did not significantly alter the relative abundances of dominant bacterial phyla, it may still indirectly impact crop growth by modulating the activity of specific functional groups (e.g., Crenarchaeota) involved in soil nitrogen transformation and supply (67).While these results demonstrate the clear benefits of the CRU3 formulation under the conditions tested, the single-season, single-site nature of this experiment warrants caution. Future multi-site, multi-year trials, combined with metagenomic and functional activity assays, are required to validate the generalizability and mechanistic basis of these microbial community shifts.

5 Conclusion

The single application of a CRBF with a ratio of 30% conventional urea to 70% coated urea successfully synchronizes nitrogen supply with the nitrogen demand of mustard, while also yielding the greatest enhancements in both agronomic characteristics and physiological performance. This fertilizer regimen significantly increased leaf photosynthetic efficiency and nitrogen metabolism enzyme activities, accompanied by elevated soil ammonium nitrogen content and enhanced catalase and urease activities. It also optimizes soil microbial community structure, with increased relative abundance of phyla such as Crenarchaeota and reduced abundance of potentially harmful groups like Olpidiomycota, although the functional significance of these shifts requires further investigation. These changes coincided with improved nutrient availability and plant physiological status, which collectively contributed to enhanced mustard yield, nitrogen recovery efficiency, and net economic benefit. Thus, the single application of a CRBF with a ratio of 30% conventional urea to 70% coated urea representing an effective nitrogen management model for mustard production, with potential implications for sustainable nutrient management in leafy vegetable production.

Nevertheless, the conclusions of this study are subject to an important limitation: the field experiment was conducted at a single site over only one growing season. The long-term environmental fate of the coated urea, the functional stability of the observed microbial community shifts, and the consistency of these responses under varying soil textures, climatic conditions, and mustard cultivars remain to be determined. Consequently, multi-site and multi-season trials are essential to validate the generalizability of these findings before large-scale agronomic recommendations can be made. Despite these limitations, our work still offers a mechanistic basis for developing simplified single application fertilization strategies in intensive leafy vegetable systems.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsoil.2026.1916307/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Tariq Aziz, University of Agriculture, Pakistan

Reviewed by: Fan Zhang, Yan’an University, China

Clayton Baravelli De Oliveira, Federal Institute of São Paulo, Brazil

Footnotes

NH₃, Ammonia; NUE, Nitrogen use efficiency; CRBF, Controlled-release blended fertilizer; RS, Rosette stage; PRS, Peak rosette stage; TS, Tillering stage; MTS, Middle tillering stage; PTS, Peak tillering stage; HP, Harvest period; Pn, Net photosynthetic rate; Gs, Stomatal conductance; Ci, Intercellular CO₂ concentration; Tr, Transpiration rate; GS, Glutamine synthetase; NADH-GOGAT, NADH-glutamate synthase; GDH, Glutamate dehydrogenase; NR, Nitrate reductase; NRE, Nitrogen recovery efficiency; NO₃^--N, Nitrate nitrogen; NH₄⁺-N, Ammonium nitrogen; SOM, Soil organic matter; TN, Total nitrogen; AP, Available phosphorus; AK, Available potassium; UE, Urease; SC, Sucrase; ACP, Acid phosphatase; CAT, Catalase; PPO, Polyphenol oxidase; BG, β-1, 4-glucosidase; NAG, β-1, 4-N-acetylglucosaminidase; LAP, L-leucine aminopeptidase; PCoA, Principal coordinate analysis.