Split nitrogen fertilizer application improved grain yield in winter wheat (Triticum aestivum L.) via modulating antioxidant capacity and 13C photosynthate mobilization under water-saving irrigation conditions

A water-saving cultivation technique of supplementary irrigation based on soil moisture levels has been adopted for winter wheat production in the Huang-Huai-Hai Plain of China, due to the enhanced water-use efficiency. However, appropriate split nitrogen management may further improve crop growth and grain yield. Here, we conducted a 2-year field experiment to determine if split nitrogen management might improve wheat productivity by enhancing 13C photosynthate mobilization and the antioxidant defense system under water-saving conditions. Split nitrogen management involved a constant total nitrogen rate (240 kg ha−1) split in four different proportions between sowing and jointing stage, i.e., 10:0 (N1), 7:3 (N2), 5:5 (N3), and 3:7 (N4). The N3 treatment significantly enhanced “soil-plant analysis development” values, superoxide dismutase antioxidant activity, soluble protein content, sucrose content, and sucrose phosphate synthetase activity, although it reduced the accumulation of malondialdehyde (MDA). The N3 treatment ultimately increased the amount of dry matter assimilation after anthesis significantly. In addition, the 13C isotope tracer experiment revealed that the N3 treatment promoted the assimilation of carbohydrates after anthesis and their partitioning to the developing grains. Compared to the unequal ratio treatments (N1, N2, and N4), the equal ratio treatment (N3) increased grain yield by 5.70–16.72% via increasing 1000-grain weight and number of grains per spike in both growing seasons. Therefore, we recommend the use of a 5:5 basal-topdressing split nitrogen fertilizer application under water-saving irrigation conditions to promote antioxidant enzyme activity and the remobilization of photosynthate after anthesis for improving wheat grain yield.


Introduction
Winter wheat is a widely cultivated crop in the Huang-Huai-Hai Plain of China (HPC), a very important area of wheat production in China (He et al. 2017). The identification of methods to increase wheat grain yield in HPC is increasingly crucial for grain productivity and food security in China (Liu et al. 2020a). Water is an important factor for wheat development and grain yield. However, recurrent water shortages in the region of the HPC, mainly due to climate change and increased water demand, threatens the sustainability of winter wheat production. Therefore, reasonable irrigation management recommendations are urgently needed.
Traditionally, farmers in the region of the HPC irrigate winter wheat by flooding four to six times during the growth season, for a total irrigation water use of 300-400 mm. However, serious water consumption affects the sustainable development of this irrigation method in the HPC (Dar et al. 2017;Mostafa et al. 2018;Jha et al. 2019). Methods that reduce irrigation water consumption require urgent attention. Our research group previously reported a water-saving cultivation technique of supplementary irrigation based on measuring soil moisture at critical stages. Compared to traditional flood irrigation, this reported technique maintained high grain yield and water use efficiency and decrease water consumption (Man et al. 2014). Our previous research also showed that this irrigation method can promote wheat root growth to a greater soil depth and, consequently, improve the absorption and utilization of soil moisture (Guo et al. 2015). Compared with flood irrigation, the total irrigation water amount was reduced by 200-300 mm under this water-saving irrigation technique .
As nitrogen is an essential macronutrient for wheat growth and yield, fertilizer management is generally the most effective way to increase grain yield in winter wheat production (Azam et al. 2020). However, unreasonable nitrogen fertilizer application has led to serious environmental problems, higher fertilizer loss, and lower returns for farmers (Singh et al. 2014;Tan et al. 2017). Therefore, appropriate nitrogen fertilizer management is essential for a sustainable strategy to improve crop yield while reducing environmental risk (Trost et al. 2016). An appropriate basal-topdressing ratio of nitrogen application is the most important component of nitrogen fertilizer management . In many areas of the world, farmers commonly tend to apply nitrogen fertilizer at pre-planting due to the convenience of such an application and its priming effect on wheat seedling growth (Santis et al. 2020). However, the nitrogen requirements of wheat have been shown to vary among growth stages (Santis et al. 2020). In addition, nitrogen fertilizer availability can be affected by wheat processes (Lollato et al. 2019). Therefore, wheat nitrogen requirements may not be satisfied if the total amount of nitrogen fertilizer is applied at once, at pre-planting. Alternatively, split nitrogen management at the jointing may contribute towards satisfying the plant nitrogen demand via soil supply, thereby, enhancing nitrogen use efficiency, and ultimately improving grain yield. In addition, the response of wheat to nitrogen fertilization is dependent on the soilwater condition. Li et al. (2012) reported that an appropriate soil moisture level led to improved nitrogen availability and the simultaneous utilization of water and that supplemental irrigation is the main factor that affects the soil water condition. Previous studies reported the effects of the nitrogen application rate, fertilization ratio, and fertilization period on wheat yield under the sufficient irrigation condition . However, the effect of split nitrogen fertilization on wheat yield under watersaving irrigation condition requires further investigation. The appropriate nitrogen application under water-saving irrigation may contribute significantly to the sustainable production of winter wheat in the region of the HPC.
Leaf senescence is a developmental process that involves the degradation of macromolecules, such as nucleic acids and proteins, the accumulation of reactive oxygen species, and the transition from nutrient assimilation to nutrient remobilization (Nehe et al. 2020). The commencement of grain growth from anthesis onwards coincides with leaf senescence (Kitonyo et al. 2018a(Kitonyo et al. , 2018b. The plant leaves are the major sites for photosynthesis, which, together with the pre-anthesis reserves, supply the assimilates for grain-filling (Kitonyo et al. 2018a(Kitonyo et al. , 2018b. Premature leaf senescence can lead to a reduction in wheat canopy size ), loss of light use efficiency, and decreased crop yield (Thapa et al. 2018). A delayed onset of senescence with longer functional photosynthesis (stay green) may produce more assimilates for developing grains and, thus, has the potential to maximize grain yield Li et al. 2019). Nitrogen is a key factor in regulating the leaf senescence of crops (Roche et al. 2017). Previous studies have demonstrated that nutrient stress due to nitrogen deficiency induces the premature senescence of flag leaves (Jiang et al. 2020). In contrast, high nitrogen levels delay the initiation of flag leaf senescence and prolong the leaf functional period (Luo et al. 2018). However, previous nitrogen management studies have focused predominantly on the agronomic and physiological bases of yield potential under sufficient irrigation conditions. More research is needed to develop an optimal water and nitrogen management system to support the use of water-saving irrigation, especially involving fertigation techniques, for winter wheat production in the HPC. Therefore, in this study, the working hypothesis is that split nitrogen management techniques would significantly affect the grain yield by impacting the antioxidant capacity and 13 C photosynthate mobilization under water-saving irrigation condition. The objectives of the present study were to (1) examine and compare the antioxidant capacity of flag leaves between different nitrogen treatments after anthesis; (2) investigate differences in 13 C photosynthate assimilation, distribution, and redistribution between different nitrogen treatments under water-saving irrigation; and (3) determine the optimal nitrogen management technique under watersaving irrigation.

Experimental site
Field experiments were conducted during the winter wheat growing seasons in 2016-2017 and 2017-2018 at the experimental station in Shijiawangzi Village, Xiaomeng Town, Yanzhou, Jining City, Shandong Province, China (35°40′ N, 116°41′ E) (Fig. 1). This region is a typical plain region of the HPC, which has a warm temperature continental climate. The soil in the region is classified as clay loam. The top 0-20 cm soil layer contained 14.20 g kg −1 organic matter concentration, 1.13 g kg −1 total nitrogen, 122.60 mg kg −1 available nitrogen, 129.44 mg kg −1 available potassium, and 38.11 mg kg −1 available phosphorus. The crop previously grown at the study was corn, and all residual straw was incorporated into the soil after harvest.

Experimental design and crop management
Winter wheat variety "Jimai 22", a cultivar widely used in the HPC, was selected for the field experiments. The experiment included four treatments at a nitrogen application rate of 240 kg ha −1 , with fertilizer ratios of base to topdressing of fertilizer, namely, 10:0 (N1), 7:3 (N2), 5:5 (N3), and 3:7 (N4). A randomized complete block design was used in a split-plot arrangement with three Fig. 1 The study site map replications. All plots were 20 m 2 in size. The distance between two adjacent plots receiving different application treatments was 2 m to avoid the interference of nitrogen and irrigation factors. Soil moisture management in the experimental plot was based on the water-saving cultivation technique of supplementary irrigation by measuring soil moisture. Briefly, the relative water content in the 0-40 cm soil layer was supplemented to 70% at the jointing and anthesis stages of wheat.
The irrigation amount was calculated using the formula: M=10 × r × H × (βi−βj), where M represents the amount of irrigation and r represents the bulk density of the quasi-moist soil depth, H represents the pseudomoist soil depth, βi represents the design water content, and βj represents the water content of the soil before irrigation. Irrigation was conducted with a hose and measured using a meter (Man et al. 2014).
Urea (N 46%), calcium superphosphate (P 2 O 5 12%), and potassium chloride (K 2 O 60%) were used as the nitrogen, phosphate, and potassium fertilizers, respectively. The basal nitrogen fertilizer and the phosphate (150 kg ha −1 ) and potassium (112.5 kg ha −1 ) fertilizers were spread on the soil surface prior to sowing. Immediately after application, the fertilizers were mixed into the top 0-20 cm layer of soil using a rotary cultivator. At the joining stage of the cultivated wheat, nitrogen fertilizer was applied in furrows, which were immediately covered with soil. After fertilizer application, wheat seeds were sown at a depth of 4 cm on October 12, 2016, and October 24, 2017, and the three-leaf basic determining was 1.8 million ha −1 . Plants were harvested on June 8, 2017, and June 7, 2018. All other agronomic management protocols were maintained consistent in all treatments following the recommendations of local farmers for high yield.

Sample collection, analyses, and calculations SPAD value determination
The soil-plant analysis development (SPAD) value of the flag leaves was measured in 10 randomly selected plants per treatment, at 7-day intervals from anthesis to maturity (measuring 5 times altogether, every year from 2016 to 2018), using an SPAD-502 chlorophyll meter model (SPAD-502, Soil-plant Analysis Development Section, Minolta Camera Co., Osaka, Japan) (Ren et al. 2020).
Superoxide dismutase, malondialdehyde content, soluble protein content, sucrose content, and sucrose phosphate synthetase activity determination One hundred single stems that flowered on the same day in each plot were tagged and sampled from anthesis to maturity within a 7-day interval, while 20 flag leaves were sampled at each sampling stage (sampling 5 times altogether, every year from 2016 to 2018). Sampled flag leaves were frozen in liquid nitrogen for 1 min and then stored at −80°C. One half of the sampled flag leaves was used to measure superoxide dismutase (SOD), malondialdehyde (MDA) content, and soluble protein content, while the other half was used to measure the sucrose content and sucrose phosphate synthetase (SPS) activity.
For SOD enzyme extraction, 500 mg of the fresh leaves was ground in 5 mL extraction buffer consisting of 100 mmol L −1 potassium phosphate buffer (pH 7.0), 1 mmol L −1 ethylenediaminetetraacetic acid (EDTA), and 1% polyvinylpolypyrrolidone (PVPP). The extract was centrifuged at 20,000 rpm for 20 min at 4°C. The supernatant was collected and used to determine SOD activity, MDA content, and soluble protein content, following the methods described by Lv et al. (2017). SOD activity was evaluated based on its ability to inhibit the photoreduction of nitroblue tetrazolium (NBT). In turn, MDA content was calculated using the thiobarbituric acid reaction. Soluble protein content was calculated by Coomassie Brilliant Blue staining.

Dry matter accumulation and translation
Overall, 20 single stems were randomly sampled from each plot by cutting at ground level at anthesis and maturity stages, and separated into vegetative organs and grains (only at maturity). All plant samples were ovendried to a constant weight at 70°C to determine dry mater biomass. Dry matter accumulation and translocation indices were calculated using the following equations (Latifmanesh et al. 2018): where DMT is the dry matter translocation amount (kg ha −1 ), DMA is the dry matter accumulation amount at anthesis, DMM is the dry matter of vegetative plant parts at maturity, CDMT is the contribution of preanthesis assimilates to grain (%), GY is the grain yield, DMAA is the dry matter accumulation amount after anthesis, and CDMAA is the contribution of dry matter accumulation amount after anthesis to grains.

Carbon isotope analysis
We conducted a leaf isotope-tagging experiment using 13 CO 2 in both growing seasons. Ten representative single stems in each experimental plot were selected at anthesis. We encased the flag leaf of each selected single stem in a 0.1-mm-thick Mylar plastic bag, which permitted sunlight to pass at levels up to 95% of the natural intensity. The bags were sealed at the base with adhesive tape and subsequently injected with 3.5 mL of 13 CO 2 . After allowing photosynthesis to proceed for 30 min, 13 CO 2 in each bag was extracted through a KOH washer to absorb the remaining 13 CO 2 , and the bag was removed. This experiment was conducted from 09:00 to 11:00 a.m. on sunny days. At 72 h after processing from anthesis and maturity stages, the wheat plants were randomly sampled from each plot by cutting at the ground level. All plants were separated into stems and sheaths, leaves, glumes (spike axis and kernel husks), and grains (only at maturity), and oven-dried to a constant weight at 70°C to determine the aboveground biomass. All samples were milled to a fine powder using a ball mill, for use in the carbon isotope analysis. The carbon isotope content of milled samples (5 mg) was determined using an online system composed of an elemental analyzer, a TripleTrap, and a mass spectrometer (Carlo Erba 2100, Milan, Italy). The distribution of 13 C photosynthates among different organs was determined (Gao et al. 2017).

Grain yield and yield component determination
Grain yield and yield components were evaluated at the harvest of winter wheat stage (Gaju et al. 2014).

Statistical analysis
Correlations between the photosynthesis performance parameters and grain filling after anthesis were analyzed using the SPSS 13.0 software (α = 0.05). An analysis of variance was used to assess the effects of irrigation and nitrogen fertilizer treatments on wheat photosynthetic performance and grain-filling parameters using GLM in SPSS 13.0. A logistic equation of grain filling was modeled using SPSS 13.0. All graphics were produced using Excel and SigmaPlot 12.5.

Results
The chlorophyll content (SPAD value) Changes in SPAD value of flag leaves after anthesis are shown in Fig. 2. Split nitrogen treatment had significant effects on the SPAD value in both growth seasons (Additional file 1: Table 1). In 2016-2017, the mean SPAD value of the N3 treatment was significantly higher than that of the N1 and N2 treatments, and no significant differences were detected between the N3 and N4 treatment. Similar results were obtained in both growth seasons. These results indicated that plants under the N3 treatment maintained a longer effective duration of photosynthesis, compared with the other treatments.

Antioxidation indexes of flag leaves
Changes in MDA content, SOD activity, and soluble protein content in the flag leaves after anthesis are shown in Fig. 3. Split nitrogen treatments had no significant effects on MDA content, SOD activity, and soluble protein content at 0 DAA after anthesis (Additional file 1: Table 2). However, significant differences were detected among the different treatments at 7, 14, 21, and 28 DAA. Compared with the N1 and N2 treatments, the N3 treatment showed significant increases in SOD activity and soluble protein content at 7, 14, 21, and 28 DAA after anthesis and a significant decrease in MDA content. No significant differences were detected between N3 and N4 treatments. These results indicated that the Fig. 2 Effects of four split nitrogen fertilizer treatments on flag leaf SPAD value at 0 days, 7 days, 14 days, 21 days, and 28 days after anthesis. **P < 0.01; *P < 0.05 N3 treatment extended the duration of high SOD activity and soluble protein content, which was conducive to delaying the senescence and apoptosis of leaf cells.

Sucrose content and SPS activity of flag leaves
Changes in the sucrose content and SPS activity levels in the flag leaves after anthesis are shown in Fig. 4. The sucrose content and SPS activity first increased and then decreased as senescence progressed from 0 to 28 DAA.
Sucrose content and SPS activity were highest at 14 DAA. Split nitrogen treatments had significant effects on sucrose content and SPS activity (Additional file 1: Table  3). Significantly higher levels were detected in the N3 treatment compared to the N1 treatment, whereas no significant differences were detected between N3 and N2 treatments or the N3 and N4 treatments at 0 DAA. At 7 DAA, the sucrose content in the N3 treatment was significantly higher than that in the N1 and N2 treatments, while no significant differences were detected between N3 and N4 treatments. From 14 to 28 DAA, sucrose content was highest in the N3 treatment, followed by the N2 and N4 treatments, while the lowest content was detected in the N1 treatment. SPS activity was similar across treatments. Overall, sucrose content and SPS activity in the flag leaves were improved by the N3 treatment.

Translocation after anthesis in vegetative organ
In 2016-2017, total dry matter biomass at anthesis and maturity and the DMT did not differ significantly among treatments (Table 1). The CDMT did not differ significantly among the N2, N3, and N4 treatments, but the CDMT from the N1 treatment was always higher than that of the other nitrogen treatments. The DMAA was the highest in the N3 treatment, followed by N2 and N4 treatments, and lowest in the N1 treatment. The CDMAA was significantly higher in the N3 treatment compared to the N1 treatment but did not differ significantly with respect to the N2, or the N4 treatments. The results obtained in 2017-2018 showed almost the same trends as those obtained in 2016-2017. Altogether, our data indicated that the dry matter assimilation after anthesis was higher under the N3 treatment and that the dry matter redistribution from the vegetative organs to the developing grain remained at a higher level after anthesis.

C photosynthate distribution in different organs of wheat
The changes in distribution of 13 C photosynthate among different organs at 72 h after labeling and at maturity are shown in Table 2. In 2016-2017, stems, sheaths, spike axes, and husks under the N3 treatment had higher 13 C photosynthate contents by averages of 15.80% and 10.52% and 11.99% and 8.17% compared to those under the N1 and N2 treatments, respectively. However, no significant differences were detected between the N3 and N4 treatments. Additionally, the leaf 13 C photosynthate content did not differ significantly among treatments at 72 h after labeling. At maturity, the 13 C Fig. 4 Effects of four split nitrogen fertilizer treatments on flag leaf source content and SPS activity at 0 days, 7 days, 14 days, 21 days, and 28 days after anthesis. ns, not significant; **P < 0.01; *P < 0.05 photosynthate contents in the leaves, stems, and sheaths were significantly lower in the N3 treatment than in the other split nitrogen treatments. The spike axis and husk 13 C photosynthate content were significantly higher in the N3 treatment than in the N1 and N2 treatments; however, no significant differences were detected between the N3 and N4 treatments. The grain 13 C photosynthate content was higher under the N3 treatment by averages of 24.43%, 11.83%, and 6.44% compared to the N1, N2, and N4 treatments, respectively. Similar responses to treatments were observed in both growth seasons. Overall, the N3 treatment significantly increased carbohydrate assimilation after anthesis and their distribution from vegetative organs to the grains, which likely accounts for the higher grain yield observed for plants in the N3 treatment.

C photosynthate translocation after anthesis
In both growth seasons, split nitrogen treatments had no effect on the translocation amount of the spike axis and husk 13 C photosynthates but did significantly affect the translocation amounts of leaf and stem 13 C DMT Dry matter translocation amount, CDMT Contribution of pre-anthesis assimilates to grain, DMAA The dry matter accumulation amount after anthesis, CDMAA The contribution of dry matter accumulation amount after anthesis to grains photosynthates (Fig. 5). In 2016-2017, the translocation amounts of leaf and stem 13 C photosynthates under the N3 treatment were higher by averages of 16.03%, 12.63%, and 11.51%, and 27.97%, 17.00%, and 8.52% compared to the N1, N2, and N4 treatments, respectively. The second growth season showed similar results. Collectively, these results indicated that the N3 treatment substantially promoted the translocation of photosynthates to the grain, especially from the leaves and stems.

Grain yield and yield components
The effects of split nitrogen fertilizer treatments on grain yield and its components in both growth seasons are shown in Table 3. Split nitrogen fertilizer treatments significantly affected grain yield in both the 2016 and 2017 seasons. Compared with N1, N2, and N4 treatments, the N3 treatment increased grain yield by 16.27%, 7.09%, and 5.70% in 2016-2017 and 16.72%, 8.88%, and 5.89% in 2017-2018, respectively. The 1000-grain weight and the number of grains per spike were significantly higher in the N3 treatment, compared with any other split nitrogen treatments. Spike number did not differ significantly among the treatments in both growth seasons. These results indicated that the N3 treatment led to a significant increase in grain yield that was mainly attributed to the increase in 1000-grain weight and the number of grains per spike.

Effects of split nitrogen fertilization on flag leaf antioxidant capacity
Chlorophyll is essential for leaf photosynthesis, and its content in wheat leaves reflects their photosynthetic capacity (Hlavacova et al. 2018). Nitrogen affects and participates in chlorophyll synthesis (Christopher et al. 2014;Gaju et al. 2014). Moderate split nitrogen fertilizer application has been shown to increase leaf chlorophyll content, which promotes the growth of aboveground parts, such as leaves, and thus leads to an increase in biomass (Xie et al. 2016;Tian et al. 2020). In our study, moderate split nitrogen ratios readjusted the flag leaf antioxidant capacity, such that plants under the N3 treatment showed the highest SOD activity and  soluble protein content after anthesis, compared with the other split nitrogen treatments. In contrast, in the unequal ratio treatments (N1, N2, N4), the increase in antioxidant enzyme activity was not sufficient for detoxification of reactive oxygen species, which caused damage to photosystems and contributed to chlorophyll degradation . The results of this study showed that flag leaf SPAD values decreased significantly under excessive basal or topdressing nitrogen fertilizer treatments. However, SPAD values significantly improved, and MDA content was significantly reduced under the N3 treatment, which indicated that the N3 treatment was the most appropriate split nitrogen treatment for the protection of chlorophyll integrity and functionality. These responses and the low level of chlorophyll degradation under the N3 treatment may be associated with plants maintaining a higher water status as a result of an improved root system under this treatment (Kitonyo et al. 2018a(Kitonyo et al. , 2018b. Overall, these findings indicated that the equal ratio treatment resulted in the flag leaves having a longer duration of photosynthetic capacity than those in any of the unequal ratio treatments (Kitonyo et al. 2017), which favored higher levels of photosynthetic assimilation.
Effects of split nitrogen fertilization on dry matter accumulation amount, 13 C-photosynthate translocation, and distribution One objective of the experiment was to determine how split nitrogen fertilizer treatment affects photosynthate assimilation, distribution, and redistribution. Our experiment allowed to observe positive effects of the split nitrogen treatments on assimilation in vegetative organs, photosynthesis capacity, and grain yield. SPS enzyme transforms photosynthesis assimilates into sucrose and contributes to grain yield (Liu et al. 2020b). The results of our study showed that sucrose content of flag leaves was higher in the N3 treatment than in the other split nitrogen treatments from 14 DAA to 28 DAA, which coincided with the N3 treatment having the highest SPS activity. The final grain yield of winter wheat is determined by the ability of dry matter assimilation in the grain-filling stage (Kamiji et al. 2014). Our results showed that the dry matter assimilation after anthesis was significantly reduced under excessive basal or topdressing nitrogen fertilizer regimens, which was in accordance with previous observations (Liang et al. 2017). Our assessment of the amount of 13 CO 2 fixed by wheat plants and translocated to each organ showed that the leaves provided an efficient model to appraise the contribution of each vegetative organ to the grain under different split nitrogen treatments. Our 13 C-isotope tracer experiment revealed that the equal ratio treatment (N3) significantly increased the 13 C photosynthate accumulation post-anthesis and its contribution to grain compared to the unequal ratio treatments (N1, N2, and N4). This is likely the result of an improved source-sink relationship because dry matter partitioning was biased towards the reproductive organs. The equal ratio treatment resulted in a higher accumulation of dry matter compared to the unequal ratio treatments, especially during anthesis. This higher accumulation of dry matter seemingly occurred because a higher abundance of assimilates at anthesis enabled plants grown under the equal ratio treatment to produce more leaves with enhanced longevity, as needed to increase the source capacity and thus, optimize leaf cover for light interception and use (Lv et al. 2017;Si et al. 2020). Photosynthates are produced by photosynthesis, and nutrient deficiencies at grain-filling stage can accelerate leaf senescence by increasing ROS production (Okamura et al. 2018), while the supply of nitrogen fertilizer during grain-filling can delay leaf senescence and increase wheat yield ).
In addition, our findings demonstrated that split nitrogen application altered the dry matter allocation among plant organs in wheat plants. Total 13 C assimilates translocated to the grain, the stems, and sheaths made the highest contribution, followed by the spike axes and husks, while leaves made the lowest contributions. In this study, as basal or topdressing nitrogen fertilizer increased, wheat plants showed an increase in vegetative dry matter accumulation and a decrease in dry matter allocation to spikes. These effects negatively affected dry matter accumulation in the grain at grain-filling stage, which led to a reduced dry matter accumulation in the grain at maturity (Finnan et al. 2019). However, it is noteworthy that the N3 treatment not only increased the amount of photosynthesis assimilates produced in the post-anthesis period, but also promoted the post-anthesis translocation of carbohydrates from the vegetative organs to the grains as well, thus, further supporting the notion that equal ratio treatments (such as that of N3) can increase grain yield under conditions of water-saving irrigation.

Effects of split nitrogen fertilization on grain yield and yield components
Previous studies have shown that wheat yield was highly correlated with the number of grains per spike and spike number under the sufficient irrigation condition (Rivera-Amado et al. 2019). Current approaches to increase grain yield are mainly focused on increasing the number of grains per spike or 1000-grain weight, but it has proven difficult to further increase the capacity of the grain sink (Agami et al. 2018). Therefore, 1000-grain weight is the most important restrictive factor affecting grain yield. The results of our experiment under conditions of water-saving irrigation further demonstrated that the N3 treatment significantly improved grain yield, which was attributed to a higher number of grains per spike concomitantly with a higher 1000-grain weight.
The unequal ratio treatments (N1, N2, and N4) showed significantly lower yield potential than the equal ratio treatment (N3). This decrease in grain yield can be explained by two factors, mainly: (1) excessive basal nitrogen fertilizer application might have led to an insufficient supply of dry matter for grain filling at the grain-filling stage because the initiation of grain filling requires a large supply of photoassimilate, and (2) excessive topdressing nitrogen fertilizer application delayed plant maturity, as it favored continued vegetative growth. Zhang et al. (2018) reported that nitrogen fertilizer application significantly improved yield by increasing spike number. Interestingly, split nitrogen fertilizer treatments in the present study had no significant effect on spike number ). This discrepancy may be attributed to different climatic conditions and the effects of soil fertility on spike number between the two studies ). Nonetheless, our study provides substantial evidence that split nitrogen management can help towards a significant improvement of wheat production. However, the evaluation of the effects of fertilizer management practice on wheat soil environment and wheat yield warrants further research.

Conclusion
Here, we investigated the effects of split nitrogen fertilizer management on 13 C photosynthate mobilization and antioxidant enzyme activity in winter wheat under watersaving irrigation conditions. Altogether, our findings confirmed that the 240 kg ha −1 of total nitrogen fertilizer treatment applied in a 5:5 basal-topdressing ratio significantly increased the SPAD values, antioxidant enzymes activities, sucrose content, and SPS activity but reduced the accumulations of MDA in wheat plants. Furthermore, the treatment significantly increased the assimilation of carbohydrates after anthesis and promoted their distribution from vegetative organs to the developing grains. The 5:5 top-dressing N fertilizer treatment effectively increased grain yield by increasing 1000-grain weight and the number of grains per spike. Therefore, combined water-saving irrigation and split nitrogen management are recommended for grain yield improvement of winter wheat in the region of the HPC.