Variation of ¹³C and ¹⁵N enrichments in different plant components of labeled winter wheat (Triticum aestivum L.)

Treatments and ¹⁵N labeling procedure

In the current study, we used winter wheat (Triticum aestivum L. ‘Luyuan 502’), a staple crop grown worldwide (Zhao et al., 2016). To examine the impacts of wheat growth stage and N fertilization on variation in ¹³C enrichment of winter wheat, two N treatments (0 mg N kg⁻¹ soil, −N; and 90 mg N kg⁻¹ soil, equivalent to 250 kg N ha⁻¹, +N) were applied, and wheat plants grown in a calcareous soil (fluvo-aquic sandy loam) were pulse-labeled with ¹³CO₂ at the elongation (168 days after sowing; DAS) and grain filling (205 DAS) stages (Figs. 1A and 1B). To examine the degree of variation in ¹⁵N enrichment among aboveground compartments, ¹⁵N-labeled fertilizer (¹⁵N-labeled urea; Shanghai Research Institute of Chemical Industry, Shanghai, China) was applied by direct addition to the soil for the +N treatment (¹⁵N 10.14 atom %; about US$10 per gram of ¹⁵N labeled urea). Additional wheat plants were selected as ¹⁵N natural abundance controls and were fertilized with unlabeled urea as for +N treatment. Ten pre-germinated (24–48 h) winter wheat seeds were sown directly into PVC pots (30 cm ×20 cm i.d.) containing 8.7 kg air-dried soil, and then the soil was moistened to 70% water-holding capacity. One week after germination (20 October, 2014), all except the six strongest seedlings were removed from each pot (equivalent to a field planting density of 1.5 million plants ha⁻¹). The experiment was conducted under field conditions. For the +N treatment, half of the ¹⁵N labeled or unlabeled urea was applied as a basal dressing at sowing (0.84 g ¹⁵N labeled or unlabeled urea was applied to each pot) and the remaining half was used to topdress plants at the elongation stage (15 March, 2015). The wheat plants for the first ¹³CO₂ labeling were fertilized as a single basal application and received 45 mg N kg⁻¹ air-dried soil. The wheat plants for the second ¹³CO₂ labeling received a total of 90 mg N kg⁻¹ air-dried soil in two applications during the experiment (1.68 g ¹⁵N labeled or unlabeled urea was applied to each pot, which corresponded to a total fertilization of 250 kg N ha⁻¹), which was split into 2  × 45 mg N kg⁻¹air-dried soil applications (basal urea N and topdressing urea N). Monopotassium phosphate (KH₂PO₄) fertilizer was applied as a basal dressing at the rate 17 mg P and 18 mg K kg⁻¹ air-dried soil, which conformed with the local agricultural practice. To simulate plant litter similar to that occurring under the field-grown conditions in this study, we applied ¹⁵N-urea to the soil of pot-grown winter wheat in the field rather than supplying ¹⁵N-fertilizer solution to a soil-free culture system.

Growth conditions and ¹³C pulse-labeling procedure

Soil was collected from the plough layer (0–20 cm) of farmland near the Agro-Ecosystem Experimental Station of China Agricultural University, Huantai County, Shandong Province (36°57′N, 117°59′E, 18 m elevation). The soil of the experimental field was derived from Yellow River alluvial sediments and was classified as an Aquic Inceptisol (a calcareous, fluvo-aquic sandy loam; Shi et al., 2013). Soil properties were as follows: soil organic C = 8.4 g kg⁻¹, soil inorganic C = 5.2 g kg⁻¹, total N = 0.67 g kg⁻¹, NH₄-N = 1.6 mg kg ⁻¹, NO₃ −N = 17.2 mg kg⁻¹, pH = 8.2 (soil:water = 1:2.5), available K = 174 mg kg⁻¹, and Olsen P = 5.2 mg kg⁻¹. After sampling, the soil was air-dried, homogenized, and sieved using a five mm screen prior to the experiment. Each pot (height = 30 cm; inner diameter = 20 cm) was filled with 8.7 kg of air-dried soil to a bulk density of 1.38 g cm⁻³. Wheat roots were mainly concentrated in the 0–20 cm soil layer and contributed 57.6%–78.6% of the 0–100 cm total root weight in typical farmland on the North China Plain (Hou, 2011). Hence, pots with 30 cm soil depth were considered sufficient for normal wheat root development in this study. The pot-grown winter wheat seedlings were transferred to an experimental field at the Agro-Ecosystem Experimental Station. The pots containing wheat plants were placed in cropland holes (30 cm depth; 24 cm diameter) to simulate local wheat-growing conditions. The total growth period was 230 days, with six developmental stages recognized: (i) seeding (0–17 DAS), (ii) tillering (18–150 DAS), (iii) elongation (151–179 DAS), (iv) anthesis (180–193 DAS), (v) grain filling (194–214 DAS), and (vi) dough ripening (215–230 DAS). Precipitation was 147 mm and the soil temperature ranged from −1.6 to 29.7 °C during the wheat growing season (Zhao et al., 2017). The soil water content of each pot, which was controlled gravimetrically to simulate local wheat production, was adjusted daily to 65% (during the seedling stage), 70% (tillering), 80% (elongation), 80% (anthesis), and 70%–75% (grain-filling) of the field water-holding capacity in accordance with the amount of rainfall and evaporation.

Labeling with ¹³CO₂ was performed at the elongation and grain filling stages (i.e., 168 and 205 DAS). On each occasion, four pots were randomly selected for ¹³CO₂ labeling from each of the −N and +N treatments, and an additional eight pots were selected as ¹³CO₂ unlabeled controls and were maintained separately from the labeled plants. Hence, we used 32 pots in the experiment, with three replicates for each N treatment, for destructive sampling at 28 days after labeling during each of the developmental stages (three from four pots per sampling for each treatment): eight pots were dual labeled with ¹³C and ¹⁵N in the +N treatment; eight pots were labeled with ¹³C for the −N treatment; and an additional 16 pots were selected as unlabeled controls (eight pots for −N and eight pots for +N). We considered each pot to be a technical replicate for each treatment (three from four pots per sampling for each treatment), and took each plant within a pot to be a biological replicate (three plants chosen within the pot). A chamber (0.6 m long ×1.0 m high; Fig. 1C), adapted from Swinnen, Van Veen & Merckx (1994) and Meng et al. (2013), was used for ¹³CO₂ labeling. On the previous day, the soil surface was covered with a PVC board and sealed with silicon, including around the stems. To remove all ¹²CO₂ before labeling, the air inside the chamber was continuously circulated for 30 min through 10 M NaOH solution using a membrane pump. The ¹³C-labeled Na₂¹³CO₃ at 98 atom % ¹³C (about US$100 per gram of Na₂¹³CO₃) was provided by the Shanghai Research Institute of Chemical Industry. A beaker containing Na₂¹³CO₃ (8.0 g Na₂¹³CO₃ for each labeling event) was placed inside the chamber. To determine the CO₂ concentration within the chamber using an infrared gas analyzer (CI301PS; CID Bio-Science, Camas, WA, USA), an unlabeled control treatment with an additional four pots was set up in a separate chamber under identical conditions except that Na₂CO₃ was used to produce non-labeled CO₂. This arrangement was used because of the different wavelengths for maximum absorption of ¹³CO₂ and ¹²CO₂ in the infrared detector (Meng et al., 2013). If the rate of decline in CO₂ concentration slowed considerably (below 200 µL L⁻¹) in the chamber containing the unlabeled control treatment, 1 M H₂SO₄ was injected until the CO₂ concentration increased to approximately 360 µL L⁻¹. The same volume of H₂SO₄ solution was also injected into the chamber housing the plants undergoing labeling. A fan was used to homogenize the atmosphere within each chamber. After 7 h of labeling, plants were removed from the chamber to prevent re-assimilation of shoot-respired ¹³CO₂.

Sampling and chemical analyses

The ¹³C-labeled winter wheat plants and soils were destructively sampled 28 days after each labeling. No significant difference between the mean aboveground biomass and root biomass of the ¹³C pulse-labeled plants and the unlabeled control plants was observed when they were harvested 28 days after labeling during each of the developmental stages. This observation indicated that ¹³C pulse labeling did not affect plant growth. To examine the uniformity of labeling within a wheat plant, individual wheat plants were collected at the final harvest and the shoots (stem and leaves) were divided into three equal-length portions (upper, middle, and lower portions; Fig. 1D). Separate analyses of ¹³C and ¹⁵N enrichment were undertaken on the leaf and stem portions as well as the ears. Three plants per pot were assessed for variation in ¹³C and ¹⁵N enrichment. Shoots (stems and leave) and ears were oven-dried at 60 °C to a constant weight. All samples were ground to a fine powder (<500 µm) in a ball mill (Restol MM2000, Retsch, Haan, Germany) prior to analysis.

The ¹³C and ¹⁵N values of wheat plant materials were determined using isotope ratio mass spectrometry (DELTAplus XP, ThermoFinnigan, Bremen, Germany). The abundance of ¹³C was expressed as parts per thousands (‰) relative to the international standard (Pee Dee Belemnite; 0‰) expressed as delta units (δ) (Craig, 1953). The ¹⁵N enrichment of ¹⁵N-labeled plant parts fertilized with labeled urea was reported as atom % ¹⁵N excess calculated by subtracting the values for the ¹⁵N natural abundance control samples fertilized with unlabeled urea (Coplen, 2011).

Data processing and statistical analysis

To estimate the weight-average ¹³C and ¹⁵N values of the aboveground parts and three equal-length portions (upper, middle, and lower portion) of the shoot of winter wheat plants, a simple isotopic mass balance mixing equation was used (Thompson, 1996): (1)${\displaystyle {\delta }^{13}{\mathrm{C}}_{\text{abveground}}={\delta }^{13}{\mathrm{C}}_{\mathrm{ear}}\times {\mathrm{F}}_{\mathrm{ear}}+{\delta }^{13}{\mathrm{C}}_{\text{leaves}}\times {\mathrm{F}}_{\text{leaves}}+{\delta }^{13}{\mathrm{C}}_{\text{stem}}\times {\mathrm{F}}_{\text{stem}}}$ (2)${\displaystyle {}^{15}{\mathrm{N}}_{\text{abveground}}{=}^{15}{\mathrm{N}}_{\mathrm{ear}}\times {\mathrm{F}}_{\mathrm{ear}}{+}^{15}{\mathrm{N}}_{\text{leaves}}\times {\mathrm{F}}_{\text{leaves}}{+}^{15}{\mathrm{N}}_{\text{stem}}\times {\mathrm{F}}_{\text{stem}}}$ (3)${\displaystyle 1={\mathrm{F}}_{\mathrm{ear}}+{\mathrm{F}}_{\text{leaves}}+{\mathrm{F}}_{\text{stem}}}$ where δ¹³C_aboveground, δ¹³C_ear, δ¹³C_leaves, and δ¹³C_stem is the C isotope value (δ) for aboveground parts, ear, leaves, and stem of winter wheat, respectively; ¹⁵N_aboveground, ¹⁵N_ear, ¹⁵N_leaves, and ¹⁵N_stem is the ¹⁵N excess (in atom %) for aboveground parts, ear, leaves, and stem of winter wheat, respectively; and F_ear, F_leaves, and F_stem is the total aboveground dry weight of winter wheat plants in the ear, leaves and stem fractions, respectively. (4)${\displaystyle {\delta }^{13}{\mathrm{C}}_{i\text{portion}}={\delta }^{13}{\mathrm{C}}_{i\text{leaves}}\times {\mathrm{F}}_{i\text{leaves}}+{\delta }^{13}{\mathrm{C}}_{i\text{stem}}\times {\mathrm{F}}_{i\text{stem}}}$ (5)${\displaystyle {}^{15}{\mathrm{N}}_{i\text{portion}}{=}^{15}{\mathrm{N}}_{i\text{leaves}}\times {\mathrm{F}}_{i\text{leaves}}{+}^{15}{\mathrm{N}}_{i\text{stem}}\times {\mathrm{F}}_{i\text{stem}}}$ (6)${\displaystyle 1={\mathrm{F}}_{i\text{leaves}}+{\mathrm{F}}_{i\text{stem}}}$

where δ¹³C_iportion and ¹⁵N_iportion are the C isotope value (δ) and ¹⁵N excess (in atom %) for the upper, middle, and lower portion, respectively, of the shoot of winter wheat plants; and F_ileaves and F_istem are the respective leaves and stem fractions in the same portion of the shoot.

Statistical analysis was carried out with SPSS (version 11.0, 2002; SPSS, Chicago, IL, USA). Two-way analysis of variance was performed to analyze the effect of N fertilization and/or developmental stage on ¹³C and ¹⁵N distribution in different aboveground components of the winter wheat plants. The least significant difference (LSD) test was used to examine differences in ¹³C and ¹⁵N isotope values between the N treatments or different aboveground components. The significance level of 0.05 was chosen to indicate statistical significance.

Winter wheat biomass and total C and N contents

The interactive effect of plant age and N fertilizer treatment on increasing the biomass of the ear and whole plant was significant (Table 1 and Fig. 2A), which indicated that the effect of N fertilization tended to be stronger as wheat growth progressed. Straw, ear, and whole-plant biomass were significantly higher in the +N treatment than in the −N at the elongation and grain filling stages (Fig. 2A). With progression of growth, the C and N contents of wheat straw and the ear significantly increased (Table 1 and Figs. 2B, 2C), whereas no significant differences in the roots were observed from the elongation to the grain filling stages (Table 1 and Fig. 2A). Compared with those of the −N treatment, N fertilization significantly decreased the C content of straw, and significantly increased the N content of wheat straw and the ear at the elongation and grain filling stages (Table 1 and Figs. 2B, 2C).

δ¹³C values of aboveground components

In the ¹³C-labeled wheat plants, the highest δ¹³C values were observed in leaves, followed by stems and ears at the elongation stage. However, significantly higher ¹³C enrichment was observed in the ears compared with the other aboveground parts at the grain filling stage (Table 2). As growth progressed, the δ¹³C values of ears significantly increased by 130% and 170% for the −N and +N treatments, respectively. In contrast, δ¹³C values of leaves and stems significantly decreased by 82.3% (−N) and 72.2% ( +N) in leaves and by 72.4% (−N) and 65.5% (+N) in stems. These results indicated that growth stage had a significant impact on ¹³C distribution among ears, leaves, and stems (Table 3). Nitrogen fertilization decreased δ¹³C values of aboveground components at the elongation stage and had no significant effect at the grain filling stage; however, the change was only significant in leaves at the elongation stage (36.9%; Tables 2 and 3).

Shoots (leaves and stems) of single plants were divided into three equal-length portions (upper, middle, and lower), to analyze the ¹³C distribution within the shoot. Following pulse labeling at the elongation stage, δ¹³C values of leaves and stems in the different portions showed significant differences. The leaves of the lower portion was the least ¹³C enriched for the +N and −N treatments, and ¹³C enrichment was highest in the middle and upper portions under the −N and +N treatments, respectively. However, δ¹³C values of stems were highest in the lower portion, followed by the middle and upper portions, for the −N and +N treatments (Fig. 3). A significant difference was observed at the grain filling stage: for both leaves and stems, δ¹³C values were significantly higher in the upper portion, with no significant differences between the middle and lower portions observed (Fig. 3).

When we combined δ¹³C values from the same portions of leaves and stems at the elongation stage, no significant differences were observed among the different portions for the +N treatment, and δ¹³C values of the upper and middle portions were significantly higher than that of the lower portion for the −N treatment. At the grain filling stage, the δ¹³C values of the upper portion was highest, followed by those of the middle and lower portions under the −N and +N treatments (Table 4).

Labeled plants showed coefficients of variation (CVs) for intra-pot (between individual plants) δ¹³C of 9.4%–76.4% and 6.3%–31.5% at the elongation and grain filling stages, respectively. Inter-pot CVs (among plants across different pots) for δ¹³C ranged from 16.5%–59.0% at the elongation stage and 3.8%–70.9% at the grain filling stage (Table 5). Generally, CVs at the grain filling stage were lower than those at the elongation stage, and N addition tended to increase δ¹³C variation, particularly at the grain filling stage.

¹⁵N enrichment variation in aboveground components of winter wheat

Following ¹⁵N-labeling, the rank order of ¹⁵N excess (in atom %) of ears, leaves, and stems was similar at the elongation and grain filling stages, with ¹⁵N excess highest in ears, followed by stems and leaves (Table 2). With regard to the three shoot portions (without ears), combined ¹⁵N excess of leaves and stems were similar among different portions at the elongation and grain filling stages (Table 4). The ¹⁵N excess of stems and leaves was similar at the elongation stage, but was highest in the upper portion and lowest in the lower portion at the grain filling stage (Fig. 4). As growth progressed, i.e., with increase in biomass, ¹⁵N excess increased by 32.1%–41.8% (Table 2). Developmental stage had a significant impact on ¹⁵N excess for all shoot components except leaves in the lower portion (Table 3). Labeled wheat exhibited an intra-pot CV of 7.7%–18.1% at the elongation and grain filling stages; the inter-pot CV was 1.7%–23.4% at the elongation stage, with considerably less variation at the grain filling stage (0.1%–6.2%; Table 5). At both stages, the variation in ¹⁵N excess was less than that of δ¹³C. For both ¹⁵N excess and δ¹³C values, intra-pot and inter-pot CVs were comparable, which implied that the degree of variation in the ¹³C and ¹⁵N distributions was similar.

Variation in ¹³C enrichment

Average total δ¹³C values of aboveground parts of winter wheat plants, resulting from ¹³CO₂ release from Na₂¹³CO₃ (98 atom % ¹³C) in the field labeling chamber, ranged from 81‰ to 295‰ (Table 2). Given that the majority of previous studies that involved pulse labeling of wheat used ¹⁴C rather than ¹³C, we compared our findings with studies that used ¹³C labeling in other plant species. By injection of ¹³CO₂ gas (90 atom% ¹³C) in a field-labeling chamber, Tahir et al. (2018) achieved a high degree of δ¹³C enrichment in wheat with overall δ¹³C enrichment of 493‰ in leaves and 357‰ in stems. Meng et al. (2013) enriched maize shoots to approximately 125‰–178‰ via ¹³CO₂ release from Ba¹³CO₃(98 atom % ¹³C) in a laboratory labeling chamber. The direct ¹³CO₂ labeling method has also been used to generate highly ¹³C-enriched crop plants, such as rice (533‰–927‰) (Liu, Jiang & Li, 2015) and clover (>1,000‰) (Thompson, 1996). The ¹³C enrichment of plant materials is dependent on the net ¹²C and ¹³C assimilation rates as well as the quantity of ¹³C per gram dry weight of plant labeled, which is influenced by the quantity of ¹³C added to the labeling chamber.

The development of the winter wheat ear during the entire growth season was reflected in the increasing allocation of ¹³C to that reproductive organ. As growth advanced, δ¹³C values of ears significantly increased, whereas those of leaves and stems significantly decreased (Tables 1 and 2). This pattern, i.e., an increase in the carbohydrate sink strength during plant development, is typical of cereal crops. In a ¹⁴C pulse-labeling experiment, Swinnen, Van Veen & Merckx (1994) observed that as much as 85% of the ¹⁴C allocated to aboveground parts of spring wheat plants was recovered in the ears at the ripening stage. In the present study, the proportion of ¹⁴C in ears relative to all aboveground portions increased from 30% at the elongation stage to 50% at the grain filling stage (Fig. 2A). Among aboveground components, δ¹³C values were also highest in ears at the grain filling stage (Table 2). At the grain filling stage, the ears were conditioned by the size and availability of the C pools. At the grain filling stage, the predominant C sources for wheat ears are primarily from remobilization of labeled photosynthesized C from leaves and stems, and from photosynthesis by the ear itself (Palta et al., 1994; Aranjuelo et al., 2013; Zhou et al., 2016). Previous studies that used the natural abundance of ¹³C and ¹³C-labeling techniques revealed that photosynthesis in the ear may account for 40%–80% of the C accumulated in the ears (Palta et al., 1994; Aranjuelo et al., 2011; Sanchez-Bragado et al., 2014). Given that the leaves and stems contained the lowest quantities of labeled C in wheat, these organs were indicated to act as a C source for grain filling, with ears as the sink containing higher amounts of labeled C. This result is in agreement with a previous C pulse-labeling study of wheat during grain filling (Aranjuelo et al., 2013). However, at the elongation stage, higher ¹³C-enriched values in the shoots (leaves and stems) were observed in the present study (Table 2). These results suggested that shoots acted as major labeled-C sinks owing to the poor sink strength of ears at the elongation stage, whereas ears represented an important labeled-C sink that competed with the shoots as the C sink at the grain filling stage (Palta et al., 1994; Aranjuelo et al., 2013; Zhou et al., 2016).

At the elongation stage, N addition tended to reduce δ¹³C values of aboveground components through a dilution effect (Table 2). Compared with those of the −N treatment, net ¹³C assimilation in plant biomass increased by 7.8% under the +N treatment (76 mg ¹³C pot⁻¹ for −N vs 82 mg ¹³C pot⁻¹ for +N), while the +N treatment showed a 35% increase in plant biomass (Fig. 2A). An additional likely relevant factor was reduced carbohydrate accumulation under over-supplied N conditions, i.e., under the +N treatment in the current study (Palta et al., 1991; Gooding & Davies, 1992). At the grain filling stage, N addition tended to increase δ¹³C values of aboveground components (Table 2). This result probably reflects the presence of a higher number of green leaves in the +N treatment at an advanced stage of maturity (grain filling), which leads to stronger photosynthetic capacity and hence can affect the ¹³C-isotopic signature of aboveground components.

At both labeled stages, ¹³C enrichment was higher in upper (newly developed) leaves than in lower (aged) leaves (Fig. 3), which may be explained by the more rapid uptake of ¹³CO₂ by younger leaves. Leaf age influences photosynthetic capacity and the ¹³C-isotopic signature (Girardin et al., 2009). In addition, young expanded leaves provide the majority of labeled C assimilates for stem growth (Ryle & Powell, 1972; Subrahmanyam & Rathore, 2004). Additional factors, such as light intensity (Ryle & Powell, 1972; Subrahmanyam & Rathore, 2004), air temperature (Meharg & Killham, 1989), and relative humidity (Farquhar, O’leary & Berry, 1982), also influence the photosynthetic rate and alter the C-isotopic signature. Wheat leaves showed higher δ¹³C values than those of stems (Table 2), which is consistent with previous ¹³CO₂pulse-labeling studies, such as a study of grain sorghum (Berg et al., 1991). Leaf and stem positions (upper, middle, and lower) also influence the δ¹³C values of aboveground components (Thompson, 1996). In the present study, the upper leaves tended to be more enriched in ¹³C compared with that of the upper stems, whereas the lower leaves were generally less ¹³C-enriched than stems at the same position (Table 2) (Thompson, 1996). This result was due to the remobilization of photosynthate and translocation of higher amounts of recently fixed C from leaves to stems during reproductive (grain filling) and vegetative (elongation) developmental stages. The strong variation in δ¹³C values in leaves and stems indicated that a single ¹³CO₂-labeling pulse did not produce uniformly ¹³C-labeled plant material. Repeat-pulse labeling has been suggested to overcome the technical constraints of continuous labeling and the high variation of the single-pulse labeling approach, and may generate adequately homogeneous ¹³C-labeled plant material (Roper et al., 2013; Tahir et al., 2018).

With respect to isotope signals, labeled plants also exhibited different patterns of variation among individual plants within a pot and among the three replicate pots (Table 5). Approximately 75% of the leaf and stem components had a CV for intra-pot δ¹³C higher than 30% at the elongation stage, whereas 67% had a CV less than 20% at the grain filling stage. This result demonstrates that the variability between individual wheat plants at the grain filling stage was smaller than that at the elongation stage, especially for the −N treatment. The variation in ¹³C-enrichment in different portions of the shoot at the elongation stage were markedly higher than those at the grain filling stage, which may reflect that young leaves are the most frequently labeled organs in wheat (Girardin et al., 2009; Nguyen Tu et al., 2013). The stronger labeling of starch is associated with its rapid synthesis during photosynthesis in leaves compared with that of stems during vegetative growth stages. The redistribution of labeled C from source organs (leaves and stems) to sink organs (ears), consisting of C assimilated during the grain filling period (Aranjuelo et al., 2013; Zhou et al., 2016), may lead to a decreased extent of variation in ¹³C enrichment among plant organs. The difference in variation patterns between the +N and −N treatments is likely because C and N remobilization starts earlier and is stronger when plants are grown under low N fertilization compared with the high N fertilization condition (Aranjuelo et al., 2013). Approximately 83% of leaf and stem components had an inter-pot δ¹³C CV higher than 20% at the elongation stage, whereas 67% had an intra-pot δ¹³C CV lower than 20% at the grain filling stage (Table 5). This result might indicate that the photosynthesized C in the leaves and stem is translocated to many destinations at the elongation stage, whereas the reallocation of labeled C from the leaves and stem is predominantly directly to the ears at the grain filling stage (Palta et al., 1994; Aranjuelo et al., 2013; Zhou et al., 2016). We therefore suggest that labeling of winter wheat at the grain filling stage without N fertilization may produce more uniformly ¹³C-labeled materials, whereas labeling of winter wheat at the elongation stage without N fertilization may produce more highly ¹³C-enriched plant materials, albeit with greater variability.

Variation in ¹⁵N enrichment

Compared with that under ¹³C labeling, differences in ¹⁵N excess between aboveground components under ¹⁵N labeling conditions were much smaller (Table 2). Variation among individual plants (intra-pot) and among different pots (inter-pot) was also much lower (Table 5). No significant differences in ¹⁵N excess were observed among wheat aboveground components; the greatest difference, between ears and leaves, was only 11% at the grain filling stage (Table 2). This result suggests that the ¹⁵N-enriched urea application used in this study can produce relatively uniformly labeled straw tissues at an advanced (grain filling) developmental stage in wheat. Nitrogen is a highly mobile nutrient in the plant, whereas certain other nutrients (i.e., C) are not translocated so readily within the plant. This outcome reflects that N is utilized as a nutrient by plants, whereas C is a basic element of plant photosynthesis. In contrast to plant ¹³C labeling, which requires complex environmental conditions, such as temperature, humidity, and ¹³CO₂ concentration, to be maintained throughout the labeling period (Soong et al., 2014), ¹⁵N labeling is relatively easy to achieve by means of ¹⁵N fertilization. Ideally, studies of the decomposition rate of straw tissue should be conducted, both to select uniformly enriched materials (Crozier, King & Volk, 1993; Soong et al., 2014) and to choose straw tissue similar to that of field-grown plant tissues (Crozier, King & Volk, 1993). The chemical composition of straw tissue at late maturity differs substantially from that at an early stage of maturity (Niu et al., 2014). For instance, the majority of plant N at advanced stages of maturity is associated with proteins, nucleic acids, and amino sugars (Wagger, Kissel & Smith, 1985; Girardin et al., 2009). Some nitrogenous components may exist in a form more resistant to microbial decomposition, such as N associated with cell wall material (Wagger, Kissel & Smith, 1985; Girardin et al., 2009).

As wheat growth progressed, i.e., with increase in biomass, ¹⁵N excess increased by 30.4%–41.1% (Table 2), which indicated that plant N demand was satisfied by available soil N derived from ¹⁵N fertilizer between the elongation and grain filling stages. Nitrogen demand per plant is predominantly satisfied through N uptake by roots prior to anthesis, and ear N content is augmented by the amount of N absorbed by the root system between anthesis and ear maturity (Kichey et al., 2007).

The ¹⁵N enrichment of shoots and ears differed among the developmental stages in winter wheat. At the elongation stage, no significant differences in ¹⁵N enrichment among the leaves, stems, and ears were observed. This result reflects that the redistribution of labeled N from source organs to ears was low and N in the heading ears was mainly absorbed from the soil during the vegetative growth stages (Aranjuelo et al., 2013). However, at the grain filling stage, wheat ears labeled with ¹⁵N tended to accumulate significantly higher quantities of ¹⁵N compared with other aboveground components (Table 2). Previous researchers have described similar phenomena following ¹⁵N labeling, such as in wheat (Ladd, Oades & Amato, 1981; Palta et al., 1994) and in winter wheat and grain sorghum (Wagger, Kissel & Smith, 1985). The ears are a strong N sink that affect N accumulation in the ears during grain filling, whereas the shoots and roots serve to supply nutrients to the ears (Palta et al., 1994; Aranjuelo et al., 2013; Zhou et al., 2016; Sun et al., 2018; Zhou et al., 2018). In the present experiment, 65% of the total N in grains was remobilized from the N accumulated in the shoots and roots during vegetative growth stages, whereas the remaining 35% was derived from N absorbed from the soil during the grain filling stage (Sun et al., 2018). Although ears exhibited significantly higher ¹⁵N enrichment compared with that of leaves and stems in winter wheat at the grain filling stage, the differences in ¹⁵N enrichment between leaves and stems were only slight (Table 2), which is suggestive of relatively uniform labeling in the plants because N accumulation and redistribution to the grain occurs many days after N accumulation in the leaves and stems (Butler & Ladd, 1985; Wagger, Kissel & Smith, 1985).