Introduction

Staying green in plants denotes delayed senescence, characterized by the persistence of green foliage and the gradual retention of green color1. For monocarpic plants (such as rice, wheat, and corn), “stay green” refers to the condition in which the stems and upper leaves remain green when the seeds are physiologically mature2,3. During agricultural production, the number of green leaves remaining on the plant when the seeds are physiologically mature can be counted to obtain a rough estimate of the stay-greenness of the variety4. Many studies have reported on the stay-greenness of sorghum and corn2,3,4,5,6,7,8,9,10, and research on stay-green wheat has commenced11,12,13,14,15. In stay-green crops, the leaves reach senescence slowly, resulting in large green leaf areas during the late stages, which in turn lead to longer functional and grain-filling periods, higher photosynthetic activity, and higher biological and grain yields2,3,7,10,11,14. Leaf stay-green and senescence are opposite processes. To a certain extent, senescence is an inevitable and absolute outcome, while stay-green (delayed senescence) is relative. The mechanisms and causes of stay-green are typically attributed to plant hormones and nitrogen (N). Studies have shown that the xylem of stay-green sorghum contains more cytokinin5, and leaf stay-green seems to be related to N balance, as stay-green plants display a higher N content or higher N uptake than senescent plants3,9. Stay-green corn varieties exhibit elevated cytokinin levels and diminished abscisic acid content, enhanced N uptake, improved N assimilation, increased N content, and reduced N translocation rates in leaves and stalks relative to other types4,6,7. However, some studies have shown that stay-greenness in corn only occurs under sufficient N supply, and it is unrelated to N uptake or translocation8. The consensus in the current literature is that N plays a key physiological role in delaying leaf senescence, increasing yield, and improving grain quality16. However, the relationship between stay-green and N uptake, translocation, and distribution is still unclear, and few comprehensive and systematic studies have been conducted.

Foliar feeding, also known as top dressing, has several advantages, including rapid and effective fertilization, low dosage requirements, and ease of use. It is a common fertilization technique in agricultural production, particularly when the root function of crops is impaired during late growth stages. Several studies conducted on 15N uptake and translocation using foliar feeding have shown that the15N applied to wheat, rice, and other crops using this methodology is translocated along the “leaf-sheath-stem-rachis-grain” route17,18,19,20,21. The 15N labeling technology is widely used in studies on the mechanisms of N absorption, utilization, and distribution in plants, especially in food crops such as wheat, rice, corn, and fruit trees17,19,20,21. Wang et al.20 used 15N foliar feeding to study the distribution dynamics of N uptake in different leaf positions of winter wheat. They found that the contribution of flag leaves to grain N was higher than that of other leaves. Shen19 showed that urea-N can be absorbed effectively and translocated rapidly in wheat and corn leaves. During the late growth stage of wheat, the center of growth shifts to the stalks and ears, with little nutrient flow to the root system. At this stage, the flag leaf is the main functional leaf. Compared with ammonium-N and nitrate-N, urea-N is more effectively absorbed and assimilated by plants, where it is first converted to NH3 by urease and then assimilated into amino acids by glutamine synthetase (GS) and glutamine oxoglutarate aminotransferase (GOGAT)22,23.

However, to our knowledge, no studies have been conducted on the N uptake and translocation of foliar-applied N during the late growth stages of wheat with different senescence rates. The N uptake in grain is divided into the pre-anthesis and post-anthesis phases, which contribute to the N content in the grain. Senescence usually occurs in the late stage, when root function has declined. In the present study, 15N-urea foliar feeding was performed in stay-green and early-senescent wheat varieties during the late growth stages to intensively investigate the absorption, translocation, and distribution dynamics of foliar-applied N. This experimental design aimed to examine the uptake and translocation of N fertilizer in wheat with different senescence patterns during the late growth stages, providing baseline information on the physiological relationship between N metabolism and stay-green of grain characteristics. We expected that stay-green wheat would demonstrate superior foliar 15N absorption and enhanced translocation of ingested N to grains compared to early-senescent wheat during the late development stage.

Materials and methods

Experimental design

Based on previous studies24,25, the stay-green wheat (Triticum aestivum L.) variety Yumai 66 (YM66) and the early-senescent wheat variety Wenmai 6 (WM6) were selected as the experimental materials. Seeds for both genotypes were purchased from commercial seed providers in China. Before this experiment, the research team grew and evaluated more than ten wheat cultivars purchased from the market, and YM66 and WM6 were chosen because they displayed distinct senescence phenotypes: YM66 consistently showed a strong stay-green habit, whereas WM6 exhibited pronounced early senescence. A pot experiment was conducted in a net house on the experimental farm of the Northwest A&F University, Yangling, PR China, using polyethylene barrels with a diameter of 23 cm and a height of 24 cm. The soil utilized in the experiment was categorized as loam according to the USDA soil texture classification, sourced from the 10–20 cm topsoil layer, and exhibited the following properties: pH 7.3, organic matter 15.8 g·kg−1, total N 0.66 g·kg−1, total phosphorus (P₂O₅) 0.21 g·kg−1, alkaline hydrolyzable N 68 mg·kg−1, accessible phosphorus 16 mg·kg−1, rapidly available potassium 145 mg·kg−1, and cation exchange capacity 18.4 cmol·kg−1. After air-drying, the soil was sieved and mixed with sand (soil: sand = 7:3 by weight). This was followed by adding urea (0.347 g·kg− 1 soil) and monopotassium phosphate (0.2 g·kg−1 soil). Eight kilograms of soil were added to each pot and saturated with water, and 14 wheat seeds were then planted and covered with 1.29 kg of soil. Thirty replicate pots were prepared for each variety. Seeds were sown on October 19, 2021, and an adequate water supply was maintained (soil water content was approximately 21%, reaching 70–75% of the field capacity). The tillers were removed during the growing stages, leaving only the main stem. The plants were harvested in late May or early June of the subsequent year, aligning with physiological grain maturity (about 45–50 days post-anthesis).

15N feeding experiments

The N uptake in plants is divided into two phases: pre-anthesis and post-anthesis. The flag leaf is the central functional leaf in the late growth stage of wheat. Therefore, 15N-urea foliar feeding was performed on wheat flag leaves 5–7 days before anthesis, and samples were collected after anthesis to study the translocation and distribution patterns of the 15N absorbed before anthesis. The 15N-urea used for foliar feeding (15N abundance of 10.28%) was purchased from the Shanghai Research Institute of Chemical Industry, Shanghai, PR China. All treatments utilized a singular concentration of 15N-urea solution, derived from a 20 g·L⁻¹ 15N-urea stock solution, with no additional concentrations employed. For the pre-anthesis experiment, 15N feeding was performed using the method described by Shen19. To maintain consistency among samples, all flag leaves were tagged at the same developmental stage (completely expanded flag leaf at anthesis) during a brief time frame, just before sunset. Each plant was administered the same concentration of 15N-urea, a uniform solution volume, and a consistent brushing technique. The labeling sequence was maintained uniformly for YM66 and WM6, and ambient conditions were steady throughout the feeding process. Following sampling, all tissues from the same time point were desiccated, pulverized, and analyzed for15N in the same batch to minimize technical variability. Starting from 4:00 pm on a cloudy day, the flag leaves of the wheat plants were fed with an 8 mL solution containing 37.28 mg N (from 4 mL of 20 g·L− 1 15N-urea), 18.21 mg P, and 22.98 mg K (from 4 mL of 20 g·L− 1 KH2PO4). N, K, and P were combined to improve foliar feeding absorption. The solution was placed in a 15-mL tube, and a small amount of surfactant was added. An ink brush was then dipped in the solution and used for repeated applications. The solution was applied to the front and back surfaces of the flag leaf (better absorption occurred on the back) from the base to the tip of the leaf until the solution was spent. After four h of absorption, the leaves were washed sequentially with water, detergent, water, and 1% HCl, and subsequently washed three times with deionized water to remove the N, P, and K from the leaf surfaces26. Twenty plants of each variety were labeled and sampled at anthesis, at 10 and 20 days after anthesis, and at physiological grain maturity (approximately 45–50 days after anthesis. Five plants were sampled each time to measure several parameters (as indicated in the following sub-section). For post-anthesis 15N feeding trials, plants that bloomed concurrently and had analogous growth patterns during the flowering phase were chosen and labeled. The administration of 15N to flag leaves was conducted as a single foliar treatment 10 days post-anthesis, using the approach described above. A volume of 10 mL of 15N-urea solution was administered per plant to saturate the flag leaf surface thoroughly. Samples were taken at 12-, 24-, and 48-hours post-labeling, in addition to at maturity, to assess several parameters (as indicated in the following subsection).

Parameter measurements

Green leaf area

From the start of the flowering period, the length and width of all green leaves of the tagged plants were measured with a ruler every 7 days, and the leaf area was calculated using the coefficient method (leaf area = leaf length × leaf width × 0.83)27.

Leaf chlorophyll content

Chlorophyll was extracted by shaking in a 1:1 (v/v) ethanol: acetone mixture overnight, and the chlorophyll content was determined as described by Holden28.

Fresh and dry weights

Grain, leaf, and stalk (including rachis and husk) samples of the15N-labeled plants were collected regularly, and their fresh weights were measured. Dry weights were measured after 10 min of sterilization at 105 °C, followed by drying at 80 °C until the samples reached a constant weight.

Total N content and15N abundance

The leaf, grain, and stalk (including rachis and husk) samples collected at each time point were dried, weighed, ground, and passed through a 0.3-mm sieve. The N content of each sample was determined using the Kjeldahl29 method. A 0.1000 g sample was weighed and placed into a digestion tube containing 5 mL of 18.4 mol·L− 1 H2SO4 with catalyst, and then heated at 200–300 °C for three hours to ensure that total N was converted to (NH4)2SO4. The digested solution was transferred to a micro-Kjeldahl distillation flask and then alkalized with 10 mL of 40% NaOH. The resulting ammonia (NH3) was distilled into 5.0 mL of standard H3BO3 (2% w/v). Using a micro-burette, NH3 was determined by titration with 0.0100 mol·L− 1 HCl. The determined value was referred to as the ammonium content.

To determine15N abundance, 0.3 g of each sample was weighed and placed in a digestion tube. After adding 4 mL concentrated H2SO4 and 2 g of catalyst mixture (K2SO4:Se = 500:1), the blend was digested for eight h and then absorbed with 0.01 mol·L− 1 dilute H2SO4. The 15N abundance was then measured using an elemental analyzer-isotope ratio mass spectrometer (Finnigan-MAT-25, Finnigan MAT Corp., Bremen, Germany).

As the proportion of leaf-absorbed nutrients flowing to the root system during the late growth stage in wheat is very low, and our experiment only investigated the transport of leaf-absorbed N to the grain, the sum of the 15N content in the grains, stalks (including rachis and husk), and leaves was regarded as the total amount of 15N absorbed by the plant30,31.

Relevant parameters were calculated as follows:

$$\:\text{Ndff\%=}\frac{\text{\%\:abundance\:of\:}\text{15}\text{N\:in\:the\:sample-\%\:natural\:abundance\:of\:}\text{15}\text{N}}{\text{\%\:abundance\:of\:}\text{15}\text{N\:in\:fertilizer-\%\:natural\:abundance\:of\:}\text{15}\text{N}}\times \text{100}$$

where the Ndff% is the mass% of N in the sample that originated from labeled N

$$\:\text{The\:cumulative\:amount\:of\:N\:in\:the\:sample\:(g)=sample\:dry\:weight\:(g)}\times{\text {N content in the sample}}$$
$$\:\text{Absorbed\:}\text{15}\text{N\:(mg)\:}\text{=cumulative\:amount\:of\:N\:(g)}\times {Ndff\%\times 1000}$$
$$\:\text{15}\text{N\:distribution\:percentage\:(\%)\:=}\frac{\text{15}\text{N\:absorbed\:by\:the\:sample\:(mg)}}{\text{total\:}\text{15}\text{N\:absorbed\:by\:the\:plant\:(mg)}}\times{\:100}$$
$$\:\text{15N-reserve\:translocation\:percentage}\text{=\:}\text{15N\:anthesis}\text{-}\text{15N\:maturity}.$$

where 15N (anthesis) and 15N (maturity) represent the percentages of 15N distribution in the same organ at anthesis and at physiological maturity, respectively.

The rate of decline in N content in each organ from anthesis to maturity was determined as follows:

$$\:\text{Decline\:rate}\text{=}\frac{\text{N\:anthesis-N\:maturity}}{\text{Days\:from\:anthesis\:to\:maturity}}$$

Data processing

The results were analyzed for variance using the SAS statistical analysis package (SAS Institute, Cary, NC, USA). Data from each sampling date were analyzed separately. A one-way ANOVA was employed to assess variations among types at each time interval. Mean values were evaluated for significant differences using the least significant difference test, with a significance threshold of P < 0.05. Before conducting ANOVA, the data were assessed for normality using the Shapiro–Wilk test and for homogeneity of variances via Levene’s test to ensure that the requisite assumptions for ANOVA were met.

Results

Green leaf area and chlorophyll content after anthesis

The green leaf area and chlorophyll concentration progressively diminished following anthesis (Fig. 1 and Table S1). At 0 days post-anthesis, YM66 exhibited a superior chlorophyll content of around 40 mg g−1 dry weight, in contrast to roughly 30 mg g−1 in WM6. After 7 days, chlorophyll content decreased to approximately 20 mg g−1 in YM66 and 15 mg g−1 in WM6, and further diminished to around 10 mg g−1 and 8 mg g−1, respectively, at 14 days. Chlorophyll content reached its nadir at 21 days post-anthesis, averaging approximately 5 mg g−1 in YM66 and 3 mg g−1 in WM6. The mean values for green leaf area exhibited a similar pattern, with YM66 consistently demonstrating greater values at all time points, especially at 0 and 7 days post-anthesis. Nonetheless, the green leaf area and chlorophyll content were continuously superior in YM66 compared to WM6 at every sampling interval. WM6 exhibited a quantifiable green leaf area and chlorophyll concentration during the early phase of grain filling; however, the significant decrease after 14 days suggests that WM6 had progressed to a more advanced state of senescence by the mid-to-late grain filling period.

Fig. 1
figure 1

Green leaf area and chlorophyll content of stay-green (YM66) and early-senescent (WM6) wheat varieties after anthesis. Values are mean ± SD. Different symbols denote significance at P < 0.05.

Full size image

Nitrogen accumulation and content

At anthesis, total N accumulation in the aboveground organs of each plant was about 30 mg in WM6 and 45 mg in YM66 (Fig. 2A and Table S2), and it increased continuously after anthesis, reaching 38 mg and 70 mg at maturity, respectively. Thus, after anthesis, cumulative total N accumulation in each plant was 8 mg in WM6 and 25 mg in YM66. Moreover, YM66 exhibited a prolonged N absorption and accumulation period, which could last up to 20 days after anthesis. In comparison, WM6 had a short N absorption and accumulation period, lasting only 10 days after anthesis. N absorbed and assimilated before anthesis is stored in the vegetative organs as a reserve and subsequently translocated to the grain after anthesis. As N was continuously translocated from the leaves and stalk into the grain after anthesis, the N content of the leaves, stalk, and rachis decreased continuously after anthesis (Fig. 2B-C). In contrast, the N content of the grain gradually increased in both wheat varieties (Fig. 2D). From anthesis to maturity, the leaf N concentration in YM66 decreased from 4.8% to 2.3%, representing a fall of 2.5% points and a decline rate of 0.083% points per day. In WM6, leaf N content decreased from 4.6% to 1.6%, representing a decline of 3.0% points and a rate of 0.100% points per day. The N content in the stalk and rachis of YM66 diminished from 1.6% to 0.9%, reflecting a fall of 0.7% points and a daily decline rate of 0.023% points. In WM6, the N content in the stalk and rachis diminished from 1.1% to 0.8%, representing a decline of 0.3% points and a decline rate of 0.010% points per day. A minor reduction in the N concentration of the grain during maturity may be ascribed to carbohydrate buildup and an increase in overall weight. The drop rates indicate that WM6 depletes leaf N more swiftly than YM66, whereas YM66 sustains elevated N concentrations in the stalk and rachis during the grain-filling period. Consequently, YM66 maintains higher vegetative N stores during the initial stages of grain filling, while WM6 exhibits a more pronounced decrease in vegetative N following anthesis.

Fig. 2
figure 2

Variations in total nitrogen (N) following anthesis in stay-green (YM66) and early-senescent (WM6) wheat. (A) the total nitrogen accumulation in the entire aboveground portion of the plant (mg plant⁻¹). Panels B, C, and D illustrate the total nitrogen concentration (%) in the leaf (B), stem and rachis (C), and grain (D). Measurements were conducted at 0, 10, and 20 days post-anthesis and at physiological maturity. Values are expressed as mean ± standard deviation (SD). Distinct symbols denote substantial differences among variants at P < 0.05.

Full size image

15N absorption with pre-anthesis foliar feeding

Pre-anthesis 15N absorption and assimilation in the flag leaves of stay-green YM66 were significantly higher than those of early-senescent WM6 (Table 1). About 45–60% of the 15N absorbed before anthesis was stored in the stalk and rachis, and 40–55% was stored in the leaves at anthesis (Table 2). The distribution percentage of 15N in the leaves, stalk, and rachis continued to decrease after anthesis, and the 15N distribution percentage in the grain continued to increase, indicating that the 15N reserves in the leaves and stalk were translocated to the grain after anthesis. This pattern suggests a distinct decrease in 15N maintained in vegetative organs from anthesis to maturity, accompanied by a corresponding increase in 15N recovered in the grain, exemplifying post-anthesis N remobilization. At maturity, over 50% of the 15N-reserve had been translocated to the grain, with about 19%–30% retention in the leaves and 18%–24.6% in the stalk (Table 2). This indicates that the N reserve assimilated before anthesis contributed more to the grain yield. Further comparisons between varieties showed that 59.44% of the 15N absorbed by the leaves before anthesis was already stored in the stalk of YM66, whereas only 45.37% was stored in the stalk of WM6 (Table 2). The 15N-reserve translocation percentage from a given organ is the difference between the 15N distribution percentage at anthesis and maturity. The 15N-reserve translocation percentage in the stalk of YM66 (about 34.8%) was higher than that of WM6 (27.2%). The 15N-reserve translocation percentage in the leaves of both wheat varieties was low (21–24%) (Table 2).

Table 1 Total 15N assimilation in the flag leaves of stay-green (YM66) and early-senescent (WM6) wheat varieties before and after anthesis.
Full size table
Table 2 Distribution percentage of 15N reserves in the stay-green (YM66) and early-senescent (WM6) wheat varieties after anthesis.
Full size table

15N absorption and translocation with post-anthesis foliar feeding

The application of 15N-urea to the flag leaves on day 10 post-anthesis indicated that 15N absorption was inferior to that recorded before anthesis, but YM66 exhibited more 15N absorption than WM6 (Table 1). At 12 h after labeling, approximately 32% of the 15N was retained in the leaves, and 60–65% had been translocated to the stalk and rachis. As the 15N distribution percentage in the stalk and rachis continued to decrease, the 15N distribution percentage in the grain continued to increase. At maturity, 38–46% of 15N was contained in the leaves, and only about 12% of 15N was included in the stalk and rachis. The 15N distribution percentage in the grain was about 40–48.7% (Table 3). This showed that the post-anthesis absorption and assimilation of N made a minor contribution to the grain’s N content compared to that during pre-anthesis, and that the translocation of assimilated N occurred primarily in the stalk. From 12 to 24 h after labeling, the 15N distribution percentage in the stalk and rachis decreased sharply, reaching approximately 23–30%. The 15N distribution percentage in the grain increased by only 9% and in the leaves by about 16–21%. Thus, the 15N in the stalk and rachis was translocated into the grain and back to the leaves. The 15N distribution percentage in each organ did not change significantly between 24 and 48 h after the labeling period, indicating that the translocation of N absorbed after anthesis occurred in a 24-hour cycle. Further comparisons between the two varieties showed that, at maturity, about 48.7% of the 15N absorbed after anthesis was in the grain of YM66, and 38.8% was retained in the leaves of YM66. In contrast, about 40% of the 15N absorbed after anthesis was in the grain of WM6, and 45.5% was retained in the leaves of WM6. The 15N distribution percentage in the stalk and rachis was about 12% in both varieties. These results indicated that the translocation percentage of 15N absorbed post-anthesis in YM66 was relatively high. In contrast, the 15N absorbed post-anthesis in WM6 had a higher retention percentage in the leaves and a lower translocation percentage to the grain than those in YM66. In addition, the 15N-reserves absorbed before anthesis were at almost the same level in the leaves and stalk, whereas the 15N absorbed after anthesis were retained primarily in the leaves and less so in the stalk (Tables 2 and 3). The retention percentage of 15N in the leaves of WM6 was higher than that of YM66, regardless of whether it was absorbed before or after anthesis.

Table 3 Distribution percentage of 15N assimilates in the stay-green (YM66) and early-senescent (WM6) wheat varieties after anthesis.
Full size table

Comparison of yield component factors

Stay-green YM66 exhibited a substantial kernel count per spike, along with elevated grain weight, yield, and total biomass, signifying its superiority as a variety (Table 4). YM66 yielded 29.6 kernels per spike, in contrast to 26.5 in WM6, and its grain weight attained 37.83 g per 1000 grains, compared to 32.03 g in WM6. The yield of YM66 was 11.15 g per 10 stems, surpassing the 8.71 g recorded for WM6. Total aboveground biomass varied, with YM66 yielding 25.42 g per 10 stems, whilst WM6 produced 21.44 g. YM66 exhibited a superior harvest index of 43.86%, whereas WM6 attained 40.62%, signifying enhanced efficiency in converting biomass into grain. This performance was attributed to its enhanced N absorption capacity and increased N buildup both before and following anthesis. Conversely, WM6 exhibited diminished kernel yield, total biomass, and harvest index, aligning with its impaired N absorption and accumulation capabilities. These yield advantages in YM66 were directly related to its better stay-green features, as higher post-anthesis leaf retention and chlorophyll preservation were associated with more kernels, larger grains, and higher total biomass when compared to WM6.

Table 4 In the pot experiment, yield components of stay-green (YM66) and early-senescent (WM6) wheat varieties.
Full size table

Discussion

In plants, N uptake is divided into two phases: pre-anthesis and post-anthesis. Most studies have reported that the N in wheat grain primarily derives from the N absorbed and assimilated before anthesis, while N absorbed and assimilated after anthesis provides a relatively small contribution to N content in the grain32,33. Prior studies have definitively demonstrated that wheat continues to assimilate N post-anthesis; however, the impact of this uptake on final grain N content is limited. Numerous independent studies indicate that N uptake after anthesis generally contributes just 10 to 30% of the total N in grain, while pre-anthesis reserves account for the bulk34,35,36. Similar results were obtained in the present study. Of the 15N absorbed by the flag leaves before anthesis, 51–56% was translocated to the grain, and of the 15N absorbed after anthesis, 40–48% was translocated to the grain at maturity. This indicates that it was effectively translocated, thereby increasing the grain’s N content. However, the translocation of the N reserve absorbed before anthesis to the grain remained more efficient. The N metabolism in wheat occurs in a coordinated sequence, commencing with root absorption of nitrate and ammonium, followed by reduction and assimilation in leaves via the nitrate reductase and GS–GOGAT pathways, where inorganic N is incorporated into amino acids and proteins37,38. GS and GOGAT are essential enzymes in the N assimilation process following anthesis in wheat, particularly during the grain-filling stage39,40. GS facilitates the transformation of inorganic N (ammonium) into glutamine, an essential amino acid that serves as a N donor in multiple metabolic pathways41. GOGAT collaborates with GS to transfer the amino group from glutamine to 2-oxoglutarate, resulting in the formation of glutamate, a pivotal amino acid in N metabolism40. Both enzymes are vital for sustaining the N supply necessary for protein synthesis and chlorophyll formation, which are critical for ongoing photosynthetic activity and plant growth during grain development. Research indicates a strong correlation between GS activity and grain protein content in wheat42, while GOGAT activity facilitates continuous N assimilation into amino acids during phases of restricted N uptake, especially in stay-green genotypes40. In wheat varieties such as YM66, characterized by extended leaf greenness, the persistent activity of GS and GOGAT helps maintain chlorophyll levels and enhances N remobilization to the grain, thereby sustaining elevated yield and grain N content even in late growth stages. These absorbed N compounds are subsequently transferred from vegetative organs to developing grains through the phloem, especially during grain filling, when root uptake declines and stem and leaf N reserves become the dominant source of grain N43,44. During grain development, proteins and amino acids stored in leaves, sheaths, and stems are progressively degraded and remobilized to the spike, providing the majority of grain N43,44. Thus, uptake, assimilation, translocation, and remobilization are closely interconnected processes controlling grain N supply. The15N distribution patterns observed in leaves, stalks, rachis, and grains in the present study align with these established phases. Higher15N incorporation in YM66 indicates more efficient N assimilation, larger15N pools in its stalk and rachis reflect stronger temporary storage and translocation capacities, and greater final grain15N recovery compared with WM6 demonstrates superior remobilization efficiency. This mechanistic sequence explains why the stay-green genotype exhibited improved N dynamics throughout the grain-filling process.

With the post-anthesis foliar application of 15N-urea, 15N was translocated to the stalk and rachis 12 h (nighttime) after labeling and translocated back to the leaves 12–24 h (daytime) after labeling. We deduced that such reverse translocation might be due to the large number of vascular bundles in the stalk and rachis, which contain many thin-walled cells capable of storing large amounts of substances. The 15N temporarily stored in the stalk and rachis at night was translocated back to the leaves for rapid assimilation during the day, when ATP and NADPH were produced via photosynthesis. Prior research utilizin g15N tracers has demonstrated that wheat swiftly incorporates absorbed15N into amino acids and proteins within leaves and stems, before remobilizing these compounds to the grain during the filling stage43,44. Therefore, the stalk is a temporary reserve of N, and the leaves are the primary assimilation site of N aboveground. The change in the distribution percentage of 15N at 24–48 h after labeling was insignificant, indicating that the translocation of N absorbed by the leaves post-anthesis occurred in a 24 h cycle. This reverse flow indicates a coordinated interaction between the xylem and phloem, rather than just a unidirectional transit. During the night, diminished transpiration facilitates the xylem’s transport of nitrogenous chemicals into the stem for temporary storage. Conversely, photosynthesis during daylight presumably facilitates phloem loading and the transport of nitrogen back to the leaves for reassimilation45. This interpretation, bolstered by the unique 24-hour isotopic pattern observed, would benefit from future research that includes complementary physiological data—such as diurnal assessments of xylem and phloem sap N composition or expression profiles of critical N transporter genes—to offer direct mechanistic validation beyond isotopic tracing. This hypothesis posits that N can shift between the two circulatory systems through xylem–phloem transfer zones in the stem, facilitating the reloading of stored N from the xylem into the phloem for transport to active assimilation sites46. The interplay between xylem and phloem provides a physiological basis for the documented cyclical reversal of 15N47 and supports the concept of the stem acting as a dynamic buffer rather than a passive reservoir during grain filling45,48,49.

When the 15N distribution percentage in different organs was investigated, it was found that the15N reserves absorbed before anthesis were retained in the stalk and rachis (18%–25%) as much as in the leaves (19%–30%) at maturity. However, the 15N absorbed after anthesis was retained primarily in the leaves (38.8%–45.5%), with less retention in the stalk and rachis (about 12%). In plants, the central role of N is to maintain cell structure and function, and to carry out protein turnover. It is possible that the plant uses the newly absorbed N for protein turnover and primarily transports the N stored before anthesis to the grain22. As the N absorbed by the wheat before anthesis, whether stored in the stalk or leaves, was mobilized and translocated to grains as much as possible during the late growth stages, similar amounts of N were retained in the two vegetative organs. In our experiment, the amount of 15N collected in the stalk and rachis was strongly related to its later remobilization to the grain. This trend is consistent with previous research demonstrating that the stem is a large temporary N reservoir in wheat, and that the size and mobility of this pool greatly influence the N transported to the grain48,50,51. YM66 stored similar or slightly greater amounts of 15N in the stalk and rachis during the early post-labeling period, and by maturity, it had exported a larger proportion of this stored N to the grain, consistent with the behavior of efficient remobilizing wheat genotypes48. In contrast, WM6 preserved a larger proportion of 15N in the stalk and rachis at maturity, suggesting restricted remobilization and poorer sink strength. These findings show that the effective export of 15N from the stalk and rachis is a significant physiological benefit of the stay-green phenotype. Wheat significantly depends on remobilization during grain filling, with 70 to 90% of grain N generally derived from N reserves in leaves, stems, and sheaths; the efficacy of this remobilization markedly affects the ultimate grain N content and yield52. This study assessed 15N remobilization by calculating the 15N distribution and the 15N-reserve translocation percentages, which collectively reflect the transfer of absorbed 15N from vegetative organs to the grain. These measures delineate the remobilization behavior linked to pre- and post-anthesis labeling34,53. However, we did find significant differences between the two wheat varieties. While YM66 exhibited increased total N accumulation and 15N absorption before and after anthesis, these findings indicate N acquisition and remobilization rather than direct biochemical proof of improved assimilation. The lack of assays for key enzyme markers of N assimilation, such as GS and GOGAT activity, requires that this finding be considered an indirect inference rather than a direct physiological evaluation. The elevated residual 15N in WM6 post-anthesis indicates a diminished remobilization efficiency relative to YM66. Premature senescence in WM6 presumably diminishes sink strength in the developing grain, impairing its ability to draw and absorb N from vegetative tissues54. As sink demand diminishes, N becomes sequestered in the leaves instead of being effectively translocated to the grain, resulting in increased retention after anthesis55. This corresponds with the notion that expedited leaf senescence constrains phloem loading and transport capacity, thereby limiting N remobilization during grain filling56.

The N absorbed after anthesis is primarily retained in the leaves and can be used for protein turnover or maintaining cell function, which is markedly different from the pattern of carbon translocation after assimilation12. In stay-green genotypes like YM66, enhanced N turnover may be facilitated by the prolonged activity of essential N-assimilating enzymes, notably GS and GOGAT, which persist in assimilating and recycling ammonium into amino acids throughout the grain-filling phase37,57. Prolonged GS/GOGAT activity maintains chlorophyll concentrations, protein turnover, and metabolic function, hence preserving leaf photosynthetic efficiency and facilitating continuous N remobilization to the grain33. Conversely, the premature senescence in WM6 likely correlates with a decrease in GS and GOGAT activity, which diminishes N recycling and results in elevated residual 15N in leaves due to impaired assimilatory capacity33,57.

At least in sorghum, the stay-green phenotype has been linked to greater N uptake during the post-anthesis period2. Numerous previous studies have shown that N uptake in wheat and other cereals is strongly correlated with biomass growth and early vigor58,59,60,61,62. The present study demonstrated that stay-green wheat exhibited a large biomass and high yield, as well as a strong 15N absorption and assimilation ability before and after anthesis, indicating not only early vigor but also strong late vigor. This demonstrates a positive functional link between the stay-green trait and yield formation. Extended leaf greenness in YM66 enabled prolonged N absorption and translated into increased grain yield components compared to WM6. Therefore, the high amount of N in the plants and the high N content of the stalk and rachis from anthesis to maturity (Fig. 2) may form the physiological basis for stay-green in YM66. As the grain prioritizes and translocates N from the stalk, this nutrient is mobilized and translocated from the leaves only during conditions of insufficient N22. Therefore, the N content of YM66 leaves decreased slowly (Fig. 2), consistent with studies on corn and sorghum2,6,9,63,64. The translocation percentage of 15N absorbed both before and after anthesis by the leaves in stay-green YM66 was higher than that in early-senescent WM6 during the grain-filling stage. This may be because YM66 absorbed and assimilated total N, and the excess parts that are not needed to maintain cell function must be transported away. However, this result is contrary to the results obtained for corn6,63,64. Differences in the research methods used may cause such inconsistency. If the 15N label is not removed after labeling, the plant will continue to absorb and translocate N. The translocation percentage is calculated using the total amount of 15N absorbed over a relatively long period or the plant’s total N as the denominator. A more plausible explanation is the significant disparity in source and sink activities between the two genotypes. YM66 exhibits an expanded green leaf area and elevated chlorophyll concentration post-anthesis (Fig. 1), accompanied by increased N content in both leaves and stems (Fig. 2), indicating enhanced photosynthetic and remobilization capabilities. Concurrently, YM66 generates a more robust sink, evidenced by its elevated kernel count, grain weight, and overall biomass (Table 4), thereby augmenting N requirements during the grain-filling phase. A robust sink promotes N withdrawal from source organs, leading to enhanced 15N translocation. Conversely, WM6 exhibits early senescence, experiences a more rapid loss of chlorophyll, and possesses a diminished grain sink, thereby constraining its capacity to remobilize stored N. The disparities in source strength, sink demand, and transport efficiency elucidate the elevated 15N translocation percentage noted in YM66. As stay-green wheat has a stronger ability to absorb and assimilate N, it has a lower translocation percentage due to a larger denominator. In the present study, 15N was fed to the leaves for a short period, and the residual labeling material was subsequently washed away to fix the total amount of 15N absorbed. Therefore, the N translocation percentage measured here should be more accurate. That is why we did not apply 15N to the soil in our experiment.

Although applying urea to the soil is a common practice in agricultural settings, this research can provide valuable insights for agricultural applications. The application of N fertilizer during wheat planting can increase grain yield and grain N content. In contrast, N fertilizer applied before and after anthesis cannot increase yield but can increase grain N content, which is beneficial for bread wheat65. As early-senescent WM6 had weaker N absorption and assimilation abilities during the late growth stages and less total N than YM66 (Fig. 2), its 15N applied via foliar feeding before or after anthesis was retained mainly in the leaves (Table 4), which might have been used for protein turnover and maintaining leaf function. Therefore, N foliar feeding of WM6 around anthesis enables the delay of senescence and the maintenance of leaf function. In YM66, which has a relatively low grain N content (Fig. 2), most of the absorbed from 15N pre-anthesis foliar feeding could be transported to the grain (Table 2). This study did not assess grain N concentration with supplementary foliar N treatments, thereby failing to demonstrate that foliar N administration can enhance the grain N content of YM66. Our findings solely delineate the translocation pattern of the 15N that the leaves have absorbed. Administering 15N-urea to a single flag leaf is labor-intensive and must occur at dusk, resulting in a limited sample size. The single-leaf feeding method facilitates exact regulation of the administered 15N dosage, and we employed uniform solution concentrations, volumes, and feeding protocols for all plants. We selected plants exhibiting analogous growth patterns and processed all samples from each time point in the same analytical batch to enhance comparability and minimize technical variance. Despite the limited sample sizes in the current study, due to the extensive time required to administer urea to a single leaf and the necessity of conducting this procedure at dusk, we employed uniform 15N-feeding conditions across all plants to ensure consistency, as feeding urea to a single leaf produces more accurate results than alternative methods. To clarify, all flag leaves were designated at the equivalent developmental stage and within the same temporal window, using uniform solution concentrations and volumes, while selecting plants with similar growth patterns to ensure comparability. A further weakness of the study was the lack of assays of N-assimilating enzymes, specifically GS and GOGAT. While these assays were beyond the parameters of the current experiment, we concur that they are crucial for mechanistically elucidating variations in N assimilation capacity. Subsequent investigations will incorporate enzyme-level analysis to corroborate the physiological conclusions derived from the 15N data. The adaptability of stay-green wheat to abiotic stress will be investigated in future studies to promote the cultivation of stay-green wheat in agricultural production. The stay-green phenomenon in wheat is highly complex, as it involves intricate genetic mechanisms and is influenced by numerous factors. The stay-green characteristic of YM66 is attributed to its strong N assimilation ability at both early and late stages, resulting in a high amount of N and enabling the maintenance of a high N content in stalks. Except for a certain amount of N for protein turnover, N should be mobilized for plant productivity. That has led to a high translocation and remobilization percentage of 15N from the stalk into the grain of YM66. In contrast, early-senescent wheat had poorer N assimilation ability, which led to more 15N retained in the leaves for the maintenance of leaf function, lower N content in the stalk, and a lower translocation and remobilization percentage of 15N during the late stages of growth (Fig. 3).

Fig. 3
figure 3

Illustration of the foliar feeding treatment in wheat plants, where the green circle represents chlorophyll in the flag leaves, the blue circle indicates the presence of nitrogen, and the yellow circle represents the applied 15N-urea solution. Dark arrows mean high absorption rate while light one mean low absorption rates. This treatment allows for rapid absorption of nitrogen through the leaf surface, facilitating translocation to other parts of the plant during late growth stages. The spike numbers reveal the fraction of absorbed 15N that was remobilized to the grain, not the overall spike nitrogen buildup.

Full size image

Conclusion

This research examined foliar 15N absorption, redistribution, and remobilization in stay-green (YM66) and early-senescent (WM6) wheat throughout the grain-filling phase by isotope tracing. The findings indicate that, in both genotypes, grain N primarily originated from N assimilated before anthesis, while N absorbed post-anthesis accounted for a lesser fraction of the final grain N. In comparison to WM6, YM66 exhibited higher foliar ¹⁵N absorption before and following anthesis, enhanced retention of absorbed N in vegetative tissues, and superior redistribution to growing grains during the grain filling period. The stay-green genotype exhibited an expanded green leaf area, elevated chlorophyll content, and prolonged N reserves in leaves and stems, which correlated with unique patterns of ¹⁵N allocation and remobilization compared to the early-senescent genotype. The findings highlight distinct physiological variations in whole-plant N dynamics across different senescence types, emphasizing the importance of pre-anthesis N stocks for grain N synthesis. The current work focuses on isotope-based N fluxes. It does not directly measure source-sink dynamics or the activities of essential N-assimilating enzymes, such as glutamine synthetase and glutamate synthase. Consequently, the physiological mechanisms responsible for the observed differences are inferred from patterns of nitrogen (N) absorption, distribution, and remobilization, rather than direct biochemical assessments. Notwithstanding this restriction, the ¹⁵N tracer methodology offers compelling evidence for genotype-specific differences in N use efficiency throughout the late development phase. Future research incorporating enzyme-level analyses, source–sink activity assessments, and molecular methodologies will be essential to elucidate the biochemical and regulatory mechanisms governing N metabolism in stay-green wheat and to further assess its potential for enhancing N use efficiency and grain yield.