Optimal Nitrogen Accumulation and Remobilization Can Synergistically Improve Maize Yield and Nitrogen-Use Efficiency Under Low-Nitrogen Conditions

Abstract

Increasing the nitrogen (N) use efficiency (NUE) of modern high-yield maize hybrids is essential for food security and reducing environmental risks. However, the relationship between dry matter (DM), N accumulation, and reallocation among different high-yield maize hybrids and NUE, particularly under various N fertilization levels, is not well understood. The field experiment was conducted in Jilin Province, Northeast China. In this study, two maize hybrids, Zhengdan958 (ZD958) and Tie 391 (T391), were grown under four N fertilizer levels: 0 (NN), 120 (LN), 240 (MN), and 360 (HN) kg ha⁻¹. We examined the effects of N input on grain yield, NUE, DM, and N accumulation, partitioning, and reallocation of these two high-yielding maize hybrids during the 2023–2024 growing season. The results showed that N input significantly increased grain yield but reduced NUE. There was no significant difference in yield and NUE between the two maize hybrids at the HN level. However, under LN conditions, the grain yield and NUE of ZD958 were higher by 16.2% and 15.6%, respectively, compared to T391. Meanwhile, ZD958 exhibited greater per-silking and post-silking DM (5.0% and 7.9%) and N accumulation (11.6% and 32.7%), as well as a higher amount of reallocated DM (45.6%) and N (17.5%) compared to T391. Moreover, 15.5–38.1% of grain N for ZD958 and 17.2–46.7% for T391 still needed to be reallocated from vegetative organs, with a larger fraction coming from the stem rather than the leaves. The middle leaves and lower stems of the canopy tended to reallocate more N to the grain, and lower-layer stem N reallocation was significantly related to grain yield. In conclusion, higher accumulation of DM and N, along with greater N reallocation—especially from the lower-layer stem—could be regarded as important traits in maize breeding to improve the NUE of high-yield maize hybrids under insufficient N supply.

1. Introduction

Maize (Zea mays L.) is the second most produced cereal crop and serves as both a staple food and a feed source [1]. With the anticipated growth of the global population and the expansion of the livestock and poultry industries, the demand for maize production is steadily increasing. In the next two decades, this demand is projected to rise at a rate of 4% to 6% per year [2]. Nitrogen (N) is an essential element for maize growth, and the application rate of N fertilizer is critical for achieving high grain yields [3,4]. However, the excessive application of N fertilizer not only leads to higher production costs and lower nitrogen-use efficiency (NUE), but also results in environmental pollution [4,5]. Currently, researchers in molecular genetics and biotechnology have modified plants to overexpress nitrate transporter genes, thereby increasing N absorption and assimilation [3]. Additionally, manipulating the expression of genes that encode crucial enzymes in the N assimilation pathway, such as glutamine synthetase and nitrate reductase, has shown promise in improving the N efficiency in plants [6]. In terms of cultivation methods, selecting low-N tolerant maize hybrids in combination with optimized N input is the most effective strategy to increase yield while reducing N fertilizer application.

Genotypic differences in maize yield exist, particularly in response to various N levels [1,3]. Modern high-yield maize hybrids typically exhibit enhanced abilities for dry matter (DM) accumulation, especially showing higher DM accumulation during the post-silking period [7,8,9]. This results in increased DM acculturation in the grains, enabling them to achieve higher grain yields compared to older maize hybrids, particularly under high-N conditions [9,10]. Moreover, modern maize hybrids typically experience delayed leaf senescence, allowing them to maintain a high leaf area index for a longer duration. This, in turn, may lead to the N uptake capacity of modern high-yield maize hybrids being likely higher than that of older maize hybrids [11,12]. However, less is known about the differences in the accumulation and reallocation of DM and N among modern high-yield maize hybrids with different N efficiencies, particularly under varying N levels.

Nitrogen is an essential nutrient for maize [13,14,15]. N levels significantly influence maize grain yield and grain N content by affecting photosynthetic capacity and plant growth status [16,17,18]. For instance, increased N availability can prolong the leaf stay-green period and expand the leaf area, which in turn enhances N accumulation and reallocation. However, both excessive and deficient N levels can inhibit N reallocation [13]. Moreover, it is well known that the distribution of N in leaves is uneven; the middle-layer leaves exhibit the highest N content, which gradually decreases toward the top leaves, resembling a bell-shaped curve based on leaf rank [17]. There is no doubt that the N distribution pattern among leaves at varying N levels is influenced by N concentration, as this concentration directly affects physiological characteristics such as leaf area and leaf senescence [18,19,20,21]. Nevertheless, the differences in the accumulation and reallocation of N in different parts (upper, middle, and lower) and in various vegetative organs (leaves and stems) among high-yielding maize hybrids in response to N levels remain unclear.

In our research, we identified two high-yielding modern maize hybrids (ZD958 and T391) that exhibited similar grain yields and NUE under high-N conditions. However, under low-N conditions, ZD958 demonstrated higher grain yield and NUE compared to T391. In the present study, the differences in their accumulation and reallocation of DM and N in various parts (upper, middle, and lower) and in different vegetative organs (leaves and stems), as well as leaf area index, were investigated. The objective of this study was to quantify the accumulation and reallocation of DM and N in different vegetative organs and canopy positions, and to understand how these factors influence grain yield and NUE between the two maize hybrids.

2. Materials and Methods

2.1. Experimental Site

A field experiment was conducted during the maize growing season from April to October in 2023 and 2024 at the Experimental Station of Jilin Agricultural University (45°26′ N, 124°87′ E) in Changchun City, Jilin Province, Northeast China. The soil type is typical black soil: organic matter 26.9 g·kg⁻¹ (potassium dichromate oxidation-external heating method), total nitrogen 1.645 g·kg⁻¹ (sulfuric acid-catalyst digestion, kjeldahl method), total phosphorus, 0.85 g·kg⁻¹ (mo-sb anti-spectroscopy), alkali-hydrolyzable nitrogen 120 mg·kg⁻¹ (alkaline hydrolysis diffusion method), available phosphorus 16.5 mg·kg⁻¹ (Mo-Sb colorimetric method), available potassium 122.0 mg·kg⁻¹ (cold nitric acid leaching-flame photometry), and pH 6.8 (soil-to-water ratio of 1:2.5 (NY/T1121.2-2006)). These values were obtained from soil sampled from the 0 to 20 cm soil profile before this study. During maize growing seasons, air temperatures above 10 °C were summed to calculate the effective cumulative temperature, which was 432.0 mm in 2023 and 778.7 mm in 2024, an annual accumulated temperature of 1656.6 °C in 2023 and 1587.3 °C in 2024 (Figure 1).

2.2. Experiment Design

In a split-plot design, the experiment utilized two maize hybrids, Zhengdan 958 (ZD958) and Tie 391 (T391), both of which are high-yield cultivars widely promoted in Jilin Province, and these were planted in the main plots. Four N fertilization (N fertilizer applied was urea) treatments were used: a high-fertilization treatment of 360 kg N ha⁻¹ (HN); a middle-fertilization treatment of 240 kg N ha⁻¹ (MN); a low-fertilization treatment of 120 kg N ha⁻¹ (LN); and a control with no N added (NN), which were individually applied in the subplots. A total of 24 plots with three replications. Each plot had an area of 60 m² (10 m length × 6 m width), with a distance of 0.6 m between rows. Maize was planted at a density of 75,000 plants ha⁻¹. All plots received uniform applications of phosphorus (100 kg P₂O₅ ha⁻¹ supplied as calcium superphosphate) and potassium (100 kg K₂O ha⁻¹ supplied as potassium chloride) as basal fertilizers were broadcast before sowing. N fertilizer was broadcast in a ratio of 5:3:2 at the maize sowing, jointing, and silking stages, respectively. There was no control over environmental conditions during the experiment, including rainfall, water stress, anomalous temperature, and other soil corrections, in addition to N, P, and K. Pests, weeds, and diseases were effectively controlled using pesticides, herbicides, and fungicides, and no irrigation was applied during the experiment. Soil preparation and seed planting of maize were carried out on 15 April and 30 April in 2023, and on 15 April and 1 May in 2024. The respective harvest dates were 1 October and 3 October for both 2023 and 2024.

2.3. Data Collection

2.3.1. Leaf Area Index (LAI)

At 40 d, 60 d, 90 d, and 120 d after sowing, three plants in each plot were selected to measure the leaf area (LA), and LA was calculated as the equation:

$$\mathrm{L}\mathrm{A}=\mathrm{L}\mathrm{e}\mathrm{a}\mathrm{f}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\times \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{f}\mathrm{w}\mathrm{i}\mathrm{d}\mathrm{t}\mathrm{h}\times 0.75$$

(1)

where 0.75 is the leaf area coefficient.

Leaf area index was computed as the ratio between green leaf area and the corresponding land area [8,14].

2.3.2. Dry Matter and N Accumulation and Reallocation

In 2024, at the silking stage (R1) and physiological maturity stage (R6), three plants per plot were selected and divided into three layers: the upper layer (from the top to the second leaf above the ear leaf), the middle layer (the three leaves representing the ear leaf and the first leaves above and below the ear leaf), and the lower layer (from the second leaf below the ear leaf to the base leaf) [5]. The specific segmentation methods as (Figure 2). The upper and lower samples were separated into two parts: leaf and stem (sheath + stem), while the middle layer was divided into grain, cob (R6), leaf, stem (sheath + stem), and other. These samples were heated at 105 °C for 0.5 h and then dried to constant weights at 80 °C. The DM weight of each part was weighed and recorded.

The following indexes were calculated to investigate post-silking DM, the reallocation amount of DM in leaf and stem, harvest index (HI), and $\mathrm{D}\mathrm{M}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}$ (Equations (2)–(5)) [6]:

$$\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}-\mathrm{s}\mathrm{i}\mathrm{l}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{D}\mathrm{M}(\mathrm{g}/\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t})=\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}{\mathrm{D}\mathrm{M}}_{\mathrm{R}6}-\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}{\mathrm{D}\mathrm{M}}_{\mathrm{R}1}$$

(2)

$$\mathrm{R}\mathrm{e}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{o}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{D}\mathrm{M}\left(\mathrm{g}\right)=\mathrm{L}\mathrm{e}\mathrm{a}\mathrm{f}/\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}{\mathrm{D}\mathrm{M}}_{\mathrm{R}1}-{\mathrm{L}\mathrm{e}\mathrm{a}\mathrm{f}/\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\mathrm{D}\mathrm{M}}_{\mathrm{R}6}$$

(3)

$$\mathrm{H}\mathrm{I}=\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{D}\mathrm{M}/\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}{\mathrm{D}\mathrm{M}}_{\mathrm{R}6}$$

(4)

$$\mathrm{D}\mathrm{M}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{D}\mathrm{M}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{v}\mathrm{e}\mathrm{g}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{o}\mathrm{r}\mathrm{g}\mathrm{a}\mathrm{n}}{\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{D}\mathrm{M}}}\times 100$$

(5)

The weighed DM samples were ground into a powder using a grinder, then passed through a 1.00 mm mesh sieve, and subsequently digested with concentrated H₂SO₄. The N content in different positions (upper, middle, and lower layers) of the canopy and in various organs (leaves, stems, and others) [5] was quantified using the Kjeldahl method [22].

The post-silking N, reallocation amount of N in leaf and stem, N harvest index (NHI), and $\mathrm{N}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}$ were calculated as follows (Equations (6)–(9)):

$$\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}-\mathrm{s}\mathrm{i}\mathrm{l}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{N}\left(\mathrm{g}/\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\right)=\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}{\mathrm{N}}_{\mathrm{R}6}-{\mathrm{N}}_{\mathrm{R}1}$$

(6)

$$\mathrm{R}\mathrm{e}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{o}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{a}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{N}\left(\mathrm{g}\right)={\mathrm{N}}_{\mathrm{R}1}-{\mathrm{N}}_{\mathrm{R}6}$$

(7)

$$\mathrm{N}\mathrm{H}\mathrm{I}=\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{N}/\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}{\mathrm{N}}_{\mathrm{R}6}$$

(8)

$$\mathrm{N}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{N}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{v}\mathrm{e}\mathrm{g}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{o}\mathrm{r}\mathrm{g}\mathrm{a}\mathrm{n}}{\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{N}}}\times 100$$

(9)

When considering the leaves or stems as a whole, DM_R1, N_R1, DMR₆, and N_R6 represent the DM or N content of the leaves or stems of plants at the R1 and R6 stages, respectively. When calculating the leaves or stems across different positions, DM_R1, N_R1, DM_R6, and N_R6 indicate the DM or N content in different positions of the canopy at the R1 and R6 stages, respectively.

2.3.3. Grain Yield Measurement

At the physiological maturity stage (R6), maize harvested manually in a bordered area (2.4 m × 10 m length) was selected from undisturbed plots to determine grain yield. The grain yield was adjusted to 14% moisture content. Twenty spikes were randomly selected, and determination of the number of kernels per ear and kernels per row was made and recorded. The ears were then threshed, and 1000 kernels were sampled (in triplicate) to determine kernel weight.

$$\mathrm{N}\mathrm{U}\mathrm{E}={\displaystyle \frac{({\mathrm{G}\mathrm{Y}}_{\mathrm{f}\mathrm{e}\mathrm{r}\mathrm{t}}-{\mathrm{G}\mathrm{Y}}_{\mathrm{u}\mathrm{n}\mathrm{f}\mathrm{e}\mathrm{r}\mathrm{t}})}{\Delta \mathrm{N}\mathrm{a}\mathrm{p}\mathrm{p}\mathrm{l}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}}$$

(10)

where GY_fert. is the grain yield (standardized to 14% moisture) of N-fertilized plants (LN, MN, or HN) at physiological maturity; GY_unfert is the grain yield of plants in the NN treatment.

2.4. Statistical Analysis

All data were tested for the assumptions of the mathematical model, which include the normality of errors and homogeneity of variance of residuals. The normality of errors was assessed using the Kolmogorov–Smirnov test [23], while the homogeneity of residual variance was evaluated using Levene’s test [24]. Statistical analysis (ANOVA) was performed using SPSS 20.0 software (SPSS Inc., Chicago, IL, USA). Differences were compared using the least significant difference test at a significance level of 0.05. Data recording and processing were conducted using Microsoft Excel 2016 (Microsoft Corp., Redmond, WA, USA). Figures for grain yield, leaf area index, nitrogen, and dry matter proportion were generated using Origin 2019b (OriginLab, Northampton, MA, USA), and correlation analyses were conducted using R version 3.4.1.

3. Results

3.1. Grain Yield

According to the ANOVA results, the factors nitrogen level (N) and genotype (G) had significant effects on 1000-kernel weight (TKW), kernel number per ear (KNP), grain yield, and nitrogen-use efficiency (NUE) (Figure 3). Over the two years, N fertilizer input significantly increased kernel number per plant (KNP) and 1000-kernel weight (TKW) regardless of hybrid (p < 0.05), with increases of 12.8% (KNP) and 7.3% (TKW) for ZD958, and 19.2% (KNP) and 7.4% (TKW) for T391, respectively. The grain yield of ZD958 was significantly higher than that of T391 under NN (16.2%), LN (16.2%), and MN (10.2%) levels, but there was no significant difference between the two maize hybrids under high-nitrogen HN (3.6%) conditions. ZD958 exhibited a significantly higher TKW (7.9%) than T391, regardless of N fertilizer application. The KNP of ZD958 was also higher than that of T391 by 14.9% on average under NN, LN, and MN conditions, but there was no difference between the two maize hybrids under HN conditions. Moreover, the highest NUE was observed in ZD958 under LN conditions.

3.2. Leaf Area Index

The leaf area index (LAI) for both hybrids exhibited similar patterns, changing from a rapid increase (40–90 days) to a slight decrease (90–120 days) over time (Figure 4). Additionally, the ZD958 hybrid demonstrated a significantly higher LAI than the T391 hybrid during this period, particularly at 90 and 120 days after sowing, regardless of N level. For instance, in terms of absolute values, under NN, LN, MN, and HN conditions, the LAI of ZD958 was 15.0%, 19.3%, 13.2%, and 3.6% greater than that of T391 at 120 days after sowing, respectively. Notably, there was no significant difference in LAI between the two hybrids under HN conditions at 120 days after sowing.

3.3. DM and N Accumulation

Nitrogen application (LN, MN, and HN) significantly increased dry matter (DM) and N accumulation at both the silking (R1) and maturity (R6) stages, as well as the harvest index (HI), compared to the NN treatment, regardless of maize hybrids (

Table 1. Whole-plant dry matter (DM) and N at silking (R1) and maturity (R6), harvest index (HI), N harvest index (NHI), dry matter, and N accumulation during post-silking (Post DM and Post N) of maize hybrids ZD958 and T391 under four N fertilization levels (NN, LN, MN, HN) in 2024.

N Level	Genotype	Dry Matter per Plant (DM)				Post DM/Grain DM (%)	HI	N Content per Plant (N)				NHI	Post N/Grain N (%)
		DMR1	DMR6	Post DM (g)	Grain			NR1	NR6	Post N	Grain N (g)
		(g)	(g)		DM (g)			(g)	(g)
N0	ZD958	115.2 g	274.9 f	159.7 e	141.0 f	113.3 ab	0.51 d	1.16 f	1.92 g	0.76 f	1.22 e	63.67 a	61.93 d
	T391	112.1 h	254.8 g	142.7 f	125.1 g	114.1 a	0.49 e	1.09 g	1.64 h	0.55 g	0.95 f	58.26 c	57.40 e
N1	ZD958	132.8 e	305.9 d	173.1 d	163.0 d	106.2 d	0.53 c	2.03 d	3.25 e	1.22 e	1.79 c	55.21 d	67.91 c
	T391	123.2 f	287.5 e	164.2 e	145.7 e	112.7 ab	0.51 d	1.68 e	2.43 f	0.76 f	1.42 d	58.34 c	53.31 f
N2	ZD958	146.5 b	352.0 b	205.5 b	193.1 b	106.5 d	0.55 b	2.44 b	4.25 c	1.81 c	2.39 b	56.37 d	75.37 b
	T391	134.5 d	331.8 c	197.3 c	180.7 c	109.1 c	0.54 b	2.22 c	3.81 d	1.59 d	2.29 b	60.24 b	69.42 c
N3	ZD958	148.4 a	389.8 a	241.3 a	219.5 a	110.0 c	0.56 a	2.53 a	5.18 a	2.65 a	3.13 a	60.46 b	84.51 a
	T391	140.7 c	381.4 a	240.8 a	215.0 a	112.0 b	0.56 a	2.43 b	4.96 b	2.53 b	3.06 a	61.56 b	82.81 a
ANOVA
N level (N)		***	***	***	***	***	***	***	***	***	***	***	***
Genotype (G)		***	***	***	***	***	***	***	***	***	***	NS	***
N × G		***	**	**	**	***	***	***	***	***	***	***	***

Note: NN, LN, MN, and HN indicate N applied at 0, 120, 240, and 360 kg ha⁻¹ levels, respectively. Values followed by a different letter within the same column are significantly different at p < 0.05. The values shown are the mean ± SE (n = 3). ** p < 0.01, *** p < 0.001, NS, no significance.

). The increase in post-silking DM was slightly greater than that of grain DM, resulting in a post-DM/grain DM ratio exceeding 100% for both maize hybrids. Furthermore, compared with T391, ZD958 exhibited relatively higher total DM and N content at both the R1 and R6 stages, as well as in post-silking DM and N accumulation, grain DM, grain N, and HI across all cases, including the NN treatment. For instance, in terms of post-silking N accumulation, ZD958 was significantly higher by 27.6% (NN), 37.7% (LN), 12.2% (MN), and 4.5% (HN) compared to T391. Notably, the ratio of post-N to grain N was less than 100%, indicating that N accumulation during the post-silking phase could not meet grain N demand, necessitating the reallocation of N from vegetative organs to grains.

3.4. DM and N Reallocation

The largest amounts of DM and N reallocation in stems and leaves were obtained in LN (ZD958) and MN (T391) conditions, with the amount of stem DM and N reallocated being higher than that of leaf DM and N reallocated (

Table 2. The reallocation amount of DM and N in leaf and stem, reallocation fraction of DM and N in leaf and stem, and contribution of N reallocation in leaf and stem to grain N in maize hybrids ZD958 and T391 under four N fertilization levels (NN, LN, MN, HN) in 2024.

N Level	Genotype	Dry Matter Reallocation per Plant (DM)						N Reallocation per Plant (N)
		Amount (g)		Fraction (%)		Contribution to Grain DM (%)		Amount (g)		Fraction (%)		Contribution to Grain N (%)
		Leaf	Stem	Leaf	Stem	Leaf	Stem	Leaf	Stem	Leaf	Stem	Leaf	Stem
N0	ZD958	3.23 b	4.83 b	9.32 b	6.88 b	6.66 a	4.88 b	0.25 e	0.32 e	45.57 a	68.93 a	37.38 b	56.52 b
	T391	1.29 c	3.89 c	3.94 cd	5.63 bc	3.17 c	4.50 b	0.18 f	0.28 f	37.85 b	61.45 c	39.70 a	64.51 a
N1	ZD958	5.58 a	7.94 a	13.34 a	9.98 a	8.19 a	6.13 a	0.45 a	0.55 a	43.52 a	65.26 b	24.26 c	36.39 d
	T391	1.16 c	3.51 c	3.21 cd	4.67 cd	2.20 c	3.21 c	0.27 d	0.46 bc	32.07 cd	65.86 b	22.61 c	46.44 c
N2	ZD958	4.50 a	5.05 b	9.46 b	5.91 b	4.90 b	3.06 c	0.42 b	0.48 b	33.09 c	48.59 d	13.84 d	20.32 e
	T391	1.52 c	4.99 b	3.59 cd	6.21 b	1.99 c	3.43 c	0.35 c	0.47 b	30.50 d	49.00 d	13.29 d	21.36 e
N3	ZD958	2.90 b	3.42 c	6.00 c	3.93 de	2.74 c	1.79 d	0.35 c	0.44 cd	26.84 e	40.65 e	8.57 e	12.98 f
	T391	1.27 c	2.44 d	2.72 d	3.04 e	1.27 c	1.41 d	0.28 d	0.42 d	22.04 f	41.93 e	7.21 e	13.72 f
ANOVA
N level (N)		*	***	**	***	***	***	***	***	***	***	***	***
Genotype (G)		***	***	***	***	***	***	***	***	***	**	NS	***
N × G		**	***	*	***	**	***	***	**	***	***	**	***

Note: NN, LN, MN, and HN indicate N applied at 0, 120, 240, and 360 kg ha⁻¹ levels, respectively. Values followed by a different letter within the same column are significantly different at p < 0.05. The values shown are the mean ± SE (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001, NS, no significance.

). Under the same N input conditions, ZD958 exhibited relatively higher DM and N reallocation than T391. For instance, the N reallocation of leaf and stem in ZD958 was higher by 28.0% (leaf) and 12.5% (stem) (NN), 40.0% (leaf) and 16.4% (stem) (LN), 16.7% (leaf) and 2.1% (stem) (MN), and 20.0% (leaf) and 4.5% (stem) (HN) compared to T391, respectively. ZD958 had a lower stem N fraction and contribution to grain N than T391 under NN and LN conditions, but there was no significant difference between the two maize hybrids in MN and HN treatments.

3.5. Fraction of DM and N Content in Different Organs

At the silking stage (R1), similar DM and N distribution patterns were observed in the same N-treated plants of both hybrids (Figure 5a,c). Generally, the largest fractions of DM and N were found in the stem (45–60%), followed by leaves (30–40%), with the remaining organs (bracts and tassels) accounting for about 10%. Compared with NN, increasing N application significantly decreased the proportion of stem DM to whole-plant DM, but increased the proportions of leaf DM to whole-plant DM, and leaf and stem N to whole-plant N.

At maturity (R6), the DM and N in stem and leaf tissues were transferred to grains in large quantities, leading to the grain organs containing the most DM (Figure 5b). Maize plants tended to allocate less DM to grains and more to vegetative parts at NN levels compared to HN levels. For instance, the proportions of grain DM to whole-plant DM decreased from 56.3% in the HN treatment to 50.2% in the NN treatment (Figure 5b). Under NN and LN conditions, ZD958 exhibited relatively higher proportions of grain DM to whole-plant DM, while under MN and HN conditions, there was no significant difference between the two hybrids. Moreover, significantly higher amounts of grain N to whole-plant N, but lower leaf and stem N, were distributed in ZD958 compared to T391 under all treatments.

3.6. Reallocation of Leaf and Stem DM and N in Different Canopy Positions

The leaf and stem reallocation DM in the middle layer was significantly higher than that in the lower and upper layers of the canopy. The negative values of the reallocated DM in the upper and middle leaves/stems indicate that there was net DM accumulation rather than reallocation between the R1 and R6 stages in these leaves. Generally, increasing N levels promoted net DM accumulation in the upper and middle layers of leaves/stems for both maize hybrids, especially for the leaves. ZD958 exhibited a higher amount of middle-layer canopy leaf and stem reallocated DM compared to T391 across all treatments (Figure 6a,c). For instance, the middle-layer stem DM reallocation was greater by 64.6%, 78.6%, 63.8%, and 94.3% for ZD958 compared to T391 in the NN, LN, MN, and HN treatments (Figure 6c).

The reallocation of leaf N showed that the middle layer had a significantly higher N content than both the lower and upper layers (Figure 6b). This may be due to the senescence characteristics of the lower leaves, which result in lower N reallocation. In contrast, N accumulation in the middle leaves is greater than in the upper leaves, giving the middle leaves the highest N reallocation capacity. However, the largest amount of stem-reallocated N was found in the lower stem, followed by the middle stem, and then the upper stem, especially under NN conditions (Figure 6d). These results indicate that increasing N reallocation in the lower stem is an important strategy for enhancing grain N under low soil N conditions. ZD958 exhibited a higher middle leaf N reallocation than T391 across all treatments (Figure 6b). The reallocation of lower stem N in ZD958 was significantly higher than that of T391 under NN (24.3%), LN (13.0%), MN (8.6%), and HN (18.8%) treatments (Figure 6d).

3.7. Correlation Analysis

As shown in Figure 7, the correlation analysis among the accumulation of pre-silking and post-silking DM and N, as well as the different positions of canopy leaf and stem DM and N reallocation with grain yield, revealed that DM and N accumulation were positively correlated with grain yield. It was found that the reallocation of DM in the upper leaves was negatively correlated with grain DM and N accumulation as well as grain yield. However, there was no significant correlation between leaf DM reallocation in the middle and lower layers of the canopy and grain DM and grain yield. Moreover, upper- and middle-leaf N reallocation showed no correlation with grain DM and N accumulation, while lower-layer leaf N reallocation showed a significant positive correlation with grain DM and N accumulation. These findings indicate that N in the upper and middle leaves remains in the leaves to maintain photosynthesis, while the reallocation of lower leaf N is the key factor in further increasing grain DM and N accumulation. For stem N reallocation, there was a negative correlation between N reallocation in the middle-layer stem and grain DM and N accumulation, but there was a significant positive correlation between N reallocation in the upper and lower stems. Therefore, the reallocation of N in lower leaves, upper stems, and lower stems may be the main indicators for improving N-efficiency, especially under low-N conditions.

U-L-DM, M-L-DM, and L-L-DM, leaf dry matter reallocation in the upper, middle, and lower layers of the canopy, respectively; U-S-DM, M-S-DM, and L-S-DM, stem dry matter reallocation in the upper, middle, and lower layers of the canopy, respectively; U-L-N, M-L-N, and L-L-N, leaf N reallocation in the upper, middle, and lower layers of the canopy, respectively; U-S-N, M-S-N, and L-S-N, stem N reallocation in the upper, middle, and lower layers of the canopy, respectively; DMR1, DMR6, and Post DM, dry matter accumulation at silking stages, at maturity stages and post-silking dry matter accumulation, respectively; NR1, NR6, and Post N, N accumulation at silking stages, at maturity stages and post-silking N accumulation, respectively; GN, grain N content; GY, grain yield.

4. Discussion

4.1. Nitrogen × Hybrids: Grain Yield and Nitrogen-Use Efficiency

Nitrogen (N) fertilizer application is crucial to obtain higher maize grain yield, while farmers usually apply excessive amounts of N fertilizer to improve grain yield [25,26,27]. Excess N fertilizer input not only caused environmental pollution but also a reduction in nitrogen-use efficiency (NUE) [28,29,30]. Cultivated high-N efficiency hybrids, characterized by stay-green, can obtain superior grain yield and NUE under low soil N conditions than that of conventional hybrids [11,31]. Stay-green maize cultivars exhibit high efficiency in N uptake, but they show lower efficiency in N remobilization [10]. The two maize hybrids of ZD958 and T391 in this study are high-yield modern maize hybrids [12,16]. However, there are still differences in sensitivity between the two high-yield maize hybrids. For instance, high-yield maize hybrids ZD958 and T391 had the same yield level and NUE at HN conditions, but the grain yields and NUE of ZD958 were higher than those of T391 at LN and MN conditions (Figure 3). Therefore, under conditions of reduced N fertilizer application, we suggest that attention be paid to maize hybrids of the ZD958 genotype, which can achieve higher NUE with minimal yield loss compared to other high-yield hybrids.

4.2. Nitrogen × Hybrids: DM Accumulation and Reallocation

Achieving high maize grain yields relies heavily on higher vegetative-DM remobilization and post-silking DM accumulation, especially higher post-silking DM, since the major grain DM was found to come from the photosynthates produced during the post-silking stage [25,26,32]. In this study, under HN conditions, there was no significant difference in post-silking and total DM between the two maize hybrids. However, under N-deficient conditions (NN and LN), ZD958 exhibited significantly greater total DM, post-silking DM, and pre-silking DM compared to T391 (Table 1). Thus, the synergy of higher total DM, post-silking DM, and pre-silking DM contributed to the increased grain yield of ZD958 under NN and LN conditions. Notably, post-silking DM accounted for 106.2–111.3% of the grain DM in ZD958 and 109.1–114.1% in T391, regardless of N input (Table 1). These results indicate that both modern high-yield maize hybrids, ZD958 and T391, exhibit optimal canopy photosynthesis, providing sufficient assimilates for grain DM during post-silking stages [13,32,33]. Furthermore, for different maize hybrid genotypes, yield depends not only on total DM accumulation but also on its partitioning to grain, which is equally important [34]. The harvest index (HI) is commonly used to represent the ability of plants to transport assimilates from source organs to sinks [16,34]. Our research demonstrated that ZD958 had significantly higher HI and grain DM/whole DM ratios than T391 under NN and LN conditions. However, there were no significant differences in HI or grain DM/whole DM ratios between ZD958 and T391 under MN and HN conditions (Table 1 and Figure 5), indicating that more DM was allocated to grain in ZD958 under NN and LN conditions.

During the post-silking period, grain N originates from both vegetative-N remobilization and N uptake during the reproductive stage [10,35]. Previous studies have proposed that high pre-silking and post-silking N accumulation would contribute to achieving high grain yields [36]. Our results showed that ZD958 exhibited both higher pre-silking and post-silking N accumulation compared to T391, especially under N-deficient conditions (NN and LN) (Table 1). This suggests that the higher N accumulation in ZD958 was beneficial for increasing its yield compared to T391. Moreover, the post N/grain N ratio was found to be less than 100, indicating that the N extracted during the post-silking phase was insufficient to meet the needs of grain growth, necessitating the relocation of N from vegetative organs for both maize hybrids (Table 1). In this study, we observed that the highest amount of N reallocation in the leaves and stems of the two maize varieties occurred at LN levels for ZD958 and at MN levels for T391. This may be attributed to the fact that both N deficiency and high-N levels inhibit N reallocation from vegetative organs to grain [13,37]. The reasons behind the largest N reallocation in the two maize hybrids at different N levels are as follows: ZD958 can tolerate lower N levels, allowing for substantial N reallocation under LN conditions, while T391 requires higher N application to achieve greater N reallocation from vegetative organs.

Leaves and stems are the two largest vegetative organs for N reallocation [18]. In this study, we found that N reallocation in the stem was greater than that in the leaves for both maize hybrids (Table 2). These results are consistent with the findings of Ning [35], who, using ¹⁵N isotopic analysis, indicated that N remobilization initially occurred preferentially from stems compared to leaves, and that the amount of N reallocated was greater in the stems than in the leaves. Furthermore, our results indicate that, under the same N level, the amount of N reallocation in the leaves and stems of ZD958 was higher than that of T391, particularly under LN and MN conditions (Table 2). Although previous research indicated that higher leaf N reallocation might reduce leaf photosynthesis—since N in leaves not only needs to be supplied for grain growth but also must remain in the leaves to maintain photosynthesis—there is a need to balance N reallocation and leaf photosynthesis rates [38,39]. In this study, ZD958 had higher leaf N reallocation compared to T391, while still maintaining a greater leaf area index, especially during the milking stages (Figure 4). These results indicate that, compared to T391, ZD958 not only had sufficient leaf N reallocation to support an increase in grain N but also retained enough N in the leaves to sustain its photosynthesis. Therefore, the higher leaf area index in the late stages of grain filling may be one of the main reasons for supporting ZD958 in having a higher N than the common high-yield hybrids (T391).

4.3. The Different Patterns of DM and N Reallocation in Various Positions and Organs Within the Maize Canopy

There are differences in the accumulation and reallocation of DM and N at different positions in the canopy [40,41]. According to previous descriptions, we divide the canopy into three parts: upper, middle, and lower [5,8,17]. We found that the amount of DM and N reallocation in these three layers was significantly different (Table 2). In the leaves, the largest amount of DM and N reallocation occurred in the middle layer, which was significantly higher than that in the upper and lower parts (Figure 6). This is mainly because the middle leaves have higher N content and a larger leaf area, allowing for greater DM and N reallocation, which is consistent with previous findings [8,13]. Moreover, the results of the correlation analysis showed that the reallocation of N from the lower-layer leaves was positively correlated with grain N content and grain yield (Figure 7). This may be due to a trade-off for leaves between sustaining photosynthesis and remobilizing leaf N. If upper and middle leaves maintain their own photosynthesis and promote N uptake during the grain-filling stages of the plant, this may be sufficient to compensate for their own N remobilization and supply sufficient N concentration for the grain [13,14]. In this study, the amount of N reallocation in the lower-layer leaves of ZD958 was higher than that of T391 under LN conditions; however, there was no significant difference between the two maize hybrids under MN and HN conditions (Figure 6b). For stem N, the amount of reallocation exhibited a pattern of lower > middle > upper layers (Figure 6). This may be due to the acceleration of N reallocation in the lower stems, which could stimulate the roots to uptake soil N [42]. Furthermore, for the two N-efficient maize hybrids, the N reallocation in the stems of ZD958 was significantly higher than that of T391, especially in the lower-layer stems (Figure 6). This could be attributed to the fact that N-efficiency maize hybrids still maintain a greater number of larger vascular bundles under low-N conditions [12]. This may also be one of the main reasons why the NUE of ZD958 was higher than that of T391 under both NN and LN conditions (Figure 3). Therefore, the lower-layer stem N reallocation can be explored as an important physiological index affecting NUE. These results are consistent with previous findings that N reallocation at different levels of stems and leaves in the canopy may significantly influence yield production [14]. The higher stem N reallocation in the lower layer may enhance plant N uptake, while greater N reallocation in the middle- and lower-layer leaves regulates the effective use of N and DM throughout the canopy [13,42,43,44].

5. Conclusions

Modern high-yielding hybrids accumulate sufficient post-silking dry DM to achieve higher yields. However, 15.5% to 46.7% of the grain N comes from the reallocation of N in vegetative organs. Among the two maize hybrid genotypes, the low-N tolerant hybrid ZD958 shows greater DM and N reallocation in the lower stems and leaves, especially under low-N conditions. These factors contribute to N accumulation both pre- and post-silking and enhance N reallocation, leading to increased grain yield and NUE compared to the commonly used high-yielding hybrid T391. This is crucial for farmers’ economic benefits and for reducing environmental risks. Therefore, future breeding of N-efficient high-yielding maize cultivars should explore the genetic control of N remobilization from lower stems and leaves, particularly under reduced N treatment.