Integrating Heterosis for Root Architecture and Nitrogen Use Efficiency of Maize: A Comparison between Hybrids from Different Decades

Abstract

Exploring the biological potential of maize root architecture is one of the most important ways to improve nitrogen use efficiency (NUE). The NUE and its heterosis in maize hybrids have improved remarkably over decades. Yet, there is little research on maize hybrid heterosis for root architecture and its possible physiological relationship to heterosis for NUE. A field study lasting two years was carried out on four typical maize hybrids from old and new eras, including their parental inbred lines with two levels of nitrogen (0, 150 kg N ha⁻¹). Compared to old-era maize hybrids, the brace root angle (BA) and crown root angle (CA) of new-era maize hybrids increased by 19.3% and 8.0% under 0 N, and by 18.8% and 7.9% under 150 N, which exhibited a steeper root architecture; the crown root number (CN) of new-era maize hybrids increased by 30.5% and 21.4% under 0 N and 150 N, respectively, which showed a denser root system; meanwhile, the depth of 95% cumulative root weight (D₉₅) of new-era maize hybrids separately increased by 10.5% and 8.5% under 0 N and 150 N, which exhibited a deeper root distribution. This steeper-denser-deeper root architecture enhanced pre-anthesis N uptake and provided a premise of greater post-anthesis N remobilization. All maize hybrids displayed significant heterosis for root architecture compared to their parental inbred lines. The brace root branching (BB) and crown root branching (CB) of new-era maize hybrids and D₉₅ have positive heterosis, while the BA, CA, and CB of old-era maize hybrids, brace root number (BN), and CN have negative heterosis. Regardless of whether root architecture heterosis was positive or negative, new-era maize hybrids showed an overall elevated trend compared to old-era maize hybrids. Structural equation modeling (SEM) showed that heterosis for nitrogen internal efficiency (NIE) was the primary reason for NUE heterosis in maize, influenced by heterosis for root architecture, which was steeper, denser, and deeper. Our results indicated that, compared with old-era maize hybrids, new-era maize hybrids had stronger heterosis for root architecture, which was beneficial to pre-silking nitrogen absorption and is an important physiological basis for the higher NIE heterosis and NUE heterosis in new-era maize hybrids.

1. Introduction

Maize (Zea mays) is a significant staple crop that contributes 36% of the total cereal calories worldwide [1,2]. Additionally, it is known for its efficient utilization of heterosis [3,4]. Increasing maize yield with minimal nitrogen fertilizer input is the inevitable choice of sustainable agricultural development [5,6]. Maize can thrive by enhancing nitrogen recovery efficiency though changes of root system traits, leading to increased nitrogen use efficiency and grain yield [7,8]. The root system has a determining importance for water absorption and nutrient uptake, as it influences crop growth and grain yields [9]. By regulating the growth and development of the root system, improving the ability to absorb and utilize water and nutrients, it plays an important role in improving maize yield and resource utilization efficiency [9].

In maize, there are five principal categories of root: primary, lateral, seminal, crown, and brace roots [10]. The bulk of the mature root biomass is comprised of post-embryonic shoot-borne roots, crown roots which develop beneath the soil surface, and brace roots which emerge above it [10]. Lynch [11] suggested the “steep, cheap, and deep” ideotype for optimal nitrogen acquisition, which showed that maize hybrids with steeper root angles [12,13], smaller crown production [14,15], and fewer but longer laterals [16,17] resulted in deeper rooting and enhanced capture of mobile nutrients from deeper soils [11,18]. Wang [19] found that the evolution of maize varieties over time has led to increased longitudinal extension and transverse contraction. Chen et al. [20] studied root growth, nitrogen uptake, and yield formation using maize varieties that were bred for large-scale promotion in China before and after 1990, and the findings indicated that the modern varieties showed a significant increase in nitrogen accumulation, dry matter production, and yield compared to the pre-1990 varieties at high and low nitrogen levels, while decreasing surface root length density, but there was no significant change in root length density in deeper soils. Liu et al. [21] suggested that the nitrogen absorption and utilization efficiency of modern maize varieties were significantly higher than those of earlier ones. With the variety in turnover, there was an increase in the number of seminal roots, the proportion of roots in the deep soil, and the rooting ability, which was conducive to the absorption of nitrogen. However, the characteristics of evolution over time in root architecture for maize hybrids has not been obtained due to the absence of a dependable phenotypic screening system. Therefore, reviewing the changes in the root system during maize variety turnover was important for an in-depth analysis of the physiological mechanisms of nitrogen use efficiency evolution for maize hybrids.

Nitrogen use efficiency (NUE) is governed by a complex interplay among genotype, environmental conditions, and agronomic strategies [22]. To decipher the impact of root morphology on NUE, an enhanced understanding of specific soil environments and their dynamic changes during the crop lifecycle is crucial [23]. Sciarresi et al. [24] investigated whether newer maize hybrids develop roots more rapidly and extensively than older hybrids, and assessed the influence of management practices and environmental factors on root trait expression. They found that environmental conditions had a significant influence on root trait expression (>41%), with variations in rainfall anomalies and soil bulk density accounting for some of this variability. Meanwhile, Wang et al. [25] examined the correlations between maize yield, nitrogen use efficiency, and inherent soil productivity. They suggested that improving inherent soil productivity could concurrently increase both yield and NUE. Nitrogen use efficiency (NUE) is influenced by both genetic and physiological factors [26,27]. The maize germplasm exhibits substantial genetic diversity, which is beneficial for traits associated with improved nitrogen uptake [26]. Mueller et al. [27] demonstrated that enhancements in NUE were facilitated by increased nitrogen remobilization in the stem and the retention of nitrogen in leaves during the reproductive phase. Their study analyzed seven maize hybrids, commercialized from 1946 to 2015 by a single seed company, under various nitrogen fertilizer regimes. A comprehensive understanding of the physiological mechanisms underlying NUE—such as the dynamics and form of nitrogen storage and remobilization in the stem, along with transporter activity during the grain-filling period—is crucial. This knowledge will aid in devising breeding strategies aimed at selecting genotypes with enhanced nitrogen recovery efficiency (NRE) and NUE.

Heterosis (hybrid vigor) is the phenomenon where heterozygous hybrids outperform their parents in terms of yield, biomass, and resistance [3,28]. Duvick et al. [29] analyzed heterosis changes in maize hybrids for yield and related characteristics using typical hybrids and their parental lines in the U.S. Corn Belt from the 1930s to the 1980s, and the results showed that genetic improvement synchronously contributed to increase in yield of both hybrids and inbreds, and that there was no increase in absolute heterosis of yield. Tollenaar and Liu [30,31,32] systematically analyzed yield heterosis using Canadian maize hybrid combinations and their inbred lines, which, due to the heterosis of leaves, increases dry matter accumulation and harvest index. Previous studies [5,33] have shown that maize nitrogen use efficiency has heterosis, which mainly comes from nitrogen internal efficiency heterosis, and which was ascribed to (i) the gradual improvement of the nitrogen accumulation heterosis in the pre-silking stage, and nitrogen remobilization heterosis in the post-silking stage; (ii) the plant height, the stalk diameter, and the specific stalk and leaf weight at the silking stage in response to an increase in nitrogen application; and (iii) an improvement in the heterosis of grain number and grain nitrogen yield. However, heterosis for root architecture in maize hybrids has not been studied. Hund et al. [34] found that, in the presence of bacteria and other microorganisms, the seedling root growth of hybrids was superior to that of their parental inbred lines, and believed that root growth had heterosis for some hybrids. Wagner et al. [35] observed that environmental factors were also one of the important factors affecting heterosis, with the microbial environment playing a critical role in the development of maize roots.

The NUE and its heterosis have improved remarkably over decades in maize hybrids [27,33,36]. Yet, there is little research on maize hybrid heterosis for root architecture and its possible physiological relationship to the heterosis of NUE [37,38,39,40]. Testing whether root structure has heterosis, and its relationship with heterosis for nitrogen use efficiency, is of great significance for further exploring yield heterosis [41,42] and heterosis for nitrogen use efficiency [43,44]. In this paper, the “Shovelomics” method [45] was adopted to study heterosis for root architecture with four typical maize hybrids from old and new eras, and their parental inbred lines under two N levels (0, 150 kg N ha⁻¹) in a two-year field. We hypothesized that (i) heterosis for root architecture existing in maize hybrids would change over time; and (ii) heterosis for root architecture would be closely related to heterosis for nitrogen use efficiency. The results will provide a scientific foundation for further clarifying the physiological basis behind NUE evolution and its heterosis of maize hybrids.

2. Materials and Methods

2.1. Experimental Site

From 2016 to 2017, the field experiment was conducted at an irrigated system on Tumed Right Banner County (40°32′ N, 110°30′ E), Inner Mongolia, China. During the two-year growth period, the total precipitation was 423.1 mm and 350.5 mm, the active accumulated temperature ≥ 10 °C was 3118.7 °C and 3312.0 °C, and the total solar radiation was 3660.1 MJ m⁻² and 3976.3 MJ m⁻². The previous crop on the test site was maize (Zea mays L.) and the soil was sandy loam.

The top 30 cm of soil was taken before sowing. The soil organic matter, available nitrogen, available phosphorus, available potassium, and pH were 24.5 g kg⁻¹, 21.2 mg kg⁻¹, 26.7 mg kg⁻¹, 120.4 mg kg⁻¹, and 7.5 in 2016, respectively; and 20.5 mg kg⁻¹, 29.3 mg kg⁻¹, 24.5 mg kg⁻¹, 102.4 mg kg⁻¹, and 7.4 in 2017, respectively.

2.2. Experimental Design

Taking the year 2000 as the boundary, old-era typical maize hybrids Zhongdan 2 (ZD2) and Danyu 13 (DY13), and new-era typical maize hybrids Zhengdan 958 (ZD958) and Xianyu 335 (XY335), and their parental inbred lines with an extension area of more than 5 million acres, were selected as test materials (

Table 1. Details of old- and new-era maize hybrids, and parental inbred lines.

Era of Hybrids	Hybrids	Parental Combination	Year of Hybrid Release	Institution That Developed the Cultivar
Old-era hybrids	Zhongdan 2 (ZD2)	Mo17 × Zi 330	1972 [46,47]	Chinese AAS, Beijing, China
	Danyu 13 (DY13)	Mo17 × E 28	1981 [46,47]	Dandong AAS of Liaoning Province, Dandong, China
New-era hybrids	Zhengdan 958 (ZD958)	Zheng 58 × Chang 7-2	2000 [46,47]	Luohe AAS of Henan Province, Luohe, China
	Xianyu 335 (XY335)	PH6WC × PH4CV	2004 [46,47]	The Tieling Pioneer limited company, Tieling, China

).

Fertilization treatment (150 N) and starter fertilizers were uniformly applied with 105 kg ha⁻¹ P₂O₅, 41 kg ha⁻¹ N (Diammonium phosphate) and 45 kg ha⁻¹ K₂O (Potassium sulfate), and a 0–15 cm soil layer was evenly tilled into the soil before planting, while 237 kg ha⁻¹ of urea was applied at the jointing stage of maize (6-leaf stage of maize).

Without nitrogen fertilizer (0 N), potassium chloride and superphosphate were applied, and the amount of phosphorus and potassium fertilizer was the same as that produced with other treatments.

The experiment was a randomized block design with three replications, the plot area was 27 m⁻², with a length of 5 m and a row spacing of 0.6 m planted in 9 rows with a planting density of 75,000 plants ha⁻¹. Flood irrigation ensured that the soil field water capacity reached 75%.

2.3. Sampling and Measurements

2.3.1. Root Architecture

The ‘Shovelomics’ method was used to determine root architecture [45]. At the silking stage (R1), 3 maize plants with uniform growth marked after seeding emergence were selected from each plot, and a cylindrical soil block with a 40 cm diameter and 35 cm depth was excavated around it. Afterwards, the unearthed root cap was shaken to eliminate most adhering soil, and then remaining soil particles were washed away with water [48]. Three maize roots were dug for each plot, meanwhile, the index of root architecture was determined [12]: (1) brace root angles (BA) and crown root angles (CA): the BA (the angle between brace roots and horizontal plane) and CA (the angle between crown roots and horizontal plane) were measured with a protractor and recorded as 0° parallel to the horizontal plane, and 90° perpendicular to the horizontal plane; (2) brace root numbers (BN: the number of aerial roots above the surface) and crown root numbers (CN: the number of secondary roots below the surface); (3) brace root branches (BB: 1 cm segments below 5 cm of brace roots and counted the number of branches) and crown root branches (CB: 1 cm segments below 5 cm of crown roots and counted the number of branches); (4) root depth of 95% cumulative root weight (D₉₅: the depth of the soil layer when the root weight reaches 95% from top to bottom) between maize plants, a soil drill with a 5 cm diameter and 60 cm depth was drilled, taking 10 cm as a layer, and then roots were picked up, dried, and weighed, and the following formula was used to calculate D₉₅ [12]:

$$RLCu{m}_{Segi}={\displaystyle \sum }_{k=1}^{i}R{L}_{Seg }k$$

(1)

where RLCum_segi is the cumulative dry weight of roots in layer i, RL_seg k is the root weight of k layer.

2.3.2. Nitrogen and Heterosis Indices

Absolute heterosis (AH) is the specific parameter difference between the F1 generation hybrids and the average value of the parental inbred lines [5,29]:

AH = F1 − MP

(2)

MP = (P1 + P2)/2

(3)

where P1, P2 are the parameter values of parental inbred lines [5], and F1 is the parameter value of the maize hybrids.

Mid-parent heterosis (MPH) is a method to measure the average performance of the maize hybrids relative to its parental inbred lines [5,30]:

MPH = [(F1 − MP)/MP] ×100

(4)

At the R1 stage, 5 continuous and uniform plants were separated into leaves, stalks (stems, leaf sheaths, and tassels), and ears (husk and cobs) [5], while those at the R6 stage were divided into leaves, stalks (stems, leaf sheaths, tassels, ear shanks, husks, and cobs), and grain [5]. The plants were placed in an oven at 105 °C for 30 min, 75 °C to constant weight, and weighed. The dried samples were crushed and sieved, and the nitrogen concentration was determined by the Kjeldahl method.

Nitrogen use efficiency (NUE) was calculated as the ratio of incremental grain yield response [5] (the difference between the yield of the plot with nitrogen application and that of the plot without nitrogen application) to the application of nitrogen fertilizer [49]. According to Ciampitti and Vyn [50,51], other nitrogen parameters such as Nitrogen recovery efficiency (NRE) and Nitrogen internal efficiency (NIE) were also calculated [5,33]:

NUE = (GY_fert − GY_unfert)/N fertilizer applied

(5)

NRE = (Nupt_fert − Nupt_unfert)/N fertilizer applied

(6)

NIE = (GY_fert − GY_unfert)/(Nupt_fert − Nupt_unfert)

(7)

where GY_fert and GY_unfert are grain yield under 150 N and 0 N, respectively; Nupt_fert and Nupt_unfert are N uptake in 150 N and 0 N, respectively.

2.3.3. Yield and Yield Components

During the physiological maturity stage, two rows in the middle of the measured production area were selected. After removing any side plants, all plants within these selected rows were harvested. The total number of harvested ears was counted. Ten plants exhibiting uniform ear growth were chosen for further analysis to determine ear rows, row grains, 1000-grain weight, and grain water content. The moisture content of the grains was measured using an LDS-1G moisture content detector (Jiangsu WKT Instrumentation Co., LTD, Suzhou, China), and the measurements were then used to calculate the maize yield (converted into hectare yield with 14% water content).

2.4. Statistical Analysis

Since most variables of 4 maize hybrids and 8 parental inbred lines were not significantly different, the experimental data were statistically analyzed by averaging the value of two years by SPSS Statistics 22.0 software (IBM, Inc., Chicago, IL, USA). The least significant difference (LSD) was used for the significance test. Sigma Plot 12.5 software (Systat, Inc., Richmond, CA, USA) was used for drawing the linear model regression and conducting the statistical test of difference (p < 0.05). Origin 2021 software (Origin Lab, Northampton, MA, USA) was used for the heat map. The structural equation model (SEM) was constructed to explore the direct and indirect effects of NUE heterosis in maize hybrids, which was based on a multivariate approach using AMOS software (IBM SPSS AMOS 24.0).

3. Results

3.1. Heterosis for Root Architecture and Its Component Processes

Significant differences were observed in the root architecture (BA, CA, CN, BB, CB, and D₉₅) of both old- and new-era maize hybrids, and BA and BN had significant differences in the parental inbred lines (

Table 2. The related components of root architecture in old- and new-era maize hybrids, and parental inbred lines under 0, 150 N.

Index	BA (°)		CA (°)		BN		CN		BB		CB		D95 (cm)
	0 N	150 N	0 N	150 N	0 N	150 N	0 N	150 N	0 N	150 N	0 N	150 N	0 N	150 N
Old-era hybrids	45.0c	43.1c	53.8b	51.6b	12.2c	14.3b	19.7c	22.9c	15.6b	17.1b	13.4c	14.7c	50.6b	47.1b
New-era hybrids	53.7b	51.2b	58.1a	55.7a	13.1bc	15.2b	25.7b	27.8b	18.2a	19.8a	19.0a	19.0a	55.9a	51.1a
Old-era inbred lines	52.3b	50.4b	59.2a	56.6a	13.7b	15.1b	25.3b	29.6a	13.6c	15.6c	15.4b	16.9b	47.8c	44.8c
New-era inbred lines	56.2a	54.3a	59.9a	58.1a	15.6a	16.5a	28.6a	30.6a	15.0bc	16.4c	15.2b	15.9b	49.3bc	47.0b
Source of variation
Hybrids (H)	**	**	**	**	ns	ns	**	**	**	**	**	**	**	**
Inbred lines (I)	*	*	ns	ns	*	*	*	ns	ns	ns	ns	ns	ns	*
N rates (N)	*		*		*		*		*		ns		**
H × N	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns
I × N	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns	ns

BA: brace root angles. CA: crown root angles. BN: brace root numbers. CN: crown root numbers. BB: brace root branches. CB: crown root branches. D₉₅: root depth of 95% cumulative root weight. Different letters in a column indicated significant variance under 0 N and 150 N, respectively (p < 0.05). ns, not significant variance; *, p < 0.05; **, p < 0.01.

). Compared to old-era maize hybrids, the new-era maize hybrids exhibited an increase of 19.3% and 8.0% in BA and CA under 0 N, and 18.8% and 7.9% under 150 N, respectively, indicating steeper root architecture; the CN of new-era maize hybrids increased by 30.5% and 21.4% under 0 N and 150 N, respectively, and the BB and CB of new-era maize hybrids increased by 16.7% and 41.8% under 0 N, and by 15.8% and 29.3% under 150 N, demonstrating a denser root system. Additionally, the D₉₅ of new-era maize hybrids separately increased by 10.5% and 8.5% under 0 N and 150 N, respectively, indicating a deeper root distribution.

The effect of AH and MPH for BA, CA, BB, CB, CN, and D₉₅ was significant (r = 5.0–121.4, p < 0.05), suggesting that heterosis largely depended on the eras of maize hybrids (

Table 3. Mean squares from the ANOVA of AH and MPH for root architecture of old- and new-era maize hybrids (F value).

Heterotic Index	Source of Variation
	Eras (E)	N Rates (N)	(E × N)
Absolute heterosis (AH)
BA	20.14 **	0.09	0.07
CA	9.94 **	0.01	0.21
BB	4.98 *	0.04	0.25
CB	121.44 **	1.01	0.33
BN	0.23	7.95 *	1.51
CN	11.69 **	0.24	0.29
D95	11.32 *	1.03	0.43
Mid-parent heterosis (MPH)
BA	27.75 **	0.34	0.01
CA	10.77 **	0.02	0.07
BB	6.38 *	0.34	0.36
CB	108.60 **	0.79	0.78
BN	1.07	8.16 *	1.16
CN	16.65 **	0.02	0.02
D95	13.10 *	0.08	0.56

*, indicates slope significance at p < 0.05; **, indicates slope significance at p < 0.01.

). All maize hybrids exhibited significant heterosis for root architecture compared to their parental inbred lines. Among the index of root architecture, BB, CB of new-era maize hybrids, and D₉₅ showed positive heterosis, while BA, CA, and CB of old-era maize hybrids, and BN and CN, had negative heterosis (Figure 1, Figure 2, Figure 3 and Figure 4). The absolute heterosis for brace root angles (AH_BA), crown root angles (AH_CA), brace root branches (AH_BB), crown root branches (AH_CB), root depth of the 95% cumulative root weight (AH_D95), and mid-parent heterosis for brace root angles (MPH_BA), crown root angles (MPH_CA), brace root branches (MPH_BB), crown root branches (MPH_CB), and root depth of the 95% cumulative root weight (MPH_D95) of new-era maize hybrids were higher than those of old-era maize hybrids by 61.1%, 59.1%, 82.2%, 165.3%, and 100.6%, and 64.1%, 59.7%, 49.6%, 50.2%, and 92.9%, respectively (Figure 1, Figure 3 and Figure 4). Furthermore, compared with old-era maize hybrids, the absolute heterosis for crown root numbers (AH_CN) and mid-parent heterosis for crown root numbers (MPH_CN) of new-era maize hybrids increased by 54.1% and 57.3%, respectively (Figure 2).

3.2. Heterosis for Nitrogen Use Efficiency and Its Component Processes

There were significant differences in the NUE, NRE, NIE, Veg_N, and Rep_N of old- and new-era maize hybrids, while there were no significant differences in the eras of maize parental inbred lines (

Table 4. The GY, Veg_N, Rep_N, Plant N, NUE, NRE, and NIE of old- and new-era maize hybrids and parental inbred lines under 0, 150 N.

Index	GY (Mg ha−1)		VegN (kg ha−1)		RepN (kg ha−1)		Plant N (kg ha−1)		NUE (kg kg−1)	NRE (kg kg−1)	NIE (kg kg−1)
	0 N	150 N	0 N	150 N	0 N	150 N	0 N	150 N	150 N	150 N	150 N
Old-era hybrids	10.8a	12.8b	115.6b	153.3b	40.3b	66.4a	155.9b	219.7b	13.5b	0.43c	31.7b
New-era hybrids	11.2a	14.7a	137.9a	193.5a	39.5b	53.7b	177.4a	247.2a	23.1a	0.47c	50.6a
Old-era inbred lines	6.7b	8.2c	61.1c	125.7c	41.1a	68.8a	102.2c	194.5c	9.5c	0.62a	14.5d
New-era inbred lines	7.4b	9.0c	69.9c	130.9c	42.9a	60.9a	112.8c	191.8c	10.5c	0.53b	21.2c
Source of variation
Hybrids (H)	ns	**	**	**	ns	**	**	**	**	ns	*
Inbred lines (I)	ns	ns	ns	ns	ns	ns	ns	ns	ns	*	*
N rates (N)	**		**		**		**		-	-	-
H × N	ns	ns	ns	ns	ns	ns	ns	ns	-	-	-
I × N	ns	ns	ns	ns	ns	ns	ns	ns	-	-	-

GY: grain yield. Veg_N: vegetative-stage N uptake. Rep_N: reproductive stage N uptake. Plant N: N content at maturity. NUE: nitrogen use efficiency. NRE: nitrogen recovery efficiency. NIE: nitrogen internal efficiency. Different letters in a column indicate significant variance under 0 N and 150 N, respectively (p < 0.05). ns, no significant variance; *, p < 0.05; **, p < 0.01.

). Compared to old-era maize hybrids, the NUE, NRE, and NIE of new-era maize hybrids increased by 71.1%, 9.3%, and 59.6%, respectively, indicating that NUE was mainly influenced by NIE. The Veg_N of new-era maize hybrids increased by 19.3% and 26.2% under 0 N and 150 N, respectively, suggesting a significant improvement in Veg_N. Additionally, the GY of new-era maize hybrids increased by 14.8% under 150 N, indicating a basis for NUE improvement.

NUE, NIE, GY, Veg_N, and Plant N had positive heterosis, while NRE and Rep_N had negative heterosis (

Table 5. The AH and MPH for GY, Veg_N, Rep_N, Plant N, NUE, NRE, and NIE of old- and new-era maize hybrids under 150 N.

Index	AH		MPH (%)
	Old-Era Hybrids	New-Era Hybrids	Old-Era Hybrids	New-Era Hybrids
GY	4.1a	3.8a	55.4a	61.7a
VegN	27.6b	62.6a	21.9b	47.8a
RepN	−2.4a	−7.2a	−3.5a	−11.8a
Plant N	25.2b	55.4a	13.1b	28.9a
NUE	4.0b	12.6a	42.1b	120.0a
NRE	−0.2b	−0.1a	−22.3a	−11.3a
NIE	17.2b	30.4a	118.4b	143.6a

Different letters indicate significant variance among new- and old-era maize hybrids (p < 0.05).

). The AH and MPH for NUE, NIE, Veg_N, and Plant N of new-era maize hybrids were higher than those of old-era maize hybrids by 215.0%, 76.7%, 126.8%, 119.8%, 185.0%, 21.3%, 118.3%, and 120.6%, respectively.

A linear regression analysis was conducted to explore the contribution of MPH_NRE and MPH_NIE on MPH_NUE (Figure 5). The coefficient of determination (R²) of MPH_NIE on MPH_NUE (53.5–84.0%) was significantly higher than that of MPH_NRE on MPH_NUE (38.3–64.4%), indicating that NIE played a more substantial role in heterosis for NUE.

3.3. Correlation between Root Architecture Heterosis and Nitrogen Use Efficiency Heterosis

The AH_NUE and MPH_NUE exhibited significant positive correlations with AH_BA and MPH_BA, AH_CN and MPH_CN, AH_CB and MPH_CB, AH_D95 and MPH_D95, AH_NRE and MPH_NRE, and AH_NIE and MPH_NIE, respectively. Additionally, the AH_NRE and MPH_NRE significantly positively correlated with AH_BA and MPH_BA, AH_CA and MPH_CA, AH_CN and MPH_CN, AH_BB and MPH_BB, and AH_CB and MPH_CB, respectively. The AH_BA and MPH_BA significantly positively correlated with AH_CA and MPH_CA, AH_CN and MPH_CN, AH_BB and MPH_BB, AH_CB and MPH_CB, and AH_D95 and MPH_D95, respectively. Furthermore, the AH_CA and MPH_CA significantly positively correlated with AH_BB and MPH_BB, and AH_CB and MPH_CB, respectively, whereas the AH_BN and MPH_BN significantly negatively correlated with AH_CN and MPH_CN, and AH_BB and MPH_BB, respectively (Figure 6), indicating that root architecture heterosis affected nitrogen use efficiency heterosis.

The SEM was utilized to assess the direct and indirect effects on root architecture and NUE of old- and new-era maize hybrids under 150 N (Figure 7). The analysis revealed that the total variation in AH_NUE and MPH_NUE were found to be 96% and 88%. Of the considered factors, AH for NIE, D₉₅, BN, CN, BB, and CB directly contributed to AH_NUE (Figure 7A), while MPH for NIE and D₉₅ directly contributed to MPH_NUE (Figure 7B). The AH_NRE and MPH_NRE (p < 0.001), as affected by AH and MPH for BA and CA, had negative effects on AH_NIE and MPH_NIE, respectively. These results indicated that AH and MPH for steeper-denser-deeper root architecture enhanced the effect of AH_NIE and MPH_NIE, ultimately affecting AH_NUE and MPH_NUE.

4. Discussion

Root architecture plays a crucial role in water and nutrient uptake [52,53,54,55,56]. Generally, the greater total length, the larger the surface area, and the more nutrients absorbed from the root [57]. During the silking stage, 90% of maize roots are concentrated in the 0–30 cm soil layer, with 80–90% of roots biomass distributed in the 0–20 cm soil layer [58,59]. A robust root system with a rational spatial distribution can delay the senescence of the root system in post-silking stage, thereby fulfilling the nitrogen demand of maize during the filling stage [60,61,62,63]. Chen et al. [64] found a gradual decrease in length, number, and dry weight of root in maize varieties since the 1980s, indicating that maize may decrease the consumption of photosynthetic products by reducing redundant root systems. This study indicates that under the two nitrogen levels, the BB of new-era maize hybrids was higher than old-era maize hybrids, and higher than their parental inbred lines. However, the BN and CN of root architecture in new-era maize hybrids were significantly lower than for the parental inbred lines, indicating that maize hybrids reduced root redundancy compared with parental inbred lines. Nonetheless, the CN of new-era maize hybrids was significantly higher than that of old-era maize hybrids, indicating that there is a limit to the reduction of redundancy (Table 2). The quantity of roots in new-era maize hybrids exhibited an increasing trend, potentially indicating that a lack of roots may not meet the yield demands, and only by increasing the root quantity appropriately can sufficient nutrients be provided for maize yield.

With the passage of time, the increase in maize yield is accompanied by the optimization of root architecture, which indicates that the evolution of maize varieties has led to increased longitudinal tension and lateral contraction [19]. Larry et al. [65] evaluated 16 maize hybrids widely planted in the United States in the past century under the condition of high and low nitrogen, and showed that the root angle of modern maize varieties decreased. Their experimental materials included wild, farmed, and local varieties, which differed from the experiment, which caused inconsistent results. This study demonstrated that the BA, CA, and D₉₅ of new-era maize hybrids were significantly higher than those of old-era maize hybrids under both nitrogen levels, indicating that new-era maize hybrids possessed stronger root piercing ability. Hund et al. [66] used UH005 and UH250 to hybridize at the optimum temperature for lateral root growth at 34 °C and 28 °C. Results showed that when the temperature was 31 °C or lower, the root growth of the hybrid at the seedling stage was better than that of the parent, indicating heterosis in root growth. This study observed heterosis in root architecture of both old- and new-era maize hybrids, with the heterosis in root architecture of new-era maize hybrids being more pronounced compared to that of old-era maize hybrids. These differences were found to be stable, which was manifested by the fact that the BB and CB of new-era maize hybrids, and D₉₅, had positive heterosis, while the BA, CA, and CB of old-era maize hybrids, and the BN and CN, had negative heterosis (Figure 1, Figure 2, Figure 3 and Figure 4). Both positive and negative heterosis can be considered positive as long as they facilitate nitrogen uptake and accumulation.

Root architecture is a promising breeding target, which has the potential to increase maize yields and food security, while also reducing environmental impacts. The “steep, cheap, and deep” ideotype is beneficial for capturing nitrogen and water [18]. Maintaining an optimized root architecture, and increasing surface area and the volume and dry matter of the root, enhances nitrogen absorption [2]. A study by Chen X et al. [67] indicated that the grain nitrogen content of modern maize varieties decreased, but nitrogen uptake increased, which led to delayed leaf senescence and improved nitrogen use efficiency (NUE), by studying maize varieties from 1973 to 2000. Wang et al. [5] showed that the NUE of maize hybrids and their parental inbred lines increased synchronously, and the evolution of the NUE in maize hybrids was mainly attributed to nitrogen internal efficiency (NIE), while the evolution of the NUE in maize parental inbred lines was mainly attributed to nitrogen recovery efficiency (NRE). The present study showed that vegetative-stage whole plant N uptake, NUE, and NIE of new-era maize hybrids were significantly higher than those of old-era maize hybrids, and all of them were higher than their parental inbred lines (Table 4). The increase in NUE in maize hybrids was mainly due to the increase in NIE. However, the higher NIE was reliant on sufficient pre-anthesis nitrogen accumulation, and nitrogen transport in post-anthesis was mainly related to nitrogen accumulation in pre-anthesis [68,69]. Li et al. [33] found that the significant enhancements in NUE of new-era maize hybrids, and NUE heterosis, increased. In this study, the heterosis for NUE of new-era maize hybrids was significantly higher than that of old-era maize hybrids, and it mainly depended on the heterosis for NIE (Table 5). The coefficients of determinations of MPH_NIE on MPH_NUE (71.5% of old-era maize hybrids and 84.0% of new-era maize hybrids) were significantly higher than those of MPH_NRE on MPH_NUE (38.3% of old-era maize hybrids and 64.4% of new-era maize hybrids), which was 28.6% higher on average (Figure 5) and indicated that the NUE heterosis mainly depended on the NIE heterosis.

Establishing an effective root architecture in maize is a crucial approach for enhancing both yield and nitrogen use efficiency [9]. This requires a comprehensive understanding of the intricate relationship between root development and nutrient uptake in maize plants. The SEM demonstrated that AH_NIE and MPH_NIE, and AH_D95 and MPH_D95 directly contribute to AH_NUE and MPH_NUE (Figure 7), which indicated that there were two pathways by which AH_NIE and MPH_NIE could influence AH_NUE and MPH_NUE. One pathway was heterosis for denser root architecture (AH_BN and MPH_BN, AH_CN and MPH_CN, AH_BB and MPH_BB, and AH_CB and MPH_CB), which significantly influenced AH_NIE, MPH_NIE, AH_NUE, and MPH_NUE (p < 0.001). The other pathway was heterosis for steeper root architecture (AH_BA and MPH_BA, and AH_CA and MPH_CA), which mediated through AH_NRE, MPH_NRE, and thus AH_NIE and MPH_NIE, and AH_NUE and MPH_NUE (p < 0.05). Therefore, heterosis for steeper-denser-deeper root architecture affected heterosis for NIE and NUE.

Increased yields of maize have typically been linked to higher planting densities during the past several decades [46,47]. Rinehart et al. [70] suggested that the pursuit of higher yields has indirectly led to a decrease in the size of the root system in maize over the course of 80 years of breeding. Studies have shown that high planting density can result in a reduction in the root dry weight of maize [71]. As planting density increases, competition for nutrients and water among plants intensifies, potentially leading to higher energy consumption due to excessive crown roots [72]. Shao et al. [73] further demonstrated that, under high planting density conditions, maize plants tend to decrease nodal root number and lateral root growth while maintaining axial root elongation to cope with limited photosynthesis supply from the shoot to the root. Given the current trend towards very high planting densities, it may be worth considering further adaptive changes in root structure to optimize nutrient uptake efficiency. This could include the development of a single layer of brace roots, a moderate number of crown roots, and deeper, thicker roots to enhance penetration and transport capacity for efficient nitrogen uptake [72]. Overall, these findings suggest that there is potential for enhancing maize root structure to improve nutrient uptake and overall plant productivity, particularly in the context of high planting densities.

In this study, only heterosis for the root architecture of maize hybrids from different decades was compared and analyzed, but heterosis for root-and-shoot co-ordination should be further analyzed. Additionally, this study lacks research on heterosis for root physiology, such as heterosis for root exudates and root metabolism. Future studies should consider analyzing heterosis for root physiology and metabolism. Furthermore, examining the complex interactions between plant roots and soil microbes could help in manipulating soil microbial composition and function to enhance nitrogen use efficiency (NUE), crop productivity, and environmental sustainability [74].

5. Conclusions

Compared with old-era maize hybrids, the root architecture of new-era maize hybrids was steeper, deeper, and denser. The increase in NUE in maize hybrids was mainly due to the increase in NIE. The BB and CB of new-era maize hybrids, and D₉₅, showed positive heterosis, whereas BA, CA, and CB of old-era maize hybrids, BN, and CN, showed negative heterosis. With the change in maize hybrids, the increase of BA, CA, and CN not only enhanced root penetration, but also promoted pre-anthesis N absorption and post-anthesis N transport efficiency. The heterosis in steeper-denser-deeper root architecture directly affected the NIE heterosis, ultimately affecting the NUE heterosis. This physiological aspect serves as a crucial foundation for further enhancing NUE heterosis in the new-era maize hybrids.