Abstract
Maize (Zea mays L.) production is often limited by edaphic stresses, viz., drought, salinity and nutrient deficiencies. The root system architecture (RSA) in is essential for efficient water and nutrient acquisition and overall plant stability under adverse conditions. Thus, identifying maize genotypes with desirable root characteristics under stress is a valuable strategy for breeding maize with enhanced abiotic stress tolerance and improved resource-use efficiency. The hydroponics-based root phenotyping offers control over the root environment, reduced labour, and almost damage-free phenotyping over field-based phenotyping. Here, we have investigated the adaptive RSA plasticity of maize under drought, low nitrogen, salinity, and non-stress environments. The results revealed significant variability for all the RSA-associated and related traits. The salinity-stressed genotypes showed adaptive plasticity with enhanced average root diameter (AD), whereas the enhanced total root length (TRL) and surface area (SA) were reduced along with average root diameter under low-N stress. The RSA traits TRL (total root length), SA (surface area), RV (root volume), SFW (shoot fresh weight), SDW (shoot dry weight), RFW (root fresh weight) and RDW (root dry weight) showed positive and significant correlations across the stress conditions (0.47 to 0.99; p < 0.001 to 0.05). Conversely, AD showed significantly negative correlations with SRL (-0.70 to -0.95; p < 0.001) of genotypes, stress (environment) and genotypes × stress (environment) interactions suggested the possibility of developing stress-specific maize cultivars by selecting inbred lines with high adaptive plasticity to specific stress environments or broad adaptability by selecting lines with high stability across stress environments. The results suggested that the maize RSA adapts not only specific root types and traits to cope with various environmental stress conditions but also the strength and directions of trait association. Notably, inbreds SKV671 and CML597 exhibited superior performance for various RSA traits under drought and low nitrogen conditions, while PML93 and MG42 excelled in salinity stress.
Similar content being viewed by others
Introduction
Maize (Zea mays L.) is an important crop in national and global agricultural systems owing to its contribution to food and nutritional security, besides serving as raw material for various industries1. However, maize production faces numerous challenges, ranging from abiotic stress factors such as drought and nutrient deficiency to biotic pressures from pests and diseases. Particularly, when evaluating the effects of climate change on crops, maize appears to be the most adversely impacted2, especially stresses such as drought, salinity, and nutrient deficiencies are recognized as major factors that adversely impact growth and productivity. In nature, the often occurrence of these stresses simultaneously in the crop ecologies have synergistic adverse effects on crop growth and biological and economic yield3,4. During drought, salinity and nutritional deficiencies, the root system is the first to encounter these stresses and undergoes adaptive modifications to optimize water, ions and nutrient uptake and minimize stress effects. Complex and diverse root system plays a crucial role in water and nutrient uptake, plant stability, and overall growth in maize. The RSA exhibits remarkable plasticity in response to environmental changes and could be critical for improving crops with efficient root systems5,6.
Recent studies have shown that maize responds to drought by altering total root length (TRL), root surface area (SA), root volume (RV) and other root traits7,8,9. The genotypes known to show reduced total root length under drought stress in the vermiculate10 and PEG treated medium11. Similarly, the salinity stress alters the root morphology, reducing root elongation and branching, while triggering mechanisms such as ion homeostasis and osmotic adjustment to mitigate stress effects12. The enhanced average diameters of plants under salinity due to thicker root periderms with salt ion accumulation or increased size of xylem vessels in their roots to improve their ability to transport water and nutrients13,14. However, accumulated salt ions inhibit root elongation by reducing the size of the root meristem, as well as the new cells produced for root hair formation, thus affecting the overall root growth13,14. In contrast to salinity stress, the N-deficiency enhances TRL, SA, RV and root biomass through activation of “root foraging response” in search of N. Maize roots exhibit high plasticity under nitrogen-deficient conditions by increasing root length and surface area to explore larger soil volumes for nutrient acquisition15. Notably, maize roots show continued growth for root dry weight (RDW) and total root length (TRL) throughout their development, even in low nitrogen conditions16. Despite the prominent role of roots in adapting the plant responses to abiotic stress tolerance, selection in maize breeding is typically centred on shoot performance, as evaluating root characteristics is more complicated and requires specialized, often elaborate phenotyping systems. Furthermore, much of the research on roots tends to focus on their development, with a notable gap in understanding how root plasticity directly contributes to tolerance mechanisms. Thus, understanding root system architecture (RSA) and the mechanisms of its development will allow manipulation and exploitation of different root traits to improve plants’ adaptation to changing climate and increase the yield for the growing population17.
To breed maize for edaphic stresses, it is crucial to accurately phenotype root traits and examine the structural changes in root architecture under target stress conditions. Historically, most researchers have focused on traits such as flowering time (anthesis and silking) and yield-related attributes to identify stress-tolerant lines for drought, salinity and nitrogen stresses. The root traits for these traits are often overlooked owing to the difficulties in phenotyping, due to the labour-intensive and time-consuming nature of root system evaluation in the field, and limited understanding of RSA genetics. Moreover, maintaining consistent stress levels across field conditions and effectively characterizing root traits remain significant challenges despite root characteristics playing a vital role in survival under various abiotic stress conditions17.
The efforts were made on a limited scale in cereals to capture the RSA in response to stress environments using field phenotyping using rhizotrons18, mini-rhizotrons19,20 and Shovelomics21. However, these methods are time-consuming and labour-intensive and cannot be used in experiments with large numbers of lines22,23. For these concerns, hydroponics is widely regarded as an efficient system for studying plant responses to various abiotic stresses, particularly for root screening. This method offers a uniform growth environment by ensuring adequate nutrient availability tailored to the crop’s requirements. Furthermore, the hydroponics-based RSA phenotyping captures a snapshot of RSA with minimal mechanical damage and allows a non-destructive way of phenotyping. As a rapid and effective screening approach, researchers worldwide have extensively utilized hydroponics to explore root trait variability in major crops at the seedling stage24.
Investigating natural variations in root responses to edaphic stresses can offer valuable insights to identify the maize line with specific allelic combinations that show adaptive responses25. Additionally, the RSA plasticity to these abiotic stresses is a function of its genetic background and its ability to modify its architecture in response to environmental stimuli. Genetic studies have highlighted the importance of root-related traits in breeding for stress resilience26. Despite the proven relevance of RSA for stress resilience and adaptation, root traits have historically been understudied due to the complexity of their measurement and observation under field conditions. Recent studies on maize root systems under stress conditions revealed significant heritability and genetic variation for key root traits9,27. However, there are limited studies on maize RSA plasticity in response to multiple abiotic stresses, although few are available for single stresses28,29,30,31. Thus, the present experiment aimed to investigate the plasticity of root systems’ architecture traits under hydroponically simulated drought, salinity, and low-nitrogen stress, to identify genotypes and traits with potential for stress-resilient breeding.
Materials and methods
Plant materials
A set of 20 diverse maize inbred lines showing variable responses to abiotic stress tolerance and belonging to international (CIMMYT maize inbred lines) and Indian breeding programmes (IARI and AICRP-Maize) was used to study the RSA under drought, salt, and nitrogen starvation stresses under hydroponic conditions (Table S1).
Experimental design, treatment and growing conditions
The experiment used a randomised complete block design (RCBD) for three abiotic stresses (drought, salinity, and low nitrogen) and control treatments. The seeds of each inbred line were surface sterilised with 1% NaClO solution for 5 min, followed by seven rinses with double-distilled water. The surface-sterilised seeds were germinated on a double-distilled water-soaked germination paper roll. For this, the surface-sterilised 50 seeds were placed on double-distilled water-soaked germination paper and properly rolled for putting in the germination chamber. The seedlings were germinated and grown for eight days in the germination chamber at 25 °C. The 25 uniform seedlings were transferred into plastic trays (40 cm [length] × 34 cm [width] × 15 cm [height]) containing 15 L of half-strength nutrient solution with constant aeration created with aquarium air pumps and grown the seedlings for seven days (Table S2; Figure S1). The hydroponic experiment was conducted in the glasshouse with a controlled environment at the National Phytotron Facility, ICAR-IARI, New Delhi. The glasshouse chambers were maintained at a temperature of 25–30 °C, and the relative humidity was around 60% with natural light of 13 and 12 h of light and dark period. The pH of the medium was checked during the growth period and maintained at a stable level of around pH 6.0. The seedlings were maintained for one week under half-strength nutrient solution hydroponics for better adaptation.
Creation of stress treatments
After seven days, the best five plants per genotype per stress (environment) were transferred into plastic trays (40 cm [length] × 34 cm [width] × 15 cm [height]) containing 15 L of full-strength nutrient solution with constant aeration with to a pH of 6.0 (Table S2). The salinity stress was created by adding 5 M sodium chloride (NaCl) to the above complete-strength nutrient solution (Table S2). The nitrogen stress was created by replacing the 4mM Ca(NO3)2 (control) with 2 µM Ca(NO3)2 (nitrogen starvation) (Table S2). To mimic the gradual occurrence of a drought environment, the drought stress was created with polyethylene glycol (PEG 6000; SRL Pvt. Ltd., India) at two levels. At the first level, 10% of PEG 6000 treatment was imposed for five days, followed by 15% PEG 6000 for the subsequent ten days (Table S2). The respective nutrient solution was replaced every five days in control and all the stress-imposed trays. The pH was monitored daily and adjusted the same with 0.01 M NaOH and 1 M HCl.
Phenotyping of root system architecture and shoot parameters
After 15 days of stress treatments, the RSA-associated traits were measured from three plants randomly selected from five plants per genotype in each stress condition. Each plant was carefully taken out from the tray without any mechanical damage to the root system. The root of each plant was properly spread in the transparent scanning tray half-filled with water and placed the same on the scanner. After checking and adjusting the spread of roots, the WinRHIZO software installed in the root scanning system was used to scan the root system. Through the root scanning facility, the RSA-associated traits, viz., TRL-total root length (cm), SA-surface area (cm2), AD-average diameter (mm), RV-root volume (cm3, TN-total number of roots (N), RL-root length (cm), SL-shoot length (cm), SFW-shoot fresh weight (g), RFW-root fresh weight (g), SDW-shoot dry weight (g), RDW-root dry weight (g), SPADBS-SPAD value before stress imposition, SPADLP-SPAD value at 15th day after stress imposition, SRL-specific root length, RLDp5-root length distribution with less than 0.5 mm were measured (Table S3).
Statistical analyses
Descriptive statistics and multivariate analysis
Descriptive statistics, viz., the mean, range (minimum to maximum), coefficient of variation and standard errors, were worked with the stats32 package of R. The multivariate analyses were performed with the stats32 and FactoMineR33 packages. The dissimilarity matrices were calculated for RSA and related traits with 1000 bootstraps using Darwin v. 6.0.021 software for studying genetic diversity among the genotypes34, and the dendrogram was generated through a weighted neighbour-joining method based on the usual Euclidean distance employing Darwin v. 6.0.021 34 to visualize the grouping patterns of genotypes.
Analysis of variances
The analysis of variance (ANOVA) for each stress treatment and control was performed as per the following model:
Where µ is the overall mean of the target trait, αi is the effect due to the ith treatment on the target trait, βj is the effect due to the jth block, and εij is the error term. The ANOVA was computed using the stats32 package of R.
Additive main effects and multiplicative interactions (AMMI) for root plasticity traits
The additive main effect and multiplicative interaction (AMMI) model was used to study the effects of G × Stress interaction on maize root phenome. The AMMI model combines both the main effects of genotypes and stresses with their interaction effects using a singular value decomposition approach. This model is particularly suitable for assessing the stability of genotypes across diverse environmental conditions35. The AMMI analysis for root plasticity was executed with the following model:
Where Yij = target trait response of ith genotype (i = 1, 2, …., I) in jth (j = 1, 2, ……, J) environment/stress; µ is the general mean; gi = main effect associated to the ith genotype; ej = main effect associated to the jth environment; N = max (G-1, E-1), i.e. the number of principle axes (principal components) retained in the model to describe the pattern of the interaction between the ith genotype with the jth environment; λn = singular value of the nth principle interaction axis; γin = ith element of the singular column vector associated to axis n; αjn is the jth element of the singular row vector associated to axis n; ρij is the AMMI residue; and ϵij = pooled error.
Realization and visualization of root plasticity traits through GGE biplots
The GGE biplot integrates genotype main effects and genotype-by-environment (stress) interaction effects into a graphical representation using singular value decomposition (SVD). The GGE biplot was used to interpret the “mean vs. stability” plot for genotype evaluation and the “discriminating power vs. representativeness” plot for environment evaluation. The plots were generated using the “which-won-where” pattern to determine the discriminating ability and representativeness of the test environment. The GGE biplot analysis was performed as per the following model:
In the above equation, the Yij and \(\:{\stackrel{-}{Y}}_{.j}\) represents an average mean of the ith genotype in the jth environment and the average mean over all genotypes in the jth environment, respectively. Whereas λ1ξi1ηj1 and λ2ξi2ηj2 are the collective first (PC1) and second (PC2) principal components; λ1 and λ2 and singular values for PC1 and PC2, respectively; ξi1, ξi2, ηj1 and ηj2 are the PC1 score for ith genotype, PC2 score for ith genotype, PC1 score for jth environment and PC2 score for jth environment, respectively.
Results
Phenomic plasticity of maize root system architecture (RSA) of diverse maize inbred lines showed visible adaptive plasticity to various RSA-associated and related traits under target environments, viz. control, drought, low-nitrogen and salinity stress (Figure S2-S6; Fig. 1).
Heatmap depicting the response of maize RSA and associated traits in response to control, drought, low-nitrogen and salinity environments under hydroponic conditions. The colour code from darker blue to darker red represents the lower to higher values for target traits.
Descriptive statistics of RSA and associated in maize
The analysis of variances for comparing RSA-associated and related traits means showed significant divergence for all the target traits (p < 0.05) except the SPADBS (SPAD value before stress initiation; p = 0.19) under target treatments, viz. control, drought, low-nitrogen and salinity stress (Table S4; Fig. 2). The mean performances of genotypes under control conditions were highest among all the treatments for TRL (1626.28 cm), SA (225.93 cm2, RV (2.55 cm3, TN (2271.30), shoot length (18.96 cm), root fresh (4.01 g) and dry (0.20 g) weights, and shoot fresh (11.91 g) and dry (0.60 g) weights. However, the genotypes in control conditions showed the lowest AD (0.46 mm). Interestingly, drought stress showed the lowest means for TRL (391.59 cm), SA (78.31 cm2, RV (1.27 cm3, TN (476.35), shoot length (10.50 cm), root fresh (2.02 g) and dry (0.11 g) weights, and shoot fresh (2.67 g) and dry (0.14 g) weights, where the means of AD (0.63 mm) were found highest in the drought environment compared to other conditions. Overall, the mean plasticity of RSA target traits, viz., TRL, SA, RV, TN, RL, SRL and RLDp5 under low-N were closer to the mean values under control environments. Meanwhile, the mean values of the same RSA target were closer between the salinity and drought stress. Further, the biomass-related trait means were reduced in all the stress environments (SL: 10.50–12.79 cm; SFW: 2.67–5.56 g; SDW: 0.14–0.27 g; RFW: 2.02–3.62 g; RDW: 0.11–0.18 g) as compared to control (SL: 18.96; SFW: 11.91 g; SDW: 0.60 g; RFW: 4.01 g; RDW: 0.20 g) environments (Table S4). Further, the genotypes CML579, CML597, PML69, PML93 and SKV671 showed high mean values under control, low-N and drought environments for various RSA adaptive traits, viz., TRL, SA, RV, TN and RL (Tables S5-S8), where MG42, PML47, PML93, and PML69 showed high means under salinity stressed environment (Table S8).
The box plots depict the descriptive statistics of RSA-associated and related traits. The red, sea serpent blue, green, and Chinese blue boxplots represent the control, drought, low-nitrogen and salinity stress environments. (A), Total root length (cm). (B), Surface area (cm2. (C), Average diameter (mm). (D), Root volume (cm3. (E), Total number of roots (N). (F), Root length (cm). (G), Shoot length (cm). (H), Shoot fresh weight (g). (I), Root fresh weight (g). (J), Shoot dry weight (g). (K), Root dry weight (g). (L), SPAD value after stress. (M), Specific root length. (N), Root length distribution. (O), SPAD-value before stress.
Multivariate analysis of RSA and associated in maize
Correlation analysis
The correlation coefficients were computed to quantify and visualize the association among RSA-associated and related traits under various environments (Table 1). Among all the target traits, TRL, SA, RV, SFW, SDW, RFW and RDW showed positive and significant correlations among each other (Table 1). The strong correlation between the shoot fresh and dry weights with these RSA-traits suggests that the robust root system supports better above-ground growth. Similarly, SRL and RLDp5 showed a significant positive correlation (0.65 to 0.90; p < 0.01) in control and stress conditions. These results indicate that plasticity responses of these traits are highly interrelated and conserved, suggesting that improvements in one trait could also improve the other linked traits. Further, the AD traits showed significantly negative correlations with SRL (-0.70 to -0.95; p < 0.001) and RLDp5 (AD: -0.87 to -0.98; p < 0.001) in control and stress conditions. These significant negative correlations highlight potential trade-offs in root trait selection for simultaneous improvement. Besides, the SPAD contents did not show any association with RSA-traits in stress environments, where the TRL (0.47; p < 0.05) and RL (0.54; p < 0.05) from drought stressed plants showed positive correlation with SPAD content.
Further, the strength of association among several RSA-associated and related traits was impacted by various stress environments. For instance, the correlations between RL with TRL, SA and AD were not significant under control, drought and low-N, whereas the significant positive (TRL: 0.77, p < 0.001; SA: 0.70, p < 0.01) and negative (-0.87, p < 0.001) correlations were observed under salinity stress conditions. Similarly, the RL showed significant positive correlations with TN under control (0.55, p < 0.05), low-N (0.60, p < 0.01) and salinity (0.77, p < 0.001), whereas the associations were non-significant under drought stress (0.35, p = 0.07). The AD showed a positive and significant correlation with RV under drought stress (0.63, p < 0.01), whereas the same was found non-significant under control (0.30, p = 0.17), low-N (0.15, p = 0.09) and salinity (-0.04, p = 0.16) conditions. The altered strength of association among the traits could be associated with stress-induced altered RSA associated with root phenome plasticity of component traits.
Additionally, the stress environment affects not only the association’s strength but also the association’s direction. The RLDp5 showed a significant positive and negative correlation with SA under salinity (0.54, p < 0.05) and drought (-0.55, p < 0.05) conditions, respectively, whereas RLDp5 association with SA were found to be non-significant under low-N (0.01, p = 0.08) and control (0.04, p = 0.11) conditions. Similarly, drought showed significant negative correlations between RLDp5 and SFW (-0.53, p < 0.05) and RLDp5 and SDW (-0.66, p < 0.01), whereas the same associations were found significantly positive (RLDp5 and SFW: 0.63, p < 0.01; RLDp5 and SDW: 0.61, p < 0.01) under salinity stress condition.
Cluster analysis
The genetic distances were computed using RSA-associated and related traits to examine the relationships among the target genotypes based on Euclidean distances generated with 1000 bootstraps (Fig. 3A). The hierarchical clustering of genotypes based on the weighted pair group method with averaging (WPGMA) revealed three main clusters in the dendrogram. The cluster-I emerged as the most significant cluster with nine genotypes encompassing four indigenous (MG42, PML35, PML47 and PML69) and five exotic (CML503, CML563, CML579, CML580 and CML601) maize inbred lines. The cluster-II comprises seven genotypes, comprising four Indigenous (LM17, MG50, PML52 and UMI1201) and three exotic (CML566, CML597 and SKV671) maize lines, placed in the second position, followed by cluster-III with three Indigenous maize lines (PML46, PMl71 and HKI1105) in the third place. Interestingly, the elite breeding line PML93 emerged as an independent line without showing any grouping pattern with the other genotypes. The mean genetic distance of the dendrogram was recorded as 1785.13, with the minimum genetic distance between PML35 and CML563 (224.57) and the maximum genetic distance between PML93 and PML46 (6205.69).
The dendrogram and PCA plot show the grouping of maize genotypes for RSA-related traits. (A), The dendrogram shows the clustering of diverse maize inbred lines for the RSA-related and associated traits mean values. The genotypes in red and blue colour indicates the exotic and indigenous inbred lines, respectively. (B), The PCA plot shows the distribution of genotypes for RSA-associated traits in maize. The X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2).
Further, the individual dendrograms were generated for each stress condition to compare the clustering patterns of genotypes. Six tanglegrams were generated pairwise among the stress environments, and cophenetic correlation coefficients were worked out for the same (Figure S7A-G). The analysis of tanglegrams with cophenetic correlation coefficients showed a significant association between the clustering pattern of control and low-nitrogen stress conditions (with a cophenetic correlation coefficient of 0.52 (p < 0.01) (Figure S7G). Further, control and low-nitrogen stress conditions shared a common clustering pattern of HKI1105 and MG42 (Figure S7B). Although control and salinity (0.06, p = 0.14), drought and salinity (0.05, p = 0.11), and low-N and salinity (0.02, p = 0.10) tanglegrams showed common clustering patterns of PML69 and CML580, SKV671 and LM17 and CML597 and CML 579, respectively; however, the correlation coefficients of these tanglegrams were low and non-significant. On the other hand, the tanglegrams between control and drought (Fig. 3A) and low-N and drought (Figure S7E) showed no common clustering pattern for any of the genotypes with non-significant correlation coefficients of -0.13 (p = 0.19) and − 0.06 (p = 0.21), respectively (Figure S7G).
Principal component analysis
Although 16 principal components captured the total variation, only the first four principal components showed eigenvalues > 1. The first principal component (PC1) explains 48.58% of the total variance in the dataset (Figure S8A). This suggests that nearly half of the variability in the data can be captured by a single component, indicating its significant contribution toward explaining total variation. The second principal component (PC2) had an eigenvalue of 3.39, contributing 21.18% to the cumulative variance. Together, PC1 and PC2 account for 69.76% of the total variance, which is a substantial portion of the variation that exists among the target genotypes. The traits SA (12.26%), TRL (11.58%), RDW (11.42%), SFW (11.37%) and SDW (11.02%) were the top five contributors to the total variations explained by PC1 (Figure S8B). Similarly, AD (27.12%), RLDp5 (25.37%), SRL (22.74%), TN (6.18%) and SPADLP (3.35%) were the five leading contributors to the total variations explained by PC2 (Figure S8C).
The PCA plot showed a similar grouping pattern of genotypes in the dendrogram (Fig. 3B). The PCA plot represented the highest genetic distance between PML93 and PML46 in the distance matrix of the dendrogram. Furthermore, PML46, PML71 and HKI1105 of cluster-III of the dendrogram were closely distributed on the PCA plot. Interestingly, we didn’t observe any grouping pattern based on the origin or source of genotypes, except for cluster-III, which showed only indigenous genotypes.
Analysis of variances for RSA and associated traits in maize
Analysis of variances (ANOVA) for all the target traits was computed to decipher the source of variations within each stress environment. The ANOVA revealed the highly significant contribution of genotypes for all the target traits in each stress environment (Tables 2, 3, 4 and 5). For control conditions, all the traits showed the contribution of a higher mean sum of squares (MSS) value with higher significance (p < 0.001). The block effects were found significant (p < 0.05) for AD and TN (Table 2). Similarly, for drought, low-N and salinity stress conditions, the genotypes explained significant variances for the target traits (Tables 3, 4 and 5). However, the block effects for SRL and SPADLP under drought, AD, SL, SFW, SDW and RDW under low-N (p < 0.05) showed significance at p < 0.05 (Table 4), although these block effects were generally more minor than the genotype effects. Further, the residuals showed minimal variation with non-significance across the traits. This indicates that most of the variance for RS-associated and related traits in maize is attributed to the genotypes.
Analysis of variances for additive main effects and multiplicative interactions of maize RSA traits
Analysis of variance for AMMI was performed for all the RSA-related and associated traits to know the contribution of genotypes, stress (environment) and genotypes × stress (environment) interactions. The analysis showed significant (p < 0.001) contributions of genotypes, stress (environment) and genotypes × stress (environment) interactions towards the total variations. Among these, the impact of stress components was found to be more pronounced, followed by genotypes and interaction components (Table 6).
Assessing maize RSA traits plasticity through additive main effects and multiplicative interaction biplots
Genotypes × stress (environment) interactions for RSA- RSA-associated and related traits were analyzed using AMMI biplots. The AMMI biplots were constructed employing IPC1 and IPC2 to decipher the magnitude of interactions of each genotype with different stress conditions.
For total root length, the first principal component (PC1) explained 50.40% of the variance, while the second principal component (PC2) explained 38.50%. Similarly, the genotypes MG42, PML47 and CML566, positioned in the direction of the salinity vector, are likely to show more adaptive plasticity for salt tolerance. PML69 and PML93 were placed between control and saintly vectors, suggesting their better adaptability under normal or salinity conditions (Fig. 4A). Further, the PC1 and PC2 components in the AMMI biplot explained 45.3% and 30.3% of the total variance for the surface area (SA) of roots. The genotypes MG42 and PML47 plotted near the salt vector, suggesting their adaptive plasticity to salinity stress tolerance. The genotypes, UMI1201 and PML52 have better surface area plasticity for no-stress (control) and drought-stress conditions. However, the mean surface areas of PML52 and CML503 were high under drought stress compared to remaining stress and no-stress conditions. Additionally, the PML93 and PML69 showed better root surface area plasticity for both low-nitrogen stress and salinity stress (Fig. 4B). Interestingly, for both total root length and root surface area, the genotypes PML46 and PML71 are positioned closely to the drought vector, suggesting that these genotypes have more substantial adaptive plasticity for both traits under drought conditions compared to low-N, salinity and no-stress conditions (Fig. 4A, B). On the other hand, the genotype SKV671, which falls between control and nitrogen (low-N) vectors is likely to show adaptive plasticity of total root length and root surface area for non-stress and low-N conditions.
AMMI genotype × stress (environment) interaction biplots for RSA-associated traits. (A), Total root length (cm). (B), Surface area (cm2. (C), Average diameter (mm). (D), Root volume (cm3. (E), Total number of roots (N). (F), Root length (cm). In each AMMI biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2).
Plants with smaller root diameters and deeper roots can adapt to various edaphic stresses and access more resources from deep soil layers. Often, the average root diameters negatively correlate with other key adaptive root traits. Thus, the genotypes with larger root diameters may pose negative plasticity for stress adaptability. Further, the entities falling away or on the opposite side of the vectors on biplots represent the least adaptive plasticity to the target vector. The AMMI biplot for root average diameter (AD) explained 95.4% of the total variation (PC1 = 75.30%; PC2 = 20.10%) (Fig. 4C). The genotypes MG42 and MG50 are aligned with the nitrogen vector, indicating a significant negative RD plasticity under low-N conditions. Similarly, CML579, PML69, and PML47 showed negative plasticity to drought stress. Where HK1105 and SKV671 were placed between salt and no-stress (control) vectors, showing negative plasticity to both conditions (Fig. 4C).
The AMMI biplot for root volume (RV) explained a total variation of 82.60%, with 48.10% from PC1 and 34.50% from PC2 (Fig. 4D). The PML47 showed plasticity to accumulate more RV under drought and salinity stress conditions than other genotypes and growing conditions (control and low-N). Similarly, CML579 and CML597 showed better adaptive plasticity for drought and low-N conditions. Where, SKV671, MG50, PML93 and LM17 showed better adaptive plasticity to low-N stress and control (non-stress) conditions (Fig. 4D).
For the total number of roots (TN), the AMMI biplots explained 94.30% of the variation, comprising 56.2% and 37.10% for PC1 and PC2, respectively (Fig. 4E). Genotypes plotted between the salt and control vector, such as MG50 and PML93, indicate their high adaptive plasticity to salinity stress and non-stress conditions. Additionally, MG42 and PML47 showed high adaptive plasticity to salinity stress (Fig. 4E). The genotypes SKV671 and UMI1201, plotted between drought and control vectors, indicated their adaptive plasticity to drought and non-stress conditions. Further, genotypes falling near the origin and base of the salt vector, viz., PML69, CML563 and CML579, suggest their adaptive plasticity across the stress conditions.
The first two PCs of the AMMI biplot for linear root length (RL) explained 55.60 and 28.40% of variations for PC1 and PC2, respectively, accounting for 84% of the total variation (Fig. 4F). The closeness of MG50 and CML580 to the control vector indicated their better RL performance for non-stress conditions. Whereas CML579 and LM17 were plotted near the nitrogen stress vector, indicating their linear root length plasticity to low-N conditions. Further, the petite length of the drought vector in the biplot shows their less discriminative ability than the rest of the conditions (Fig. 4F).
The shoot length AMMI biplot was plotted with 49.30% variance from PC1 and 37.00% from PC2, encompassing a total variation of 85.30% (Fig. 5A). The AMMI biplot for shoot fresh weight (SFW) explained total variations of 92.25% (PC1: 67.00%; PC2: 25.20% (Fig. 5B). Genotypes plotted near the control vector, viz. MG50 and CML580 showed better plasticity with enhanced shoot biomass accumulation under non-stress conditions. Further, the genotypes MG42, PML47 and PML35 showed adaptability to both low-N and salinity-stressed environments. The lines PML71, SKV671 and UMI1201 plotted between control and drought vectors, suggesting their capacity to accumulate higher shoot length and fresh weight for non-stress and drought environments (Fig. 5A, B). Similarly, the shoot dry weight (SDW) AMMI biplot captured 86.40% of total variations, with 75.30% from PC1 and 21.10% from PC2 for shoot dry weight. Whereas MG50, PML93, CML580 and PML47 showed adaptive plasticity for both shoot length, shoot fresh weight and shoot dry weight under non-stress and salinity-stressed conditions (Fig. 5A-C).
AMMI genotype × stress (environment) interaction biplots for RSA-associated traits. (A), Shoot length (cm). (B), Shoot fresh weight (g). (C), Shoot fresh weight (g). (D), Root fresh weight (g). (E), Root dry weight (g). (F), SPAD value after stress. In each AMMI biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2).
The AMMI biplots for RFW and RDW explained 80.60% (PC1: 46.70%; PC2: 33.90%) and 81.50% (PC1: 42.90%; PC2: 38.60%) of variations, respectively. The genotypes PML93 and CML579 showed adaptive plasticity for tolerance for root fresh and dry weights under salinity and low-N stress, respectively (Fig. 5D, E). Similarly, the genotypes LM17, MG50 and SKV671 showed better root fresh weight and dry weights plasticity to control (non-stress) and low-N conditions. Whereas, the genotypes UMI1201, CML563 and CML580 showed comparatively better adaptive plasticity to non-stress and salinity-stressed conditions (Fig. 5D, E). For the SPADLP values, the AMMI biplot explained 84.40% of variations from PC1 (55.8%) and PC2 (28.60%) components (Fig. 5F). The genotypes CML563, CML580 and MG42 explained better adaptive plasticity for chlorophyll content in non-stress and low-N stress environments, whereas PML69 showed high chlorophyll accumulation under non-stress conditions (Fig. 5F). Further, the genotypes CML566, MG50, HKI1105 and PML46 showed comparatively better chlorophyll synthesis under salinity and drought-stressed conditions (Fig. 5F).
The specific root length (SRL) AMMI biplot was plotted with 86.20% variance from PC1 (49.60%) and PC2 (36.60%) (Fig. 6A). Similarly, the relative root length distribution (RLDp5) explained a total of 93.40% of variations from PC1 (73.20%) and PC2 (20.20%), respectively (Fig. 6B). For SRL, the genotype PML93 showed adaptive plasticity to non-stress (control) and low-N stress conditions, whereas PML35 for the non-stress environment (Fig. 6A). Similarly, for low-N and drought stress conditions, the genotypes SKV 671, UMI1201, CML 601 and PML52 showed adaptive plasticity to both SRL and RLDp5 (Fig. 6A, B). Further, MG42 and MG50 showed adaptive plasticity for drought and salinity stress conditions (Fig. 6A, B).
AMMI genotype × stress (environment) interaction biplots for RSA-associated traits. (A), Specific root length. (B) Root length distribution. In each AMMI biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2).
GGE biplots analysis for RSA and associated traits in maize
The main effect of genotype (G) + G × S (GSI) interactions is the principal source of variation in assessing the genotypes under multi-stress trials. The GGE helps to elucidate genotypes showing the best adaptive plasticity to one or few target growing environments using ‘which-won-where’ biplots, which is a practical approach to visualize the pattern of GSI based on the correlation between G and GS. The GGE biplot comprises an asymmetrical polygon with stripes or lines running vertically from the biplot’s centre to the polygon at a right angle. All the genotypes apart from the biplot centre are linked with the polygon, thus covering all genotypes in the polygon marker. The vertical stripe that runs perpendicular to the polygon from the centre of the biplot represents an expected environment in which the two genotypes on opposite sides of the polygon are expected to behave similarly. Furthermore, it divides the biplot into different parts, each with its own enticing or winning genotypes. The captivating genotype is always positioned at the vertex of the polygon where both sides of the polygon meet that vertical stripe, generating a borderline of that segment or section.
The GGE biplots of TRL, SA, SRL, RLDp5, SFW, and SDW grouped the non-stress (control), low-N (nitrogen) and drought stress conditions as one mega-condition and salinity as a separate one (Figs. 7A and B and 8C-F). The RV, RL, RFW and RDW grouped non-stress (control), salinity and drought stress conditions as mega-conditions and low-N (nitrogen) as separate ones (Figs. 7D and F and 8A and B). Furthermore, the AD identified drought as one mega-condition and the rest (control, salinity, and low-N) as a separate mega-condition (Fig. 7C).
GGE-which-won-where biplots for RSA-associated traits plasticity. (A), Total root length (cm). (B), Surface area (cm2. (C), Average diameter (mm). (D), Root volume (cm3. (E), Total number of roots (N). (F), Root length (cm). In each GGE biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2). The genotypes are represented by blue solid circle, while environments are indicated by solid green diamond shapes.
GGE-which-won-where biplots for RSA-associated traits plasticity. (A) Root fresh weight (g), (B) Root dry weight (g), (C) Specific root length, (D) Root length distribution, (E) Shoot fresh weight (g). (F) Shoot dry weight (g). In each GGE biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2). The genotypes are represented by blue solid circle, while environments are indicated by solid green diamond shapes.
The SPADLP parameter grouped control (no-stress) and low-N (nitrogen) as one mega-condition, and drought and salinity-stressed conditions are the separate mega-conditions (Fig. 9B). However, the TN variable grouped all the target growing conditions as a single mega-condition (Fig. 7E).
GGE-which-won-where biplots for RSA-associated traits plasticity. (A), Shoot length (cm). (B), SPAD value after stress. In each GGE biplots, X-axis represents the first principal component (PC1) and the Y-axis represents the second principal component (PC2). The genotypes are represented by blue solid circle, while environments are indicated by solid green diamond shapes.
The GGE biplots explained 94.95% (PC1: 77.44%; PC2: 17.51%) and 90.65% (PC1: 80.55%; PC2: 10.20%) of variations for TRL (Fig. 7A) and SA (Fig. 7B). The “which–won–where” pattern showcases higher adaptive plasticity of the genotype SKV671 for mega-condition-1 comprising no-stress (control), drought and low-nitrogen (nitrogen) conditions for both TRL and SA (Fig. 7A, B). The PML93 showed higher adaptive plasticity to salinity stress for TRL and SA (Fig. 7A, B). For SA, additional prominent genotypes UMI1201 and PML47 showed adaptive plasticity for mega-conditions 1 of TRL (drought, low-N, no-stress) and mega-condition-2 of the SA biplot, respectively (Fig. 7B). For TRL, the CML579 and MG50 showed the slightest deviations for mega-conditions-1 (drought, low-N, no-stress) and mega-condition-2 (salinity), respectively (Fig. 7A). Similarly, for SA, MG50 showed the least deviations for the SA. Here, the performance of SKV671 ranked first, followed by UMI1201, based on the surface area of roots.
The enhanced AD is associated with negative plasticity for stress adaptability. The genotypes showing higher AD are the least adaptive to stress conditions. The GGE biplot pattern for AD of maize seedlings recorded 86.44% of total variations, comprising 69.72% from PC1 and 16.72% from PC2 (Fig. 7C). Two mega-conditions were identified for four conditions, i.e. control (no-stress), nitrogen (low-N) and salt (salinity) in mega-condition-1 and drought stress positioned in mega-condition-2. HKI1105 and PML46 show high negative plasticity for mega condition-1 and preferably with low-N and control vectors. However, the root plasticity response of HKI1105 was more than PML46. Further, CML597 showed higher negative plasticity under drought stress (Fig. 7C). For RV, the GGE biplot explained 89.20% (PC1: 75.63%; PC2: 13.57%) of variation with mega-condition-1 (control, drought and salt) and − 2 (nitrogen) (Fig. 7D). For RV, SKV671 showed high adaptive plasticity to mega-condition-1 and CML597 for mega-condition-2. The genotypes CML 563 and PML69 showed the least plasticity for root volume (Fig. 7D).
The GGE biplot captured 96.20% of variations for TN with PC1 explaining 81.01% of the variation and 15.21% with PC2 (Fig. 7E). All the environmental vectors grouped into single mega conditions suggest the lowest discriminative capacity of stress conditions for TN. The genotype PML93 showed a maximum number of roots with high plasticity, while genotypes PML69 and CML563 showed comparatively stable performance compared to other genotypes (Fig. 7E). The polygon view of the GGE biplot pattern for RL explained 94.95% of variations that comprise PC1, showing 59.5% of the variation, and PC2, showing 26.46% (Fig. 7F). Two segments have been divided into four mega conditions. The mega-condition-1 encompassed environmental control, drought and salt, and nitrogen stress positioned in another segment. CML579 has higher plasticity for RL under low-N stress (mega-condition-2). The plasticity of genotypes PML93 is least affected under target stress conditions. Further, MG50 shows higher adaptive plasticity to mega-condition-1 (control, drought and salt) (Fig. 7F).
The RFW and RDW GGE biplots explained 86.58% (PC1: 72.58%; PC2: 14.00%) (Fig. 8A) and 87.37% (PC1: 75.36%; PC2: 12.01%) (Fig. 8B) of variations, respectively. For both traits, SKV671 and CML579 showed higher adaptive plasticity to mega-condition-1 (drought, salinity, control) and − 2 (nitrogen), conditions, respectively. PML69 showed the least plasticity for both RFW and RDW traits (Fig. 8A, B). Further GGE biplot for SRL, explained 87.58% of variations with 66.91% from PC1 and 20.67% from PC2 (Fig. 8C). The genotypes PML93, CML601 and PML35 showed adaptive plasticity mega-condition-1 that comprises non-stress (control), low-N (nitrogen) stress and drought stress. Among these lines, PML 93 showed more adaptive plasticity to low-N stress and no-stress conditions, whereas CML601 showed more adaptive plasticity to drought stress. For salinity stress (mega-condition-2), the PML47 and PML69 showed quite adaptive plasticity (Fig. 8C).
For RLDp5, the GGE biplots were formed with PC1 (66.96%) and PC2 (17.61%), explaining a total variation of 84.57% (Fig. 8D). The biplot showed the genotypes CML601 and PML93 showing maximum adaptive plasticity to mega-condition-1 that includes low-N (nitrogen) stress, drought and no-stress (control) conditions. The PML35 and CML566 showed adaptive plasticity to salinity conditions that make a mega-condition-2 (Fig. 8D). Further, the CML601 in mega-condition-1 shows plasticity to low-N and drought stress, whereas PML93 to no-stress condition.
Similarly, for SFW and SDW, the GGE biplots explained 89.87% (PC1: 72.42%; PC2: 11.45%) (Fig. 8E) and 90.67% (PC1: 81.54%; PC2: 9.13%) (Fig. 8F). For both the traits, SKV671 and CML579 showed higher adaptive plasticity to mega-condition-1 (drought, nitrogen, salinity, control) and PML93 for mega-conditions-2 (salinity) conditions, respectively. PML69 showed the least plasticity for both RFW and RDW traits (Fig. 8E, F). PML69 and CML597 showed comparatively the least plasticity for both traits.
The GGE biplots of SL (Fig. 9A) and SPADLP (Fig. 9B) explained 85.41% (PC1: 68.18; PC2: 17.23%) and 90.71% (PC1: 77.36%; PC2: 13.35%) of variations, respectively. The SKV671 and CML579 showed the highest adaptive plasticity for SL in the mega-conditions-1 (drought, low-N and non-stress), whereas PML93 showed for mega-condition-2 (salinity) (Fig. 9A). Similarly, for SPADLP, the low-N and non-stress included under mega-conditions-1, and drought and salinity stress under mega-condition-2. The PML69 showed adaptive plasticity for chlorophyl content (SPAD value after stress) mega-condition-1 and genotypes CML566 and MG50 for mega-condition-2 (Fig. 9B).
Discussion
Amid global challenges like climate change, soil degradation, and resource shortages, enhancing crop resilience to abiotic stresses has become vital for agricultural research. Maize one of the most widely cultivated and economically important cereals, is highly vulnerable to environmental stresses such as drought, salinity, and low nutrient availability. These stress factors can severely affect maize growth, development, and yield, primarily by altering root system architecture (RSA), which is critical for water and nutrient acquisition from the soil. Understanding the phenological and genetic mechanisms behind maize RSA under varying environmental conditions is crucial for developing stress-resilient maize varieties. Root plasticity plays a key role in the plant’s adaptive strategies. The current investigation employed the hydroponic system to study the RSA in maize seedlings as a rapid and controlled methodology, although it possesses the setback of not simulating complex soil rhizosphere conditions and natural mechanical support by the edaphic environment and their interactions with genotypes. However, the RSA studies based on hydroponics provide preliminary and indicative information for selecting a few genotypes for field-level studies, besides allowing precise phenotyping of a large number of germplasm for RSA traits.
Our study examined the variability in RSA-related traits to uncover how maize respond to stress conditions. Additionally, the research explores shoot traits to better understand how the plant trade off with RSA traits. The RSA plasticity studies underline the significance of specific genotypes that demonstrate promising adaptability to stress conditions, indicating their potential utility for future maize breeding. The genotype-specific responses and the complex interaction of traits under stress highlights the need for precision breeding strategies tailored to specific environments. The current investigation reported significant variability in root and shoot traits under various environmental stress conditions, such as drought, low N, and salinity, demonstrating their effects on growth and resource acquisition (Fig. 2; Table S4). Among these, PEG induced drought stress is the most damaging, causing major reductions in key root traits like total root length (TRL), surface area (SA) and root volume (RV) (Fig. 2; Table S4). This is consistent with previous findings36,37, which show that root traits decline sharply under water-deficit conditions, reflecting the reduced ability to explore the soil for essential resources. Similarly, the TRL and RL reduction was also reported in PEG-treated lentil38 and maize30,39. The significant reduction in root biomass, as seen in root fresh weight (RFW) and root dry weight (RDW), further indicates that limited water availability impairs cellular turgor pressure and resource allocation for root growth40. Low-N stress also negatively affects root traits, though to a lesser degree than drought and salinity stresses (Fig. 2; Table S4). Nitrogen deficiency disrupts chlorophyll synthesis and photosynthetic capacity, which in turn influences root development41. Interestingly, specific root length (SRL) increases significantly under low-nitrogen stress, likely as a compensatory mechanism to enhance nutrient uptake by producing thinner and longer roots, which is less evident under drought and salinity stresses42. Salinity stress exhibits more genotypic-based variability in its impact on root traits. This suggests that some genotypes may possess mechanisms such as osmotic adjustment, which allow them to maintain root growth under salinity stress43. The results indicate that drought stress typically leads to an increase in average root diameter, though it also exhibits the greatest variability for a range of responses among the target genotypes44. Regarding shoot traits, the study finds that drought stress causes the most significant reductions in shoot length (SL), shoot fresh weight (SFW), and shoot dry weight (SDW), consistent with previous studies where the shoot growth is highly sensitive to water availability due to its reliance on turgor-driven cell expansion45,46. Low-nitrogen stress significantly lowers SPAD values, which is a measure of chlorophyll content and photosynthetic capacity, though genotypes like PML69 and CML601 maintain relatively higher SPAD values, indicating better chlorophyll retention under stress47.
The stability, stress-specific adaptability and plasticity are key aspects of the genotype’s response to the environment. The stability of genotypes emphasises the consistent performance of genotypes across the target environments, where the stress-specific adaptability is mainly associated with adaptive plasticity of genotypes (Figs. 4, 5, 6, 7, 8 and 9). For instance, the genotypes PML35, CML566 and CML579 fall near the origin and show minimal deviation from the axis in AMMI biplots (Fig. 4A, B), indicating their high stability performance for the traits TRL and SA, although they may not outperform on genotypes like PML93, SKV671 and UMI1201 genotypes (Table S5-S8). On the other hand, the genotypes PML93, SKV671 and UMI1201 showed the best performance for important RSA, viz. TRL, SA, RV, RL, RFW and RDW across the target environments, although the stress environments showed lower mean values (Table S5-S8), suggesting these genotypes have better adaptability for target traits. The genotypes with near-absolute stable performance generally possess genes or gene combinations that are least influenced by environmental conditions, and having this kind of stability with high performance is not viable in practical breeding. However, obtaining a genotype that maintains its yield in all environments is almost impossible48. Therefore, the performance of a stable genotype with a dynamic concept across stress environments is preferred as the genetic background of such genotypes possesses genes that allow them to adjust to various environmental conditions, leading to adaptability48,49,50.
In some cases, the genotype shows adaptive plasticity to specific stress or environmental conditions, although it may not perform its best under other stress or optimum environments. In the current investigation, CML597 showed adaptive plasticity for TRL, SA, RV and RL under low-N stress and AD under drought stress (Fig. 1; Table S5-S8). Further, the genotypes UMI1201, SKV671, PML52 and HKI1105 showed increased AD; where UMI1201, SKV671, PML93 and PML46 showed adaptive plasticity for enhanced SPADLP under drought and salinity stresses (Fig. 1; Table S5-S8). Thus, UMI1201 and SKV671 showed adaptive plasticity for both above- and below-ground traits for drought and salinity stresses. The root vascular system is an essential component for water and nutrient uptake, and any type of stress results in changes in these functional processes due to changes in vascular tissue in response. Under salinity stress, the root periderm accumulates salt ions and becomes thicker with an increased size of xylem vessels, resulting in increased AD for maintaining water and nutrient transport, which is an adaptive plasticity under a salinity-stressed environment13,14, where the low root AD diameters having low axial hydraulic conductivity for water uptake and mobile nutrient forage is an adaptive plasticity for drought and low-N stresses30,51. Supporting this, PML93 showed very little increase in AD and enhanced plasticity for other root traits under drought and low-N is plasticity as compared to other genotypes (Table S5-S8). Most interestingly, the mean plasticity values of SL. SFW and SDW were low for all the target genotypes across the stress environments compared to the control environment (Fig. 1; Table S5-S8). Similar results were also reported in rice, where stressed plants showed lower shoot length compared to non-stressed plants52. Although there is a trade-off with biomass, the reduced SL improves water use efficiency and reduces injury under drought stress53.
Further, the investigation also highlights genotype-specific stress responses. For example, under control conditions, genotypes like PML69 and SKV671 show higher TRL and SA, indicating robust root systems that may confer an advantage in resource acquisition. However, under drought stress, all genotypes experience a significant reduction in TRL, with UMI1201 and CML579 exhibiting the most substantial decrease. These findings align with previous studies showing that drought reduces root growth to minimize water loss while increasing the root-to-shoot ratio as a survival strategy54. Conversely, genotypes like CML579 and SKV671 maintain relatively higher TRL and SA under low-nitrogen conditions, suggesting they have root architecture traits that enable more effective soil exploration and N-absorption55. Additionally, genotypes such as MG50 and CML579 show higher RFW and RDW under low-N stress, indicating a genetic predisposition for sustaining biomass under low-N condition through enhanced carbon allocation to roots56. CML601 performs better for RLDp5 under drought and low-nitrogen stress, though its performance declines significantly under salinity stress, reflecting reduced root system architecture and soil exploration capacity. Overall, the variation in genotype responses across different stress conditions underscores the potential for selecting and breeding stress-resilient genotypes. Genotypes such as PML69 and CML597 demonstrate strong performance across multiple stress conditions, such as CML597’s better performance under drought and low-N for root length. This suggests the presence of adaptive traits such as deeper root systems, efficient water and nutrient uptake, and better maintenance of photosynthetic capacity57. The significant genotype-environment interactions observed in this study emphasise the importance of considering specific stress environments and focusing on traits that enhance both root and shoot performance in future breeding programs58. This approach is aligned with current precision breeding trends aimed at developing climate-resilient crops that can thrive under diverse environmental conditions59.
The analysis of RSA in maize under different environmental stress conditions provides significant insights into root trait plasticity and adaptability. Positive correlations among root traits like TRL, SA, RV, SFW, SDW, RFW, and RDW were found, indicating a coordinated response to environmental challenges and the potential for simultaneous improvements60,61. Negative correlations, particularly between AD and SRL, highlight trade-offs in resource allocation that are essential for selecting traits that enhance stress tolerance62. Environmental stresses significantly altered trait correlations; for instance, under salinity stress, RL showed strong positive correlations with TRL and SA but a significant negative correlation with AD, indicating that traits enhancing water uptake are prioritized. In contrast, those associated with root depth are less favourable under high salinity63. AD also correlated positively with RV under drought, consistent with findings that certain traits become more pronounced under drought to improve water uptake efficiency64. The direction of trait associations shifted under different stress conditions, reflecting the complexity of root trait interactions and the impact of environmental conditions on stress-responsive gene expression61. Hierarchical clustering and principal component analysis revealed significant genetic relationships and clustering patterns among maize inbred lines under various stress conditions. The dendrogram showed mixed clustering of genotypes, including both indigenous (MG42, PML35, PML47, PML69) and exotic lines (CML503, CML563, CML579, CML580, CML601), except for cluster III, which consisted solely of three indigenous lines (PML46, PML71, HKI1105). However, most clusters did not show a clear grouping based on geographical origin, suggesting a complex interplay of selection, breeding practices, and genetic drift. These findings have important implications, highlighting the genetic diversity within and between clusters and the potential to select lines with complementary traits. The variation in genetic responses to different stresses suggests a need for targeted breeding strategies that are tailored to specific stress conditions. Moreover, the elite breeding line PML93 stood apart from the other genotypes, clustering independently due to its unique genetic makeup, which might be resulted from targeted breeding efforts to enhance RSA traits.
The variance analysis demonstrated that genotype contributions were significant for all measured traits across different conditions, underscoring the importance of genetic variation in shaping phenotypic traits in both control and stress environments. Significant genotype effects on TRL across all conditions indicate diverse root growth patterns crucial for adaptation to environmental stresses like drought and low nitrogen, where larger root systems improve water and nutrient uptake65,66. Similarly, genetic variation in root architecture is vital for drought tolerance67. The significant impact of genotypes on SA suggests that genetic differences enable varied capacities to maximize water and nutrient absorption, particularly under stress68. Traits like AD and RV under salt stress demonstrate that root thickness and volume contribute to adaptation to mechanical stresses, such as soil compaction, with thicker roots better able to penetrate compacted soils and tolerate salinity42,69. The strong genotype effect on traits like total number (TN) and RL under stress emphasizes the importance of RSA in stress adaptation. Larger root number and length improve water and nutrient uptake under drought and nutrient-deficient conditions6,61. Significant genotype effects on root biomass under low-N stress suggest that genetic variation influences root biomass accumulation, which is crucial for resource acquisition and stress tolerance70. Furthermore, genotype significantly impacted shoot traits, which aligns with findings that both root and shoot growth are genetically influenced71. Differences in SPAD values before and after stress indicate that chlorophyll content and photosynthetic efficiency are genetically determined, with higher SPAD values indicating better maintenance of photosynthetic capacity under stress47. Significant genotype effects on SRL and RLDp5 under stress reflect genetic variation in root size, distribution, and efficiency, which enhance nutrient and water uptake in challenging environments72,73.
Conclusion
The current investigation deciphered adaptive root plasticity of maize inbred lines to various edaphic stresses. The RSA adaptive plasticity is prominently governed by genetic contribution, which has immense practical significance while breeding edaphic stress resilient and resource-use efficient maize cultivars. Furthermore, we have shown that edaphic factors affect not only the strength of the RSA traits association but also the direction of the association. The higher TRL, SA, RV and RL provide adaptive plasticity to drought, low-N stresses and salinity stresses, whereas the higher AD to salinity stress specific plasticity. For TRL, SA, RV and RL traits, SKV671 and CML597 exhibited superior adaptability under drought and low-N stress, while PML93 and MG42 under salinity stress. Similarly, for high AD plasticity for salinity stress is observed in HKI1105 and PML46. These inbred lines showing stress stability and stress specific plasticity could be used in the development of maize inbred lines to improve RSA plasticity for target stresses and resource use efficiencies.
Data availability
All data generated or analysed during this study are included in this published article and its supplementary information files.
References
Mallikarjuna, M. G. et al. Stability performance of inductively coupled plasma mass spectrometry-phenotyped kernel minerals concentration and grain yield in maize in different agro-climatic zones. PLoS One. 10, e0139067 (2015).
Tebaldi, C. & Lobell, D. Estimated impacts of emission reductions on wheat and maize crops. Clim. Change. 146, 533–545 (2018).
Zandalinas, S. I., Casal, J., Rouached, H. & Mittler, R. Stress combination: from genes to ecosystems. Plant. J. 117, 1639–1641 (2024).
Mittler, R. Abiotic stress, the field environment and stress combination. Trends Plant. Sci. 11, 15–19 (2006).
Khan, M. A., Gemenet, D. C. & Villordon, A. Root system architecture and abiotic stress tolerance: current knowledge in root and tuber crops. Front. Plant. Sci. 7, 209303 (2016).
Koevoets, I. T., Venema, J. H., Elzenga, J. T. M. & Testerink, C. Roots withstanding their environment: exploiting root system architecture responses to abiotic stress to improve crop tolerance. Front. Plant. Sci. 7, 215317 (2016).
Sharp, R. E. et al. Root growth maintenance during water deficits: physiology to functional genomics. J. Exp. Bot. 55, 2343–2351 (2004).
Ober, E. S. & Sharp, R. E. Regulation of root growth responses to water deficit. Advances in Molecular Breeding Toward Drought and Salt Tolerant Crops 33–53 (2007). https://doi.org/10.1007/978-1-4020-5578-2_2
Pace, J., Gardner, C., Romay, C., Ganapathysubramanian, B. & Lübberstedt, T. Genome-wide association analysis of seedling root development in maize (Zea mays L). BMC Genom. 16, 1–12 (2015).
Li, R. et al. Genetic variation for maize root architecture in response to drought stress at the seedling stage. Breed. Sci. 65, 298–307 (2015).
Elizabeth Mustamu, N. et al. Drought stress induced by polyethylene glycol (PEG) in local maize at the early seedling stage. Heliyon 9, e20209 (2023).
Claes, B. et al. Characterization of a rice gene showing organ-specific expression in response to salt stress and drought. Plant. Cell. 2, 19–27 (1990).
Robin, A. H. K., Matthew, C., Uddin, M. J. & Bayazid, K. N. Salinity-induced reduction in root surface area and changes in major root and shoot traits at the phytomer level in wheat. J. Exp. Bot. 67, 3719–3729 (2016).
Liu, W. et al. Salt stress reduces root meristem size by nitric oxide-mediated modulation of auxin accumulation and signaling in Arabidopsis. Plant. Physiol. 168, 343–356 (2015).
Kumar, B., Abdel-Ghani, A. H., Reyes-Matamoros, J., Hochholdinger, F. & Lübberstedt, T. Genotypic variation for root architecture traits in seedlings of maize (Zea mays L.) inbred lines. Plant. Breed. 131, 465–478 (2012).
Warncke, D. D. & Barber, S. A. Root development and nutrient uptake by corn grown in solution culture. Agron. J. 66, 514–516 (1974).
Smith, S. & de Smet, I. Root system architecture: insights from Arabidopsis and cereal crops. Philos. Trans. R Soc. B Biol. Sci. 367, 1441–1452 (2012).
Cope, J. E. et al. Precrop-treated soil influences wheat (Triticum aestivum L.) root system architecture and its response to drought. Front Plant. Sci 15, (2024).
Postic, F., Beauchêne, K., Gouache, D. & Doussan, C. Scanner-based minirhizotrons help to highlight relations between deep roots and yield in various wheat cultivars under combined water and nitrogen deficit conditions. Agronomy 9, 297 (2019).
Liedgens, M. & Richner, W. Minirhizotron observations of the Spatial distribution of the maize root system. Agron. J. 93, 1097–1104 (2001).
Le Marié, C. A. et al. Shovelomics root traits assessed on the EURoot maize panel are highly heritable across environments but show low genotype-by-nitrogen interaction. Euphytica 215, 173 (2019).
Uddin, M. S. et al. High-throughput root network system analysis for low phosphorus tolerance in maize at seedling stage. Agronomy 11, 2230 (2021).
Jones, J. B. Jr. Complete Guide for Growing Plants Hydroponically (CRC, 2014).
Kumar, B. et al. Rapid method of screening for drought stress tolerance in maize (Zea mays L). Ind. J. Genet. Plant. Breed. 80, 16–25 (2020).
Julkowska, M. M. & Testerink, C. Tuning plant signaling and growth to survive salt. Trends Plant. Sci. 20, 586–594 (2015).
Bruce, W. B., Edmeades, G. O. & Barker, T. C. Molecular and physiological approaches to maize improvement for drought tolerance. J. Exp. Bot. 53, 13–25 (2002).
Li, P. et al. A genetic relationship between nitrogen use efficiency and seedling root traits in maize as revealed by QTL analysis. J. Exp. Bot. 66, 3175–3188 (2015).
Hu, D. et al. Maize (Zea mays L.) responses to salt stress in terms of root anatomy, respiration and antioxidative enzyme activity. BMC Plant. Biol. 22, 602 (2022).
Yu, P., Hochholdinger, F. & Li, C. Root-type-specific plasticity in response to localized high nitrate supply in maize (Zea mays). Ann. Bot. 116, 751–762 (2015).
Wajhat-Un-Nisa, Sandhu, S., Ranjan, R. & Sharda, R. Root plasticity: an effective selection technique for identification of drought tolerant maize (Zea mays L.) inbred lines. Sci. Rep. 13, 5501 (2023).
Cai, Q. et al. Morphological plasticity of root growth under mild water stress increases water use efficiency without reducing yield in maize. Biogeosciences 14, 3851–3858 (2017).
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing. Preprint at, 2024).
Lê, S., Josse, J., Husson, F. & FactoMineR An R package for multivariate analysis. J. Stat. Softw. 25, 1–18 (2008).
Xavier & Perrier Albert flori & François Bonnot. Genetic Diversity of Cultivated Tropical Plants. (CRC, doi:https://doi.org/10.1201/9781482280043. (2003).
Zobel, R. W., Wright, M. J. & Gauch, H. G. Statistical analysis of a yield trial. Agron. J. 80, 388–393 (1988).
Ru, C., Hu, X., Chen, D., Wang, W. & Song, T. Heat and drought priming induce tolerance to subsequent heat and drought stress by regulating leaf photosynthesis, root morphology, and antioxidant defense in maize seedlings. Environ. Exp. Bot. 202, 105010 (2022).
Canales, F. J., Nagel, K. A., Müller, C., Rispail, N. & Prats, E. Deciphering root architectural traits involved to Cope with water deficit in oat. Front. Plant. Sci. 10, 485835 (2019).
Muscolo, A., Sidari, M., Anastasi, U., Santonoceto, C. & Maggio, A. Effect of PEG-induced drought stress on seed germination of four lentil genotypes. J. Plant. Interact. 9, 354–363 (2014).
Dar, I. A., Sofi, P. A., Dar, Z. A., luddin, K. & Lone, A. A. Screening of maize genotypes for drought tolerance related trait variability. Int. J. Curr. Microbiol. Appl. Sci. 7, 668–682 (2018).
Mohanapriya, B. et al. Root phenological and physiological response of maize (Zea mays L.) for adaptation under drought stress at vegetative and reproductive stages. Res. Crops. 24, 287–294 (2023).
Machado, J. et al. Drought and nitrogen stress effects and tolerance mechanisms in tomato: a review. Plant. Nutr. Food Secur. Era Clim. Change. 315–359. https://doi.org/10.1016/B978-0-12-822916-3.00014-7 (2022).
Franco, J. A., Bañón, S., Vicente, M. J., Miralles, J. & Martínez-Sánchez, J. J. Review article:root development in horticultural plants grown under abiotic stress conditions – a review. J. Hortic. Sci. Biotechnol. 86, 543–556 (2011).
Bernstein, N. in & Uzi Kafkafi. Plant Roots. (eds Waisel, Y., Eshel, A., Beeckman, T. & Kafkafi, U.) (CRC, 2002). https://doi.org/10.1201/9780203909423
Li, W. et al. The effects of simulated nitrogen deposition on plant root traits: A meta-analysis. Soil. Biol. Biochem. 82, 112–118 (2015).
Ahmed, M., Kheir, M. S., Zeeshan Mehmood, A., Ahmad, M., Hasanuzzaman, M. & S. & Changes in germination and seedling traits of Sesame under simulated drought. Phyton (B Aires). 91, 713–726 (2022).
Ahmed, H. G. M. D. et al. Conferring of drought tolerance in wheat (Triticum aestivum L.) genotypes using seedling indices. Front. Plant. Sci. 13, 961049 (2022).
Shah, S. H., Houborg, R. & McCabe, M. F. Response of chlorophyll, carotenoid and spad-502 measurement to salinity and nutrient stress in wheat (Triticum aestivum L). Agron. 2017. 7, 61 (2017).
Alipour, H., Abdi, H., Rahimi, Y. & Bihamta, M. R. Dissection of the genetic basis of genotype-by-environment interactions for grain yield and main agronomic traits in Iranian bread wheat landraces and cultivars. Sci. Rep. 11, 17742 (2021).
Meena, V. K., Sharma, R. K., Chand, S., Kumar, S. & Choudhary, K. Comparative study of stability models for identifying stable spring wheat genotypes in diverse conditions. Discover Agric. 3, 16 (2025).
Kurt, D. Adaptability and stability models in promising genotype selection for hybrid breeding of sun cured tobacco. South. Afr. J. Bot. 154, 190–202 (2023).
Feng, W., Lindner, H., Robbins, N. E. & Dinneny, J. R. Growing out of stress: the role of cell- and organ-scale growth control in plant water-stress responses. Plant. Cell. 28, 1769–1782 (2016).
Comas, L. H., Becker, S. R., Cruz, V. M. V., Byrne, P. F. & Dierig, D. A. Root traits contributing to plant productivity under drought. Front Plant. Sci 4, (2013).
Ramezanifar, H., Yazdanpanah, N., Yazd, G. H., Tavousi, H., Mahmoodabadi, M. & M. & Spinach growth regulation due to interactive salinity, water, and nitrogen stresses. J. Plant. Growth Regul. 41, 1654–1671 (2022).
El Soda, M., Nadakuduti, S. S., Pillen, K. & Uptmoor, R. Stability parameter and genotype mean estimates for drought stress effects on root and shoot growth of wild barley pre-introgression lines. Mol. Breed. 26, 583–593 (2010).
Gaudin, A. C. M., Mcclymont, S. A., Holmes, B. M., Lyons, E. & Raizada, M. N. Novel Temporal, fine-scale and growth variation phenotypes in roots of adult-stage maize (Zea mays L.) in response to low nitrogen stress. Plant. Cell. Environ. 34, 2122–2137 (2011).
Shi, Q., Hua, J., Creech, D. & Yin, Y. Biomass Estimation and carbon storage of taxodium hybrid Zhongshanshan plantations in the Yangtze river basin. Forests 2022. 13, 1725 (2022).
Sandhu, N. et al. Rice root architectural plasticity traits and genetic regions for adaptability to variable cultivation and stress conditions. Plant. Physiol. 171, 2562–2576 (2016).
Lopez-Cruz, M. et al. Leveraging data from the Genomes-to-Fields Initiative to investigate genotype-by-environment interactions in maize in North America. Nat. Comm. 14 (1), 1–14 (2023).
Cvejić, S. et al. Innovative approaches in the breeding of climate-resilient crops. Clim. Change Agriculture: Perspect. Sustain. Resil. 111–156. https://doi.org/10.1002/9781119789789.CH6 (2022).
Woods, P., Lehner, K. R., Hein, K., Mullen, J. L. & McKay, J. K. Root pulling force across drought in maize reveals genotype by environment interactions and candidate genes. Front. Plant. Sci. 13, 883209 (2022).
Karlova, R., Boer, D., Hayes, S. & Testerink, C. Root plasticity under abiotic stress. Plant. Physiol. 187, 1057–1070 (2021).
Wajhat-Un-Nisa, Sandhu, S., Ranjan, R. & Sharda, R. Root plasticity: an effective selection technique for identification of drought tolerant maize (Zea mays L.) inbred lines. Sci. Rep. 13 (1), 1–13 (2023).
Schneider, H. M. et al. Spatio-temporal variation in water uptake in seminal and nodal root systems of barley plants grown in soil. Front. Plant. Sci. 11, 543281 (2020).
Rogers, E. D. & Benfey, P. N. Regulation of plant root system architecture: implications for crop advancement. Curr. Opin. Biotechnol. 32, 93–98 (2015).
Ge, Y., Fang, X., Liu, W., Sheng, L. & Xu, L. Adventitious lateral rooting: the plasticity of root system architecture. Physiol. Plant. 165, 39–43 (2019).
Hannan, A., Hassan, L., Hoque, M. N., Tahjib-Ul-Arif, M. & Robin, A. H. K. Increasing new root length reflects survival mechanism of rice (Oryza sativa L.) genotypes under peg-induced osmotic stress. Plant. Breed. Biotechnol. 8, 46–57 (2020).
Hammer, G. L. et al. Can changes in canopy and/or root system architecture explain historical maize yield trends in the U.S. Corn belt? Crop Sci. 49, 299–312 (2009).
Meng, D. & Fricke, W. Changes in root hydraulic conductivity facilitate the overall hydraulic response of rice (Oryza sativa L.) cultivars to salt and osmotic stress. Plant. Physiol. Biochem. 113, 64–77 (2017).
Castillo, S. R., Dowdy, R. H., Bradford, J. M. & Larson, W. E. Effects of applied mechanical stress on plant growth and nutrient uptake. Agron. J. 74, 526–530 (1982).
Nagar, C. K. et al. Nitrogen stress induced changes in root system architecture (RSA) in diverse wheat (T. aestivum L.) genotypes at seedling stage. J Cereal Res 10, (2018).
Lozano, Y. M., Aguilar-Trigueros, C. A., Flaig, I. C. & Rillig, M. C. Root trait responses to drought are more heterogeneous than leaf trait responses. Funct. Ecol. 34, 2224–2235 (2020).
Lovelli, S., Perniola, M., Tommaso, T., Di, Bochicchio, R. & Amato, M. Specific root length and diameter of hydroponically-grown tomato plants under salinity. J. Agron. 11, 101–106 (2012).
Tylová, E., Pecková, E., Blascheová, Z. & Soukup, A. Casparian bands and Suberin lamellae in exodermis of lateral roots: an important trait of roots system response to abiotic stress factors. Ann. Bot. 120, 71–85 (2017).
Acknowledgements
The first Author thanks the Indian Council of Agricultural Research and Post Graduate School, Indian Agricultural Research Institute, New Delhi, for a Junior Research Fellowship during his graduation programme.
Funding
MGM and VC are grateful to the “Network Project on Computational Biology and Agricultural Bioinformatics” (Agril.Edn.14(44)/2014-A&P) for financial support. The funding agencies had no role in designing the study, data collection and analyses, the decision to publish, or the preparation of the manuscript.
Author information
Authors and Affiliations
Contributions
GMK: Experimentation, Methodology, Formal analysis, Visualization, Writing – original draft. MGM: Conceptualization, Methodology, Formal analysis, Visualization, Writing – original draft, Writing – review & editing, Funding acquisition, Project administration, Supervision. SKJ: Conceptualization, Writing – review & editing. Rakesh Pandey: Methodology, Writing – review & editing. PV: Experimentation. HCL: Conceptualization, Writing – review & editing. VC: Conceptualization, Writing – review & editing, Funding acquisition, Supervision.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Keerthi, G.M., Mallikarjuna, M.G., Jha, S.K. et al. Unravelling root system architecture plasticity in response to abiotic stresses in maize. Sci Rep 15, 20433 (2025). https://doi.org/10.1038/s41598-025-04123-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-025-04123-z
