The molecular mechanisms by which multiple metabolic pathways regulate soybean responses to Mn stress

Abstract

Background

Manganese (Mn) is an essential micronutrient for plants. Soybean, a globally important food and feed crop, shows high sensitivity to Mn stress. This study applied an integrated multi-omics approach to elucidate the mechanisms underlying soybean responses to Mn toxicity stress.

Result

The results showed that Mn toxicity significantly inhibited soybean growth, with a 35.68% decrease in leaf glutathione (GSH) content and a 53.54% decrease in hydrogen peroxide (H₂O₂) content, whereas phenylalanine ammonia-lyase (PAL) activity in roots increased by up to 83.15%. When soybean plants were treated with 100 µM Mn, it significantly affected phenotypic and physiological data and avoided lethality. Therefore, this concentration was selected as the Mn stress treatment concentration for subsequent ion content determination and omics studies. Under Mn toxicity stress (100 µM Mn), the Mn content in the roots and leaves of soybean plants increased by 7.4 times and 7.7 times respectively. Transcriptomic and metabolomic analyses were performed on soybean plants treated with 5 µM and 100 µM Mn. A total of 5045 differentially expressed genes (DEGs), including 3373 upregulated and 1672 downregulated genes, and 3259 differentially accumulated metabolites (DAMs), including 1505 upregulated and 1754 downregulated metabolites, were identified in leaves and roots. Integrated transcriptional and metabolomic analyses indicated that flavonoid biosynthesis and phenylalanine metabolism in leaves, and glycolysis/gluconeogenesis in roots, represent the core response pathways to Mn toxicity.

Conclusion

Coordinated regulation of genes and metabolites establishes a buffering system that mitigates Mn-induced toxicity. These findings provide a theoretical basis for breeding Mn-tolerant soybean cultivars and for strategies aimed at remediation of Mn-contaminated environments.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12870-026-09721-1.

Introduction

Due to the continuous severe threat to global agricultural production from non-biological stressors, crop yields and quality have significantly declined [1]. Among various non-biological stress factors, metal ion pollution has emerged as a key limiting category due to its persistence, toxicity, and extensive agricultural impact [2, 3].

Mn is an essential element for plant growth and metabolism, serving as a metal cofactor for approximately 6% of all known metalloenzymes [4]. Although many enzymes can be activated by Mn²⁺, only a limited number exhibit absolute and irreplaceable catalytic dependence on this metal [4]. In plants, the most extensively studied Mn-dependent enzymes include Mn superoxide dismutase (Mn-SOD), oxalate oxidase (OxOx), and the Mn cluster of photosystem II (PSII) [5]. The Mn cluster catalyzes water splitting, supplies electrons to drive photosynthesis, and supports ATP synthesis, Rubisco activity, and the biosynthesis of fatty acids, amino acids, and proteins [5]. Mn also participates in chlorophyll biosynthesis, nitrate assimilation, and hormone activation [6]. However, excessive Mn accumulation is detrimental to plant growth [7]. Typical symptoms of Mn toxicity include leaf chlorosis, leaf spotting, and growth retardation [7]. To cope with Mn toxicity, plants have developed multiple adaptive strategies, including activation of the antioxidant system, regulation of Mn uptake and homeostasis, and sequestration of Mn into subcellular compartments such as vacuoles, the endoplasmic reticulum, Golgi bodies, and cell walls [8]. Mn toxicity induces oxidative stress in rice (Oryza sativa), leading to alterations in chloroplast and mitochondrial ultrastructure, inhibition of photosynthesis, and downregulation of numerous proteins associated with photosynthesis and carbohydrate metabolism [7].

Under Mn stress, plants can adjust the activities of key glycolytic enzymes, such as hexokinase (HK) and pyruvate kinase (PK), to accelerate glucose catabolism and compensate for ATP deficiency caused by impaired photosynthesis [9]. At the same time, gluconeogenesis may be activated to convert non-carbohydrate precursors, including specific amino acids or glycerol, into glucose, thereby maintaining intracellular sugar homeostasis and supplying emergency energy and reducing equivalents, such as NADPH, to sustain antioxidant defenses under stress conditions [10]. Mn toxicity has a significant adverse effect on the amino acid metabolic network, particularly on the pathways of alanine, aspartate, and glutamate metabolism [11]. These amino acids are central to protein synthesis and serve as major carriers for nitrogen transport and redistribution [12]. Under Mn-induced metabolic imbalance, the coordination of carbon and nitrogen fluxes within cells is disrupted [13]. Changes in glutamate metabolism, which acts as a hub for multiple transamination reactions, directly influence the synthesis of other amino acids [14]. Its metabolic allocation shifts from predominant use in glutamine synthesis for nitrogen assimilation to increased participation in γ-aminobutyric acid production and transamination reactions with alanine and aspartate [14]. For example, rice roots enhance glutamine synthetase activity to accelerate assimilation of excess ammonium, thereby preventing secondary toxicity caused by ammonia accumulation [14]. In parallel, alanine and aspartate metabolism undergoes significant reprogramming, which is considered a dynamic mechanism for nitrogen storage and redistribution [14]. This mechanism prioritizes limited nitrogen resources for the synthesis of glutathione (GSH), disease-related proteins, and other defense components, thereby enhancing stress tolerance [14].

Upregulation of the phenylalanine metabolic pathway further promotes the synthesis of antioxidants, including flavonoids, and directly strengthens oxidative stress defenses [15]. Phenylalanine acts as a key metabolic node connecting primary metabolism with extensive secondary metabolic networks [16]. Mn stress, as a strong biotic and abiotic signal, induces increased expression and activity of enzymes such as phenylalanine ammonia-lyase (PAL), leading to activation of the phenylpropanoid pathway [17]. Activation of this pathway promotes lignin biosynthesis and cross-linking within the cell wall, enhancing mechanical strength and restricting the diffusion of Mn ions and other harmful substances between cells [18]. It also supplies precursors for a wide range of downstream defense metabolites [18]. One major branch of the phenylpropanoid pathway is flavonoid biosynthesis [19]. Under Mn stress, expression of chalcone synthase (CHS) and flavanone-3-hydroxylase is significantly induced in rice, leading to the accumulation of quercetin and other flavonols that enhance antioxidant capacity [20]. This response results in the accumulation of flavonoids, flavonols, and proanthocyanidins [19]. These secondary metabolites directly scavenge excessive reactive oxygen species (ROS), protecting membrane integrity and photosynthetic structures from oxidative damage [21]. They may also function as signaling molecules that participate in the regulation of stress-responsive gene networks, thereby contributing to overall stress resistance [21].

Soybean (Glycine max) is one of the most important foods and feed crops worldwide [19]. It represents a major source of high-quality plant protein and edible oil and is a key economic crop in China [22]. However, high concentration Mn stress (such as 50 mg/L) significantly reduces the maximum photochemical efficiency, photosynthetic rate and stomatal conductance of soybean leaves [23]. At the same time, it increases non-photochemical quenching and intercellular CO₂ concentration, thereby leading to energy loss and growth inhibition [23]. Hence, soybean is highly sensitive to Mn stress, for instance, the concentration threshold at which soybean plants exhibit typical manganese toxicity symptoms is usually within the range of 50–200 µM (approximately 2.7–11 mg/L) [25]. When the available Mn concentration in soil exceeds 80–100 mg/kg, which represents the critical toxicity threshold, characteristic Mn spots appear on leaves, photosynthetic efficiency declines, and antioxidant enzyme activities in roots, including superoxide dismutase and peroxidase, increase by approximately 2–4 fold [25]. Concurrent accumulation of osmotic adjustment substances, such as proline and malondialdehyde, ultimately results in reduced pod number per plant, decreased hundred-grain weight, and yield losses of up to 25–35% [24]. In acidic red soil regions of southern China, Mn toxicity has become a major abiotic factor limiting stable and high soybean yield [18]. Despite its agronomic importance, the molecular regulatory mechanisms underlying soybean responses to Mn stress remain insufficiently understood.

To date, no integrated transcriptomic–metabolomic analysis has systematically characterized Mn toxicity responses in soybean [8, 25]. Here, a scientific hypothesis is proposed: Mn stress may have activated the tissue-specific coordination of secondary metabolic pathways in roots and leaves, particularly the biosynthesis of phenylpropanoids and flavonoids. Moreover, this coordinated response may enhance the clearance of ROS and strengthen the cell wall barrier, thereby playing a central role in alleviating Mn-induced stress. To clarify the molecular regulatory mechanisms of soybean under Mn stress, this study integrates metabolomic, transcriptomic, and physiological and biochemical analyses to systematically examine dynamic changes in key metabolic pathways [5]. The objective is to elucidate how coordinated metabolite interactions and regulatory networks mitigate Mn toxicity and explore simultaneous root-leaf metabolite-gene module mapping under Mn toxicity in soybean [10]. These results provide a theoretical basis and technical support for the breeding of Mn-resistant soybean cultivars and for ecological restoration of Mn-contaminated agricultural soils [21].

Materials and methods

Plant materials

The experimental material Yuchun03-3 (YC03-3), a soybean cultivar, was developed by South China Agricultural University. Plants were cultivated at the experimental base of the Institute of Agricultural Biotechnology, Guangdong Ocean University (longitude 110.300995, latitude 21.147989). Uniform, fully developed soybean seeds without visible cracks were selected, germinated, and grown in sand. Seedlings were subsequently transferred to 4 L hydroponic containers. Each container is planted with 6 seedlings, and each container is defined as an independent experimental unit (i.e., a biological replicate). The composition of the nutrient solution followed a previously reported formulation [26] (Table S1). All reagents used were of analytical grade (Solarbio Corporation, Beijing, China). MnSO₄·H₂O (Guangzhou Reagent, Guangzhou, China) was used to establish different Mn treatment concentrations (5, 50, 100, 150, and 200 µM). For each Mn treatment concentration, three biological replicates were set up (i.e., four separate hydroponic containers). The growth chamber was maintained at 25–30℃ during the day and 18–22℃ at night, with a 12 h photoperiod and a light intensity of 200 µmol·m⁻²·s⁻¹. The nutrient solution was renewed every 5 days. The pH value of the culture solution was regulated to 5.0 by 1 M potassium hydroxide or sulfuric acid every 2 days (Solarbio corporation, China). After 15 days of treatment, roots and the second fully expanded leaves from the apex were harvested for subsequent analyses.

Measurement of physiological indicators

To determine reduced GSH content, PAL activity, total antioxidant capacity, and H₂O₂ content in soybean roots and leaves under different Mn treatments, 0.1 g of fresh tissue was rapidly frozen in liquid nitrogen and ground into a fine powder [27]. Measurements were performed using commercial assay kits (Boxbio Co., Ltd., Beijing, China) according to the manufacturer’s instructions. Each treatment included three biological replicates. Root morphology was scanned using a ScanMaker i800 Plus image scanner (Shanghai Zhongjing, China). The images were analyzed with LA-S plant root analysis instrument (Wanshen, Zhejiang, China) to obtain root surface area, total root length, and average root diameter [25].

Determination of ion contents

After 15 days of Mn treatment, soybean roots and leaves were collected for ion content analysis. Approximately 0.1 g of each sample was transferred into a 50 mL microwave digestion vessel (Anton Paar, Graz, Austria), and a small amount of ultrapure water was added. Subsequently, 6 mL of 65% nitric acid and 0.5 mL of 30% H₂O₂ (Guangzhou Reagent Company, Guangzhou, China) were added. Samples were digested using a microwave digestion system (Anton Paar Multiwave PRO 41HVT56, Graz, Austria). After digestion, 1 mL of 65% nitric acid was added, and the volume was adjusted to 50 mL with ultrapure water using a glass cuvette (Beibow, Beijing, China). Mn, magnesium (Mg), iron (Fe), and sodium (Na) contents were determined using an inductively coupled plasma mass spectrometer (ICP-MS; 7500cx, Agilent, Santa Clara, CA, USA) according to a previously reported method [27]. To ensure the accuracy of the measurement results, a multi-element standard solution was used to establish an external standard curve for quantitative analysis, and a reagent blank was set up for background correction.

cDNA library preparation and transcriptome analysis

Fresh root and leaf samples from soybean plants treated with 5 µM and 100 µM Mn for 15 days were used for transcriptome analysis, with three biological replicates per treatment. RNA was isolated from the roots and leaves of soybean plants using the ultra-pure RNA kit (MolPure^® Plant RNA Kit, Yeasen, China). The cDNA library was constructed using Takara (Maebashi, Japan) reagents in accordance with the manufacturer’s instructions. The cDNA libraries were constructed following previously reported procedures and sequenced on an Illumina platform in paired-end 150 bp (PE150) mode [28]. Raw sequencing data were filtered to obtain high-quality clean reads. Clean reads were aligned to the soybean reference genome (Williams 82.a4.v1) using HISAT2 v2.0.5. Gene-level read counts, representing original expression levels, were calculated using HTSeq v0.9.1. Differential expression analysis was conducted using DESeq, with differentially expressed genes (DEGs) defined by |log2FoldChange| > 1 and P < 0.05. After the gene expression levels were standardized, the Benjamini–Hochberg method was used to correct for the false discovery rate (FDR). The screening criteria for DEGs were set as a corrected P value (FDR) < 0.05 and |log2Fold Change| ≥ 1. Gene Ontology enrichment analysis was performed using topGO, and kyoto encyclopedia of genes and genomes (KEGG) pathway enrichment analysis was conducted with clusterProfiler v3.4.4. The data were supplemented with the integrated gene expression database, with entry number PRJNA1232013 (https://www.ncbi.nlm.nih.gov/sra/PRJNA1232013).

Metabolomics analysis

Fresh root and leaf samples from soybean plants treated with 5 µM and 100 µM Mn for 15 days were subjected to metabolomic analysis, with four biological replicates per treatment. Sample preparation and data processing were performed according to established protocols by Suzhou Panomic Biomedical Technology Co., Ltd. Metabolite detection was carried out using a Thermo Vanquish ultra-high-performance liquid chromatography system coupled with a Thermo Orbitrap Exploris 120 mass spectrometer (Thermo Fisher Scientific, USA). Principal component analysis and orthogonal partial least squares discriminant analysis were applied for data dimensionality reduction and regression analysis. Metabolite identification was based on spectral databases, including HMDB, MassBank, LipidMaps, mzCloud, KEGG, and Nomi Metabolism [28]. Metabolites with secondary identification confidence were retained, and those meeting the criteria of P < 0.05 and variable importance in projection > 1 were defined as differentially accumulated metabolites (DAMs). Data analysis was conducted using the Nomibio Cloud platform (https://www.biodeep.cn/home). Integrated analyses combining metabolomic and transcriptomic datasets were subsequently performed.

Quantitative real-time PCR analysis

Total RNA was extracted from soybean roots and leaves using an ultrapure RNA kit (CWBIO, China). After removal of genomic DNA, complementary DNA was synthesized using a reverse transcription kit (Takara, Maebashi, Japan). Quantitative real-time PCR (qRT‒PCR) was performed on a Bio-Rad real-time PCR system (Bio-Rad, Hercules, CA, USA). The reagent of qRT‒PCR used was Hieff UNICON^® Universal Blue qPCR SYBR Green Master Mix (Yeasen, Shanghai, China). The primers for qRT‒PCR were obtained from Guangzhou IGE Biotechnology Co., Ltd. (Guangzhou, China). The cDNA sample was diluted 3 times, and qRT‒PCR was performed using a 20 µL reaction system: H₂O: 8 µL; Mix: 10 µL; Primer F: 0.5 µL; Primer R: 0.5 µL; cDNA: 1 µL. The reaction programs were setting up as follows: 95 ℃ for 3 min, 95 ℃ for 10 s, 60 ℃ for 30 s, and for 40 cycles. Finally, the data were calculated for relative expression using the 2^−ΔΔCT method. The soybean elongation factor gene GmEF (Glyma.17G186600) was used as the internal reference, as reported previously [27].

Data analysis

Statistical analyses were performed using IBM SPSS v26.0. Before the analysis, normality (Shapiro-Wilk test) and homogeneity of variance (Levene test) were conducted for all the data [29]. The mean of the plant measurement values in each container was taken as the representative value of that biological replicate for subsequent comparisons. An independent-samples t-test was used to assess significant differences between two Mn treatment concentrations, with P < 0.05 indicating significance and P < 0.01 indicating high significance. One-way analysis of variance followed by the Waller–Duncan test was applied to evaluate differences among more than two treatment concentrations. Different lowercase letters denote statistically significant differences at P < 0.05.

Results

Mn stress inhibits soybean growth

Compared with the control, the Mn concentration (5 µM), root and leaf growth of soybean plants was significantly inhibited under elevated Mn concentrations (50, 100, 150, and 200 µM) (Fig. 1). Specifically, total root length decreased by 11.39%, 13.08%, 22.54%, and 46.25%, respectively (Figure S1A). Root surface area decreased by 15.03%, 25.78%, 34.03%, and 71.02%, respectively (Figure S1B). In contrast, average root diameter increased by 5.47%, 11.63%, 10.83%, and 14.71%, respectively (Figure S1C).

Fig. 1.

Effects of different concentrations of Mn on the morphology of soybean plants. A-E: from left to right: 5, 50, 100, 150 and 200 µM (bar = 8 cm)

Results of physiological index detection in soybean leaves and roots

Using the 5 µM Mn treatment as the control, physiological indices were measured after 15 days under Mn concentrations of 50, 100, 150, and 200 µM (Figs. 2A–H). Reduced GSH content in leaves decreased by 19.93%, 12.67%, 23.83%, and 35.68%, respectively (Fig. 2A), whereas in roots it increased by 16.37%, 62.02%, 7.37%, and 15.31%, respectively (Fig. 2B). PAL activity in leaves increased by 1.31%, 21.98%, and 0.17% under 50, 150, and 200 µM Mn, respectively, but decreased by 13.7% under 100 µM Mn (Fig. 2C). In roots, PAL activity increased by 39.74%, 83.15%, 61.75%, and 63.01%, respectively (Fig. 2D). Total antioxidant capacity(T-AOC) in leaves increased by 12.82%, 27.88%, 22.09%, and 26.74% (Fig. 2E), whereas in roots it increased by 24.07%, 32.71%, 28.49%, and 21.91% (Fig. 2F). H₂O₂ content in leaves decreased by 38.56%, 37.77%, 42.46%, and 53.54% (Fig. 2G), whereas in roots it increased by 7.25%, 54.81%, 16.54%, and 20.37% (Fig. 2H). When soybean plants were treated with 100 µM Mn, it significantly affected phenotypic and physiological data and avoided lethality. Therefore, this concentration was selected as the Mn stress treatment concentration for subsequent ion content determination and omics studies. Under Mn toxicity stress (100 µM Mn), the Mn content in the roots and leaves of soybean plants increased by 7.4 times and 7.7 times respectively (Figure S2A).

Fig. 2.

Results of physiological response indicators of soybean leaves and roots under Mn stress. Four physiological indicators of fresh samples of soybean leaves and roots treated with 5, 50, 100, 150, and 200 µM Mn for 15 days. Glutathione (GSH) content of (A) leaves and (B) roots; Phenylalanine ammonia-lyase (PAL) activity of (C) leaves and (D) roots; Total antioxidant capacity (T-AOC) of (E) leaves and (F) roots; H₂O₂ content of leaves (G) and (H) roots

Transcriptome result analysis

After removing low-quality data, an average of 51,580,955 high-quality reads were obtained from the samples of the control group (5 µM Mn) and the treatment group (100 µM Mn) of soybean roots and leaves (Table S2). The number of high-quality reads in all tested samples ranged from 40,687,474 to 63,193,498 (Table S2). The average percentage of high-quality reads was 98.8% (Table S2).

To further identify core genes responding to Mn stress in soybean roots and leaves, whole-genome transcriptome sequencing was performed using tissues exposed to control Mn concentration (5 µM) and high Mn concentration (100 µM). The analysis identified 5045 DEGs in leaves, including 3373 up-regulated and 1672 down-regulated genes, and 3259 DEGs in roots, including 1505 up-regulated and 1754 down-regulated genes (Figs. 3 and S3). PCA and cluster heatmap analysis were conducted on the DEGs in soybean plants. The results showed that the response of soybean plants to Mn stress was highly tissue-specific (Figures S4 and S5).

Fig. 3.

Changes in the transcriptome of soybean induced by Mn toxicity. A Distribution pattern of DEGs; B Venn diagram analysis of upregulated and downregulated genes in leaves (5 µM VS 100 µM Mn) and roots (5 µM VS 100 µM Mn)

KEGG enrichment analysis was conducted to examine the effects of Mn toxicity on gene expression, and 30 significantly enriched pathways (P < 0.05) were selected for further analysis. In leaves, enriched pathways were mainly assigned to environmental information processing, metabolism, and biological systems (Fig. 4A). Representative pathways included phenylpropanoid biosynthesis, flavonoid biosynthesis, α-linolenic acid metabolism, isoflavone biosynthesis, and the mitogen-activated protein kinase signaling pathway in plants (Fig. 4A). In roots, enriched pathways were primarily associated with environmental information processing and metabolism (Fig. 4B). Major pathways included nitrogen metabolism, glycolysis/gluconeogenesis, phenylpropanoid biosynthesis, cysteine and methionine metabolism, and cutin, suberin, and wax biosynthesis (Fig. 4B).

Fig. 4.

KEGG enrichment analysis diagram of DEGs in soybean under Mn stress. A Leaves; B Roots

Metabolomic result analysis

To elucidate metabolic regulation under Mn stress, metabolomic profiling was conducted on soybean leaves and roots treated with control Mn concentration (5 µM) and high Mn concentration (100 µM). In leaves, 164 DAMs were identified, including 101 up-regulated and 63 down-regulated metabolites. In roots, 136 DAMs were detected, comprising 46 up-regulated and 90 down-regulated metabolites (Figure S6).

The DAMs in the leaves and roots were classified separately. In the leaves, the top five categories of DAMs were: amino acids and their metabolites, carbohydrates and their metabolites, lipid and fatty acid-related metabolites, nucleotides and their metabolites, and organic acids and their derivatives (Table S3). In the roots, the top four categories of DAMs were: lipid and fatty acid-related metabolites, nucleotides and their metabolites, amino acids and their metabolites, and carbohydrates and their metabolites (Table S3).

KEGG pathway enrichment analysis identified the top 20 pathways enriched by DAMs in soybean leaves and roots (Fig. 6). In leaves, these metabolites were mainly enriched in flavonoid biosynthesis, oxidative phosphorylation, the citrate cycle (TCA cycle), ATP-binding cassette transporters, and alanine, aspartate, and glutamate metabolism (Fig. 5A). In roots, enriched pathways included linoleic acid metabolism, glycolysis/gluconeogenesis, purine metabolism, α-linolenic acid metabolism, and pyrimidine metabolism (Fig. 5B).

Fig. 6.

Key regulatory pathways responding to Mn stress in soybean leaves and roots. Results of KEGG enrichment association after combined analysis of transcriptome and metabolome in (A) leaves and (B) roots. Different color areas represented the number of KEGG enriched pathways in different omics, and overlapping areas represented the number of enriched pathways in both omics. KEGG enrichment histogram of combined analysis of transcriptome and metabolome in (C) leaves and (D) roots

Fig. 5.

KEGG enrichment analysis of DAMs in soybean. A Leaves; B Roots. The horizontal axis represented the impact values of enrichment in different pathways, with higher impact values indicating a greater contribution of the detected differential metabolites in the pathway. The vertical axis represented the metabolic pathways. This number represented the corresponding metabolite quantity on the pathway

The PCA analysis results of the DAMs in soybean roots and leaves showed that the experimental data quality was high; and the preset groups (control vs. treatment) showed significant differences in the overall metabolic profiles (Figure S7). The OPLS-DA analysis results showed that Mn stress treatment caused significant changes in the metabolite profile of soybean plants (Figure S8). The separation between groups was good and the separation model was statistically significant, not caused by random noise or overfitting (Figure S8).

Cluster analysis of metabolite accumulation patterns under control and high Mn treatments showed that metabolites in roots and leaves could be grouped into two major clusters (Figure S9). In roots, metabolites in cluster I accumulated at lower levels under control Mn conditions than under high Mn stress, whereas metabolites in cluster II showed higher accumulation under control Mn conditions than under high Mn stress (Figure S9A). In leaves, metabolites in cluster I accumulated at higher levels under control Mn conditions than under high Mn stress, whereas metabolites in cluster II displayed the opposite pattern, with lower accumulation under control Mn conditions than under high Mn stress (Figure S9B).

Joint analysis of transcriptome and metabolome

To examine the associations between genes and metabolites under Mn stress, pathway enrichment analyses were conducted for DEGs and DAMs. A total of 14 pathways were significantly enriched by both datasets in soybean leaves and roots (Fig. 6A and 6B).

Based on the combined enrichment results, pathways containing both DEGs and DAMs were visualized for integrated transcriptome–metabolome analysis (Figures S8 and S9). In soybean leaves, three pathways were enriched under Mn toxicity stress, namely flavonoid biosynthesis, alanine, aspartate and glutamate metabolism, and phenylalanine metabolism (Tables S4 and S5). Among these, the biosynthesis of flavonoids and the phenylpropanoid biosynthesis pathway have undergone the most significant changes, while alanine, aspartate and glutamate metabolism showed relatively weaker enrichment (Tables S4-S6). In soybean roots, only one pathway was enriched, namely glycolysis/gluconeogenesis (Fig. 6C and 6D).

Accordingly, flavonoid biosynthesis and phenylalanine metabolism in leaves, and glycolysis/gluconeogenesis in roots, were selected as the core pathways for integrated analysis of transcriptomic and metabolomic responses to Mn stress (Figs. 7A–C and Tables S7-S8).

Fig. 7.

Analysis of key pathways in soybean response to Mn stress. A: Phenylalanine metabolic pathway in leaves; B: Flavonoid biosynthesis pathway in leaves; C: Changes in genes and metabolites in the glycolysis/glyoxylate pathway in roots. Different colors of the rectangles in the pathways represented the LOG2FC values of corresponding metabolites, and different colors of the squares represented the LOG2FC values of corresponding genes

Within the phenylalanine metabolic pathway in leaves under Mn stress, 6 DAMs and 12 DEGs were detected, including 11 upregulated genes and 1 downregulated gene (Fig. 7A). At the metabolite level, four metabolites showed increased accumulation, including phenethylamine, phenylacetic acid, 3-(2-hydroxyphenyl) propanoate, and fumarate, whereas two metabolites showed decreased accumulation, namely D-phenylalanine and phenylethanol. As illustrated in the phenylalanine metabolic pathway (Fig. 7A), primary metabolism centered on L-phenylalanine was redirected toward secondary metabolism. L-phenylalanine was first deaminated to form phenylpyruvic acid, which could be decarboxylated to phenylaldehyde and subsequently reduced to phenylacetic acid. In a parallel branch, oxidation yielded trans-cinnamate, which served as a key precursor for phenylpropanoid metabolism. Cinnamic acid was further activated to form cinnamoyl-CoA, which could directly contribute to the synthesis of 3-(2-hydroxyphenyl) propionic acid and enter the tricarboxylic acid (TCA) cycle, linking secondary metabolism with energy metabolism. Cinnamoyl-CoA also acted as a central precursor for subsequent flavonoid and lignin biosynthesis (Fig. 7A).

Flavonoid biosynthesis represented one of the core pathways of plant secondary metabolism (Fig. 7B). In this pathway, cinnamoyl-CoA derived from phenylalanine metabolism was condensed with malonyl-CoA. CHS catalyzed the formation of chalcones, such as pinocembrin chalcone, which were subsequently isomerized by chalcone isomerase (CHI) to form naringenin, the core flavonoid skeleton. Naringenin then undergoes successive enzymatic reactions catalyzed by enzymes such as flavanone-3-hydroxylase, flavonoid 3′-hydroxylase (F3’H), flavonoid 3′,5′-hydroxylase, and dihydroflavonol 4-reductase, corresponding to genes including Glyma.06G039600, leading to the formation of simple flavonoids, flavonols, anthocyanin precursors, and dimeric flavonoids (Fig. 7B).

In roots, the glycolysis/gluconeogenesis pathway showed a complex regulatory pattern characterized by reduced metabolite accumulation and extensive transcriptional modulation (Fig. 7C). A total of four DAMs and 45 DEGs, including 29 upregulated and 16 downregulated genes, were identified. All detected metabolites, including salicin, glucose-1-phosphate, oxaloacetate, and glycerate-3-phosphate, exhibited decreased accumulation, suggesting altered metabolic flux through glycolysis intermediates. Examination of the glycolysis/gluconeogenesis-related network (Fig. 7C) indicated that monosaccharides such as glucose and fructose supported secondary metabolism through glycolysis, the pentose phosphate pathway (PPP), and the TCA cycle by supplying carbon skeletons and energy. During glycolysis, glucose-6-phosphate was converted through the actions of HK and phosphofructokinase to generate pyruvate, accompanied by ATP and NADH production. Glucose-6-phosphate was also diverted into the pentose phosphate pathway via glucose-6-phosphate dehydrogenase, generating ribulose-5-phosphate, which supplied precursors for aromatic amino acid synthesis (phenylalanine, tyrosine, etc.), thereby indirectly supporting phenylpropanoid metabolism. Pyruvate entered mitochondria to form acetyl coenzyme A and proceeds through the TCA cycle, producing intermediated such as citric acid and α-ketoglutarate. In parallel, metabolites such as 3-(2-hydroxyphenyl) propionic acid could enter the TCA cycle, establishing metabolic exchange between secondary metabolism and energy metabolism (Figs. 7C).

Phenylpropanoid metabolism (Fig. 7A) and flavonoid biosynthesis (Fig. 7B) shared common precursors such as cinnamoyl-CoA. Central carbon metabolism (Fig. 7C) supplied both carbon frameworks and energy in the form of ATP and NADPH for secondary metabolic processes. In turn, secondary metabolites, including flavonoids, enhanced plant adaptation to environmental stress through antioxidant activity and signal transduction, forming an integrated regulatory loop in which primary metabolism supported secondary metabolism and secondary metabolism contributes to cellular homeostasis.

Verification of transcriptome results

To validate the reliability of the transcriptome data, eight genes associated with flavonoid biosynthesis, phenylalanine metabolism, and glycolysis/gluconeogenesis were selected for quantitative real-time PCR analysis (Fig. 8; Table S9). Five genes were significantly upregulated in leaves, including CHS2 (Glyma.08G109500), CHS4 (Glyma.09G075200), CHI (Glyma.20G241700), F3’H (Glyma.06G202300), and PAL2 (Glyma.02G309300). Two genes were significantly upregulated in roots, namely pyruvate decarboxylase (PDC; Glyma.13G231700) and PK (Glyma.19G000700), whereas HK (Glyma.14G218800) showed downregulation without statistical significance. The expression patterns of these genes under Mn stress were generally consistent with the transcriptome results, supporting the reliability of the sequencing data.

Fig. 8.

qRT-PCR results of some DEGs in the key pathways. A: DEGs in flavonoid biosynthesis and phenylalanine metabolism pathways of leaves; B: DEGs in glycolysis/glyoxylate metabolism pathway of roots. CHS2: Chalcone synthase 2; CHS4: Chalcone synthase 4; CHI: Chalcone-flavone isomerase; PAL2: Phenylalanine ammonia-lyase 2; F3’H: Flavonoid 3’-hydroxylase; HK: Glucose kinase; PDC: Pyruvate decarboxylase; PK: Phosphoenolpyruvate kinase

Discussion

Effect of Mn stress on the oxidative defense system of soybean

Under Mn stress, Mn is absorbed by roots and transported to aerial tissues through the xylem, leading to its accumulation in leaves [30]. In acidic soils, leaf H₂O₂ content rises markedly, and the production rate of superoxide anions increases by three- to fivefold, thereby disrupting cellular redox homeostasis [31]. The results of this study indicated that under Mn toxicity, the H₂O₂ level in the leaves of soybean plants decreased significantly while it increased in the roots. Soybean plants may employ a differentiated redox strategy between roots and leaves to cope with Mn stress: the roots accumulate H₂O₂ to activate systemic defense signals, while the leaves efficiently remove H₂O₂ to prioritize the protection of photosynthetic structures. At the metabolic level, glycolysis undergoes significant reprogramming [32]. Root HK activity declines, whereas phosphoenolpyruvate carboxykinase activity increases, which slows glucose catabolism but enhances the conversion of pyruvate into the TCA cycle. This shift supplies carbon skeletons for the synthesis of antioxidant compounds, as indicated by the positive association between triosephosphate isomerase activity and GSH content [33]. The biosynthetic pathway of flavonoids might be activated, and the activity changes of PAL showed tissue-specific differences between roots and leaves: its activity significantly increased in roots, while the change pattern was more complex in leaves [34]. This difference may be related to the differences in the metabolic flow of phenylpropanoids in different tissues (such as roots, which focus on lignification/defensive compound synthesis, and leaves, which focus on flavonoid accumulation), ultimately leading to an increase in flavonoid production [34]. In parallel, activation of the flavonoid biosynthesis pathway is observed, accompanied by increased PAL activity, resulting in enhanced flavonoid production [34]. Glutamate family metabolism functions as a central regulatory hub [35]. Increased glutamate dehydrogenase activity supplies precursors for GSH synthesis, whereas aspartate transaminase activity shows a positive association with GSH levels [35]. Alanine content also increases, contributing to energy supply for antioxidant reactions [35]. GSH participates in a coordinated defense process characterized by synthesis, chelation, and sequestration [35]. Enhanced synthesis of GSH is accompanied by γ-glutamylcysteine synthetase–mediated chelation of Mn²⁺, whereas GSH peroxidase regulates H₂O₂ levels to maintain redox balance within a safe range [35]. Transcriptomic analysis shows upregulation of genes associated with these metabolic pathways, supporting activation of the antioxidant defense system.

Key pathways identified by integrated transcriptomic and metabolomic analysis in soybean leaves

Integrated omics analysis identified three common KEGG pathways at both transcriptional and metabolic levels in leaves under Mn stress, including flavonoid biosynthesis and alanine, aspartate, and glutamate metabolism.

Phenylalanine serves as a precursor for a wide range of phenolic compounds, including flavonoids [36]. Its key metabolic enzyme, PAL, catalyzes the deamination of L-phenylalanine to form trans-cinnamic acid, representing a major route by which plants respond to abiotic stress [36]. In this study, seven PAL genes were significantly upregulated in soybean leaves under Mn stress. Following conversion of L-phenylalanine to trans-cinnamic acid, downstream metabolites such as 3-(2-hydroxyphenyl) propanoate and fumaric acid accumulated. Combined analysis (Fig. 5) suggests that this metabolic adjustment contributes to detoxification through two mechanisms [37]. First, proton consumption during organic acid synthesis contributes to intracellular pH regulation [37]. Second, the esters and organic acids produced can form stable complexes with Mn ions, thereby reducing the bioavailability of free Mn²⁺ [37]. Such regulation can increase intracellular pH by 0.2–0.6 units, which effectively alleviates oxidative damage caused by Mn stress [38].

Mn stress also induces substantial accumulation of phenylacetic acid within the phenylalanine metabolic pathway [39]. Phenylacetic acid, a phenylalanine-derived compound, promotes tissue repair and growth regulation in plants [39]. Under regulation by amine oxidase AOC3, phenylethylamine content also increases concurrently [40]. Phenylacetic acid promotes cell elongation and tissue repair [41]. Under Mn stress in soybean, activation of this pathway likely facilitates repair of Mn-induced leaf damage and supports recovery of injured tissues [42]. Experimental evidence shows that phenylacetic acid treatment may increase the elongation rate of damaged cells by approximately 45%, thereby accelerating wound healing [42]. Flavonoids are key metabolites for plant tolerance to abiotic stress [42]. In this study, Mn stress induced upregulation of genes encoding CHS, CHI, and F3’H. These enzymes catalyze flavonoid biosynthesis, with CHS and CHI functioning as rate-limiting steps that convert 4-coumaroyl-CoA and malonyl-CoA into chalcone intermediates and subsequently generate flavanones such as hesperetin [43]. Flavonoids, including hesperetin, perform multiple defensive functions [43]. First, they may effectively scavenge excessive ROS, thereby reducing oxidative damage [43]. Under Mn stress, flavonoid accumulation may decrease intracellular ROS levels by approximately 50% [43]. Flavonoids containing hydroxyl groups may also form stable complexes with metal ions, further limiting free radical generation [43]. Second, flavonoids interact with membrane phospholipids through hydrogen bonding, which stabilizes membrane structure and integrity [43]. Experimental evidence indicates that flavonoid accumulation reduces membrane permeability by 25–30%, thereby limiting Mn ion entry into cells [43]. Activation of flavonoid metabolism is accompanied by enhancement of the PPP, which supplies additional NADPH for antioxidant reactions [44]. Under Mn stress, the flux of the pentose phosphate pathway may increase significantly, providing sufficient reducing power and carbon skeleton for flavonoid synthesis [44]. Collectively, these metabolic adjustments establish a multi-level defense system that regulates intracellular homeostasis, supports tissue repair, enhances antioxidant capacity, and preserves cellular structures, thereby mitigating the adverse effects of Mn toxicity on soybean growth and development [45].

Key pathways revealed by integrated transcriptomic and metabolomic analysis of soybean roots

In this study, genes associated with the glycolysis pathway in soybean roots showed an overall trend toward upregulated expression under Mn stress. Mn exposure may contribute to accelerate glycolytic flux, promoting consumption of glycolytic substrates such as glucose-1-phosphate and salicin, as well as intermediates associated with gluconeogenesis, including glycerate-3-phosphate [46]. This shift may support the synthesis of stress-responsive metabolites that contribute to Mn tolerance [46]. For example, isotope-labeling experiments have shown that pyruvate phosphate dikinase catalyzes the conversion of pyruvate to phosphoenolpyruvate, which participates in the conversion of L-alanine to sucrose [46]. This process increases soluble sugar content and enhances tolerance to abiotic stress. These sugars provide osmotic protection and can also participate in ROS scavenging as signaling molecules that regulate gene expression [47, 48].

Mn stress exerts an inhibitory effect on the TCA cycle, resulting in reduced oxidative phosphorylation [49]. In the present study, five pyruvate decarboxylase (PDC) genes were upregulated, and two glyceraldehyde-3-phosphate dehydrogenase (GAPDH) genes associated with NADH and H⁺ production and consumption also showed increased expression. These results suggest enhanced sugar anabolic metabolism in roots, which may contribute to regulate intracellular acidity through consumption of H⁺ generated during glycolysis, thereby alleviating Mn stress [50]. Within the glycolytic pathway, differential responses between gene expression and metabolite accumulation were observed. HK expression was downregulated, whereas PK and PDC expression was upregulated. Concurrently, the accumulation of metabolites at both ends of the pathway, including glucose-1-phosphate and salicin, decreased significantly, whereas the intermediate metabolite pyruvate showed no significant change. Together, these observations indicate that Mn stress disrupts the dynamic balance between glycolysis and sugar anabolic pathways in soybean roots, leading to altered steady-state regulation of this metabolic network [50].

Notably, five copies of the PDC gene were significantly upregulated. PDC catalyzes anaerobic fermentation of pyruvate and contributes to regulation of intracellular redox status by consuming NADH produced during glycolysis [51]. This process prevents cytoplasmic acidification and supports sustained glycolytic activity [51]. In parallel, two GAPDH genes associated with NADH and H⁺ turnover were also upregulated [51]. These findings suggest that under Mn stress, soybean roots may contribute to enhance sugar anabolic metabolism through coordinated regulation of PDC and GAPDH [52]. This coordination is likely to accelerate consumption of reducing equivalents and associated protons, stabilizes intracellular pH, alleviates acid accumulation induced by Mn toxicity, and enhances Mn tolerance [52].

During glycolysis, NAD⁺ is reduced to NADH [53]. In this study, except for Glyma.04G008300, expression of aldolase and phosphoglycerate kinase genes showed an increasing trend, suggesting a possible rise in NAD⁺ levels in root cells [53]. This increase may reflect a cellular requirement for NAD⁺ to maintain redox balance under stress conditions that exceed the demand for ATP production [54]. ATP can be generated through substrate-level phosphorylation during glycolysis, supporting basic cellular functions [54]. However, although phosphoglycerate kinase expression increased, accumulation of glycerate-3-phosphate declined, indicating that substrates for gluconeogenesis may gradually become limiting, whereas ATP accumulation may increase due to reduced downstream consumption [55]. Previous studies indicate that ATP generation is critical for metal ion detoxification processes in plants. Therefore, soybean roots may accumulate energy reserves to mitigate Mn toxicity, although the underlying mechanisms require further investigation [54].

Molecular mechanisms underlying soybean leaf and root responses to Mn stress

This study investigated the molecular responses of glycolysis, phenylpropanoid metabolism, and flavonoid biosynthesis pathways in soybean roots and leaves under Mn stress (Fig. 9). Under high Mn concentration (100 µM), roots appear to perceive stress signals first, leading to altered activities of key glycolytic enzymes. Glucose-1-phosphate is converted by HK to α-glucose-6-phosphate and enters downstream metabolic processes. During this transition, ATP production and intracellular H⁺ levels change, indicating that glycolysis serves a central role in energy regulation during Mn stress. Simultaneously, phenylpropanoid metabolism and flavonoid biosynthesis in leaves were markedly activated. L-phenylalanine was converted by PAL to cinnamoyl-CoA, which was subsequently transformed into intermediates such as p-coumaroyl-CoA. Coumaroyl-CoA then entered the flavonoid biosynthesis pathway, where CHS, CHI, and related enzymes catalyzed formation of flavonoids such as luteolin and quercetin, as well as other phenylpropanoid compounds including resveratrol. Alterations in phenylalanine-associated pathways affected flavonoid synthesis and accumulation of additional secondary metabolites in leaves. These metabolic adjustments were closely associated with hormonal regulation. Differential expression of genes related to auxin and gibberellin signaling was observed, suggesting that these hormones contribute to regulation of stress-responsive metabolic pathways [56]. Additionally, glycolytic activity in roots appears to be functionally associated with phenylpropanoid and flavonoid metabolism in leaves [57]. Although the precise signaling mechanisms remain unclear, it is proposed that metabolites or signaling molecules generated through root glycolysis are transported to leaves, where they influence metabolic reprogramming under Mn stress [58, 59]. Further investigation of signal exchange between roots and leaves under Mn toxicity will provide insight into systemic stress responses and support the development of Mn-tolerant soybean cultivars.

Fig. 9.

Physiological responses and molecular regulatory mechanisms of soybean plants under Mn stress

Conclusion

This study demonstrates that Mn toxicity significantly inhibits soybean seedling growth. Under high Mn concentrations, intracellular ROS increase, leading to activation of antioxidant responses in both roots and leaves and an elevation of total antioxidant capacity. Concurrent changes in H₂O₂ levels are associated with enhanced GSH synthesis. PAL activity also increases. As a rate-limiting enzyme of the phenylpropanoid pathway, PAL promotes conversion of intermediates derived from phenylalanine metabolism, including organic acids, into corresponding esters and salts. This process contributes to regulation of intracellular acid–base balance and strengthens soybean tolerance to Mn stress. By integrating multi-omics analyses, this study systematically clarifies the physiological basis of Mn-induced growth inhibition in soybean seedlings and the regulatory functions of key metabolic pathways. The results deepen understanding of the molecular framework linking Mn toxicity, oxidative stress, and metabolic pathway regulation. They also identify glycolysis/gluconeogenesis, phenylalanine metabolism, and flavonoid biosynthesis as central components of the regulatory network underlying soybean responses to Mn stress. These findings provide a theoretical foundation for developing Mn-tolerant soybean cultivars through molecular breeding strategies, such as targeted modulation of key enzyme genes including PAL and CHS, and through agronomic optimization, including adjustment of root-zone pH and carbon–nitrogen balance. The results further offer guidance for ecological restoration of metal-contaminated farmland and for advancing sustainable agricultural practices.

Authors’ contributions

L.W., S.Y. and Y.L.: Conceived and designed the experiments. L.W., S.Y., Y.H., Y.P., R.Z., Q.X., H.H., Y.X., and Y.L.: Performed the experiments and analyzed the data. L.W., S.Y. and Y.L.: Wrote the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Key-Area Research and Development Program of Guangdong Province, China (2025B0202060002); the Project of Rural Science and Technology Team “One-to-One” Service to Help the Town Full Coverage of Action Funding in Zhanjiang City (A23081), the Project of Social Services of Rural Science and Technology Specialists of Guangdong Ocean University (080503052212), the Graduate Education Innovation Program of Guangdong Ocean University (202401 and 202504), College Students’ Innovation and Entrepreneurship Training Program Project (CXXL2025055); Annual Outstanding Young Researcher Project of the Coastal Agriculture College of Guangdong Ocean University (BH2025JCQN004); the Innovation Team Project of Guangdong Ocean University for College Students (CXTD2021009).

Data availability

The raw sequencing data of the transcriptomics in this study can be found in the NCBI database under registration number PRJNA1232013 (https://www.ncbi.nlm.nih.gov/sra/PRJNA1232013).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.