A cytochrome P450 gene, GmSUR2 ^a , confers submergence tolerance and improves yield in soybean by modulating auxin homeostasis

ABSTRACT

Flooding is one of the most devastating abiotic stresses in agriculture, and enhancing flood tolerance is critical for sustainable crop production. Soybean (Glycine max) is highly susceptible to flooding stress. However, the molecular mechanism underlying soybean response to flooding stress remains largely unknown, and the genes available for improving soybean flood tolerance are relatively limited. In this study, we characterized the function of the predicted cytochrome P450 gene, GmSUR2^a , in regulating submergence tolerance in both soybean and Arabidopsis. The Arabidopsis SUR2 gene played a positive role in regulating submergence tolerance. The GmSUR2 ^a , which localizes to a previously identified quantitative trait locus (QTL) for waterlogging tolerance on soybean chromosome 03, shared conserved functions with AtSUR2 in regulating submergence tolerance. Transgenic soybean plants overexpressing GmSUR2 ^a displayed enhanced submergence tolerance accompanied by reduced auxin levels. Further experiments revealed that auxin negatively regulates submergence tolerance: Exogenous auxin application decreased survival rates, and auxin signaling mutants exhibited improved tolerance. We also identified that GmAGL15 transcription factor directly repressed GmSUR2 ^a expression. Importantly, field trials confirmed that GmSUR2 ^a overexpression not only improved submergence tolerance but also increased the yield of transgenic soybean plants. Our findings provide new insights into the molecular mechanisms of submergence tolerance and establish GmSUR2 ^a as a promising target for breeding flood‐resistant soybean varieties.

The key cytochrome P450 gene GmSUR2^a enables soybeans to withstand destructive submergence stress. By lowering the level of the plant hormone indole‐3‐acetic acid, this gene improves soybean survival and increases field yield. It offers an important tool for breeding stress‑resilient soybeans, securing food production against extreme weather conditions.

INTRODUCTION

As global warming intensifies, extreme weather is becoming increasingly frequent, posing severe threats to agricultural production. Among them, flooding is a major abiotic stress. Research indicates that approximately 12% of global croplands are affected by flooding, resulting in crop yield losses ranging from 40% to 80% (Shabala et al., 2014; Li et al., 2021b; Zhou et al., 2021b; Toulotte et al., 2022). Soybean (Glycine max), one of the world's most critical oilseed and forage crops, is highly susceptible to flooding stress (Hou and Thseng, 1991; Mustafa and Komatsu, 2014; Du et al., 2023; Shahzad et al., 2026). It has been reported that flooding stress causes soybean yield reductions of approximately 17%–43% at the vegetative stage and 50%–56% at the reproductive stage (Oosterhuis et al., 1990; Mustafa and Komatsu, 2014; Xu et al., 2025).

Flooding stress is generally categorized into “waterlogging” and “submergence” based on the water level relative to the plant tissues (Voesenek and Bailey‐Serres, 2015; Fukao et al., 2019; Gong et al., 2019). Waterlogging describes conditions where soil is fully saturated and water covers only root tissues, while aboveground organs remain exposed to air (Voesenek and Bailey‐Serres, 2015; Fukao et al., 2019). In contrast, submergence occurs when the entire aerial portion of the plant is completely covered by deep water (Voesenek and Bailey‐Serres, 2015; Fukao et al., 2019), making it one of the most severe flooding conditions encountered by plants (Mommer and Visser, 2005; Colmer and Voesenek, 2009). The primary damaging effect of flooding on crops is hypoxia (low oxygen stress), which disrupts multiple aspects of plant growth and development (Bailey‐Serres and Voesenek, 2008; Ahmed et al., 2013; Voesenek and Bailey‐Serres, 2015; Loreti et al., 2016; Zhou et al., 2020; Wang and Komatsu, 2022). Under anaerobic or low oxygen conditions, the oxidative phosphorylation pathway in plants is inhibited; instead, the glycolysis and fermentation pathways are activated, drastically reducing ATP production efficiency (Brownstein et al., 2013; Tamang et al., 2014; Loreti et al., 2018). Additionally, the low‐light environment caused by submergence stress decreases photosynthetic rates and CO₂ assimilation, impairing carbohydrate accumulation (Pedersen et al., 2013). Flooding also triggers excessive production of toxic metabolites (e.g., acetaldehyde, hydrogen sulfide) via anaerobic respiration, while inhibiting the electron transfer of mitochondria (Gill and Tuteja, 2010; Irfan et al., 2010). These influences induce reactive oxygen species (ROS) overproduction, leading to membrane lipid peroxidation and damage to proteins and nucleic acids and ultimately compromising normal plant growth and development (Gill and Tuteja, 2010; Irfan et al., 2010; Liu et al., 2020; Yuan et al., 2023).

During long‐term evolution, plants have evolved two primary adaptive strategies to cope with flooding stress: low oxygen escape syndrome (LOES) and low oxygen quiescence syndrome (LOQS) (Loreti et al., 2016). LOES is characterized by traits that facilitate escape from hypoxia, including aerenchyma formation, adventitious root (AR) development, upward leaf bending, stem elongation to breach the water surface, and formation of leaf‐surface air films (Voesenek and Bailey‐Serres, 2015; Loreti et al., 2016). For example, in deepwater rice, the ERF‐VII transcription factors SNORKEL1 (SK1) and SNORKEL2 (SK2) mediate hypoxia escape via the gibberellin signaling pathway (Hattori et al., 2009). In contrast, LOQS minimizes energy expenditure by suppressing energy‐intensive processes (e.g., ribosome biogenesis, cell wall synthesis), and generates only a small amount of ATP to maintain basic life activities via starch degradation, glycolysis, and ethanol fermentation (Voesenek and Bailey‐Serres, 2015; Loreti et al., 2016). In rice, the ERF‐VII transcription factor Submergence1 (Sub1) regulates hypoxic quiescence by inhibiting gibberellin signaling and enhancing expression of ADH1, a key gene in the ethanol fermentation pathway (Xu et al., 2006).

Adventitious roots (ARs), which originate from aboveground plant organs, are a critical adaptive trait for flooding tolerance (Jia et al., 2021). Increased AR formation enables plants to sustain water and nutrient uptake, as well as oxygen supply, under flooded conditions (Visser et al., 1996; Striker, 2012). Studies have shown that plants with more ARs exhibit significantly enhanced flooding resistance (Mano et al., 2005; Kim et al., 2015; Ye et al., 2018). Auxin, a pivotal phytohormone, regulates various aspects of plant development and stress responses (Yang et al., 2020; Wang et al., 2021; Zhang et al., 2022; Jing et al., 2023; Salehin, 2024; Vanneste et al., 2025; Jiang et al., 2026), and has been implicated in AR formation under flooding stress (Wang and Komatsu, 2022). For instance, in wheat, AR emergence from waterlogged stem nodes is associated with elevated indole‐3‐acetic acid (IAA) levels and is regulated by the IAA biosynthesis/transport genes TDC, YUC1, and PIN9 (Nguyen et al., 2018; Zhao et al., 2023). In pre‐etiolated flooded Arabidopsis seedlings, loss of function of AUX (an auxin influx carrier) reduces AR numbers—a phenotype that cannot be reversed by exogenous IAA (Bai et al., 2020; da Costa et al., 2020; Zhuang et al., 2021). In cucumber, flooding‐induced auxin accumulation regulates AR formation via crosstalk with ethylene and ROS (Qi et al., 2019; Zhang et al., 2025). Ethylene (ET) is a key regulator of plant responses to flooding stress (Geng et al., 2026). Restricted gas exchange in plant roots under flooding stress leads to ethylene accumulation, which acts as an early signal promoting the formation of aerenchyma and adventitious roots (ARs) in plants (Daniel and Hartman, 2024; Yan et al., 2026). In Arabidopsis, the SUR2 gene (encoding a cytochrome P450) modulates auxin homeostasis by directing indole‐3‐acetaldoxime (IAOx) flux toward indole glucosinolate biosynthesis (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001). Loss‐of‐function sur2 mutants exhibit blocked IAOx‐to‐indole glucosinolate conversion, leading to IAA accumulation and excessive AR formation (Bak and Feyereisen, 2001). Conversely, SUR2 overexpression (OE) results in typical auxin‐deficient phenotypes (Bak and Tax, Feldmann, et al., 2001). Despite preliminary insights into the link between auxin and AR formation, the role of auxin in regulating plant response to flooding remains largely uncharacterized.

To date, approximately 27 quantitative trait loci (QTLs) associated with soybean flooding tolerance have been mapped to nearly all chromosomes, except chromosomes 16 (Chr.16), Chr.17, and Chr.20 (VanToai et al., 2001; Reyna et al., 2003; Cornelious et al., 2005; Githiri et al., 2006; Sayama et al., 2009; Rizal and Karki, 2011; Nguyen et al., 2012, 2021; Van Nguyen et al., 2017; Ye et al., 2018; Dhungana et al., 2020, 2021). Notably, a QTL on Chr.03 has been linked to root system architecture under flooding stress, where the tolerant allele of this QTL is thought to enhance root growth and thereby improve flooding tolerance (Ye et al., 2018). Within this QTL region, 23 genes have been identified (Ye et al., 2018). In this study, we investigated the role of Glyma.03G031000 (designated GmSUR2 ^a ), one of 23 genes in the QTL region on Chr.03, in regulating submergence tolerance. For comparative purposes, we also characterized AtSUR2 function in Arabidopsis. Our results clarify the role of GmSUR2 ^a , reveal a negative correlation between auxin content and submergence tolerance, and provide insights into the molecular mechanisms underlying flooding tolerance in crops.

RESULTS

The soybean GmSUR2 ^a and Arabidopsis AtSUR2 function as positive regulators of submergence tolerance

Adventitious root formation is widely recognized to enhance plant survival under submergence (hypoxia) stress (Mano et al., 2005; Striker, 2012; Kim et al., 2015; Ye et al., 2018; Jia et al., 2021). In Arabidopsis, loss‐of‐function mutants of AtSUR2 (also known as AtCYP83B1) have been reported to exhibit increased adventitious root production (Barlier et al., 2000; Bak and Tax, Feldmann, et al., 2001; Figure S1A), which prompts us to investigate their responses to submergence stress. Unexpectedly, atsur2 mutants displayed a significant reduction in survival rate compared to the wild type Col‐0 under submergence stress (Figure 1A). Conversely, overexpression of AtSUR2 could enhance plant submergence tolerance (Figures 1A, S1C, D). In addition, AtSUR2 expression was upregulated at 3 and 6 h under submergence, followed by a gradual decrease at 12, 24, and 48 h (Figure S1B). These results demonstrated that AtSUR2 plays a positive rather than a negative role in regulating plant response to submergence stress.

Figure 1.

GmSUR2 ^a positively regulates submergence tolerance in soybean

(A) Phenotypes and survival rates of 5‐week‐old Col‐0, atsur2, and 35S::AtSUR2 overexpression transgenic plants (in Col‐0 background) after 3 d submergence treatment followed by 1‐week recovery (Bar = 5 cm). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 5) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (B) Phylogenetic analysis of AtSUR2 and GmSUR2^a homologs in Arabidopsis and soybean. (C) Relative expression levels of GmSUR2 ^a in Vegetative stage 1 (V1) Williams 82 (W82) plants before treatment, 1 day (D), 2 D, 3 D submergence treatment, and 1 D recovery (Recovery). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to one‐way ANOVA. (D) Relative expression levels of GmSUR2 ^a in Tianlong1 (TL1) and 35S::GmSUR2 ^a plants. Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to one‐way ANOVA. (E) Phenotypes and survival rates of Vegetative stage 2 (V2) (4‐week‐old) TL1 and 35S::GmSUR2 ^a overexpression transgenic plants after 7 d submergence treatment followed by 3 d recovery under greenhouse conditions (Bar = 10 cm). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 10) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (F) Phenotypes and survival rates of Vegetative stage 1 (V1) (4‐week‐old) TL1 and 35S::GmSUR2 ^a overexpression transgenic plants after 7 d submergence treatment and 3 D recovery under field conditions (Bar = 5 cm). Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA.

We then searched for the AtSUR2 homologs in soybean. The phylogenetic tree analysis revealed an expansion of the AtSUR2 homologous gene family in soybean, with eight members identified (Figure 1B). All eight genes clustered within a single branch, among which Glyma.03G031000 (designated GmSUR2 ^a ) shared the highest homology with AtSUR2 (Figure 1B). Gene mapping showed all these eight genes were located in a major QTL region on Chr.03 that was previously reported to be associated with soybean waterlogging tolerance (Figure S1E, F; Ye et al., 2018). The expression levels of eight homolog genes in the QTL region were measured by quantitative reverse transcription PCR (RT‐qPCR) under submergence treatment for 0 day (D), 1 D, 2 D, 3 D, and during recovery (Figure S2A). Among these genes, GmSUR2 ^a exhibited the most significant differential expression in response to flooding, which was upregulated on Days 1 and 3 after submergence treatment, followed by a return to baseline levels after recovery from submergence (Figures 1C, S2A). To determine GmSUR2 ^a 's role in submergence tolerance, we generated transgenic soybean lines overexpressing GmSUR2 ^a under the control of the constitutive 35S promoter. We then evaluated submergence tolerance of these OE lines under both greenhouse and field conditions. Consistent with the function of AtSUR2, GmSUR2 ^a ‐OE plants exhibited a significant increase in survival rate compared to non‐transgenic controls in both environments (Figures 1D–F, S3). In addition, GmSUR2 ^a ‐OE plants exhibited a longer total root length and a larger root area compared to the control (Figure S2B–D), as well as a significant increase in total leaf chlorophyll content, while showing a decrease in malondialdehyde (MDA)—a product of membrane lipid peroxidation under oxidative stress that is often used as an indirect indicator of cellular oxidative damage (Yeung et al., 2018; Figure S2E, F). These results confirm that GmSUR2 ^a also acts as a positive regulator of submergence tolerance in soybean.

GmSUR2 ^a enhances submergence tolerance in soybean by reducing endogenous indole‐3‐acetic acid (IAA) contents

To investigate whether GmSUR2 ^a and AtSUR2 share conserved functions, we heterologously overexpressed GmSUR2 ^a under the 35S promoter in the Arabidopsis atsur2 mutant background, aiming to determine whether it could rescue the submergence susceptibility of atsur2 mutants. As a control, we also overexpressed AtSUR2 in the atsur2 background. Under submergence stress, GmSUR2 ^a ‐OE plants, like AtSUR2‐OE plants, exhibited a significantly higher survival rate compared to atsur2 mutants (Figure 2A, B). Additionally, the characteristic phenotypic defects of atsur2 mutants (dwarf stature, increased adventitious roots, and shortened primary roots) were rescued by GmSUR2 ^a overexpression (Figure S4). These results demonstrated that GmSUR2 ^a and AtSUR2 play conserved roles in regulating both plant submergence tolerance and development.

Figure 2.

The GmSUR2 ^a enhances plants' submergence tolerance by negatively regulating IAA accumulation

(A) Phenotypes and survival rates of 4‐week‐old Col‐0, atsur2, 35S::AtSUR2/atsur2, and 35S::GmSUR2 ^a /atsur2 overexpression transgenic plants (in atsur2 background) after 3 d dark submergence treatment followed by 1‐week recovery (Bar = 5 cm). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 6) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (B) Relative expression of GmSUR2 ^a in 4‐week‐old Col‐0 and 35S::GmSUR2 ^a /atsur2 Arabidopsis plants. Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to one‐way ANOVA. (C) The content of indole‐3‐acetic acid (IAA) in Tianlong1 (TL1) and 35S::GmSUR2 ^a overexpression transgenic soybean plants (n = 3, **P < 0.01, Student's t‐test). (D) Phenotypes and survival rates of Vegetative stage 1 (V1) (4‐week‐old) TL1 plants after 9 days submergence followed by 2 days recovery with 0 and 5 μM p‐Chlorophenoxyisobutyric acid (PCIB) under growth chamber conditions (Bar = 10 cm). Before, before treatment. 0, 5 μM, recovery after submergence with 0 and 5 μM PCIB. Values are means ± SD (n = 6) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to one‐way ANOVA. (E) Phenotypes and survival rates of Vegetative stage 1 (V1) (4‐week‐old) TL1 and 35S::GmSUR2 ^a overexpression transgenic soybean plants after 3 days submergence followed by 3 d recovery with 0 mg/L, 0.25 mg/L, and 0.5 mg/L IAA under growth chamber conditions (Bar = 10 cm). Before, before treatment. 0 mg/L, 0.25 mg/L, and 0.5 mg/L, recovery after submergence with 0 mg/L, 0.25 mg/L and 0.5 mg/L IAA. Values are means ± SD (n = 6) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (F) Phenotypes and survival rates of 4‐week‐old Arabidopsis Col‐0, atsur2, and 35S::GmSUR2 ^a (before treatment (Before); after 48 h submergence treatment followed by 1‐week recovery) under submergence treatment with 0, 0.25 and 0.5 mg/L exogenous IAA, respectively (Bar = 5 cm). Values are means ± SD (n = 5) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (G) Phenotypes and survival rates of 4‐week‐old Col‐0, atiaa19 mutants after 3 days dark submergence treatment followed by 3 days recovery (Bar = 5 cm). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 5) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA.

In Arabidopsis, AtSUR2 has been proven to mediate auxin homeostasis, where atsur2 mutants exhibit elevated indole‐3‐acetic acid (IAA) levels while AtSUR2 overexpression (OE) lines show reduced IAA content (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001). Given the functional conservation between GmSUR2 ^a and AtSUR2, we hypothesized that GmSUR2 ^a similarly regulates auxin homeostasis in soybean, like AtSUR2's role in Arabidopsis (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001). To test this hypothesis, we quantified endogenous IAA levels in GmSUR2 ^a ‐OE transgenic soybean plants. Consistent with our hypothesis, GmSUR2 ^a ‐OE lines displayed a significant reduction in IAA content (Figure 2C). The auxin responsive genes, GmARF10, GmIAA2, and GmSAUR23, were also downregulated in GmSUR2 ^a ‐OE lines (Figure S5). These results confirm that GmSUR2 ^a retains a conserved function in mediating auxin homeostasis in soybean.

To investigate the role of auxin in response to submergence stress, we treated Arabidopsis and soybean plants with exogenous IAA and auxin signaling inhibitor p‐Chlorophenoxyisobutyric acid (PCIB) and assessed their survival under submergence stress (Ye et al., 2018). In both species, IAA‐treated plants exhibited a significant decrease in survival rate compared to untreated controls, while PCIB‐treated plants showed a significant increase in survival rate (Figure S2D–F). To further validate this relationship, we analyzed the Arabidopsis iaa19 mutant, a line defective in auxin signaling (Maki et al., 2022; Lee et al., 2024), for submergence tolerance. The iaa19 mutants showed a significant increase in survival rate under submergence (Figure 2G), reflecting enhanced tolerance to submergence stress. Taken together, these results demonstrate that auxin negatively regulates submergence tolerance across Arabidopsis and soybean.

The GmSUR2 ^a could be directly repressed by the MADS‐box transcription factor GmAGL15

To identify upstream regulators of GmSUR2 ^a , we performed yeast one‐hybrid screenings using a ~2 kb fragment of the GmSUR2 ^a promoter as bait. One gene encoding a predicted MADS transcription factor, Glyma.11G158812 (designated as GmAGL15), was identified and confirmed in yeast (Figures 3A, S6). To map the specific GmSUR2 ^a promoter region mediating its interaction with GmAGL15, we generated two truncated promoter fragments (spanning −703 bp to ATG and −1,589 bp to ATG, respectively) and performed Y1H binding assays (Figure 3B). The results revealed that the promoter region from −1,589 bp to −703 bp was responsible for the interaction between GmAGL15 and GmSUR2 ^a (Figure 3B). A single abscisic acid (ABA)‐responsive element (ABRE) motif was identified within this region, whose functional relevance was further validated by introducing site‐directed mutations (converting CACGT to AAAAA) (Figure 3B). Electrophoretic mobility shift assays (EMSA) were subsequently conducted to confirm the direct binding of GmAGL15 to the ABRE motif (Figure 3C). To investigate the regulatory effect of GmAGL15 on GmSUR2 ^a expression, we conducted Agrobacterium‐mediated transient transformation assays in tobacco leaves, using a luciferase (LUC) reporter gene driven by the GmSUR2 ^a promoter. Co‐expression of GmAGL15 with the reporter construct significantly suppressed LUC activity, demonstrating that GmAGL15 directly inhibits GmSUR2 ^a transcription (Figure 3D).

Figure 3.

GmAGL15 directly suppresses the expression of GmSUR2 ^a

(A) Yeast one‐hybrid assays of the interaction between GmAGL15 and GmSUR2 ^a promoter. (B) The interaction between GmAGL15 and the ABRE motif in the GmSUR2 ^a promoter by yeast one‐hybrid assay. Three bait constructs, pHis‐GmSUR2 ^a promoter1 (−703 to 0), pHis‐GmSUR2 ^a promoter2 (−1,589 to 0), and pHis‐GmSUR2 ^a promoter2 mut (ABRE motif CACGT mutated to AAAAA), and the prey construct pGADT7‐GmAGL15 were co‐transformed into the Y187 yeast strain and then plated onto specific nutrient‐deficient media. The combinations of pHis‐GmSUR2 ^a pro1 with the empty prey vector (pGADT7) and pHis‐GmSUR2 ^a with pGADT7 served as negative controls. The red letters indicate mutated nucleotides. (C) The direct binding of GmAGL15 to the ABRE‐motif elements in the GmSUR2 ^a promoter by Electrophoretic Mobility Shift Assay (EMSA). The bound protein‐DNA complex and the free probe are indicated by arrows. “+” or “−”, presence (+) or absence (−) of the corresponding component shown on the left. (D) The inhibition of the activity of the GmSUR2 ^a promoter by the luciferase (LUC) reporter assay in Nicotiana benthamiana leaves (n = 3, **P < 0.01, Student's t‐test).

Gene expression analysis showed that GmAGL15 was gradually downregulated in response to submergence treatment at 1 D, 2 D, and 3 D, and returned to baseline levels after recovery from submergence (Figure S7B). To assess the biological role of AGL15 in submergence tolerance, we generated GmAGL15 overexpression (OE) soybean lines. Under submergence stress, GmAGL15‐OE plants exhibited a significant reduction in survival rate compared to wild‐type (WT) plants (Figure 4A), indicating a negative regulatory role for GmAGL15. Consistent with its repressive effect on GmSUR2 ^a , RT‐qPCR analysis showed that GmSUR2 ^a transcript levels were significantly downregulated in GmAGL15‐OE lines with or without submergence treatment (Figures 4B, S7A). The IAA content was markedly higher in GmAGL15‐OE plants (Figure 4E). Upon submergence exposure, IAA levels in these OE plants declined progressively at 1, 3, and 5 d of treatment, and subsequently rebounded to an elevated level following recovery from submergence stress (Figure S7C). Measurements of total leaf chlorophyll and malondialdehyde (MDA) showed that GmAGL15‐OE plants had a significant decrease in chlorophyll content (Figure 4C) and an increase in MDA levels (Figure 4D). All the above results indicated that GmAGL15 functions negatively in regulating submergence tolerance by directly repressing GmSUR2 ^a in vivo.

Figure 4.

The GmAGL15 negatively regulates submergence tolerance in soybean

(A) Phenotypes and survival rates of Vegetative stage 2 (V2) (4‐week‐old) soybean Tianlong1 (TL1) and 35S::GmAGL15 overexpression transgenic plants (in TL1 background) after 5 days submergence treatment followed by 3 days recovery under greenhouse conditions (Bar = 10 cm). Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 5) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (B) Relative expression of GmAGL15 and GmSUR2 ^a in GmAGL15 overexpression transgenic plants. Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to one‐way ANOVA. (C, D) The chlorophyll (C) and malondialdehyde (MDA) contents (D) of TL1 and 35S::GmAGL15 overexpression transgenic soybean plants before submergence and recovery after submergence. Before, before treatment. Recovery, recovery after submergence. Values are means ± SD (n = 3) from three independent experiments. Different lowercase letters indicate significant differences (P < 0.05) according to two‐way ANOVA. (E) The content of indole‐3‐acetic acid (IAA) in TL1 and 35S::GmAGL15 overexpression transgenic soybean leaves (n = 3, *P < 0.05, Student's t‐test).

We also evaluated the developmental phenotypes of GmAGL15‐OE plants. The results showed that GmAGL15‐OE plants exhibited significant reductions in branch number, pod number, and plant height, whereas 100‐seed weight showed no significant difference. Consequently, yield was decreased (Figure S8).

Overexpression of GmSUR2 ^a in soybean enhances yield performance under field conditions

To investigate the application potential of GmSUR2 ^a in breeding, we evaluated yield‐related agronomic traits of GmSUR2 ^a ‐OE transgenic soybean plants in field trials conducted in 2023 and 2024 (Figure 5A). The results showed that GmSUR2 ^a ‐OE plants exhibited a significant increase in the number of branches per plant (Figure 5B), pods per plant (Figure 5C), and seeds per plant (Figure 5D). Meanwhile, there were no significant differences in plant height (Figure 5E), 100‐seed weight (Figure 5F), protein content (Figure 5G), or oil content (Figure 5H) between GmSUR2 ^a ‐OE plants and controls. Consequently, GmSUR2 ^a ‐OE plants have a significant improvement in overall yield per plant (Figure 5I). These findings suggest GmSUR2 ^a has substantial application prospects for the development of submergence‐tolerant soybean cultivars via transgenic breeding approaches.

Figure 5.

GmSUR2 ^a improves key agronomic traits in soybean

(A) The phenotype of Tianlong1 (TL1) and GmSUR2 ^a overexpression transgenic plants (Bar = 5 cm). (B) Branch number. (C) Pod number. (D) Seed number. (E) Plant height. (F) 100‐seed weight. (G) Protein content. (H) Oil content. (I) Yield per plant. Bars represent the means ± SD. Different lowercase letters indicate significant differences (n ≥ 12, P < 0.05) according to one‐way ANOVA.

DISCUSSION

In this study, we elucidated the positive role of GmSUR2 ^a and AtSUR2 in regulating submergence tolerance by modulating auxin biosynthesis in soybean and Arabidopsis, respectively. The negative role of auxin in the regulation of submergence tolerance was revealed. The GmSUR2 ^a gene was found to hold considerable potential in breeding submergence‐tolerant soybean varieties. Our results advance the current understanding of molecular mechanisms underlying submergence tolerance and provide a valuable theoretical foundation and candidate genetic resource for improving flood tolerance in soybean and other crops.

Increased adventitious root formation does not necessarily confer enhanced submergence tolerance

Adventitious root formation represents a key morphological adaptation to flooding stress (Voesenek and Bailey‐Serres, 2015; Loreti et al., 2016; Jia et al., 2021; Wang et al., 2025). It is widely accepted that a greater number of adventitious roots helps sustain water and nutrient uptake as well as oxygen supply under flooding conditions (Visser et al., 1996; Striker, 2012). Kim et al. (2015) reported that the waterlogging‐tolerant soybean line PI408105A produced more adventitious roots than the susceptible line S99‐2281 (Kim et al., 2015). Mano et al. (2005) demonstrated that enhanced adventitious root formation contributes to high flooding tolerance in teosinte (Zea mays ssp. huehuetenangensis) (Mano et al., 2005). Ye et al. (2018) observed the differences in adventitious root number between flood‐tolerant and susceptible soybean near‐isogenic lines (NILs), with tolerant NILs forming more adventitious roots than the susceptible ones (Ye et al., 2018). However, does a greater number of adventitious roots always translate to stronger flooding tolerance? In our study, although Arabidopsis atsur2 mutants are known to produce more adventitious roots (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001), we unexpectedly found that they displayed reduced submergence tolerance (Figure 1A). This result indicates that adventitious root number and submergence tolerance do not always exhibit a positive correlation. Plant adaptation to flooding involves multiple morphological modifications, including aerenchyma formation, adventitious root development, shoot elongation, stomatal closure, and leaf petiole movement (Bailey‐Serres and Voesenek, 2008; Voesenek and Bailey‐Serres, 2015; Huang et al., 2020). Concurrently, flood responses are tightly and coordinately regulated by a suite of physiological and biochemical processes, such as carbon/nitrogen metabolism, photosynthesis and respiration, programmed cell death, and phytohormone signaling (Bailey‐Serres and Voesenek, 2008; Voesenek and Bailey‐Serres, 2015; Wang and Komatsu, 2022; Guan et al., 2025). Adventitious root formation is merely one component of this complex adaptive network. Improving crop flood tolerance thus relies on the comprehensive and coordinated optimization of all adaptive traits, rather than the enhancement of a single feature. Strengthening one trait in isolation may not boost flood resistance, and can even exert adverse effects, especially if such enhancement is accompanied by impairments in other adaptive processes. In atsur2 mutants, for example, excessive adventitious root formation likely imposes a heavy demand for nutrients and energy, diverting resources that would otherwise support stress resistance. Additionally, the excessive accumulation of auxin in these mutants may disrupt the balance of phytohormone signaling pathways, further compromising stress tolerance. Therefore, a “tradeoff threshold” likely exists for flooding tolerance‐related traits: Enhancing a trait within this threshold may produce beneficial effects, whereas exceeding it may undermine overall stress adaptation.

The phytohormone auxin plays a negative role in regulating submergence tolerance in soybean

In Arabidopsis, AtSUR2 encodes a cytochrome P450 that maintains auxin homeostasis by directing the flux of indole‐3‐acetaldoxime (IAOx) to indole glucosinolates (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001). IAOx is the metabolic branch point between indole glucosinolates and IAA biosynthesis (Bak and Tax, Feldmann, et al., 2001). Loss‐of‐function mutations in AtSUR2 blocked the conversion of IAOx into indole glucosinolates, resulting in reduced indole glucosinolate levels and concomitant accumulation of endogenous IAA. Conversely, overexpression of AtSUR2 increases indole glucosinolate levels while reducing endogenous IAA content (Bak and Feyereisen, 2001; Bak and Tax, Feldmann, et al., 2001). These results indicate that AtSUR2 acts as an indirect negative regulator of auxin biosynthesis.

The phytohormone auxin acts as a master regulator of plant growth and development (Kepinski and Leyser, 2005; Zhao, 2010; Shirley et al., 2019; Xi et al., 2019; Zhang et al., 2020b; Mazzoni‐Putman et al., 2021; Yu et al., 2022; Ma et al., 2023; Li et al., 2023b; Wang et al., 2024a), and also plays important roles in plant responses to various abiotic stresses (Jing et al., 2023). For instance, exogenous auxin application protects pollen from heat shock in wheat, rice, and barley (Sakata et al., 2010; Sharma et al., 2018; Abeysingha et al., 2021; Jing et al., 2023); in Arabidopsis, the auxin biosynthesis gene YUC8 is directly targeted by the thermomorphogenesis regulator PIF4 under high temperature conditions (Sun et al., 2012); the expressions of auxin biosynthesis genes including NIT1, NIT2, and members of YUC family, are significantly induced in response to NaCl treatment in Arabidopsis, cucumis sativus, and potato (Kim et al., 2013; Yan et al., 2016; Cackett et al., 2022); overexpression of auxin biosynthesis genes, YUC6 and YUC7, enhances drought resistance (Lee et al., 2012; Kim et al., 2013), whereas the yuc1yuc2yuc6 triple mutants with reduced endogenous IAA levels exhibits decreased drought tolerance (Shi et al., 2014); overexpression of the bacterial tryptophan‐2‐monooxygenase (iaaM) increases endogenous auxin levels and thereby enhances drought tolerance (Shi et al., 2014). These findings highlight the positive roles of auxin in regulating plant responses to abiotic stresses.

In contrast to these previous studies, our results suggest that the auxin pathway may play a negative role in mediating plant submergence tolerance, supported by multiple lines of evidence: Arabidopsis atsur2 mutants with elevated IAA levels showed reduced submergence tolerance, whereas AtSUR2 overexpression (OE) lines with reduced IAA levels exhibited enhanced tolerance (Figure 1A); transgenic soybean plants overexpressing GmSUR2 ^a exhibited submergence tolerance accompanied by lower endogenous IAA levels (Figure 2C); exogenous IAA treatment reduced submergence tolerance in both soybean and Arabidopsis (Figure 2E, F); and, the Arabidopsis auxin signaling‐deficient mutant iaa19 displayed enhanced submergence tolerance (Figure 2G). Future studies are needed to further elucidate the molecular mechanisms underlying auxin‐mediated regulation of plant submergence responses.

Soybean GmSUR2 ^a , which functions conservatively with Arabidopsis AtSUR2, is hypothesized to be the causal gene underlying a previously reported waterlogging tolerance QTL on soybean chromosome 03

To investigate the functional similarity between GmSUR2 ^a and AtSUR2, we heterologously overexpressed GmSUR2 ^a in an atsur2 mutant background and found that GmSUR2 ^a overexpression rescued both flood susceptibility and developmental deficits in atsur2 mutants, similar to the effect of AtSUR2 overexpression (Figures 2A, S4). In addition, IAA content was significantly reduced in GmSUR2 ^a ‐OE transgenic soybean plants (Figure 2C), consistent with previous reports for AtSUR2‐OE Arabidopsis plants (Bak and Tax, Feldmann, et al., 2001). These results confirmed the functional conservation between GmSUR2 ^a and AtSUR2. We attempted to generate GmSUR2 ^a knockout plants using CRISPR/Cas9 technology. However, due to high sequence similarity among members of the GmSUR2 ^a gene family (Figure S1F), we were unable to isolate homozygous single‐knockout plants for GmSUR2 ^a . Consequently, submergence tolerance and root phenotypes of GmSUR2 ^a knockout plants could not be evaluated in this study. We hypothesized that GmSUR2 ^a single knockout may not lead to obvious phenotypic abnormalities due to functional redundancy among its homologs. Further attempts, such as generating multi‐gene knockouts targeting redundant GmSUR2 ^a homologs, will be necessary to test this hypothesis and clarify gene function.

The GmSUR2 ^a gene is located within a previously mapped flood tolerance QTL interval on Chr.03, which contains more than 20 genes, including SAUR family proteins, YABBY family transcription factors, plant invertase/pectin methylesterase inhibitors, glycosyl hydrolases, GmSUR2 ^a , and its homologs (Ye et al., 2018). In this study, we clarified the function of GmSUR2 ^a in regulating submergence tolerance. The expression of GmSUR2 ^a was induced by submergence treatment (Figure 1C); overexpression of GmSUR2 ^a in soybean enhanced submergence tolerance by reducing IAA content in vivo (Figures 1E, 2C); exogenous IAA treatment decreased soybean submergence tolerance (Figure 2E). Collectively, these results support the idea that GmSUR2 ^a contributes to the flooding tolerance phenotype conferred by the QTL on Chr. 03 (Ye et al., 2018). Nevertheless, we cannot exclude the possibility that other genes within this QTL interval also modulate flood tolerance, as several additional genes in the region show differential expression under flooding stress (https://plantrnadb.com/) (Zhang et al., 2020a). Future studies should therefore include functional characterization of these genes to define their roles in flooding tolerance regulation.

In summary, our study characterized the functions of GmSUR2 ^a and AtSUR2 in regulating submergence tolerance. These findings not only expand current knowledge of plant flooding tolerance mechanisms but also provide a promising molecular target for the genetic improvement of flooding tolerance in crops.

MATERIALS AND METHODS

Plant materials and growth conditions

The Arabidopsis thaliana T‐DNA insertion mutants atsur2 (SALK_028573) used in this study were acquired from the AraShare resource center. Overexpression and complementation lines (35S::AtSUR2/Col‐0, 35S::GmSUR2 ^a /Col‐0, 35S::AtSUR2/atsur2, 35S::GmSUR2 ^a /atsur2) were produced via Agrobacterium tumefaciens‐mediated floral dip transformation of wild‐type or mutant plants, respectively.

The Arabidopsis seed germination and seedling culture were performed following a previously reported protocol with minor modifications (Li et al., 2023a): 4‐week‐old plants were completely submerged in distilled water under continuous dark conditions, with the water depth maintained at least 5 cm above the shoot apex for 3 d, followed by a 7 d recovery period under the original growth conditions before survival rate assessment. To prevent pot flotation during submergence, five glass beads (2 cm in diameter) were placed at the bottom of each container prior to planting. Each experimental replicate included nine plants per genotype.

Soybean cultivar Tianlong1 (TL1) was used as the genetic background to generate the GmSUR2 ^a overexpression line (35S::GmSUR2 ^a ). Homozygous T3 plants were subjected to submergence stress treatment in a controlled‐environment greenhouse at Henan Agricultural University, Zhengzhou, China. Fully mature seeds were selected, germinated, and transplanted into soil. Plants were grown under controlled conditions (25°C, 16‐h light/8‐h dark photoperiod). Submergence stress was imposed on 4‐week‐old plants by maintaining the water level 5 cm above the shoot apex for 1 week. Survival rates were recorded after a 3 d recovery period. The survival rate was calculated using the formula: Survival rate (%) = (Number of surviving plants after recovery/Number of plants before submergence treatment) × 100%.

The field submergence stress trial was conducted at the Yuanyang Experimental Farm of Henan Agricultural University, Xinxiang, China, using two soybean genotypes: Tianlong1 (TL1) and 35S::GmSUR2 ^a . Prior to sowing, the field was carefully leveled to ensure uniform water depth and synchronous initiation and termination of stress across all plots. The experiment was arranged in a randomized complete block design with three replicates. Seeds were sown at a depth of 3–5 cm, with a row spacing of 20 cm, a within‐row plant spacing of 5 cm, and 20 cm alleyways between plots. Submergence was initiated at the V1 growth stage by maintaining a water depth of approximately 5 cm above the shoot apex for 7 d. After drainage, plant survival rates were evaluated following a 3 d recovery period under well‐drained conditions (Zhou et al., 2021a).

In addition, other agronomic traits of the transgenic lines planted in the field were investigated at the corresponding growth stages, including yield, plant height, branch number, pod number, seed number, and 100‐seed weight at full maturation. Seed quality‐related traits (oil content and protein content) were determined using a near‐infrared spectrophotometer (NIR) seed analyzer (DA7200, Perten Instruments, Huddinge, Sweden) (Wang et al., 2020; Li et al., 2021a; Hu et al., 2022; Wang et al., 2023; Yang et al., 2024).

Submergence treatment with exogenous IAA and PCIB application

For exogenous indole‐3‐acetic acid (IAA) treatments, a stock solution was prepared by dissolving IAA in 0.1% (v/v) dimethyl sulfoxide (DMSO), which was then diluted with distilled water to final concentrations of 0, 0.25, and 0.5 mg/L. The control solution contained the same concentration of DMSO (0.1% v/v) without IAA. Four‐week‐old Col‐0 and TL1 plants were completely submerged in the respective treatment solutions, following the submergence protocol described in the previous section (Du et al., 2025).

For exogenous p‐chlorophenoxyisobutyric acid (PCIB) treatments, a 0.1 mol/L stock solution was prepared by dissolving PCIB in absolute ethanol, which was then diluted with distilled water containing 0.1% (v/v) Tween‐20 to final concentrations of 0 and 5 μM. The control solution contained equivalent concentrations of ethanol and Tween‐20 without PCIB. Four‐week‐old TL1 soybean plants were sprayed with the treatment solutions until leaf runoff occurred. After 24 h of treatment, the plants were subjected to submergence as described previously.

Root phenotype analysis

Surface‐sterilized seeds of Arabidopsis genotypes (Col‐0, atsur2, 35S::AtSUR2/atsur2, and 35S::GmSUR2 ^a /atsur2) were stratified and germinated on 1/2MS plates. After 3 d of vertical growth, root phenotypes were observed and imaged using a stereo microscope (Zeiss Stemi508, Carl Zeiss AG, Oberkochen, Germany) (Du et al., 2025).

Four‐week‐old plants of soybean cultivar TL1 and transgenic line 35S::GmSUR2 ^a were harvested, and their roots were carefully washed to remove adhering vermiculite. Root system architecture was quantified using an automated root scanning system (EPSON Expression 11000XL) equipped with WinRHIZO software (Regent Instruments Inc, Canada). First, to minimize overlap and crossing, the cleaned root systems were placed in transparent root trays containing 1‐cm‐deep purified water and carefully spread with forceps according to their natural growth orientation. Second, root images were acquired at 400 dpi resolution and saved in TIFF format. Then, root architectural traits were determined via image analysis using WinRHIZO software (Yang et al., 2023).

Measurement of soybean leaf chlorophyll and malondialdehyde (MDA) content

The total chlorophyll content in soybean leaves was determined using a modified ethanol extraction method (Ritchie, 2006). Briefly, fresh leaf samples were washed, dried, and cut into 2 mm strips after removing the midrib. Approximately 0.1 g of fresh leaf tissue was completely submerged in 10 mL of 95% ethanol. The leaf tissue was incubated in the dark for 40 h until it turned colorless. After incubation, the absorbance of the extract was measured at 649 and 665 nm using a Lambda 365 UV‐Vis spectrophotometer with a 10 mm path length cuvette, with 95% ethanol as the blank control.

Malondialdehyde (MDA) content was determined following the thiobarbituric acid (TBA) method (Yeung et al., 2018). Leaf tissue (0.5 g) was ground in 5 mL of 10% (w/v) trichloroacetic acid (TCA) and centrifuged at 10,000 × g for 10 min. Next, 2 mL of 0.6% (w/v) TBA was added to 2 mL of the supernatant, mixed thoroughly, boiled for 15 min, rapidly cooled, and centrifuged at 3,000 × g for 10 min. The absorbance of the resulting supernatant was measured at 450, 532, and 600 nm. MDA concentration was calculated using the formula: 6.45 × (A₅₃₂ – A₆₀₀) – 0.56 × A₄₅₀.

Endogenous IAA quantification

Soybean cultivar TL1, 35S::GmSUR2 ^a transgenic lines, and 35S::GmAGL15 transgenic lines were grown under standard conditions until the V2 stage. Samples consisting of the second trifoliate leaf and younger tissues (3 g fresh weight) were harvested, immediately frozen in liquid nitrogen, and ground into a fine powder using a pre‐chilled mortar and pestle (Fu et al., 2025). For endogenous IAA extraction, approximately 170 mg of this powdered material was accurately weighed and extracted with methanol and ²H₂‐IAA for 24 h. Endogenous IAA was purified and measured as previously described, with modifications to the detection conditions (Fu et al., 2012; Chu et al., 2017; Wang et al., 2024b). LC‐MS/MS analysis was performed on a UPLC system (Waters) coupled to a 5500 Qtrap system (AB SCIEX). LC separation was achieved using a BEH C18 column (1.7 μm, 100 × 2.1 mm; Waters) with mobile phase A (0.1% (v/v) formic acid in water) and mobile phase B (acetonitrile). The gradient program was as follows: 0–0.5 min, 5% B; 0.5–14 min, linear gradient from 5% B to 25% B; 14–16 min, linear gradient from 25% B to 100% B. IAA was detected in multiple reaction monitoring (MRM) mode. The transitions for IAA and [²H₂]‐IAA were 174.1 > 130.1 and 176.1 > 132.1, respectively. The levels of IAA in TL1 and GmAGL15‐OE plants with submergence treatment were measured using an enzyme‐linked immunosorbent assay (ELISA) kit (Cat. YJ147100, Shanghai Yuanji Biotechnology Co. Ltd., China) (Li et al., 2025).

Gene expression profiling

Soybean cultivar Williams 82 (W82) was grown until the V1 stage and then subjected to submergence stress. Tissue samples were collected at 0, 1, 2, and 3 days after the onset of stress, as well as after a 1 day recovery period, with three independent biological replicates per time point. Total RNA was extracted from the samples using an RNAprep Pure Plant Kit (Tiangen, cat. #DP441, China). cDNA was synthesized according to the manufacturer's instructions (Clontech, cat. #6110 A, Japan), and RT–qPCR was performed on an ABI 7500 real‐time PCR system using the SYBR Green Mix Kit (Bio‐Rad, Hercules, CA, USA). Sequences of all gene‐specific primers used in this study are provided in Table S1.

Dual‐luciferase reporter assays

A 2,095 bp fragment spanning the promoter region of GmSUR2 ^a was amplified from soybean genomic DNA. Primer sequences were derived from the G. max Williams 82 (W82) genome assembly in Phytozome (https://phytozome-next.jgi.doe.gov/). The purified fragment was cloned into the pGREENII 0800‐LUC vector. The coding sequence of GmAGL15 was inserted into a GFP‐tagged overexpression vector via homologous recombination. Both constructs were introduced into the Agrobacterium tumefaciens strain GV3101. Bacterial cultures were harvested and resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES, 150 μM acetosyringone, pH 5.6). Bacterial suspensions carrying different constructs were mixed at appropriate ratios, incubated at 28°C for 2–3 h, and then infiltrated into leaves of 4‐week‐old Nicotiana benthamiana plants. After 48 h, luciferin substrate was applied to the infiltrated leaves, and luminescence signals were detected using an in vivo imaging system (NightSHADE LB985, Berthold Technologies, Germany). Protein extracts were prepared from the infiltrated leaf discs, and luciferase activity was quantified using a dual‐luciferase reporter assay kit (Yeasen Biotechnology, Cat# 11402ES08, Shanghai, China) according to the manufacturer's instructions (GloMax® 20/20, Promega Corporation, USA) (Chao et al., 2024).

Yeast one‐hybrid assay

The soybean cDNA library was constructed using various tissues of the Williams 82 (W82) cultivar, including roots, root nodules, stems, shoot apices, compound leaves, trifoliate leaves, flowers, unpodded pods, filled pods, immature seeds, and mature seeds. The cDNA was directionally cloned into the pGADT7‑AD vector.

The yeast one‐hybrid (Y1H) assay was performed in accordance with the manufacturer's instructions (Clontech, Japan; catalog nos. #630491, #630466, #630499). Briefly, the bait sequences (segments within the SUR2 promoter) were first amplified and cloned into the pAbAi vector. The constructed vectors were then linearized using the restriction enzyme BbsI and transformed into the yeast strain Y1H Gold. The bait sequences were integrated into the yeast genome through homologous recombination. Following selection on synthetic defined (SD) medium lacking uracil (SD/−Ura), healthy colonies were picked for PCR validation. To eliminate false‐positive results, the selected yeast colonies were further screened on SD/−Ura medium supplemented with an appropriate concentration of aureobasidin A (AbA), a compound that inhibits the growth of yeast cells.

Next, the constructed pGADT7‐AGL15 and pHis‐SUR2 vectors were co‐transformed into the Saccharomyces cerevisiae Y187 (MATα) strain. SD/−Leu/−Trp selective medium was used to select positive colonies, which were subsequently validated by PCR amplification. Finally, Y187 yeast strains harboring both the pGADT7‐AGL15 and pHis‐SUR2 vectors were plated on SD/−His/−Leu/−Trp medium supplemented with 150 mM 3‐amino‐1,2,4‐triazole (3‐AT) to examine the direct interaction between AGL15 and the SUR2 promoter (Bu et al., 2025).

Electrophoretic mobility shift assay (EMSA)

The recombinant plasmid was transformed into Escherichia coli (E. coli) BL21 competent cells. Isopropyl β‐D‐1‐thiogalactopyranoside (IPTG) was added to the culture medium to a final concentration of 1.0 mM, and the MBP‐GmAGL15 fusion protein was obtained after induction at 16°C for 16 h. The MBP‐GmAGL15 fusion protein was purified using an MBP‐Tag Protein Purification Kit (CWbio, Beijing, China) in strict accordance with the manufacturer's instructions. 5’‐biotin‐labeled wild‐type and mutant DNA probes were synthesized and annealed. Binding reactions and detection were performed using a Chemiluminescent EMSA Kit (Coolaber Co. Ltd., Beijing, China) following the manufacturer's protocol. The resulting bands were visualized with a TOUCH IMAGER system (E‐BLOT, Shanghai, China). All primers used in this study are listed in Table S1.

Sequence and phylogenetic analysis

Genomic sequences of soybean (G. max [Williams 82.a4.v1]) and Arabidopsis (A. thaliana [TAIR10]) were retrieved from Phytozome v14 (https://phytozome-next.jgi.doe.gov/). Multiple sequence alignment was conducted with DNAMAN software under default parameters. Phylogenetic trees were constructed using the neighbor‐joining method in MEGA 6.06 with 1,000 bootstrap replicates for statistical support. Sequence specificity was verified via BLAST analysis against the whole genome database of SoyBase (https://www.soybase.org/).

Statistical analysis

All experiments were performed with at least three independent biological replicates. Data are presented as the means ± standard deviation (SD). Statistical analyses were conducted using one‐way analysis of variance (ANOVA) followed by Fisher's least significant difference (LSD) post hoc test in GraphPad Prism version 8.0.2 (GraphPad Software, San Diego, CA, USA). Differences were considered statistically significant when P < 0.05.

Generative AI usage statement

During the preparation of this work, the authors used the Generative AI tool Doubao to improve the language of this manuscript. After using this tool, the authors carefully reviewed and revised the content where necessary and took full responsibility for the final version of the publication.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.

AUTHOR CONTRIBUTIONS

Y.J. conceived and supervised the study. Y.C. and P.F. performed most of the experiments. M.Z., L.Z., Y.L., Y.J., Y.W., B.Z., C.L., C.W., S.C., and E.X. helped with vector construction, transformation, field planting, and data analysis. Y.J., E.X., Y.C., and P.F. analyzed the data and wrote and revised the article. All authors read and approved the content.

Supporting information

Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70299/suppinfo

Figure S1. Characterization of AtSUR2 and GmSUR2 ^a

Figure S2. The GmSUR2 ^a positively regulates soybean tolerance to submergence stress

Figure S3. Field evaluation of submergence tolerance for GmSUR2 ^a overexpression soybean plants

Figure S4. The complementation of the atsur2 mutant phenotype by GmSUR2 ^a

Figure S5. Expression levels of auxin‐related genes in GmSUR2 ^a overexpression transgenic soybean plants

Figure S6. Phylogenetic analysis of the AGL15 family in Arabidopsis and soybean

Figure S7. Expression levels of GmSUR2 ^a and GmAGL15 in TL1 and 35S::GmAGL15 transgenic plants

Figure S8. Statistical analyses of key agronomic traits in TL1 and 35S::GmAGL15 overexpression lines

Table S1. List of primers used in PCR analysis

Biographies

Chen, Y. , Feng, P. , Zheng, M. , Zhao, L. , Li, Y. , Jing, Y. , Wang, Y. , Zhang, B. , Liu, C. , Wei, C. , et al. (2026). A cytochrome P450 gene, GmSUR2 ^a , confers submergence tolerance and improves yield in soybean by modulating auxin homeostasis. J. Integr. Plant Biol. 68: 3628–3643.