A Trp-574-Leu mutation in acetolactate synthase confers imazamox resistance in barnyardgrass (Echinochloa crus-galli) from China

Abstract

Barnyardgrass [Echinochloa crus-galli (L.) P. Beauv.] is increasingly infesting imidazolinone-tolerant (IMI-T) rice (Oryza sativa L.) fields in China, and E. crus-galli imazamox resistance has become the major concern for weed management in IMI-T rice fields. In this study, the susceptible population JLGY-3 (S) and the suspected resistant population JHXY-2 (R) collected from IMI-T rice fields were used as research subjects. When treated with imazamox, the JHXY-2 (R) population showed a high level of herbicide resistance with a resistance index of 31.2. JHXY-2 (R) was cross-resistant to all five acetolactate synthase (ALS) inhibitors from different chemical families, but sensitive to herbicides inhibiting acetyl-CoA carboxylase. To understand the reason why JHXY-2 (R) was resistant to imazamox, we performed experiments to characterize potential target-site resistance (TSR) and non–target site resistance (NTSR) mechanisms. A Trp-574-Leu amino acid mutation in ALS and low imazamox ALS sensitivity were the main mechanisms underlying imazamox resistance in this JHXY-2 (R) population. There was no significant difference in ALS gene expression and ALS protein abundance between R and S populations. High-performance liquid chromatography–tandem mass spectrometry analysis showed enhanced metabolism of imazamox in JHXY-2 (R), which was in contrast to the results of pretreatment with a metabolic enzyme inhibitor. Treatments with cytochrome P450 monooxygenase/glutathione S-transferase (P450/GST) inhibitors did not alter the resistance level of JHXY-2 (R) against imazamox. Transcriptome sequencing showed that there was almost no significant difference in the expression of P450 and GST metabolic enzyme genes between R and S populations, and only GSTU1 showed a significant upregulation in the R population, further clarifying the NTSR mechanism of JHXY-2 (R). In conclusion, amino acid mutation and higher enzyme activity of ALS are the main causes of imazamox resistance in JHXY-2 (R). However, given the differences in imazamox residues in the leaves of E. crus-galli, there may still be undetectable NTSR mechanisms that are causing imazamox resistance in the R population.

Introduction

Weed control has always been an important issue in agricultural production. If weeds are not controlled in time and effectively, grain yield will be seriously reduced (Singh et al. 2015). Weedy rice (Oryza sativa L.) is a harmful weed in rice fields and is extremely difficult to control with the herbicides registered in rice fields because of its botanical similarity to cultivated rice (De Leon et al. 2019). Weedy rice competes with cultivated rice for water and nutrients in paddy fields, resulting in serious loss of rice yield (Burgos et al. 2008). In response to this situation, the BASF Company initially developed the technology of planting rice tolerant to imidazolinone herbicides combined with the use of imazethapyr (Clearfield^®) in the United States in 2003. Since then, Clearfield^® rice has been rapidly popularized and planted in the United States and has brought considerable income to Clearfield^® rice growers (Burgos et al. 2014). In recent years, this technology has also been developed in China, where farmers have begun to grow imidazolinone-tolerant (IMI-T) rice cultivars in combination with the use of imazamox to control weeds in paddy fields. Imazamox is one of the imidazolinone herbicides that can effectively control annual grasses and broadleaf weeds. In China, the main weeds in paddy fields include weedy rice, Echinochloa spp., Chinese sprangletop [Leptochloa chinensis (L.) Nees], large crabgrass [Digitaria sanguinalis (L.) Scop.], heusenkraut (Ludwigia prostrata Roxb.), and variable flatsedge (Cyperus difformis L.) (Liu et al. 2021). Most of the weeds, including Echinochloa spp., can be controlled by spraying imazamox in the IMI-T rice fields in the early years after introduction of the technology. However, after continuously planting IMI-T rice and using imazamox for a few years, some farmers found that Echinochloa spp. became rampant in the rice fields. After observing the distribution of Echinochloa spp. in many places growing IMI-T rice, we found that E. crus-galli is the main species infesting IMI-T rice fields among the sites.

Barnyardgrass [Echinochloa crus-galli (L.) P. Beauv.] is a very common malignant weed in the rice fields, and its control mainly depends on the use of various herbicides (Fang et al. 2019b). In specialized cultivation practices, such as the integration of IMI-T rice with imazamox, herbicides are applied repeatedly on an annual basis. Under this continuous selection pressure of a single herbicide, it is easy for E. crus-galli to evolve herbicide resistance (Chen et al. 2016).

Echinochloa crus-galli is an allopolyploid weed species and has a more intricate genomic architecture compared with diploid species (Panozzo et al. 2013, 2021). Due to its substantial genome size and the delayed initiation of genetic research on weeds, the genome sequence of E. crus-galli remained elusive until Guo et al.’s study (2017) unveiled the complexities of its genome. In polyploid species, the gene-copy number at each locus can vary (Panozzo et al. 2021); for instance, Fang et al. (2019a) reported differences in ALS gene-copy numbers across various populations of E. crus-galli, while Iwakami et al. (2017) observed similar variability in ALS gene-copy numbers within shortawn foxtail (Alopecurus aequalis Sobol.).

Imazamox is an acetolactate synthase (ALS) inhibitor, inhibiting the biosynthesis of the branched-chain amino acids valine, leucine, and isoleucine and eventually leading to arrested growth and plant death (Duggleby et al. 2008). At present, research on herbicide-resistance mechanisms is mainly concentrated on target-site resistance (TSR) or non–target site resistance (NTSR) (Yu and Powles 2014a). Research on TSR mainly focuses on the activity of the target enzyme (Fang et al. 2019b), the mutation of the target gene (Yu and Powles 2014b), and the change in target gene expression (Gao et al. 2017). Most studies have shown that the ALS activity of resistant weeds with ALS gene mutations is higher than that of susceptible weeds (Fang et al. 2019b; Gao et al. 2023; Li et al. 2017; Panozzo et al. 2013). Herbicide resistance mediated by target gene mutation has been investigated extensively. At present, a total of 31 amino acid substitutions at nine amino acid sites located in the ALS gene have been found in resistant weeds (Fang et al. 2022; Liu et al. 2021). The relationship between the target gene expression and weed resistance remains a mystery for researchers to unravel. The relationship between the accumulation of target proteins and herbicide resistance has been unclear, as well as whether the expression of target genes and the accumulation of target proteins are synchronized. Although antibodies can be used to determine protein accumulation, their application in herbicide-resistance research remains very limited (Chen et al. 2020; Lowe et al. 2024).

In comparison to TSR, NTSR presents greater management challenges, complicates investigative efforts, and necessitates the allocation of additional resources (Fang et al. 2019b). When weeds develop TSR to herbicides, farmers can effectively manage weed populations by employing herbicides with alternative modes of action. (Liu et al. 2021). However, when weeds develop NTSR to herbicides, the development of weed control strategies becomes more complex (Liu et al. 2021). For instance, research on a rice barnyardgrass [Echinochloa phyllopogon (Stapf) Koso-Pol.] population with NTSR has demonstrated that this weed population has developed resistance to penoxsulam, cyhalofop-butyl, and florpyrauxifen-benzyl. A resistant E. crus-galli population has also been found to be resistant to penoxsulam, pinoxaden, and quinclorac. The study of NTSR mainly involves examination of three aspects: enhancement of herbicide metabolism (Délye 2013; Pan et al. 2022), decrease of herbicide penetration and transport (Riar et al. 2013), and alleviation of herbicide-induced oxidative stresses (Pan et al. 2021). Enhanced herbicide metabolism was identified as the most common NTSR mechanism (Yu and Powles 2014a). At present, the genes related to herbicide metabolism have been classified into eight main categories: cytochrome P450 monooxygenase (P450), glutathione S-transferase (GST), ATP-binding box transporter, glycosyltransferase, oxidase, esterase, peroxidase, and hydrolase (Délye 2013). Transcriptome sequencing is often used in the investigation of metabolic resistance. With this technique, researchers can quickly identify the metabolic enzyme genes that may play a key role in herbicide resistance (Pan et al. 2016; Zhao et al. 2022). The metabolic genes that have been characterized are mainly P450 family genes and GST family genes. For example, it was found that CYP81A68 mediates herbicide resistance of E. crus-galli to penoxsulam and other herbicides (Pan et al. 2022); CYP81A12 and CYP81A21 mediates the resistance of E. phyllopogon to bensulfuron-methyl and penoxsulam (Iwakami et al. 2014); and GSTF1 mediates the resistance of blackgrass (Alopecurus myosuroides Huds.) to various herbicides, including chlorotoluron, fenoxaprop-p-ethyl, and clodinafop-propargyl (Cummins et al. 2013).

Therefore, this study aimed to (1) test whether the E. crus-galli population collected from IMI-T rice fields is resistant to imazamox, (2) determine the extent of resistance to other ALS inhibitors and acetyl-CoA carboxylase (ACCase)-inhibiting herbicides in this R population, and (3) understand the TSR and NTSR basis of this imazamox-resistant population. The results of this study may clarify the resistance mechanism of E. crus-galli to imazamox under this IMI-T rice-planting system while providing a new avenue for managing imazamox-resistant weeds.

Materials and Methods

Plant Materials

The JHXY-2 (R) population seeds were collected from IMI-T rice fields (33.01°N, 118.62°E) in the Jiangsu Province of China in 2021, where imazamox has been used for many years. Seeds of the susceptible population JLGY-3 (S) were collected from uncultivated land (34.83°N, 119.12°E) that had never been exposed to herbicides. All seeds were collected by hand, air-dried in the shade, and stored in paper bags at 4 C until use.

Sensitivity to Imazamox and Other Herbicides

Twenty seeds from each of the two populations were sown in plastic pots (9-cm diameter by 10-cm height), which were filled with a 2:1 (w/w) mixture of nutrient matrix and sandy soil. Seedlings were grown in a greenhouse and thinned to 10 plants per pot before herbicide treatment. At the 3- to 4-leaf stage, herbicides were applied using a laboratory sprayer equipped with a flat-fan nozzle, delivering 280 L ha⁻¹ at 230 kPa. The commercial preparation of imazamox (Zhongqi Technology, NanJing, China) was used in this experiment. The subsequent use of imazamox was the same commercial preparation, diluted to the corresponding dose or concentration. Based on a preliminary experiment (data not shown), imazamox was applied at 0, 3.75, 7.5, 15, 30, 60, and 120 g ai ha⁻¹ to the JLGY-3 (S) population. For the JHXY-2 (R) population, the dose of imazamox was set to 30, 60, 120, 240, 480, and 960 g ai ha⁻¹. After 21 d of imazamox treatment, the aboveground parts of the E. crus-galli plants were collected and weighed. Sensitivity to other herbicides was also determined using the same method. The application doses were based on the results of a preliminary experiment (data not shown); detailed dose information is shown in Supplementary Table S1. This experiment was conducted twice in a completely randomized design, with three replications for every dose, and each replication contained 10 plants.

Effect of Metabolic Inhibitors on Imazamox Sensitivity

Two P450 inhibitors (PBO and malathion) and one GST inhibitor (NBD-Cl) were used to evaluate the effect of metabolic inhibitors on resistance level. The applied doses and methods of PBO (4,200 g ai ha⁻¹), malathion (1,000 g ai ha⁻¹), and NBD-Cl (270 g ai ha⁻¹) were previously reported (Gao et al. 2023). PBO and malathion were applied 1 h before herbicide application, and NBD-Cl was applied 48 h before herbicide application. The experiments were performed as described earlier.

ALS Enzyme Activity Assay

According to a method described by Yu et al. (2004), we assessed ALS enzyme responses of S and R populations to imazamox with slight modifications. The stems and leaves (3 g) of E. crus-galli were harvested from each population and powdered in liquid nitrogen. The ground plant tissue was collected into a 50-ml centrifuge tube, 8 ml of enzyme extraction buffer (100 mM potassium phosphate buffer, pH 7.5, 10 mM sodium pyruvate, 0.5 mM MgCl₂, 0.5 mM Thiamine Pyrophosphate (TPP), and 10 μM Flavin Adenine Dinucleotide (FAD)) was added, then shaken for 3 min and left on ice for 15 min. The plant homogenate was filtered with gauze mesh, and the filtrate was centrifuged and treated with ammonium sulfate to obtain protein precipitate. The protein precipitates of S and R populations were dissolved in 4-ml of enzyme assay buffer (100 mM potassium phosphate buffer, pH 7.5, 200 mM sodium pyruvate, 20 mM MgCl₂, 2 mM TPP, 20 μM FAD, and 50 mM HEPES) to obtain crude enzyme solution. Each reaction system consists of 100 μl of crude enzyme solution and 100 μl of ALS inhibitor (imazamox at 0.001, 0.01, 0.1, 1, 10, 100, and 1,000 μM; penoxsulam at 0.001, 0.01, 0.1, 1, 10, 100, and 1,000 μM). The reaction mixture was incubated at 37 C for 60 min. The reaction was terminated by adding 10 μl of 6 N H₂SO₄ at 60 C for 15 min. The color reaction was performed by adding 190 μl of creatine solution (0.55%) and 190 μl of α-naphthol solution (5.5% in 5 N NaOH) at 60C for 15 min. ALS activity was determined colorimetrically (530 nm) by measuring the acetoin production (pure acetoin was used as the standard; Aladdin, Shanghai, China) using a microplate photometer (Thermo Fisher, Waltham, MA). The assay was performed twice with independent extractions and three replicates per herbicide concentration, and each replication contained 100 μl of crude enzyme solution and 100 μl of ALS inhibitor.

Gene Cloning and Sequence Analysis

The DNA of E. crus-galli was extracted using the Plant Genomic DNA Kit (Tiangen Biotech, Beijing, China). The ALS sequences ALS1 and ALS2 of the S population JLGY-3 (S) were described in previous studies (Fang et al. 2019a). Based on the conserved regions of these two sequences, we designed a primer pair (forward: TCTTCGCCTACCCCGGCG; reverse: TCAATACACGGTCCTGCCATCACC) with primer Premier v. 5.0 (Premier Biosoft International, Palo Alto, CA, USA); the amplified ALS gene fragment can cover the nine mutation sites (Ala-122, Pro-197, Ala-205, Phe-206, Asp-376, Arg-377, Trp-574, Ser-653, and Gly-654) reported to cause herbicide resistance (Fang et al. 2022; Liu et al. 2021). The polymerase chain reaction (PCR) mixture contained 20 ng of template DNA, 2 μl of each primer (10 μM), 25 μl of 2× Rapid Taq Master Mix (Vazyme Biotech, Nanjing, China), and ddH₂O to a final volume of 50 μl. The amplification reaction was set as follows: 3 min at 95 C for DNA denaturation; 35 cycles of 15 sec at 95 C for DNA denaturation, 15 sec at 63 C for annealing, and 1 min 10 sec at 72 C for DNA elongation; and a final elongation for 6 min at 72 C. ALS genes from two resistant E. crus-galli were used for gene cloning. Gene cloning was performed using the universal pMD19-T Rapid Cloning Kit (TaKaRa Biotech, Shiga, Japan). Briefly, the ALS gene from two resistant E. crus-galli was amplified utilizing the aforementioned method. The resulting amplified product was subsequently ligated to the pMD19-T cloning vector. The reaction mixture comprised 0.3 pmol of insert DNA, 1 μl of pMD19-T vector, and 5 μl of Solution I (ligase solution). This reaction was incubated at 16 C for 30 minutes before transforming 10 μl of the reaction product into Escherichia coli DH5α. Plates were then incubated overnight, followed by blue-white colony screening on the following day, during which white colonies were selected for further analysis. These bacteria were cultured at 37 C for an additional 6 h, after which bacterial liquid samples were collected for positive identification. Positive samples were subsequently sent for sequencing using universal primers M13F and M13R specific to the vector. At least 14 transformed clones from each plant were selected to obtain ALS sequences, which were aligned and compared using SnapGene v. 6.1 (GSL Biotech LLC, Boston, MA, USA).

Determination of ALS Gene Expression by Real-Time Quantitative Reverse Transcriptase PCR (RT-qPCR)

Plant tissues from S and R populations were collected at five time points (0 h, 12 h, 1 d, 3 d, and 5 d) after imazamox treatment, and three replicates were established for each time point, each containing 0.1 g of leaves. Total RNA of E. crus-galli was extracted using the RNA Simple Total RNA Kit (Tiangen Biotech, Nanjing, China) according to the manufacturer’s instructions. All RNA samples were reverse transcribed into cDNA according to the instructions of the Reverse Transcription Kit (Vazyme Biotech, Nanjing, China). The β-actin gene of E. crus-galli (GenBank accession no.: HQ395760) was used as an internal reference gene, and the expression level of the ALS gene in JLGY-3 (S) and JHXY-2 (R) was determined by RT-qPCR. RT-qPCR analyses were performed on an ABI-7500 Fast Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) using the ChamQ Universal SYBR qPCR Master Mix (Vazyme Biotech, Nanjing, China), following the manufacturer’s instructions. The RT-qPCR primers were sourced from Fang et al. (2022), with the specific primer sequences provided in Supplementary Table S2. The fold changes in gene expression were determined using the 2^−ΔΔCT method. Significant differences in expression levels were analyzed using Welch’s t-test. The entire experiment was repeated twice.

Protein Extraction and Immunoblot Analysis

Samples from S and R populations were collected at two time points (0 h and 1 d) after imazamox treatment, with three replicates at each time point. Leaves of E. crus-galli were ground into powder using a mortar and pestle in liquid nitrogen and then transferred into a 2-ml plastic centrifuge tube. The plant samples were vortexed with 400 μl of protein extraction buffer (10% glycerol, 25 mM Tris, pH 7.5, 1 mM Ethylene Diamine Tetraacetic Acid (EDTA), 150 mM NaCl, 10 mM Dithiothreitol (DTT), 1× protein inhibitor cocktail, 2% Polyvinylpolypyrrolidone (PVPP)) for 2 min. Samples were centrifuged at 12,000 rpm for 5 min at 4 C, and the supernatant was collected. Protein loading buffer was added to the supernatant and the mixtures was boiled for 5 min. The protein samples were separated using 10% SDS-PAGE and then transferred to PVDF membrane. The primary ALS antibody against E. crus-galli ALS was prepared in our previous study (Liu et al. 2024b), and it was applied at a dilution of 1:5,000. Goat anti-mouse immunoglobulin G horseradish peroxidase was applied at a dilution of 1:10,000. The protein strips were exposed using enhanced chemiluminescence, and the gray scale of the bands was quantified using Image J software (NIH, Bethesda, MD, USA).

High-Performance Liquid Chromatography–Tandem Mass Spectrometry (HPLC-MS/MS) Analysis of the Imazamox Residue

When E. crus-galli reached the 3- to 4-leaf stage, each plant was treated with 4 μl of imazamox with a concentration of 0.03675 g ml⁻¹ applied evenly as a liquid on the surfaces of fully opened leaves. The imazamox application dose on each E. crus-galli plant was equivalent to using 120 g ha⁻¹ in the IMI-T rice fields. Samples from S and R populations were collected at three time points (12 h, 24 h, and 48 h) after imazamox treatment, with three replicates set at each time point, and each replicate containing five E. crus-galli plants. When collecting samples, we employed acetonitrile to cleanse the leaf surface, with the aim of eliminating as much of the remaining imazamox on the leaf surface. The samples were prepared with a modified QuEChERS (quick, easy, cheap, effective, rugged, and safe) method according to Yasuor et al. (2009). Leaves of E. crus-galli were ground into powder using a mortar and pestle in liquid nitrogen and then transferred into a 50-ml plastic centrifuge tube. The plant samples were vortexed with 10 ml of acetonitrile for 5 min, then 2.0 g of NaCl and 4.0 g of MgSO₄ were added, with vortexing continued for 30 sec, and samples were then centrifuged at 4,000 rpm for 3 min. The extract (2 ml) was transferred to a purification tube containing 400 mg of C₁₈ and 400 mg of MgSO₄. The tube was shaken for 5 min to complete the purification and centrifuged for 3 min at 12,000 rpm. One milliliter of supernatant was transferred with a syringe and then filtered using a 0.22-μm organic filter membrane. The imazamox amount was determined by HPLC-MS/MS (Agilent 6460 Triple Quad, Agilent, California, USA). Chromatography was performed on a Waters BEH C18 column (1.7 μm, 2.1 mm by 50 mm) (Waters, Milford, Delaware, USA). The injection volume was 1.0 μl. The mobile phase was an acetonitrile/water solution (0.01% formic acid + 5 mM of ammonium acetate). Gradient elution conditions are shown in Supplementary Table S3. Multiple reaction monitoring setting conditions are shown in Supplementary Table S4. The calibration curve of imazamox is provided in Supplementary Table S5. The residual imazamox concentration was determined by substituting the peak area of imazamox detected into the calibration curve. Significant differences in imazamox residue were analyzed using Welch’s t-test.

Sample Preparation for RNA-seq

Echinochloa crus-galli plants were cultivated to the 3- to 4-leaf stage. Plant tissues from S and R populations were collected at 0 h, 12 h, and 24 h after treatment with imazamox (120 g ha⁻¹). The imazamox application method was the same as described earlier. Three biological replicates were set up for each time point, with each replicate containing 0.5 g of leaf tissue. After leaf samples were collected, they were immediately frozen in liquid nitrogen and stored at −80 C until use.

RNA Extraction, cDNA Library Preparation, Transcriptome Sequencing, and Bioinformatics Analysis

Total RNA was extracted from the R and S leaf material by using the RNA Simple Total RNA Kit (Tiangen Biotech) according to the manufacturer’s instructions. cDNA library preparation and transcriptome sequencing were performed as previously reported (Liu et al. 2015; Zhong et al. 2011). Clean reads were mapped to E. crus-galli genome (https://ngdc.cncb.ac.cn/gwh/ncbi_assembly/56777/show) using TopHat2 software (Wu et al. 2022), and only unique mapping reads were retained for calculating gene expression. RNA-seq analysis was performed according to previous protocols (Trapnell et al. 2010, 2013). Contigs were selected based on statistical significance (P < 0.05), the magnitude of expression differences, and annotations related to known herbicide metabolism genes and signaling functions using the E. crus-galli genome. Differentially expressed genes were identified using the package (http://www.r-project.org) with an False Discovery Rate (FDR) < 0.05 and an absolute log₂ ratio value ≥1.

RT-qPCR Validation

RNA samples returned from transcriptome sequencing were reverse transcribed into cDNA using HiScript III QRT SuperMix for qPCR (+gDNA wiper) (Vazyme Biotech); 14 genes from the metabolizing enzyme library (Pan et al. 2016) were selected to design primers for RT-qPCR, and the primer sequences are listed in Supplementary Table S6. RT-qPCR analyses were performed according to Determination of ALS Gene Expression by RT-qPCR. The E. crus-galli β-actin gene was used as the internal control gene. Total RNAs of S and R populations collected at the 0-h, 12-h, and 24-h time points were extracted using RNA Simple Total RNA Kit (Tiangen Biotech), following the manufacturer’s instructions. The expression changes of 14 genes picked from the metabolizing enzyme library were verified by RT-qPCR.

Data Analysis

All whole-plant dose–response data were collected for ANOVA using SPSS v. 21.0 (SPSS Inc., Chicago, IL, USA). The results showed no significant difference between parallel experiments (P > 0.05). Data were processed using SigmaPlot v. 12.5 (Systat, San Jose, CA, USA). In simple terms, a four-parameter nonlinear logistic regression model was used to merge the data to determine the herbicide dose that resulted in a 50% fresh weight loss of weeds:

(1) $$y = c + (d - c)/\left[ {1 + {{\left( {x/g} \right)}^b}} \right]$$

where y is the aboveground fresh weight of weeds, x is the herbicide dose, b is the slope of the curve, c is the lower limit, d is the upper limit, and g is the herbicide dose at the point of inflection halfway between the upper and lower limits. The same methods were used to calculate the herbicide concentration that caused 50% inhibition of ALS activity (IC₅₀). Resistance indexes (RIs) were calculated by dividing the GR₅₀ (or IC₅₀) of the R population by the GR₅₀ (or IC₅₀) of the S population. The classification of resistance level refers to supporting information.

Results and Discussion

Sensitivity to Imazamox

The GR₅₀ of the presumptive resistant population JHXY-2 (R) (166.71 g ha⁻¹) was higher than the recommended application dose (120 g ha⁻¹), while the GR₅₀ of the susceptible population JLGY-3 (S) was only 5.19 g ha⁻¹ (Table 1; Figure 1). Whole-plant bioassays showed that the JHXY-2 (R) population still survived tenaciously imazamox treatment at a much higher dose (960 g ha⁻¹). The RI of the JHXY-2 (R) population to imazamox was up to 32.1.

Table 1. Sensitivities of the resistant and susceptible populations to imazamox with/without three metabolic inhibitors.

^a RI, resistance index.

Figure 1. Fresh weight of Echinochloa crus-galli treated with imazamox. Vertical bars represent the mean ± SE.

Sensitivity to Other Herbicides

After being treated with a variety of herbicides, it was observed that the R population had developed resistance to four ALS inhibitors from different chemical families in comparison to the S population. The RI for flucarbazone-sodium reached 8.5, indicating a medium level of resistance. The other three ALS inhibitors showed RI of 14.9 (penoxsulam), 21.8 (propyrisulfuron), and 18.5 (bispyribac-sodium), which indicated a high level of resistance. However, the R population JHXY-2 (R) did not show any resistance to ACCase inhibitors, with an RI of 1 for metamifop and 1.6 for cyhalofop-butyl (Table 2).

Table 2. Sensitivity of the two Echinochloa crus-galli populations to other herbicides ^a .

^a Abbreviations: ACCase, acetyl-CoA carboxylase; ALS, acetolactate synthase; APP, aryloxy phenoxy propionate; PTB, pyrimidinylthiobenzoate; SCT, sulfonylamino-carbonyl-triazolinone; SU, sulfonylurea; TP, triazolopyrimidine.

^b RI, resistance index.

Sensitivity Change to Imazamox under Pretreatment with Three Metabolic Inhibitors

When three metabolic inhibitors (PBO, malathion, and NBD-Cl) were applied before imazamox treatment, the growth of the S population was hardly affected (Table 1). The GR₅₀ value and RI of the R population increased under treatment with the two P450 inhibitors. Under treatment with the GST inhibitor (NBD-Cl), the GR₅₀ value of the R population increased; however, the RI decreased by 18.06% (32.1 to 26.3).

In Vitro ALS Activity Assay

The sensitivity of the ALS enzyme to imazamox and penoxsulam was assessed in vitro, revealing that the extent of inhibition against the R population was comparatively lower than that observed in the S population (Figure 2). The IC₅₀ value of JLGY-3 (S) was 0.8 μM, while the IC₅₀ value of JHXY-2 (R) was 9.87 μM under treatment with imazamox (Figure 2A). There was a 12.3-fold difference in ALS activity between the two populations. The IC₅₀ values for JLGY-3 (S) and JHXY-2 (R) were 0.61 μM and 4.03 μM, respectively, under treatment with penoxsulam, and there was also a 6.6-fold difference (Figure 2B).

Figure 2. In vitro acetolactate synthase (ALS) activity of two Echinochloa crus-galli populations treated with (A) imazamox and (B) penoxsulam. Vertical bars represent the mean ± SE.

Gene Cloning and Sequence Analysis

The JLGY-3 (S) population sequences have been identified in previous studies (Fang et al. 2019a). Echinochloa crus-galli is a hexaploid weed characterized by multiple copies of the ALS gene. Utilizing cloning sequencing, we acquired a substantial collection of ALS sequences from JHXY-2, subsequently translating these sequences into amino acid formats for comparative analysis. Through the analysis of this substantial number of sequences, we distinctly identified single-nucleotide polymorphisms at several conserved sites across different gene copies and established that the JHXY-2 population possesses ALS gene copies (Supplementary Table S7). Comparative analysis of codons at positions 217, 322, and 441 allows for the differentiation of ALS3 from ALS1 and ALS2. Furthermore, a distinction can be made between ALS1 and ALS2 by examining codons at positions 265 and 441 (Supplementary Table S7). We identified three copies of the ALS gene in the JHXY-2 (R) population, whereas, two copies of the ALS gene were observed in the JLGY-3 population. The observed variation in gene-copy number in the present study is consistent with that previously reported in E. crus-galli, A. aequalis, and Japanese foxtail (Alopecurus japonicus Steud.) (Fang et al. 2019a; Feng et al. 2016; Iwakami et al. 2017). Analysis of the ALS sequence in the R population revealed the same mutation (TGG to TTG) in all three ALS sequences, which resulted in the substitution of Trp with Leu at position 574 (Trp-574-Leu) (Figure 3A). Following our classification, we observed that the Trp-574-Leu mutation can manifest in two copies within an individual plant and may occur in either ALS1 or ALS3 or concurrently in ALS2 and ALS3. In addition, we detected the ALS sequences of the R population in a total of 80 individual plants, of which 64 contained Trp-574-Leu mutations, accounting for 80% (Figure 3B).

Figure 3. Sequence analysis of ALS gene in Echinochloa crus-galli resistant population JHXY-2 (R). (A) Nucleotide sequence alignment of ALS gene fragments in R and S populations. (B) Mutation frequency of the JHXY-2 (R) population.

Gene Expression Level and Protein Abundance of ALS between R and S

The expression levels of ALS genes in the S and R populations were monitored at five time points (0 h, 12 h, 1 d, 3 d, and 5 d) after treatment with imazamox (120 g ha⁻¹) (Figure 4A). It was found that there was no significant difference in ALS gene expression between the two populations, and the variation trend of ALS gene expression in the two populations tended to be consistent within 5 d. In addition, anti-ALS immunoblotting was used to determine whether changes in ALS protein abundance were associated with imazamox resistance (Figure 4B). The results showed that there was no positive correlation among transcription expression, ALS Trp-574-Leu mutation, and ALS protein abundance. Imazamox resistance in JHXY-2 (R) population with the ALS Trp-574-Leu mutation did not relate with changes in ALS protein abundance.

Figure 4. (A) Relative mRNA level of ALS gene in Echinochloa crus-galli JLGY-3 (S) and JHXY-2 (R) populations with imazamox treatment (120 g ha⁻¹). (B) The ALS (69 kDa) abundance in different populations determined by immunoblotting and the grayscale analysis of ALS protein. Protein levels are relative to S-0 h. Vertical bars represent the mean ± SE.

HPLC-MS/MS Analysis of the Imazamox Residue

The residues of imazamox in stems and leaves of E. crus-galli were detected within 48 h after treatment with imazamox. In the S population, the residual amounts of imazamox at 12 h, 24 h, and 48 h were 84.4 μg, 99.8 μg, and 79 μg, respectively, while, the residual amounts were 72.8 μg, 81.6 μg, and 75 μg in the R population, respectively (Figure 5). We found that the amount of imazamox absorbed by E. crus-galli increases over time and may begin to decrease at 24 h. In addition, the imazamox residues of R population were consistently lower than those of the S population. A significant difference was found at 12-h and 24-h time points in the residual amounts of imazamox; thus the possibility of metabolic resistance was suspected.

Figure 5. Imazamox residue in susceptible (S) and resistant (R) Echinochloa crus-galli at 12 h, 24 h, and 48 h after treatment. Means and SEs from three biological replicates are shown. An asterisk (*) indicates significant differences between R and S populations: P-value < 0.05.

RNA-seq Data

The sample in RNA-seq experiment is labeled S/R0/12/24-1/2/3. Eighteen cDNA samples were sequenced using the Illumina sequencing platform (Illumina, San Diego, California, USA), and each sample produced at least 5.4 G of clean data (Supplementary Table S8). The percentage of Q20 exceeded 98.13%, the percentage of Q30 exceeded 94.64%, and the GC (guanine / cytosine) content of each sample varied between 52.44% and 53.65% (Supplementary Table S8). Each population had a different number of single gene clean reads (36756570 to 58280754) (Supplementary Table S9), indicating a high degree of accuracy in sequencing. The read segment comparison tool in Bowtie 2 v. 2.2.8 (Johns Hopkins University, Washington, USA) was used to conduct a comparative analysis of high-quality clean read segments against the ribosomal sequences of the species. After the rRNA read segment was removed, the retained data were used for subsequent analysis.

Identification and Validation of Differentially Expressed P450 and GST Genes Using RNA-seq and RT-qPCR

Transcriptome sequencing was conducted on the two populations, resulting in a total of 110,617 sequences being identified using the Paspalum genome as a reference. Among these were 990 P450 contig sequences and 283 GST contig sequences. Subsequently, genes exhibiting significant differences in expression levels as well as stably expressed genes were selected for validation (refer to Table 3). The findings revealed no significant difference in P450 gene expression between the JLGY-3 (S) and JHXY-2 (R) populations; however, one gene (AH01.132, GSTU1) within the GST gene family exhibited statistically significant upregulation.

Table 3. Identification of the upregulated genes potentially related with herbicide metabolism in Echinochloa crus-galli imazamox-resistant population via RNA-seq and real-time quantitative reverse transcriptase PCR (RT-qPCR).

^a R_T means the R population was treated with imazamox. R_CK means the R population was not treated with imazamox. An asterisk (*) indicates that the gene expression level of the R population was significantly higher than that of the S population: P-value < 0.05.

ALS inhibitors are the most commonly used herbicides in rice fields, but when weeds become resistant to one ALS inhibitor, growers may apply a different ALS inhibitor or herbicides with different modes of action (ACCase inhibitors, 4-hydroxphenylpyruvate dioxygenase inhibitors, or auxin mimic herbicides) (Liu et al. 2021). However, in China, farmers plant IMI-T rice and use imazamox to control weeds in rice fields, which leads to the continuous use of imazamox in IMI-T rice fields. This behavior undoubtedly accelerates the development of weed resistance. In a region of Jiangsu, China, we collected a suspected imazamox-resistant E. crus-galli population JHXY-2 (R) from IMI-T rice fields and determined its resistance to imazamox through whole-plant bioassays. The results showed that the population had a high level of resistance to imazamox, with an RI of 32.1. Based on our research findings, we do not advocate the use of alternative ALS inhibitors in IMI-T rice fields. Given that the weeds exhibiting the Trp-574-Leu mutation demonstrate resistance to all five classes of ALS inhibitors currently employed, it can be inferred that they may possess resistance to all ALS inhibitors. However, employing a rotation or mixture of herbicides with diverse modes of action may improve weed management efficiency.

The mechanisms of herbicide resistance in weeds are currently divided into TSR and NTSR (Yu and Powles 2014a). Many studies have shown that both TSR and NTSR coexist in resistant weeds, so studying either TSR or NTSR alone is not enough to fully understand the causes of herbicide resistance (Yu and Powles 2014a). Therefore, this study investigated both TSR and NTSR mechanisms of imazamox resistance in weeds from IMI-T rice fields. With respect to TSR, after determining the enzyme activity in vitro, we found that the ALS enzyme of the R population with the Trp-574-Leu mutation was less sensitive to imazamox and penoxsulam than that of the S population, resulting in relative resistance ratios of 12.33 and 6.6, which is basically consistent with the previous studies (Palma-Bautista et al. 2020; Panozzo et al. 2013). For example, it has been found that ALS in E. crus-galli is simultaneously less sensitive to penoxsulam and imazamox (Panozzo et al. 2013). Some studies have also found that the ALS of wild poinsettia (Euphorbia heterophylla L.) is less sensitive to imazamox (Palma-Bautista et al. 2020). In previous studies, a total of nine amino acid substitutions have been found to potentially cause herbicide resistance in weeds (Fang et al. 2022; Liu et al. 2021), and the frequency of the Trp-574-Leu substitution is much higher than that of other amino acid substitutions (Heap 2024). E. crus-galli is a hexaploid weed species possessing three copies of the ALS gene (Iwakami et al. 2015). Upon examining the ALS sequences in the R population, we indeed identified three distinct copies of the ALS gene. Furthermore, we observed variations in the number of ALS copies between susceptible and resistant populations. This observation of gene-copy number variation within a single species is not unprecedented; Iwakami et al. (2017) previously reported copy-number diversity among A. aequalis populations exhibiting resistance to thifensulfuron-methyl, with some populations harboring two copies while others contained four, but no differences in resistance levels were observed in these populations. In a polyploid species with multiple ALS copies, “a dilution effect” will occur if only some of the copies carry ALS mutations (Panozzo et al. 2021). However, this phenomenon does not seem to occur in resistant populations. Bioassay experiments indicated that the R population exhibited a higher RI compared with the S population under treatment with imazamox, and ALS enzyme activity assays corroborated this trend.

Changes in the expression level of target genes are often thought to be the cause of herbicide resistance in weeds (Laforest et al. 2017; Ngo et al. 2018). However, it was difficult to find a consistent change in the expression level of the ALS gene in weeds after herbicide treatment in many previous studies (Fang et al. 2020; Laforest et al. 2017; Liu et al. 2022; Ngo et al. 2018). To determine whether there is a difference in the expression of the ALS gene between the R and S populations after imazamox treatment, we monitored the expression of the ALS gene at five time points after imazamox treatment. The results showed that the expression of the ALS gene in the R population was only higher than that of the S population for a short period of time before and after 12 h. The expression of the ALS gene in both R and S populations increased significantly within 1 d after imazamox treatment, and the overall trend was almost the same. We hypothesize that after herbicide treatment, weeds will experience abiotic stresses; this phenomenon likely occurs in all weeds, and there is no difference between the resistant and sensitive populations. The changes in gene expression may not be the cause of herbicide resistance. Nevertheless, the primers employed in RT-qPCR were capable of amplifying all ALS genes from both populations; thus, it remained plausible that a single copy of ALS in the resistant population may exhibit higher expression level compared with others.

At the same time, we monitored the changes of ALS protein abundance with anti-ALS antibody in JHXY-2 (R) and JLGY-3 (S) populations after imazamox treatment, and no significant difference was found. Changes in ALS protein abundance were compared with the changes in ALS gene expression and amino acid mutation, and the results showed no direct correlation. Therefore, we hypothesized that the change in ALS protein abundance is not the cause of the resistance of JHXY-2 (R) to imazamox. Trp-574-Leu substitution did not affect ALS protein accumulation. However, it is still not ruled out that changes in ALS protein abundance in other resistant weeds can lead to resistance, and further studies are needed to clarify this phenomenon (Chen et al. 2020).

In NTSR studies, researchers often used some metabolic enzyme inhibitors to treat weeds, and the change in the RI was used to evaluate whether metabolic resistance was involved (Gao et al. 2023; Pan et al. 2022). A decreased RI may be interpreted as indicating metabolic resistance, and further research may be undertaken. However, there have been few studies on the metabolic resistance of weeds that show no significant change in RI under treatment with metabolic enzyme inhibitors (Liu et al. 2024a). In this study, when S and R populations were treated with malathion, PBO, or NBD-Cl, no significant RI changes were detected in either the S or R populations. Nevertheless, the possibility remains that NTSR may be present in JHXY-2 (R). The results of HPLC-MS/MS showed that the residual amount of imazamox in the leaves of the JHXY-2 (R) population was lower than that of the JLGY-3 (S) population at the 12-h, 24-h, and 48-h time points. We suspect that there may be a small number of metabolic genes playing roles in improving the survival ability of the JHXY-2 (R) population under imazamox application. Transcriptome sequencing is often used to examine weed metabolic resistance, and this technology can identify a large number of differentially expressed genes and metabolic pathways (Dimaano and Iwakami 2021; Pan et al. 2019; Wang et al. 2023). Through transcriptome sequencing, it was discovered that P450 genes CYP99A44 and CYP704A177 are associated with mesosulfuron-methyl resistance in American sloughgrass [Beckmannia syzigachne (Steud.) Fernald] (Bai et al. 2022), while P450 gene CYP81A68 is linked to penoxsulam resistance in E. crus-galli (Pan et al. 2022). Additionally, the GST gene AmGSTF1 is associated with fenoxaprop-p-ethyl resistance in A. myosuroides (Cummins et al. 2013). A total of 990 P450 sequences and 283 GST sequences were identified through transcriptome sequencing. The FPKM (fragments per kilobase of exon model per million mapped fragments) of P450 and GST genes was compared between S and R populations, revealing only a few genes with an expression difference ratio greater than 2. After selecting some genes for expression verification, it was found that only the gene expression of GSTU1 was significantly upregulated in the R population in comparison with S populations. However, considering the differences in residues of imazamox in the leaves of E. crus-galli, there may still be some unidentified metabolic enzyme genes that mediate the metabolism of imazamox by E. crus-galli. Alternatively, there may be differences in the uptake of imazamox between S and R populations, with the R population reducing its uptake of imazamox (Yu et al. 2013).

In conclusion, we found that E. crus-galli collected from IMI-T fields has evolved imazamox resistance. Our study revealed that a Trp-574-Leu mutation in the JHXY-2 (R) population and decreased sensitivity of ALS to imazamox in JHXY-2 (R) are the main resistance mechanisms. Importantly, we further provided key evidence that the Trp-574-Leu mutation did not lead to the change in ALS protein abundance. Our research on NTSR has revealed differences in the residues of imazamox between R and S populations. Identifying new genes that may mediate the metabolism of imazamox in E. crus-galli, and detecting the uptake of imazamox between populations between S and R are worth investigating in future studies to clarify the potential NTSR in resistant E. crus-galli.