Article Cite This: J. Agric. Food Chem. 2018, 66, 9850−9857
pubs.acs.org/JAFC
Enhanced Herbicide Metabolism and Metabolic Resistance Genes Identified in Tribenuron-Methyl Resistant Myosoton aquaticum L. Shuang Bai,†,∥ Weitang Liu,*,†,∥ Hengzhi Wang,† Ning Zhao,† Sisi Jia,‡ Nan Zou,† Wenlei Guo,§ and Jinxin Wang*,†
Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on September 26, 2018 at 08:19:45 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
Key Laboratory of Pesticide Toxicology and Application Technique, College of Plant Protection, Shandong Agricultural University, Tai’an 271018, Shandong, China ‡ Taian Customs, Tai’an 271000, Shandong, China § Plant Protection Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, P.R. China S Supporting Information *
ABSTRACT: The evolved resistance of Myosoton aquaticum L. to acetolactate synthase (ALS) inhibitors is well established, but most research has focused on target-site resistance, while nontarget-site resistance remains neglected. Here, we investigated mechanisms of the latter. The pretreatment with the P450 inhibitor malathion significantly increased the sensitivity of resistant plants to tribenuron-methyl. The rapid P450-mediated tribenuron-methyl metabolism in resistant plants was confirmed by LCMS/MS analysis. Besides, GST activity was higher among resistant than susceptible individuals. The next transcriptome analysis generated 544,102,236 clean reads from RNA sequencing libraries. De novo assembly yielded 102,529 unigenes with an average length of 866 bp, annotated across seven databases. Digital gene expression selected 25 differentially expressed genes, further validated with qRT-PCR. Three P450 genes, two GST genes, two glucosyltransferase genes, four ABC transporter genes, and four additional contigs were constitutively up-regulated in resistant individuals. Overall, our research confirmed that enhanced herbicide metabolism drives tribenuron-methyl resistance in M. aquaticum. KEYWORDS: abiotic stress, Myosoton aquaticum L., acetolactate synthase, tribenuron-methyl, metabolic resistance, transcriptomics, molecular mechanisms
■
costs linked to NTSR were also reported.13 Nontarget-site resistance to ALS inhibitors even increased fitness in Apera spica-venti.14 A better understanding of NTSR mechanisms on the molecular level would prove beneficial for transgenic breeding of resistant crops (especially the mechanism with no fitness cost), while also allowing the formulation of improved weed control.4 RNA-sequencing (RNA-seq) technology is a powerful tool for investigating the genetic basis of abiotic stress response (including herbicide resistance) in plants, especially among nonmodel species without genomic resources.1,15 Specifically, RNA-seq has clarified some of the genetic differences between herbicide-resistant (R) and susceptible (S) plants, identifying NTSR-related genes in Lolium rigidum Gaudin,16,17 Alopecurus myosuroides Huds.,18 Lolium perenne,19 Beckmannia syzigachne Steud.,20 Descurainia sophia L.,21 and Alopecurus aequalis Sobol.22 These data have revealed that genes involved in metabolic resistance can vary according to plant species and their history of herbicide application.2,4 Widespread in China’s winter fields, Myosoton aquaticum L. is a diploid dicot broadleaf weed that has evolved serious resistance to many ALS inhibitors. Previously, we identified a population (JS17) with high resistance to herbicide tribenur-
INTRODUCTION Evolved herbicide resistance among weeds is an increasingly intractable problem affecting crop production worldwide. Resistance mechanisms fall into two main categories: targetsite (TSR) and nontarget-site (NTSR). The two may coexist in plants. The former results from mutations in a single gene encoding herbicide target protein sites (monogenic resistance), while the latter is any form of polygenic resistance that reduces the amount of herbicide reaching its target site.1−4 One major mechanism of NTSR is metabolic resistance or enhanced rates of herbicide metabolism, often involving the action of cytochrome P450 monooxygenase (P450), ATP-binding cassette (ABC) transporter, glutathione S-transferase (GST), glycosyltransferase (GT), and peroxidase (POD).4,5 Metabolic resistance is a serious threat to herbicide sustainability because the process unpredictably confers tolerance to existing, new, or undiscovered herbicides. Furthermore, this form of resistance is increasing among both monocot and dicot weed species.4,6−9 Despite this danger, potentially coexistent NTSR mechanisms are often neglected after TSR is identified in a given weed. The relative lack of attention is largely due to technical difficulties associated with exploring mechanisms as complex and polygenic as NTSR. Recent research does provide us with some idea of NTSR resistance mechanisms in an increasing number of weed species. Herbicide resistance genes may cause pleiotropic effects on plant growth and fitness.10 While fitness costs associated with P450 action have been observed,11,12 no fitness © 2018 American Chemical Society
Received: Revised: Accepted: Published: 9850
May 25, 2018 August 18, 2018 August 24, 2018 August 24, 2018 DOI: 10.1021/acs.jafc.8b02740 J. Agric. Food Chem. 2018, 66, 9850−9857
Article
Journal of Agricultural and Food Chemistry
LC-MS/MS Analysis of the Tribenuron-Methyl Residue. Technical-grade tribenuron-methyl (95%) was provided by Jiangsu Longdeng Chemistry Company. Micropipettes were used to apply 1.56 μg of tribenuron-methyl S and HR plants at the 3−4 leaf stage (1.56 μg per individual; 150 plants total, divided into three replicates per population). At 1, 3, 5, 7, and 9 d after herbicide treatment, 30 plants each from S and HR were randomly selected for tribenuronmethyl extraction. At 7 and 9 d, tribenuron-methyl extraction was also performed on 30 malathion-and-herbicide-treated plants per population. Tribenuron-methyl was washed off leaves with 30 mL of acetonitrile before ∼0.2 g of pooled samples was ground to powder in liquid nitrogen and transferred into a 50 mL sharp-bottomed centrifuge tube. After adding 10 mL of acetonitrile containing 0.1‰ glacial acetic acid, the tube was shaken for 1 min before 3 g of NaCl was added. The tube was shaken vigorously for another minute before centrifugation at 4000 rpm for 5 min. One milliliter of supernatant was collected in a 2 mL centrifuge tube containing 30 mg of primary secondary amine (PSA) and 50 mg of anhydrous magnesium sulfate (MgSO4), then vibrated for 1 min before a second centrifugation at 13000 rpm for 2 min. The final supernatant (1 mL) was filtered through a 0.22 μm filter into an autosampler vial for LC-MS/MS. A flow rate of 0.3 mL/min at 30 °C was maintained for separation using a C18 colum. Mobile phase A was 0.1% formic acid + 0.01 M ammonium formate/methanol, while mobile phase B was 0.1% formic acid + 0.01 M ammonium formate/water. The gradient elution program (10 min run time) was 0−2 min, linear gradient 10−30% A; 2−9 min, linear gradient 30−100% A; return to initial composition within 30 s; and then equilibration for 2.5 min before the next injection (10 μL). Detection conditions were desolvation gas temperature and flow of 350 °C, respectively; nebulizer gas (N2) pressure of 35.0 psi; capillary voltage of 4000 V; as well as MS 1 and MS 2 heater temperature of 100 °C. Target compounds were quantified using the MS/MS acquisition parameters (MRM mode) described previously.25 Tribenuron-methyl retention time was 8.46 min. For validation of analysis methods, see Table S1. Sample Collection and Preparation for RNA-seq. Leaf samples were taken from tribenuron-methyl-treated and untreated plants 24 h after spraying (11.25 g ai ha−1). Each condition had three biological replicates per population, yielding twelve 10-seedling HR and S combinations (three biological replicates × two treatments × two populations). Collected samples were immediately frozen in liquid nitrogen and stored at −80 °C until RNA extraction. RNA Extraction, Library Construction, Sequencing, and Bioinformatics Analysis. Total RNA extraction and quality control were performed according to standard protocols.22 Annoroad Gene Technology Co. (Beijing, China) constructed the cDNA library and performed Illumina sequencing on a HiSeq 4000 (Illumina, San Diego, CA) to yield 150-bp paired-end reads. Raw reads (transformed by the Consensus Assessment of Sequence and Variation (CASAVA, v.1.8.2)) containing the adapter poly-N (>5% of the unknown sequences designated as “N”) and low-quality reads (>15% bases with quality value 200bp no. of unigenes >500bp maximum length minimum length average unigene length N50 value N90 value total nucleotides of unigenes annotated in NR annotated in NT annotated in KEGG annotated in Uniprot annotated in PFAM annotated in GO annotated in COG annotated in all databases annotated in at least one database
595,393,828 544,102,236 81.62G 102,529 44,712 102,348 201 866 1,642 321 88,809,713 38,204 10,667 16,723 34,129 24,165 32,795 13,505 2,276 44,540
Table 4. Number of Differentially Expressed Genes between Different Treatment Groupsa groups/samples HR_T versus S_T HR_C versus S_C HR_T versus HR_C S_T versus S_C
upregulated number
down-regulated number
total number
5910 1549 59
6134 1629 68
12044 3178 127
5276
4233
9509
a
HR_T, resistant M. aquaticum seedlings sprayed with tribenuronmethyl; S_T, susceptible M. aquaticum seedlings sprayed with tribenuron-methyl; HR_C, resistant M. aquaticum seedlings without tribenuron-methyl; and S_C, susceptible M. aquaticum seedlings without tribenuron-methyl.
transcripts with an average length of 1121 bp (Figure S2). We obtained 102,529 unigenes >200 bp and 44,712 unigenes >500 bp; based on the longest transcript per locus per gene, the mean length was 866 bp, and the N50 length was 1642 bp. GO analysis assigned 32795 unigenes into 62 functional categories: 23 for “biological process,” 19 for “cellular component,” and 20 for “molecular function” (Figure S3). The most enriched subgroups were “cellular process” in “biological process” (20,094 unigenes, 61.27%), “cell parts” in “cellular component” (24,911 unigenes, 75.96%), and “binding” in “molecular function” (19,612 unigenes, 59.80%). Next, KEGG analysis classified 16,723 unigenes into 33 pathways comprising five terms: metabolism (10209 unigenes, 61.05%), organismal systems (10144 unigenes, 60.66%), environmental information processing (8096 unigenes, 48.41%), cellular processes (5751 unigenes, 34.39%), and genetic information processing (4386 unigenes, 26.23%) (Figure S4). These unigenes were also annotated using the following five databases: Nucleotide Sequences (NT), Nonredundant Protein Sequences (NR), Universal Protein Resource (UniProt), Protein Family (PFAM), and Cluster of Orthologous Groups (COG).
Next, GO and KEGG enrichment analyses were performed on annotated DEGs (Tables S5 and S6). Notably, among the DEGs between treated HR and S, 7 and 8 upregulated genes were significantly enriched in the “drug metabolizingcytochrome CytP450” and “metabolism of xenobiotics by CytP450” pathways. Overall, annotations and enrichment analyses suggest that P450 genes and other genes involved in metabolic/signaling pathways are vital to the metabolic resistance of M. aquaticum against tribenuron-methyl, especially given the known presence of such mechanisms in weeds.1,2,4,29 Selection of Candidate Metabolic-Resistance Genes and Their Relative Expression. To identify and validate candidate genes involved in metabolic resistance to tribenuronmethyl, we selected genes upregulated in all experimental groups (treated HR and S) that had similar functional annotations (i.e., encoding P450, GST, GT, and ABC transporters). We also included contigs with annotations 9853
DOI: 10.1021/acs.jafc.8b02740 J. Agric. Food Chem. 2018, 66, 9850−9857
Article
Journal of Agricultural and Food Chemistry
Table 5. Identification of Upregulated Unigenes Linked to Metabolic Resistance in M. aquaticum, Verified Using RNA-seq and qRT-PCR (2−ΔCt)a FoldChage: qRT-PCR (2−ΔCt)
RNA-seq gene ID
PFAM ID
function annotation
c39007_g1 c46918_g1 c35633_g1 c35929_g1 c48123_g1 c23506_g1 c41497_g1 c47413_g1 c46086_g1 c41408_g1 c32104_g1 c32196_g1 c44793_g4
PF00067.17 PF00067.17 PF00067.17
P450, CYP75B2 P450, CYP74A P450, CYP703A2 P450, CYP710A1 P450, CYP71B23 P450, CYP90A1 cytochrome c oxidase GST, MGST1 GST, DHAR2 GST, GSTF11 GT-CSLD5 U-GT78D2 ABC transporter, ABCB19 ABC transporter, ABCB2 ABC transporter, ABCC10 ABC transporter, ABCC3 ABC transporter, ABCG14 ABC transporter, ABCG14 ABC transporter, ABCG8 peroxidase 57 peroxidase 70 peroxidase 72 oxidase oxidase hydrolase_4
PF00067.17 PF00067.17 PF01124.13 PF00043.20 PF03552.9 PF00201.13 PF00664.18
c49741_g6 c50054_g1
PF00005.22
c39205_g1 c35573_g1
PF00664.18 PF01061.19
c35573_g3
PF00005.22
c69288_g1 c42442_g1 c40142_g1 c39115_g1 c23330_g1 c25837_g1 c25979_g1
PF00005.22 PF00141.18 PF00141.18 PF00141.18 PF07732.10 PF03171.15 PF12146.3
FoldChange(HRT_ST)
RNA-seq samples (HRT_ST)c
additional samples (HRT_ST)c
2.69 4.96 5.56 5.02 5.28 3.75 7.49 12.80 7.09 4.20 4.42 3.39 3.24
2.01*d 2.25* 1.98 4.07* 1.84 3.17* 1.87 2.24* 5.67* 2.59* 2.61* 3.48* 2.14*
6.41* 10.51* 0.44 25.85* 1.12 1.23 1.73 3.38* 41.81* 0.51 8.28* 5.91* 7.38*
8.55 × 10−21 1.37 × 10−04
3.37 2.27
1.87 8.41*
1.68 0.28
8.92 × 10−34 3.23 × 10−10
3.83 5.33
2.81* 3.47*
4.78* 28.29*
3.46 × 10−36
8.37
1.84
9.56*
10−04 10−46 10−30 10−28 10−31 10−24 10−22
4.94 4.75 3.87 4.07 3.71 3.35 4.56
2.48* 2.35* 5.27* 1.93 1.99 2.42* 2.75*
18.57* 2.33* 3.40* 15.17* 39.00* 5.78* 5.21*
q-valueb 1.73 1.73 2.47 5.24 2.41 8.11 4.36 8.17 1.07 8.71 6.33 3.48 1.24
5.10 4.64 5.31 5.58 5.58 1.10 5.16
× × × × × × × × × × × × ×
× × × × × × ×
10−03 10−23 10−07 10−06 10−06 10−06 10−03 10−05 10−22 10−08 10−32 10−09 10−84
a
HR_T, resistant M. aquaticum seedlings sprayed with tribenuron-methyl; S_T, susceptible M. aquaticum seedlings sprayed with tribenuronmethyl. bThe resulting p-value was adjusted and expressed as the q-value by the Benjamini-Hochberg procedure for controlling the false discovery rate. cMeans were separated using Fisher’s protected least significant difference (LSD) test at the 5% level of probability (from SPSS analysis). dA P-value of
