The Resistance Levels and Target-Site Based Resistance Mechanisms to Glyphosate in Eleusine indica from China

Abstract

The Dongting and Poyang Lakes are the important rice growing areas, and the Bohai Rim and Loess Plateau are the main producing areas of apples in China, where glyphosate has been used continuously to control weeds including Eleusine. indica for many years. In this study, the resistance levels and target-site based resistance (TSR) mechanisms to glyphosate in E. indica populations, which were collected from above areas were investigated. A total of 35 out of 50 (70%) E. indica populations have evolved resistance to glyphosate with resistance index (RI) of 2.01~10.43. The glyphosate-resistant (GR) E. indica accumulated less shikimic acid than glyphosate-susceptible (GS) populations, when treated by 1.0 mg/L, 10 mg/L or 100 mg/L glyphosate. There was no mutation at Thr102 and Pro106 in 5-enolpyruvate shikimate-3-phosphate synthase (EPSPS), which endowed glyphosate resistance in E. indica and other weed species. A Pro-381-Leu was found in EPSPS in GR populations. In contrast, the expression level of EPSPS gene was highly correlated with glyphosate resistance in E. indica with a determination coefficient of 0.88. These indicate that the glyphosate resistance in aforementioned E. indica populations was mainly caused by the overexpression of EPSPS, not by amino acid mutation in EPSPS.

1. Introduction

Glyphosate is considered a once-in-a-century herbicide, and is used for nonselective controlling of the majority of weeds in agricultural and nonagricultural situations. It kills weeds by inhibiting 5-enolpyruvate shikimate-3-phosphate synthase (EPSPS), which is a key enzyme in the biosynthesis of the aromatic amino acids tryptophan (Try), tyrosine (Tyr) and phenylalanine (Phe) [1]. The glyphosate inhibits EPSPS by competing with phosphoenolpyruvate for the binding site. In addition, the EPSPS only occurs in plants and microorganisms, so glyphosate displays low toxicity to animals [2].

Overreliance on glyphosate has made more and more weeds evolve resistance to glyphosate since the middle of 1990s [3]. The glyphosate resistance was first reported in Lolium rigidum L. from Australia in 1998, which was about 20 years after glyphosate commercialization [4]. To date, at least 56 weed species in 28 different countries have evolved resistance to glyphosate [5]. The glyphosate resistance can be endowed by the target-site based (TSR) mechanisms due to amino acid mutation or overexpression of the EPSPS. The resistance mutation has been documented at the site of Pro106 (changing Pro106 to any of Ala/Leu/Ser/Thr) and Thr102 in EPSPS, which usually causes weeds in field to evolve moderate resistance to glyphosate [6,7]. A double amino acid mutation (Thr-102-Ile + Pro-106-Ser) was identified in glyphosate-resistant (GR) E. indica, which showed high level resistance to glyphosate [8]. Recently, a triple amino acid mutation (Thr-102-Ile + Ala-103-Val+Pro-106-Ser) was found in GR Amaranthus hybridus [9]. In addition, the over-expression of the EPSPS gene was reported in GR L. rigidum and Conyza spp. [10,11], and EPSPS gene amplification was demonstrated in GR Amaranthus palmeri, Amaranthus tuberculatus, Kochia scoparia, Lolium perenne ssp. multiflorum, Chloris truncata and Bromus diandrus et al. [12,13]. In contrast to the TSR mechanisms, the glyphosate resistance can be conferred by multiple non-target-site based (NTSR) mechanisms. For example, reduced glyphosate foliar uptake and/or translocation in GR Sorghum halepense and L. rigidum, vacuolar sequestration and enhanced metabolism by an aldo-keto reductase in GR Echinochloa colona [14,15,16,17,18,19].

Goosegrass (E. indica) is an annual, self-pollinating and diploid grass specie. It is considered one of the “world’s worst weeds” and has been reported to be a problematic weed in a large number of crops all over the world [10]. E. indica has been reported to evolve resistance to the herbicides glyphosate, glufosinate-ammonium, trifuluralin, paraquat, clethodim, cyhalofop-butyl, fluazifop-butyl, haloxyfop-methyl, sethoxydim, imazapyr et al. [5]. Just as in other countries, E. indica is also a notorious weed and widely distributed in both crop and non-crops fields in China. The E. indica also have evolved resistance to various herbicides. For example, the E. indica in direct-seeded rice showed resistance to metamifop, cyhalofop-butyl, sethoxydim, haloxyfop-R-methyl and fenoxaprop-p-ethyl [20]. The E. indica from cotton fields in the Hunan province was reported to develop resistance to quizalofop-p-ethyl [21]. Although the glyphosate has been used to control weeds in orchards, plantations and other non-crop fields for more than forty years, the extensive investigations on glyphosate resistance in E. indica have not been carried out. In this study, the glyphosate resistance level of E. indica infesting farmland around Dongting (DT) and Poyang (PY) lakes, apple fields around Bohai Rim and Loess Plateau was investigated. The shikimic acid accumulation and relative expression level of EPSPS mRNA were determined. In addition, the amino acid mutation of EPSPS in GR and glyphosate-susceptible (GS) populations was identified.

2. Materials and Methods

2.1. Plant Materials

The E. indica seeds were collected from apple orchards in provinces of Beijing (BJ), Hebei (HB), Liaoning (LN) and Shandong (SD) around Bohai Rim; provinces of Henan (HN), Shanxi (SX) and Shannxi (SHX) around Loess Plateau; and from farmlands around Dongting (DT) and Poyang (PY) lakes in China, where glyphosate was used to control E. indica for many years (Figure 1, Table S1).

Seeds were germinated on a filter paper in a Petri dish, then transplanted into pots containing moist loam soil and grown in an artificial climate chamber. The conditions of germination and growth in the chamber were 25 °C/20 °C (light/dark), 16 h photoperiod and luminous intensity of about 15,000~20,000 Lux.

2.2. Petri Dish Dose-Response Bioassay

The test concentration of glyphosate was 0.1, 0.5, 2.5, 12.5, 62.5, 125 or 250 mg/L, respectively. Firstly, 5.0 mL glyphosate solution was added into each Petri dish containing one layer of filter paper. Then, germinated E. indica seeds were placed evenly in each Petri dish, and grown in an artificial climate chamber with the same conditions described in Section 2.1. The root length of each plant was measured 6 days later, and the GR₅₀ value (herbicide rate causing 50% growth reduction of the root lengths) was estimated by logistic regression. Three replications were set for each glyphosate concentration. The experiment was repeated at different times.

2.3. Extraction and Accumulation Assay of Shikimic Acid

The shikimic acid was measured using the leaf disc method described by Mei et al. (2018) [11]. First, the Petri dish was filled with 5 mL NaH₂PO₄ solution (pH 4.4) containing 1.0, 10 or 100 mg/L glyphosate. Next, two leaves with a length of 5 mm were cut from each of the five E. indica plants and immersed into glyphosate solutions. The Petri dishes were transferred into an artificial climate chamber for 16 h under the conditions of 25 °C, continuous 16 h illumination. All leaves were ground in liquid nitrogen, and then 1 mL of 1.25 M HCl was added to extract the shikimate. Finally, the samples were incubated at 37 °C with agitation for 30 min after mixing vigorously for 5 min, and then centrifuged at 12,000 g for 10 min. The supernatants were collected and used for shikimic acid determination.

The reaction mixtures were incubated at 37 °C for 60 min, which contained 250 µL supernatants, 1.0 mL 0.25% (w/v) periodic acid and 0.25% (w/v) sodium metaperiodate. The reaction was stopped by adding 1 mL 0.6 M NaOH and 0.22 M Na₂SO₃. The optical density (OD) at 383 nm was determined colorimetrically using microplate photometer (BioTek, Winooski, VT, USA). Six replicates were set for each glyphosate concentration. The shikimic acid concentration was calculated based on a standard curve. The experiments were repeated four times with independent extracts. The shikimic acid concentration was expressed in µg cm⁻² leaves.

2.4. RNA Extraction and cDNA Synthesis

Pooled samples of six plants at the stage of 6-leaf were used for RNA extraction. Total RNA was extracted according to the instruction of RNApre Pure Plant Kits (Tiangen, Beijing, China). First-strand cDNA was reverse-transcribed using 2 μg total RNA (Tiangen, Beijing, China). The cDNA was synthesized in a 25 µL reaction mixture, which included 1 μL cDNA, 0.5 µL primer pairs, 12.5 µL 2 × PCR Mix and 10.5 μL ddH₂O. The primer pairs (5′-GGTGGATAACCTTTTAAACAGTGAG-3′ and 5′-TTAGTTCTTGACGAAAGTGCTGAG-3′) were designed according the EPSPS sequence of E. indica (HQ403647). The PCR programs consisted of one denaturation step of 3 min at 94 °C, 30 cycles of 30 s at 94 °C, 30 s at 56 °C, 72 °C for 30 s, followed by a final extension step of 5 min at 72 °C.

2.5. Identification of Mutation in EPSPS

The purified PCR product was cloned into the pLB vector (Tiangen, Beijing, China), and then transformed into TOP10 Chemically Competent Cells (Tiangen, Beijing, China). The plasmids containing the insert were extracted and sequenced directly with the primer pairs (5′-CGACTCACTATAGGGAGAGCGGC-3′ and 5′-AAGAACATCGATTTTCCATGGCAG-3′). Six clones were selected for sequencing. Ant potential resistance mutation was identified by comparing the EPSPS sequence with that of wild E. indica specie.

2.6. EPSPS Gene Expression Level

The expression level of the EPSPS gene in E. indica was determined using Super Real Pre Mix Plus (SYBRGreen). The primer pairs of EPSPS (5′-CTGATGGCTGCTCCTTTAGCTC-3′ and 5′-CCCAGCTATCAGAATGCTCTGC-3′) and reference gene β-actin (5′-AACAGGGAGAAGATGACCCAGA-3′ and 5′-GCCCACTAGCGTAAAGGGACAG-3′) were designed. The reaction mixtures contained 10 µL 2 × Super Real Pre Mix Plus, 0.6 µL primers (0.1 μM), 1 µL diluted cDNA (25 ng/μL), 0.4 μL 50 × ROX Reference Dye and 7.4 μL RNase-free ddH₂O. The RT-qPCR program started with an incubation at 95 °C for 15 min, followed by 40 cycles of 95 °C for 10 s, 60 °C for 20 s and 72 °C for 32 s. Under this condition, the internal reference and the target gene displayed high amplification efficiency (92~100%). Three biological replicates and four technical replicates were performed for each E. indica population.

The relative expression ratio (as 2^−△△C_T) was calculated by the comparative CT method, where △C_T = [C_T target gene − C_T internal control gene] and △△C_T = [△C_T resistant − △C_T susceptible] [22].

2.7. Statistical Analysis

Dose-response curves and GR₅₀ (herbicide rate causing 50% growth reduction of plants) were obtained by a non-linear regression using the log-logistic Equation (1) [23]:

$$\mathrm{y}=\mathrm{C} +(\mathrm{D}-\mathrm{C})/{(1+(\mathrm{x}/{\mathrm{GR}}_{50}\left)\right)}^{\mathrm{b}}$$

(1)

where y represents shoot biomass (percentage of control) at herbicide rate x; C and D are the lower and upper limits, respectively; and b is the slope at GR₅₀. The expression difference of the EPSPS gene between GR and GS (BJ05) E. indica populations was analyzed by an independent-sample T test at p = 0.05 or 0.01 significance level.

3. Results

3.1. Glyphosate Resistance of E. Indica

The sensitivity of 50 E. indica populations (34 populations from Bohai Rim and Loess Plateau, and 16 populations Dongting and Poyang lakes) to glyphosate was determined. The results indicated that 35 out of 50 (70%) populations have developed resistance to glyphosate according to the resistance standard of an RI value greater than 2.0. The proportion of E. indica populations with an RI value of 2~4 and 4~10 is 32% and 36%, respectively. One (2%) population (SX03) evolved a high resistance level to glyphosate with an RI value of 10.42. Among them, 81.25% of the populations around Dongting and Poyang Lakes have evolved resistance to glyphosate, while the proportion of GR populations around Dongting (DT) and Poyang (PY) Lakes is 64.7% (Figure 2, Table S2).

3.2. Shikimate Accumulation

In this study, the shikimate accumulation in 27 E. indica populations was tested. The results indicated that the GR populations accumulated less shikimate than GS populations, when E. indica was treated by glyphosate at concentrations of 1.0 mg/L, 10 mg/L or 100 mg/L (

Table 1. Shikimate accumulation in E. indica leaves treated by glyphosate at concentrations of 1.0, 10 and 100 mg/L. Data were sorted from largest to smallest according the shikimic acid concentration treated by 100 mg/L glyphosate.

Populations	Glyphosate (mg/L)
	0 (CK)	1.0		10		100
	Shikimic Acid
	(µg/cm2)	(µg/cm2)	R/S	(µg/cm2)	R/S	(µg/cm2)	R/S
BJ05	18.30 ± 2.39	21.88 ± 1.15	1.00	23.55 ± 1.25	1.00	57.76 ± 7.73	1.00
PY07	14.12 ± 1.93	24.92 ± 1.39 *	1.14	46.96 ± 2.46 ***	1.99	56.23 ± 2.10	0.97
SD04	26.74 ± 1.23	29.02 ± 2.07 **	1.33	36.17 ± 3.57 **	1.54	53.35 ± 3.32	0.92
BJ01	25.91 ± 3.66	31.00 ± 1.74 ***	1.42	36.17 ± 4.27 **	1.54	52.66 ± 3.12	0.91
HN01	11.08 ± 1.17	15.72 ± 1.67 **	0.72	23.09 ± 5.74	0.98	41.56 ± 2.80 *	0.72
HN03	10.78 ± 1.30	19.67 ± 0.82 *	0.9	25.68 ± 2.57	1.09	40.65 ± 1.24 *	0.7
LN02	22.03 ± 2.17	24.01 ± 3.27	1.1	26.51 ± 1.12 *	1.13	40.65 ± 6.08 *	0.7
SHX04	27.12 ± 2.60	31.07 ± 4.30 *	1.42	33.20 ± 0.53 ***	1.41	39.51 ± 3.11 *	0.68
DT01	28.72 ± 3.52	29.02 ± 3.09	1.33	34.04 ± 1.17 ***	1.45	37.23 ± 4.24 **	0.64
SD08	23.02 ± 1.63	25.22 ± 0.63 *	1.15	28.19 ± 1.24 **	1.2	36.40 ± 4.43 **	0.63
DT11	9.26 ± 0.67	12.53 ± 2.14 ***	0.57	13.06 ± 0.87 ***	0.55	34.49 ± 1.65 **	0.6
DT09	14.73 ± 1.30	20.13 ± 2.01	0.92	26.36 ± 2.45	1.12	33.51 ± 2.88 **	0.58
PY08	19.22 ± 0.18	21.42 ± 2.25	0.98	31.30 ± 0.87 ***	1.32	31.91 ± 3.66 **	0.55
SD10	9.64 ± 0.66	13.90 ± 1.38 ***	0.64	15.11 ± 0.96 ***	0.64	30.54 ± 1.91 **	0.53
LN07	18.99 ± 1.71	19.22 ± 1.82 *	0.88	20.05 ± 1.91 *	0.85	27.27 ± 3.62 **	0.47
SD05	17.24 ± 2.37	17.62 ± 3.76	0.81	19.06 ± 3.47	0.81	26.29 ± 2.28 **	0.46
HB08	11.84 ± 0.91	13.59 ± 2.03 ***	0.62	20.74 ± 3.69	0.88	24.16 ± 4.18 **	0.42
SX05	11.61 ± 0.95	13.06 ± 1.90 ***	0.6	14.35 ± 1.35 ***	0.61	22.71 ± 2.64 **	0.39
PY06	8.50 ± 2.05	10.17 ± 1.32 ***	0.46	11.46 ± 0.25 ***	0.49	21.88 ± 6.39 ***	0.38
LN04	10.02 ± 0.67	12.45 ± 1.00 ***	0.57	15.34 ± 0.91 ***	0.65	21.57 ± 1.90 **	0.37
LN05	12.98 ± 0.56	12.98 ± 1.08 ***	0.59	14.73 ± 3.12 **	0.63	19.82 ± 6.56 ***	0.34
SX03	8.95 ± 1.30	9.11 ± 0.38 ***	0.42	9.26 ± 0.67 ***	0.39	19.29 ± 4.24 ***	0.33
DT13	17.32 ± 0.52	17.01 ± 1.00 **	0.78	18.23 ± 1.23 **	0.77	18.76 ± 1.51 **	0.32
PY02	11.08 ± 1.67	12.38 ± 1.14 ***	0.57	16.18 ± 1.83 **	0.69	18.08 ± 1.03 **	0.31
HB03	7.81 ± 0.25	9.26 ± 2.28 ***	0.42	12.68 ± 1.79 ***	0.54	17.01 ± 1.99 **	0.29
SD12	9.94 ± 0.43	11.46 ± 0.96 ***	0.52	11.84 ± 0.91 ***	0.5	15.34 ± 1.83 **	0.27
SHX01	8.35 ± 0.29	10.55 ± 0.25 ***	0.48	10.93 ± 1.54 ***	0.46	11.92 ± 0.72 **	0.21

Data are means ± SE of four replicates of four plants. The *, ** or *** indicated, respectively, the means of shikimic acid are significantly different at p = 0.05, 0.01 or 0.001 significance level compared to that of the susceptible (BJ05) biotype.

). The Pearson coefficient between the GR₅₀ and shikimate concentration was −0.73, −0.74 and −0.69, when the E. indica plants were treated by glyphosate with concentrations of 1.0 mg/L, 10 mg/L or 100 mg/L, respectively. This indicated that the level of shikimate had a significant negative relationship with E. indica resistance to glyphosate. In addition, all E. indica populations can accumulate more shikimate with the increasing of glyphosate concentration (Table 1).

3.3. EPSPS Gene Expression

The results indicated that the expression level of the EPSPS gene in the GR E. indica population of DT11, PY08, SX05, SD05, DT09, LN07, SHX01, HB08, SD12, LN05, DT13, HB03, PY06 and SX03 was significantly higher than the GS population (BJ05). However, the EPSPS expression level of SD04, BJ01, PY07, HN03, LN02, HN01, SHX04, DT01 and SD08 showed no obvious difference with the BJ05 population. The EPSPS expression had a significant positive relationship with the resistance index (RI) of E. indica to glyphosate (R² = 0.88), which demonstrated the EPSPS overexpression conferring the E. indica resistance to glyphosate (Figure 3).

3.4. EPSPS Gene Sequencing

The E. indica populations with different resistance levels to glyphosate from different areas were selected for EPSPS gene sequencing. The results indicated that there was no amino acid mutation at the site of Thr102 and Pro106 in EPSPS, which was reported to endow glyphosate resistance in E. indica and other weed species. It was interesting that a Pro381 Leu mutation was found in GR E. indica populations (HN01, PY08, DT09, SHX01, HB08, DT13, PY06 and SX03), but not in GS populations (SX01, PY07 and BJ04). However, it is not clear whether the Pro381 Leu mutation mediates GR E. indica resistance to glyphosate.

4. Discussion

The whole-plant experiments are usually used for the bioassay of herbicide efficacy to weeds. The results obtained by this method can better reflect the true effects of herbicides in the fields. However, this method is expensive, time-consuming and takes up a lot of space to test a large number of samples. The Petri dish is a simple, quick, reliable and inexpensive method that has been used before to identify GR L. multiflorum and L. rigidum [24,25,26,27]. Hence, the Petri dish bioassay was selected to test the glyphosate resistance of E. indica populations in this study due to huge E. indica populations.

At present, the weed of E. indica, C. canadensis, Digitaria sanguinalis, Amaranthus retroflexus and Alternanthera philoxeroides in China have evolved resistance to glyphosate due to the extensive and intensive use of glyphosate [11,28,29]. Dongting (DT) and Poyang (PY) Lakes are the important planting areas of rice, and the areas around Bohai Rim and Loess Plateau are the main producing areas of apples in China, where glyphosate was widely used for many years. In this study, the glyphosate resistance levels of E. indica populations from the above areas were investigated. The results indicated that 81.25% of populations in farmland around Dongting and Poyang Lakes, and 64.7% of populations in apple orchards around Bohai Rim and Loess Plateau have evolved resistance to glyphosate. Among them, 37.5% and 43.75% populations around DT and PY lakes evolved low (2 ≤ RI < 4) and moderate (4 ≤ RI ≤ 10) resistance levels to glyphosate, respectively. Additionally, 29.4%, 32.4% and 2.9% of populations around Bohai Rim and Loess Plateau evolved low, moderate and high (RI > 10) resistance levels to glyphosate, respectively (Figure 2). Overall, the proportion of GR populations around DT and PY lakes was higher than that around Bohai Rim and Loess Plateau, which may be due to the difference of E. indica samples from two areas. In addition, farmers living around Bohai Rim and Loess Plateau began to stop using glyphosate to control weeds in apple orchards in recent years, because the apple trees can suffer from glyphosate absorbed through exposed roots.

Glyphosate caused the shikimate accumulation in plants by inhibiting EPSPS activities. The GR weeds tend to accumulate less shikimate than GS populations, which was confirmed in various weed species, such as C. truncata, Amaranthus spinosus, L. perenne ssp. Multiflorum, Hordeum glaucum and C. canadensis [11,30,31,32,33]. Hence, the accumulation of shikimate is an important indicator of whether weed species are sensitive to glyphosate. Similar results were also obtained in GR E. indica populations from farmland around Dongting (DT) and Poyang (PY) lakes, apple orchards around areas of Bohai Rim and Loess Plateau. Obviously, the higher the resistance level to glyphosate, the less shikimic acid accumulated (Table 1). It indicated that the GR E. indica populations likely contained one or more TSR mechanisms, such as gene mutation, over expression or amplification. The results of EPSPS gene sequencing showed that there was no amino acid mutation at Thr102 or Pro106 in EPSPS, which was reported to confer glyphosate resistance in E. indica and other weed species [5,6,10]. In contrast, the EPSPS gene overexpressed in most GR E. indica populations, and the expression level of EPSPS was highly correlated with glyphosate resistance levels in E. Indica (Figure 3). Hence, we speculated that the over expression of EPSPS was mainly responsible for the glyphosate resistance of E. indica from areas around Bohai Rim, Loess Plateau, Dongting and Poyang lakes in China. It was interesting that a Pro381 Leu mutation was found in GR E. indica populations (HN01, PY08, DT09, SHX01, HB08, DT13, PY06 and SX03), but not in GS populations (SX01, PY07 and BJ04). The Pro381 Leu was also found in the E. indica population from Malaysia [10]. The X-ray crystallography of E. coli EPSPS, shikimate-3-phosphate and glyphosate ternary complex demonstrated that the residue at position 381 would likely reside on the EPSPS enzyme’s outer surface within a turn between two beta-sheets, which was away from the active site of EPSPS. Thus, it is unlikely to be directly involved in catalysis, substrate binding, or domain closure [2,10]. In addition, EPSPS gene amplification, which is a common mechanism conferring glyphosate resistance, cannot be excluded.

5. Conclusions

The E. indica populations, which infested farmland around Dongting and Poyang Lakes, and apple orchards around Bohai Rim and Loess Plateau, have evolved resistance to glyphosate. Therefore, the glyphosate resistance in E. indica was mainly caused by over expression of the EPSPS gene, but not by gene mutation in EPSPS.