Increased Activity of 5-Enolpyruvylshikimate-3-phosphate Synthase (EPSPS) Enzyme Describe the Natural Tolerance of Vulpia myuros to Glyphosate in Comparison with Apera spica-venti

: Rattail fescue ( Vulpia myuros (L.) C.C. Gmel.) is a self-pollinating winter annual grassy weed of winter annual crops. The problems with V. myuros are mostly associated with no-till cropping systems where glyphosate application before sowing or emergence of the crop is the most important control measure. Ineffective V. myuros control has been reported following glyphosate applications. Experiments were performed to study the effectiveness of glyphosate on V. myuros , and determine the causes of the lower performance of glyphosate on V. myuros compared to other grass weeds. Estimated GR 50 values demonstrated that V. myuros was less susceptible to glyphosate than Apera spica-venti regardless of the growth stage. Within each species, glyphosate efﬁcacy at different growth stages was closely related to spray retention. However, the low susceptibility to glyphosate in V. myuros was not caused by lower retention as previously suggested. A signiﬁcantly lower shikimic acid accumulation in V. myuros compared to A. spica-venti was associated with a higher activity of the EPSPS enzyme in V. myuros . Nevertheless, the relative responses in EPSPS activity to different glyphosate concentrations were similar in the two grass species, which indicate that EPSPS from V. myuros is as susceptible to glyphosate as EPSPS from A. spica-venti suggesting no alternation in the binding site of EPSPS. The results from the current study indicate that V. myuros is less susceptible to glyphosate compared to A. spica-venti , and the low susceptibility of V. myuros is caused by an increased EPSPS enzyme activity.


Introduction
Rattail fescue (Vulpia myuros (L.) C.C. Gmel.) is considered a problematic weed in Northern European countries [1]. Since first reported in Denmark in 1990, areas infested with V. myuros have significantly increased [2]. Initially, V. myuros was primarily found in grass seed crops, in particular red fescue; however, with increasing adoption of no-till practices and repeated cropping of winter cereals, V. myuros is now considered a common weed in winter cereal crops as well [3,4].
Vulpia myuros is a winter annual grass weed with a life cycle very similar to winter wheat [5]. Vulpia myuros is a prolific seed producer and can form a large seed-bank in a single season under poor weed management scenarios [6]. A recent study reported up to 50% winter wheat yield losses at a density of 405 V. myuros plants m −2 [7]. V. myuros has a shallow root system rendering it sensitive to soil disturbance. Furthermore, it has short seed-longevity in soil and persist longer in no-till cropping systems where seeds remain on the soil surface [8]. With the wide adoption of no-till practices, glyphosate has become a widely used herbicide for pre-plant and -emergence weed control [9]. However, V. myuros In the summer of 2017, V. myuros seeds were collected from non-agricultural areas at Flakkebjerg, Denmark (55.3 • N, 11.4 • E), where there was no known history of herbicide application. Seeds of susceptible A. spica-venti populations originating from six locations across Denmark were mixed to form a meta-population [21].

Dose-Response Study
Twenty seeds of V. myuros and A. spica-venti were planted in 1-L pots, filled with potting mixture containing soil, peat and sand (2:1:1 by weight). Pots were placed in an unheated glasshouse or on outdoor tables depending on the season in which the experiment was conducted. Groups of pots were sown on different dates to ensure that different growth stages could be sprayed simultaneously. After seedling emergence, plant numbers per pot were thinned to a pre-set number (8 plants per pot). Plants were treated with glyphosate (Roundup Bio, 360 g/L, Monsanto Crop Sciences, Hellerup, Denmark) rates ranging from 11.3 to 720 g ha −1 using a spray cabinet equipped with a boom and two flat-fan nozzle (HARDI ISO F-110-02). The nozzles were operated at a pressure of three bars and velocity of 5.2 km h −1 delivering a spray volume of a 152 L/ha. The dose-response study was repeated three times, plants were sprayed at different plant growth stages in three repeats (Table 1). Experiment 1 was sprayed on 24-01-2019 at growth stages BBCH 11, BBCH 13 and BBCH 21, Experiment 2 was sprayed on 25-04-2019 at BBCH  pots were placed either in an unheated glasshouse (Experiments 1 and 2) or outdoors (Experiment 3). Each experiment was performed with three replications per treatment plus six untreated controls using a complete randomized design. Dry and fresh foliage weights were recorded 4 to 5 weeks after herbicide treatment. Spray retention was measured at three different growth stages using similar experimental conditions as for the dose-response study. Spray retention study was repeated three times. As plants were sprayed at different plant growth stages and under different environment in three repeats, therefore, to make the interpretations of results straightforward terms Experiments 1-3 will be referred to as three runs of the study (Table 1). Two experiments (Experiments 1 and 2) were conducted on the plants grown in the unheated glasshouse and one (Experiment 3) on outdoor grown plants. Experiment 1 was performed as a part of dose-response Experiment 1, while Experiments 2 and 3 were performed as separate experiments. In Experiments 1 and 2, plants were sprayed at BBCH 11, BBCH 13, BBCH 21. In Experiment 3, plants were sprayed at BBCH 22, BBCH 23, BBCH 24. The spray solution consisted of 90 g ha −1 glyphosate in mixture with the fluorescent dye brilliant sulfoflavin at a concentration of 200 g ha −1 . Following spray application, five plants of each plant species were cut at soil level and washed in glass bottles containing 50 mL Milli Q water + 0.2% non-ionic surfactant (Contact, AgroDan, Brabrand, Denmark) and shaken well. A representative sample of the solution was taken for analysis. Within 12 h, post treatment the amount of dye in repetitive samples was measured using a luminescence spectrometer (Perkin Elmer model LS50B). The samples were excited at 420 nm and after excitation emission was measured at 518 nm. The actual amount of deposition on the plants was calculated from a standard curve showing the response of concentrations from 3 to 800 µg per liter of the dye. The equation for the standard curve was linear with R 2 of 0.99. Herbicide treated plant samples (sprayed with glyphosate without the fluorescent dye) and untreated plant samples (not sprayed but washed with demineralized water) were also included for comparison. The leaf area of plant samples was measured using a Licor 3100 area meter and was used for the calculation of per unit tracer deposition. The experiment included 10 replicates per treatment.

Whole Plant Shikimic Acid Accumulation
V. myuros and A. spica-venti plants at BBCH 23 were treated with glyphosate at the rate of 210 g ae ha −1 and 420 g ae ha −1 (Table 1). Application method was the same used in the dose-response experiment. Plant material was harvested 24, 48, 72, and 96 h after treatment, and stored in liquid nitrogen at −20 • C until further analysis. Accumulation of shikimic acid was measured using a method described by Singh and Shaner [22]. A total of 450 mL HCl (0.25 M) was added to 150 mg of plant tissue in a 2 mL Eppendorf tube. The plant material in liquid nitrogen was ground with glass beads using a FastPrep instrument (FastPrep ® FP120, Thermo Savant, CA, USA). Samples were then gently vortexed and centrifuged at 10,606× g for fifteen minutes at 4 • C. Twenty µL of the supernatant was mixed with 0.5 mL of periodic acid (1%) and incubated at room temperature for 3 h. Then, 0.5 mL of 1 M NaOH and 0.3 mL of 0.1 M glycine were added by pipette to the solution. Spectrophotometric reading of 200 µL samples was performed at 380 nm using a 96 well plate (Epoch, BioTek, Winooski, VT, USA). Six replications were used for each glyphosate concentration and species. Plant morphology for V. myuros and A. spica-venti differs [5], which could affect spray deposition. Thus, to account for differences in spray deposition on the two species, herbicide spray deposition was determined at the same growth stage that was used to study shikimic acid accumulation, using the method described above. Results were presented as µmol of shikimate per µg glyphosate per g of fresh weight per cm 2 .

Accumulation of Shikimic Acid in Excised Leaves
Six leaf segments of 5 cm length were harvested from the youngest fully expanded leaves of V. myuros and A. spica-venti plants at BBCH 23 to BBCH 25. Fifty Mg of fresh plant material was transferred into a 2 mL Eppendorf tube containing 1 mM NH 4 H 2 PO 4 (pH 4.4). Glyphosate was added to the Eppendorf tubes at the range of 0, 0.1, 0.5, 2, 5, 10, 100, 200, 400, 500 and 600 µM (Table 1). Samples were incubated in a growth chamber for 24 h at 24/16 • C day/night with 16 h photoperiod with photosynthetic photon Flux density of 850 µmol m −2 s −1 . After 24 h of incubation, the Eppendorf tubes were kept at −20 • C until analysis.
Prior to analysis Eppendorf tubes were thawed at 60 • C for 30 min. Then 250 µL of 1.25 N HCl was added, and incubated at 60 • C for fifteen minutes. Thereafter, a 125 µL aliquot from each Eppendorf tube was transferred into a 2 mL Eppendorf tube, and mixed with 500 µL of periodic acid and sodium metaperiodate (0.25% w/v). Following the incubation at room temperature for 90-min, 500 µL of 0.6 N sodium hydroxide, and 0.22 M sodium sulfite were added to the reaction mixture. The absorbance of samples was measured using a spectrophotometer (Epoch, BioTek, Winooski, VT, USA) at 380 nm. A 96 well plate was used for analysis using 200 µL of each sample. The experiment was performed in three replicates per species and was repeated twice.

EPSPS Enzyme Sensitivity
V. myuros and A. spica-venti plants were established in 1-L pots in an unheated glasshouse. Five g of plant material was collected from the two youngest fully expanded leaves from V. myuros and A. spica-venti plants at the three-tillering stage (BBCH 23) ( Table 1). EPSPS enzyme extraction was conducted using the method described by Sammons and Gaines [14]. Five g of plant material was grounded to a fine powder by pestle and chilled mortar using liquid nitrogen. The powdered plant material was transferred to tubes containing 100 mL cold extraction buffer (100 mM MOPS, 5 mM EDTA, 10% glycerol, 50 mM KCl, and 0.5 mM benzamidine), 70 µL of fresh β-mercaptoethanol, and 1% in polyvinylpolypyrrolidone. Tubes containing samples were continuously stirred and then centrifuged for 40 min (18,000× g) at 4 • C. Thereafter, supernatants were decanted into a beaker through a cheesecloth. Ammonium sulfate ((NH 4 ) 2 SO 4 ) was slowly added to the solution to obtain 45% (w/v) concentration by constant stirring for 30 min and centrifuging at 20,000× g and 4 • C for 30 min. The ammonium sulfate precipitation step was repeated using 80% (w/v) (NH 4 ) 2 SO 4 . The extract was precipitated with gentle stirring, and the precipitate was then collected with centrifugation (20,000× g, at 4 • C for 30 min). Pellets were dissolved in 3 mL of extraction buffer and dialyzed overnight in 2 L of dialysis buffer by 30 mm, 1000-MWC dialysis tubing at 4 • C on stir plate.
The activity of EPSPS from V. myuros and A. spica-venti plants was determined using the protocol described by Dayan et al. [13] with on EnzCheckQR phosphate assay Kit (Invitrogen, Carlsbad, CA, USA). The specific activity of EPSPS was determined in the absence and presence of glyphosate. To determine the inhibition of EPSPS activity, the following glyphosate concentrations were used: 0, 0.1, 10, 100, 1000 µM ( Table 1). The assay buffer contained 1 mM of MgCl 2 , 100 mM of MOPS, 10% glycerol, 2 mM sodium molybdate, and 200 mM of NaF. The assay was performed in three replicates for each species, and experiments were repeated twice. The enzyme activity was calculated to measure the phosphate amount (µmol) liberated per µg of total soluble protein (TSP) per minute.

Statistical Analysis
Glyphosate dose-response, and EPSPS enzyme activity data were analyzed using a log-logistic model [23].
where Y is the response that represents the percent fresh biomass of untreated control or EPSPS enzyme inhibition relative to the control treatment; and c and d are regression parameters representing the equation's lower and upper asymptotes, respectively. Parameter e is the glyphosate rate dose providing 50% reduction in fresh biomass (GR 50 ) or inhibition of EPSPS enzyme activity (I 50 ) (midway between the d, and c parameters), b denotes the slope of the curve around parameter e. The model was tested with lack of fit test (p > 0.05). If the lower parameter value (c) was equal to zero, the four-parameter equation was reduced to a three-parameter equation: Non-linear regressions were performed using R version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria), with drc package [24]. The species were compared in terms of the parameter e (GR 50 ) using post hoc t-tests. Tolerance index (TI) was calculated as V. myuros-to-A. spica-venti GR 50 ratios to compare the responses from a population of V. myuros with a meta-population of A. spica-venti.
Two-way ANOVA was performed to test differences between V. myuros and A. spica-venti, and among studied growth stages with respect to spray retention and accumulation of shikimic acid. Means were compared by Tukey HSD test at p < 0.05. Assumption of normality and homogeneous variance were visually examined. Data analysis was performed using R version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria).

Dose-Response and Spray Retention Study
The estimated GR 50 values showed that V. myuros was less susceptible to glyphosate than A. spica-venti at all growth stages studied ( Figure 1; Table 2). The tolerance index (TI), i.e., GR 50 of V. myuros relative to GR 50 of A. spica-venti indicated that V. myuros required 2.8, 2.0 and 4.0 times higher doses than A. spica-venti for a 50% reduction of biomass at BBCH 11, BBCH 13 and BBCH 21, respectively in Experiment 1. TI's were between 1.3 and 5.4 in Experiments 2 and 3 ( Table 2). The TI was not influenced by the plant growth stage studied but plant growth stage had effect on glyphosate effectiveness. For V. myuros the significant effect of plant growth stage on glyphosate activity was noticed only in Experiment 1, where the estimated GR 50 value was 1.8 and 1.5 fold lower at BBCH 13 and BBCH 21, respectively compared to BBCH 11 (Table 2). Generally, glyphosate activity on A. spica-venti was higher at more advanced than at earlier growth stages in all three experiments (BBCH 13 and BBCH 21 than BBCH 11 in Experiment 1; BBCH 23 compared to BBCH 22 in Experiment 2; and BBCH 26 and BBCH 29 compared BBCH 23 in Experiment 3) ( Table 2).

Spray Retention Assay
There was no treatment-by-experiment interaction between the two glasshouse experiments. Therefore, data from the glasshouse studies were pooled and presented together ( Figure 2), while the results from the experiment conducted outdoors are presented separately. Spray deposition was significantly higher on V. myuros compared to A. spicaventi at all growth stages. For instance, in the glasshouse experiment spray deposition was 3.6, 6.4 and 5.5 times higher on V. myuros than on A. spica-venti at BBCH 11, BBCH 13 and BBCH 21, respectively. Under outdoor conditions, spray retention was 3.6, 2.3 and 2.0 times higher on V. myuros compared to A. spica-venti at BBCH 22, BBCH 23 and BBCH 24 stage, respectively. Plant growth stage also had a significant effect on spray deposition. For instance, in the glasshouse study (Experiments 1 and 2) spray retention on V. myuros at BBCH 11 stage was significantly lower than at BBCH 13 and BBCH 21, which may explain the lower susceptibility at BBCH 11 compared to the later growth stages ( Table 2). A similar trend was observed on A. spica-venti; however, significant differences were only detected between BBCH 11 and BBCH 21. In contrast, under outdoor conditions (Experiment 3) spray retention on V. myuros tended to decline with increasing growth stages. For example, the spray retention was significantly higher at BBCH 22 stage than at BBCH 23 and BBCH 24, while no difference was observed between BBCH 23 and 24.

Spray Retention Assay
There was no treatment-by-experiment interaction between the two glasshouse experiments. Therefore, data from the glasshouse studies were pooled and presented together (Figure 2), while the results from the experiment conducted outdoors are presented separately. Spray deposition was significantly higher on V. myuros compared to A. spica-venti at all growth stages. For instance, in the glasshouse experiment spray deposition was 3.6, 6.4 and 5.5 times higher on V. myuros than on A. spica-venti at BBCH 11, BBCH 13 and BBCH 21, respectively. Under outdoor conditions, spray retention was 3.6, 2.3 and 2.0 times higher on V. myuros compared to A. spica-venti at BBCH 22, BBCH 23 and BBCH 24 stage, respectively.

Shikimic Acid Accumulation Assays
In the whole plant assay, the shikimate concentration in plant tissue of V. myuros and A. spica-venti increased over time after glyphosate application. The increase in shikimate concentration was significantly higher in A. spica-venti than in V. myuros. At 96 h after glyphosate application, A. spica-venti accumulated approximately 3 and 5 times more shikimic acid than V. myuros following treatment with 210 and 420 g ha −1 (Figure 3).
Similar findings were observed in the assay using leaf segments (Figure 4). At low glyphosate concentrations (0.1 and 0.5 μmol), shikimate accumulation in V. myuros and A. spica-venti was similar but with increasing glyphosate concentration shikimate accumula- Plant growth stage also had a significant effect on spray deposition. For instance, in the glasshouse study (Experiments 1 and 2) spray retention on V. myuros at BBCH 11 stage was significantly lower than at BBCH 13 and BBCH 21, which may explain the lower susceptibility at BBCH 11 compared to the later growth stages ( Table 2). A similar trend was observed on A. spica-venti; however, significant differences were only detected between BBCH 11 and BBCH 21. In contrast, under outdoor conditions (Experiment 3) spray retention on V. myuros tended to decline with increasing growth stages. For example, the spray retention was significantly higher at BBCH 22 stage than at BBCH 23 and BBCH 24, while no difference was observed between BBCH 23 and 24.

Shikimic Acid Accumulation Assays
In the whole plant assay, the shikimate concentration in plant tissue of V. myuros and A. spica-venti increased over time after glyphosate application. The increase in shikimate concentration was significantly higher in A. spica-venti than in V. myuros. At 96 h after glyphosate application, A. spica-venti accumulated approximately 3 and 5 times more shikimic acid than V. myuros following treatment with 210 and 420 g ha −1 (Figure 3).
Agriculture 2021, 11, x 10 of 16 tion was higher in A. spica-venti than in V. myuros. Depending on glyphosate concentration, V. myuros accumulated 1.2-2.5 fold less shikimate than A. spica-venti. Findings from both the whole plant and leaf-segment shikimic acid bioassays were consistent and showed that the susceptible A. spica-venti accumulates more shikimic acid than the tolerant V. myuros when exposed to glyphosate.   Similar findings were observed in the assay using leaf segments (Figure 4). At low glyphosate concentrations (0.1 and 0.5 µmol), shikimate accumulation in V. myuros and A. spica-venti was similar but with increasing glyphosate concentration shikimate accumulation was higher in A. spica-venti than in V. myuros. Depending on glyphosate concentration, V. myuros accumulated 1.2-2.5 fold less shikimate than A. spica-venti. Findings from both the whole plant and leaf-segment shikimic acid bioassays were consistent and showed that the susceptible A. spica-venti accumulates more shikimic acid than the tolerant V. myuros when exposed to glyphosate.

EPSPS Enzyme Sensitivity
A log-logistic equation was fitted to the EPSPS enzyme activity data for V. myuros, and A. spica-venti, and regression parameters were estimated ( Figure 5; Table 3). The regression parameter values for the slope (b), and lower asymptote (c) were not significantly different between the two species studied. In contrast, a comparison of the estimates for the upper asymptote (d), reflecting EPSPS enzyme activity in the absence of glyphosate (baseline activity) revealed a significant difference between the two species (p < 0.001). EPSPS baseline activity for V. myuros was 1.3 fold higher than for A. spica-venti. The glyphosate rates needed to inhibit EPSPS enzyme activity by 50% (I 50 ) were 264 ± 190 µM and 81 ± 75 µM for V. myuros and A. spica-venti, respectively (p = 0.56) ( Table 3).

EPSPS Enzyme Sensitivity
A log-logistic equation was fitted to the EPSPS enzyme activity data for V. myuros, and A. spica-venti, and regression parameters were estimated ( Figure 5; Table 3). The regression parameter values for the slope (b), and lower asymptote (c) were not significantly different between the two species studied. In contrast, a comparison of the estimates for the upper asymptote (d), reflecting EPSPS enzyme activity in the absence of glyphosate (baseline activity) revealed a significant difference between the two species (p < 0.001). EPSPS baseline activity for V. myuros was 1.3 fold higher than for A. spica-venti. The glyphosate rates needed to inhibit EPSPS enzyme activity by 50% (I50) were 264 ± 190 μM and 81 ± 75 μM for V. myuros and A. spica-venti, respectively (p = 0.56) ( Table 3).

Discussion
V. myuros management strategies in no-till systems primarily relies on pre-sowing Figure 5. EPSPS activity of Vulpia myuros and Apera spica-venti exposed to different concentrations of glyphosate. Table 3. Parameter estimates of log-logistic equations (1 or 2) applied to determine the sensitivity of EPSPS enzyme activity to glyphosate in leaf extracts of Vulpia myuros and Apera spica-venti. Standard errors are presented in the parenthesis.  (2)) were compared by t-tests at the 5% level of significance between Vulpia myuros and Apera spica-venti.

Discussion
V. myuros management strategies in no-till systems primarily relies on pre-sowing glyphosate applications [4]. Farmers have reported that glyphosate is only marginally effective in controlling V. myuros, in contrast to other grass weeds, such as A. spica-venti. Among scientists and advisors, it is often assumed that ineffective control of V. myuros with glyphosate can be attributed to low spray retention due to its narrow and erect leaves [4,10,11]. To elucidate the cause(s) of the higher tolerance of V. myuros to glyphosate in comparison to other common winter annual grass weed species, we conducted a series of studies on a V. myuros population that has never been exposed to glyphosate and compared the responses to a those of a meta-population of A. spica-venti.

Dose-Response Study
Results obtained in dose-response studies showed that the GR 50 values for V. myuros were 2 to 3 times higher than for A. spica-venti, irrespective of growth stage ( Figure S1). However, the actual doses needed to control the two grass weed species were lower than the recommended field doses, which illustrates that pot-grown grass weeds are generally more susceptible than plants in the field. Interestingly, V. myuros susceptibility to glyphosate was consistently lower than A. spica-venti in three experiments. Field studies have shown that glyphosate is not highly effective in controlling V. myuros and that the doses needed for satisfactory control are often higher than the recommended doses [20,25]. Yu et al. [17] studying the closely related species V. bromoides, proved its tolerance to ACCase and several ALS inhibitors and concluded that an insensitive ACCase and enhanced metabolism were the likely mechanisms of tolerance to the two modes of action. Hull et al. [3] confirmed that V. myuros was tolerant to the two herbicide groups. Here, we report for the first time that V. myuros is also tolerant to glyphosate.

Spray Retention Study
Previous studies have shown that variation in plant architecture and leaf characteristics can contribute to differences in herbicide deposition on treated plants [26,27]. In V. myuros, the presence of several pubescent veins on the leaf surface and rough leaf margins may result in a higher herbicide retention than on the light and hairless leaves of A. spica-venti [28,29]. Differences in the chemical composition and/or structure of the cuticle could also influence spray retention [30]. In contrast to many studies relating the poor performance of glyphosate on V. myuros to low spay retention [4,10,11,19], this study showed that reduced spray retention is not the cause of the low performance of glyphosate on V. myuros.
In the current study, no correlation was found between spray deposition and herbicide efficacy on the two grass weed species. In contrast, a close relationship was found between glyphosate activity and spray retention comparing growth stages of the same species. For instance, in Experiment 1, V. myuros control with glyphosate was low if sprayed at the early plant growth stage (BBCH 11) while control was higher when sprayed at later growth stages (BBCH 13 and BBCH 21), and a similar trend was observed for A. spica-venti in three experiments. Overall, our results are in line with those previously reported for V. myuros, and other grass weeds in the literature [11,31,32]. In contrast to dicot weeds, grass weeds, due to lower spray retention on the more erect leaves, are generally less susceptible to foliar applied herbicides at very early growth stages than at later growth stages [32]. Ball et al. [11] suggested that the erect leaf orientation of young V. myuros seedlings are limiting spray coverage on the leaf surface. The higher retention at later growth stages resulted in a better control than at earlier growth stages in Experiments 1 and 2 ( Figure 2; Table 2). Similar to our findings, Koo et al. [33] also found a close relationship between foliar retention of pyribenzoxim and its activity on Echinochloa crus-galli.
It can be argued that dose-response and spray retention experiments were carried out at different growth stages and under different growing conditions, but it is important to emphasize here that the primary aim of these experiments was to study the level and causes(s) of V. myuros tolerance to glyphosate. V. myuros susceptibility to glyphosate was lower than of A. spica-venti, and the reason of low susceptibility to glyphosate in V. myuros was not caused by lower retention, as previously suggested. Interestingly, these results were consistent across experiments that were carried out at different growth stages and growing conditions.

Shikimic Acid Accumulation Assays
Inhibition of the EPSPS enzyme is the mechanism of glyphosate action in plants, which results in the accumulation of shikimic acid in glyphosate sensitive weed plants [15,34]. Shikimate accumulation and EPSPS enzyme activity analysis are considered appropriate parameters to determine tolerance to glyphosate [13]. The findings from the whole plant and leaf-segment shikimic acid bioassays suggest that glyphosate incurs lower inhibition of EPSPS in V. myuros than in A. spica-venti. The somewhat smaller difference between the two plant species using leaf segments, compared to whole plants, is likely due to the fact that the exposure to glyphosate was the same in contrast to the whole plant assay, where more glyphosate was retained on the V. myuros than on the A. spica-venti plants. The presence of 200 µM or more of glyphosate did not significantly increase the accumulation of shikimic acid in V. myuros. In contrast, in the case of A. spica-venti, there was a significant increase in the accumulation of shikimic acid even at concentration of 500 µM. In glyphosate-resistant species, the lower accumulation of shikimic acid can be explained by either a functional feedback control mechanism of the shikimic pathway by the precursor that regulates the 3-deoxy-d-arabino-heptulosonate-7-phosphate (DAHP) synthase activity, and/or a lower interaction with the EPSPS enzyme by glyphosate [35]. The former mechanism prevents further accumulation of shikimic acid and limits the reaction rate early in the pathway, and this scenario can explain the results observed in the leaf-segment shikimic acid bioassays. In vitro methods measuring shikimic acid accumulation have been widely used for detecting glyphosate resistance in plants, where leaf segments are immersed in a glyphosate medium and then incubated [36]. This method has proven to be an effective and quick way to evaluate the differences among glyphosate-resistant and susceptible plants [37]. In our study, the whole plant shikimic acid assay also provided evidence that the interaction of glyphosate with EPSPS was lower in V. myuros than in A. spica-venti.

EPSPS Enzyme Sensitivity
Several studies reported high activity of EPSPS as a mechanism of glyphosate resistance [15,34,38]. The larger concentration of EPSPS enzyme in the total protein extracted from sample tissue as a whole/or per unit fresh weight limits glyphosate activity and prevents blocking of the shikimic pathway. Previous studies reported a strong relation between baseline EPSPS enzyme activity and gene copy number [34,39]. Higher baseline EPSPS activity could contribute to V. myuros tolerance to glyphosate. Although results from the current study have shown an association between glyphosate tolerance and higher EPSPS enzyme activity in V. myuros, it remains to be understood why the higher spray deposition on V. myuros did not overcome the effect of the difference in EPSPS activity.
Genome duplication might contribute to gene multiplication, which can lead to the evolution of genes with modified functions [34,40]. The presence of multiple genomes (polyploidy) can provide novel traits to plants [40,41]. For instance, Bunnell et al. [42] has reported that tetraploid individuals of bahiagrass (Paspalum notatum) were tolerant to metsulfuron while diploid individuals were susceptible. It is possible that a higher baseline EPSPS activity could be attributed to the higher genome size of the hexaploid V. myuros (2n = 42) with potentially multiple functional EPSPS alleles compared to the diploid A. spica-venti (2n = 14). Our study measured the EPSPS enzyme activity at several glyphosate doses in order to determine if a target-site mechanism may have caused the low sensitivity of V. myuros [34]. The I 50 values for A. spica-venti were lower but not statistically different from V. myuros (Table 3), indicating that EPSPS from V. myuros is as susceptible to glyphosate as EPSPS from A. spica-venti suggesting no alternation in the binding site of EPSPS [15]. According to Salas et al. [34] the amino acids present in the catalytic site of EPSPS are very conserved, hence, target-site mutation in plants from natural populations is very rare compared to other herbicide target sites. As there is no evidence of differences in susceptibility within populations of V. myuros to glyphosate (Kudsk, pers. comm.), it is very unlikely that any selection for higher EPSPS activity has occurred in natural populations of this species. This assumption is further supported by the fact that V. myuros is a new weed species that has been intensively exposed to glyphosate for only a few years. Hence, we suggest that the higher level of EPSPS activity is an intrinsic property of V. myuros that might be due to the higher ploidy level, and it is less likely that any resistance mechanisms, such as target-site mutation or reduced absorption, translocation, metabolism and vacuolar sequestration, is the cause of low glyphosate susceptibility in V. myuros.

Conclusions
The current study confirmed that V. myuros is more tolerant than other grass species to glyphosate as it is to other herbicide modes of action and it showed that the tolerance is not attributed to low spray retention as previously anticipated. The EPSPS activity in V. myuros was elevated compared to A. spica-venti and could explain the lower accumulation of shikimic acid in V. myuros and the observed difference in the susceptibility between the two grass weed species. Despite innate tolerance of V. myuros to glyphosate, the relative difference in the accumulation of shikimic acid and the estimated GR 50 values between the two species suggest that V. myuros can be controlled by glyphosate but its susceptibility is lower compared to other grass weeds, such as A. spica-venti. Because of its tolerance to many selective graminicides, the chemical control of V. myuros largely depends on glyphosate. Overuse of glyphosate, due to a lack of other chemical options, may trigger the evolution of resistance to glyphosate in V. myuros. To avoid this, the use of glyphosate should be combined with other control strategies, such as a diversified crop rotation and cultivation where possible [42,43].