Different Non-Target Site Mechanisms Endow Different Glyphosate Susceptibility in Avena Species from Spain

Abstract

In recent decades, cereal agriculture across main producing areas in Spain has progressively adopted direct sowing, associated with an increased use of herbicides in pre-sowing. Weedy species from genus Avena have been observed after the application of glyphosate in wheat. Here, Avena fatua (two accessions), Avena byzantina and Avena sterilis subsp. sterilis, one accession each, were taxonomically characterized by a biometric study. Dose-response trials confirmed that one A. fatua accession evolved to resistance, because it was four times more resistant (R) than the others, ascribed as susceptible (S). In addition, based on LD₅₀, A. byzantina and A. sterilis had low susceptibility to glyphosate, with 604 and 458 g ae ha⁻¹. Shikimic acid accumulation was able to discriminate between high susceptible (A. fatua (S)), low susceptible (A. byzantina and A. sterilis), and low resistant (A. fatua (R)) populations. On the other hand, the study revealed that A. fatua (R), A. byzantina and A. sterilis had low foliar uptake and decreased movement of glyphosate. In addition, the metabolism study showed less metabolite accumulation in A. byzantina and A. fatua (S). However, at 96 h after glyphosate application, A. fatua (R) and A. sterilis were able to convert more than 30% of glyphosate to aminomethylphosphonic acid (AMPA) and sarcosine. Enzyme basal activity and I₅₀ values predicted high affinity between the herbicide and the target enzyme for all accessions, thus indicating that a target-site (TS) mechanism is probably not involved in the differences in glyphosate susceptibility. In closing, non-target site (NTS) mechanisms could participate both in A. fatua R to glyphosate, and low susceptibility in A. byzantina and A. sterilis from Spain.

1. Introduction

Weeds are an important biological constraint to the production of grain crops [1]. In the Western Mediterranean, Avena sterilis L. subsp. sterilis and Avena fatua L. are among the most important grass weeds, causing revenue losses to grain growers [2]. A recent study reported that 15 to 16 plants/m² of Avena reduced wheat yield by 50% [3]. In Spain, direct sowing has been commonly used in annual crops for more than 20 years, with a concomitant increased use of herbicides [4]. Compared to traditional intensive tillage, this technique switches mechanical operations for herbicide application, mainly glyphosate, to control early germinating weeds that are already resistant to acetolactate synthase (ALS) and Acetyl CoA carboxylase (ACCase) herbicides (i.e., Phalaris spp. and Avena spp.) [5,6].

The controversial glyphosate herbicide interrupt biosynthesis inhibition of three essential amino acids (phenylalanine, tyrosine and tryptophan) by inhibiting the enzyme 5-enolpyruvylshikimate-3 phosphate synthase (EPSPS) [7]. Unlearned lessons by farmers in integrated weed management (IWM) have led to extensive use of glyphosate without rotation of herbicides and/or crops. As a result of this incorrect practice, the selection of tolerant/resistant weeds has been favored [8]. Because of the different agronomic practices implemented in cropping systems, there are two major problems: (a) the composition of the weed flora has progressively changed not only in annual crops managed by direct sowing, but also in Mediterranean perennial crops such as olive groves, fruit orchards and vineyards [9,10]; (b) on the other hand, direct sowing in annual crops could also increase the residuality of herbicides in the soil [11,12].

The term “tolerance” has been defined by the Weed Science Society of America (1998) as the inherent capacity of a species to survive and reproduce after herbicide treatment at field doses. This means that no preselection of herbicide led to low sensibility; therefore, it is naturally an innate tolerant accession. Tolerance to herbicides (in this case to glyphosate) could result from different morphological features or physiological processes, collectively known as non-target site (NTS) mechanisms. These mechanisms participate in reducing or preventing the accumulation of the herbicide in the target meristematic tissues [13]. NTS mechanisms include limited absorption or translocation of the herbicide and/or its sequestration into cellular organelles such as vacuoles. Plant metabolic pathways capable of ending herbicide detoxification also belong to this important group of mechanisms [14,15].

To support the development of IWM strategies in Mediterranean annual crops relying on recurrent herbicide use, the objectives of this study were (a) to determine the level of susceptibility to glyphosate in four accessions of the main weedy Avena species sampled in non-cultivated areas and crops of northern Spain and (b) to elucidate the resistance mechanisms involved in these accessions.

2. Materials and Methods

2.1. Plant Collection

Glyphosate susceptibility level was evaluated in three different Avena species, including two main weeds of winter annual crops in Mediterranean Spain: two A. fatua accessions (one putative resistant (R) and one putative susceptible (S)) and one Avena sterilis subsp. sterilis population (hereafter A. sterilis). In addition, a cultivated A. byzantina C. Koch accession, widely used in Spain as livestock feed, was included in the study. In June 2020, seeds of two A. fatua and one A. sterilis accessions were collected in northeastern Spain. Avena fatua (R) was collected under cereal field (41.9102772, 1.005761); A. fatua (S) from a field margin (41.864485, 1.007657); and A. sterilis was collected around the University of Lleida (non-cultivated habitats). Seeds of all accessions were preserved in a laboratory at 4 °C. Seeds of A. byzantina were supplied by the repository of the Department of Agricultural Chemistry, Edaphology and Microbiology of the University of Cordoba.

Seeds of the different accessions were germinated in Petri dishes containing filter paper moistened with distilled water. The Petri dishes were placed in a growth chamber at 28/18 °C (day/night) with a photoperiod of 16 h, 350 μmol s⁻¹ of photosynthetic photon flux and 80% relative humidity. After three days, the resulting seedlings were placed in pots (one plant per pot) containing sand/peat in a ratio of 1:1 (v/v), and the pots were taken into a greenhouse under the same temperature and photoperiod conditions.

2.2. Biometric Study and Taxonomic Identify

Seven to eight plants per population were grown until flowering to confirm species identity. A biometric study was conducted on spikelet and floret diagnostic traits, including lower and upper glume length, number of fertile florets per spikelet, form of disarticulation of the rhachilla at fruit maturity and, for the lemma of the basal floret, length, distribution of hairs, presence of apical awns, and total length and length of the column (basal twisting portion) of the dorsal awn. Traits were measured in one randomly selected spikelet per plant.

2.3. Dose-Response Assays

Glyphosate dose-response assays were conducted to determine the susceptibility level of weedy and cultivated Avena species. At third to four leaf stage (BBCH 13–14), the accessions were treated with glyphosate using a laboratory sprayer system (SBS-060 De Vries Manufacturing, Hollandale, MN, USA) equipped with 8002 flat fan nozzles delivering 250 L ha⁻¹ at 50 cm height above plants. Glyphosate (Roundup Ultimate^® SL, 480 g ae L⁻¹ as isopropyl amine salt, Monsanto) doses applied (10 plants per dose) were: 32.25, 62.5, 125, 250, 375, 500, 625, 750, 1000 and 1500 g ae ha⁻¹. Untreated plants were used as control. Plant mortality was recorded after 28 days to determine the effective mean dose suppressing 50% of accession (LD₅₀). In addition, the plants were harvested and subsequently dried (60 °C for two weeks) to determine the effective mean dose reducing growth by 50% (GR₅₀). Data were expressed as percentage in relation to the untreated controls.

2.4. Spray Retention Assays

To determine the amount herbicide retained on leaf surfaces, six plants of each accession of Avena were assayed in a completely randomized design. The plants at BBCH 13–14 stage were sprayed with 360 g ae ha⁻¹ of glyphosate plus 100 mg L⁻¹ Na-fluorescein in the spraying system described in Section 2.3. Subsequently, we followed the methodology described by Yanniccari et al. [16]. Between 1.5 and 2 h after treatment (HAT), Avena plants were cut at ground level, and the tissue was submerged in test tubes with 50 mL of 5 mM NaOH for 30 s to remove the spray solution. The washing solution was recovered in glass flasks. Subsequently, plants were placed on labelled filter paper strips, oven dried at 60 °C for 48 h and weighed. Fluorescein absorbance was determined using a spectrofluorometer (Hitachi F-2500, Tokyo, Japan) at 490 nm wavelength and an absorbance at 510 nm wavelength. The experiment was repeated, and the results, expressed as µL of sprayed solution retained per g dry weight, were combined for statistical analysis.

2.5. Shikimate Accumulation Assay

For each Avena accession, 25 fully extended young leaves (BBCH 13–14) were taken from at least 10 plants and 4-mm diameter discs were cut. Samples of 50 mg leaf discs (pool foliar tissue) were placed in 2-mL tubes containing 1 mL of 10-mM ammonium phosphate monobasic solution (pH adjusted to 4.4 with 0.1 HCl) at different concentrations of glyphosate (0, 0.1, 0.2, 0.4, 0.6, 0.8 and 1 mM). The extraction and quantification of shikimic acid was performed according to Vázquez-García et al. [17]. Three samples of each Avena accession were assessed per herbicide concentration. The experiment was completely randomized and was repeated. Results were expressed as micrograms of shikimate per mL of HCl solution (µg shikimic acid mL⁻¹) in relation to a calibration curve of shikimic acid.

2.6. Glyphosate Absorption and Mobilization Assays

Ten plants at the BBCH-13,14 stage of each Avena accession were used for ¹⁴C-glyphosate absorption and mobilization assays, whereas another five plants at the same stage were used for a digital visualization assay. Both experiments were arranged in a completely random design.

A mix (mix A) with labeled glyphosate (¹⁴C-glyphosate (glycine-2-¹⁴C, 95% radiochemical purity, 273 Mbq mmol⁻¹ specific activity, Institute of Isotopes Co., Ltd., Budapest, Hungary) plus a commercial formula (Roundup ultimate 48%, Monsanto, Spain) was prepared with a final specific activity of 100,000 dpm μL⁻¹ and 360 g ae ha⁻¹. Thereafter, a drop (1 μL) of mix A was applied on the second leaf of each plant.

A second mix (mix B) of distilled water plus acetone (1:1 v/v) was prepared to wash the non-absorbed ¹⁴C-glyphosate at 48 and 96 h after treatment (HAT), and this procedure was repeated three times per sample (1 mL of mix B per time). The treated leaves of five plants from each population were washed at each evaluation time. The wash solution was recovered in scintillation vials, and 2 mL of scintillation cocktail was added and analyzed by liquid scintillation spectrophotometry (LSS). ¹⁴C-glyphosate absorption was expressed as a percentage of recovered radioactivity using the following equation: % of ¹⁴C-herbicide absorbed = (¹⁴C in combusted tissue/(¹⁴C in combusted tissue + ¹⁴C in shoot washes)).

Prewashed plants used to determine absorption were removed from the pots, and the root system was washed immediately to remove the substrate. Whole plants were fractioned into treated leaf, shoots and roots, saved in filter paper cones, dried in a stove for two days at 60 °C for 96 h, and thereafter samples were combusted in a sample oxidizer (Packard 307), for 180 s. Evolved ¹⁴CO₂ was trapped and counted in Carbosob/Permafluor E+ (3/7 v/v) (Packard Instruments Co., Downers Grove, IL, USA). The radioactivity of plant tissues was quantified by LSS (10 min). Mobilization percentages were calculated according to Vázquez-García et al. [17].

To achieve digital visualization of ¹⁴C-glyphosate movement through the plant, Avena plants were treated with mix “A” and washed with mix “B”, as described above. Whole plants, fixed on filter paper, were oven-dried for four days at 60 °C. Then, the plants were pressed, placed adjacent to phosphor storage film (25 cm × 12.5 cm) and scanned (three hours after) for herbicide movement visualization in a phosphor imager (Cyclone, Perkin-Elmer, Packard Bioscience BV, Waltham, MA, USA).

2.7. Glyphosate Metabolism

The study of glyphosate metabolism in Avena plants comprised three relevant steps: herbicide application, sample harvest and quantification of herbicide and metabolites.

The first step was performed under the same conditions described in the dose-response curves section. A glyphosate dose of 360 g ae ha⁻¹ was applied over ten plants of each accession at the BBCH-14 growth stage. Ten additional untreated plants were saved and used as blanks. At 96 HAT, the plants were cut at ground level and washed with distilled water to eliminate herbicide excess from leaf surfaces and substrate residues from roots. Thereafter, the plants were dried with paper towel for three minutes and frozen by liquid nitrogen (N₂). Treated and untreated plant samples were stored in −80 °C until metabolite analysis. Amino methyl phosphonic acid (AMPA), glyoxylate, and sarcosine extraction was performed according to Rojano-Delgado et al. [18]. Calibrations curves were obtained using metabolite standards (Sigma-Aldrich, St. Louis, MI, USA). This study was arranged in a completely randomized design, with ten replications (one replica-one plant) per Avena accession. Data were expressed as percentages of the sum of glyphosate plus metabolites recovered.

2.8. Activity of Target Enzyme of Glyphosate

The EPSPS activity was assayed in the four Avena accessions. Five grams of young foliar tissue (BBCH-13,14) was collected and immediately macerated in a mortar until obtaining a fine powder. Then, the samples were stored at −40 °C until analyses. The extraction of the target enzyme of the glyphosate, as well as the determination of the total soluble protein (TPS, basal activity without glyphosate) and the EPSPS inhibition rate by adding increased concentrations of glyphosate (0.1, 1, 10, 100, and 1000 μM) were performed following the detailed methodology by Dayan et al. [19]. For each glyphosate concentration, three technical replicates of each accession were assayed. The experiment was repeated (finally n = 6 for statistical analysis), and the results were given as a percentage relative to the control (0 μM glyphosate) of the amount (μmol) of inorganic phosphate (Pi) released per μg of TSP min⁻¹ (μmol Pi μg⁻¹ TSP min⁻¹).

2.9. Statistical Study

For dose response data, the three-parameter non-linear function, y = d/{1 + exp[b(log x − log e)]}, was fitted using the “drc” package for the R environment according Ritz et al. [20] to estimate the glyphosate dose necessary to reduce plant growth, lead plant mortality, or inhibit enzyme activity by 50% for each Avena accession (GR₅₀, LD₅₀, and I₅₀, respectively). In the above function, y is percentage of growth, survival or enzyme inhibition, parameter b controls the steepness of the curve, d is the upper limit of y, and e is GR₅₀, LD₅₀ or I₅₀ values. A resistance index (Ri) was calculated for each accession as the quotient of its GR₅₀/LD₅₀/I₅₀ value and the corresponding values of the less tolerant/susceptible accession.

For foliar retention, shikimic acid accumulation, herbicide absorption and movement, and metabolism data, normal error distribution and the homogeneity of the variance were graphically verified for each set. Then, ANOVAs were performed to test for significance effects. Post hoc multiple comparisons to separate means were performed using the Tukey’s HSD test.

3. Results

3.1. Biometric Study and Taxonomic Identify

Disarticulation of the rhachilla between florets leading to independent dispersal of caryopses separates A. fatua from the other two Avena species. One-awned lemmas separate A. fatua from Avena barbata Pott. Ex Link, a species in the flora of the Iberian Peninsula, which also exhibits the above trait. Avena sterilis can be easily separated from co-generic species based on its large-sized spikelets and florets, large dorsal awns in lemmas, and rhachillas disarticulating only above the glumes, i.e., with the caryopses produced within a spikelet dispersing as a unit. On the other hand, traits identifying the domesticated species A. byzantina include lack of disarticulation of the rhachilla at fruit maturity and dorsal awns in lemmas lacking column, i.e., the basal twisting portion of the awn (Figure 1,

Table 1. Biometric analysis of diagnostic traits on plants individually grown in pots of the four Avena accession studied. Mean values (in cm) and standard deviations are offered for continuous quantitative traits. One spikelet per plant was randomly selected. The number of plants sampled per accession is indicated between parentheses.

	Avena fatua (R) (n = 7)	Avena byzantina (n = 7)	Avena sterilis (n = 7)	Avena fatua (S) (n = 8)
Spikelet
Lower glume length	2.5 ± 0.12	2.4 ± 0.08	3.1 ± 0.16	2.5 ± 0.16
Upper glume length	2.6 ± 0.08	2.7 ± 0.13	3.1 ± 0.08	2.6 ± 0.18
Number of fertile florets	2	2	2	2–3
Disarticulation of the rhachilla at maturity	between florets	no	below the basal floret only	between florets
Lemma of the basal floret
Length	1.6 ± 0.06	1.6 ± 0.07	2.5 ± 0.11	1.8 ± 0.14
Hairiness	lower half	only at floret callus	lower half	lower half
Apical awns	absent	absent	absent	absent
Dorsal awn, length of the basal twisting portion (column)	1.4 ± 0.12	without a column	2.0 ± 0.14	1.5 ± 0.14
Dorsal awn, total length	3.6 ± 0.33	2.5 ± 0.14	5.7 ± 0.41	3.9 ± 0.20

n = number of individuals.

).

3.2. Dose-Response Assay

All Avena accessions exhibited differential susceptibility to glyphosate (Figure 2). Avena fatua labelled as R showed the highest Ri and GR₅₀ values (4.2 and 560.5 g ae ha⁻¹), followed by A. byzantina (Ri = 3.4, GR₅₀ = 467.5 g ae ha⁻¹), and the A. sterilis accession (Ri = 2.4, GR₅₀ = 288.3 g ae ha⁻¹), which was considered as low susceptible compared to the accession of A. fatua labelled as S (117.1 g ae ha⁻¹). Only the A. fatua R population had some plants (around 15%) surviving the field dose of glyphosate (1080 g ae ha⁻¹), while all Avena accessions were controlled at 1.5x the field dose of glyphosate. Compared to the A. fatua (S) accession, A. fatua R, A. byzantina and A. sterilis accessions required a 2–3.5 fold higher dose for achieving 50% mortality (

Table 2. Parameters of the log-logistic regression to estimate the mean dose (g ae ha⁻¹) of glyphosate required to reduce the dry weight (GR₅₀) and plant mortality (LD₅₀) by 50% in different Avena spp.

Accession	b	d	GR50	95% CI		Ri a
				Lower	Upper
A. fatua (R)	3.1	93.1	560.5	525.5	595.6	4.2
A. byzantina	3.5	93.3	467.5	439.7	495.2	3.4
A. sterilis	2.5	97.4	314.2	288.3	340.0	2.4
A. fatua (S)	2.9	100.5	130.5	117.1	143.8	-
			LD50
A. fatua (R)	6.5	97.2	788.8	764.9	812.8	3.5
A. byzantina	5.4	95.7	604.0	582.8	625.3	2.7
A. sterilis	3.9	99.0	458.2	437.5	479.0	2
A. fatua (S)	3.1	98.0	222.1	201.5	242.74	-

^a Resistance index (GR₅₀ Avena spp./GR₅₀ A. fatua (S)).

).

3.3. Herbicide Retention

Foliar retention ranged from 641.5 to 729.8 µL of glyphosate solution per g dry weight among Avena accessions. Although the A. fatua populations retained ~80–100 µL less than A. byzantina and A. sterilis, there were no differences in leaf retention (Figure 3).

3.4. Shikimic Acid Accumulation

Shikimic acid accumulation increased as glyphosate concentration increased and differed among Avena accessions. Avena fatua (R) and A. byzantina exhibited similarly low levels of shikimate (20–31.5 and 18.5–36.3 µg shikimic acid, respectively). Avena sterilis showed intermediate levels (70.8), whereas the A. fatua (S) accession accumulated the highest amounts of shikimate (161.4) (Figure 4).

3.5. ¹⁴C-Glyphosate Absorption, Mobilization and Visualization

Different patterns of absorption and mobilization of ¹⁴C-glyphosate were observed at both evaluation times (48 and 96 HAT) (Figure 5). Avena fatua (S) absorbed ~60% of applied ¹⁴C-herbicide at 48 HAT, whereas the remaining Avena accessions absorbed less than 28%. At 96 HAT, ¹⁴C-glyphosate absorption increased between 20 and 30% in all biotypes. In addition, the amount of herbicide moved from the treated leaf to the shoots and roots was higher in A. fatua (S) and A. byzantine, while A. sterilis and A. fatua (R) kept most of ¹⁴C-herbicide restricted to the treated leaf, i.e., low amounts of ¹⁴C-glyphosate were moved to shoots and roots in these accessions (Figure 5). The qualitative results obtained in the autoradiographs corroborated these patterns of ¹⁴C-glyphosate mobilization within Avena plants of different accession (Figure 6).

3.6. Metabolism Study

Presence of glyphosate metabolites was confirmed in plants of the four accessions. In plants of A. fatua (R) and A. sterilis, the extent of metabolization of glyphosate to AMPA and sarcosine at 96 HAT was much higher (35–45% of herbicide was metabolized). This resulted in higher percentages (~90%) of unmetabolized glyphosate in A. fatua (S) and A. byzantina plants (Figure 7).

3.7. Activity of Glyphosate Target Enzyme

The basal activity of EPSPS (in the absence of glyphosate) was similar among Avena populations/species, ranging from 0.048, for A. fatua (S), to 0.055 μmol μg⁻¹ protein min⁻¹, for A. byzantina (Figure 8A). In addition, the EPSPS enzyme activity was inhibited by glyphosate in a similar proportion among populations (Figure 8B). The I₅₀ values ranged from 30.7 to 43.7 μM glyphosate. Although the lowest I₅₀ was found for A. sterilis, to determine the Ris the I₅₀, A. fatua (S) was used; thus the Ris varied from 0.8 to 1.2 (

Table 3. Parameter values and resistance index of the log-logistic function used to model enzyme activity (EPSPS) as a function of glyphosate concentration in the four Avena accessions.

Accession	b	d	I50	95% CI		Ri a
				Lower	Upper
A. fatua (R)	0.9	96.7	43.7	34.0	53.4	1.19
A. byzantina	0.6	99.9	37.7	26.3	49.0	1.03
A. sterilis	0.8	100.2	30.7	26.8	44.1	0.84
A. fatua (S)	0.8	94.5	36.5	23.7	37.7	-

^a Resistance index (I₅₀ Avena spp./I₅₀ A. fatua (S)).

).

4. Discussion

4.1. Glyphosate-Resistance Confirmation

The four Avena accessions characterized in this study showed different levels of susceptibility to glyphosate, as follows (from lowest to highest): A. fatua (R) > A. byzantina > A. sterilis > A. fatua (S). The differences between the two A. fatua accessions demonstrated an evolution towards a low-level resistance in the R one, while what happened between the three Avena species can be explained in terms of inherent differential susceptibility. It is important to consider herbicide tolerance for these cases, because it is necessary to increase the glyphosate doses to kill 50% of the population. On the other hand, it is also true that some studies demonstrate that natural tolerance to glyphosate can evolve into resistance in few years if the required selection pressure is exerted on field populations [21,22]. Differences between other Avena spp. accessions were found in Australia; however, resistance mechanisms were not characterized [3]. Furthermore, the levels of resistance to glyphosate in A. fatua (R), 789 g ae ha⁻¹ and 561 g ae ha⁻¹ (for LD₅₀ and GR₅₀, respectively), were higher than those reported in Australia (556 g ae ha⁻¹ and g ae ha⁻¹), representing a 42% and 60% increase of LD₅₀ and GR₅₀, respectively. In addition, the low glyphosate susceptibility level found in A. byzantina and A. sterilis suggests that both species have a certain level of natural tolerance to the herbicide, which could evolve into resistance in coming years, as found in previous works [3,8].

EPSPS is the target enzyme of glyphosate that blocks the shikimate pathway [7]; therefore, the presence of glyphosate causes an increase of shikimic acid. The low shikimic acid contents reveal no or low interaction between enzyme/herbicide, i.e., the shikimate pathway is not affected and vice versa. This parameter is used as an early indicator of interaction of glyphosate with its target enzyme EPSPS [23,24]. Here, based on shikimic acid accumulation (at 48 and 96 HAT), we characterized one accession susceptible (A. fatua (S)) and one resistant (A. fatua (R)), as well as two species with low glyphosate susceptibility (A. byzantina and A. sterilis). Even if the separation of resistant and susceptible accessions was performed, this result cannot by itself define the mechanisms involved in the response of our Avena spp. [25].

4.2. Unraveling Resistance Mechanisms to Glyphosate

For glyphosate to be toxic in plants, it must be present in the cytoplasm and in the growing apical meristems, so mechanisms that reduce the amount of glyphosate to a sublethal level are likely to confer resistance [26]. A lower foliar retention of glyphosate solution is the first event that affects the herbicide effectiveness, because it can reduce the amount that is subsequently absorbed by the plant [27,28]. Here, we provide evidence of no differential foliar retention of glyphosate solution among the Avena spp. tested, possibly due to the similarity of the cuticle of the leaves between species/accessions [8,29].

The limited absorption and impaired translocation of herbicide may be involved in different susceptibility of glyphosate in plants [30]. These mechanisms have been identified as responsible for glyphosate resistance in several grass weeds [17,27,31], as well as natural tolerance in different plant species (i.e., Ribeiro et al. [32]). The limited uptake could be due to poor penetration of glyphosate through plant cuticles [29]. In our work, we can separate the tested populations into two groups: the first one corresponds to A. fatua (R), A. byzantina and A. sterilis, and the second group to A. fatua (S), which absorbed more glyphosate. The impaired translocation generally prevents the movement of glyphosate within the plant from the treated leaves towards growing apical meristems, as corroborated in digital visualization of plants treated with ¹⁴C-glyphosate, showing a small amount of herbicide translocated from 48 to 96 HAT (Figure 6).

An interesting discovery in our work was the ability of A. fatua (R) and A. sterilis to detoxify glyphosate. Previously, A. sterilis from Andalusia was also able to support and metabolize sublethal doses of glyphosate [8]. Our results show that this characteristic is being distributed throughout the Iberian Peninsula in other populations/species of Avena. In addition, this is the first work with evidence of enhanced metabolism contributing to glyphosate resistance in A. fatua. Different enzyme(s) have been characterized by metabolizing glyphosate: glyphosate oxidoreductase (GOX), aldo-keto reductase (AKR) and Gly oxidase in microbes and plants, both broadleaf and grass weeds [24,26,33,34,35]. Our results confirm that A. fatua (R) and A. sterilis can degrade glyphosate into AMPA, glyoxylate and sarcosine; nevertheless, further studies need to be carried out to identify which enzymes could be involved in the detoxification.

Resistance based on target-site alterations resulting in reduced sensitivity to glyphosate could be a point mutation or EPSPS overexpression [36]. We discarded the presence of target-site resistance mechanisms, based on both the basal enzyme activity and the inhibition rate of EPSPS with glyphosate, which were both similar among the four Avena spp. accessions. Previous studies have shown that when TSR mechanisms to glyphosate are not present, differences in basal enzyme activity or I₅₀ are not expected, both in broadleaf grass weeds [37,38]; however, further studies are necessary to confirm this in our accessions.

5. Conclusions

The results of this work demonstrate different NTS resistance mechanisms governing Avena species. First, the low susceptibility in A. fatua (R) and A. sterilis is due to a detoxification of glyphosate to AMPA, glyoxylate and sarcosine. In addition, reduced uptake and mobilization were confirmed in these two accessions. The different responses between A. fatua (S) and (R) accessions confirm the first case of glyphosate resistance evolution of this species occurring in Spain and Europe. Regarding A. byzantina accession, this work confirms that the limited uptake and low mobilization endow it low susceptibility to glyphosate. In short, chemical fallow and glyphosate in pre-seeding is an important tool in arable crops for weed control; however, if they are not properly managed, weed accession with high resistance to glyphosate or more modes of action could evolve.