Target Site Resistance to Acetolactate Synthase Inhibitors in Diplotaxis erucoides and Erucaria hispanica –Mechanism of Resistance and Response to Alternative Herbicides

: Diplotaxis erucoides and Erucaria hispanica are common weeds of the Mediterranean region; they infest various habitats including cultivated ﬁelds and roadsides. In several ﬁelds across Israel, farmers have reported on poor control of D. erucoides and E. hispanica plants using acetolactate synthase (ALS) inhibitors. Greenhouse experiments were conducted to determine the e ﬀ ect of various ALS inhibitors on plants from two potentially resistant D. erucoides and E. hispanica populations. Additionally, alternative management strategies using auxinic herbicides were studied. Plants from both populations exhibited resistance to all tested ALS inhibitors, up to 20-fold the label ﬁeld rate, as compared with ALS sensitive populations of D. erucoides and E. hispanica . Sequencing of the ALS gene revealed Trp574 to Leu substitution in ALS-resistant D. erucoides plants, whereas a Pro197 to Ser substitution was detected in ALS-resistant E. hispanica plants. Although high levels of resistance were observed in individuals from both putative resistant populations, sensitive individuals were also detected, suggesting the evolution of resistance in these two populations is still in progress. Auxinic herbicides, 2,4-D, and mecoprop-P, provided excellent control of plants from both ALS-resistant populations. This study documents and conﬁrms the ﬁrst case of evolution of resistance to ALS inhibitors in D. erucoides and E. hispanica populations.


Introduction
Diplotaxis erucoides (white wall-rocket) and Erucaria hispanica (Spanish pink mustard) are common weeds of the Brassicaceae family spread across the Mediterranean region [1]. These species are frequently found as native weeds mostly in Portugal, Spain, Italy, Cyprus, Greece, Turkey, and Israel; recently, they had also invaded as alien species other countries, such as Romania, Norway, Switzerland, and Slovakia [2]. High germination rate and reproductive success may contribute to the high abundance of these two species [3,4]. Furthermore, both species have high seed fecundity with reproduction mechanisms that prevent self-pollination [5,6]. These species can germinate in several flushes in a season, enabling them to better compete with crop seedlings, dominate the entire field, and cause high yield losses (B Rubin, personal communication). The fact that these weed species are exogamous (i.e., they rely mainly on outcross-pollination) may enhance the spread of beneficial adaptive traits, especially if they are inherited as a single dominant gene, such as in the case of target-site resistance [7].

Plant Material
Seeds of two potentially resistant populations of D. erucoides (DER) and E. hispanica (EHR) were collected at Kibbutz Dvir (31 • 24 45" N 34 • 49 30" E) and Kibbutz Beeri (31 • 25 26" N 34 • 29 28" E), respectively. Samples were collected from fields (20-30 hectares) where farmers reported on weed control failure using ALS inhibitors. Seeds of sensitive populations of D. erucoides (DES) and E. hispanica (EHS) were collected from an uncultivated area where no herbicides have been used at Kibbutz Re'em (31 • 23 08" N 34 • 27 34" E). To ensure appropriate representation of field population, mature seedpods from 30 to 40 randomly selected plants in each population were collected and pooled. Collected seeds were air-dried and stored at 4 • C until used. Seeds from each population were germinated in flats filled with commercial potting media (Tuff, Marom Golan, Israel), including Osmocote®(The Scotts Company, Marysville, OH, USA) slow release fertilizer. Seedlings of D. erucoides and E. hispanica populations at the fully developed cotyledons stage were transplanted into pots, 7 by 7 by 6 cm (one plant per pot), filled with the same potting media. Pots were kept in a greenhouse (22/16 • C, day⁄night), at natural light conditions, and watered daily.

Plant Responses to ALS Herbicides
At the stage of three to four true leaves, plants were treated with 0, 1/32, 1/16, 1/8, 1/4 1/2, 1, 2, or 4 times the label field rate of each ALS inhibiting herbicides using a chain-driven sprayer delivering 300 L ha −1 , with a flat-fan 8001E nozzle (TeeJet®, Spraying Systems Co., Wheaton, IL, USA). Different ALS inhibitors (commercial formulations) and their label field rate are detailed in Table 1. The level of resistance in both DER and EHR populations was determined in a separate experiment when plants from all populations were treated with high rates of 10 and 20 times the label field rate of each herbicide as previously described. All experiments were arranged in a completely randomized-factorial design, with three to five replicates for each treatment, and repeated three times. No significant treatment by experimental run were observed; therefore, the data obtained from the repeated experiments for each herbicide were pooled. Shoot fresh weight (FW) and survival rate (alive or dead) were recorded 21 days after treatment (DAT).

Plant Responses to Auxinic Herbicides
At the stage of three to four true leaves, plants from both DER and EHR populations were treated with two different auxinic herbicides, as detailed in Table 1. Each of the auxinic herbicides was applied at the following rates: 1/4, 1/2, 1, 2, or 4 times the label field rate as described above. The experiment was arranged in a completely randomized-factorial design, with three to five replicates of each treatment and was repeated three times. No significant treatment by experimental run were observed; therefore, the data obtained from the repeated experiments for each herbicide were pooled. Shoot FW and survival rate (alive or dead) were recorded 21 DAT.

DNA Extraction and Molecular Studies
Leaf tissue (3 cm 2 ) was excised from individual plants of all four populations (DER, EHR, DES, and EHS). Each sample was placed separately in a microtube. DNA was extracted using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN, USA) according to the manufacturer's instructions and diluted 10-fold before further use. Specific primers were used to identify the ALS gene and locate common point mutations previously shown to endow resistance to ALS inhibitors. Primers were designed according to the known sequences of the ALS genes from Arabidopsis thaliana (X51514) and Lolium rigidum (DQ184640.1) ( Table 2). Sequence analyses and alignment were performed using the Bioedit software [20]. The obtained sequences were compared to the known sequence of the ALS gene of A. thaliana (X51514).

Statistical Analyses
Data of FW were analyzed using ANOVA in JMP (ver. 13) statistical package (SAS Institute Inc., Cary, NC, USA) as the percentage of FW reduction compared to the untreated control. Means were compared using Student's t-test (α = 0.05). Dose-response curve experiments were conducted to evaluate the effect of different herbicides on the average shoot FW and survival rate of plants from each population. For each treatment, data is presented and plotted as a percentage of untreated control. Data were fit to a nonlinear sigmoidal logistic three-parameter model [21] as described in Equation (1): In this model, if b > 0, then a describes the upper limit of Y; X 0 describes the herbicide rate causing 50% shoot FW reduction (ED 50 ) or 50% mortality (LD 50 ), and b describes the slope of the curve. For each species, the resistance index (RI) was calculated as the R: S (resistant: susceptible) ratio.

Response to ALS Iinhibitors
Although they are very common in the Mediterranean region, apart from this case, herbicide resistance has never before been reported for both D. erucoides or E. hispanica [9]. Plants from the DER population exhibited resistance in response to imazamox, propoxycarbazone-sodium, and tribenuron-methyl (RI > 10). EHR plants, were highly resistant to all tested ALS inhibitors except imazamox (Table 3; Figures 1 and 2). A wide range of herbicide responses among DER and EHR populations were recorded, including highly resistant, as well as highly sensitive individuals were found in both populations. This was mainly pronounced in imazamox-treated DER plants and tribenuron-methyl treated EHR plants, as increasing rates did not show the same response as some of the lower rates did (Figure 1b). No significant reduction in FW and survival percentage of EHR plants was recorded in response to tribenuron-methyl; thus, the ED 50 and LD 50 values have exceeded the highest applied rate (60 g ha −1 ), which eventually resulted in failure to fit a correct model to describe this treatment (Figure 2a). Table 3. Effects of post-emergence treatments on the shoot fresh weight (FW) of plants from D. erucoides (DES) and E. hispanica (EHS) ALS-sensitive and -resistant populations (DES, D. erucoides (DER), EHS, and E. hispanica (her), respectively). Herbicides were applied at 1 and 4 times the label field rate (X).
ED 50 value represent herbicide rate reducing plant growth by 50% were extracted from the dose response curves. RI was calculated as the ratio of the ED 50 value of the resistant population compared with the ED 50 of the sensitive one. Each comparison was between a resistant and sensitive population from the same species under the same treatment. * or ** asterisks indicate on significant differences between treatments, as determined using Student t test (P ≤ 0.05 and 0.01) (n = 12). Label field rate for each herbicide (X): a 24 g ai ha −1 , b 45.5 g ai ha −1 , c 15 g ai ha −1 , and d 4 g ai ha −1 .  The fact that both resistant populations (DER and EHR) exhibited high level of variability in the response to ALS inhibitors may indicate that these populations are still in transition, comprising a mixture of resistant and sensitive individuals. The data concerning survival rates show that even though shoot fresh weight was not significantly different from sensitive populations (DES and EHS), more than 25% of plants from both resistant populations survived the highest rate (4X) of all four ALS inhibitors (Table 4). When treated with the highest rate of tribenuron-methyl, DER showed a higher survival rate than EHR; however, LD 50 values of both populations exceeded the highest rate (>60). In contrast, EHR had higher survival rates following treatment with imazamox, as presented also by higher LD 50 values (36.67 vs. 19.42; Table 4). When both sets of data (shoot FW and survival rate) are viewed side by side, it appears that the DER and EHR populations are at a transitional stage and have not yet developed homogenic response to the tested ALS inhibitors. Using high herbicide rates, 10 and 20 times the label field rate, we have detected high number of survivors in both species alongside susceptible individuals (data not shown), reinforcing our hypothesis that the evolution of resistance to ALS inhibitors in these two populations is still in progress. Previous studies on D. erucoides or E. hispanica mainly deal with their ecological traits and adaptation ability to thrive in different environments [3,4,6,22]. These highly adaptive traits emphasize the risk in the evolution of herbicide resistance of these two species. The fact that both resistant populations (DER and EHR) exhibited high level of variability in the response to ALS inhibitors may indicate that these populations are still in transition, comprising a mixture of resistant and sensitive individuals. The data concerning survival rates show that even though shoot fresh weight was not significantly different from sensitive populations (DES and EHS), more than 25% of plants from both resistant populations survived the highest rate (4X) of all four ALS inhibitors (Table 4). When treated with the highest rate of tribenuron-methyl, DER showed a higher survival rate than EHR; however, LD50 values of both populations exceeded the highest rate (>60). In contrast, EHR had higher survival rates following treatment with imazamox, as presented also by higher LD50 values (36.67 vs. 19.42; Table 4). When both sets of data (shoot FW and survival rate) are viewed side by side, it appears that the DER and EHR populations are at a transitional stage and have not yet developed homogenic response to the tested ALS inhibitors. Using high herbicide rates, 10 and 20 times the label field rate, we have detected high number of survivors in both species alongside susceptible individuals (data not shown), reinforcing our hypothesis that the evolution of resistance to ALS inhibitors in these two populations is still in progress. Previous studies on D. erucoides or E. hispanica mainly deal with their ecological traits and adaptation ability to thrive in different environments [3,4,6,22]. These highly adaptive traits emphasize the risk in the evolution of herbicide resistance of these two species.

Mechanism of Resistance to ALS Inhibitors
Several different ALS gene point mutations have been previously identified to endow resistance to ALS inhibitors [14]. Specific point mutation may alter plant response to ALS inhibitors, as will different amino acid substitutions at the same point mutation [23].
Alignment of sequences from DER plants revealed a Trp574 to Leu substitution in the ALS gene sequence (Figure 3a). In EHR plants, a different substitution, Pro197 to Ser was detected (Figure 3b). To better understand mutations rate and their nature (homozygous and heterozygous) in both DER and EHR populations, random individuals from each population were sequenced. In the DER population, the homozygous (R/R) form of the Trp574 to Leu substitution was found in four out of 15 individuals, two of these individuals exhibited heterozygosity (R/S) and the rest were homozygous for the ALS-sensitive allele (S/S). Out of 12 individual plants from the EHR population, six were found to be homozygous (R/R) for the Pro197 to Ser substitution, four exhibited heterozygosity (R/S), and two were homozygous for the ALS-sensitive allele (S/S). Differences in allelic frequency between both resistant populations can also be correlated with the variation in response to ALS inhibitors. In most studies, mutations in Trp574 confers high level of resistance to all ALS inhibitors [24][25][26][27], while mutations in position Pro197 are mainly associated with high level of resistance to SU herbicides [18,28,29]. The fact that resistance to IMI and SCT herbicides was higher in DER, while resistance to SU was higher in EHR can be associated with the different target site mutations detected in the ALS gene sequence of plants from both populations. It can be suggested that the observed differences in resistance levels and the abundance of mutations are characteristic of a weed population in a transitional stage toward homogenic resistance.

Mechanism of Resistance to ALS Inhibitors
Several different ALS gene point mutations have been previously identified to endow resistance to ALS inhibitors [14]. Specific point mutation may alter plant response to ALS inhibitors, as will different amino acid substitutions at the same point mutation [23].
Alignment of sequences from DER plants revealed a Trp574 to Leu substitution in the ALS gene sequence (Figure 3a). In EHR plants, a different substitution, Pro197 to Ser was detected (Figure 3b). To better understand mutations rate and their nature (homozygous and heterozygous) in both DER and EHR populations, random individuals from each population were sequenced. In the DER population, the homozygous (R/R) form of the Trp574 to Leu substitution was found in four out of 15 individuals, two of these individuals exhibited heterozygosity (R/S) and the rest were homozygous for the ALS-sensitive allele (S/S). Out of 12 individual plants from the EHR population, six were found to be homozygous (R/R) for the Pro197 to Ser substitution, four exhibited heterozygosity (R/S), and two were homozygous for the ALS-sensitive allele (S/S). Differences in allelic frequency between both resistant populations can also be correlated with the variation in response to ALS inhibitors. In most studies, mutations in Trp574 confers high level of resistance to all ALS inhibitors [24][25][26][27], while mutations in position Pro197 are mainly associated with high level of resistance to SU herbicides [18,28,29]. The fact that resistance to IMI and SCT herbicides was higher in DER, while resistance to SU was higher in EHR can be associated with the different target site mutations detected in the ALS gene sequence of plants from both populations. It can be suggested that the observed differences in resistance levels and the abundance of mutations are characteristic of a weed population in a transitional stage toward homogenic resistance.

Alternative Management Using Auxinic Herbicides
Several chemicals that confer auxinic activity are used as herbicides [30]. Epinasty, shoot reorientation, curly leaf, and early senescence have been documented as plant response to auxinic herbicide activity [31,32]. When treated with auxinic herbicides, 2,4-D, or mecoprop-P, ALS-resistant D. erucoides and E. hispanica plants were severely and rapidly damaged, and no survivors were detected at any tested rate. Even at the rate of 1/4X, none of the treated plants survived (Figure 4). Auxinic herbicides may be used to control ALS-resistant individuals to contain and reduce the problem we currently face, preventing further buildup of resistance, such as in the case of Raphanus raphanistrum [33] in Australia, Kochia scoparia [34] in Western Canada, and Amaranthus tuberculatus

Alternative Management Using Auxinic Herbicides
Several chemicals that confer auxinic activity are used as herbicides [30]. Epinasty, shoot reorientation, curly leaf, and early senescence have been documented as plant response to auxinic herbicide activity [31,32]. When treated with auxinic herbicides, 2,4-D, or mecoprop-P, ALS-resistant D. erucoides and E. hispanica plants were severely and rapidly damaged, and no survivors were detected at any tested rate. Even at the rate of 1/4X, none of the treated plants survived (Figure 4). Auxinic herbicides may be used to control ALS-resistant individuals to contain and reduce the problem we currently face, preventing further buildup of resistance, such as in the case of Raphanus raphanistrum [33] in Australia, Kochia scoparia [34] in Western Canada, and Amaranthus tuberculatus [35] in the USA.
To reduce the probability of evolution of resistance to auxinic herbicides, proper crop rotation, as well as other components of integrated weed management, should be applied. [35] in the USA. To reduce the probability of evolution of resistance to auxinic herbicides, proper crop rotation, as well as other components of integrated weed management, should be applied. As was shown before, the over-use of one herbicide can eventually lead to resistance to several other herbicides from the same MOA [36,37]. According to the records of herbicide applications received from the farmers, extensive and prolonged use of tribenuron-methyl and iodosulfuron was documented in fields were DER and EHR populations were found. The evolution of ALS-resistant populations of D. erucoides and E. hispanica in pea and wheat fields may be associated with the repeated use of ALS inhibitors.

Conclusions
Populations of D. erucoides and E. hispanica collected from pea and wheat fields were found to be resistant to ALS inhibitors. In both populations, resistance was found to be associated with target site mutations in the ALS gene sequence. Auxinic herbicides provided excellent control of plants from both ALS-resistant populations. According to the record of herbicide applications received from the farmers, ALS inhibitors have been extensively used in these fields, which in turn may explain the evolution of resistance to ALS inhibitors. Different resistance levels may suggest these populations are in a transitional stage toward homogenic resistance. This report should increase the awareness and alert farmers and researchers in the region and beyond in order to proactively challenge the evolution of herbicide resistance in D. erucoides and E. hispanica in arable crops.  As was shown before, the over-use of one herbicide can eventually lead to resistance to several other herbicides from the same MOA [36,37]. According to the records of herbicide applications received from the farmers, extensive and prolonged use of tribenuron-methyl and iodosulfuron was documented in fields were DER and EHR populations were found. The evolution of ALS-resistant populations of D. erucoides and E. hispanica in pea and wheat fields may be associated with the repeated use of ALS inhibitors.

Conclusions
Populations of D. erucoides and E. hispanica collected from pea and wheat fields were found to be resistant to ALS inhibitors. In both populations, resistance was found to be associated with target site mutations in the ALS gene sequence. Auxinic herbicides provided excellent control of plants from both ALS-resistant populations. According to the record of herbicide applications received from the farmers, ALS inhibitors have been extensively used in these fields, which in turn may explain the evolution of resistance to ALS inhibitors. Different resistance levels may suggest these populations are in a transitional stage toward homogenic resistance. This report should increase the awareness and alert farmers and researchers in the region and beyond in order to proactively challenge the evolution of herbicide resistance in D. erucoides and E. hispanica in arable crops.