Glyphosate- and Imazapic-Resistant Chloris virgata Populations in the Southeastern Cropping Region of Australia

Abstract

Chloris virgata is one of the most problematic summer grass species in southeastern Australia. A total of 40 populations of C. virgata were evaluated in the spring–summer season of 2021–2022 in an open environment at the Gatton Farms of the University of Queensland, Queensland, Australia, for their response to two acetyl-coenzyme-A carboxylase (ACCase) inhibitors (clethodim and haloxyfop), a 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) inhibitor (glyphosate), and an acetolactate synthase (ALS) inhibitor (imazapic). In the first experiment, all populations were screened at the field rate of each herbicide, and the second experiment evaluated the response of seven populations of C. virgata to different rates of glyphosate and imazapic. There were three replications of each treatment and both experiments were repeated over time. None of the populations survived the field rate of clethodim and haloxyfop, possibly suggesting a low exposure of the populations to these herbicides. Individuals in about 90% of populations survived (1% to 100% of individuals surviving) the field rates of glyphosate and individuals in all populations survived (1% to 100%) the imazapic field rate. The dose-response study revealed up to 14- and 5-fold glyphosate resistance in C. virgata populations based on survival and biomass values, respectively, compared to the most susceptible population. Imazapic resistance was up to 2.3- and 16-fold greater than the most susceptible population in terms of survival and biomass values, respectively. The increased cases of glyphosate- and imazapic-resistant C. virgata warrant a nationwide survey and diversified management strategies.

1. Introduction

Weeds are the most important biological constraints to crop production throughout the world. In Australia, they cost more than AU$ 3 billion per annum to grain growers [1] in addition to cotton (Gossypium hirsutum L.) and vegetable growers. Among these weeds, Chloris virgata Sw. is one of the most problematic summer grass species in the southeastern cropping region of Australia. It causes a loss of 40,000 tonnes of grain per year [1]. In mungbean (Vigna radiata (L.) Wilczek), about 35 C. virgata plants m⁻² can reduce grain yield by 50% [2]. In a recent field trial, C. virgata infestation reduced mungbean yield by 73% compared to the best herbicide option [3], indicating the high competitive ability of this weed species. In the same study, the weed produced more than 64,000 seeds m⁻².

In addition to the problem in grain crops, C. virgata is also a troublesome weed in summer fallow and cotton. In fallow, this species can produce more than 140,000 seeds per plant [4]. Seeds of this weed species can germinate at temperatures varying from 15/5 C to 35/25 C (alternate day/night temperatures) [5], suggesting the ability of this weed to germinate throughout the year in southeast Australia. Seeds are small, exhibit hairs, and are lightweight, which helps them to disperse. While C. virgata has a short-lived seed bank [6,7], its germination is stimulated by light and its emergence is greatest for seeds on the soil surface [7,8], supporting its dominance in no-till farming systems. In fallow conditions, weeds are managed using herbicides because of the adverse effect of tillage. Due to the overreliance on the nonselective herbicide, glyphosate, to manage weeds in fallow situations, several weeds have evolved resistance to this herbicide. The first glyphosate-resistant C. virgata case in Australia was reported in 2015 [9] and now there are several glyphosate-resistant populations in Australia [10,11]. A survey conducted in the cotton-growing region of southeast Australia placed C. virgata in the top 10 most common weed species [12]. Compared to previous surveys, this survey reported a significant jump in the ranking of C. virgata and the authors indicated overreliance on glyphosate to manage weeds in glyphosate-tolerant cotton as the possible reason for this change in the ranking.

In cotton and fallow situations, growers use ACCase-inhibiting herbicides, such as haloxyfop and clethodim, to manage grass weed species. The first case of ACCase-inhibiting herbicide-resistant C. virgata was reported in 2019 [13] from the southeastern grain region. ACCase-inhibiting herbicides are highly prone for selecting resistance and therefore the use of these herbicides needs to be monitored. Due to the increased cases of glyphosate-resistant weed populations, growers have turned to the use of ALS-inhibiting herbicides, such as imazapic, to manage grass weeds in fallow [3]. Imazapic-resistant C. virgata has not been reported, however, this herbicide is of high resistance risk. Personal observations suggest imazapic-resistant C. virgata in the southeastern region of Australia, but little is reported of the level of resistance in different populations of this weed. The objective of this study was to evaluate the level of resistance in C. virgata to ACCase- (haloxyfop and clethodim), ALS-(imazapic), and EPSPS-inhibiting (glyphosate) herbicides in the southeastern cropping region of Australia.

2. Material and Methods

2.1. Seed Collection

Seeds of 60 C. virgata populations were collected from January to March of 2017, 2019, and 2021 (20 populations each year) from different locations in Queensland, Australia (Figure 1). Permission to collect seeds was obtained from the landlord. Seeds were collected from sorghum [Sorghum bicolor (L.) Moench), mungbean [Vigna radiata (L.) Wilczek], corn (Zea mays L.), or fallow fields. Seed heads from 20 to 30 plants from each field were mixed and considered a population. Seeds were stored at room temperature (25 ± 2 °C) in plastic containers until January 2021 when all the populations were planted (fresh seeds for the 2021 batch) and grown in the same environment at the Gatton farm of the University of Queensland, Australia. For this purpose, seeds were initially planted in trays filled with potting mix (Centenary Landscape, Darra, QLD, Australia) and transplanted in pots (5 plants per pot; three pots per population) at the 2–3 leaf stage. Of the 60 populations, plants of 40 populations (11 from 2017, 13 from 2019, and 16 from 2021) emerged (Figure 1). Pots with different populations were separated using plastic sheets. Seeds from each population were collected in April 2021 and stored in separate containers at room temperature (25 ± 2 °C) until the commencement of the study in September 2021.

2.2. Herbicide Screening

Herbicide screenings were performed during the spring–summer seasons of 2021–2022 at the weed science research facility of the University of Queensland, Gatton, Queensland, Australia. All 40 populations of C. virgata were evaluated for their response to two ACCase-inhibiting herbicides (clethodim and haloxyfop), one ALS-inhibiting herbicide (imazapic), and the EPSPS-inhibiting herbicide (glyphosate). Herbicides were applied at the recommended field rates: clethodim at 90 g a.i. ha⁻¹, haloxyfop at 78 g a.i. ha⁻¹, glyphosate at 741 g a.e. ha⁻¹, and imazapic at 96 g a.i. ha⁻¹. Adjuvants at 1% were used with haloxyfop and imazapic (Hasten^TM), and clethodim (Supercharge^®). Plastic pots (14 cm in diameter) were filled with potting mix and 12–15 seeds were planted per pot at a depth of 2 mm. After emergence, plants were thinned and 6 plants per pot were maintained thereafter. There were three replications of each treatment and two experimental runs. The second run was commenced within a week after concluding the first run. Pots were kept on benches outdoor and watering was done using an automated sprinkler system. At the 4–5 leaf stage, herbicides were applied using a research track sprayer fitted with Teejet XR110015 flat nozzles that delivered a spray volume of 108 L ha⁻¹ at a pressure of 200 kPa. Plants were not watered until 24 h after the spray. At 3 weeks after herbicide application, plants were evaluated for survival with the criterion of at least one new leaf.

A second pot trial was conducted to evaluate the response of seven populations of C. virgata to different herbicide rates. The first trial found that clethodim and haloxyfop killed seedlings of all populations at their field rates; therefore, these two herbicides were not included in the dose–response experiment. Based on the seedling survival data, populations 12/21 and 25/21 were selected as glyphosate-susceptible populations and populations 7/17, 11/17, 12/19, 3/21, and 19/21 were selected as the glyphosate-resistant populations. The glyphosate dose-response study was conducted with seven rates (0, 0.25, 0.5, 1, 2, 4, and 8× for resistant populations and 0, 0.0625, 0.125, 0.25, 0.5, 1, and 2× for susceptible populations). Populations 9/17 and 12/21 were selected as the imazapic-susceptible populations and populations 11/17, 7/19, 12/19, 16/21, and 19/21 were selected as the imazapic-resistant populations. Imazapic was applied at seven rates (0, 0.0625, 0.125, 0.25, 0.5, 1 and 2×). Herbicides were applied as described above. There were three replications of each treatment, and the trial was conducted twice. Seedling survival (%) data were taken 4 weeks after herbicide application and aboveground biomass of plants were harvested. Plant samples were dried in an oven at 70 C for 72 h and weighed.

2.3. Statistical Analyses

Experiments were conducted in a randomized complete block design. Data from the herbicide screening and dose-response trials were subjected to the analysis of variance (ANOVA) to determine treatment by experimental run interaction [14]. The interactions were not significant; therefore, the data were pooled across the two experimental runs for the final analysis. Seedling survival data from the first trial were plotted as a box plot using Sigmaplot 14.5 (Systat Software, Inc., Point Richmond, CA, USA) [15]. Populations with no survival were considered susceptible, with 1% to 20% survival considered slightly resistant, with 21% to 80% survival considered moderately resistant, and with 81% to 100% survival considered highly resistant. In the dose-response trial, seedling survival and biomass (percent of nontreated control treatments) data were regressed against glyphosate or imazapic doses using a three-parameter logistic model using Sigmaplot 14.5. The fitted model was:

S = a/[1 + (x/H₅₀)^b]

where S is seedling survival (%) or biomass (% reduction of control) at herbicide dose x, a is the maximum seedling survival or biomass, H₅₀ is glyphosate or imazapic dose (g a.e. or a.i. ha⁻¹) that would cause a 50% reduction in seedling survival (LD₅₀) or biomass (GR₅₀), and b is the slope of the model. Resistant indices were calculated as the ratio between LD₅₀ or GR₅₀ of each population and the LD₅₀ or GR₅₀ of the most susceptible population (i.e., 12/21 for glyphosate and 9/17 for imazapic).

3. Results and Discussion

3.1. ACCase-Inhibitors

None of the 40 populations tested in this study survived the field rate of clethodim and haloxyfop (

Table 1. Resistance levels of Chloris virgata populations (n = 40) collected from Queensland, Australia.

Herbicide	Highly Resistant a	Moderately Resistant b	Slightly Resistant c	Susceptible d
	–––––––––––% of the population–––––––––––
Clethodim	0	0	0	100
Glyphosate	57.5	27.5	5	10
Haloxyfop	0	0	0	100
Imazapic	2.5	85	12.5	0

^a Highly resistant defined as populations with 81 to 100% survivors. ^b Moderately resistant defined as populations with 21% to 80% survivors. ^c Slightly resistant defined as populations with 1% to 20% survivors. ^d Susceptible was defined as populations with 0% survivors.

). No case of clethodim- and haloxyfop-resistant C. virgata has been reported from overseas. In Australia, the first case of ACCase-inhibiting herbicide-resistant C. virgata was reported in 2019, but the report did not mention the name of the herbicide [13]. The low occurrence of ACCase-inhibiting herbicide-resistant populations could be attributed to low exposure of C. virgata to these herbicides in Australia. Other grass species, such as Lolium rigidum Gaud. and L. perenne ssp. multiflorum, have been widely reported to be resistant to ACCase-inhibitors in Australia and other countries [15,16,17] due to the increased selection pressure for these herbicides.

Due to the widespread glyphosate-resistant populations of C. virgata across Australia, farmers are encouraged to use alternate herbicide options, including clethodim and haloxyfop [18]. ACCase-inhibitors are highly prone to developing resistance; therefore, growers need to rotate these herbicides with other herbicide modes of action. In the current study, herbicides were applied at the four–five leaf stage, but growers may not be able to apply herbicides at a young seedling stage because of environmental constraints or available resources. In such situations, effective herbicides may also not work. In a recent study, for example, clethodim spray at the 24-leaf stage of C. virgata plants at the field rate (90 g a.i. ha⁻¹) resulted in 30% plant survival [18].

3.2. Glyphosate

Only 10% of the populations (n = 40) were highly susceptible (0% survivors) and 5% of the populations were slightly resistant to glyphosate at the field rate (Table 1). Out of the 40 populations, 58% of the populations were highly resistant (>80% survivors) and 28% of the populations were moderately resistant to glyphosate (21 to 80% survivors). The box plot shows that 50% of the populations had at least 94% seedling survival frequency (Figure 2).

The dose-response assays revealed that the LD₅₀ value of the most susceptible population (12/21) was 190 g a.e. ha⁻¹ (Figure 3;

Table 2. Estimated glyphosate doses to kill 50% of the plants (LD₅₀) of seven populations of Chloris virgata, glyphosate doses required to reduce plant biomass by 50% (GR₅₀), and resistance indices (RI).

Population	LD50	RI	GR50	RI
	––g a.e. ha−1––		––g a.e. ha−1––
7/17	1692	9.0	355	2.8
11/17	1351	7.2	262	2.0
12/19	1979	10.6	308	2.4
3/21	1789	9.6	662	5.1
12/21	187	1.0	129	1.0
19/21	2689	14.4	471	3.7
25/21	262	1.4	148	1.1

RI were calculated as the ratio between the LD₅₀ or GR₅₀ of each population and the LD₅₀ or GR₅₀ of the most susceptible population (i.e., 12/21).

). The LD₅₀ values for the other six populations ranged from 260 to 1980 g a.e. ha⁻¹, exhibiting up to 11-fold resistance compared to the 12/21 population, based on the resistance indices. The GR₅₀ value for the most susceptible population (12/21) was 130 g a.e. ha⁻¹ whereas, for the other six populations, the GR₅₀ values ranged from 150 to 660 g a.e. ha⁻¹, exhibiting up to five-fold resistance based on biomass values (Figure 4; Table 2).

Glyphosate resistance in C. virgata is only present in Australia [9]. Similar to our study, a previous study in South Australia reported 2- to 10-fold resistance (based on LD₅₀ values) in glyphosate-resistant C. virgata populations compared with a susceptible population [11]. Another study in Queensland reported LD₅₀ values for glyphosate of up to 5.8 kg a.e. ha⁻¹ [10]. In both studies, target-site EPSPS mutations conferred resistance to glyphosate in C. virgata [10,11]. It was also suggested that the resistance was not due to alteration in glyphosate or translocation [11]. The poor control of this weed species by glyphosate is still not clearly understood as C. virgata is known to have a level of natural tolerance to glyphosate [19]. In the current study, differential levels of glyphosate resistance among different populations could be attributed to the exposure history to glyphosate.

Globally, the number of glyphosate-resistant weeds has been increasing [9], mainly due to overreliance on glyphosate for presowing weed control and the introduction of glyphosate-resistant crops [20]. Closely related species of C. virgata, Chloris truncata R. Br, Chloris barabata Sw., and Chloris polydactyla (L.) Sw., also have evolved resistance to glyphosate [11,21,22]. Chloris truncata is an Australian native grass species and it coexists with the introduced species, C. virgata. Some other Australian summer weed species that have evolved resistance to glyphosate are Sonchus oleraceus L., Eleusine indica (L.) Gaertn., Sorghum halepense (L.) Pers., and Echinochloa colona (L.) Link. [9].

The sole reliance on glyphosate for weed control in summer fallow and glyphosate-resistant cotton in Australia may result in a further increase in resistant cases. In addition to resistance, glyphosate efficacy is also reduced by high temperature, low soil moisture, and low relative humidity [23,24,25,26,27]. These observations suggest that poor efficacy of glyphosate may continue to occur in glyphosate-resistant C. virgata populations during periods of suboptimal environmental conditions. The occurrence of high numbers of glyphosate-resistant C. virgata populations means that strategies that rely on glyphosate alone will not be successful [11]. In glyphosate-resistant cotton crops, growers may have to spray other herbicides to control glyphosate-resistant C. virgata.

3.3. Imazapic

Of the 40 populations, about 3% of the populations were highly resistant (>80% survivors) and 85% of the populations were moderately resistant (21% to 80% survivors) (Table 1). Only 13% of the populations were slightly resistant (1% to 20% survivors) and none were susceptible. Fifty percent of the populations had at least 38% seedling survival frequency (Figure 2).

The LD₅₀ value of the most susceptible population (9/17) was 59 g a.i. ha⁻¹ (Figure 5;

Table 3. Estimated imazapic doses to kill 50% of the plants (LD₅₀) of seven populations of Chloris virgata, imazapic doses required to reduce plant biomass by 50% (GR₅₀), and resistance indices (RI).

Population	LD50	RI	GR50	RI
	––g a.i. ha−1––		––g a.i. ha−1––
9/17	59	1.0	4	1.0
11/17	85	1.4	44	11.0
7/19	135	2.3	64	16.0
12/19	95	1.6	15	3.8
12/21	65	1.1	15	3.8
16/21	92	1.6	24	6.0
19/21	83	1.4	22	5.5

RI were calculated as the ratio between the LD₅₀ or GR₅₀ of each population and the LD₅₀ or GR₅₀ of the most susceptible population (i.e., 9/17).

). The LD₅₀ values for other populations ranged from 65 to 135 g a.i. ha⁻¹. The GR₅₀ value for the population 9/17 was 4 g a.i. ha⁻¹, whereas for other populations, the values ranged from 15 to 64 g a.i. ha⁻¹, exhibiting up to 16-fold resistance to imazapic based on biomass reductions (Figure 6; Table 3). In general, the GR₅₀ values were greater than the LD₅₀ values for different populations. This response could be due to differential levels of intracompetition among survived plants of C. virgata.

Imazapic, an imidazolinone herbicide, is an ALS-inhibitor and the selection for resistance to herbicides with this mode of action is generally quicker than other herbicide modes of action [28]. In Australia, this herbicide is registered for use in fallow conditions, sugarcane (Saccharum officinarum L.), and peanuts (Arachis hypogaea L.) for the control of Echinochloa colona and Eleusine indica, Digitaria ciliaris (Retz.) Koel., and other grass species, but C. virgata is not on the imazapic label. However, growers commonly use imazapic to control a mixture of grass weed species, including C. virgata, in fallow conditions. While no imazapic-resistant C. virgata has been reported, our results suggest that there are widespread imazapic-resistant populations in Queensland. The maximum recommended field rate of imazapic is 96 g a.i. ha⁻¹ and our study suggests that there is a population that needs up to 135 g a.i. ha⁻¹ of imazapic to kill 50% of the seedlings (i.e., LD₅₀).

The results of the current study are similar to the results obtained in our previous study [3]. In the previous study, imazapic was applied as PRE, but in the current study, the herbicide was applied as POST (imazapic is recommended as PRE as well as POST). In the previous study, imazapic at 96 g a.i. ha⁻¹ did not provide effective control of C. virgata. These results suggest the need for a survey to collect and screen C. virgata populations across Australia for ALS-inhibiting herbicides, including imazapic. Due to the increasing cases of glyphosate resistance, growers have started using ALS inhibitors to manage weeds in summer fallow. The continuous use of imazapic may result in a shift towards harder-to-control weeds, including C. virgata.

In summary, this study evaluated resistance levels in 40 populations of C. virgata to clethodim, haloxyfop, glyphosate, and imazapic. While weed populations are prone to evolving resistance quickly to ACCase inhibitors, all 40 populations were found to be highly susceptible to clethodim and haloxyfop. C. virgata is not listed on the label of glyphosate and imazapic products, but these herbicides are commonly used to control C. virgata in summer fallows. The current study and previous studies found that C. virgata populations are susceptible to imazapic [3] and glyphosate [10,11], suggesting that C. virgata populations with differential resistance levels to these herbicides exist in Queensland. While glyphosate-resistant cases of C. virgata have been widely reported, there have been no cases reporting resistance to imazapic. In Australia, weed surveys are conducted at the national level with the help of the Grains Research and Development Corporation (GRDC) and to our knowledge, C. virgata has not been included in such surveys for quantifying its resistance status to imazapic. Our results suggest the need to evaluate the response of C. virgata populations to imazapic in future surveys. Our study was conducted in an outdoor environment and temperatures may affect the efficacy of herbicides. Therefore, future studies may need to be conducted in glasshouse conditions.

The widespread increase in glyphosate-resistant C. virgata will see the use of alternate herbicides, such as ACCase and ALS inhibitors. While the 40 populations of C. virgata were found to be highly susceptible to clethodim and haloxyfop, their continuous use may result in the evolution of resistance in C. virgata to these herbicides. Therefore, integrated weed management strategies need to be developed to delay the evolution of resistance that take into account nonchemical methods of weed control, such as tillage and harvest weed seed control practices [2,11,29]. Burying seeds below the maximum depth of emergence could help manage C. virgata populations resistant to multiple herbicides [6]. The use of effective residual herbicides (e.g., S-metolachlor, pendimethalin, etc.) also needs to be integrated with other management practices to control herbicide-resistant C. virgata. In addition, the use of the double knock technique (i.e., sequential herbicide applications) can provide effective control of glyphosate-resistant C. virgata. In this technique, glyphosate application is followed by paraquat (7–10 days after glyphosate spray). In a recent study, haloxyfop followed by paraquat applications were found highly effective for the control of C. virgata [30]. Future research should focus on understanding the combined use of glyphosate and glufosinate in Xtendflex^TM cotton crops [31], which may become available soon in Australia. Future research should also focus on understanding the molecular mechanisms responsible for imazapic resistance in C. virgata populations.