Introduction
Kochia [Bassia scoparia (L.) A.J. Scott] is a tumbleweed that is native to Eurasia and was introduced to North America as an ornamental plant in the mid- to late-1800s (Friesen et al. Reference Friesen, Beckie, Warwick and Van Acker2009). Bassia scoparia can be an invasive and troublesome weed in cropping systems, pastureland, and ruderal areas due to its summer annual life cycle, early seedling emergence (Kumar et al. Reference Kumar, Jha, Dille and Stahlman2018; Schwinghamer and Van Acker Reference Schwinghamer and Van Acker2008), abiotic stress tolerance (Friesen et al. Reference Friesen, Beckie, Warwick and Van Acker2009), competitiveness (Geddes and Sharpe Reference Geddes and Sharpe2022), prolific seed production (Beckie et al. Reference Beckie, Blackshaw, Hall and Johnson2016), and short-lived seed persistence in both aerial (Geddes and Pittman Reference Geddes and Pittman2023) and soil seedbanks (Beckie et al. Reference Beckie, Blackshaw, Leeson, Stahlman, Gaines and Johnson2018; Dille et al. Reference Dille, Stahlman, Du, Geier, Riffel, Currie, Wilson, Sbatella, Westra, Kniss, Moechnig and Cole2017; Schwinghamer and Van Acker Reference Schwinghamer and Van Acker2008). High genetic diversity (Martin et al. Reference Martin, Benedict, Sauder, Beckie and Hall2020) combined with efficient pollen- and seed-mediated gene flow (Beckie et al. Reference Beckie, Blackshaw, Hall and Johnson2016) cause rapid evolution of B. scoparia in response to management practices. Herbicides remain the primary method used to manage this weed, and in response, B. scoparia populations have evolved resistance to up to four herbicide sites of action (Beckie et al. Reference Beckie, Hall, Shirriff, Martin and Leeson2019; Varanasi et al. Reference Varanasi, Godar, Currie, Dille, Thompson, Stahlman and Jugulam2015).
Multiple herbicide–resistant B. scoparia is a widespread issue throughout the Great Plains of North America (Kumar et al. Reference Kumar, Jha, Jugulam, Yadav and Stahlman2019), where it can cause substantial crop yield losses if left unmanaged (Geddes and Sharpe Reference Geddes and Sharpe2022). Bassia scoparia was the most-abundant herbicide-resistant broadleaf weed in a 2019/2020 survey of Saskatchewan, where uncontrolled populations occupied an estimated 39% of annual-cropped fields (Geddes et al. Reference Geddes, Pittman, Sharpe and Leeson2024). Herbicide resistance in B. scoparia dates back to 1976, when photosystem II (PSII) inhibitor (Herbicide Resistance Action Committee [HRAC] Group 5) resistance was reported in Kansas (Heap Reference Heap2024). Bassia scoparia resistant to acetolactate synthase (ALS)-inhibiting (HRAC Group 2) herbicides was reported first in Kansas and North Dakota in 1987. ALS inhibitor−resistant B. scoparia is widespread and was present in all survey samples tested in Canada in recent decades (Beckie et al. Reference Beckie, Gulden, Shaikh, Johnson, Willenborg, Brenzil, Shirriff, Lozinski and Ford2015; Hall et al. Reference Hall, Beckie, Low, Shirriff, Blackshaw, Kimmel and Neeser2014). Bassia scoparia with resistance to auxin mimics (HRAC Group 4) was reported first in Montana in 1993/1994 and only recently in Canada since 2015 (Beckie et al. Reference Beckie, Hall, Shirriff, Martin and Leeson2019; Cranston et al. Reference Cranston, Kern, Hackett, Miller, Maxwell and Dyer2001; Geddes et al. Reference Geddes, Ostendorf, Owen, Leeson, Sharpe, Shirriff and Beckie2022a, Reference Geddes, Owen, Ostendorf, Leeson, Sharpe, Shirriff and Beckie2022b, Reference Geddes, Pittman, Gulden, Jones, Leeson, Sharpe, Shirriff and Beckie2022c, Reference Geddes, Pittman, Hall, Topinka, Sharpe, Leeson and Beckie2023; Heap Reference Heap2024). Bassia scoparia resistant to glyphosate (HRAC Group 9) was first documented in Kansas in 2007 and later in multiple states and provinces (Beckie et al. Reference Beckie, Blackshaw, Low, Hall, Sauder, Martin, Brandt and Shirriff2013; Hall et al. Reference Hall, Beckie, Low, Shirriff, Blackshaw, Kimmel and Neeser2014; Heap Reference Heap2024). After only a single decade since the first report of glyphosate-resistant B. scoparia in Canada (2011), this biotype was present in about three-quarters of B. scoparia samples tested (n = 889) between 2018 and 2021 (Geddes et al. Reference Geddes, Pittman, Gulden, Jones, Leeson, Sharpe, Shirriff and Beckie2022c, Reference Geddes, Pittman, Hall, Topinka, Sharpe, Leeson and Beckie2023; Sharpe et al. Reference Sharpe, Leeson, Geddes, Willenborg and Beckie2023).
Interest in protoporphyrinogen oxidase (PPO)-inhibiting herbicides has grown recently, due in part to the continued evolution and spread of glyphosate-resistant weeds (Barker et al. Reference Barker, Pawlak, Duke, Beffa, Tranel, Wuerffel, Young, Porri, Liebl, Aponte, Findley, Betz, Lerchl, Culpepper, Bradley and Dayan2023; Dayan et al. Reference Dayan, Barker and Tranel2018). Herbicides targeting this site of action have been commercialized for more than a half century, despite their mechanism of action only being elucidated in recent decades (Matringe et al. Reference Matringe, Camadro, Labbe and Scalla1989a, Reference Matringe, Camadro, Labbe and Scalla1989b). In susceptible plants, PPO inhibitors cause chlorosis, wilting, and necrosis; they have been referred to colloquially as bleaching or peroxidizing herbicides. Following plant uptake, the PPO-inhibiting active ingredient enters photosynthetically active parenchyma cells, where it inhibits PPO isoforms, PPO1 and PPO2, located in the chloroplast. Protogen is then leaked into the cytoplasm, where it is converted to photodynamic protoporphyrin IX (proto). Proto generates a flush of reactive oxygen species under light, which ultimately causes membrane lipid peroxidation (Barker et al. Reference Barker, Pawlak, Duke, Beffa, Tranel, Wuerffel, Young, Porri, Liebl, Aponte, Findley, Betz, Lerchl, Culpepper, Bradley and Dayan2023). There are currently 21 unique herbicide active ingredients commercialized, spanning four different chemical families, that inhibit PPO (HRAC 2024b).
Rapid evolution and spread of multiple herbicide resistance traits in B. scoparia, and widespread glyphosate resistance in particular, resulted in greater reliance on PPO-inhibiting herbicides for B. scoparia control. Several studies document excellent foliar- and soil-applied activity of PPO inhibitors on B. scoparia (Kumar and Jha Reference Kumar and Jha2015; Torbiak et al. Reference Torbiak, Blackshaw, Brandt, Hall, Hamman and Geddes2021a, Reference Torbiak, Blackshaw, Brandt, Hamman and Geddes2021b, Reference Torbiak, Blackshaw, Brandt, Hamman and Geddes2022, Reference Torbiak, Blackshaw, Brandt, Hamman and Geddes2024; Yadav et al. Reference Yadav, Kumar and Jha2020). For example, preemergence sulfentrazone (105 g ai ha−1) controlled glyphosate-resistant B. scoparia in spring wheat (Triticum aestivum L.) by 95% to 99% 3 wk after postemergence herbicides were applied (Torbiak et al. Reference Torbiak, Blackshaw, Brandt, Hamman and Geddes2021b). Carfentrazone + sulfentrazone (9 + 105 g ai ha−1) applied preemergence controlled glyphosate- and ALS inhibitor–resistant B. scoparia in field pea (Pisum sativum L.) by 94% on average 3 wk after the postemergence herbicide treatment timing (Torbiak et al. Reference Torbiak, Blackshaw, Brandt, Hamman and Geddes2022). In chemical fallow, glyphosate (450 g ae ha−1) mixed with saflufenacil (18 or 50 g ai ha−1), carfentrazone (18 g ai ha−1), or carfentrazone + sulfentrazone (9 + 53 or 9 + 105 g ai ha−1) resulted in ≥90% control of glyphosate-resistant B. scoparia in Alberta (Torbiak et al. Reference Torbiak, Blackshaw, Brandt, Hall, Hamman and Geddes2021a). In Montana, saflufenacil (25 g ai ha−1) applied postemergence controlled B. scoparia by 90% 1 wk after treatment (WAT), which decreased to 67% by 5 WAT absent crop interference (Kumar and Jha Reference Kumar and Jha2015). Glyphosate + sulfentrazone (1,261 + 210 g ae/ai ha−1) resulted in near-complete control and 97% to 100% biomass reduction of B. scoparia in glyphosate/dicamba-resistant soybean [Glycine max (L.) Merr.] grown in Montana and Kansas (Yadav et al. Reference Yadav, Kumar and Jha2020). Excellent B. scoparia control with PPO-inhibiting herbicides resulted in extensive adoption of glyphosate and PPO-inhibiting herbicide mixtures to control glyphosate-resistant B. scoparia before crop planting in the conservation tillage systems that dominate the Great Plains region. However, due to widespread glyphosate resistance in this species, this resulted in only a single herbicide site of action with sufficient activity on B. scoparia. When this is combined with widespread ALS-inhibitor resistance in B. scoparia resulting in no effective postemergence herbicides in many pulse crops grown in the region, and auxinic herbicide resistance limiting postemergence weed control in small grain cereals, heavy reliance on PPO inhibitors for B. scoparia control could increase risk of selection for PPO-inhibitor resistance (Sharpe and Novek Reference Sharpe and Novek2024).
Poor control of B. scoparia with PPO-inhibiting herbicides was identified in a brown mustard [Brassica juncea (L.) Czern.] field located near Kindersley, SK, Canada, in 2021. Glyphosate and sulfentrazone either alone or mixed with carfentrazone were applied preemergence during the previous three growing seasons (Table 1). Similarly, poor B. scoparia control with carfentrazone + sulfentrazone was noted in a sunflower (Helianthus annuus L.) field near Mandan, ND, USA, and in research plots near Minot, ND, USA, in 2022 (Table 2). The objectives of this research were to determine (1) if the B. scoparia accessions collected from Saskatchewan and North Dakota were resistant to the foliar-applied PPO-inhibiting herbicides saflufenacil and carfentrazone, and (2) if so, the level of resistance observed.
Table 1. Recent herbicide use history in the Kindersley, SK, Canada, field where protoporphyrinogen oxidase (PPO) inhibitor–resistant Bassia scoparia was confirmed in 2021.
a POST, postemergence; PRE, preemergence.
b Brown mustard, Brassica juncea (L.) Czern.; chickpea, Cicer arietinum L.; flax, Linum usitatissimum L.
c PPO-inhibiting active ingredients are underlined.
Table 2. Recent protoporphyrinogen oxidase (PPO)-inhibiting herbicide use in the Mandan and Minot, ND, USA, fields where PPO inhibitor–resistant Bassia scoparia was confirmed in 2022.
a Corn, Zea mays L.; field pea, Pisum sativum L.; sunflower Helianthus annuus L.; wheat, Triticum aestivum L
b These herbicides were applied in research plots in various parts of the field, and research plots were moved around the field each year. The rest of the field was typically seeded to wheat where no PPO-inhibiting herbicides were used.
Materials and Methods
Plant Material
Mature seeds from at least 20 uncontrolled B. scoparia plants were collected at random from the fields of interest (Figure 1). The Saskatchewan fields were sampled in October 2021 and the North Dakota fields were sampled in October 2022. The putative-resistant sample from Saskatchewan was collected from a field planted to brown mustard near Kindersley, SK, and designated “KindersleyR” (coordinates not provided to protect farmer identity). Two susceptible control accessions were also collected, one from a field near Eastend, SK (hereafter “EastendS”) and another being a lab-maintained ALS inhibitor–resistant, but glyphosate- and auxin mimic–susceptible control collected near Rosetown, SK (hereafter “RosetownS”). Two previously collected accessions were used as susceptible controls in the North Dakota experiments. A field near Fargo, ND (hereafter “FargoS”) was sampled in 2012, and a field in Minot, ND (hereafter “MinotS”) was sampled in approximately 2010. The putative-resistant accessions collected near Mandan and Minot, ND, in 2022 were designated “MandanR” and “MinotR”, respectively.
Figure 1. Map of Canada and the United States showing the collection locations of the protoporphyrinogen oxidase (PPO) inhibitor–resistant Bassia scoparia accessions and the susceptible control accessions used for the first confirmations of PPO inhibitor–resistant Bassia scoparia in 2021 and 2022. Collection locations are adjusted to the nearest city or town.
The field-collected samples from Saskatchewan were subjected initially to single-dose screening with saflufenacil (Heat® LQ, BASF Canada, Mississauga, ON, Canada) at 50 g ai ha−1. The single-dose screening was unreplicated and consisted of three B. scoparia accessions (KindersleyR, EastendS, and RosetownS) and two herbicide regimes (treated and untreated). The B. scoparia accessions were planted in 24 by 24 by 5 cm greenhouse flats filled with Cornell soilless potting medium (Sheldrake and Boodley Reference Sheldrake and Boodley1966) targeting 40 plants flat−1. The flats were placed in the greenhouse at the Agriculture and Agri-Food Canada Lethbridge Research and Development Centre where they were watered daily. The greenhouse followed a 20/18 C day/night temperature regime with 16-h photophase and 8-h scotophase. Fluence RAZR 3 light-emitting diode bulbs (Fluence, Austin, TX, USA) provided 230 µmol m−2 s−1 supplemental light. The herbicide was applied at 5- to 8-cm plant height using a moving-nozzle cabinet sprayer with a TeeJet® (Wheaton, IL, USA) flat-fan 8002VS nozzle calibrated to deliver 200 L ha−1 spray solution at 275 kPa when traveling at 2.4 km h−1. To limit the potential impact of parental environment on the phenotypic expression of resistance and to demonstrate transfer of the resistance trait to subsequent generations (HRAC 2024a), survivors from the treated KindersleyR accession and untreated EastendS and RosetownS accessions were transplanted separately into larger containers 21 d after treatment (DAT) and grown for seed under pollination bags created from 10-µm nylon mesh (Miami Aqua-culture, Boynton Beach, FL, USA). Four surviving plants from each accession were placed under a pollination bag where they were allowed to cross-pollinate within each accession to avoid inbreeding depression. The second-generation seeds were hand harvested and threshed, and the seed was stored at 4 C until used for the dose–response experiments. The North Dakota accessions (MandanR, MinotR, FargoS, and MinotS) were field-collected samples that were not regenerated under controlled-environment conditions.
Dose Response
Separate dose–response bioassays were conducted for the Saskatchewan (KindersleyR, EastendS, and RosetownS) and North Dakota (MandanR, MinotR, FargoS, and MinotS) accessions in the greenhouses at the Lethbridge Research and Development Centre and North Dakota State University, respectively. Each experiment included either one (Saskatchewan) or two (North Dakota) putative PPO inhibitor–resistant B. scoparia accessions and two locally relevant susceptible control accessions. At each location, separate dose–response experiments were conducted for saflufenacil and carfentrazone (Aim® EC, FMC of Canada, Mississauga, ON, Canada) and were repeated once. Each experiment followed a factorial randomized complete block design in which the first factor consisted of B. scoparia accession and the second factor was herbicide rate. The rate structure followed 0 (untreated), 0.01, 0.1, 1, 3.16, 10, 31.6, and 100× the U.S. field rate for each active ingredient (25 g ai ha−1 for saflufenacil and 17.5 g ai ha−1 for carfentrazone; Ikley et al. Reference Ikley, Christoffers, Dalley, Endres, Gramig, Howatt, Jenks, Law, Lim, Ostlie, Peters, Robinson, Thostenson and Valenti2024). The Saskatchewan accession experiments (Figure 2) included nine blocks consisting of one B. scoparia plant in each 10 by 10 cm plastic greenhouse pot. The Saskatchewan experiments used the same potting medium, growth environment, and herbicide treatment methodology described previously. The North Dakota accession experiments (Figure 3) included 10 blocks consisting of one B. scoparia plant in each 4-cm diameter by 21-cm deep Cone-tainer™ (Stuewe & Sons, Tangent, OR, USA) filled with four parts potting soil (Promix BX, Premier Horticulture, Quakertown, PA, USA) and one part sandy loam soil. In the North Dakota experiments, plants were treated when they reached 2 to 3 cm in height using a moving-nozzle cabinet sprayer equipped with a TeeJet® XR 8002E nozzle calibrated to deliver 140 L ha−1 spray solution at 207 kPa when traveling 5.4 km h−1. Ammonium sulfate and methylated seed oil were added to the spray solution at 10 g L−1 and 1% v/v, respectively. Greenhouse temperatures were maintained between 24 and 27 C with a 16-h photophase and 8-h scotophase supplemented with light from 1,000-W high-pressure sodium lamps (P.L. Light Systems, Beamsville, ON, Canada). All experiments were watered from above daily and did not receive fertilization other than that provided by the potting medium.
Figure 2. One replicate of the (A) saflufenacil and (B) carfentrazone dose–response experiments at 21 d after treatment (DAT) for one putative-resistant (KindersleyR) and two susceptible (RosetownS and EastendS) Bassia scoparia accessions from Saskatchewan, Canada.
Figure 3. One replicate of the (A) saflufenacil at 28 d after treatment (DAT) and (B) carfentrazone at 21 DAT dose–response experiments for two putative-resistant (MandanR and MinotR) and two susceptible (MinotS and FargoS) Bassia scoparia accessions from North Dakota, USA.
The B. scoparia measurements included visible control at 7 and 21 DAT and shoot biomass fresh weight (FW) and dry weight (DW) at 21 DAT; save for the North Dakota saflufenacil experiments that included visible control, FW, and DW measurements at 28 DAT. Bassia scoparia visible control was estimated as a percentage from 0% to 100% control relative to the untreated control within each accession and block following the rating scale reported by the Canadian Weed Science Society (2018). Bassia scoparia shoot biomass was determined by harvesting all living and dead tissue above the soil surface and weighing (FW), followed by drying in an oven at 60 C until constant weight and weighing again (DW). Both biomass FW and DW were included as response variables to account for the impacts of dead B. scoparia tissue at high herbicide rates on the dose–response relationship due to differential moisture retention between living and dead plant tissue.
Statistical Analyses
The experiments using the Saskatchewan and North Dakota B. scoparia accessions were analyzed separately following the same two-stage procedure including ANOVA followed by nonlinear regression. Visible control (at 7 and 21/28 DAT) and biomass (FW and DW at 21/28 DAT) data were subjected to ANOVA using PROC MIXED in SAS software v. 9.4 (SAS Institute, Cary, NC, USA). Bassia scoparia accession, herbicide rate, experimental run, and their interactions were considered fixed factors, while block nested within run was considered a random factor. The model assumptions were assessed using PROC UNIVARIATE based on the Shapiro-Wilk statistic and by plotting the residuals and fitted values (Littell et al. Reference Littell, Milliken, Stroup, Wolfinger and Shabenberger2006). Variance component analyses were used to determine the percentage of total model sums of squares allocated to each factor. All main and interaction effects including experimental run accounted for <5% of the total sums of squares, and this factor was therefore removed from the final analysis after confirming homogeneous variance across runs.
The B. scoparia visible control (at 7 and 21/28 DAT) and biomass (FW and DW at 21/28 DAT) data were analyzed using nonlinear regression in the drc package of R v. 4.3.1 (R Core Team 2023). The analysis used the three-parameter log-logistic function (Equation 1)
$$y = {{d}\over{{1 + {\rm{exp}}\left\{ {b\left[ {\log \left( x \right) - \log \left( e \right)} \right]} \right\}}}}$$
where
$y$
is the response variable,
$d$
is the upper asymptote,
$b$
is the slope of the regression line at dose
$e$
,
$e$
is the regression line inflection point, and
$x$
is the herbicide rate (in g ai ha−1) (Ritz et al. Reference Ritz, Baty, Streibig and Gerhard2015). A four-parameter log-logistic function was fit initially, but the lower asymptote did not differ from zero (α = 0.05), and so a common lower asymptote was fit based on model parsimony. A similar approach was taken to fit a common upper asymptote when modeling visible control data only when the upper asymptote for each regression curve did not differ from 100% control, in which case the
$d$
parameter was fit individually for each B. scoparia accession. The ED and EDcomp functions were used to determine herbicide effective doses resulting in 50% and 80% visible control (ED50 and ED80, respectively) or biomass reduction (GR50 and GR80, respectively) and compare among them (α = 0.05). The effective doses for visible control were determined relative to the limits of 0% and 100% control, while the effective doses for biomass were determined relative to the predicted biomass for the untreated control within each B. scoparia accession. The resistance index was calculated by dividing the ED50 or GR50 value for each putative-resistant accession by that for each corresponding susceptible control accession. The putative PPO inhibitor–resistant B. scoparia accessions were considered to be highly resistant if the resistance index was ≥10 (HRAC 2024a).
Results and Discussion
The putative PPO inhibitor–resistant B. scoparia accessions from Saskatchewan (KindersleyR) and North Dakota (MandanR and MinotR) were highly resistant to foliar-applied saflufenacil and carfentrazone. These represent the first reports of PPO inhibitor–resistant B. scoparia globally and that the issue was present in the Northern Great Plains region at sites located up to 790 km apart and on either side of the Canada/U.S. border (Figure 1).
Saflufenacil Resistance
Saskatchewan
The B. scoparia accession collected near Kindersley, SK, Canada, in 2021 was highly resistant to foliar-applied saflufenacil. KindersleyR exhibited 57.0- to 87.2-fold resistance to saflufenacil based on biomass DW (Table 3; Figures 2 and 4). The differential response of KindersleyR to saflufenacil compared with the susceptible control accessions was evident visually by 7 DAT and extended to at least 21 DAT (Table 3; Figure 4). Visible control resistance indices ranged from 46.9- to 47.4-fold resistance at 7 DAT and increased to 56.5- to 101.1-fold resistance by 21 DAT (Table 3; Figure 4). The estimated rate of saflufenacil causing 80% reduction in biomass DW was 126.9 g ai ha−1 (Table 4). This was well above the high field rate registered in western Canada (50 g ai ha−1) (Anonymous 2024a, 2024b). It was also >125-fold greater than the saflufenacil rate causing 80% decline in biomass DW for the susceptible control accessions, EastendS and RosetownS (0.8 and 1.0 g ai ha−1, respectively).
Table 3. Saflufenacil and carfentrazone resistance indices for one putative protoporphyrinogen oxidase (PPO) inhibitor–resistant Bassia scoparia accession collected from Saskatchewan in 2021 and two putative PPO inhibitor–resistant Bassia scoparia accessions collected from North Dakota in 2022 compared with two locally relevant susceptible control accessions based on visible control at 7 and 21/28 d after treatment (DAT) and biomass fresh weight (FW) and dry weight (DW) at 21/28 DAT.
a Significant difference of the resistance index from unity at *P < 0.05; **P < 0.01; ***P < 0.001.
b R/S1 indicates the resistance index relative to the first susceptible control accession; EastendS for Saskatchewan and FargoS for North Dakota.
c R/S2 indicates the resistance index relative to the second susceptible control accession; RosetownS for Saskatchewan and MinotS for North Dakota.
Figure 4. The response of one putative-resistant (KindersleyR) and two susceptible (RosetownS and EastendS) Bassia scoparia accessions from Saskatchewan, Canada, to a range of foliar-applied saflufenacil rates based on visible control at (A) 7 and (B) 21 d after treatment (DAT) and shoot biomass (C) fresh weight (FW) and (D) dry weight (DW) at 21 DAT. Dots indicate treatment means; bars represent standard errors. Embedded text indicates the resistance index (R/S ratio) for the putative-resistant accession relative to each susceptible accession.
Table 4. Regression parameter estimates for the three-parameter log-logistic model fit to describe the response of three Saskatchewan Bassia scoparia accessions to a rate titration of saflufenacil or carfentrazone based on visible control at 7 and 21 d after treatment (DAT) and shoot biomass fresh weight (FW) and dry weight (DW) at 21 DAT a .
a
Abbreviations:
$b$
, slope of the response curve at inflection point;
$d$
upper asymptote;
$e$
, response curve inflection point considered ED50 for visible control or GR50 for biomass; ED80, effective dose of herbicide resulting 80% visible control (ED80) or biomass reduction (GR80); RSE, residual standard error.
North Dakota
The B. scoparia accessions collected near Mandan and Minot, ND, USA, in 2022 were highly resistant to saflufenacil, similar to the KindersleyR accession. The saflufenacil resistance indices for MinotR were about one-quarter that of MandanR. For example, MandanR exhibited 204.0- to 320.5-fold resistance, while MinotR exhibited 45.4- to 71.3-fold resistance to foliar-applied saflufenacil based on biomass DW (Table 3; Figures 3 and 5). While both accessions were highly resistant based on guidelines recommended by the Global HRAC (HRAC 2024a), the difference in resistance indices between these two accessions was due to very low GR50 values for the susceptible control accessions that were ≤0.7 g ai ha−1 of saflufenacil (Table 5). Like the Saskatchewan accessions, differential response of the resistant from the susceptible North Dakota accessions was evident by 7 DAT and extended to at least 28 DAT (Table 3; Figure 5). However, resistance indices based on visible control ratings were not statistically different from 1 despite R/S ratios that were ≥292.8 (Table 3). This was due, in part, to variability around the dose–response model inflection point (Table 5), which could reflect the variable nature of the North Dakota field-collected samples that were absentregeneration under controlled environment or incomplete (70% to 72%) visible control of the resistant accessions at the highest saflufenacil rate (2,500 g ai ha−1) (Figure 5). Mean visible control at 28 DAT for the susceptible accessions increased from 3% at 0.25 g ha−1 of saflufenacil to 88% at 2.5 g ai ha−1, also contributing to inaccurate estimation of the ED50 values and no statistical difference when determining the visible control resistance indices. The estimated saflufenacil rate causing 80% reduction in biomass DW was 485.6 and 387.8 g ai ha−1 for MandanR and MinotR compared with 0.9 and 0.4 g ai ha−1 for FargoS and MinotS (Table 5), about 16 to 18 times the typical U.S. burndown rate of 25 g ai ha−1 (Ikley et al. Reference Ikley, Christoffers, Dalley, Endres, Gramig, Howatt, Jenks, Law, Lim, Ostlie, Peters, Robinson, Thostenson and Valenti2024).
Figure 5. The response of two putative-resistant (MandanR and MinotR) and two susceptible (FargoS and MinotS) Bassia scoparia accessions from North Dakota, USA, to a range of foliar-applied saflufenacil rates based on visible control at (A) 7 and (B) 28 d after treatment (DAT) and shoot biomass (C) fresh weight (FW) and (D) dry weight (DW) at 28 DAT. Dots indicate treatment means; bars represent standard errors. Embedded text indicates the resistance index (R/S ratio) for the putative-resistant accession relative to each susceptible accession.
Table 5. Regression parameter estimates for the three-parameter log-logistic model fit to describe the response of four North Dakota Bassia scoparia accessions to a rate titration of saflufenacil or carfentrazone based on visible control at 7 and 21/28 d after treatment (DAT) and shoot biomass fresh weight (FW) and dry weight (DW) at 21/28 DAT a .
a
Abbreviations:
$b$
, slope of the response curve at inflection point;
$d$
upper asymptote;
$e$
, response curve inflection point considered ED50 for visible control or GR50 for biomass; ED80, effective dose of herbicide resulting 80% visible control (ED80) or biomass reduction (GR80); RSE, residual standard error.
Carfentrazone Resistance
Saskatchewan
The KindersleyR accession was highly resistant to foliar-applied carfentrazone. KindersleyR exhibited 97.0- to 120.9-fold resistance to carfentrazone based on biomass DW, compared with the two susceptible control accessions (Table 3; Figures 2 and 6). Like the response to saflufenacil, differential response to carfentrazone was obvious by 7 DAT and extended to 21 DAT (Table 3; Figure 6). The estimated carfentrazone rate causing 80% reduction in biomass DW of KindersleyR was 157.9 g ai ha−1, which was well above that for the susceptible accessions (2.0 to 3.0 g ai ha−1) (Table 4) and 6 to 18 times the registered burndown field rates (9 to 28 g ai ha−1) for carfentrazone in western Canada (Anonymous 2024a, 2024b).
Figure 6. The response of one putative-resistant (KindersleyR) and two susceptible (RosetownS and EastendS) Bassia scoparia accessions from Saskatchewan, Canada, to a range of foliar-applied carfentrazone rates based on visible control at (A) 7 and (B) 21 d after treatment (DAT) and shoot biomass (C) fresh weight (FW) and (D) dry weight (DW) at 21 DAT. Dots indicate treatment means; bars represent standard errors. Embedded text indicates the resistance index (R/S ratio) for each putative-resistant accession relative to each susceptible accession.
North Dakota
The MandanR and MinotR accessions were also highly resistant to carfentrazone. For example, resistance indices based on biomass DW ranged from 110.5- to 330.1-fold for MandanR and from 88.4- to 264.1-fold for MinotR (Table 3; Figures 3 and 7). However, resistance indices based on GR50 values were not significantly different from 1 (α = 0.05) for these North Dakota accessions (Table 3). Despite this, resistance indices based on visible control at 21 DAT were significantly different from 1 and ranged from 515.7- to 1,008.1-fold resistance for MandanR and from 107.0- to 210.9-fold resistance for MinotR. Like the response of these accessions to saflufenacil, high resistance indices but lack of significant differences from unity for some response variables and not others were caused by a combination of incomplete (50% to 76%) control of the resistant accessions at the highest carfentrazone rate (1,750 g ai ha−1), high variability around the model inflection point, and natural variability in the first-generation field-collected samples (Table 5; Figure 7). Nevertheless, taken together, observations across response variables, accessions, and locations suggest that these putative PPO inhibitor–resistant B. scoparia accessions were highly resistant to carfentrazone (HRAC 2024a).
Figure 7. The response of two putative-resistant (MandanR and MinotR) and two susceptible (FargoS and MinotS) Bassia scoparia accessions from North Dakota, USA, to a range of foliar-applied carfentrazone rates based on visible control at (A) 7 and (B) 21 d after treatment (DAT) and shoot biomass (C) fresh weight (FW) and (D) dry weight (DW) at 21 DAT. Dots indicate treatment means; bars represent standard errors. Embedded text indicates the resistance index (R/S ratio) for the putative-resistant accession relative to each susceptible accession.
Untreated B. scoparia plants in the Saskatchewan experiments grew approximately 5 to 10 times larger than those in the North Dakota experiments (Figures 2–7), which may have contributed to the larger resistance indices observed in North Dakota than Saskatchewan (Table 3). These differences in biomass were caused by earlier herbicide treatment (2- to 3-cm height) and more restricted B. scoparia growth in the 4-cm-diameter Cone-tainers used in North Dakota compared with slightly later treatment (5- to 8-cm height) and larger (10 by 10 cm) pots used in Saskatchewan. The smaller size of the Cone-tainers may have restricted growth of the North Dakota plants and potentially also led to nutrient deficiency by 21/28 DAT. Indeed, the untreated plants grown in Saskatchewan appeared visually healthy at 21 DAT (Figure 2), while those at 21/28DAT in North Dakota appeared less so (Figure 3). Despite this, apparent stress to the B. scoparia plants in the North Dakota experiments did not seem to influence herbicide efficacy, as the effective doses for control of the susceptible accessions remained similar between experiments conducted at both locations (Tables 4 and 5; Figures 4–7). Interestingly, untreated plants from both susceptible accessions accumulated less biomass than the resistant accessions in the carfentrazone but not the saflufenacil experiments in North Dakota (Figures 5 and 7). This difference could be explained, in part, by the difference in timing of the biomass measurements in North Dakota, which took place at 28 DAT for saflufenacil and 21 DAT for carfentrazone, while greater heterogeneity of these field-collected samples likely also played a role.
To date, 17 different weed species have evolved resistance to PPO-inhibiting herbicides globally, and the majority of cases reporting PPO-inhibitor resistance in the international database also report resistance to other herbicide sites of action (i.e., cross- or multiple resistance) (Barker et al. Reference Barker, Pawlak, Duke, Beffa, Tranel, Wuerffel, Young, Porri, Liebl, Aponte, Findley, Betz, Lerchl, Culpepper, Bradley and Dayan2023; Heap Reference Heap2024). PPO inhibitor–resistant waterhemp [Amaranthus tuberculatus (Moq.) Sauer] was documented in Kansas in 2001, representing the first case of resistance to this site of action among weed species (Shoup et al. Reference Shoup, Al-Khatib and Peterson2003). Since then, PPO inhibitor–resistant weeds have been documented in 10 countries worldwide and in a range of crop species (Heap Reference Heap2024). Bassia scoparia represents the fourth and seventh weed species to evolve PPO-inhibitor resistance in Canada and the United States, respectively. It remains unknown whether the three PPO inhibitor–resistant B. scoparia accessions identified in the current study also exhibit resistance to other herbicide sites of action. This knowledge gap is one focal point of several new questions regarding PPO inhibitor–resistant B. scoparia that warrants further investigation.
Similar to cases of PPO-inhibitor resistance in some other weed species (Dayan et al. Reference Dayan, Barker and Tranel2018), the resistant B. scoparia plants exhibited initial necrosis after foliar treatment with saflufenacil or carfentrazone followed by healthy new regrowth shortly thereafter (CMG and QDL, personal observations). The initial symptomology typical of foliar treatment with PPO-inhibiting herbicides could make field diagnostics difficult if scouting is conducted shortly after application. However, differential control of the resistant and susceptible accessions was evident visually by 7 DAT under controlled-environment conditions (Figures 4–7), which may also translate to a field scenario. Results from the current study suggest that field-scouting efforts to identify PPO inhibitor–resistant B. scoparia should be effective when conducted between 1 and 3 wk after foliar treatment, but may be more obvious later given the initial necrosis injury observed after treatment of resistant plants.
The current study showed that the PPO inhibitor–resistant B. scoparia accessions exhibited cross-resistance to two chemical families of PPO-inhibiting herbicides (Table 3; Figures 2–7); saflufenacil belonging to the N-phenylimides and carfentrazone belonging to the N-phenyltriazolinones (HRAC 2024b). Cross-resistance in B. scoparia among the other families of PPO inhibitors warrants further research. Indeed, variable cross-resistance to PPO-inhibiting chemical families has been noted in other weed species and depends on herbicide application method and timing, the weed species, and the resistance mechanism (Barker et al. Reference Barker, Pawlak, Duke, Beffa, Tranel, Wuerffel, Young, Porri, Liebl, Aponte, Findley, Betz, Lerchl, Culpepper, Bradley and Dayan2023). Further research aimed at understanding the mechanism conferring PPO-inhibitor resistance in B. scoparia may help further elucidate the associated pattern of cross-resistance.
Practical Implications
Spread of PPO inhibitor–resistant B. scoparia could limit options for herbicidal control, especially given the likely stacking of multiple resistance traits in this species resulting in resistance across a wide range of herbicide sites of action. Herbicide resistance traits can evolve and spread efficiently in B. scoparia, as demonstrated by the rapid increase in frequency of populations resistant to glyphosate and ALS-inhibiting herbicides in recent decades (Geddes et al. Reference Geddes, Pittman, Gulden, Jones, Leeson, Sharpe, Shirriff and Beckie2022c, Reference Geddes, Pittman, Hall, Topinka, Sharpe, Leeson and Beckie2023; Kumar et al. Reference Kumar, Jha, Jugulam, Yadav and Stahlman2019; Sharpe et al. Reference Sharpe, Leeson, Geddes, Willenborg and Beckie2023; Westra et al. Reference Westra, Nissen, Getts, Westra and Gaines2019). In B. scoparia, rapid evolution and spread of these traits is due to a combination of ample selection pressure due to heavy use of herbicides across a large area, high genetic diversity (Martin et al. Reference Martin, Benedict, Sauder, Beckie and Hall2020), and efficient seed- and pollen-mediated gene flow (Beckie et al. Reference Beckie, Blackshaw, Hall and Johnson2016). PPO-inhibitor resistance in B. scoparia will create a gap, particularly during the preplant/preemergence weed control window in several field crops grown in the Northern Great Plains (Tables 6 and 7). As multiple herbicide resistance traits continue to stack in this species, the available options for herbicidal control become limited, causing reliance on contact-type herbicides like glufosinate (HRAC Group 10) postemergence in crops engineered to resist this glutamine synthetase inhibitor or the PSII inhibitor bromoxynil (HRAC Group 6) alone or mixed with an inhibitor of 4-hydroxyphenylpyruvate dioxygenase (HRAC Group 27). One key difference between Canada and the United States, among others, is the commercial availability of the photosystem I–inhibiting herbicide paraquat (HRAC Group 22) in the United States (Ikley et al. Reference Ikley, Christoffers, Dalley, Endres, Gramig, Howatt, Jenks, Law, Lim, Ostlie, Peters, Robinson, Thostenson and Valenti2024) but not Canada (Anonymous 2024a, 2024b), which further limits herbicidal control options north of the Canada/U.S. border.
Table 6. Herbicide options registered for Bassia scoparia control or suppression in western Canada assuming blanket resistance to all active ingredients within Herbicide Resistance Action Committee (HRAC) Groups 2, 4, 9, and 14a,b .
a Adapted from Anonymous (2024a, 2024b); C indicates control (≥80% control), S indicates suppression (60–79% control).
b Wheat, Triticum aestivum L.; barley, Hordeum vulgare L.; oat, Avena sativa L.; corn, Zea mays L.; canola, Brassica napus L.; mustard, Brassica juncea (L.) Czern. or Sinapis alba L.; flax, Linum usitatissimum L.; soybean, Glycine max (L.) Merr.; field pea, Pisum sativum L.; lentil, Lens culinaris Medik.
c Preplant incorporated or late fall applied.
d Yellow mustard only.
e Mixed with glyphosate.
f Glufosinate-resistant varieties.
g Must be applied with tank-mix partner.
Table 7. Herbicide options registered for Bassia scoparia control or suppression in the United States assuming blanket resistance to all active ingredients within Herbicide Resistance Action Committee (HRAC) Groups 2, 4, 9, and 14a,b .
a Adapted from Ikley et al. (Reference Ikley, Christoffers, Dalley, Endres, Gramig, Howatt, Jenks, Law, Lim, Ostlie, Peters, Robinson, Thostenson and Valenti2024); C indicates control (≥80% control), S indicates suppression (60–79% control).
b Wheat, Triticum aestivum L.; barley, Hordeum vulgare L.; oat, Avena sativa L.; corn, Zea mays L.; canola, Brassica napus L.; mustard, Brassica juncea (L.) Czern. or Sinapis alba L.; flax, Linum usitatissimum L.; soybean, Glycine max (L) Merr.; dry pea, Pisum sativum L.; lentil, Lens culinaris Medik.
c Preplant incorporated or late fall applied.
d Glufosinate-resistant varieties.
Careful stewardship of herbicides that remain effective on multiple herbicide–resistant B. scoparia is warranted through further and more targeted implementation of integrated weed management programs. Past research has shown that B. scoparia responds to competitive crop scenarios by substantially reducing seed production (Mosqueda et al. Reference Mosqueda, Lim, Sbatella, Jha, Lawrence and Kniss2020). Management practices targeting B. scoparia seed production and return to the soil seedbank represent a key choking point in the life cycle of this weed (Geddes and Davis Reference Geddes and Davis2021) due to short seed longevity once it enters the soil seedbank (Beckie et al. Reference Beckie, Blackshaw, Leeson, Stahlman, Gaines and Johnson2018; Dille et al. Reference Dille, Stahlman, Du, Geier, Riffel, Currie, Wilson, Sbatella, Westra, Kniss, Moechnig and Cole2017; Geddes Reference Geddes2021; Schwinghamer and Van Acker Reference Schwinghamer and Van Acker2008). Integrating nonchemical practices, such as competitive crops (Mosqueda et al. Reference Mosqueda, Lim, Sbatella, Jha, Lawrence and Kniss2020), alternative crop life cycles (Geddes and Davis Reference Geddes and Davis2021), higher crop seeding rates (Geddes and Kimmins Reference Geddes and Kimmins2021), strategic yet judicious tillage (Obour et al. Reference Obour, Holman, Simon and Schlegel2021), or cutting B. scoparia for animal feed (Nair et al. Reference Nair, Lima, Abdalla, Molnar, Wang, McAllister and Geddes2021), may serve to reduce B. scoparia seedbank replenishment. In addition, cutting or mowing B. scoparia plants could help mitigate the globe-shaped growth structure characteristic of tumbleweeds by physically disrupting unfettered growth and development, thereby preventing B. scoparia movement beyond its source location. Physical barriers like fence lines or shelterbelts may also serve to mitigate seed-mediated gene flow by catching B. scoparia plants that move beyond field boundaries (Beckie et al. Reference Beckie, Blackshaw, Hall and Johnson2016; Geddes and Sharpe Reference Geddes and Sharpe2022). Mitigation efforts should employ the core foundational principles of integrated weed management wherein multiple cultural, physical, and biological tactics are implemented along with strategic herbicide use to limit B. scoparia proliferation. Of utmost importance will be continued investment in the design and understanding of sustainable integrated weed management strategies that target the unique biology of this troublesome and highly elastic species.
Acknowledgments
We thank Bailey Maurstad for assisting with the research at North Dakota State University. We also thank the Saskatchewan agronomist who brought the initial observation of lack of control to our attention and the greenhouse staff at the Lethbridge Research and Development Centre for maintaining the growth environment.
Funding statement
Funding for the research conducted at the Lethbridge Research and Development Centre was provided by the Prairie Oat Growers Association, Manitoba Crop Alliance, Saskatchewan Oilseeds Development Commission, Saskatchewan Pulse Growers, Saskatchewan Wheat Development Commission, Western Grains Research Foundation, ADAMA Agricultural Solutions Canada Ltd., BASF Canada Inc., Bayer CropScience Inc., FMC of Canada Ltd., Gowan Agro Canada, Nufarm Agriculture Inc., Valent Canada Inc., and the Saskatchewan Ministry of Agriculture–Agriculture Development Fund through project no. ADF20230188. Research conducted at North Dakota State University was partially funded by the North Dakota Agricultural Experiment Station and USDA-NIFA Hatch Project no. ND01532.
Competing interests
IM and AP are employees of BASF. CMG received grant funding for this research from the crop protection companies: ADAMA Agricultural Solutions Canada Ltd., BASF Canada Inc., Bayer CropScience Inc., FMC of Canada Ltd., Gowan Agro Canada, Nufarm Agriculture Inc., and Valent Canada Inc.
Open access
