Introduction
From 2012 to 2022, U.S. growers planted more than 30 million hectares of soybean and 4 million hectares of cotton each year (USDA-NASS 2022). Weeds are the largest threat to United States soybean and cotton production, with the potential to decrease yields by ≥36% if left uncontrolled (Oerke Reference Oerke2006). In 2016, new transgenic cultivars became commercially available for soybean and cotton producers, with resistance to 2,4-D or dicamba, in addition to glufosinate and glyphosate, thereby increasing the number of over-the-top herbicide options for growers. Corn and soybean plants with resistance to 2,4-D- and dicamba were developed through the insertion of the AAD-1 (aryloxyalkanoate dioxygenase-1) transgene and dicamba monooxygenase gene, respectively, resulting in herbicide detoxification (Behrens et al. Reference Behrens, Mutlu, Chakraborty, Dumitru, Jiang, LaVallee, Herman, Clemente and Weeks2007; Braxton et al. Reference Braxton, Richburg, York, Culpepper, Haygood, Lovelace, Perry and Walton2017; Inman et al. Reference Inman, Jordan, York, Jennings, Monks, Everman, Bollman, Fowler, Cole and Soteres2016). The year after commercialization, herbicide formulations of 2,4-D and dicamba with low volatility received approval from the Environmental Protection Agencyfor use in these new soybean and cotton weed management systems. These two synthetic auxinic herbicides selectively control broadleaf weeds such as Palmer amaranth, and when applied in a timely manner, are effective at controlling weeds after they emerge (Cahoon et al. Reference Cahoon, York, Jordan, Everman, Seagroves, Culpepper and Eure2015; Manuchehri et al. Reference Manuchehri, Dotray and Keeling2017).
Before 2017, total dicamba use in the United States was estimated at less than 6 million kg per year (USGS 2021). Since the commercialization of dicamba-resistant crops and subsequent labeling of the herbicide for in-crop use, more than 15 million kg of dicamba is now applied across the United States; 10 of those 15 million kg were used in cotton and soybean fields in 2019. This is nearly 10 times the amount used in these cropping systems prior to 2017. The state of Tennessee accounts for approximately 5% of this dicamba use, despite planting fewer hectares in soybean and cotton compared with other states. Overreliance on a specific herbicide site of action can lead to increased selection for herbicide-resistant biotypes (Beckie and Rebound Reference Beckie and Rebound2009; Powles et al. Reference Powles, Preston, Bryanand and Jutsum1997).
In 2019, Tennessee growers began reporting to the Extension Service that dicamba was no longer controlling Palmer amaranth in their fields. A preliminary field screen for dicamba resistance was conducted at a grower’s field in Lauderdale County in 2020 and 2021. Additionally, in 2020, a field screen was conducted in Gibson County. The herbicide had been applied when Palmer amaranth reached 10 cm in height. Treatments consisted of dicamba applied at 0.56 (1×) and 1.12 (2×) kg ae ha−1. Dicamba was applied using a CO2-pressurized backpack sprayer equipped with Turbo TeeJet Induction 11002 (TeeJet® Technologies, Glendale Heights, IL) nozzles calibrated to deliver 140 L ha−1 at 4.8 kph using 220 kPa. In addition, a preliminary greenhouse screen of Palmer amaranth from Lauderdale County (two sites) and Gibson County, TN, was conducted in spring 2020. In that screen, dicamba was applied as described earlier but only at one rate (0.56 kg ha−1) to 10-cm-tall Palmer amaranth. In each of these screens, control of Palmer amaranth was ≤50% following timely applications of dicamba, prompting the larger survey and dose-response experiment.
A survey and seed collection of Palmer amaranth escapes in dicamba-resistant cotton and soybean fields in Tennessee was conducted in fall 2021 with the objective of determining whether poor control was due to an environmental phenomenon or to the development of dicamba-resistant Palmer amaranth in western Tennessee.
Materials and Methods
A greenhouse dose-response experiment was conducted in 2021 and 2022 at the West Tennessee AgResearch and Education Center in Jackson, TN (35.632003°N, 88.855874°W). Palmer amaranth seeds from 15 locations in western Tennessee where dicamba failures were reported were collected in fall 2021 (Table 1). A specific field history for most individual commercial field locations was unknown; however, extensive use of dicamba for the past two decades in burndown due to widespread no-till practices across the state and more recently in-crop use in Xtend® crops suggests heavy dicamba use regardless of location (USDA-NASS 2018). The specific field history for the Lauderdale County 1 location was known and consisted of Xtend® cotton planted from 2016 to 2021. The Gibson County 1 and 2 sites were planted with Xtend® cotton from 2016 to 2020, Enlist® cotton in 2021, and Xtend® cotton again in 2022. In 2019, both growers noticed a small area of escaped Palmer amaranth after multiple applications of dicamba at 0.56 kg ae ha−1. Seeds were collected from these fields after being brought to the authors’ attention by extension agricultural agents or crop consultants, and a preliminary greenhouse screen for dicamba resistance was conducted in 2020 prior to the survey at hand (results not shown).
Table 1. Palmer amaranth accessions screened for dicamba resistance.
Seeds from all 15 survey sites were processed and stored at 4 C for 4 wk before greenhouse trials were initiated. A known susceptible population of Palmer amaranth purchased from Azlin Seed Services (Leland, MS) was included for comparison. Palmer amaranth seeds were sprinkled on top of premoistened potting mix (Sta-Green Moisture Max Potting Mix) in 28 cm by 55 cm by 6 cm greenhouse trays (Greenhouse Megastore, Danville, IL). Seeds were covered with 0.5 cm of potting mix and received overhead watering. Trays were kept moist throughout the experiment using subsurface irrigation, and supplemental lighting was used to ensure a 16-h photoperiod; daytime temperature was set to 33 C, and nighttime temperature was 26 C. Once plants emerged, Palmer amaranth plants were thinned to one plant per 30 cm2, or approximately 50 per tray. Trays were arranged in a randomized complete block design, and each try was considered one plot, or experimental unit. The experiment was repeated two times with three replications, or trays, per population in each run.
Herbicide treatments were applied using a stationary greenhouse spray chamber (Devries Manufacturing, Hollandale, MN) calibrated to deliver 140 L ha−1 at 4.8 kph using 200 kPa from a boom set up with two Turbo TeeJet Induction 11002 (TeeJet® Technologies) nozzles. The herbicide was applied when Palmer amaranth plants reached 10 cm in height. Treatments consisted of dicamba (Xtendimax® with VaporGrip® Technology; Bayer CropScience, St. Louis, MO) applied at 0.14 (0.25×), 0.28 (0.5×), 0.56 (1×), and 1.12 (2×) kg ae ha−1. The 1× rate was based on the XtendiMax label in which 0.56 kg ha−1 (Anonymous 2022) is designated as the labeled over-the-top use rate for tolerant cotton and soybeans. Plants were placed in the greenhouse after application and grown for 21 d, after which the number of dead and live plants per flat were counted to calculate a percent mortality (control) and fresh weight of surviving plants was measured in grams.
Percent control and fresh weights were subjected to ANOVA using the GLIMMIX procedure in SAS software, version 9.4 (SAS Institute, Cary, NC) with Tukey’s honestly significant difference at α = 0.05 for means separation. Location, herbicide rate, and location*herbicide rate interactions were tested for significance. Single degree-of-freedom contrast statements were conducted to compare each suspected resistant accession to the susceptible check-by rate. Percent control was fit to a three-parameter sigmoidal curve using SigmaPlot 14.5 (Systat Software Inc, San Jose, CA) as suggested by Thornley and Johnson (Reference Thornley and Johnson1990), where Parameter a described the upper limit of control, Parameter b estimates the slope, and Parameter c represents the EC50 rate (Equation 1). The EC50 value was then subjected to ANOVA using the same methodology as the percent control and fresh weight values. Both replication and run were considered random effects in the model. Relative resistance factor was then calculated by dividing the EC50 estimate for each population by the EC50 estimate for the susceptible population.
$${\rm{y = a/}}\{ {\rm{1 + exp}}[ - ({\rm{rate}} - {\rm{c}}){\rm{/b}}]\} $$
Results and Discussion
Contrast statements used to compare the response of 15 Palmer amaranth accessions from Tennessee to a known susceptible check following increasing rates of dicamba showed a decrease in control at 0.14 kg ae ha−1 for 10 of 15 accessions (Table 2). Four Tennessee accessions (Carroll, Lauderdale 1, Lauderdale 2, and Dyer counties) were not effectively controlled at the 0.28 kg ae ha−1 rate. When using the 1× field rate (0.56 kg dicamba ha−1), Lauderdale 1 (1%), Lauderdale 2 (72%), Tipton (81%), and Gibson 3 (80%) County accessions exhibited less control than the susceptible check (95%). At 1.12 kg ae ha−1, dicamba controlled Palmer amaranth by 20%, 79%, and 82% at Lauderdale 1, Madison 1, and Dyer counties, respectively, while control for the susceptible check was 100%.
Table 2. Contrast statements comparing percent Palmer amaranth mortality of 15 accessions with a susceptible accession following increasing rates of dicamba.
Dicamba dose-response curves suggest that Palmer amaranth populations in Tennessee are segregating based on their relative susceptibility to dicamba (Figure 1). Eight accessions responded with higher tolerance or resistance to dicamba compared with the known susceptible check. Of those eight accessions, three showed less control at rates two to four times above the 0.56 kg ha−1 rate. The Lauderdale County 1 accession represented by the grey line showed an order of magnitude greater resistance to dicamba than all other accessions.
Figure 1. Dicamba dose response of 15 Tennessee accessions. The responses of Palmer amaranth to increasing rates of dicamba as described by Equation 1: y = a/{1+exp[-(rate-c)/b]} in which Parameter a described the upper limit of control, Parameter b estimates the slope, and Parameter c represents the EC50 rate.
The EC50 value for the susceptible check was 0.1262, indicating that this amount of dicamba per hectare would control 50% of the population (Table 3). Four Tennessee Palmer amaranth accessions had higher EC50 values than the susceptible check: Carroll County (0.2338), Lauderdale County 1 (1.7978), Lauderdale County 2 (0.3140), and Dyer County (0.2398). The relative resistance factor for Carroll, Lauderdale 2, and Dyer counties was between 1.85 and 2.49, whereas the relative resistance factor for the Lauderdale County 1 accession was 14.25, indicating that this population has developed a high level of resistance and can no longer be effectively controlled using dicamba. These results are consistent with reports from the grower who manages this field. Lauderdale and Tipton counties in Tennessee have been the epicenter for Palmer amaranth resistance to herbicides in previous years and is where one of the first Palmer amaranth populations that have resistance to glyphosate and protoporphyrinogen oxidase were discovered in the state (Copeland et al. Reference Copeland, Giacomini, Tranel, Montgomery and Steckel2018; Steckel et al. Reference Steckel, Main, Ellis and Mueller2008).
Table 3. Response of Tennessee Palmer amaranth accessions to dicamba in 2022.
a Abbreviations: EC50, half-maximal effective concentration; RRF, relative resistance factor.
b Means not followed by a common letter are significantly different (P < 0.05).
Fresh weight of surviving plants was measured 21 d after application. At less than 0.56 kg ae ha−1 of dicamba, an increase in biomass was observed in some accessions compared with the nontreated control of those same accessions. Because the location*rate interaction was not significant for fresh weights, but location was significant, fresh weight was averaged for each location and compared to the susceptible check (Table 4). Lauderdale County 1 (106%) and Carroll County (40%) were the only accessions to exhibit higher overall biomass as a percent of the nontreated control compared with the susceptible check (20%). These findings support the control results with the Lauderdale County 1 accession showing an actual biomass increase after a dicamba application compared to the same accession that was not treated.
Table 4. Palmer amaranth accession fresh weights 21 d following dicamba application.
a Means not followed by a common letter are significantly different (P < 0.05).
These data document a segregating population of Palmer amaranth to dicamba in Tennessee. It ranges from 11 accessions with control similar to the susceptible check to three accessions (Caroll, Dyer, Lauderdale 2) showing resistance ratios of of 1.85 to 2.49. The Lauderdale 1 accession is confirmed to be highly resistant with a resistant ratio of 14.25. Another step to confirm resistance is documenting heritability of the resistance between generations. The history of Palmer amaranth escaping dicamba in the Lauderdale 1 location from 2019 to 2021 in the grower’s field, preliminary field research, and this greenhouse dose response would indicate that the dicamba resistance has passed between generations. This demonstrates the dicamba resistance allele or alleles were passed from the 2019 Palmer amaranth generation to the 2020 and the 2021 generations. This research documents the first findings of Palmer amaranth control failures in cotton and soybean fields due to the evolution of dicamba resistance.
Dicamba resistance in Palmer amaranth greatly limits control options in cotton and soybean. Glyphosate-resistant Palmer amaranth was first documented in Tennessee in 2008 (Steckel et al. Reference Steckel, Main, Ellis and Mueller2008). By 2013, the glyphosate-resistant biotype had become predominant in western Tennessee and was becoming established in central Tennessee (Steckel Reference Steckel2013). Recent documentation of glufosinate-resistant Palmer amaranth (Priess et al. Reference Priess, Norsworthy, Godara, Mauromoustakos, Butts, Roberts and Barber2022) in the Arkansas county adjacent to Lauderdale County, Tennessee, calls into question whether the XtendFlex trait (Bayer CropScience, St. Louis, MO) that provides cotton and soybean resistance to dicamba, glyphosate, and glufosinate will be a viable weed management tool for this weed in future years.
Future research should be conducted to determine whether dicamba-resistant Palmer amaranth accessions are cross-resistant to 2,4-D. In additon, research designed to assess the mechanism or mechanisms of resistance with the Lauderdale County 1 accession will be conducted. Finally, weed management research needs to be conducted to determine how best to integrate herbicides and nonchemical tactics to better control Palmer amaranth.
Acknowledgments
We thank Cotton Incorporated for funding for this project. No other conflicts of interest are noted.
Open access
