Glyphosate Resistance in Giant Ragweed ( Ambrosia trifida L . ) from Mississippi Is Partly Due to Reduced Translocation

A giant ragweed population from a glyphosate-resistant (GR) soybean field in Mississippi, USA was suspected to be resistant to glyphosate. Greenhouse and laboratory studies were conducted to confirm and quantify the magnitude of glyphosate resistance in a resistant biotype selected from this population and to elucidate possible physiological and molecular mechanisms of glyphosate resistance. Glyphosate dose response studies indicated that ED50 (effective dose required to reduce plant growth by 50%) values for glyphosate-resistant (GR-MS) and glyphosate-susceptible (GS-MS) biotypes, based on percent injury, were 0.52 and 0.34 kg ae/ha glyphosate, respectively, indicating a 1.5-fold level of resistance in GR-MS. The absorption pattern of 14C-glyphosate in the two giant ragweed biotypes was similar throughout the measured time course of 168 h after treatment (HAT). The amount of 14C-glyphosate that translocated out of treated leaves of the GRMS and GS-MS plants was similar up to 24 HAT. However, the GS-MS biotype translocated more (71% and 76% of absorbed at 48 and 96 HAT, respectively) 14C-glyphosate than the GR-MS biotype (44% and 66% of absorbed at 48 and 96 HAT, respectively) out of the treated leaf. No target site mutation was identified at the Pro106 location of the EPSPS gene of the GR-MS biotype. The mechanism of resistance to glyphosate in giant ragweed from Mississippi, at least, is due to reduced glyphosate translocation.


Introduction
Glyphosate, a nonselective, broad-spectrum, systemic, postemergence herbicide, has been used extensively throughout the world in both crop and noncrop lands since its commercialization in 1974.With the introduction of glyphosate-resistant (GR) crops in the mid-1990s, glyphosate has been used selectively and predominantly for weed control in GR crops.The widespread adoption of GR crops around the world with the associated use of repeat glyphosate applications has resulted in the evolution of several GR weed biotypes.As of July 2015, resistance to glyphosate was confirmed in 32 weed species, including giant ragweed (Ambrosia trifida L.) [1].Giant ragweed, resistant to glyphosate, has been reported in Arkansas, Indiana, Iowa, Kansas, Kentucky, Minnesota, Mississippi, Missouri, Nebraska, Ohio, Tennessee, and Wisconsin in the US and in Ontario, Canada [1]- [7].
Giant ragweed is an erect summer annual weed that is commonly found in ditches, stream and creek banks, and roadsides.It has gained importance as an agronomic weed due to its competitive ability in row crops.For instance, interference from giant ragweed at densities of 1 to 13.8 plants/m 2 reduced corn (Zea mays L.) yield by 14% to 90% [8].Several traits of sweet corn ears were affected by competition from giant ragweed [9].Cotton (Gossypium hirsutum L.) lint yield was reduced by 50% when giant ragweed was present at a density of 0.26 plants/m crop row [10].
A state-wide survey was conducted in the summer of 2009 in Mississippi to document existing GR weed populations in agronomic crops.As a part of this survey, a grower from Tunica County in northwestern Mississippi reported lack of control of a giant ragweed population from glyphosate applications in GR soybean (Glycine max (L.) Merr.).The objectives of this research were to confirm and quantify the magnitude of glyphosate resistance in giant ragweed from northwestern Mississippi and to elucidate the physiological and molecular mechanisms of glyphosate resistance in this giant ragweed.

Seed Collection, Storage, Germination, Planting, Growth, and Herbicide Treatment Conditions
In the summer of 2009 mature giant ragweed plants suspected (apical and/or one or more lateral/terminal meristems uninjured and actively growing) to be resistant to glyphosate were identified in a field in Tunica County, Mississippi that had been continuously planted to GR soybean in 2009 and the preceding six years.Six giant ragweed plants were collected along with roots and transplanted to 10 L pots containing field soil and allowed to grow outdoors until they set seed.Giant ragweed inflorescences containing seeds were collected, kept separate for each individual plant (not bulked), and air-dried in a greenhouse (25˚C/20˚C day/night, 12-h photoperiod under natural sunlight conditions) for 7 d, cleaned, and stored at 0˚C until further use.Seeds from the above six plants were planted at 1.25-cm depth in 50-cm by 20-cm by 6-cm plastic trays with holes containing a commercial potting mix (Metro-Mix 360, Sun Gro Horticulture, Bellevue, WA, USA), watered, and incubated at 2˚C to 8˚C for 30 d. Thereafter seed trays were returned to above mentioned greenhouse conditions supplemented with high pressure sodium lights providing 400 mmol/m 2 /s.Two weeks after emergence, giant ragweed seedlings were transplanted into 6-cm by 6-cm by 6-cm pots containing the potting mix.Plants were fertilized once with a nutrient solution (Miracle-Gro, The Scotts Company LLC, Marysville, OH, USA) containing 200 mg/L each of N, P 2 O 5 , and K 2 O one week after transplanting and sub-irrigated as needed, thereafter.All herbicide treatments were applied with a moving nozzle sprayer equipped with 8002E nozzles (Spraying Systems Co., Wheaton, IL, USA) delivering 140 L/ha at 280 kPa to giant ragweed plants that were 15-cm tall and at the three-leaf (node) stage.Percent control [visible estimate of injury on a scale of 0 (no injury) to 100 (plant death)] was recorded 4 weeks after treatment (WAT).Seed from a designated glyphosate-susceptible (GS-MS) giant ragweed biotype was collected from a fence row along a non-agricultural wooded area in southern Bolivar County, Mississippi, processed, and screened as described for the suspected GR plants.The natural conditions at sites from where the parent GR and GS giant ragweed seed was collected were similar (data not shown) and were not expected to have an impact on the experimental studies.All studies were conducted from 2011 to 2013 at the Jamie Whitten Delta States Research Center of USDA-ARS in Stoneville, Mississippi.

Screening with a Discriminating Glyphosate Dose
In preliminary resistance screening studies, 25 plants each from seed of six original field-collected plants, sus-pected to be GR, and GS-MS, were treated with a 0.84 kg ae/ha rate of glyphosate (potassium salt, Roundup WeatherMAX ® , Monsanto Company, St. Louis, MO, USA) (data not shown).A single plant survived 4 WAT, matured, and produced the second generation seed.Additional screening experiments indicated that all the second generation plants survived a glyphosate treatment of 0.84 kg/ha (data not shown).This second generation GR seed, henceforth referred as GR-MS biotype, was used in all subsequent studies.

Glyphosate Dose Response
GR-MS and GS-MS giant ragweed plants were treated with glyphosate at 0, 0.21, 0.42, 0.84, 1.68, and 3.36 kg/ha, representing rates overlapping the 1X field application rate of 0.84 kg•ha −1 .The above range of rates was determined based on the preliminary screening results.Percent visual injury ratings (0 = no effect on growth, 100 = complete kill) were recorded 4 WAT.There were four replications per treatment and the experiment was conducted two times.

14 C-Glyphosate Absorption, Translocation, and Phosphor Imaging
GR-MS and GS-MS giant ragweed plants were treated with glyphosate as described before, except that one of the fully expanded leaves at the second node was covered with a water-resistant paper sleeve.This sleeve was removed immediately after herbicide treatment for subsequent (within 30 min of overspray) application of solutions containing 14 C-glyphosate ( 14 C-methyl labeled with 2.0 GBq/mmol specific activity, 99.5% radiochemical purity in an aqueous stock solution of 7.4 MBq/mL, American Radiolabeled Chemicals, Inc., St. Louis, MO, USA).A solution containing glyphosate at a final concentration equivalent to 0.84 kg/ha in 140 L was prepared using 14 C-glyphosate, a commercial formulation of glyphosate, and distilled water.A 10-µL volume of the treatment solution, containing 5 kBq of 14 C-glyphosate, was applied to the adaxial surface of the non sprayed third true leaf of 15-cm tall plants in the form of 25 droplets with a micro applicator.
Plants were harvested at 1, 4, 8, 24, 48, 96, and 168 h after 14 C-glyphosate treatment (HAT) for absorption and translocation measurements.The treated leaf was removed and immersed in 10 mL 10% methanol in a glass vial and gently shaken for 20 s to remove nonabsorbed 14 C-glyphosate remaining on the leaf surface.The washed leaf was rewashed with an additional 10 mL of 10% methanol.Two 1-ml aliquots of each leaf wash were mixed with 10 mL scintillation cocktail (Ecolume, ICN, Costa Mesa, CA, USA) to measure nonansorbed 14 C-glyphosate.
Phosphorimaged (see below) plants were carefully removed from the mounting paper and divided into treated leaf, shoot above treated leaf (SATL), shoot below treated leaf (SBTL), and roots for measuring translocation.The plant parts were wrapped in a single layer of tissue paper (Kimwipes, Kimberly-Clark Corporation, Roswell, GA), placed in a glass vial, and oven dried at 60˚C for 24 h.Oven-dried plant samples were combusted in a biological oxidizer (Packard Instruments Company, Downers Grove, IL) and the evolved 14 CO 2 was trapped in 10 mL Carbosorb E (Packard BioScience Company, Meridian, CT, USA) and 10 mL Permafluor E+ (Packard Bi-oScience).Radioactivity from leaf washes and oxidations was quantified using liquid scintillation spectrometry.The average recovery of applied 14 C-glyphosate was 93%, based on the sum of the radioactivity measured in all plant parts (absorption, expressed as percent of applied 14 C) and leaf washes.Total radioactivity recovered in all plant parts except the treated leaf was designated as translocated 14 C and expressed as percent of absorbed.There were four replications per treatment (harvest) time per plant type (GR-MS/GS-MS) and the experiment was conducted once.
GR-MS and GS-MS plants were processed for phosphorimaging prior to translocation analysis.The treated leaves from the giant ragweed plants were removed at respective harvest times to wash off unabsorbed radioactivity and set aside.The remaining above ground part of the plant was excised from roots and mounted on a 27 by 21.25 cm piece of plain white paper.Shoot parts were evenly spread and kept in place with thin strips of clear office tape.Care was taken to avoid contact of the washed treated leaf with other parts of the plant.Roots were gently rinsed with water to remove soil, blotted dry with paper towels, and mounted on a separate sheet as with the shoot.The mounted plant parts were pressed between one or more layers of newspaper, and bound with two hard cardboard sections.The assembled plant press was held together with large binder clips and stored at −20˚C for later drying.The plant samples were dried in a gravity convection oven at 60˚C for 24 h.Phosphorimaging was used to develop an image of the plant samples.After cooling the dried sample to room temperature, the plant was placed in a 20 by 40 cm exposure cassette (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA) and brought into contact with a storage phosphor screen (BAS IPSR 2025 E, GE Healthcare Bio-Sciences Corp.) under diffused lighting.The apparatus was placed in a dark cabinet for 24 h.A phosphorimager (Typhoon FLA 7000, GE Healthcare) was used to detect distribution of 14 C-glyphosate and develop an image.

EPSPS Sequence Analysis
Leaf tissue was removed from three each of GR-MS and GS-MS plants (15-cm tall and at the three-leaf (node) stage, nontreated with glyphosate) for RNA extraction.The tissue was incubated in RNAlater ® (Life Technologies, Grand Island, NY, USA) for five hours at 4˚C.The tissue was transferred to a 2 mL microcentrifuge tube containing 450 µL RLT buffer (Qiagen, Valencia, CA, USA) and a 6.35 mm ball bearing (Boca Bearing, Boynton Beach, FL, USA).The tissue was homogenized by placing the tubes in a 5 G high speed paint shaker (Fluid Management, Wheeling, IL, USA) for 6 min.The lysate was transferred to a fresh tube and the RNA extracted using an RNeasy plant mini-kit (Qiagen, Valencia, CA, USA).The quality and quantity of the RNA was determined by gel electrophoresis, A260, and A260/280 ratios.A 197 bp fragment of the EPSPS containing the Pro 106 codon was amplified using 1-step RT-PCR master mix (Fisher, Pittsburgh, PA, USA).The Pro 106 locus, standardized to Arabidopsis thaliana EPSPS sequence, represents the amino acid proline at location 106 in wild type/sensitive/susceptible biotypes of selected weeds.Amino acid substitutions (http://weedscience.org/mutations/mutationdisplayall.aspx?MOAID=12), due to mutations, in place of proline (and in addition, threonine at position 102 in goosegrass (Eleucine indica L.)] result in biotypes resistant to glyphosate.Primer sequences used were AtF1-5' ACATGCTTGGGGCTCTAAGAA 3' and AtR1-5' TTGAATTACCACCAGCAGCGGT 3'.The reaction was prepared as follows: 25 -50 ng RNA, 5 µmol forward and reverse primers, 1× Ready mix, 0.5 µL RT enhancer, 0.4 µL enzyme mix, and water to 10 µL.Cycle conditions were as follows: 50˚C for 15 min, 95˚C for 2 min, 30 cycles of 95˚C for 20 s, 56˚C for 30 s, and 72˚C for 1 min, and 72˚C for 5 min.For TOPO TA cloning, a second round of PCR was performed using the TAKARA LA PCR kit ver 2.1 (Fisher, Pittsburgh, PA, USA).Reactions were prepared as follows: 0.5 µL of the initial PCR reaction, 4 µmol forward and reverse primers, 2.5 mM Mg 2+ , 1x buffer, 400 µM dNTPs, 2.5 U of polymerase, and water to 25 µL.Cycle conditions were as follows: 94˚C for 1 min, 30 cycles of 94˚C for 30 s, 56˚C for 30 s, and 72˚C for 1 min, and 72˚C for 5 min.PCR products were gel extracted using a GenElute Gel Extraction kit (Sigma Aldrich, St Louis, MO, USA).The gel extracted products were examined by gel electrophoresis to assess quality.PCR products were ligated to cloning vector using a TA Cloning ® kit with pCR TM 2.1 vector (Life Technologies, Grand Island, NY, USA).Reactions were prepared as follows: 3 to 1 ratio of insert to vector (25 ng), 1× buffer, 5 units of ligase, and water to 10 µL.Reactions were incubated at room temperature for 1 hr.Chemically competent TOP10 cells were prepared and transformed according to previously described protocols [11].The transformants were screened by PCR amplification of the insert using the primers and cycle conditions described above.Glycerol stocks of positive transformants were prepared by adding 800 µL of an overnight culture to 200 µL of 80% glycerol and were stored at −80˚C.Three positive clones per ragweed biotype (GR-MS and GS-MS) were selected for sequencing.Cultures were prepared by inoculating 500 µL LB media (1% w/v tryptone, 0.5% w/v yeast extract, and 0.5% w/v NaCl) with 50 µg/mL amp with 10 µL of an overnight culture.The cultures were submitted to the Genomics and Bioinformatics Research Unit, USDA-ARS, in Stoneville, MS, USA for plasmid isolation and sequencing.Both DNA strands were covered in sequencing.The sequence data were analyzed and aligned using Geneious software [12] (Biomatters Ltd., Auckland, New Zealand).The purpose of the EPSPS sequence analysis is to acquire knowledge on the presence or absence of mutation-based target-site resistance in the GR-MS biotype of giant ragweed.

Statistical Analysis
All experiments were conducted using a completely randomized design.Data from all experiments, with the exception of EPSPS sequence analysis, were analyzed by ANOVA via the PROC GLM statement using SAS software (version 9.2, SAS Institute, Inc., Cary, NC, USA).No significant experiment effect was observed in repeated experiments; therefore, data from experiments were pooled.Nonlinear regression analysis was applied to fit a sigmoidal log-logistic curve of the form: where a is an asymptote, x and x 0 are the upper and lower response limits with the latter approaching 0, and b is the slope of the curve around x 0 , to relate the effect of glyphosate dose on giant ragweed control and HAT on 14 C-glyphosate absorption and translocation.Equation parameters were computed using SigmaPlot (version 11.0, Systat Software, Inc., San Jose, CA, USA).Treatment means in selected experiments were separated using Fisher's Protected LSD test at P = 0.05.

Glyphosate Dose Response
Response of giant ragweed biotypes to glyphosate dose is presented in Figure 1.ED 50 (dose required to reduce plant growth by 50%) values for the GR-MS and GS-MS biotypes, based on percent control, were 0.52 and 0.34 kg ae/ha glyphosate, respectively, indicating a 1.5-fold level of resistance.This resistance level is lower than previously reported cases of GR giant ragweed, 2.1-to 6.2-fold from Ohio [2], 5.3-fold from Tennessee [3], and 2.3-to 7.2-fold from Arkansas [6].GR giant ragweed from Ontario, Canada required 1.66 kg/ha of glyphosate to cause a 50% reduction in growth [7], which is more than three times that required by the GR-MS biotype from Mississippi.The GR-MS and GS-MS plants survived glyphosate up to 0.84 and 0.42 kg/ha, respectively.Two types of phenotypic responses are commonly associated with GR giant ragweed accessions treated with glyphosate.First, a rapid necrotic response results in death of mature leaves following a contact herbicide-like injury, but growing points remain uninjured [13] [14].Second, involves symptoms similar to those displayed by species susceptible to glyphosate, whereby treated plants cease growth for two weeks, turn mildly chlorotic, and resume growth [13] [14].The symptoms produced by the GR-MS biotype were similar to the slow response phenotype.The rapid necrotic response has been partly attributed to hydrogen peroxide (Tardif, personal communication) and salicyclic acid (Van Horn and Westra, personal communication) formation.Transcriptome analysis of GR and GS giant ragweed biotypes from the Midwest US indicated a rapid response similar to pathogen induced reaction in the former and a slower stress response in the latter [15].

14 C-Glyphosate Absorption, Translocation, and Phosphor Imaging
The absorption pattern of 14 C-glyphosate in the two giant ragweed biotypes was similar throughout the time course of 168 HAT (Figure 2).Glyphosate absorption measured at 24 and 72 HAT was similar between GR and GS giant ragweed biotypes from Arkansas [6], analogous to the findings reported here.The amount of 14 C-glyphosate that translocated out of the treated leaves of the GR-MS and GS-MS plants was similar up to 24 HAT (Figure 3).However, the GS-MS biotype translocated more (71% and 76% of absorbed at 48 and 96 HAT, respectively) 14 C-glyphosate than the GR-MS biotype (44% and 66% of absorbed at 48 and 96 HAT, respectively) out of the treated leaf, with translocated amounts leveling off to similar levels in both biotypes by 168 HAT (69% and 68% of absorbed in GR-MS and GS-MS, respectively).The translocation levels out of the treated leaf by 168 HAT observed here (68 to 69% of absorbed across the biotypes) are similar to the levels for giant ragweed (71% of absorbed at 168 HAT) previously reported [16].
Distribution of absorbed 14 C-glyphosate in the GR-MS and GS-MS biotypes is summarized in Table 1.The quantity of 14 C-glyphosate that remained in the treated leaf was higher in the GR-MS biotype (55% of absorbed) compared to the GS-MS biotype (30% of absorbed) at 48 HAT.The amounts of 14 C-glyphosate that accumulated in the shoot above treated leaf (SATL) were lower in the GR-MS biotype (14% and 17% of absorbed at 48 and 96 HAT, respectively) compared to the GS-MS biotype (35% and 37% of absorbed at 48 and 96 HAT, respectively).Similarly, the GR-MS biotype retained lower levels of 14 C-glyphosate (8% of absorbed) than the susceptible GS-MS biotype (16% of absorbed) at 48 HAT in SBTL.No difference in the accumulation of translocated 14 C-glyphosate in roots of the two biotypes was found, but the resistant GR-MS had higher levels (4% of absorbed) compared to 1% of absorbed in the GS-MS at 1 HAT.This difference in the roots is unlikely to have any effect on the overall physiology of the plants.
Phosphorimaging was used to generate autoradiographs of GR-MS and GS-MS plants treated with 14 C-glyphosate, shown in Figure 4, to visualize 14 C-glyphosate translocation patterns. 14C-glyphosate seemingly moved out of the treated leaf faster in susceptible plants than the resistant plants, with discernible accumulation of 14 C-glyphosate in the GS-MS biotype but not the GR-MS biotype at 4 HAT.Thereafter, 14 C-glyphosate was detectable in all tissues other than the treated leaf at all harvest times in both biotypes.The intensity of the autoradiograph of the shoot was perceptibly deeper for the GS-MS biotype compared to the GR-MS biotype 48 HAT.This observation is also reflected in the translocation and distribution data, where the GS-MS accumulated more 14 C-glyphosate than the GR-MS at 48 HAT (Figure 4).

EPSPS Sequencing
The conserved region of EPSPS containing the Pro106 location was sequenced from three plants each of GR-MS and GS-MS.Alignment of the sequences from the six plants showed that the target site mutation responsible for glyphosate resistance was absent in the resistant GR-MS (Figure 5).This indicates that a mechanism other than a point mutation in the conserved region of EPSPS is responsible for glyphosate resistance in GR-MS.

Conclusion
In summary, low level glyphosate resistance has been documented in a giant ragweed accession from Mississippi.The mechanism of resistance, at least in part, is due to reduced translocation of glyphosate and not because of a target site mutation of EPSPS.A very recent review on state of the knowledge on glyphosate resistance mechanisms in plants observes that GR giant ragweed mechanisms are not clearly understood [27].The research reported here will help clarify glyphosate resistance mechanism in some biotypes of this species.

Table 1 .
Distribution of14C-glyphosate in glyphosate-resistant and -susceptible giant ragweed biotypes.abc.05)nsnsns 5 ns ns ns ns ns 8 19 ns ns ns ns 4 ns ns 1 ns ns ns ns ns Distribution represents partitioning of absorbed14C-glyphosate between the treated leaf, shoot above treated leaf, shoot below treated leaf, and root); c LSD (0.05): A number indicates significance at the 5% level of probability and 'ns' indicates no significant difference between means within the same column.