Article Cite This: J. Agric. Food Chem. 2019, 67, 6133−6142
pubs.acs.org/JAFC
Glyphosate and Aminomethylphosphonic Acid Content in Glyphosate-Resistant Soybean Leaves, Stems, and Roots and Associated Phytotoxicity Following a Single Glyphosate-Based Herbicide Application É lise Smedbol,*,† Marc Lucotte,† Sophie Maccario,† Marcelo Pedrosa Gomes,‡ Serge Paquet,§ Matthieu Moingt,† Lila Lucero Celis Mercier,§ Millaray Rayen Perez Sobarzo,∥ and Marc-Andre ́ Blouin∥ Downloaded via GUILFORD COLG on July 23, 2019 at 02:20:34 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
Université du Québec à Montréal, GEOTOP & Institut des Sciences de l’Environnement, 201 Avenue du Président-Kennedy, H2X 3Y7 Montréal, Québec, Canada ‡ Universidade Federal do Paraná, Departamento de Botânica, Setor de Ciências Biológicas, 80050-540 Curitiba, Paraná, Brazil § Université du Québec à Montréal, Département des Sciences Biologiques, 141 Avenue du Président-Kennedy, H2X 1Y4 Montréal, Québec, Canada ∥ Université du Québec à Montréal, Département de Chimie, 2101 rue Jeanne-Mance, H2X 2J6 Montréal, Québec, Canada ABSTRACT: Glyphosate-based herbicide (GBH) applications were reported to induce physiological damages to glyphosateresistant (GR) soybean, which were mainly attributed to aminomethylphosphonic acid (AMPA). In order to study glyphosate and AMPA dynamics in plants and associated phytotoxic effects, a greenhouse experiment was set where GR soybeans were exposed to GBH (0.7 to 4.5 kg glyphosate ha−1) and sampled over time (2, 7, 14, and 28 days after treatment (DAT)). Hydrogen peroxide content increased 2 DAT, while a decrease was observed for the effective quantum yield (2, 7, 14 DAT), stomatal conductance (2 DAT), and biomass (14 DAT). Glyphosate content was higher in leaves, followed by stems, and then roots. AMPA content tended to increase with time, especially in roots, and the amount of AMPA in roots was negatively correlated to mostly all phytotoxicity indicators. This finding is important since AMPA residues are measured in agricultural soils several months after GBH applications, which could impact productivity in GR crops. KEYWORDS: glyphosate and AMPA toxicokinetics, glyphosate-based herbicides, glyphosate-resistant soybean, H2O2, stomatal conductance, biomass, effective quantum yield
1. INTRODUCTION Glyphosate-based herbicides (GBH) applications on glyphosate-resistant (GR) crops is now the main weed control strategy worldwide in field crops, and approximately 78% of all soybean crops are GR.1 While glyphosate (N(phosphonomethyl)glycine) has previously been described as slightly mobile in plants, residues of this molecule, as well as aminomethylphosphonic acid (AMPA), its main metabolite, are measured in grains several weeks and/or months after GBH applications.2−4 Glyphosate and AMPA are also detected in GR soybean leaves and stems at all growth stages, following GBH applications.5 The glyphosate fate in plants after foliar application is relatively well documented; however, very little is known about that of AMPA. Glyphosate, the active molecule of GBH, is a phosphonic acid, usually sprayed in a water-soluble form (isopropylamine salt) and is best absorbed by plants with the addition of surfactants contained in the commercial formulations.6 After foliar application, it is proposed that glyphosate is absorbed through cuticular uptake and/or stomatal infiltration at varying rates following the species tested, the applied concentration, and the surfactant concentration.7 Different degrees of glyphosate absorption are reported in the literature, ranging from 4% to 70% of the applied dose.7−9 About 40−45% of the absorbed glyphosate is then systematically translocated in the entire plant.7 After reaching the vascular © 2019 American Chemical Society
tissues, glyphosate is preferably distributed toward specific tissues following the photoassimilation pathway.10 The half-time dissipation of glyphosate measured in GR maize leaves and stems is estimated at 5.39 and 4.02 DAT in plants exposed to glyphosate doses of 0.8 and 1.6 kg ha−1, respectively.11 Glyphosate dissipation in plants can occur by its metabolization into AMPA, particulary so in the second generation of GR plants expressing glyphosate oxidoreductase (GOX) enzyme, isolated from Ochrobactrum anthropi.12 Indeed, these GR crops are resistant to GBH through its metabolization into AMPA, by the expression of GOX which cleaves the glyphosate C−N bond.13 Thus, in the case of GR plants expressing GOX, AMPA residues are expected to be present. The second resistance pathway in GR crops is the introduction of the CP4 gene isolated from Agrobacterium sp., coding for an insensitive 5-enolpyruvolshikimate 3-phosphate synthase (EPSPS) enzyme.14 All GR crops have an insentitive EPSPS, while some crops have both an insensitive EPSPS and express GOX.15 However, AMPA residues are also measured in GR plants not expressing GOX but only an insensitive EPSPS, as well as in non-GR plants,9,13 Received: Revised: Accepted: Published: 6133
February 8, 2019 May 6, 2019 May 8, 2019 May 8, 2019 DOI: 10.1021/acs.jafc.9b00949 J. Agric. Food Chem. 2019, 67, 6133−6142
Article
Journal of Agricultural and Food Chemistry
a three hour period) and rinsed, and the fresh weight (FW) was measured. The third fully expanded leaves were frozen in liquid nitrogen and stored at −80 °C while the rest of the plants was frozen at −20 °C for further measurements. Dry weight (DW) measurements were performed later by drying them to a constant weight at 60 °C. 2.2. Photosynthesis and Pigment Evaluations. Chlorophyll fluorescence measurements were performed using a portable chlorophyll fluorometer PAM-2500 (Waltz, Effeltrich, Germany). Samples were dark acclimated for 15 min and then exposed to ten actinic light intensities (11, 30, 47, 71, 113, 168, 257, 380, 570, and 846 μmol photons m−2 s−1) for 20 s intervals, with a saturating pulse at the end of each interval. The effective quantum yield was calculated as
which suggests another pathway for glyphosate metabolization into AMPA in plants. Glyphosate dissipation in plants could also occur via its excretion in root exudates, which is estimated to be in the order of 8−12% of the applied glyphosate dose.16 Moreover, nontreated plants may absorb glyphosate via the rhizosphere from treated plants.17 Whether or not this phenomenon may also occur for AMPA is actually unknown. Since GR plants are not resistant to AMPA, it is suggested that glyphosate degradation into AMPA in plant tissues could be responsible for the phytotoxity observed in GR crops, such as reduced photosynthesis, stomatal conductance and transpiration,18 or reduced chlorophyll content and biomass,19 occurring after GBH applications. Similar phytotoxic effects are also reported in non-GR plants following GBH applications, with, for example, reduced photosynthesis and chlorophyll content in phytoplankton,20,21 reduced CO2 assimilation in non-GR soybean,22 reduced ribulose 1,5-biphosphate carboxylase oxygenase (Rubisco) activity,23 or carbohydrates accumulation in leaves.24 Since these effects on photosynthesis and carbon metabolism can occur in both GR and non-GR species, the phytotoxicity observed after GBH applications could result either from the effects of glyphosate, AMPA, or both molecules simultaneously. In non-GR plants, it was proposed that AMPA could reduce chlorophyll biosynthesis by competing with δaminolevulinic acid (ALA) synthetase or glycine, while glyphosate seemed to induce chlorophyll degradation through the enhanced formation of reactive oxygen species.25 This suggests that glyphosate and AMPA may interfere with different mecanisms that both result in a decrease of the chlorophyll content. The main objective of this study is to investigate glyphosate degradation into AMPA in GR soybean, as well as to relate the distribution of both molecules among plant tissues with observed phytotoxic effects upon GBH application. The effects of GBH applications on phytotoxicity indicators, such as the effective quantum yield (ΦPSII), hydrogen peroxide content (H2O2), biomass, chlorophyll content, and stomatal conductance, have been evaluated over the first month following a single GBH application along with the determination of glyphosate and AMPA content.
ΦPSII = (Fm′ − Ft)/Fm′
(1)
where Fm′ is the maximal fluorescence in light and Ft is the fluorescence level before light illumination.27 Chlorophyll content was measured using an atLEAF PLUS chlorophyll meter (atLeaf, Delaware, United States). Data were converted in SPAD units28 with a coefficient of determination (r2) of 0.9999 and then in chlorophyll content (μg cm−2) with an r2 value of 0.9883. 2.3. Gas Exchange. Stomatal conductance (gs) measurements were performed using a SC-1 leaf porometer (Decagon Devices Inc., Washington, United States; expressed in mmol m−2 s−1). The accuracy of the leaf porometer was of ±10% of the measurement. Three measurements were performed on each leaf, and the mean value was used as the stomatal conductance value. The porometer was calibrated under the greenhouse conditions prior to measurements for each time point. 2.4. Oxidative Stress Marker. Hydrogen peroxide (H2O2) content in soybeans was measured following the method by Velikova et al.29 Briefly, 200 mg of frozen leaves was ground on ice with 2 mL of trichloroacetic acid (0.1%, w/v), and then the mixture was centrifuged for 15 min (12 000g) at 4 °C. Next, 0.5 mL of the supernatant was added to 0.5 mL of 10 mM phosphate buffer (KH2PO4 + K2HPO4, pH 7) and 1 mL of 1 M potassium iodide. Absorbance was read at 390 nm, and values were reported on a standard curve with H2O2 concentrations ranging from 0 to 9 μg mL−1. 2.5. Glyphosate and AMPA Content. Glyphosate and AMPA content were determined following the slighly modified method described in Gomes et al.25 In short, 50 mg of fresh material (leaves, stems, and roots) were weighed and homogenized in a 50 mL Falcon tube containing 10 mL of deionized water, 10 mL of HPLC grade methanol, and 5 mL of GC grade dichloromethane for 1 min (25000 rpm) using a T18 digital ULTRA-TURRAX (IKA, North Carolina, USA) instrument. Falcon tubes were then centrifuged for 20 min at 3400 rpm. Next, 50 μL (leaves and stems) or 100 μL (roots) of supernatant was transferred and evaporated to dryness under nitrogen flow. Samples were derivatized by the addition of 1 mL of trifluoroacetic anhydride (TFAA) and 500 μL of trifluoroethanol (TFE), prior to heating at 100 °C for 1 h. After they were cooled to room temperature for 15 min, samples were evaporated to dryness under nitrogen flow and then resuspended with 1 mL of ethyl acetate prior to injection (0.5 μL). A Varian CP 3800 gas chromatograph coupled with an electron capture detector and equipped with a Rxi-5Sil MS column (Restek, Pennsylvania, USA) (30 m × 0.25 mm × 0.25 μm) was used for glyphosate and AMPA quantification. The injector and detector were held at 280 and 300 °C, respectively. Hydrogen was used as the carrier gas with a column flow of 1.8 mL min−1. The oven temperature program began at an initial temperature of 70 °C which was held for 0.8 min, followed by a 5 °C min−1 increase up to 130 °C and held for 5 min, and then followed by a 60 °C min−1 increase up to 250 °C and held for 12 min, for a total run time of 31.8 min. 2.6. Statistical Analyses. Analyses of variance were performed followed by multiple mean comparisons for ΦPSII, H2O2 content, stomatal conductance, chlorophyll content, and biomass (FW and DW), using Dunnett’s parametric test (DW, H2O2) or Steel’s nonparametric test when data did not fulfill the conditions of residuals normality (FW, ΦPSII, chlorophyll content, stomatal conductance), for comparison with the control treatment (0 kg ha−1).
2. MATERIALS AND METHODS 2.1. Growth Conditions and Glyphosate Applications. GR soybean seeds (5091RR2Y, É lite) were planted in two gallon (7.57 L) pots filled with the surface soil (0−30 cm) of an eutric Brunisol26 collected in Boisbriand (Québec, Canada). The Brunisol had a sandy loam texture with 3% organic matter, a pH of 7.59 ± 0.35, and a mineral content of 4980 ± 172 μg g−1 Ca, 124 ± 48 μg g−1 Al, 118 ± 3 μg g−1 Fe, 90 ± 4. μg g−1 Mg, 75 ± 3 μg g−1 K, 40 ± 2 μg g−1 Mn, 1.6 ± 0.1 μg g−1 Cu, and 1.4 ± 0.2 μg g−1 Zn. Soybeans were fertilized with ammonium nitrate (N−P−K of 34−0−0), but not inoculated with diazotrophs, one week prior to the GBH applications. GBH (Factor 540, Interprovincial Cooperative Limited, Saskatoon, Canada) was applied manually to V3 growth stage soybeans (three trifoliate leaves) at doses of glyphosate of 0.7, 1.8, 2.3, 3.6, and 4.5 kg ha−1. For each treatment, three plants (pseudoreplicates) were grown in three pots. Soybeans were harvested 2, 7, 14, and 28 DAT, and plants from each pot were pooled (n = 3). A similar number of plants not treated with glyphosate were used as controls (n = 3). Pots were randomly placed (random number generator) in the greenhouse under natural light conditions complemented with sodium vapor lamps to maintain a 12 h/12 h light/ dark cycle and a mean photosynthetic active radiation (PAR) level of 850 μmol photons m−2 s−1. Temperature was maintained between 20 and 25 °C. In vivo measurements were performed on the third fully expanded leaf on at least two plants from each pot at the four time points (2, 7, 14, and 28 DAT). All plants were then harvested (within 6134
DOI: 10.1021/acs.jafc.9b00949 J. Agric. Food Chem. 2019, 67, 6133−6142
Article
Journal of Agricultural and Food Chemistry
Figure 1. A−D: Effective quantum yield (ΦPSII) in leaves and biomass (E−H: fresh weight, I−L: dry weight) of GR soybeans when exposed to increasing glyphosate doses (0.7 to 4.5 kg glyphosate ha−1) at four time points (2, 7, 14, and 28 DAT) (n = 3). An asterisk (*) indicates a significant difference (p-value < 0.05) from the control treatment (0 kg ha−1) using Dunnett’s or Steel’s test. Spearman’s rank correlations were performed with parameters measured in leaves (ΦPSII, H2O2 content, chlorophyll content, stomatal conductance, glyphosate and AMPA content), glyphosate and AMPA content in roots, as well as biomass (DW). The Bonferroni correction was performed, and the p-value was adjusted to 0.01 (0.05/5 variables), with the tests being significant when the p-value < 0.01. All other statistical analyses were significant when the p-value < 0.05, and all analyses were performed using the JMP 14 software from SAS.
Two-factor analyses of variance (ANOVA) were also performed (DAT, plant tissue [leaves, stems, roots], DAT × plant tissue), followed by Tukey’s post hoc tests with glyphosate and AMPA content in soybean leaves, stems, and roots, for all GBH doses. Data were previously logarithmically (ln) converted in order to fulfill the conditions of residuals normality. When the interaction between DAT and the plant compartment was nonsignificant, the two-factor ANOVA was repeated without the interaction variable (DAT × plant tissue). 6135
DOI: 10.1021/acs.jafc.9b00949 J. Agric. Food Chem. 2019, 67, 6133−6142
Article
Journal of Agricultural and Food Chemistry
Figure 2. A−D: Chlorophyll content, E−H: H2O2 content, and I−L: stomatal conductance in GR soybean leaves following exposure to increasing glyphosate doses (0.7 to 4.5 kg glyphosate ha−1) at four time points (2, 7, 14, and 28 DAT) (n = 3). An asterisk (*) indicates a significant difference (pvalue < 0.05) from the control treatment (0 kg ha−1) using Dunnett’s or Steel’s test.
control treatment, at the dose of 4.5 kg glyphosate ha−1 for fresh weight and dry weight (Figure 1G, 1K). The plants’ fresh weight increased 7 and 28 DAT at doses of 3.6 and 0.7 kg glyphosate ha−1, respectively (Figure 1F, 1H). There was no other significant difference regarding biomass (FW and DW) for all other time points (Figure 1E, 1I, 1J, 1L). GBH applications significantly reduced stomatal conductance 2 DAT at the highest dose (4.5 kg glyphosate ha−1) (Figure 2I). This was followed by a significant increase of the stomatal conductance 14 DAT,
3. RESULTS 3.1. Physiological Indicators. The effective quantum yield was significantly reduced 2 DAT, for GBH applications with doses of 1.8 to 4.5 kg glyphosate ha−1 (Figure 1A). ΦPSII was also significantly reduced 7 DAT and 14 DAT at the higher dose of 4.5 kg glyphosate ha−1 (Figure 1B, 1C). No significant difference was observed 28 DAT (Figure 1D). Plant biomass was significantly reduced 14 DAT, when compared to the 6136
DOI: 10.1021/acs.jafc.9b00949 J. Agric. Food Chem. 2019, 67, 6133−6142
Article
Journal of Agricultural and Food Chemistry Table 1. Statistical Analyses for Glyphosate and AMPA Content Measured in GR Soybean Leaves, Stems, and Rootsa,b Glyphosate
AMPA
glyphosate applied dose (kg ha−1)
model
DAT
plant tissue (leaves, stems, roots)
DAT × plant tissue
0.7 1.8 2.3 3.6 4.5 0.7 1.8 2.3 3.6 4.5
