Dissipation of the Herbicide Benzobicyclon Hydrolysate in a Model

Sep 29, 2017 - Department of Environmental Toxicology, College of Agricultural and Environmental Sciences, University of California, Davis, California...
3 downloads 25 Views 801KB Size
Article pubs.acs.org/JAFC

Cite This: J. Agric. Food Chem. 2017, 65, 9200-9207

Dissipation of the Herbicide Benzobicyclon Hydrolysate in a Model California Rice Field Soil Katryn L. Williams,*,† Joshua J. Gladfelder,‡ Lindsay L. Quigley,‡ David B. Ball,‡ and Ronald S. Tjeerdema† †

Department of Environmental Toxicology, College of Agricultural and Environmental Sciences, University of California, Davis, California 95616, United States ‡ Department of Chemistry and Biochemistry, California State University, Chico, California 95929, United States S Supporting Information *

ABSTRACT: The herbicide benzobicyclon (BZB; 3-(2-chloro-4-(methylsulfonyl)benzoyl)-2-phenylthiobicyclo[3.2.1]oct-2-en4-one) has recently been approved for use on California rice fields by the United States Environmental Protection Agency (U.S. EPA). Hydrolysis of BZB rapidly forms the active compound, benzobicyclon hydrolysate (BH), whose fate is currently not well understood. A model California rice soil was used to determine BH soil dissipation. The pKa and aqueous solubility were also determined, as experimental values are not currently available. Sorption data indicate BH does not bind tightly, or irreversibly, with this soil. Flooding resulted in decreased BH loss, indicating anaerobic microbes are less likely to transform BH compared to aerobic microorganisms. Temperature increased dissipation, while autoclaving decreased BH loss. Overall, dissipation was slow regardless of treatment. Further investigation is needed to elucidate the exact routes of loss in soil, though BH is expected to dissipate slowly in flooded rice field soil. KEYWORDS: soil dissipation, herbicide, benzobicyclon, benzobicyclon hydrolysate, rice, pesticide



anaerobic microbes tend to dominate in established flooded fields, though aerobic organisms can survive near the surface of a saturated soil column.9 Contributions of both microbial communities to rice herbicide soil degradation should therefore be considered for pesticides of concern. Triketone herbicide sorption to and degradation in soils is largely governed by soil pH, as these herbicides share an acidic moiety (Figure 2).10−13 California rice field soil pH has ranged between 4.6 and 6.8 in previous studies14,15 and shifts toward pH 7 after flooding.16 Though the pKa of sulcotrione, mesotrione, and tembotrione are known (3.22, 2.98, and 3.17, respectively),10 no value has been published to date for BH. Temperature and pesticide degradation in soil commonly share a direct relationship to a certain degree as a result of increased microbial activity with rising temperature. After a certain threshold, however, microbes can become inhibited though the exact temperature at which this occurs is species dependent. During a typical growing season, California rice fields can experience temperature fluctuations between 13 and 40 °C.17,18 Microbial activity and resulting pesticide dissipation is expected to vary within this range, as has been reported previously for other California rice pesticides.19,20 BH soil dissipation was investigated as a function of sorption and degradation in a model California rice field soil. Sorption was assessed at different temperatures (15, 25, and 35 °C, all

INTRODUCTION Weed resistance to current herbicide regimens is rapidly rising in California rice fields.1−4 A method commonly used to combat this trend is the utilization of herbicides with alternative modes of action from those currently employed. One such herbicide is benzobicyclon (Butte; Figure 1), which has recently been approved for use on flooded California rice fields by the U.S. EPA. BZB is a proherbicide that hydrolyzes to the active fo rm , ben zo bic yclo n h y dr oly sat e ( 3-[2-ch lor o-4(methylsulfonyl)benzoyl]bicyclo[3.2.1]octane-2,4-dione; Figure 1), in flooded fields.5 BH, a triketone herbicide, inhibits 4-hydroxyphenylpyruvate dioxygenase (4-HPPD) which leads to chlorophyll depletion and plant death.6 Other triketone herbicides include sulcotrione, mesotrione, and tembotrione. Although recently approved for use on California rice fields, little is currently known about the environmental fate of BH, and several parameters are derived from modeling software.7 Previous investigation of BH hydrolysis suggests recalcitrance in both acidic (pH ≤ 4) and natural rice field water (pH 8.2), with low reactivity under both neutral and basic (pH ≥ 9) conditions.5 BH is not expected to volatilize significantly from water (with an estimated Henry’s Law Constant of 2.7 × 10−11 atm m3 mol−1 and estimated aqueous solubility of 133 mg L−1);8 therefore, soil dissipation (via sorption and degradation) may play an important role in the fate of BH in flooded California rice fields. Dissipation of BH in soil may have significant implications for ecological risk to aquatic plants and microorganisms in rice fields as well as in drainage ditches and the Sacramento River. Soil properties such as microbial community, pH, and temperature can impact pesticide dissipation. Generally, © 2017 American Chemical Society

Received: Revised: Accepted: Published: 9200

August 7, 2017 September 26, 2017 September 29, 2017 September 29, 2017 DOI: 10.1021/acs.jafc.7b03679 J. Agric. Food Chem. 2017, 65, 9200−9207

Article

Journal of Agricultural and Food Chemistry

Figure 1. BH formation and hypothesized degradation pathways (A12,39,47,48 and B39,49) in soil based on previous research with a similar herbicide, sulcotrione. ROP was not analyzed directly as a standard was not commercially available and could not be synthesized. GTM China Co., Ltd. (Changzhou, China). Cyclopentane-1,3dicarboxylic acid (CDCA, 95% purity; Figure 1) was purchased through AK Scientific, Inc. (Union City, CA). 1-[2-chloro-4(methylsulfonyl)phenyl]ethane-1-one (CMAP; Figure 1) preparation is in the Supporting Information. Target degradation compounds (BOD, CDCA, CMBA, CMAP, ROP) were predicted based on consistently known sulcotrione soil degradation products.12,21−24 Determination of pKa. The pKa of BH was determined using the spectrophotometric method described by Albert and Serjeant and adapted by Tam and Takács-Novák.25,26 A solution of 0.01 M HCl and a solution of 0.01 M NaOH, both spiked to 1 × 10−4 M BH and 0.15 M KCl, were combined to produce seven solutions with pH ranging from 2.5−3.7 (expected pKa ± 0.6), which were measured using an Oakton waterproof pHTestr 30 (Thermo Fisher Scientific, Vernon Hills, IL). All mixtures, including parent solutions, were analyzed by UV/vis spectroscopy (Hewlett-Packard 8452A diode array spectrophotometer; Palo Alto, CA) in triplicate. Mixture absorbances at 258 nm were used to compute pKa:25

Figure 2. Neutral (A) and ionic species (B) of BH.

±1 °C). Dissipation of BH was investigated as a function of temperature (15, 24, and 35 °C, all ±2 °C), moisture content (flooded and approximately 50% water holding capacity) to elucidate behavior in anaerobic and aerobic conditions, respectively, and sterilization (unaltered and autoclaved soil) to assess microbial involvement. Additionally, the pKa and water solubility of BH were experimentally derived. To the best of our knowledge, there are no previously published articles reporting BH dissipation in a California rice field soil, its experimental pKa, or aqueous solubility.



pK a = pH + log

MATERIALS AND METHODS

Reagents. Acetonitrile, HPLC-grade water, magnesium sulfate (anhydrous), sodium chloride (Redi-Dri), sodium citrate dibasic sesquihydrate (SCDS), and sulcotrione (IS, Pestanal grade) were purchased from Sigma-Aldrich (St. Louis, MO). Formic acid (LC-MS grade), hydrochloric acid (ACS grade), Optima LC-MS acetonitrile, Optima water, sodium citrate dihydrate (SCD), calcium chloride dihydrate, and 2-chloro-4-(methylsulfonyl)benzoic acid (CMBA, 95% purity; Figure 1) were purchased from Fisher Scientific (Hampton, NH). Ammonium hydroxide (ACS grade) was purchased from EMD Millipore (Billerica, MA). BH (97% purity) and bicyclo[3.2.1]octane2,4-dione (BOD, 96% purity; Figure 1) were custom-synthesized by

AI − A A − AM

(1)

where AI is the absorbance of the ionized species, AM is the absorbance of the neutral species, and A is the absorbance measured in a mixture. AI and AM were obtained from the 0.01 M NaOH and 0.01 M HCl parent solutions, respectively. Aqueous Solubility. A shake-flask design proposed in OECD 105 and by Bergström et al. was used to assess the aqueous solubility of BH at pH 2, 7, and 11.27,28 Solutions were prepared at the desired pH by using 2.5 M HCl and 0.1 M NaOH. In screw-cap glass tubes, 10 mL of solution (either pH 2, 7, or 11) and 50 mg of BH (100 mg of BH for pH 11 samples) were allowed to shake at 300 rpm and 23 ± 2 °C. After 24, 71, and 95 h, three replicates of each pH were removed, 9201

DOI: 10.1021/acs.jafc.7b03679 J. Agric. Food Chem. 2017, 65, 9200−9207

Article

Journal of Agricultural and Food Chemistry

Figure 3. Redox potential over time in flooded samples (n = 3). decanted into polypropylene tubes (15 mL), and centrifuged at 3000 rpm (1982g) for 15 min. Aliquots of 1 mL were filtered via a 0.2 μm PTFE syringe filter, diluted with HPLC water to allow for quantitation, and analyzed via liquid chromatography-triple quadrupole mass spectrometer (LC-TQMS). Interday values for each pH were averaged if differences were less than 15%, which was the case for all solutions.27 Soil Collection and Characterization. Soil (classified as an Esquon-Neerdobe thermic clay loam29 with no prior exposure to BH) was obtained from the top 10 cm of a rice field at the Rice Experiment Station (Biggs, CA) in April 2016. Soil was sieved to ≤2 mm in the field and homogenized immediately after collection. Buckets were loosely capped and stored at 4 ± 2 °C for less than 3 months prior to use. The soil had a pH of 4.52 and was composed of 27% silt, 22% sand, 51% clay, 4.19% organic matter, and 1.60% total organic carbon (TOC). Autoclaved soil used for degradation experiments had a pH of 4.31 and contained 27% silt, 23% sand, 50% clay, 4.01% organic matter, and 1.45% TOC. Values were determined by the UCD Analytical Laboratory.30 Soil Isotherms. The batch equilibrium protocol provided in OECD 106 was followed for adsorption and desorption isotherms, which were constructed at 15, 25, and 35 °C (±1 °C).31 Soil used for isotherms was first air-dried and stored at room temperature. A soil-tosolution ratio of 1:10 (5 g of soil and 50 mL of 0.01 M CaCl2 solution) was selected for all isotherms. Optimized equilibration times for BH adsorption and desorption were 24 h for adsorption and 24 h for desorption (15 °C), 12 and 24 h, respectively, at 25 °C, and 12 and 48 h, respectively, at 35 °C. Sacrificial samples were constructed in 50 mL amber polypropylene tubes. Seven initial BH concentrations ranging from 0.1 mg L−1 to 10 mg L−1 were replicated four times and used for all isotherms. Samples were pre-equilibrated with 45 mL of 0.01 M CaCl2 solution (pH 7 ± 0.03) for at least 12 h and then spiked with 5 mL of BH in 0.01 M CaCl2 to achieve desired concentrations. Samples were shaken at 200 rpm until the relevant equilibration time had been reached and then centrifuged at 3000 rpm (1982g) for 15 min. Supernatant was filtered to 0.2 μm via PTFE syringe filter and analyzed via LC-TQMS. For desorption isotherms, fresh 0.01 M CalCl2 solution equivalent to the supernatant volume was added to the tube and allowed to shake at 200 rpm at the desired temperature until equilibration was reached. Samples were extracted and analyzed as per the adsorption supernatant. Isotherm Soil Extraction. Water (5 mL), 0.2 mL of 25% NH4OH, and 10 mL of acetonitrile was added to the samples, which were then sonicated for 2 h at 80 kHz, then centrifuged at 3000 rpm (1982g) for 5 min and the supernatant removed. Samples were

resuspended in 2.5 mL of water, 0.1 mL of 25% NH4OH, and 6 mL of acetonitrile and sonicated for 2 h at 80 kHz. After centrifuging at 3000 rpm (1982g) for 5 min, supernatants were combined and acidified (pH < 2) with 4 mL of 25% HCl. MgSO4 (4 g), 1 g of NaCl, 1 g of SCD, and 0.5 g of SCDS were added to each sample. Tubes were shaken for 1 min and then centrifuged at 3000 rpm (1982g) for 5 min. Acetonitrile extract (4 mL) was transferred into dSPE tubes (900 mg of MgSO4 and 150 mg of C18 (Bond Elut, Agilent)), which were shaken for 30 s and centrifuged at 3000 rpm (1560g) for 5 min. Extracts were diluted 1:10 with water, filtered to 0.2 μm via PTFE syringe filter, and spiked with sulcotrione (IS). Samples were stored at −20 °C prior to analysis via LC-TQMS. Sample recoveries are given in the Supporting Information (Table S1). Soil Degradation. Microcosms were constructed using 25 g of soil (1.5 cm depth) in sterilized 250 mL amber polypropylene bottles (Dynalon). Anaerobic microcosms also contained 50 mL of water (corresponding to 1.5 cm flood depth) in tightly capped bottles, while aerobic microcosms were maintained at approximately 50% water holding capacity in loosely capped bottles. Control soil microcosms were prepared identically except samples were autoclaved in triplicate to inhibit microbial activity. Sterilized water (either 50 mL for anaerobic samples or 5 mL for aerobic samples) was added to the samples, which were allowed to incubate at 15, 24, or 35 °C (±2 °C) for 2 weeks to equilibrate soil to the test conditions. Preliminary studies indicated 2 weeks was sufficient to produce anaerobic conditions (Figure 3). After equilibration, soils were spiked with 75 μL of a 100 mg L−1 BH standard in acetonitrile (blank soils were spiked with 75 μL of acetonitrile) to approximate field concentration of BH. At each temperature, 3 unaltered microcosms, 3 control microcosms, and 1 blank microcosm were removed at 0, 3, 7, 14, 30, 60, 92, or 120 days for immediate extraction. Three hydrolysis controls (15 μL of a 100 mg L−1 BH stock in acetonitrile into 10 mL of sterile water) were also removed at each time point and directly analyzed by LC-TQMS. Degradation Soil Extraction. To each sample, 50 mL of water (aerobic samples only) and 2 mL of 25% NH4OH were added, then samples were shaken and incubated for 15 min at room temperature. Acetonitrile (100 mL) was then added and samples were sonicated (Fisherbrand FB11201 Sonicator) at 80 kHz for 2 h. After centrifugation at 3000 rpm (1472g) for 5 min, the supernatant was removed and the soils were resuspended in 25 mL of water, 60 mL of acetonitrile, and 1 mL of 25% NH4OH and sonicated for 2 h. Samples were centrifuged at 3000 rpm (1472g) for 5 min, and the supernatants were combined and acidified (pH < 2) with 4 mL of 25% HCl. MgSO4 9202

DOI: 10.1021/acs.jafc.7b03679 J. Agric. Food Chem. 2017, 65, 9200−9207

Article

Journal of Agricultural and Food Chemistry

concentration of BH in the aqueous fraction at equilibrium (mg L−1), and N is the slope of the Freundlich isotherm. Desorption constants were identically derived, where Cs and Cw were the concentrations of BH in soil and solution, respectively, after desorption equilibration was reached. KOC were derived from Freundlich coefficients through normalization to the fraction of organic carbon ( f OC) in the soil:

(40 g), 10 g of NaCl, 10 g of SCD, and 5 g of SCDS were added to each sample, which were then shaken for 1 min and centrifuged at 3000 rpm (1472g) for 5 min. Acetonitrile extract (140 mL) was spiked with 0.1 mL of keeper solution (5% decyl alcohol in acetone) and evaporated to near dryness under vacuum (Bucchi RE 121 rotavaporator). Samples were reconstituted in 5 mL of acetonitrile and transferred to dSPE tubes containing 900 mg of MgSO4 and 150 mg of C18 (Bond Elut, Agilent), which were shaken for 30 s and centrifuged at 3000 rpm (1560g) for 5 min. Extracts were diluted 1:4 with water, filtered to 0.2 μm via PTFE syringe filter, and spiked with sulcotrione (IS). Samples were stored at −20 °C prior to analysis. Sample recoveries are in the Supporting Information (Table S1). Analysis. Analyses were conducted on an Agilent 1260 Infinity series HPLC (Santa Clara, CA) coupled to an Agilent 6420 triple quadrupole mass spectrometer (TQMS) using electrospray ionization. All compounds except for CDCA and CMBA were analyzed in positive mode. The mobile phase was 0.25% formic acid (A) and 0.25% formic acid in acetonitrile (B), which were eluted via gradient (Table S2, Supporting Information) through a Kromasil 100A C18 column (150 mm × 3.5 mm, 3.5 μm) with a Kromasil 100 A C18 guard cartridge (10 mm × 3 mm, 5 μm) with a 5 μL injection volume. BH and predicted degradation compounds were measured using multiple reaction monitoring (MRM) mode, except CDCA which was quantitatively analyzed by Q1 scan at 157 m/z. The optimized transitions used to quantitate BH, sulcotrione, CMBA, CMAP, and BOD were 355 m/z → 165 m/z, 329 m/z → 139 m/z, 235 m/z → 191 m/z, 233 m/z → 155 m/z, and 139 m/z → 111 m/z, respectively. An additional transition (355 m/z → 81 m/z) was used to qualitatively identify BH. Second-order calibration curves (R2 ≥ 0.9916, residuals ≤10%) ranging from 0.01 mg L−1 to 5 mg L−1 were constructed using mixed standards of all predicted degradation compounds and BH in solutions mirroring sample compositions. Further details regarding TQMS conditions can be located in the Supporting Information (Table S3). Method limits of detection (MLOD) and quantitation (MLOQ) were determined using blank soil extracts for both degradation and isotherm samples. The standard deviation of 12 extracts spiked to 0.01 mg L−1 was multiplied by 3 and 10 to obtain the MLOD and the MLOQ, respectively. Values can be found in Table 1.

K OC =

BH CDCA CMBA CMAP BOD

experiment

MLOD (mg L−1)

MLOQ (mg L−1)

sorption degradation degradation degradation degradation degradation

0.0027 0.0110 0.0026 0.0033 0.0062 0.0015

0.0089 0.0350 0.0086 0.0110 0.0210 0.0050

Ct = C0 e−kdegt

(4) −1

where C0 is the initial concentration of BH (mg kg ), Ct is the concentration of BH at time t (days), and kdeg is the first-order soil degradation rate constant. The half-life (DT50) can then be measured through the equation:

DT50 =

ln(2) kdeg

(5)

Statistics. Data were analyzed using JMP Pro 12 statistical software (Cary, NC) with a significance level set to P ≤ 0.01. Temperature, flooding, and sterilization effects on BH degradation were assessed using a three-way analysis of variance (ANOVA) with post hoc pairwise comparisons using Tukey HSD with a significance level of α = 0.05. Temperature effects on BH sorption were assessed in the same manner as for BH degradation.



RESULTS AND DISCUSSION BH pKa and Aqueous Solubility. The pKa of BH was determined to be 2.89 ± 0.04, which is comparable to values for similar triketone herbicides sulcotrione, mesotrione, and tembotrione (3.22, 2.98, and 3.17, respectively).10 Neutral and ionic species of BH are shown in Figure 2. The pH of the soil used in this study was 4.52; therefore, BH existed primarily as an anion in this soil, though a minor portion likely remained in its neutral form. BH aqueous solubilities at pH 2, 7, and 11 were found to be 22.6, 146, and 6490 mg L−1, respectively. A steep increase in solubility was observed as the pH of solution increased, which is expected for an acidic herbicide. BH is likely quite soluble in rice field water, as the pH normally ranges from 7−10 in flooded fields.32,33 Soil Sorption. KF and N values for both BH adsorption and desorption are characterized in Table 2. Adsorption Log KF

A Thermo Fisher Scientific LTQ Orbitrap XL with IonMax electrospray source was used to assess the presence of nontarget products of BH in degradation samples. Source parameters were 4.5 kV spray voltage, capillary temperature of 275 °C, and sheath gas setting of 15. The scan range was from 80 m/z to 400 m/z and a 10 μL injection volume was used. Solvents were 50% ACN and 50% H20 with 0.1% formic acid at 0.2 mL min−1. Spectral data were acquired in the centroid mode at a resolution setting of at least 60 000 fwhm with the lockmass feature, which typically results in a mass accuracy