Enantioselective hydrolyzation and photolyzation of dufulin in water

Background Dufulin is a novel, highly effective antiviral agent that activatives systemic acquired resistance of plants. This compound is widely used in China to prevent and control viral diseases in tobacco, vegetable and rice. Dufulin can treat plants infected by the tobacco mosaic virus and the cucumber mosaic virus. However, the achiral analysis and residue determination of dufulin remain underdeveloped because of its high enantioselectivity rates and high control costs. The enantioselectivity of an antiviral compound is an important factor that should be considered when studying the effect of chiral pesticides on the environment. The enantioselective degradation of dufulin in water remains an important objective in pesticide science. Results The configuration of dufulin enantiomers was determined in this study based on its circular dichroism spectra. The S-(+)-dufulin and R-(−)-dufulin enantiomers were separated and identified using an amylose tris-(3,5-dimethylphenylcarbamate) chiral column by normal phase high-performance liquid chromatography. The degradation of the rac-dufulin racemate and its separate enantiomers complied with first-order reaction kinetics and demonstrated acceptable linearity. The enantioselective photolysis of rac-dufulin allowed for the faster degradation of R-(−)-dufulin, as compared with S-(+)-dufulin. However, S-(+)-dufulin was hydrolyzed faster than its antipode. Conclusion The photolysation and hydrolyzation of dufulin in water samples normally complied with the first-order kinetics and demonstrated acceptable linearity (R2>0.66). A preferential photolysation of the R-(−)-enantiomer was observed in water samples. Moreover, the S-(+)-enantiomer was hydrolyzed faster than its antipode.


Background
Amino phosphonic acids are effective and environmentfriendly analogs of natural amino acids. N-(Phosphonomethyl)glycine, more commonly known as glyphosate, is a representative amino phosphonic acid pesticide; it was first introduced in 1960s but eventually became one of the most popular pesticides worldwide. Related structural features can be found on (DL)-homoalanine-4-yl-(methyl)phosphonic acid, which is known as glufosinate and used as another herbicide [1]. The use of glyphosate and glufosinate has steadily increased over several decades, thereby making both the most frequently used herbicides worldwide. The success of glyphosate and glufosinate is attributed to their effective control of herb growth and their relative environmental safety; this level of safety may be attributed to their ability to bind to soil colloids for degradation by soil microbes [2][3][4].
Based on previous reports of amino phosphonic acids, our research team designed a series of novel heterocyclic α-amino phosphonic acid esters. Our group conducted an intensive effort to discover novel antiviral lead structures for plants, by optimizing α-amino phosphonic acids as lead compounds; a commercially registered plant antiviral agent named dufulin was eventually produced [5]. [(2-Fluoro-phenyl)-(4-methyl-benzothiazol-2ylamino)-methyl]-phosphonic acid diethyl ester, also known as dufulin, has an asymmetric central carbon atom; this novel antiviral agent belongs to the α-amino phosphonate family (Figure 1). Dufulin is a highly effective against plant viruses; it functions by activating the systemic acquired resistance. This antiviral agent is used to control the tobacco mosaic virus (TMV) [6], the cucumber mosaic virus (CMV) [7], and Southern rice black-streaked dwarf virus (SRBSDV) [8] in China. A few methods for the achiral analysis and residue determination of dufulin have been reported in some matrices. However, to the best of our knowledge, the enantioselective degradation of dufulin in water has not been studied.
A number of studies have shown that the toxicity of chiral amino acid pesticides is enantiomer specific. This finding suggests that the environmental behavior of pesticides containing amino acid groups should be studied using racemic compounds and their individual enantiomers [20][21][22][23][24][25]. For example, the absorption rate of D-glufosinate was lower than that of racor L-glufosinate in sugar beet [26]. In addition, the different biotransformations of rac-glufosinate, L-glufosinate, and D-glufosinate reportedly depend on the plant species. rac-Glufosinate and L-glufosinate were metabolized in non-transgenic and transgenic plant cell cultures, whereas D-glufosinate was not metabolized [27]. However, most chiral pesticides are released into the environment in their racemic forms as equimolar mixtures of enantiomers. Therefore, the enantioselective degradation of racemic amino acid pesticides is a major topic in pesticide degradation research [28].
In this paper, we investigated the enantioselective hydrolyzation and photolyzation of the racemate racdufulin and its individual enantiomers in water, as influenced by some environmental factors. A detailed knowledge of the kinetics, hydrolysis, and photolysis pathways of pesticides is pertinent during experimental design to obtain reliable rate constants for assessing the fate and transport of pesticide pollutants in aquatic ecosystems. Our results may have some implications on the environmental risk assessment of chiral pesticides.

Absolute configuration of dufulin enantiomers
The individual enantiomers of dufulin were stereochemically analyzed by circular dichroism (CD) spectroscopy (NMR data of dufulin are shown in Additional file 1). Mirror-imaged CD curves were obtained by CD spectroscopy. The overall curves of the computed electronic circular dichroism (ECD) were obtained using timedependent density functional theory (TDDFT) calculations. The computed and experimental ECD were similar to each other. The configurations of the dufulin enantiomers eluted from the columns could be correctly assigned by comparing the observed values with the absolute configurations of the computed ECD. The enantiomers represented by peaks 1 and 2 in Figure 2D were assigned as S-(+)-dufulin (which has better biological activity) [29] and R-(−)-dufulin, respectively.

Calibration curves and assay validation
The calibration curves were plotted over the concentration range of 1.20 μg/mL to 60.0 μg/mL (n = 5) for each enantiomer in racemic dufulin. The regression equations and respective correlation coefficients were y = 12.549x + 2.5715 (R 2 = 0.9998) for the S-(+)-enantiomer and y = 12.568x + 2.1867 (R 2 = 0.9997) for the R-(−)-enantiomer. The recovery and precision data for the samples are summarized in Table 1. Each recovery value was acceptable for determining the enantiomers.

Enantioselective photolyzation of dufulin in aqueous systems
Water samples were spiked with either rac-dufulin or the individual enantiomers during the incubation period. The normal degradation of the dufulin enantiomers complied with first-order kinetics (see Tables 2 and 3, as well as Figure 3) while demonstrating acceptable linearity (correlation coefficient, R 2 of 0.8891 to 0.9817). The photolyzation of dufulin enantiomers in water was influenced by the dufulin concentration and the reaction pH. The half-life (t 1/2 ) values of the S-(+)-enantiomer in water at pH 7 were 31.4, 47.5, and 60.3 min for the racdufulin concentrations of 0.30, 1.50, and 3.00 μg/mL, respectively. By contrast, the t 1/2 values of the R-(−)-enantiomer in water at pH 7 were 27.9, 47.5, and 60.8 min for the rac-dufulin concentrations of 0.30, 1.50, and 3.00 μg/mL, respectively. Therefore, the enantiomers were degraded faster as the concentration of dufulin was increased.  The effect of pH is presented in Table 2. The t 1/2 values of the S-(+)-enantiomer of rac-dufulin (3.00 μg/mL) in water were 60.3, 71.5, and 154.0 min at pH values of 7, 5, and 9, respectively. The t 1/2 values of the R-(−)-enantiomer of rac-dufulin (3.00 μg/mL) in water were 60.8, 67.3, and 154.0 min at pH values of 7, 5, and 9, respectively. Dufulin photolyzation proceeded at higher rates under neutral conditions (with decreasing degradation rates at pH 7 > pH 5 > pH 9). Therefore, the photolytic process may be hindered in basic or acidic aqueous solutions. The photolysis of rac-dufulin was enantioselective in certain specific conditions, such as at pH 5 for 3.00 μg/mL or at pH 7 for 0.30 μg/mL, wherein the enantiomeric selectivity (ES) values were 0.030 and 0.058, respectively. The R-(−)-enantiomer was noticeably degraded faster than the S-(+)-enantiomer.
The preferential photolyzation of the single racemic dufulin enantiomer occurred at higher concentrations in neutral aqueous systems. The pure R-(−)-dufulin was photolyzed faster than its antipode under all different conditions ( Table 3). The photolyzation rates of dufulin enantiomers in the racemate mixture were slower than those in the single-enantiomer form. This phenomenon might have been caused by the competition between the two enantiomers of rac-dufulin.

Enantioselective hydrolyzation of dufulin in aqueous system
The hydrolytic degradation of dufulin enantiomers complied with first-order kinetics. The R 2 values of 0.6659 to 0.9724 demonstrated acceptable linearity ( Figure 4). The hydrolyzation of dufulin in water was strongly influenced by the pH, as shown in Tables 4 and 5. The t 1/2 values of the S-(+)-enantiomer of rac-dufulin (3.00 μg/mL) in water at 25°C were 40.8, 46.2, and 61.9 d at pH values of 9, 5, and 7, respectively. By contrast, the t 1/2 of the R-(−)-enantiomer of rac-dufulin (3.00 μg/mL) in water at 25°C were 43.6, 46.2, and 70.0 d at pH values of 9, 5, and 7, respectively. The t 1/2 of dufulin demonstrated that dufulin hydrolysis proceeded at higher rates under alkaline and acidic conditions (with degradation rates of pH 9 > pH 5 > pH 7). These results implied that the reaction was more effectively catalyzed by hydroxide or hydronium ions than by the neutral water molecules. The incubation temperature similarly influenced the hydrolyzation of dufulin. The higher temperatures caused the faster hydrolysis of the dufulin enantiomers. The S-(+)-enantiomer, in particular, was degraded faster than its antipode in the experiments (except at pH 5). The ES values ranged from −0.099 to −0.033. These ES values likewise suggested that the hydrolysis of rac-dufulin was enantioselective. The S-(+)-enantiomer was degraded more rapidly than the R-(−)-enantiomer.
Similar to rac-dufulin, the pure S-(+)-dufulin and R-(−)-dufulin compounds were hydrolyzed faster at higher temperatures in acidic or basic conditions. The S-(+) S-(+)-dufulin hydrolyzation of S-(+)-dufulin was preferred (except at pH 5). However, the t 1/2 values of the enantiomers in rac-dufulin were smaller than those of the individual enantiomers. This phenomenon could be attributed to the possible mutual promotion effects of the two enantiomers in racemic dufulin.

Experimental
Materials rac-Dufulin (purity, >99%) was synthesized in pure form in our laboratory. The products were unequivocally characterized using a JEOL ECX 500 NMR spectrometer (JEOL, Tokyo, Japan). The spectrometer was operated at 500 and 125 MHz during 1 H-NMR and 13 C-NMR spectroscopy at room temperature. DMSO-d6 was used as the solvent, whereas tetramethylsilane was the internal standard. Elemental analysis was conducted using a Elementar Vario-III CHN analyzer (Elementar, Frankfurt, Germany). The pure enantiomers of dufulin were prepared using an Agilent HPLC system (Agilent, Wokingham, UK) with a semi-preparative chiral column. The purity of both enantiomers was greater than 99%. ECD spectroscopy was performed using a Jasco J810 spectropolarimeter (Jasco, Easton, USA) at room temperature. Water was purified using a Milli-Q system (Merck Millipore, Billerica, USA). The photolytic instrument was fabricated in-house. The radiant exposure, wavelength, and intensity of the ultraviolet lamp were set at 60 UW/cm 2 , 253.7 nm, and 20 lx, respectively. All other chemicals and solvents were of analytical-reagent or HPLC grade, as obtained from common commercial sources. The stock solutions of rac-dufulin and the two pure enantiomers were prepared in methanol and stored at −20°C. The water samples were obtained directly from a Milli-Q system. The aqueous solutions were prepared at different pH levels, and 0.1 mol/L NaOH was used to modify the pH of these solutions. The potassium acid phthalate solution (0.1 mol/L) was adjusted to pH 5, the potassium dihydrogen phosphate solution (0.1 mol/L) was adjusted to pH 7, and the boric acid/potassium chloride solution (0.1 mol/L) was adjusted to pH 9. The blank water samples showed undetectable levels of dufulin (Figures 2A, 2B, and 2C).

Sample preparation Photolyzation study
The water samples were spiked with dufulin by mixing the standard solution with methanol in 250 mL Quartz glass tubes and evaporating the solvent under a nitrogen flow (the experiments were conducted at pH 5, 7, and 9).
Each tube was subsequently filled with 200 mL of each sample in water to produce the different concentrations of 0.30, 1.50, and 3.00 μg/mL. These samples were homogenized for 20 min by careful agitation, capped, and placed under the ultraviolet light (20 W). At exact sampling time schedule, a 20 mL sample was removed into  Table 4 Hydrolytic regressive functions of rac-dufulin enantiomers (3.00 μg/mL) in water (n=3) the separatory funnel and immediately used for extraction.

Hydrolyzation study
The water samples were spiked with dufulin by mixing the standard solution with methanol in 500 mL Erlenmeyer flasks and evaporating the solvent under a nitrogen flow (the experiments were conducted at pH 5, 7, and 9). Each Erlenmeyer flask was subsequently filled with 200 mL of each sample in water to produce a nominal pesticide concentration of 3.00 μg/mL. The Erlenmeyer flask content was also agitated during 20 min, capped and placed at 25°C (or 15°C, or 45°C) in the dark. At exact sampling time schedule, 20 mL samples were removed into the separatory funnel and extracted immediately.

Sample purification
Each water sample were decanted into a 250 mL separatory funnel containing 20 mL distilled water and 10 mL 25% sodium chloride solution, and extracted with 30 mL, 20 mL dichloromethane, respectively. The lower layer was dried with anhydrous sodium sulfate, and then concentrated using a rotary evaporator (35°C, 0.07 MPa) and a nitrogen blow dry instrument (40°C). The residue was dissolved in 1.0 mL methanol, after filtration, an aliquot (5 μL) was injected into the HPLC for analysis. There was no detectable dufulin residue in blank water samples. All of the water incubation experiments were carried out in triplicate.

ECD spectroscopy and chromatographic measurements
ECD spectroscopy was carried out by a Jasco-J810 spectropolarimeter at room temperature. The spectra were collected from 200 nm to 330 nm with a scan speed of 50 nm/min. The optimized geometry of two dufulin enantiomers was obtained by Gaussian 09 with density functional theory (DFT) at the B3LYP/6-31G* level. The ECD calculations of two enantiomers of dufulin were carried out by Gaussian 09 with TDDFT methods at the B3LYP/6-311++G (2d, 2p) level. The analytical HPLC was performed using an Agilent 1200 Series apparatus equipped with a quaternary pump, an autosampler, a DAD detector (detection wavelength of 270 nm, reference wavelength of 360 nm, slit length of 4 nm), a vacuum degasser, a column oven and Agilent Chemstation software. After filtration, 5 μL of the sample was injected into the Chiralpak IA column (250 mm × 4.6 mm i.d., 5 μm, Daicel Chemical Industries Ltd.) with a guard column. The temperature of the column was adjusted to 25°C. The flow rate of the n-hexane-ethanol mobile phase (90:10, v/v) was 1 mL/min. The isocratic elution was used.

Calibration curves and assay validation
For this study, a series of dufulin working standard solutions (1.20-60.0 μg/mL) for linearity of the two enantiomers were prepared for HPLC analysis. Calibration curves were generated by plotting the peak area of each enantiomer against its concentration. Linear regression analysis was performed using Microsoft Excel 2010. The precision and accuracy of the method were calculated as the ratio of the spiked concentration of each enantiomer in blank treatment (with water alone) to the predicted concentration. The recoveries of enantiomers were determined by analyzing the quality control samples at three different fortified levels. The samples were extracted, and determined as previously described. The concentration of each enantiomer in water was calculated from calibration curves of the corresponding enantiomer using an external standard.

Kinetic analysis and calculation
For the photolytic and hydrolytic treatments, the data were assumed followed first-order kinetics model, corresponding rate constant k for the (+)-enantiomer and (−)-enantiomer were calculated according to (1). The starting point was the maximum value of the concentration, and declined in following days. The half-life (t 1/2 day) was determined from (2).

Conclusion
In conclusion, the absolute configurations of dufulin enantiomers were determined by ECD spectra, the first eluted enantiomer was S-(+)-dufulin, the second one was R-(−)-dufulin. The extract and detection methods were optimized. The precision and accuracy data showed that these methods were satisfied with the experiments. The photolysation and hydrolyzation of rac-dufulin in water samples normally complied with the first-order kinetics, with acceptable linearity (R 2 >0.66). The preferential photolyzation of R-(−)-dufulin occurred in certain water samples. Moreover, S-(+)-dufulin was hydrolyzed faster than its antipode. Future investigations should focus on the mechanisms of enantioselective sorptiondesorption and the mobility behavior of dufulin in the soil, which may provide insights for improved environmental and ecological risk assessment.