Synthesis of Novel Amino Acid–Fipronil Conjugates and Study on Their Phloem Loading Mechanism

To develop a new pesticide with phloem mobility, a series of new amino acid–fipronil conjugates were designed and synthesized based on derivatization at the 3-position of the pyrazole ring of fipronil. Experiments using a Ricinus communis seedling system showed that all tested conjugates were phloem mobile except for the isoleucine–fipronil conjugate, and that the serine–fipronil conjugate (4g) exhibited the highest concentration in phloem sap (52.00 ± 5.80 μM). According to prediction with log Cf values and uptake experiments with Xenopus oocytes, the phloem loading process of conjugate 4g involved both passive diffusion and an active carrier system (RcANT15). In particular, compared with for a previously reported glycinergic–fipronil conjugate (GlyF), passive diffusion played a more important role for conjugate 4g in the enhancement of phloem mobility. This study suggests that associating a nutrient at a different position of an existing pesticide structure could still be effective in obtaining phloem-mobile derivatives, but the distinct physicochemical properties of resultant conjugates may lead to different phloem loading mechanisms.

Although proven to be feasible in conferring phloem mobility, the current derivatizations on fipronil were limited to one specific position of the parent structure: the amino group at the 5-position of pyrazole. According to the assumption that the additional phloem mobility was led by the affinity between the amino acid fragment on conjugates and amino acid transporters in plants, derivatization at other positions on the parent structure with the same nutrient substituent could also lead to phloem-mobile compounds. To support this hypothesis, in this study, a new series of amino acid-fipronil conjugates based on derivatization at the 3-position of the pyrazole ring of fipronil (4a-l) was designed and synthesized, and their phloem mobility was measured in R. communis seedlings. Additionally, nutrient-pesticide conjugates with different conjugating sites may present different physicochemical properties, which could also affect their ability to permeate the plasma membrane [16][17][18][19]. Thus, experiments including prediction with log Cf values and uptake experiments with Xenopus oocytes were conducted to compare the new and previous conjugates, and the results could help provide a deeper understanding of the phloem loading mechanism of similar xenobiotics.

Synthesis and Characterizations
The new conjugates (4a-l) were synthesized by combining fipronil and natural amino acids at the 3-position of the pyrazole ring of fipronil [20][21][22] following a three-step synthetic route. As shown in Scheme 1, the cyano group on the pyrazole ring was first hydrolyzed into a reactive carboxyl group [23], which enables easy chemical modification. Followed by condensation reactions with amino acid esters in the presence of EDC·HCl and DMAP [24,25], 12 amino acid ester-fipronil conjugates (3a-l) were obtained. Hydrolysis of the methyl ester group in 3a-l with lithium hydroxide [10,26] provided the new amino acid-fipronil conjugates 4a-l in good yields. All structures were confirmed via 1 H, 13 C NMR spectroscopy and ESI mass spectrometry.

Phloem Mobility in R. communis Seedlings
Phloem mobility of amino acid-fipronil conjugates 4a-l was measured in R. communis seedlings, which are an ideal biological model to evaluate the phloem mobility of xenobiotics because of their thin and highly permeable cuticles [16,27]. For each measurement, R. communis seedlings were incubated with 100 µM of the compound to be tested. Phloem sap was then collected and analyzed using High-Performance Liquid Chromatography (HPLC) [9].
As shown in Table 1, all conjugates could be detected in phloem sap except for conjugate 4e (isoleucine-fipronil). This verified our hypothesis that conjugation of an amino acid fragment at a different position on the pyrazole ring of fipronil is feasible for acquiring phloem mobility for the non-phloem-mobile parent compound. In general, the phloem mobility of compounds that have a small substituent at the α-position of the amino acid fragment was better than that of the ones containing branched alkyl groups, aromatic rings, or heterocycles. Among them, conjugate 4g (serine-fipronil) showed the highest phloem mobility, with the concentration in phloem sap being 52.00 ± 5.80 µM.
It is noteworthy that compared with previously reported GlyF [15], nine of the conjugates (4a, 4b, 4c, 4d, 4f, 4g, 4i, 4k, and 4l) exhibited better phloem mobility. In particular, the concentration in phloem sap of conjugate 4g was five times as high as that of GlyF. As all compounds have a free carboxylic acid function, it was possible that the ion trap mechanism was involved in the phloem transport of these conjugates. However, according to the prediction with ACD/Labs version 14.0 software (classical method), the net charges of all compounds were at −1 between pH 5.6 and 8.0. Therefore, these compounds do not match the characteristics of the ion trap mechanism [27]. Considering that passive diffusion and active transportation are the two major mechanisms involved in phloem loading of xenobiotics, two factors may have caused the observed enhancement in phloem mobility: (1) physicochemical properties that facilitate passive diffusion; and (2) higher affinity to associate with amino acid carriers which promote active transportation [28,29]. Further experiments were then conducted to analyze the respective significance of the two aspects. Glutamic acid (Glu) 20.00 ± 3.00 GlyF Glycine (Gly) 10.14 ± 0.30 [15] a Phloem sap was collected after cotyledons were soaked in buffer solution containing 100 µM of tested conjugates at pH 5.6 for 6 h. Each datapoint is the mean of 12 seedlings ± SE (n = 4); b "ND" means not detected.

Prediction of Phloem Mobility Using Log Cf Values
Predictions on phloem mobility of amino acid-fipronil conjugates were then performed based on physicochemical properties (log K o/w and pK a ) of compounds using log Cf values [1]. For most of the tested xenobiotics, the experimental data fitted well with theoretical predictions. However, the phloem mobility of many carrier-mediated xenobiotics would not fit in the prediction of log Cf values, because it lacks full consideration of biological parameters concerning penetration across the leaf cuticle, metabolism, cell compartmentation, and, in particular, the active transportation mechanism.
As a result, the predicted results mainly represent the ability of xenobiotics to enter into phloem through passive diffusion [16,27]. Thus, analysis of the predicted phloem mobility may help us to reveal if the physical properties of new conjugates enabled better passive diffusion compared with those of GlyF.
As shown in Table 2 [1], conjugates 4a-l were predicted to have different phloem mobility. The log Cf values of seven conjugates (4a, 4b, 4c, 4f, 4g, 4k, and 4l) ranged from −4 to 1, indicating that these conjugates had moderate phloem mobility, and the phloem uptake of these compounds involved the process of passive diffusion. The log Cf values of five conjugates (4d, 4e, 4h, 4i, and 4j) were below −4, suggesting that these xenobiotics have no phloem mobility. Considering that all conjugates except 4e were actually phloem mobile in R. communis seedlings, the uptake of 4d, 4h, 4i, and 4j was carrier meditated. Notably, the conjugates with log Cf values from −4 to 1 in general showed higher phloem mobility than the ones with log Cf values below −4. In particular, compared with GlyF, the uptake process of which was proved to be carrier meditated [15], all conjugates with log Cf values ranging from −4 to 1 presented higher phloem mobility. This demonstrated that the ability to penetrate the plasma membrane through passive diffusion dominated the overall phloem loading behavior for the new conjugates with high phloem mobility, and was the key factor that caused the enhanced uptake compared with GlyF ( Table 2). Uptake experiments with Xenopus oocytes were then conducted to provide experimental evidence for this deduction.

Uptake Experiments with Xenopus Oocytes
In our former research, it has been demonstrated that RcANT15 has the function of transporting GlyF [30]. Considering the structural similarity between the new series of conjugates and GlyF (both combinations of amino acid and fipronil), it is supposed that the same active-carrier-mediated mechanism may still effectively contribute to the phloem loading of the new series of conjugates. Thus, a cellular-level uptake experiment with Xenopus oocytes under the mediation of RcANT15 was conducted with conjugate 4g, which presents the highest phloem mobility in R. communis seedlings. Conjugate 4a, which contains the same glycine fragment as GlyF, was also tested under the same experimental condition as a comparison. For each measurement, cRNA of RcANT15 was injected into Xenopus oocytes, and the oocytes were treated with 0.1 mM of the conjugates being tested for one hour before analysis for the amount of conjugate absorbed. Xenopus oocytes injected with the same volume of Nuclease-Free water were used as a control [31]. Although the contribution of other carriers to the uptake process was not excluded in this experiment, the result was able to show a general scope for the uptake process.
The transport activity for endogenous substances [32], especially for amino acids [33], was weak in the plasma membrane of the Xenopus oocytes system. Thus, the ability of Xenopus oocytes without RcANT15 cRNA expression to transport xenobiotics could be ignored [34][35][36], and the uptake of the candidate compounds in the control group was mainly driven by passive diffusion. As shown in Figure 2, the uptake levels of conjugates 4g and 4a in oocytes injected without RcANT15 cRNA (0.25 and 0.15 nmol/10 cells, respectively) were both higher than that of GlyF (0.09 nmol/10 cells [30]). The experimental results were consistent with the theoretical prediction from log Cf values that, compared with GlyF, conjugates 4g and 4a have better transmembrane behavior through passive diffusion.
The injection of RcANT15 in Xenopus oocytes significantly enhanced the uptake level of all three conjugates being tested, and the results correlated positively with the phloem mobility measured in R. communis seedlings. Compared with the control group, the uptake of conjugates 4g, 4a, and GlyF (control: 0.50, 0.20, and 0.36 nmol/10 cells [30], respectively) in oocytes injected with RcANT15 cRNA was increased by about 0.25, 0.05, and 0.27 nmol/10 cells, respectively. This indicated that the amino acid carrier RcANT15 did promote the uptake of all tested conjugates in Xenopus oocytes, but the significance may be distinct for different conjugates. The transport capability of RcANT15 towards amino acid conjugates was responsible for the phloem mobility of GlyF, which was consistent with the conclusion from our previous study [15]. Although conjugated with the same amino acid substituent, conjugate 4a did not show adequate improvement in the uptake amount comparable to GlyF, indicating that the derivatization position and other factors such as linker structure may still affect its ability to be transported by carriers. For conjugate 4g, injection of RcANT15 cRNA led to a similar enhancement in the uptake amount compared to GlyF, but with a higher amount being absorbed through passive diffusion, it showed the overall highest phloem mobility. Combined with the results from theoretical prediction, we can conclude that high efficiency in passive diffusion and high affinity to bind with an active carrier were both key aspects that made conjugate 4g highly phloem mobile.

General Information
All reagents and solvents were purchased from Energy Chemical Co (Shanghai, China). Nuclear magnetic resonance (NMR) spectra were obtained using a Bruker AV-600 instrument (Bruker, Karlsruhe, Germany). Deuterated solvents were obtained from Cambridge Isotope Laboratories (Andover, MA, USA). DMSO-d 6 and CDCl 3 solvent peaks (2.50 and 7.26 ppm for 1 H and 39.52 and 77.16 ppm for 13 C, respectively) were used as internal chemical shift references. Mass spectrographic analysis was conducted on a Waters SYNAPTQ TM (Waters, Milford, MA, USA). Analytical thin-layer chromatography (TLC) was carried out on percolated plates (silica gel GF254), and spots were visualized with ZF-20D ultraviolet analyzer (Qingdao Marine Chemical Ltd., Qingdao, China). Silica gel (200-300 mesh) was used for column chromatography.
Fipronil (8.74 g, 20 mmol) was dissolved in acetic acid (50 mL). Concentrated (98%) sulfuric acid (25 mL) and water (20 mL) was then added, and the mixture was stirred at 120 • C for 1 h and then heated to reflux for an additional 2 h. After cooling to room temperature, 1 M aqueous solution of sodium hydroxide was added dropwise at 0 • C until the pH reached 10. Water was added to dissolve all precipitate, and the aqueous solution was washed with ethyl ether (3 × 200 mL). The aqueous layer was acidified with 1 M hydrochloric acid solution until the pH reached 1, and was then extracted with ethyl acetate (3 × 200 mL). The combined organic layer was washed with brine (3 × 200 mL), dried over anhydrous sodium sulphate, and evaporated under reduced pressure. The residue was recrystallized with mixture of ethyl acetate (5 mL) and petroleum ether (150 mL) twice and dried under vacuum to afford compound 2. The 1 H-NMR and 13 C-NMR spectra of compound 2 can be found in Figures S1 and S2 A mixture of amino acid methyl ester hydrochloride (15 mmol) and N-methylmorpholine (2.20 mL, 20 mmol) was dissolved in anhydrous dichloromethane (150 mL). The reaction mixture was then cooled to 0 • C, and compound 2 (4.56 g, 10 mmol), EDC·HCl (2.88 g, 15 mmol), and DMAP (0.12 g, 1 mmol) were added successively. The resulting solution was stirred at room temperature for 3 h. After completion of the reaction according to TLC tracking, water was added and the aqueous layer was extracted three times with dichloromethane (3 × 50 mL). The combined organic layer was sequentially washed with saturated sodium bicarbonate solution, water, and brine. The organic layer was dried over anhydrous sodium sulphate and evaporated to obtain the residue, which was purified by column chromatography (petroleum/ethyl acetate, v/v = 3:1−6:1) to afford 3a-l as white solids. The 1 H-NMR and 13 C-NMR spectra of compounds 3a-l can be found in Figures S3-S26.   Hz, 6H). 13 173.19, 171.82, 159.99, 151.81, 145.80, 137.19, 137.00, 135.14  Lithium hydroxide (0.63 g, 15 mmol) was added to a solution of compound 3 (5 mmol) in water (10 mL) and THF (20 mL), and the mixture was stirred for 2 h at room temperature. The solution was then adjusted to pH 2 with 1 M hydrochloric acid. The organic solvent was then removed via rotary evaporation, and the residual aqueous solution was extracted with ethyl acetate (3 × 25 mL). The combined ethyl acetate layer was sequentially washed with saturated sodium bicarbonate solution, water, and brine. The organic layer was dried over anhydrous sodium sulphate and evaporated to obtain a residue, which was purified by column chromatography (petroleum/ethyl acetate/acetic acid, v/v/v = 2:1:0.005-4:1:0.005) to afford 4a-l as white solids. The 1 H-NMR and 13 C-NMR spectra of compounds 4a-l can be found in Figures S27-S50.  13

Physicochemical Properties
Physicochemical properties [17] (molecular weight (MW), octanol/water partitioning coefficient (log K o/w )) of the twelve amino acid-fipronil conjugates were predicted using ACD/Labs version 14.0 software (classical method, ACD/Labs, Toronto, ON, Canada). The ionization constant in aqueous solution (pK a ) was calculated with Marvin Sketch (version 6.3.0, Chemaxon Ltd., Budapest, Hungary). The results are summarized in Table 2.

Plant Materials
Castor bean seeds (R. communis) No. 9 were purchased from the Agricultural Science Academy of Zibo, Shandong, China. The seeds were placed in humid cotton for 24 h at 27 • C prior to sowing in wet vermiculite. Seedlings were nurtured as previously described [37]. Seedlings after 6 days of growth were selected for further experiments.

Phloem Sap Collection
The method to collect phloem sap was similar as that recently described [38,39]. Cotyledons, from which the endosperm had been removed, were put in wells of a 12-well cell culture plate and incubated in buffer solution (containing 20 mM MES (pH = 5.6), 0.25 mM MgCl 2 , and 0.5 mM CaCl 2 ) supplemented with 100 µM amino acid-fipronil conjugates. Roots of seedlings were immersed in 0.5 mM CaCl 2 solution. After 1 h of pre-incubation, the hypocotyls were severed in the hook region for phloem exudation. Phloem sap was collected at 1 h intervals for 5 h, and was saved at 4 • C until analysis. After being diluted with pure water to four times its volume, the phloem sap was quantified using an Agilent HPLC system (1260, Agilent Technologies Inc., Santa Clara, CA, USA).

Animal Materials
The Xenopus laevis were purchased from the Institute of Biochemistry and Cell Biology, SIBS (Shanghai, China) and incubated in a clean glass sink. Feedstuff (Tianbangmeiwa 2) was used 2 to 3 times per week.

Xenopus Heterologous Expression System and Drug Uptake Experiment
Total RNA extraction of castor beans was performed using an E.A.N.A. TM Plant RNA Kit (OMEGA, Norcross, GA, USA). The ratio of A260/A280 was 1.8-2.0 during the experiment, which was suitable for subsequent experiments. The concentration of extracted RNA was diluted to 250 ng/µL. A quantity of 2 µg of extracted RNA was used for cDNA synthesis using a PrimeScript TM II 1st Strand cDNA Synthesis Kit (TAKARA, Tokyo, Japan). A full-length coding sequence of RcANT15 was cloned into eukaryotic expression vector pT7Ts. cRNAs were synthesized from linearized vectors using an mMESSAGE mMACHINETM T7 Transcription Kit (Ambion, Austin, TX, USA).
Ten oocytes that were injected with cRNA of RcANT15 or Nuclease-Free water were pre-incubated in 500 µL Kulori buffer (90 mM NaCl, 1 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , and 5 mM MES, pH = 5.6) for 5 min to ensure intracellular steady-state pH. Then, the oocytes were put into solution containing 0.1 mM of test compound and treated for 1 h. After incubation, oocytes were washed three times with Kulori buffer and solubilized in 10% SDS to dissociate the cell. The total volume was adjusted to 100 µL after being concentrated in vacuo for 4 h. The uptake of tested compounds was determined by HPLC. Each treatment was repeated at least three times [40].

Analytical Methods
A C 8 reversed-phase column (5 µm, 250 × 4.6 mm inner diameter, Agilent Technologies Inc., Santa Clara, CA, USA) were used for separations at 30 • C. The solvent system comprised acetonitrile and water (60:40, v/v) containing 0.1% TFA. The flow rate of the solvent system was 1 mL/min, and the injection volume was 10 µL.

Conclusions
A new series of phloem-mobile fipronil derivatives (4a-l) were designed and synthesized by conjugating amino acids with fipronil molecules at the 3-position on the pyrazole ring. Among them, conjugate 4g presented the best phloem mobility, and the concentration of 4g in phloem sap of R. communis seedlings was fivefold that of a previously reported glycinergic-fipronil conjugate (GlyF). Results from prediction with log Cf values and uptake experiments with Xenopus oocytes demonstrated that the phloem loading process of 4g involves both passive diffusion and active carriers such as RcANT15. However, compared with GlyF, for which the phloem loading process was primarily carrier mediated, passive diffusion may have played a more important role for conjugate 4g in addition to the involvement of carrier systems. Thus, the enhanced phloem mobility of conjugate 4g compared to that of GlyF was probably due to its higher hydrophilicity. Overall, this study provides another example that conjugating amino acids to existing pesticide structures is a feasible and efficient strategy to acquire phloem mobility in non-phloem-mobile pesticides. It also suggested that due to the distinct physicochemical properties, the phloem loading process of different series of conjugates may occur via different mechanisms. Thus, optimizations to enable better passive diffusion may be a potential strategy to obtain pesticides with enhanced phloem mobility.