Synthesis and Acaricidal Activities of Scopoletin Phenolic Ether Derivatives: QSAR, Molecular Docking Study and in Silico ADME Predictions

Thirty phenolic ether derivatives of scopoletin modified at the 7-hydroxy position were synthesized, and their structures were confirmed by IR, 1H-NMR, 13C-NMR, MS and elemental analysis. Preliminary acaricidal activities of these compounds against female adults of Tetranychus cinnabarinus (Boisduval) were evaluated using the slide-dip method. The results indicated that some of these compounds exhibit more pronounced acaricidal activity than scopoletin, especially compounds 32, 20, 28, 27 and 8 which exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite. Statistically significant 2D-QSAR model supports the observed acaricidal activities and reveals that polarizability (HATS5p) was the most important parameter controlling bioactivity. 3D-QSAR (CoMFA: q2 = 0.802, r2 = 0.993; CoMSIA: q2 = 0.735, r2 = 0.965) results show that bulky substituents at R4, R1, R2 and R5 (C6, C3, C4, and C7) positions, electron positive groups at R5 (C7) position, hydrophobic groups at R1 (C3) and R2 (C4), H-bond donors groups at R1 (C3) and R4 (C6) will increase their acaricidal activity, which provide a good insight into the molecular features relevant to the acaricidal activity for further designing novel acaricidal agents. Molecular docking demonstrates that these selected derivatives display different bide modes with TcPMCA1 from lead compound and they interact with more key amino acid residues than scopoletin. In silico ADME properties of scopoletin and its phenolic ether derivatives were also analyzed and showed potential to develop as good acaricidal candidates.


Introduction
The carmine spider mite, Tetranychus cinnabarinus (Boisduval), is considered as one of the most economically important arthropod pests [1]. This mite has been reported to infest over 100 crops or plants grown in the field or greenhouse worldwide, especially cotton, beans, eggplants, tomatoes, peppers, cucurbits and strawberries and so on [2][3][4]. Spider mites usually feed through a piercing-sucking process to remove cellular contents, resulting in reduction of photosynthesis and transpiration rates in plants [5,6]. The plants slightly infested by spider mite display discoloration of their leaves and defoliation, bud and fruit dropping and reductions in fruit yield and quality; serious plants infestations by this mite will cause whole plant death [7]. It is recognized as one of the most difficult mites to control mainly due to its small size, high reproductive potential, extremely short life cycle, and strong adaptability and ability to develop resistance [8,9]. The genetic system of spider mites is known as

Acaricidal Activity
As shown in Table 1, all the tested compounds exhibited varying degrees of acaricidal potency against female adults of T. cinnabarinus after treatment for 48 h, LC50 (mmol/L) and χ 2 values of all tested compounds are less than 6.6 and 5.6, respectively, pLC50 (mol/L) and P values of all tested compounds are more than 2.0 and 0.1, respectively. Except for compound 15, the rest of the target compounds exhibited more pronounced acaricidal activities against T. cinnabarinus than scopoletin. In particular, compounds 32, 20, 28, 27 and 8 exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency than the lead compound. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite. Target compounds 26-37 were obtained through a three-step reaction. The intermediates 6a-l were synthesized through the reaction of 2-chloroacetyl chloride with an alkylamine or substituted benzylamine. To quickly remove the by-product hydrochloric acid from this reaction, we used triethylamine as acid-binding agent, which was added before adding 2-chloroacetyl chloride into the reaction mixture. Chlorinated intermediates 6a-l were converted into iodine-substituted intermediates 7a-l (Scheme 3) to obtain high yields of the target compounds. Finally, the target compounds 26-37 were synthesized by reacting scopoletin with intermediates 7a-l in acetone. All of the target compounds provided satisfactory analytical and spectroscopic data, which were consistent with their depicted structures.

Acaricidal Activity
As shown in Table 1, all the tested compounds exhibited varying degrees of acaricidal potency against female adults of T. cinnabarinus after treatment for 48 h, LC50 (mmol/L) and χ 2 values of all tested compounds are less than 6.6 and 5.6, respectively, pLC50 (mol/L) and P values of all tested compounds are more than 2.0 and 0.1, respectively. Except for compound 15, the rest of the target compounds exhibited more pronounced acaricidal activities against T. cinnabarinus than scopoletin. In particular, compounds 32, 20, 28, 27 and 8 exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency than the lead compound. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite. Scheme 2. Synthesis of scopoletin phenolic ether derivatives . Target compounds 26-37 were obtained through a three-step reaction. The intermediates 6a-l were synthesized through the reaction of 2-chloroacetyl chloride with an alkylamine or substituted benzylamine. To quickly remove the by-product hydrochloric acid from this reaction, we used triethylamine as acid-binding agent, which was added before adding 2-chloroacetyl chloride into the reaction mixture. Chlorinated intermediates 6a-l were converted into iodine-substituted intermediates 7a-l (Scheme 3) to obtain high yields of the target compounds. Finally, the target compounds 26-37 were synthesized by reacting scopoletin with intermediates 7a-l in acetone. All of the target compounds provided satisfactory analytical and spectroscopic data, which were consistent with their depicted structures.

Acaricidal Activity
As shown in Table 1, all the tested compounds exhibited varying degrees of acaricidal potency against female adults of T. cinnabarinus after treatment for 48 h, LC50 (mmol/L) and χ 2 values of all tested compounds are less than 6.6 and 5.6, respectively, pLC50 (mol/L) and P values of all tested compounds are more than 2.0 and 0.1, respectively. Except for compound 15, the rest of the target compounds exhibited more pronounced acaricidal activities against T. cinnabarinus than scopoletin. In particular, compounds 32, 20, 28, 27 and 8 exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency than the lead compound. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite.

Acaricidal Activity
As shown in Table 1, all the tested compounds exhibited varying degrees of acaricidal potency against female adults of T. cinnabarinus after treatment for 48 h, LC 50 (mmol/L) and χ 2 values of all tested compounds are less than 6.6 and 5.6, respectively, pLC 50 (mol/L) and P values of all tested compounds are more than 2.0 and 0.1, respectively. Except for compound 15, the rest of the target compounds exhibited more pronounced acaricidal activities against T. cinnabarinus than scopoletin. In particular, compounds 32, 20, 28, 27 and 8 exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency than the lead compound. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite. The acaricidal activity of compounds 8-13 decreased as the carbon chain length of the substituent groups increased. Compounds 14-17 with a naphthenic base displayed lower acaricidal activity. When the hydrogen of a hydroxyl or amino from compounds 18-25 and 32-37 was substituted by a substituted phenyl group, different acaricidal potency was shown. This is due to the types, quantity and position of substituents on the benzene rings of these compounds. However, all of them well followed the Topliss tree rule [47]. Compound 23 with a para-chlorinated phenyl (pLC 50 = 2.3759) exhibited low potency against T. cinnabarinus compared with compound 18 with a simple phenyl (pLC 50 = 2.5922), therefore, both compound 19 with para-methylphenyl (pLC 50 = 2.4954) and compound 22 with 3,4 dichlophenyl (pLC 50 = 2.7821) don't show excellent potency against T. cinnabarinus. Compound 33 with a para-chlorinated phenyl containing an amide group (pLC 50 = 3.0716) exhibited equivalent potency against T. cinnabarinus compared with compound 31 with a phenyl-containing amide group (pLC 50 = 2.9452), compound 35 with a para-methylphenyl-containing amide group exhibited low potency against T. cinnabarinus (pLC 50 = 2.7510) compared with compound 33, therefore, the compound 32 with a 3-chlorophenyl-containing amide group (pLC 50 = 3.1835) show higher acaricidal potency. Both compound 34 with a 3,4-dichlorophenyl-containing amide group (pLC 50 = 3.0165) and compound 36 with a para-methoxyphenyl-containing amide group (pLC 50 = 2.8193) don't show good potency against T. cinnabarinus.
The 48 h LC 50 value of scopoletin in the current study was different from our previous reports [23], which may be attributed to the differences in scopoletin purity, the pesticide adjuvants, and the solvents used to prepare the tested compounds.
Compound 8 was used as typical representative of all target compounds to evaluate acaricidal activity against eggs, larval, and nymphal of T. cinnabarinus, basing on its higher acaricidal potency against female adults. As shown in Table 2, compound 8 exhibits excellent acaricidal potency against larva, low activity against nymphs, and no ovicidal activity. The different acaricidal potency maybe due to different expression of the possible target gene TcPMCA1 at different stages of T. cinnabarinus [36]. To determine whether multicollinearity existed among the descriptors in the models or not, a variable inflation factor (VIF) (VIF = 1/(1 − Rj2), where Rj2 represents the multiple correlation coefficient of one descriptor's effect on the remaining molecular descriptors) was calculated for each variable in the regression equation [48]. If VIF ranges from 1.0 to 5.0, the linked equation is suitable [49]. As shown in Table 3, the VIF of all descriptors were smaller than 2, indicating that the generated model possessed statistical significance and good stability. Table 4 gives the correlation matrix of the selected descriptors. From this table, it can be seen that the linear correlation coefficient value for each pair of descriptors was smaller than 0.6, suggesting that the selected descriptors were independent, meeting the important criterion for the model selections [50].
The reliability and statistical relevance of the attained BMLR-QSAR model is examined by internal and external validation procedures. Experimental and predicted activities (pLC 50 , mol/L) values of the compounds are shown in Table 5 and Figure 1. The residual values obtained by calculating the difference between the predicted and experimental pLC 50 are below 0.35 logarithmic units for all the compounds.
Internal validation is applied by the SPSS technique employing Leave One Out (LOO), which involves developing a number of models with one example omitted at a time. The observed correlations due to the internal validation techniques are R 2 LOO = 0.768. The R 2 LOO value was bigger than 0.5, indicating that the developed model had good stability and predictive ability [48].
The synthesized thirty target compounds were randomly divided into a 25-molecule training set with LC 50 values range from 0.655 to 6.588 mmol/L and a 5-molecule (11,16,19,27, and 33) test set with LC 50 values range from 0.815 to 3.196 mmol/L were used as an external test set for validating the attained QSAR models. The predicted/estimated acaricidal properties of the test set compounds are close to their experimentally observed values preserving their potencies. In addition, the value of R 2 pred = 0.583 for the external prediction was an acceptable result, which conformed that the generated MLR model was useful for meaningful predictions.
The QSAR model indicated that the descriptors representing polarizability (HATS5p) is main property governing acaricidal active agent of the scopoletin phenolic ether derivatives as shown by its high regression coefficient values of 12.920 (Equation (1)). The QSAR model demonstrated that high values of HATS5p, and Depressant-80, but low value of R8e, MATS6e, and HNar are required for potent activity of the compounds. Among these compounds 8-25, compounds    Table 6 shows the PLS results of the CoMFA and CoMSIA models. The results showed that the optimal CoMFA model yielded a cross-validated q 2 = 0.802 with an optimal number of principal components (ONC) of 6, non-cross-validated R 2 Table 6 shows the PLS results of the CoMFA and CoMSIA models. The results showed that the optimal CoMFA model yielded a cross-validated q 2 = 0.802 with an optimal number of principal components (ONC) of 6, non-cross-validated R 2 of 0.993, SEE = 0.029 and F value of 422.047. The contribution of steric and electrostatic fields is 70.8% and 29.2%, respectively. The best CoMSIA model yielded a q 2 of 0.735 with an ONC of 6, non-cross-validated R 2 of 0.965, SEE = 0.059 and F value of 83.553. The contribution of steric, electrostatic, hydrophobic, and hydrogen-bond acceptor are 21.5%, 28.5%, 44.9%, and 5.0%, respectively. Based on these field contributions, the steric field is the most important field in the CoMFA model, whereas the hydrophobic field is the most important field in the CoMSIA model. All the parameters in the Table 6 indicate that the CoMFA and CoMSIA models are robust and stable. The plot of experimental versus predicted acaricidal activities for CoMFA and CoMSIA models are shown in Table 7, and Figures 2 and 3. The residual values obtained by calculating the difference between the predicted and experimental pLC 50 are below 0.3 logarithmic unit for all the compounds. In addition, the values of R 2 pred = 0.999 (CoMFA) and R 2 pred = 0.787 (CoMSIA) for the external prediction were acceptable results. The CoMFA R 2 pred is higher than its R 2 , which indicated CoMFA model higher predictive ability. The prediced pLC 50 values of five test compounds by CoMFA model are very close to their experimental pLC 50 values, and their residuals are less than 0.008 logarithmic unit. These results indicate that the CoMFA and CoMSIA models are predictive.    Core structure of the studied scopoletin phenolic ether derivatives were shown in Figure 4A, the compound 8 was employed as the template molecule for the analysis of contour maps ( Figure 4B).    Core structure of the studied scopoletin phenolic ether derivatives were shown in Figure 4A, the compound 8 was employed as the template molecule for the analysis of contour maps ( Figure 4B).   Core structure of the studied scopoletin phenolic ether derivatives were shown in Figure 4A, the compound 8 was employed as the template molecule for the analysis of contour maps ( Figure 4B).  The CoMFA steric and electrostatic contour maps are shown in Figure 5 with compound 8. Those contours depict default contribution levels. In the CoMFA steric field shown in Figure 5A, A largesized and two medium-sized green contour near R4, R1R2and R5 (C6, C3, C4, and C7) indicate that bulky substituents were preferred here. It can be explained that the most of synthesized 7-position scopoletin phenolic ether derivatives have higher acaricidal activity than scopoletin. For CoMFA electrostatic map ( Figure 5B), there is one blue contour around the R5 (C7) position, which can be explain the fact that compound 8 with the smallest electronegative OCH3 groups possesses higher acaricidal activity among the synthesized target compounds. CoMSIA steric, electrostatic, hydrophobic, hydrogen-bond acceptor field contour maps are shown in Figure 6 with compound 8 as an example. Those contours also depict the default contribution levels. Since the steric and electrostatic contour are very similar with that of CoMFA, only hydrophobic, hydrogen-bond acceptor will be described as follows: in the hydrophobic contour map ( Figure 6C), a large-sized yellow contour near R1 (C3) and R2 (C4) indicates that introducing hydrophobic groups to that position could increase the acaricidal activity of the molecule. A largesized white contour near R6 (C8) suggests that hydrophilic substitutes preferentially localize at these positions. A medium-sized white contour were found surrounding the R5 (C7) which indicates that introducing hydrophilic groups to this position could improve the acaricidal activity. Therefore, the A B The CoMFA steric and electrostatic contour maps are shown in Figure 5 with compound 8. Those contours depict default contribution levels. In the CoMFA steric field shown in Figure 5A, A large-sized and two medium-sized green contour near R 4 , R 1 , R 2 and R 5 (C 6 , C 3 , C 4 , and C 7 ) indicate that bulky substituents were preferred here. It can be explained that the most of synthesized 7-position scopoletin phenolic ether derivatives have higher acaricidal activity than scopoletin. For CoMFA electrostatic map ( Figure 5B), there is one blue contour around the R 5 (C 7 ) position, which can be explain the fact that compound 8 with the smallest electronegative OCH 3 groups possesses higher acaricidal activity among the synthesized target compounds.  The CoMFA steric and electrostatic contour maps are shown in Figure 5 with compound 8. Those contours depict default contribution levels. In the CoMFA steric field shown in Figure 5A, A largesized and two medium-sized green contour near R4, R1R2and R5 (C6, C3, C4, and C7) indicate that bulky substituents were preferred here. It can be explained that the most of synthesized 7-position scopoletin phenolic ether derivatives have higher acaricidal activity than scopoletin. For CoMFA electrostatic map ( Figure 5B), there is one blue contour around the R5 (C7) position, which can be explain the fact that compound 8 with the smallest electronegative OCH3 groups possesses higher acaricidal activity among the synthesized target compounds. CoMSIA steric, electrostatic, hydrophobic, hydrogen-bond acceptor field contour maps are shown in Figure 6 with compound 8 as an example. Those contours also depict the default contribution levels. Since the steric and electrostatic contour are very similar with that of CoMFA, only hydrophobic, hydrogen-bond acceptor will be described as follows: in the hydrophobic contour map ( Figure 6C), a large-sized yellow contour near R1 (C3) and R2 (C4) indicates that introducing hydrophobic groups to that position could increase the acaricidal activity of the molecule. A largesized white contour near R6 (C8) suggests that hydrophilic substitutes preferentially localize at these positions. A medium-sized white contour were found surrounding the R5 (C7) which indicates that introducing hydrophilic groups to this position could improve the acaricidal activity. Therefore, the A B CoMSIA steric, electrostatic, hydrophobic, hydrogen-bond acceptor field contour maps are shown in Figure 6 with compound 8 as an example. Those contours also depict the default contribution levels. Since the steric and electrostatic contour are very similar with that of CoMFA, only hydrophobic, hydrogen-bond acceptor will be described as follows: in the hydrophobic contour map ( Figure 6C), a large-sized yellow contour near R 1 (C 3 ) and R 2 (C 4 ) indicates that introducing hydrophobic groups to that position could increase the acaricidal activity of the molecule. A large-sized white contour near R 6 (C 8 ) suggests that hydrophilic substitutes preferentially localize at these positions. A medium-sized white contour were found surrounding the R 5 (C 7 ) which indicates that introducing hydrophilic groups to this position could improve the acaricidal activity. Therefore, the synthesized target compouds (e.g., 27, 28 and 32) containing amide groups show higher acaricidal activity. synthesized target compouds (e.g., 27, 28 and 32) containing amide groups show higher acaricidal activity. In H-bond acceptor contour maps ( Figure 6D), two large-sized red contours near R1 (C3) and R4 (C6) indicate that introducing H-bond donors groups at those positions could increase the acaricidal activity. A medium-sized magenta contour near R5 (C7) suggests that H-bond acceptor groups at this position are favorable, and will increase the molecular activity. For example, several compounds (e.g., 8, 9 and 12) with H-bond acceptor groups display higher acaricidal activity.

3D-QSAR Analysis
In this research, only R5 (C7)-position was be modified to investigate acaricidal activity, and the contours maps of CoMFA and CoMSIA-derived models suggest that some favored group introduced to other position of scopoletin could improve acaricidal activity, which need to be further study.

Molecular Docking
Some tested compounds exhibit higher acaricidal activity than scopoletin, which prompted us to performed molecular docking study to understand the ligand-the target protein Ca 2+ -ATPase interactions in detail. Scopoletin and the synthesized derivatives 8, 9, 12, 20 and 28 possessing the higher acaricidal activity were selected for the docking study.
Docking results of scopoletin and its derivatives 8, 9, 12, 20 and 28 binding to TcPMCA1 are listed in Table 8. Scopoletin and selected compound show low binding energies of much less than 5.0 kcal/mol, which can be generally considered as specific ligands of TcPMCA1. Figure 7 shows the binding modes and orientations of scopoletin and its derivatives to TcPMCA1. The two dimensional interaction diagrams of scopoletin and its derivatives to TcPMCA1 are shown in Figure 8 In H-bond acceptor contour maps ( Figure 6D), two large-sized red contours near R 1 (C 3 ) and R 4 (C 6 ) indicate that introducing H-bond donors groups at those positions could increase the acaricidal activity. A medium-sized magenta contour near R 5 (C 7 ) suggests that H-bond acceptor groups at this position are favorable, and will increase the molecular activity. For example, several compounds (e. g., 8, 9 and 12) with H-bond acceptor groups display higher acaricidal activity.
In this research, only R 5 (C 7 )-position was be modified to investigate acaricidal activity, and the contours maps of CoMFA and CoMSIA-derived models suggest that some favored group introduced to other position of scopoletin could improve acaricidal activity, which need to be further study.

Molecular Docking
Some tested compounds exhibit higher acaricidal activity than scopoletin, which prompted us to performed molecular docking study to understand the ligand-the target protein Ca 2+ -ATPase interactions in detail. Scopoletin and the synthesized derivatives 8, 9, 12, 20 and 28 possessing the higher acaricidal activity were selected for the docking study.
Docking results of scopoletin and its derivatives 8, 9, 12, 20 and 28 binding to TcPMCA1 are listed in Table 8. Scopoletin and selected compound show low binding energies of much less than 5.0 kcal/mol, which can be generally considered as specific ligands of TcPMCA1. Figure 7 shows the binding modes and orientations of scopoletin and its derivatives to TcPMCA1. The two dimensional interaction diagrams of scopoletin and its derivatives to TcPMCA1 are shown in Figure 8 interactions with the VAL224. The N-propyl connected to the amide at the 7-position forms alkyl-alkyl interactions with the LEU217. Compound 28 interacts with ILE212, HIS223, ASP222, GLU220, ASN725, ASP724, PRO784, SER783, and SER782 through Van der Waals interactions. These selected derivatives display higher acaricidal activity that may be due to their different binding modes with TcPMCA1 from the lead compound and they interact with more key amino acid residues. Three selected homologues bind tighter with shortening of the side chain at the 7-position. The binding modes of scopoletin with TcPMCA1 in the current study were different from our previous reports [32], which may be attributed to our use of different software to find the active binding site interactions of TcPMCA1. Molecules 2018, 23, x FOR PEER REVIEW 13 of 13

ADME Study
An in silico study of scopoletin and its semisynthesed derivatives (compounds 8-37) was performed for prediction of ADME properties [51] (Table 9). From all these parameters, it can be observed that all tested compounds exhibited excellent % absorption (76.40-92.21%). It was also observed that all of these compounds followed Lipinski's rule of five and its extensions well. Four typical Lipinski's rule criteria are logP (octanol-water partition coefficient) ≤5, molecular weight ≤500, number of hydrogen bond acceptors ≤10 and number of hydrogen bond donors ≤5. Extension of Lipinski's rule of five includes the following criteria: number of rotatable bonds ≤10, topological polar surface area ≤140 A 2 . Table 9. Evaluation parameters of Lipinski's rule of five and its extensions from scopoletin and its phenolic ether derivatives .

General Information
Microwave-assisted synthesis was performed on a CW-2000 Ultrasonic Microwave Assisted Extractor (Xintuo Analytical Instruments Co., Ltd., Shanghai, China). Melting points were determined on a WRS-1B Digital Melting-Point Apparatus (Shanghai Shenguang Instrument Co., Ltd., Shanghai, China) and were uncorrected. IR spectra were obtained on a TENSOR 27 FT-IR spectrometer (Bruker Spectroscopic Instruments Co., Rheinstetten, Germany) using KBr pellets and values were presented in cm −1 . 1 H-and 13 C-NMR spectra were recorded on an Avance III 400 NMR spectrometer (Bruker Spectroscopic Instruments Co., Rheinstetten, Germany) with CDCl 3 or DMSO-d 6 as solvent. Mass spectra were carried out with a GCMS-QP2010 Ultra instrument (Shimadzu Corporation, Kyoto, Japan). Elemental analyses were performed on a Vario EL III elemental analyzer (Elementar Analysensysteme GmbH, Hanau, Germany)Propargite 90.00% TC was provided by Qingdao Hansen Biologic Science Co., Ltd., (Qingdao, China), and all other chemicals and solvents were of analytical grade and used as purchased. Analytical thin-layer chromatography (TLC) was performed on a glass plate coated with silica gel GF-254 (Qingdao Haiyang Chemical Co., Ltd., Qingdao, China) and visualized under ZF-1 ultraviolet analyzer (Shanghai Gucun Electro-optical Instrument Factory, Shanghai, China) under UV light (254 nm). Column chromatography was performed on silica gel (200 to 300 mesh). (4) Preparation of 2,4-dihydroxy-5-methoxybenzaldehyde (2) Following a literature method [22,46], aluminum (III) chloride (40 g, 0.30 mol) and CTAB (3.2 g, 8.8% mol) were added into dichloromethane (400 mL), and the reaction mixture was stirred at room temperature for 30 min, then a solution of 2,4,5-trimethoxybenzaldehyde (1, 20 g, 0.1 mol) in dichloromethane (100 mL) was added dropwise, then the mixture was refluxed for 4 h (the reaction progress was monitored by TLC with UV detection). The reaction mixture was cooled and poured onto 500 g of ice to which 100 mL of concentrated hydrochloric acid was added. The organic layer was separated and was washed with saturation salt solution, dried over anhydrous sodium sulfate, evaporated under reduced pressure to give 2,4-dihydroxy-5-methoxybenzaldehyde (2) as a light yellow solid, 60.84% yeild, m.p. 152-153 • C.
3.2.2. General Procedure for the Synthesis of 8-13 and 18-25 K 2 CO 3 (0.2073 g, 3 mmol) and CTAB (54.67 mg, 7.5% mmol) were added into a solution of scopoletin (4, 0.3843 g, 2 mmol) in acetone (30 mL), and the reaction mixture was stirred at reflux for 30 min. Then an alkyl or aromatic halide (3 mmol) was added into the mixture and maintained at reflux for 6-24 h (the reaction progress was monitored by TLC with UV detection). After cooling the reaction and filtration, the solvent was evaporated under reduced pressure, and the residue was dissolved in ethyl acetate, washed with saturation sodium bicarbonate, and saturation salt solution, successively, dried over anhydrous sodium sulfate, evaporated under reduced pressure to give the target crude products. The crude products were purified by column chromatography using petroleum ether/ethyl acetate from 10:1 to 7:1 as the gradient eluent system to yield the products 8-13 and 18-25.  Figure S5). 13  The slide-dip method [52] was adopted to evaluate the acaricidal activity of 8-37 against female adults of T. cinnabarinus. The appropriate amounts of target compounds were dissolved in 0.2 mL acetone and then diluted with water containing 0.1% Tween-80 to obtain the desired final concentration of 1000 mg/L for the preliminary screening. Based on the preliminary test results, a series of five to seven concentrations of the tested compounds were chosen to determine the median lethal concentration (LC 50 ) values of the compounds. Propargite 90.00% TC and scopoletin were used as positive controls, and water containing 0.1% Tween-80 was used as a blank control. Acaricidal activity assays were performed in triplicate and repeated thrice. The LC 50 values of the tested compounds were calculated using the probit analysis procedure of SPSS 17.0 for Windows (SPSS Inc., Chicago, IL, USA).
The leaf-dip method was used to evaluate the acaricidal activity of compound 8 against eggs, larval, and nymphal of T. cinnabarinus. The test solutions of compound 8 was prepared as above slide-dip method. Leaf discs were prepared to obtain uniform individuals at different developmental stages. Fresh cowpea leaves that had not been exposed to pesticides were washed thoroughly. Leaf discs with 3 cm diameters were placed on a corresponding size water-saturated sponge in a Petri dish (9 cm diameter) [53]. Adult females (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) were transferred to each leaf disc, allowed to lay eggs, and removed after 12 h. The leaf disc with eggs, larvae, and nymphs were then dipped in the compound 8 solutions for 5 s, taken out, and then laid on sponge in Petri dish again. The observed results were recorded after 48 h.

Data Set
The synthesized thirty target compounds and their acaricidal activities (LC 50 values) were used as data set for QSAR analysis. They were randomly divided into a 25-molecule training set for 2Dand 3D-QSAR models development and 5-molecule test set (compounds 11, 16, 19, 27 and 33) for external validation.

2D-QSAR (Multiple Linear Regression Model) Method
2D structures of the 30 target synthesized compounds were generated by ChemDraw Ultra (Cambridge Soft Corporation, Cambridge, MA, USA), and their energies were minimized using MM2 of Chem3D Ultra. Then 1666 molecular descriptors were calculated for each compound using DRAGON Web version 1.0 developed by the Milano Chemometrics and QSAR Research Group (http://www.vcclab.org/lab/edragon/start.html). These descriptors included (i) 0D constitutional (atom and group counts), (ii) 1D functional groups and atom-centred fragments, (iii) 2D topological, counts, autocorrelations, connectivity indices, information indices, topological indices, and eigenvalue-based indices, and (iv) 3D geometrical, WHIM, and GETAWAY descriptors, etc. [54]. 1302 descriptors were utilized as input values for model construction after eliminating the descriptors with constant values or mostly zero values (>90%) from the all the calculated descriptors.
2D-QSAR models were obtained using SPSS software (Version 17.0) that can run multiple linear regression. Different mathematical transformations of the observed median lethal concentration (LC 50 ) of the training set analogs, including property LC 50 (mg/L), LC 50 (mol/L), 1/LC 50 (mg/L), 1/LC 50 (mol/L), log LC 50 (mg/L), log LC 50 (mol/L), −log LC 50 (mg/L) and −log LC 50 (mol/L) values, were utilized in the present 2D-QSAR modeling to searching for the best model. −log LC 50 (mol/L) (pLC 50 ) values were used as dependent variables. Stepwise method for variable selection along with multiple linear regression was used to construct models.

3D-QSAR (CoMFA and CoMSIA) Methods
The molecular structures of synthesized compounds were generated and optimized using SYBYL 6.9 (Tripos Associates, St. Louis, MO, USA). The Gasteiger-Hückel charge, Tripos force field, and Powell method were used for structure optimization. To guarantee the obtaining of the molecular lowest energy conformation, conformation search was executed by using multisearch routin [55]. The most important component of a 3D-QSAR study is the alignment of the molecules based on the scaffold they share [56]. In this paper, the 7-oxy-6-methoxy-2H-chromen-2-one structure was selected as the common scaffold for molecular alignment. Compound 8 was used as the template molecule. All other synthesized acaricidal agents were aligned with the 7-oxy-6-methoxy-2H-chromen-2-one core.
The comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) are commonly used 3D-QSAR methods [51]. In CoMFA, the steric and electrostatic fields were calculated by setting the energy cutoff as the default value of 30 kcal·mol −1 . Five CoMSIA fields including the steric, electrostatic, hydrophobic, hydrogen-bond donor and hydrogen-bond acceptor were calculated using the default attenuation factor of 0.3 for Gaussian function. Field type "Stdev * Coeff" was used as the coefficient to analysis the contour map of each field [36]. The partial least squares (PLS) [57] was used to quantify the relationships by setting the biological activity (pLC50 values) as the dependent variables and the CoMFA/CoMSIA descriptors as independent variables.

Molecular Docking
Molecular docking studies were performed using AutoDock 4.2 and AutoDock Tools version 1.5.6 (ADT). The 3D structure of TcPMCA1 (GenBank No. KP455490), and its binding pocket were obtained from the I-TASSER server (Available online: http://zhanglab.ccmb.med.umich.edu/I-TASSER/), then water molecules were removed, polar hydrogen atoms were added, Compute Gasteiger charges were added, and AD 4 type atoms were assigned [41]. The 3D structure of ligands were constructed and their energy minimization were performed using ChemOffice 2004. Following by the structural optimization, all ligands were prepared for docking by merging non-polar hydrogen atoms, detecting rotatable bonds and adding gasteiger charges [41]. The grid box size of 60 × 60 ×60 Å was generated and allocated to center of binding cavity using x, y and z coordinates of 102.273, 100.115, and 118.080 for intend searching modality. Other parameters were set as the default. The Lamarckian genetic algorithmwas applied to calculate the possible conformation of the ligand molecule and macromolecule. Finally, the docking results were analyzed using the free version of Discovery Studio Visualizer 4.5 (Accelrys Software Inc., San Diego, CA, USA) [58].

Conclusions
Thirty phenolic ether derivatives of scopoletin including twelve compounds with amide groups were synthesized successfully using a molecular hybridization method. Their acaricidal activities, QSAR, molecular docking and a silico ADME properties were investigated. Some of these compounds exhibit more pronounced acaricidal activity than scopoletin, especially compounds 32, 20, 28, 27 and 8 exhibited about 8.41-, 7.32-, 7.23-, 6.76-, and 6.65-fold higher acaricidal potency than scopoletin. Compound 32 possessed the the most promising acaricidal activity and exhibited about 1.45-fold higher acaricidal potency against T. cinnabarinus than propargite. Statistically significant 2D-QSAR model supports the observed acaricidal activities and reveals that polarizability (HATS5p) was the most important parameter controlling bioactivity. 3D-QSAR (CoMFA: q 2 = 0.802, r 2 = 0.993; CoMSIA: q 2 = 0.735, r 2 = 0.965) results show that bulky substituents at R 4 , R 1 , R 2 and R 5 (C 6 , C 3 , C 4 , and C 7 ) positions, electron positive groups at the R 5 (C 7 ) position, hydrophobic groups at the R 1 (C 3 ) and R 2 (C 4 ), H-bond donors groups at R 1 (C 3 ) and R 4 (C 6 ) will increase their acaricidal activity, which provide a good insight into the molecular features relevant to the acaricidal activity for further designing novel acaricidal agents. Molecular docking demonstrates that these selected derivatives display different bide modes with TcPMCA1 from lead compound and they interact with more key amino acid residues than scopoletin. In silico ADME properties study of scopoletin and its phenolic ether derivatives were also analyzed and showed potential to develop these compounds as good acaricidal candidates.