Synthesis of Novel α-Trifluoroanisole Derivatives Containing Phenylpyridine Moieties with Herbicidal Activity

To find novel herbicidal compounds with high activity and broad spectrum, a series of phenylpyridine moiety-containing α-trifluoroanisole derivatives were designed, synthesized, and identified via nuclear magnetic resonance (NMR) and high-resolution mass spectrometry (HRMS). Greenhouse-based herbicidal activity assays revealed that compound 7a exhibited > 80% inhibitory activity against Abutilon theophrasti, Amaranthus retroflexus, Eclipta prostrate, Digitaria sanguinalis, and Setaria viridis at a dose of 37.5 g a.i./hm2, which was better than fomesafen. Compound 7a further exhibited excellent herbicidal activity against Abutilon theophrasti and Amaranthus retroflexus in this greenhouse setting, with respective median effective dose (ED50) values of 13.32 and 5.48 g a.i./hm2, both of which were slightly superior to fomesafen (ED50 = 36.39, 10.09 g a.i./hm2). The respective half-maximal inhibitory concentration (IC50) for compound 7a and fomesafen when used to inhibit the Nicotiana tabacum protoporphyrinogen oxidase (NtPPO) enzyme, were 9.4 and 110.5 nM. The docking result of compound 7a indicated that the introduction of 3-chloro-5-trifluoromethylpyridine and the trifluoromethoxy group was beneficial to the formation of stable interactions between these compounds and NtPPO. This work demonstrated that compound 7a could be further optimized as a PPO herbicide candidate to control various weeds.


Introduction
Species regarded as weeds generally exhibit advantages with respect to their ability to compete for water, space, light, and nutrients as compared to crops thereby resulting in crop production losses [1,2]. The Food and Agriculture Organization of the United Nations (FAO) estimates that weeds, pests, and plant diseases result in 20-40% reductions in annual global crop yields [3]. Herbicides can effectively suppress weed growth, thereby improving crop yields. FAO data suggest that sustained herbicide use would improve global food production by a minimum of 10% per year [4]. Therefore, it is necessary to continuously develop new herbicides with high efficiency and a broad spectrum to deal with various weeds.
Here, 18 novel compounds were synthesized via the introduction of substituted αtrifluoroanisole moieties into 2-phenylpyridines, and the resultant compounds were found to exhibit herbicidal activity against both broadleaf and grass weeds. The synthesis strategy for these compounds is detailed in Figure 2. Protoporphyrinogen oxidase (PPO) is the last shared enzyme in the plant chlorophyll and animal heme biosynthesis pathways, which catalyzes the oxidation of protoporphyrinogen IX to protoporphyrin IX. Inhibiting the activity of PPO will lead to the massive accumulation of protoporphyrinogen IX, which penetrates the cytoplasm and is oxidized to protoporphyrin IX via other non-enzymatic mechanisms or oxidases. Protoporphyrin IX can generate reactive oxygen species under light, thereby resulting in the death of the plant. Therefore, PPO has been an important target for a variety of herbicides. Commercially available PPO inhibitors with herbicidal activity can be classified into oxadiazoles, uracils, phenylpyrazoles, diphenyl ethers, thiadiazoles, triazolinones, oxazolidinediones, and N-phenylphthalimides based upon their structural characteristics [15][16][17][18][19][20][21]. These herbicides have several advantages, including rapid onset of action, prolonged efficacy, a low environmental impact, and minimal mammalian toxicity [22]. Phenylpyridine compounds, which are structurally similar to uracil and diphenyl ethers, have been studied in the context of pesticide development. Schaefer et al. demonstrated that synthesized substituted 2-phenylpyridines exhibited excellent PPO enzyme inhibitory activity and herbicidal activity [22,23]. Substituted 3-(pyridin-2-yl)benzenesulfonamide derivatives prepared by Liu et al. exhibited effective inhibition against various weeds, especially broadleaf weeds [24][25][26]. In addition, Matsuya et al. demonstrated that N-[4-(arylmethyloxy)phenyl]-4-methyl-3,4,5,6-tetrahydrophthalimide exhibited good herbicidal activity [27,28]. Theodoridis reported that 2-[(4-heterocyclic-substituted-3-halophenoxymethyl)phenoxy]alkanoates showed excellent herbicidal activities against broadleaf weeds for both pre-and postemergence [29]. Therefore, α-trifluoroanisole was introduced into the active structure of 2-phenylpyridines to discover novel broad-spectrum herbicides.
Here, 18 novel compounds were synthesized via the introduction of substituted αtrifluoroanisole moieties into 2-phenylpyridines, and the resultant compounds were found to exhibit herbicidal activity against both broadleaf and grass weeds. The synthesis strategy for these compounds is detailed in Figure 2.  Figure 2. Design strategy of target compounds.

Synthesis of Target Compounds
The synthesis procedures for the target compounds used in this study are outlined in Scheme 1. Intermediates 3a-3i were synthesized via a Suzuki cross-coupling reaction from 2,3-dichloro-5-trifluoromethylpyridine 1 and substituted p-hydroxyphenylboronic acid 2a-2i, while intermediates 6j-6r were synthesized using substituted 4-trifluoromethoxybenzaldehyde as starting materials in a simple two-step reaction. These approaches yielded high levels of compounds 7a-7r using intermediates 3a-3i and intermediates 6a, 6j-6r as materials via nucleophilic substitution reactions in which N,N-Dimethylformamide DMF served as a solvent. After synthesis, compounds were characterized via HRMS and NMR, with the resultant spectral and analytical data being consistent with the expected structures. X-ray diffraction crystallography was further used to confirm the structure of compound 7a (Figure 3). The crystal data for compound 7a are shown in Tables S1-S7 in the Supplementary Material, and the NMR and HRMS spectra of compounds 7a-7r are shown in Figures S1-S81 in the Supplementary Material.

Synthesis of Target Compounds
The synthesis procedures for the target compounds used in this study are outlined in Scheme 1. Intermediates 3a-3i were synthesized via a Suzuki cross-coupling reaction from 2,3-dichloro-5-trifluoromethylpyridine 1 and substituted p-hydroxyphenylboronic acid 2a-2i, while intermediates 6j-6r were synthesized using substituted 4-trifluoromethoxybenzaldehyde as starting materials in a simple two-step reaction. These approaches yielded high levels of compounds 7a-7r using intermediates 3a-3i and intermediates 6a, 6j-6r as materials via nucleophilic substitution reactions in which N,N-Dimethylformamide DMF served as a solvent. After synthesis, compounds were characterized via HRMS and NMR, with the resultant spectral and analytical data being consistent with the expected structures. X-ray diffraction crystallography was further used to confirm the structure of compound 7a (Figure 3). The crystal data for compound 7a are shown in Tables S1-S7 in the Supplementary Material, and the NMR and HRMS spectra of compounds 7a-7r are shown in Figures S1-S81

Greenhouse Herbicidal Activity Assays
Herbicidal activity levels for compounds 7a-7r, when used for the post-emergence treatment of dicotyledonous weeds Abutilon theophrasti (A. theophrasti), Amaranthus retro-Scheme 1. Synthesis of α-trifluoroanisole derivatives containing substituted phenylpyridine moieties. 6j-6r as materials via nucleophilic substitution reactions in which N,N-Dimethylforma-mide DMF served as a solvent. After synthesis, compounds were characterized via HRMS and NMR, with the resultant spectral and analytical data being consistent with the expected structures. X-ray diffraction crystallography was further used to confirm the structure of compound 7a (Figure 3). The crystal data for compound 7a are shown in Tables S1-S7 in the Supplementary Material, and the NMR and HRMS spectra of compounds 7a-7r are shown in Figures S1-S81 in the Supplementary Material. Figure 3. The X-ray crystal structure of compound 7a. Figure 3. The X-ray crystal structure of compound 7a.
Further herbicidal activities revealed that at a 37.5 g a.i./hm 2 dose, compounds 7a, 7j, and 7k achieved 100% inhibition against the broadleaf weeds A. theophrasti and A. retroflexus in post-emergence testing, performing better than fomesafen, with compound 7a additionally exhibiting > 80% inhibitory activities against the broadleaf weed E. prostrata and the grass weeds D. sanguinalis and S. Viridis (Table 2).
SAR analysis indicated that the activities of these compounds were dependent on R 1 , R 2 , R 3 , and R 4 . For R 1 and R 2 substitutions, optimal herbicidal activity was observed when both R 1 and R 2 were hydrogen or fluorine atoms. In contrast, such activity decreased markedly with R 2 substitution of bulky groups, such as CF 3 and CH 3 . With respect to the R 3 and R 4 substitutions, optimal herbicidal activity was observed when R 4 was a hydrogen atom or chlorine atom, whereas other substitutions were associated with reduced activity against grass weeds. As such, the best herbicidal activity for the synthesized compounds was observed when both R 1 , R 2 , R 3 , and R 4 were substituted with hydrogen atoms.
Compound 7a was next used in weed spectrum control and crop injury tests conducted under greenhouse conditions ( Figure 4). Compound 7a exhibited good broad-spectrum inhibitory activity against the majority of tested broadleaf weeds and several grass weeds, with over 70% herbicidal activity at a dose of 18.75 g a.i./hm 2 . In addition, compound 7a was able to fully control the growth of Medicago sativa, Portulaca oleracea, Abutilon theophrasti, Amaranthus retroflexus, Cyperus rotundus L., and Solanum nigrum. Compound 7a also exhibited slight damage to the tested crops, with 20% inhibitory effect on Zea mays (maize, Z. mays), Triticum aestivum (wheat, T. aestivum), Gossypium spp. (cotton, G. spp.), and 15% inhibitory effect on Orysa sativa (rise, O. sativa), Glycine max (soybean, G. max). atom or chlorine atom, whereas other substitutions were associated with reduced activity against grass weeds. As such, the best herbicidal activity for the synthesized compounds was observed when both R1, R2, R3, and R4 were substituted with hydrogen atoms. Compound 7a was next used in weed spectrum control and crop injury tests conducted under greenhouse conditions ( Figure 4). Compound 7a exhibited good broadspectrum inhibitory activity against the majority of tested broadleaf weeds and several grass weeds, with over 70% herbicidal activity at a dose of 18.75 g a.i./hm 2 . In addition, compound 7a was able to fully control the growth of Medicago sativa, Portulaca oleracea, Abutilon theophrasti, Amaranthus retroflexus, Cyperus rotundus L., and Solanum nigrum. Compound 7a also exhibited slight damage to the tested crops, with 20% inhibitory effect on  To evaluate the ability of compound 7a more fully, i.e., to inhibit the growth of A. theophrasti and A. retroflexus, an ED50 assay was next conducted (Table 3). Compound 7a exhibited excellent herbicidal activities against A. theophrasti and A. retroflexus, with respective ED50 values of 13.32 and 5.48 g a.i./hm 2 , performing slightly better than the con- To evaluate the ability of compound 7a more fully, i.e., to inhibit the growth of A. theophrasti and A. retroflexus, an ED 50 assay was next conducted (Table 3). Compound 7a exhibited excellent herbicidal activities against A. theophrasti and A. retroflexus, with respective ED 50 values of 13.32 and 5.48 g a.i./hm 2 , performing slightly better than the control fomesafen (ED 50 = 36.39, 10.09 g a.i./hm 2 ).

Analysis of In Vitro PPO Inhibitory Activity
Compound 7a exhibited significant inhibitory activity against the NtPPO enzyme in vitro (IC 50 = 9.4 nM), with this value obviously being better than that for fomesafen (IC 50 = 110.5 nM) ( Table 4).

Docking Analysis
To better understand the mechanism of action between the prepared compounds and PPO, molecular docking was used to predict the binding conformation of small molecule compounds and target enzymes at the active site. Specifically, compound 7a and fomesafen were selected for docking studies to predict the binding mode of all target compounds to PPO. As shown in Figure 5B, the phenyl ring-containing trifluoromethyl group of fomesafen forms a π-π stacking interaction with the amino acid residue Phe392, the fluorine atom of the trifluoromethyl group forms a hydrophobic interaction with the amino acid residue Leu356, the phenyl ring-containing nitro group forms a π-π stacking interaction with the coenzyme FAD600, the nitro group forms a hydrogen bond with the coenzyme FAD600, and the sulfonyl group forms two hydrogen bonds with the amino acid residue Arg98. As shown in Figure 5A, the phenyl ring containing trifluoromethoxy group of compound 7a forms a π-π stacking interaction with the amino acid residue Phe392 and a hydrophobic interaction with the amino acid residue Leu356, the pyridine ring forms a π-π stacking interaction with the coenzyme FAD600, the phenyl ring linked to pyridine forms a hydrophobic interaction with the amino acid residue Leu372, the fluorine atoms of the trifluoromethoxy group form a hydrophobic interaction with the amino acid residue Phe439 and Leu334, respectively. Furthermore, the difference in herbicidal activity of compound 7a and fomesafen may be attributed to the different interactions.
group of compound 7a forms a π-π stacking interaction with the amino acid residue Phe392 and a hydrophobic interaction with the amino acid residue Leu356, the pyridine ring forms a π-π stacking interaction with the coenzyme FAD600, the phenyl ring linked to pyridine forms a hydrophobic interaction with the amino acid residue Leu372, the fluorine atoms of the trifluoromethoxy group form a hydrophobic interaction with the amino acid residue Phe439 and Leu334, respectively. Furthermore, the difference in herbicidal activity of compound 7a and fomesafen may be attributed to the different interactions.

Instrumentation
All reagents and other materials were obtained from Guangdong Guanghua Scitech Co., Ltd. (Guangdong, China), Bide Pharmatech Co., Ltd. (Shanghai, China), and Accela ChemBio Co., Ltd. (Shanghai, China) and used without additional purification unless otherwise noted. A B-545 melting point instrument (Buchi, Hangzhou, China) was used to quantify melting points, which were used without further correction. A Bruker AV-400 or AV-500 MHz spectrometer (Bruker, Hangzhou, China) was used to generate NMR spectra, with CDCl 3 or DMSO-d 6 serving as solvents. An Agilent 6545 Q-TOF LCMS spectrometer (Agilent, Hangzhou, China) was used for mass spectrometry. A Bruker D8 Venture diffractometer (Bruker, Hangzhou, China) was utilized to collect crystallographic data.

Synthesis
The synthesis approach for α-trifluoroanisole derivatives in this study is outlined in Scheme 1. Among these, compounds 6a, 6j, and 6k were obtained from commercial sources and used without additional purification.

General Approach to the Synthesis of Compounds 3a-3i
To prepare these compounds, 2,3-dichloro-5-trifluoromethylpyridine 1 (5 mmol), potassium carbonate (10 mmol), triphenylphosphorus (0.5 mmol), substituted p-hydroxybenzeneboronic acid 2a-2i (5.5 mmol), and palladium(II) acetate (0.25 mmol) were mixed for 6 h with 5 mL of CH 3 OH and 10 mL of CH 3 CN at 50 • C under N 2 . Following the completion of this reaction as determined via TLC, the mixture was extracted thrice using ethyl acetate. Organic layers from these three extraction steps were then combined, rinsed with brine, and dried over MgSO 4 . Vacuum filtration was used to remove solid residues, and the solvent was removed. The remaining residue was then recrystallized with ethanol and water at 70 • C to obtain compounds 3a-3i [30].

General Approach to the Synthesis of Compounds 5l-5r, 6l-6r
Sodium borohydride (10 mmol) was slowly added to a solution of substituted ptrifluoromethoxy benzaldehyde (5 mmol) in methanol (20 mL) with constant magnetic stirring at −10 • C over a 10 min period, followed by further stirring for an 8 h period. Following the completion of this reaction as determined via TLC, the mixture was evaporated until dry at which time HCl (3 N) was added to the remaining residue to yield an acidic solution that was subsequently extracted thrice using ethyl acetate. The combined ethyl acetate layers were then rinsed using brine, after which the organic layer was evaporated to yield compounds 5l-5r for subsequent use without additional purification [31].
Phosphorus tribromide (5.4 mmol) in CH 2 Cl 2 (2 mL) was added in a dropwise manner to a substituted 4-trifluoromethoxybenzyl alcohol 5l-5r (3 mmol) in CH 2 Cl 2 (10 mL) solution at −5 • C with stirring for 5 h. Following the completion of this reaction as determined via TLC, the mixture was extracted thrice with CH 2 Cl 2 . The combined dichloromethane layers were then rinsed using NaHCO 3 and brine, dried over MgSO 4 , and concentrated to yield the substituted 4-trifluoromethoxybenzyl bromide 6l-6r for subsequent use without additional purification [31].

General Approach to Synthesis of Compounds 7a-7r
Substituted phenylpyridines 3a-3i (2 mmol), N,N-dimethylformamide (10 mL), and NaH (3 mmol, 0.12 g) were mixed and stirred in a three-necked reaction flask at 20 • C for 30 min under N 2 . Next, the substituted 4-trifluoromethoxybenzyl bromide 6a, 6j-6r (2.4 mmol) was added followed by stirring for 8 h at 60 • C. Following the completion of this reaction as determined via TLC, the mixture was extracted thrice using ethyl acetate (30 mL × 3). The pooled ethyl acetate layers were then rinsed thrice using brine and concentrated. Residues were then purified via silica gel column chromatography using ethyl acetate (EA) and petroleum ether (PE) (V EA :V PE =1:15) to yield compounds 7a-7r [30].  Note: Due to the existence of the symmetrical structure and the multiplet caused by the fluorine atom, the number of signals in the NMR carbon spectrum of all the above compounds does not correspond to the number of carbons in the molecule.
DMF was used to dissolve all test compounds, followed by their dilution with distilled water containing 0.1% Tween-80 to an appropriate dose. The application rates for postemergence treatment were 150, 75, and 37.5 g a.i./ha. Seedlings were treated with test compounds at the three-leaf stage with three replicates, treated without test compound seedlings serving as blank controls, and fomesafen-treated seedlings serving as positive controls. The herbicidal activity was assessed after a 20-day period, with the results being compiled in Tables 1 and 2. Growth inhibition rates (0-100%) were assessed based on/evaluated by a visual inspection, with 0% and 100%, respectively, corresponding to no damage and the total destruction of aboveground plant structures.

Weed Spectrum and Crop Injury Tests
Weed spectrum and crop injury tests were conducted for compound 7a using the same procedures as above in a greenhouse, with an estimated post-emergence application rate of 18.75 g a.i./hm 2 . The utilized species included dicotyledonous weeds Pharbitis nil (P.

Herbicidal Activities Glasshouse Assay against Abutilon theophrasti and Amaranthus retroflexus
To study the control effect of 7a with the highest herbicidal activities against Abutilon theophrasti and Amaranthus retroflexus, its herbicidal activities were tested using the same procedures as above in a greenhouse, with fomesafen serving as a control. The application rates for post-emergence treatment were 450, 225, 150, 75, 45, 30, 15, 7.5, 3.75, and 1.875 g a.i./hm 2 . Seedlings were treated with test compounds at the three-leaf stage with three replicates, treated without test compound seedlings serving as blank controls, and fomesafen-treated seedlings serving as positive controls. The herbicidal activity was assessed after a 14-day period, with the results being compiled in Table 3.
Then fresh weight inhibition rates were established by measuring fresh weight values for the aboveground parts of target weeds, and the ED 50 values were calculated.

In Vitro Analysis of NtPPO Inhibitory Activity
To explore the mode of action more fully, i.e., for the prepared compounds, compound 7a was selected as a representative target in effort to confirm its ability to inhibit PPO, with fomesafen serving as a control compound. NtPPO was prepared as reported previously [35]. The inhibitory activities of the NtPPO enzyme were determined using a previously reported method [36]. The NtPPO catalytic rate was calculated by measuring changes in the concentrations of the NtPPO catalytic product protoporphyrin IX over a given time period by measuring the absorbance at 410 nm via a UV-Vis spectrophotometer.

Molecular Docking Analysis
The NtPPO (PDB ID: 1SEZ) structure was obtained from the Protein Data Bank, with molecular docking analysis being performed using chain A of the NtPPO structure [37]. The crystal structures of NtPPO and the small molecule 7a were obtained via a standard approach using Dock 6.9 (University of California, San Francisco, California 94158, USA) [38], with fomesafen serving as a control compound. The optimal binding conformation was then selected in light of the results of the docking score ( Figure 5).

Conclusions
In conclusion, a series of new α-trifluoroanisole derivatives containing phenylpyridine moieties were herein developed and synthesized. Among these, compound 7a exhibited optimal structural characteristics and good herbicidal activity when utilized for the postemergence treatment of a wide spectrum of dicotyledonous weeds (at a dosage of 18.75 g a.i./hm 2 ) in greenhouse testing. In addition, the excellent inhibitory activity against PPO and the docking result of compound 7a will provide directions for further optimization as a candidate PPO herbicide.
Supplementary Materials: The supporting information can be downloaded at: https://www.mdpi. com/article/10.3390/ijms231911083/s1. Author Contributions: Z.C. and Y.C. carried out experimental work; Z.C. prepared the manuscript; X.D. designed the material, supervised the project, and revised the paper. All authors have read and agreed to the published version of the manuscript.