Synthesis, Antifungal Activity and QSAR of Some Novel Carboxylic Acid Amides

A series of novel aromatic carboxylic acid amides were synthesized and tested for their activities against six phytopathogenic fungi by an in vitro mycelia growth inhibition assay. Most of them displayed moderate to good activity. Among them N-(2-(1H-indazol-1-yl)phenyl)-2-(trifluoromethyl)benzamide (3c) exhibited the highest antifungal activity against Pythium aphanidermatum (EC50 = 16.75 µg/mL) and Rhizoctonia solani (EC50 = 19.19 µg/mL), compared to the reference compound boscalid with EC50 values of 10.68 and 14.47 µg/mL, respectively. Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were employed to develop a three-dimensional quantitative structure-activity relationship model for the activity of the compounds. In the molecular docking, a fluorine atom and the carbonyl oxygen atom of 3c formed hydrogen bonds toward the hydroxyl hydrogens of TYR58 and TRP173.


Introduction
Amide fungicides, an old type of fungicide, still play an important role in the agricultural chemistry field nowadays [1]. Some excellent commercial fungicides including boscalid (BASF, 2003), tiadinil (Nihon Nohyaku Co., Ltd., 2004), fluxapyroxad (BASF, 2012) and benzovindiflupyr (Syngenta, 2012) belong to this class. Recently, many aspects of structural transformation and structure-activity relationship studies have been reported. Wen [2] and coworkers changed the biphenyl in the structure of boscalid to a diphenyl ether, and thus obtained target compounds with some extent of antifungal activities. Liu et al. synthesized a series of N-substituted pyridine amide fungicides exhibiting good activity to Magnaporthe oryzae and Blumeria graminis f. sp. Tritici [3]. Ye et al. changed the chlorine atom of boscalid to a methylthio or aromatic sulfur, and obtained some compounds with good activity to Rhizoctonia solani and Sclerotinia sclerotiorum [4].
Amide fungicides target the complex II (succinate dehydrogenase, SDH) in the mitochondrial respiratory chain [5]. As SDH inhibitors (SDHIs), they can inhibit the growth and even cause the death of pathogens by disrupting the mitochondrial tricarboxylic acid cycle [6] and then interfering with their respiration [7]. After analyzing the structure of the known SDHIs encompassing furametpyr, boscalid, fluxapyroxad, bixafen, penthiopyrad, penflufen, sedaxane, isopyrazam, benzovindiflupyr, we found that the aromatic amines share an auxiliary group with a five to nine atoms bridge and ortho-substituents. The alignment of the three-dimensional structures of carboxin, flutolanil, boscalid and penflufen ( Figure 1) demonstrates that the common chemical features discussed above superimpose well, suggesting an identical binding mode at SDH. By summarizing the chemical structure of all commercial aromatic amides fungicides, Dehne found that they share common chemical features, which are essential for fungicidal activity, and they hence bind to their target in the same manner [8]. In most previous studies, the ortho of the amide group on the benzene ring is attached to a C atom (Figure 2a), which is an important part in some commercial fungicides. Based on bioisosterism [9,10] and computational docking experiments [8], we introduced a N atom to replace the C atom and thus designed and synthesized the target compounds ( Figure 2b). Bioassays showed that some target molecules exhibited good antifungal activity and might be useful as potential lead compounds.

Synthesis of Compounds
The synthetic route to the target compounds is shown in Scheme 1. The chlorine atom in 1-chloro-2-nitrobenzene was replaced by an amino group via an aromatic nucleophilic substitution reaction, giving the compound I. Then the nitro group in compound I was reduced with hydrazine hydrate to provide the key intermediates II, which were subsequently acylated to produce the target amides III.   Table 1 summarizes the bioassay results. Overall, the target molecules showed different levels of antifungal activity. The inhibitory activity of the target molecules to P. aphanidermatum and R. solani were higher than against the other four fungi. Compounds 1c, 1i, 3c exhibited wide spectrum antifungal activity and compound 1c displayed excellent activities against Colletotrichum orbiculare, Rhizoctonia solani, Pythium aphanidermatum and Botrytis cinerea, with 60.43%, 75.19%, 74.14% and 79.27% inhibition rate, respectively. Moreover, some compounds showed highly specific pathogen activity, whereby compounds 1h and 3e exhibited activities against Rhizoctonia solani with 67.75% and 69.08% inhibition rates, respectively. Table 1. In vitro antifungal activity of the target compounds against different phyto-pathogenic fungi (at 50 μg·mL −1 ).

Quantitative Structure-Activity Relationship (QSAR) Analyses
During biological screening, CoMFA and CoMSIA models of the new compounds were constructed using Sybyl 7.3 to find some clues about the structure-activity relationships. Compound 3c was the template molecule used for constructing models for P. aphanidermatum activity. Some important parameters of the models were as follows: (  In the three-dimensional contour maps of the CoMFA of compounds 3c to P. aphanidermatum and R. solani. Some areas were covered with yellow blocks, which indicates that a large substituent in this position will likely reduce the biological activity of the compound. Green regions substantially covered the indazole and the benzene (Figure 3a,e), meaning that the introduction of greater steric hindrance groups at these positions was beneficial to improving the activity. In the CoMSIA and CoMFA electrostatic field, carbonyl and trifluoromethyl groups on the benzene ring were red blocks ( Figure 3b,c,f,g), indicating that the introduction of negatively charged groups to the active position was beneficial for the antifungal activity. In the hydrogen acceptor field (Figure 3d,h), the results showed that more hydrogen bonds should improve the biological activity. As a hydrogen acceptor, the fluorine in the trifluoromethyl and the nitrogen atom in the pyrazole are helpful for forming hydrogen bonds, so taking together all this information, it can explain why 3c showed better antifungal activity than the other compounds.

Molecular Docking
In an effort to elucidate the possible antifungal mechanism of these compounds, molecular docking of compound 3c to the binding site of SDH (pdb code: 2FBW [13,14]) pdb was performed. Boscalid was docked as a reference at the same time. The three-dimensional schematic diagrams clearly explained the possible optimal combination between the ligands and receptor protein ( Figure 4). Boscalid and compound 3c adopted similar conformations and locations in the active site. The docking total scores were 5.25 and 6.87, respectively. They were both well bound to the receptor protein with their amino hydrogen toward the carboxyl oxygen of TRP173, and the carboxyl oxygen forming hydrogen bonds toward the hydroxyl hydrogen of TYR58. At the same time, an additional hydrogen bond was formed between the fluorine atom of compound 3c and the hydroxyl hydrogen of TYR58. Two-dimensional diagrams are displayed as Figure 4c,f in which all of the amino acid residues TRP32, MET36, SER39, ILE40, AGR43, PRO169, HIS216, ILE218 interacted with the ligand, including weak interactions such as van der Waals interactions and polar interactions. Thus, a stable complex between compound 3c and SDH was formed based on these interactions. The docking results were consistent with previous studies comparing boscalid and other SDHIs (such as 3-nitropropionic acid, 2-thenoyltrifluoroacetone, carboxin) [15].
Resistant fungal genotype analysis verified that most of those key residues involved in forming the binding cavity were related to resistance formation [16].

General Information
For all reactions, the solvents and chemical reagents were of analytical or synthetic grade obtained from Sinopharm Chemical Reagent Beijing Co., Ltd. (Beijing, China), and were used without purification. Column chromatography purification was performed using silica gel. Melting points were determined using a B-III microscope (Beijing Technical Instrument Co., Beijing China), and the thermometer was uncorrected. NMR spectra were obtained using an Avance DPX300 spectrometer (Bruker, Billerica, MA, USA) with tetramethylsilane (TMS) as the internal standard. High resolution mass spectrometry data were obtained with an Accurate-Mass-Q-TOF MS 6520 system equipped with an electrospray ionization (ESI) source (Agilent, Santa Clara, CA, USA).

Synthesis of Compounds I: 1-(2-Nitrophenyl) Piperidine
In a 100 mL flask, 1-chloro-2-nitrobenzene (3.94 g, 0.025 mol), piperidine (2.55 g, 0.030 mol-we take R1 = piperidine as the example, the same method was used to synthesize and purify the rest of the reactants), potassium carbonate (6.9 g, 0.05 mol), HMTA (0.15 g), and CuI (0.2 g) were dissolved in DMF (50 mL), and the mixture was heated to 130 °C and refluxed for 24 h [17]. A same volume of water was then slowly added to the solution which was extracted with ethyl acetate (100 mL × 3). The combined organic layer was dried over anhydrous magnesium sulfate and evaporated under reduced pressure to give a crude product that was purified by column chromatography (EtOAc-PE = 10:1) to give the title compound as an orange powder, yield 75%; mp 76-77 °C; 1

Synthesis of Compounds III
A solution of 2-(piperidin-1-yl)aniline (0.2 g, 0.025 mol), dichloromethane (25 mL) and triethylamine (0.5 mL) was added to a 50 mL flask and cooled to 0 °C in an ice bath. After ten minutes, 2-chloronicotinoyl chloride (0.22 g) was added [20,21]. The mixture was stirred for 30 min, and then concentrated under reduced pressure to give a crude product. The pure 2-chloro-N-(2-(piperidin-1-yl) phenyl) nicotinamide (1b) 165.17, 140.87,  135.92, 133.36, 132.00, 129.87, 128.14, 126.61, 126.37, 126.31, 125.81, 124.26, 120.81, 119.54, 67.10       antifungal tests. The fungicidal activity of the target compounds was tested in vitro against the six plant pathogenic fungi using the mycelia growth inhibition method [22]. The tested compounds were dissolved in DMSO to prepare a 10 mg·mL −1 stock solution before mixing with PDA. The media containing compounds at a concentration of 50 μg·mL −1 were then poured into sterilized Petri dishes for initial screening. After two days at 25 °C, the colony diameter of each strain was measured. Percentage inhibition rate was calculated as (1 − a/b) × 100%, where a represents the colony diameter in the Petri dishes with tested compounds and b is the mean colony diameter in control Petri dishes. Each test was repeated three times. The 10 mg·mL −1 solution was diluted to 200, 100, 50, 25, 12.5, 6.25 μg·mL −1 and the above experiments were repeated, the inhibition rates were calculated separately. The EC50 values were calculated using SPSS Statistics v17.0.

QSAR Analyses
3D QSAR analyses were performed to predict the favorable and unfavorable moieties for improved bioactivity using the comparative molecular field analysis (CoMFA) and the comparative molecular similarity indices analysis (CoMSIA) models in the SYBYL 7.3 software [23]. CoMFA models were generated using the Sybyl 7.3 package on a Linux system. In total, 21 compounds obtained from synthesis were used to create a data set in which the bioactivity of all compounds was determined (Table 1) against Pythium aphanidermatum and Rhizoctonia solani. The pEC50 values were used for constructing the models. Three-dimensional molecular structures were built using the SKETCH module in Sybyl 7.3, while structural energy minimization was performed with the Tripos force field until a gradient convergence of 0.05 kcal/(mol A) was achieved. Gasteiger−Hückel charges were calculated and used to construct the CoMFA models [24].

Molecular Docking
Docking was performed with Surflex-Dock (SYBYL 7.3). The molecular structures were energetically minimized using Tripos with 1000 iterations and a minimum gradient of 0.005. The Gasteiger−Hückel charges of ligands were assigned. All bound water and ligands were eliminated from the protein, and the polar hydrogen atoms and the Kollman-united charges were added to the proteins.

Conclusions
Based on bioisosterism and the computational docking experiments of commercial amide fungicides, we designed and synthesized a series of novel aromatic amides by exchanging the biphenyl group of boscalid for a nitrogen-containing heterocyclic ring. The bioassays showed that all of the target molecules exhibited considerable in vitro antibacterial activity against the six kinds of fungi. Compounds 1c and 3c with an indazolyl group exhibited higher inhibition activity than others, indicating the potential value of this moiety in agrochemical applications. A molecular docking study showed that the total docking score of compound 3c was higher than that of boscalid and the binding of the indazolyl aniline moiety to the receptor was tighter than that of a phenylaniline in the SDH ligand binding pocket. The indazolyl group was thus very favorable for the binding of compound 3c with SDH; although compound 3c exhibited slightly lower activity than boscalid, indicating that improving some physical and chemical properties of compound 3c, such as penetration, might improve its bioactivity. Further synthesis and structural optimization studies are ongoing in our laboratory.