The Synthesis, Fungicidal Activity, and in Silico Study of Alkoxy Analogues of Natural Precocenes I, II, and III

This study aimed to synthesize, characterize, and explore the eco-friendly and antifungal potential of precocenes and their derivatives. The organic synthesis of the mono-O-alkyl-2,2-dimethyl 2H-1-chromene series, including the natural product precocene I, and the di-O-alkyl 2,2-dimethyl-2H-1-chromene series, including the natural 2H-1-chromenes precocenes II and III, was achieved. The synthetic compounds were subjected to spectroscopic analysis, 1HNMR,13CNMR, and mass characterization. The antifungal activity of synthesized precocenes I, II, and III, as well as their synthetic intermediates, was evaluated by the poison food technique. Precocene II (EC50 106.8 µg × mL−1 and 4.94 µg mL−1), and its regioisomers 7a (EC50 97.18 µg × mL−1 and 35.30 µg × mL−1) and 7d (EC50 170.58 × µg mL−1), exhibited strong fungitoxic activity against Aspergillus niger and Rhizoctonia solani. Some of the novel chromenes, 11a and 11b, which had never been evaluated before, yielded stronger fungitoxic effects. Finally, docking simulations for compounds with promising fungitoxic activity were subjected to structure–activity relationship analyses against the polygalactouronases and voltage-dependent anion channels. Conclusively, precocenes and their regioisomers demonstrated promising fungitoxic activity; such compounds can be subjected to minor structural modifications to yield promising and novel fungicides.


Introduction
The problems associated with excessive pesticide use, particularly in developing countries such as Egypt, have prompted scientists to investigate eco-friendly and novel methods of plant protection. Secondary metabolites of higher plants, insects, and animals have been investigated as potential natural fungitoxicants for plant disease management. These natural chemicals are physiologically active and are produced to aid in pathogen defense, interspecies competition, and reproductive process facilitation. The biological activity of natural products against phytophagous insects, such as pesticides and crop protection compounds, has gained significant attention over the last couple of decades. Natural products are now considered as promising alternatives to the current arsenal of synthetic compounds [1]. The heterocyclic benzopyran skeleton containing oxygen, particularly its structural derivatives such as precocenes I, II, and III, constitutes a privileged group of compounds. These compounds are found in a wide range of phytochemical classes of natural products [2][3][4][5][6]. These compounds have excellent and broad-spectrum antibacterial, antifungal, antiviral, and insecticidal activities. Additionally, they have pharmacological properties, such as anti-inflammatory, anticancer, and antioxidant activities [7,8].
Phytopathogenic fungi, Rhizoctonia solani, and Aspergillus niger, especially the former, have long been recognized as a major threat to agricultural production around the globe, I, II, and III, as well as the synthesis of a series of regioisomers of those natural products, was carried out to investigate their antifungal activity prospects and their mass production via organic synthesis. In this present work, the naturally occurring precocenes and their analogues were synthesized, characterized, and evaluated for their fungitoxic activity against A. niger and R. solani.

Synthesis of Natural 2H-1-Chromene Compounds and Their Alkoxy Regioisomer Analogues
Natural chromene (precocenes II and III) analogues and their regioisomers were synthesized according to Timár and Jaszberenyi [27], with some modifications as illustrated in Scheme 1.
Our initial efforts were primarily focused on choosing one of the many strategies aimed at constructing a benzopyrane structure. The following synthetic strategy was opted: (i) the construction of benzopyranone compounds (synthetic precursors) and Scheme 1. Synthetic diagram of dialkoxy 2,2-dimethyl-2H-benzopyrane and monoal-koxy-2,2dimethyl-2H-1-benzopyran. The red marked substituent R group is positioned on carbon atoms marked by red numbers.
Our initial efforts were primarily focused on choosing one of the many strategies aimed at constructing a benzopyrane structure. The following synthetic strategy was opted: (i) the construction of benzopyranone compounds (synthetic precursors) and ben-zopyrane compounds in high yields and high purities to study their fungitoxic effects and (ii) the preparation of regioisomers of natural chromenes and their synthetic intermediated chromanones to test the effect of changing the position and type of substituent at the aromatic ring on their potential fungitoxic activity. Finally, the quality, quantity, and structure of the synthesized compounds were assessed using 1 HNMR and 13 CNMR spectroscopic data.

Evaluation of Antifungal Activity
Natural chromenes and its alkoxy derivatives, obtained by chemical synthesis, were evaluated for antifungal activities against the selected toxigenic fungi. The novelty of these compounds relies on the simple and straightforward synthesis and the absence of halogenated derivatives. This latter property makes these compounds more environmentally friendly than commercial fungicides. The antifungal activity of all synthesized natural chromenes, precocenes I, II, and III, as well as their analogues and synthetic intermediates, was assessed in vitro against R. solani and A. niger using the poison feed technique. Mycelial growth inhibition was studied at the effective concentrations of 50% (EC 50 ) and 90% (EC 90 ) ( Table 1). Almost all the compounds exhibited pronounced fungitoxic effects at varying levels. The EC 50 and EC 90 values of 3a against A. niger were 614.56 and 833.66 µg × mL −1 , respectively, and values of 3b were 1023.29 and 1591.74 µg × mL −1 . Both 3a and 3b compounds are regioisomers with different positions of the hydroxy phenolic groups. Nonetheless, the enhanced antifungal activity of 3a could be attributed to the presence of the α,β-unsaturated bond in the side chain adjacent to the C=O group [39]. Moreover, the position of the phenolic OH group was presumed to play an important role in bioactivity. The chemical nature of α,β-unsaturated aldehydes, as well as some of their toxicological effects, was based on their ability to function as direct-acting alkylating agents. Carbonyl carbon is an electrophilic site that readily reacts with nucleophiles [40,41]. Nucleophilic attack on the carbonyl moiety by primary amines, thiols, and possibly alcohols results in the formation of substituted amines, known as Schiff bases and hemiacetals, under physiological conditions. Secondary amine or thiol attack on the initial adducts can cause protein-protein, DNA-protein, or DNA-DNA cross-linking [42]. So, we speculate that the hyper-antifungal activity of 3a is due to the presence of a carbonyl moiety with a double bond. The EC 50 value of 4c was lower than 4a and 4b. The free phenolic OH groups at positions 5 and 7 in 4c boosted its antifungal activity, nearly doubling the activity compared to 4a and 4b (Table 1). It is worth noting that when the phenolic OH group in position 7 of 4c was alkylated, the resulting compound 5c demonstrated a higher antifungal activity. Likewise, methylating the phenolic OH group at position 6 in 4b enhanced the antifungal activity of the resulting product 5b. On the contrary, the antifungal activity of 5a was lowered dramatically. It was also observed that 5c showed much higher activity than its 5f analogue. When a methoxy group in 5a was substituted by an ethoxy group in 5d, the antifungal activity increased nearly sixfold. A similar observation was noticed in 5e. It could be inferred that changing the alkyl group in such compounds could improve their antifungal activity. Abrunhosa et al. [43] evaluated the antifungal activity of chromene dimers and found that the growth and activity of Asprgillus spp. in producing ochratoxin A varied with the change in chromene side chain structure, specifically when the H on benzene ring was replaced by the OCH 3 group.
The EC 50 values of 6a-f were highly variable, with 6c having the lowest (398.32 µg × m L −1 ) and 6b having the highest (1191.30 µg × m L −1 ) ( Table 1). Compound 6c was found to have higher antifungal activity than its regioisomers 6a and 6b. On the other hand, 6d had a strong inhibitory effect on the mycelial growth of A. niger, compared to 6e and 6f. An earlier report on the antifungal activity of the structurally related compound, 2-phenylchromen-4-one, showed antifungal activity at high concentrations against Penicillium spp. and Colletotrichum spp. [44]. The compound 2-(4-ethoxy-phenyl)-chromen-4-one is a potent inhibitor of energy-dependent fungicide efflux transporters in Pyrenophora triticirepentis [45]. Using this compound in conjunction with fungicides reversed P. triticirepenti fungicide resistance.
The fungitoxic effect of 7a and 7b was strong against A. niger, with EC 50 values of 97.18 µg × mL −1 and 106.8 µg × mL −1 , respectively (Table 1; Supplementary Figure S1). Meanwhile, 7c showed no activity against A. niger. Additionally, the reduced antifungal activity of 7d-f was observed, indicating that substituting a methoxy group with an ethoxy group resulted in the compromised antifungal activity of 7e and 7f, while 7d remained unaffected. The natural product precocene II had antifungal activity with an EC 50 value of 89.13 µg × mL −1 [16,46,47]. Chromene inhibited mitochondrial respiration in wildtype yeast [48] and F. graminearum by interacting with the VDAC, resulting in enhanced superoxide levels in mitochondria [49].
The antifungal activity of 9 was lower than 8 (Tables 1 and 2). The compound 10a inhibited the mycelial growth of A. niger with an EC 50 value of 221.31 µg × mL −1 and an EC 90 value of 799.7 × µg mL −1 ( Table 1). The EC 50 value of 10b, on the other hand, was found to have no activity against the tested fungus, while its EC 90 value could not be inferred. The compound 11a (precocene I) was found to have nearly twice the antifungal activity of its analogue 11b. Notably, precocene II (7b) inhibited the mycelial growth of A. niger almost three times more effectively than precocene I (11a) at the EC 50 . Our findings corroborated the previous reports that demonstrated a higher antifungal activity of precocene I, extracted from natural plant extracts, than precocene II [50][51][52]. Precocene II inhibited trichothecene production in F. graminearum by increasing superoxide levels in mitochondria after interacting with VDACs [49,53]. Based on the results obtained for the synthesized compounds against R. solani, two general observations were made: (i) the ethoxy group (which is more bulky, more electrondonating, and less polar) enhanced the fungitoxicity of the chroman-4-one compound more than methoxy group; and (ii) the more substituents on the benzene ring of chroman-4-one structure, the more active as a fungitoxic agent.
Among 7a, 7b, and 7f, both regioisomers 7a (EC 50 Figure S2). Similar findings have been demonstrated earlier by Ramadan et al. [16], who found that precocene II, extracted from A. houstonianum essential oil, had promising EC 50 values, 2.0 µg × mL −1 and 38.07 × µg mL −1 , against R. solani and P. megasperma, respectively. It was suggested that the natural precocene II has a very strong fungitoxic activity against two species of soil-borne disease fungi. The current findings, along with the earlier reports, have opened an array of myriads for using natural fungicides in managing root rot diseases, which cause substantial losses to the agricultural economy.
Likewise, 11a (EC 50 10.39 µg × mL −1 ) was tenfold more fungitoxic than its analogue 11b (EC 50 116.47 µg × mL −1 ) ( Table 2). The outstanding fungal inhibition activity of 11a and 11b has never been reported before. The inhibition of the mycelial growth of R. solani observed for compound 8 was higher (EC 50 76.62 µg × mL −1 ) than 3a (EC 50 123.77 µg × mL −1 ). Among the monoalkoxy compounds, 5c showed strong antifungal activity (EC 50 46.88 µg mL −1 and EC 90 146.16 µg × mL −1 ). Meanwhile, 6f and 6d induced fungal inhibition at higher concentrations, with EC 50 of 204.55 and 720.24 µg × mL −1 , respectively, and EC 90 values of 245.42, 406.91 µg × mL −1 , respectively ( Table 2). The observed results revealed that the presence of different moieties, including the free hydroxyl group, the methoxy group, or the ethoxy group in a different position at the benzene ring, strongly enhances the fungitoxic activity of chromenes and chromanones. In a bioassay using the yeast Saccharomyces cerevisiae, the chromene isolated from Eulypa lata either caused death or strongly inhibited the yeast growth [54]. Additionally, a respiratory assay using 2,3,5-triphenyl tetrazolium revealed that eutypinol and eulatachromene inhibited mitochondrial respiration in wild-type yeast and significantly reduced the cell growth of a mutant S. cerevisiae, lacking a thioredoxin peroxidase [54].
From the previous studies, the synthetic benzopyrones and their derivatives exhibited outstanding antifungal activity against different species of phytopathogenic fungi, e.g., Trichophyton longifusus, T. longifusum, Candida albicans, and A. flavus [55,56]. Another chromene derivative, 5-hydroxy-6-acetyl-2-hydroxymethyl-2-methyl chromene, had a strong fungitoxic effect against C. albicans and Cryptococcus neoformans [57]. The com- and [1-ethoxy-6-hydroxy 8-methoxy-3,5-dimethyl isochroman] inhibited the mycelial growth of Lasiodiplodia theobromae at 100 mg × mL −1 concentration [58]. Sariaslani et al. [59] studied the biotransformation of precocene II by microbial enzymes in Streptomyces griseus using 18 O 2 incorporation studies, concluding that precocene II was transformed into three major metabolites, including the mono-oxygenase enzyme, and could introduce possible evidence of interaction between the heterocyclic oxygen-containing compounds with the enzymatic systems and the mitochondrial respiration. Conner and Beuchat [60] and Knobloch et al. [61] proposed a mechanism for the fungitoxic effect of heterocyclic oxygen compounds, such as chromenes and chromanones, which affect the cell membrane causing increased permeability or interference with a variety of enzyme systems. Aqueous and ether extracts of Ageratum leaves, containing precocene II as a major constituent, inhibited fungal growth by halting the formation of germ tubes by spores in the presence of the tested fungi, which is crucial for the microorganism's survival because new hyphae formation can only begin with the germ tubes [62]. Precocene II retards fungal growth or stops the release of mycotoxins, such as aflatoxins (B1, B2, G1, and G2) and trichothecenes. It can reduce mRNA synthesis of deoxynivalenol, a contaminant released by F. graminearum that reduces grain utilization [63,64].

Molecular Docking
Molecular docking analyses of different PGUs and VDACs against chromenes and chromanones are shown in Figures 5 and 6. Precocene has previously been shown to inhibit fungal growth by interacting with PGUs [16]. Therefore, in molecular docking studies, various PGUs from A. niger and R. solani were included. Meanwhile, the VDAC was targeted by precocene II to affect VDAC gating (by gate closing) and inhibit microbial growth [49].
The molecular docking analysis demonstrated that 11a and 7b docked to F. solani-encoded VDACs and mouse-encoded VDACs at different cavities; the former docked at a cavity volume of 3839 and the later docked to two different cavity volumes (139 and 933). F. solani-encoded VDACs had smaller Vina scores (−6.3) than two chromanones and a chromene with Vina scores of −6.5 and −7.0, respectively (Supplementary Table S1). Although both VDACs are structurally similar, nonetheless, both docked in different places with synthetic compounds. Generally, 5c, 7b, 7d, and 11a docked to PGUs at similar positions in both A. niger and R. solani by interacting with the same AA residues ( Figure 1). Meanwhile, in the case of VDACs, a different trend was observed; 5c and 10a docked to mouse VDACs at the same position, interacting with the same AA residues, while 7b, 7d, and 11a docked to another position. However, in the case of F. solani VDACs, 5c, 7d, and 10a docked at the same position, while 7b and 11a docked at different positions ( Figure 2).
To decipher the precise role of chromenes and their derivates as promising alternative fungicides at the molecular level, it is a prerequisite to identify their exact bioreceptor. Nevertheless, the preceding findings could be validated in vivo using PGUs, VDACs, or other fungal enzymes, and could be of interest for futuristic studies.
The GC-MS spectral data for compound 5a

Synthesis of dialkoxy 2,2-dimethyl chroman-4-one (6a-f)
To obtain 6a-f, 0.1 moles of respective 5a-f in 100 mL of N,N-dimethylformamide was treated with 15.4g (0.11 moles) of anhydrous K 2 CO 3 , and 19.87 g (12.98 mL, 0.14 moles) of CH 3 I (99%) was added using a programmable syringe pump (LAMBDA-FIT, Switzerland), as described in Section 2.2. The reaction time and reaction yield are presented in Table 3. The products were refined after cooling to 20 • C, and then the mix was poured onto crushed ice and extracted twice with CH 2 Cl 2. The organic layer was washed two times with NaOH 2% and water, and dried over anhydrous sodium sulfate. After removing the solvent, the product was re-crystallized from ethanol. The GC-MS spectral data for compound 6a 3.2.5. Synthesis of dialkoxy 2,2-dimethyl 2H-1-chromene (7a-f) and monoalkoxy 2,2-dimethyl 2H-1- chromene (11a-b) The 2H-1-chromene compounds (7a-f) were prepared following the reduction and dehydration of the corresponding 6a-f compounds. In total, 0.05 moles of compounds 6a-f was dissolved in 100 mL of dry methanol (absolute) and then treated with 5 g (0.13 moles) of NaBH 4 dissolved in 50 mL of dry methanol (absolute) using a dropping funnel over one hour under a stream of argon gas condition. Similarly, 7-O-methyl-2,2-dimethylchromene (11a; precocenes I) and 7-O-ethyl-2,2-dimethylchromene (11b) were synthesized by reducing 0.07 moles of compounds 10a-b, respectively, with 5.31 g (0.13 moles) of LiAlH 4 in 50 mL of dry tetrahydrofuran (THF). After the reduction process, the reaction was stopped by adding 100 mL of water, and the product was extracted from the mixture with CH 2 Cl 2 . The reaction was monitored by TLC with hexane-ethyl acetate in a ratio of 1:1. Subsequently, the solvent was removed, and the residue was subjected to dehydration with 100 mL (4 mol × L −1 ) of HCl in dry THF at 5 • C using a dropping funnel over 1h. The time required to complete the reduction and dehydration is presented in Table 4. The crude products were extracted by diethyl ether several times, and the combined organic layer was extracted with 5% NaOH solution and then dried over anhydrous Na 2 SO 4 . The product was obtained by column chromatography (9:1 hexane: Et 2 O). Table 4. Reaction times required for reduction and dehydration of corresponding compounds (6a-f) and (11a-b), as well as the % reaction yields.

Compound
Reduction

Fungitoxic Evaluation of Synthetic Chromene and Chromanone Compounds
The antifungal activity of all synthesized compounds and the standard antifungal drug amphotericin-B was evaluated in vitro against two phytopathogenic fungi, R. solani and A. niger, using the poisoned food technique [69]. Pure fungi cultures were obtained from the Department of Plant Pathology, Faculty of Agriculture, Ain-Shams University, Egypt, and the Department of Biology, Faculty of Science, Memorial University of Newfoundland, St. John's, NL. Canada. Two cultural media, Czapek-Dox agar (CDA) and potato dextrose agar (PDA), were used in this study. Cultural media were obtained from Merck Chem. Co (Canada). CDA media were used as specific growth media for A. niger, while PDA media were used as growth media for R. solani. The fungitoxic activity of 29 compounds (chromanones and chromenes) was tested against A. niger and R. solani in vitro at various concentrations, ranging from 100 to 800 µg × mL −1 , while amphotericin-B (X-GEN, NY, USA) was used at 5-25 µg × mL −1 . An in vitro assay was performed on PDA and CDA growth media treated with gradient concentrations of all the synthesized compounds (100-800 µg × mL −1 in sterilized DMSO (5%) + Tween-20 as a dispersing agent). Then, 1 mL of a solution containing different compounds was poured into sterilized melted media, homogenized, and plated into Petri dishes (90 × 15 mm). The media-containing compounds were incubated for 48 h at 25 • C. After incubation, all plates were inoculated with agar plugs containing fungi and incubated again at 22 • C for 8 days. To assess mycelial inhibition, fungal growth diameters (mm) were measured daily, and radial inhibition was calculated when negative control plates were fully covered with the fungal mycelial. For all treatments, four replicates were used, and the percentage of mycelial growth inhibition was calculated according to the equation suggested by Pandey et al. [70]. where: • dc = the diameter of the fungal colony in the negative control; • dt = the diameter of the fungal colony in treatment.
The positive control (inoculated media with fungus and DMSO + Tween-20) was used to evaluate the toxicity of the solvent and the dispersing agent. In addition, the synthetic antibiotic drug, amphotericin-B (X-GEN, NY, USA), was also used as a standard antifungal drug at concentrations of 5-25 µg × mL −1 . The EC 50 and EC 90 for all treatments were calculated using a regression equation between the log concentrations and the probit of the percentage growth inhibition of fungi, according to Abd El-Naeem et al. [71].
To prepare protein input files, all water molecules, ligands, and ions were removed, and polar hydrogens were added from the PDB file using AutoDock Vina (version 1.10) [72]. Finally, the files were saved in the pdb format for docking processes.
Three-dimensional structures of different chromenes and chromanones were either downloaded from the Webchem webpage (https://pubchem.ncbi.nlm.nih.gov/compound) (accessed on 16 June 2022) or drawn using ChemBiochem Drew Ultra (version 12). All the ligands were used in the structure data file (sdf) format.
Blind molecular docking was performed to investigate the putative binding sites of chromenes and chromanones to PGUs and VDACs. The CB-dock2 server was used for blind docking with its default settings. Docking validation was accomplished by re-docking the original ligand into the receptor's active site and compared the binding sites. For each used ligand, the CB-dock2 was set to generate ten cavities for docking [73]. Following docking, the ligand with the lowest Vina score was considered credible and photographed.