Natural Compound-Derived Cytochrome bc1 Complex Inhibitors as Antifungal Agents

The high incidence of fungal pathogens has become a global issue for crop protection. A promising strategy to control fungal plant infections is based on the use of nature-inspired compounds. The cytochrome bc1 complex is an essential component of the cellular respiratory chain and is one of the most important fungicidal targets. Natural products have played a crucial role in the discovery of cytochrome bc1 inhibitors, as proven by the development of strobilurins, one of the most important classes of crop-protection agents, over the past two decades. In this review, we summarize advances in the exploration of natural product scaffolds for the design and development of new bc1 complex inhibitors. Particular emphasis is given to molecular modeling-based approaches and structure–activity relationship (SAR) studies performed to improve the stability and increase the potency of natural precursors. The collected results highlight the versatility of natural compounds and provide an insight into the potential development of nature-inspired derivatives as antifungal agents.


Introduction
Plant diseases caused by fungal pathogens are a major threat to the agricultural industry worldwide. Controlling plant fungal infections is therefore key to increase food production, as the planet's continuously growing population demands maximized crop yields [1].
The cytochrome bc1 complex (also known as complex III) is one of the most important fungicidal targets. The complex is an essential component of the cellular respiratory chain. It catalyzes the ubiquinol oxidation (UQH2) to quinone (UQ) and channels the resulting electrons to cytochrome c through the ubiquinone cycle (Qo cycle, Figure 1). Furthermore, the ubiquinol oxidation results in a bifurcated electron transfer. One electron is transferred, via the Rieske iron-sulfur protein and cytochrome c1, to cytochrome c, while the second electron passes through two b-type haems (b L and b H ) to reduce ubiquinone at the Qi site [2].
This process results in proton translocation and generates the proton motive force required in the production of ATP. Protons are taken up from the mitochondrial matrix when ubiquinone is reduced at the Qi site and they are released into the intermembrane space when ubiquinol is oxidized at the Qo site.
Inhibition of the activity of the bc1 complex blocks the generation of ATP, leading to cell death. For this reason, inhibitors of the cytochrome bc1 complex have aroused great interest in controlling fungal diseases. The cyt bc1 inhibitors target either the ubiquinol oxidation site (Qo or Q P ) or the ubiquinone reduction site (Qi or Q N ). In this review, we summarize the efforts towards the exploration of natural product scaffolds for the design and development of bc1 complex inhibitors, focusing on what has been observed and achieved in the past decade. In particular, we concentrate on studies focused on optimization of the core natural scaffold by simplification, direct substitution, and/or by use of isosteric modifications. Molecular modeling-based approaches and structure-activity relationship (SAR) studies performed to improve the potency, selectivity, and stability of bioactive natural products are highlighted as well.

Strobilurins
Strobilurins are natural compounds isolated for the first time in 1977 from basidiomycetes. They are the first example of natural fungicides belonging to the group of QoI (quinone outside inhibitors, (FRAC group 11)). Strobilurin-based products, obtained by activity-guided rational design, have been a milestone in the fungicidal market worldwide. Indeed, these compounds have a broad-spectrum of activity, being effective against all four major classes of phytopathogenic organisms (ascomycetes, basidiomycetes, deuteromycetes, and oomycetes). They show acute toxicity against germinating fungal spores and relatively low toxicity for terrestrial animals.
The major drawbacks of these fungicides are their toxicity to aquatic organisms and the increased number of fungal pathogens that are showing occurrence of resistance, mainly due to mutations on mitochondrial genes encoding for the Qo site [4].
Based on this evidence, the development of new strobilurin derivatives is still pressing because of their role in the market and the need of overcoming the explosive increase of fungal resistance worldwide. Natural products have played a crucial role in the investigation of cytochrome bc1 inhibitors, leading to the discovery of one of the most important classes of crop-protection agents, the strobilurins. These compounds still have a dominant position in the global market.
Natural products, with their enormous structural diversity, are an invaluable source of inspiration in the design and development of new biologically active compounds [3]. Having evolved over several millennia to acquire specific ligand-protein binding motifs, these privileged compounds serve as important, biologically pre-validated platforms for the development of new leads in medicinal chemistry and agriculture. However, the structural complexity, toxicity, and unfavorable bioavailability often associated with natural products can limit their potential, which is why as such structural modification is often required.
In this review, we summarize the efforts towards the exploration of natural product scaffolds for the design and development of bc1 complex inhibitors, focusing on what has been observed and achieved in the past decade. In particular, we concentrate on studies focused on optimization of the core natural scaffold by simplification, direct substitution, and/or by use of isosteric modifications. Molecular modeling-based approaches and structure-activity relationship (SAR) studies performed to improve the potency, selectivity, and stability of bioactive natural products are highlighted as well.

Strobilurins
Strobilurins are natural compounds isolated for the first time in 1977 from basidiomycetes. They are the first example of natural fungicides belonging to the group of QoI (quinone outside inhibitors, (FRAC group 11)). Strobilurin-based products, obtained by activity-guided rational design, have been a milestone in the fungicidal market worldwide. Indeed, these compounds have a broad-spectrum of activity, being effective against all four major classes of phytopathogenic organisms (ascomycetes, basidiomycetes, deuteromycetes, and oomycetes). They show acute toxicity against germinating fungal spores and relatively low toxicity for terrestrial animals.
The major drawbacks of these fungicides are their toxicity to aquatic organisms and the increased number of fungal pathogens that are showing occurrence of resistance, mainly due to mutations on mitochondrial genes encoding for the Qo site [4].
Based on this evidence, the development of new strobilurin derivatives is still pressing because of their role in the market and the need of overcoming the explosive increase of fungal resistance worldwide. Various reviews have been published about the design and synthesis of new strobilurin-derived compounds [5,6]. In the next section, we provide an overview of the efforts made in this field in the last 10 years.

From Natural to Synthetic Strobilurins
Strobilurin A (1) was the first compound of this family isolated from Strobilurus tenacellus by Anke et al. [7], followed by strobilurin B (2) and so on ( Figure 2). Although isolated from different fungi, the various strobilurins have a very similar structure characterized by a pharmacophore portion, an unsaturated bridge, and a side chain, varying only in the aromatic ring substituents. The interest for strobilurins increased following the discovery of their fungicidal activity by Anke et al. [7], which led scientists to search for these compounds in various fungi. However, their activity was weak and, more importantly, they exhibited stability problems because of their photo lability.
Molecules 2020, 25, 4582 3 of 28 Various reviews have been published about the design and synthesis of new strobilurin-derived compounds [5,6]. In the next section, we provide an overview of the efforts made in this field in the last 10 years.

From Natural to Synthetic Strobilurins
Strobilurin A (1) was the first compound of this family isolated from Strobilurus tenacellus by Anke et al. [7], followed by strobilurin B (2) and so on ( Figure 2). Although isolated from different fungi, the various strobilurins have a very similar structure characterized by a pharmacophore portion, an unsaturated bridge, and a side chain, varying only in the aromatic ring substituents. The interest for strobilurins increased following the discovery of their fungicidal activity by Anke et al. [7], which led scientists to search for these compounds in various fungi. However, their activity was weak and, more importantly, they exhibited stability problems because of their photo lability.
During the first 1980s, the ICI and the BASF groups simultaneously, although working separately, identified methoxyacrylate stilbene (3, MOAS, Figure 2) as a stable synthetic strobilurin with a benzene ring in place of the central double bond of the original triene system [6]. Compound 3 immediately became the lead synthetic strobilurin because of its stability and increased efficacy ( Figure 2). Successive studies led to the introduction into the market of various synthetic analogs, featuring broad structural variations on the side chain portion as well as containing bioisosteres of the (E)-β-methoxyacrylate pharmacophoric group ( Figure 3) [8].  During the first 1980s, the ICI and the BASF groups simultaneously, although working separately, identified methoxyacrylate stilbene (3, MOAS, Figure 2) as a stable synthetic strobilurin with a benzene ring in place of the central double bond of the original triene system [6].
Compound 3 immediately became the lead synthetic strobilurin because of its stability and increased efficacy ( Figure 2). Successive studies led to the introduction into the market of various synthetic analogs, featuring broad structural variations on the side chain portion as well as containing bioisosteres of the (E)-β-methoxyacrylate pharmacophoric group ( Figure 3) [8].
While the early chemical modifications of the strobilurin scaffold addressed the improvement of physical characteristics (i.e., volatility, stability, resistance to UV breakdown) and resulted in commercial strobilurins like azoxystrobin, picoxystrobin, pyraclostrobin, and trifloxystrobin ( Figure 3), in the latest years new strobilurins were developed to increase their efficacy as well as to overcome the resistance to their action on QoI-resistant strains.
For this purpose, different approaches were used, such as the me-too approach, biorational or chemorational design, fragment-based drug design (FBDD), intermediate derivatization methods based on bioisosteric replacement, and pharmacophore-linked fragment virtual screening (PFVS).
Hao et al. developed a new molecular design method based on PFVS, by integrating the advantages of FBDD and the advantages of docking methods [9]. Through this approach, Yang and coworkers [10] designed and synthesized a series of benzophenone/fluorenone-containing derivatives ( Figure 4) to obtain new strobilurin analogs with higher fungicidal activity. As shown in Figure 4, the O-bridged derivative 14b (K i = 3.28 nmol/L) is more active than its corresponding S-bridged derivative 14a (K i = 13.95 nmol/L) and the NH-bridged derivative 14c (K i > 1000 nmol/L). Interestingly, compound 14d showed an improved binding affinity (K i = 1.89 nmol/L) to the porcine cytochrome bc1 complex (porcine SCR, succinate cytochrome c reductase) compared to the commercial inhibitor azoxystrobin. Compound 3 immediately became the lead synthetic strobilurin because of its stability and increased efficacy ( Figure 2). Successive studies led to the introduction into the market of various synthetic analogs, featuring broad structural variations on the side chain portion as well as containing bioisosteres of the (E)-β-methoxyacrylate pharmacophoric group ( Figure 3) [8].  While the early chemical modifications of the strobilurin scaffold addressed the improvement of physical characteristics (i.e., volatility, stability, resistance to UV breakdown) and resulted in commercial strobilurins like azoxystrobin, picoxystrobin, pyraclostrobin, and trifloxystrobin ( Figure  3), in the latest years new strobilurins were developed to increase their efficacy as well as to overcome the resistance to their action on QoI-resistant strains.
For this purpose, different approaches were used, such as the me-too approach, biorational or chemorational design, fragment-based drug design (FBDD), intermediate derivatization methods based on bioisosteric replacement, and pharmacophore-linked fragment virtual screening (PFVS).
Hao et al. developed a new molecular design method based on PFVS, by integrating the advantages of FBDD and the advantages of docking methods [9]. Through this approach, Yang and coworkers [10] designed and synthesized a series of benzophenone/fluorenone-containing derivatives ( Figure 4) to obtain new strobilurin analogs with higher fungicidal activity. As shown in Figure 4, the O-bridged derivative 14b (Ki = 3.28 nmol/L) is more active than its corresponding S-bridged derivative 14a (Ki = 13.95 nmol/L) and the NH-bridged derivative 14c (Ki > 1000 nmol/L). Interestingly, compound 14d showed an improved binding affinity (Ki = 1.89 nmol/L) to the porcine cytochrome bc1 complex (porcine SCR, succinate cytochrome c reductase) compared to the commercial inhibitor azoxystrobin. The binding mode of compound 14d in Qo site of bc1 complex demonstrated that the pharmacophore of this new inhibitor bound in the same way of typical methoxyacrylate inhibitors, interacting with Phe128, Tyr131, Phe274, and Glu271, and forming an H-bond between the methoxy group of the methoxyacrylate moiety and the backbone nitrogen of Glu271. The presence of the fluorenone ring in 14d significantly improved the π-π interactions with Phe274 compared with that observed for the azoxystrobin, justifying the higher potency of the compound [10].
Moreover, most of the new compounds displayed excellent in vivo fungicidal activity against Sphaerotheca fuliginea at the concentration of 200 μmol/L.
Starting from the scaffold of enoxastrobin (15, Figure 5) developed by Rohm and Haas Company, Xie et al. introduced modifications in the side chain. In order to stabilize the E-styryl group, the authors firstly synthesized a small library of indene-substituted oxime ethers (16, Figure  5) [11]. Afterward, they prepared new oxime ethers featuring heterocyclic moieties, which could drive some physicochemical properties, such as lipophilicity and solubility, toward the optimal balanced range for uptake and bioavailability. In particular, benzothiophene, benzofuran, and indole analogs were developed ( Figure 5) [12][13][14]. physicochemical properties, such as lipophilicity and solubility, toward the optimal balanced range for uptake and bioavailability. In particular, benzothiophene, benzofuran, and indole analogs were developed ( Figure 5) [12][13][14].
The introduction of a chlorine atom in para position on the phenyl ring had a positive effect on the fungicidal activity of these compounds. The p-chlorophenyl derivative 20 ( Figure 6) was the most effective inhibitor against all the tested pathogens with MIC (minimum inhibitory concentration) values in the range of 16-64 µ g/L). The inhibition of resazurin (RZ) reduction indicated that this compound has a mechanism of action similar to azoxystrobin, inhibiting the mitochondrial respiration by binding to the Qo site of cytochrome b (EC50 for inhibition of RZ reduction in D. oryzae by azoxystrobin and compound 20 were 3.42 ± 0.03 µ g/mL and 3.62 ± 0.21 µ g/mL, respectively). At the same time, the authors observed that 1,2,3-triazole derivatives (i.e., compound 21) were ineffective in controlling the fungal growth at the highest concentration (512 µ g/mL), and concluded that the 1,2,4-triazole moiety contributed to the antifungal activity of these derivatives.
Among the furanyl derivatives, compounds featuring a three-or four-carbon chain (22b, 22c, Table 2) linked to the 1,2,4-triazole group had the highest inhibitory effect ( Table 2). The EC 50 values of compounds 22a (14.82 mg/L) and 22b (18.72 mg/L) against Cercospora arachidicola Hori were much lower than that of the control (40.54 mg/L). Moreover, the introduction of a bromine atom linked to the furan ring allowed to obtain compound 22d with an inhibition rate of 90% against Rhizoctonia cerealis. The EC 50 values of compounds 22d (9.89 mg/L) and 22e (8.66 mg/L) against Rhizoctonia cerealis were lower than that of azoxystrobin (10.86 mg/L). In 2017 the same research group obtained strobilurin analogs containing the 1,3,4-oxadiazole moiety [22] (Figure 6). The antifungal activity of some representative compounds (23a-c, 24) is reported in Table 3. In general, the efficacy of these new analogs in controlling the fungal growth of Sclerotinia sclerotiorum was comparable to that of azoxystrobin. Interestingly, some of them were more effective than the control against Rhizoctonia cerealis. Table 3. Inhibition of fungal growth (%) of compounds 23a-c and 24 at 50 mg/L and EC 50 (mg/L) against Sclerotinia sclerotiorum and Rhizoctonia cerealis from Reference [22]. Li and coworkers designed new strobilurin analogs combining the 2-(2-methylphenyl)-3methoxyacrylate pharmacophore with a series of 4-halo-5-aryl-1,2,3-triazoles, which had previously showed excellent fungicidal activity against Phytophthora capsici and Sclerotinia sclerotiorum [23]. The compounds exhibited moderate to good fungicidal activity against Phytophthora capsici and Alternaria alternata. In particular, compounds 25a and 25b ( Figure 6) inhibited the fungal growth of Phytophthora capsici up to 73.6%, a significantly higher value than the 42.5% observed for difenoconazole.
Based on the active substructure combination approach, Liu et al. designed and synthesized new azoxystrobin analogs with various substituted phenyl groups linked to the pyrimidine ring [24]. The analogs exhibited moderate or remarkable antifungal activities against three tested fungi, Botrytis cinerea, Colletotrichum orbiculare, and Phytophthora capsici Leonian.
Preliminary structure-activity relationships (SAR) studies showed that the introduction of an electron-withdrawing group on the phenyl ring increased the activity against all the tested fungi with the following trend: 2-substituted phenyl derivative > 4-substituted phenyl derivative > 3-substituted phenyl derivative. Conversely, the effect of the introduction of electron-donating groups was the following (in terms of activity): 4-substituted phenyl derivative > 3-substituted phenyl derivative > 2-substituted phenyl derivative. Interestingly, among the tested compounds, the 2,5-dimethylphenyl derivative 26 ( Figure 6) displayed the most promising activity, with a growth inhibition rate (%) comparable (75-78%) to that of azoxystrobin (66-71%) against Colletotrichum orbiculare, Botrytis cinerea Pers, and Phytophthora capsici Leonian.
The discovery of coumoxystrobin (27, Figure ??) [25], containing the coumarin skeleton, paved the way to the synthesis of new strobilurins. Starting from the structure of this compound, Liu et al. investigated the activity of derivatives containing the quinolin-2(1H)-one moiety as a bioisoster of coumarin (Figure ??) and tested their efficacy in controlling the growth of ten plant pathogens [26].
Molecules 2020, 25 Figure 7) [25], containing the coumarin skeleton, paved the way to the synthesis of new strobilurins. Starting from the structure of this compound, Liu et al. investigated the activity of derivatives containing the quinolin-2(1H)-one moiety as a bioisoster of coumarin ( Figure 7) and tested their efficacy in controlling the growth of ten plant pathogens [26].

28 30
Intermediate derivatization method The modification of the lactam ring resulted in crucial for the fungicidal activity. The introduction of the ethyl group at N1-position gave compounds more effective against all the tested fungi compared to the N-H-containing compounds (e.g., 29 vs. 28). Compounds featuring an amino group at C7 position of the quinolinone moiety (e.g., 30) displayed better antifungal activity against Gibberella fujikuroi, Sclerotinia sclerotiorum, Phytophthora infestans, Alternaria solani, and Fusarium graminearum than compounds containing an ether linkage, which, on the contrary, was beneficial to the antifungal activity against Fusarium oxysporum and Rhizoctonia cerealis.
In particular, the growth inhibition activity of compound 29 was comparable and in some cases higher than the activity of the reference coumoxystrobin. The authors reported the EC50 values of compounds 29 (EC50 3.4646 μg/mL) and 30 (EC50 8.9148 μg/mL) against Rhizoctonia cerealis. The results indicate that compound 29 is 4.17-and 5.07-fold more effective than coumoxystrobin (EC50 14.4593 μg/mL) and azoxystrobin (EC50 17.5804 μg/mL), respectively, against this pathogen.
In 2017 Chen et al. published two papers about the synthesis of novel fungicides containing isothiazole-, thiadiazole-, and thiazole-based structures ( Figure 8).
Using trifloxystrobin (7) structure as a template, the authors first designed and synthesized 3,4-dichloroisothiazole-containing strobilurins [27]. This heterocyclic scaffold can be found in numerous biologically active molecules [28]. In particular, the 3,4-dichloroisothiazole-5-carboxylic acid derivative isotianil was developed as a novel fungicide with activating defense responses against a wide range of plant pathogens [29].
Compound 31 ( Figure 8) exhibited excellent activity with EC50 of 0.07 μg/mL and 0.49 μg/mL against R. cerealis and P. infestans, respectively (Table 4). These values are comparable to that of the positive control azoxystrobin. Moreover, compound 31 was more effective than azoxystrobin in controlling the growth of G. zeae and B. cinerea (Table 4).
Similarly, the same authors prepared novel 1,2,3-thiadiazole and thiazole-based strobilurins The modification of the lactam ring resulted in crucial for the fungicidal activity. The introduction of the ethyl group at N1-position gave compounds more effective against all the tested fungi compared to the N-H-containing compounds (e.g., 29 vs. 28). Compounds featuring an amino group at C7 position of the quinolinone moiety (e.g., 30) displayed better antifungal activity against Gibberella fujikuroi, Sclerotinia sclerotiorum, Phytophthora infestans, Alternaria solani, and Fusarium graminearum than compounds containing an ether linkage, which, on the contrary, was beneficial to the antifungal activity against Fusarium oxysporum and Rhizoctonia cerealis.
In particular, the growth inhibition activity of compound 29 was comparable and in some cases higher than the activity of the reference coumoxystrobin. The authors reported the EC 50 values of compounds 29 (EC 50 3.4646 µg/mL) and 30 (EC 50 8.9148 µg/mL) against Rhizoctonia cerealis.
In 2017 Chen et al. published two papers about the synthesis of novel fungicides containing isothiazole-, thiadiazole-, and thiazole-based structures ( Figure 8).
Using trifloxystrobin (7) structure as a template, the authors first designed and synthesized 3,4-dichloroisothiazole-containing strobilurins [27]. This heterocyclic scaffold can be found in numerous biologically active molecules [28]. In particular, the 3,4-dichloroisothiazole-5-carboxylic acid derivative isotianil was developed as a novel fungicide with activating defense responses against a wide range of plant pathogens [29]. The most active compounds of each series were evaluated as fungicide candidates against Sphaerotheca fuliginea and Pseudoperonospora cubensis in cucumber fields.
Both the 3,4-dichloroisothiazole derivative 31 and the 1,2,3-thiadiazole derivative 32 showed better efficacy against cucumber S. fuliginea than azoxystrobin and trifloxystrobin at the same application rate (37.5 g ai/ha). Moreover, compound 32 showed significantly better efficacy (p < 0.05) against P. cubensis than trifloxystrobin and efficacy comparable to azoxystrobin (at the same application rate of 75 g ai/ha). Compounds 31 showed efficacy against P. cubensis comparable to pyraclostrobin but significantly better than that of trifloxystrobin at the same application rate (75 g ai/ha.). Table 4. EC50 value of compounds 31 and 32 from References [27,30]. An interesting approach to develop new strobilurin-based fungicides was recently reported by Su et al. [31]. The strategy is based on the combination of the pharmacophore moieties of strobilurins (β-methoxyacrylate, methoxyiminoacetamide, or methoxy-N-methylacrylamide) with a monoterpenic phenol typically found in plant essential oils (EOs).
Antifungal activity of EOs has been frequently associated with the presence of monoterpenic phenols, such as thymol, carvacrol, and paeonol [32]. However, EOs components are generally volatile, unstable to light or heat. Additionally, their short-term fungicidal efficacy and slow action considerably restrict their possible practical application.
The authors prepared seventeen new compounds and tested their potential fungicidal activity against eleven species of plant pathogen fungi. The structure of representative analogs containing carvacrol and thymol moiety and EC50 values against Sclerotinia sclerotiourum are reported in Figure  9. Compound 33 exhibited a very interesting activity against Pestalotiopsis theae, Phomopsis adianticola, Sclerotinia sclerotiorum, and Magnapothe grisea as well. Compound 31 (Figure 8) exhibited excellent activity with EC 50 of 0.07 µg/mL and 0.49 µg/mL against R. cerealis and P. infestans, respectively (Table 4). These values are comparable to that of the positive control azoxystrobin. Moreover, compound 31 was more effective than azoxystrobin in controlling the growth of G. zeae and B. cinerea (Table 4). Similarly, the same authors prepared novel 1,2,3-thiadiazole and thiazole-based strobilurins [30]. The 1,2,3-thiadiazole derivative 32 ( Figure 8) exhibited excellent activities against G. zeae, S. sclerotiorum, and R. cerealis with EC 50 values of 2.68, 0.44, and 0.01 µg/mL, respectively.
The most active compounds of each series were evaluated as fungicide candidates against Sphaerotheca fuliginea and Pseudoperonospora cubensis in cucumber fields.
Both the 3,4-dichloroisothiazole derivative 31 and the 1,2,3-thiadiazole derivative 32 showed better efficacy against cucumber S. fuliginea than azoxystrobin and trifloxystrobin at the same application rate (37.5 g ai/ha). Moreover, compound 32 showed significantly better efficacy (p < 0.05) against P. cubensis than trifloxystrobin and efficacy comparable to azoxystrobin (at the same application rate of 75 g ai/ha). Compounds 31 showed efficacy against P. cubensis comparable to pyraclostrobin but significantly better than that of trifloxystrobin at the same application rate (75 g ai/ha.).
An interesting approach to develop new strobilurin-based fungicides was recently reported by Su et al. [31]. The strategy is based on the combination of the pharmacophore moieties of strobilurins (β-methoxyacrylate, methoxyiminoacetamide, or methoxy-N-methylacrylamide) with a monoterpenic phenol typically found in plant essential oils (EOs).
Antifungal activity of EOs has been frequently associated with the presence of monoterpenic phenols, such as thymol, carvacrol, and paeonol [32]. However, EOs components are generally volatile, unstable to light or heat. Additionally, their short-term fungicidal efficacy and slow action considerably restrict their possible practical application.
The authors prepared seventeen new compounds and tested their potential fungicidal activity against eleven species of plant pathogen fungi. The structure of representative analogs containing carvacrol and thymol moiety and EC 50 values against Sclerotinia sclerotiourum are reported in Figure 9. Using pyraclostrobin (10) as a lead compound, Liu and coworkers [33] obtained 44 new strobilurin analogs by introducing the methoxyiminoacetate or the methoxyacrylate pharmacophore into halo-or (un)substituted arylpyrazole scaffolds ( Figure 10). In particular, the authors focused their attention on three structural aspects: the effect of different pharmacophores and their position; the effect of nature and position of the substituent on the terminal benzene ring; the effect of halogen substituent on the pyrazole ring.
The SARs study highlighted that the combination of both methoxyiminoacetate pharmacophore and electron-withdrawing substituent R on phenyl ring improved the fungicidal activity. Moreover, the type and size of the halogen on the pyrazole ring resulted crucial for the efficacy. In particular, the presence of a chloro substituent had a positive effect on the fungicidal activity, whether it was on the pyrazole ring or on the phenyl ring. The collected biological data and the comparative molecular field analysis allowed the authors to build up a 3D-QSAR model for these new strobilurin analogs with high correlative and predictive abilities [33].
Among the obtained analogs, compounds 38-41 ( Figure 10) exhibited 98.94%, 83.40%, 71.40%, and 65.87% inhibition rates at 0.1 μg/mL against Rhizoctonia solani, respectively, which are better than commercial pyraclostrobin (35.73%). Using pyraclostrobin (10) as a lead compound, Liu and coworkers [33] obtained 44 new strobilurin analogs by introducing the methoxyiminoacetate or the methoxyacrylate pharmacophore into halo-or (un)substituted arylpyrazole scaffolds ( Figure 10). In particular, the authors focused their attention on three structural aspects: the effect of different pharmacophores and their position; the effect of nature and position of the substituent on the terminal benzene ring; the effect of halogen substituent on the pyrazole ring.
The SARs study highlighted that the combination of both methoxyiminoacetate pharmacophore and electron-withdrawing substituent R on phenyl ring improved the fungicidal activity. Moreover, the type and size of the halogen on the pyrazole ring resulted crucial for the efficacy. In particular, the presence of a chloro substituent had a positive effect on the fungicidal activity, whether it was on the pyrazole ring or on the phenyl ring. The collected biological data and the comparative molecular field analysis allowed the authors to build up a 3D-QSAR model for these new strobilurin analogs with high correlative and predictive abilities [33].
Among the obtained analogs, compounds 38-41 ( Figure 10) exhibited 98.94%, 83.40%, 71.40%, and 65.87% inhibition rates at 0.1 µg/mL against Rhizoctonia solani, respectively, which are better than commercial pyraclostrobin (35.73%). In 2018, Wang et al. [34] introduced a pyridinyl group as bridge portion and side chain modifications, obtaining a library of N-ortho substituted pyridine analogs of pyraclostrobin ( Figure  11). Besides the in vitro fungicidal activities against five important plant pathogens (Botrytis cinerea, Phytophthora capsici, Fusarium sulphureum, Gloeosporium pestis, and Sclerotinia sclerotiorum), the authors evaluated the percentage of inhibition, the IC50 values, and predicted the total binding free energy of representative compounds against porcine SCR (Table 5). Compound 43 exhibited the lowest IC50 value (0.95 µ M). In contrast, compound 42, without the methylene linker, displayed only 11% inhibitory activity at 100 µ M concentration, suggesting that the flexible side chain is favorable for the enzymatic inhibition activity. In general, the presence of a hydrophobic side chain is critical for the antifungal activity of these analogs. Moreover, biaryl rings at the side chain are preferable and certain flexibility is beneficial to the inhibitory activity against SCR bc1 complex.
Interestingly, docking studies into the binding pocket of the cytochrome bc1 complex showed that the pyridinyl group can form stable arene-H interaction with residue proline-271, thus improving the binding to cytochrome bc1 complex. In 2018, Wang et al. [34] introduced a pyridinyl group as bridge portion and side chain modifications, obtaining a library of N-ortho substituted pyridine analogs of pyraclostrobin ( Figure 11). Besides the in vitro fungicidal activities against five important plant pathogens (Botrytis cinerea, Phytophthora capsici, Fusarium sulphureum, Gloeosporium pestis, and Sclerotinia sclerotiorum), the authors evaluated the percentage of inhibition, the IC 50 values, and predicted the total binding free energy of representative compounds against porcine SCR (Table 5). Compound 43 exhibited the lowest IC 50 value (0.95 µM). In contrast, compound 42, without the methylene linker, displayed only 11% inhibitory activity at 100 µM concentration, suggesting that the flexible side chain is favorable for the enzymatic inhibition activity. In general, the presence of a hydrophobic side chain is critical for the antifungal activity of these analogs. Moreover, biaryl rings at the side chain are preferable and certain flexibility is beneficial to the inhibitory activity against SCR bc1 complex.
Interestingly, docking studies into the binding pocket of the cytochrome bc1 complex showed that the pyridinyl group can form stable arene-H interaction with residue proline-271, thus improving the binding to cytochrome bc1 complex.    [35]. Although no new molecules resulted more active than the lead compound, they isolated compound 45 (Figure 12), which showed a broad spectrum of activity, as a byproduct of their synthetic pathway. Recently, Matsuzaki and coworkers [36] obtained new Qo inhibitors characterized by a tetrazolinone pharmacophore and a methyl group at position 3 on the central phenyl ring. These compounds were designed to overcome the resistance developed by different pathogenic species and associated with the G143A mutation at the target site. The presence of the methyl group of the alanine residue at position 143 seems to cause severe steric hindrance at the central linking rings of QoIs.
Through an initial random screening on 200 QoI-like compounds to identify the key structural moieties to retain the activity against G143A mutant, the authors found that compound 46 ( Figure  13), featuring the tetrazolinone pharmacophore and the same side chain of trifloxystrobin, was less affected by this mutation, showing a resistant factor RF = 2 ( Figure 13, Table 6). After a careful structure-activity relationship investigation, the authors obtained compound 47 (Figure 14) with the same side chain of pyraclostrobin (4-chlorophenylpyrazole structure). The EC50 of compound 47 in Figure 11. Structure of pyraclostrobin analogs from Reference [34].   [35]. Although no new molecules resulted more active than the lead compound, they isolated compound 45 (Figure 12), which showed a broad spectrum of activity, as a byproduct of their synthetic pathway.    [35]. Although no new molecules resulted more active than the lead compound, they isolated compound 45 (Figure 12), which showed a broad spectrum of activity, as a byproduct of their synthetic pathway. Recently, Matsuzaki and coworkers [36] obtained new Qo inhibitors characterized by a tetrazolinone pharmacophore and a methyl group at position 3 on the central phenyl ring. These compounds were designed to overcome the resistance developed by different pathogenic species and associated with the G143A mutation at the target site. The presence of the methyl group of the alanine residue at position 143 seems to cause severe steric hindrance at the central linking rings of QoIs.
Through an initial random screening on 200 QoI-like compounds to identify the key structural moieties to retain the activity against G143A mutant, the authors found that compound 46 ( Figure  13), featuring the tetrazolinone pharmacophore and the same side chain of trifloxystrobin, was less affected by this mutation, showing a resistant factor RF = 2 ( Figure 13, Table 6). After a careful structure-activity relationship investigation, the authors obtained compound 47 (Figure 14) with the same side chain of pyraclostrobin (4-chlorophenylpyrazole structure). The EC50 of compound 47 in Recently, Matsuzaki and coworkers [36] obtained new Qo inhibitors characterized by a tetrazolinone pharmacophore and a methyl group at position 3 on the central phenyl ring. These compounds were designed to overcome the resistance developed by different pathogenic species and associated with the G143A mutation at the target site. The presence of the methyl group of the alanine residue at position 143 seems to cause severe steric hindrance at the central linking rings of QoIs.
Through an initial random screening on 200 QoI-like compounds to identify the key structural moieties to retain the activity against G143A mutant, the authors found that compound 46 (Figure 13), featuring the tetrazolinone pharmacophore and the same side chain of trifloxystrobin, was less affected by this mutation, showing a resistant factor RF = 2 ( Figure 13, Table 6). After a careful structure-activity relationship investigation, the authors obtained compound 47 (Figure 14) with the same side chain of pyraclostrobin (4-chlorophenylpyrazole structure). The EC 50 of compound 47 in the wild-type sensitive strain (WT) and G143A mutant were both 0.02 ppm, showing a 10-to 20-fold increase of activity compared to compound 46 (Table 6).
Molecules 2020, 25,4582 13 of 28 the wild-type sensitive strain (WT) and G143A mutant were both 0.02 ppm, showing a 10-to 20-fold increase of activity compared to compound 46 (Table 6). Finally, the introduction of the 3-methyl group on the central phenyl ring led to obtain methyltetraprole 48, showing a 10-fold higher activity than that of compound 47 ( Table 6). The authors hypothesized that the presence of the methyl group might have a role in limiting the rotation of the side chain and in avoiding steric hindrance between the QoI and the mutated target site.
Interestingly, only the modification of the conjugated double bond geometry (cyrmenin B1 vs. its (8E,10E)-geometrical isomer 50a) was tolerated. The complete lack of activity of compounds 50b and 50c underlined that maintaining the lipophilicity of the molecule was not enough to have active compounds. The exo double bond of the parent molecule was also crucial for the antifungal activity. In fact, both compound 51, bearing an alanine residue, and the compound bearing a serine residue (52) were inactive against all the tested strains. The modification of the expected crucial β-methoxyacrylate group was also deleterious (compound 53). All these results evidenced a limitation for the development of new fungicide compounds derived from natural cyrmenins.

Mixothiazols, Melithiazols and Fulvuthiacens
Myxobacteria are a rich source of various biologically active secondary metabolites. These include the antifungal agents myxothiazol and melithiazol, which are very potent inhibitors of the electron transport through the cytochrome bc1 complex of the eukaryotic respiratory chain. Finally, the introduction of the 3-methyl group on the central phenyl ring led to obtain methyltetraprole 48, showing a 10-fold higher activity than that of compound 47 ( Table 6). The authors hypothesized that the presence of the methyl group might have a role in limiting the rotation of the side chain and in avoiding steric hindrance between the QoI and the mutated target site.
The bioassay results clearly indicated the relevance of each portion of the parent molecule (lipophilic unsaturated chain, dehydroalanine moiety, and β-methoxyacrylate system), leaving very little space for synthetic modifications.
Interestingly, only the modification of the conjugated double bond geometry (cyrmenin B 1 vs. its (8E,10E)-geometrical isomer 50a) was tolerated. The complete lack of activity of compounds 50b and 50c underlined that maintaining the lipophilicity of the molecule was not enough to have active compounds. The exo double bond of the parent molecule was also crucial for the antifungal activity. In fact, both compound 51, bearing an alanine residue, and the compound bearing a serine residue (52) were inactive against all the tested strains. The modification of the expected crucial β-methoxyacrylate group was also deleterious (compound 53). All these results evidenced a limitation for the development of new fungicide compounds derived from natural cyrmenins.

Mixothiazols, Melithiazols and Fulvuthiacens
Myxobacteria are a rich source of various biologically active secondary metabolites. These include the antifungal agents myxothiazol and melithiazol, which are very potent inhibitors of the electron transport through the cytochrome bc1 complex of the eukaryotic respiratory chain.
In 1980, Reichenbach et al. and Hofle et al. reported the isolation of a novel myxobacterial antibiotic named myxothiazol A (54) [40,41], whose structure is characterized by a central bis-thiazole unit linked to a β-methoxyacrylate moiety and a heptadienyl side-chain bearing a stereogenic center at the α-position of the thiazole ring ( Figure 15). Acting as potent inhibitors of the electron transport through the cytochrome bc1 complex, both myxothiazol A (54) and its corresponding methyl ester, myxothiazol Z (55), exhibit broad antifungal activity as well as significant cytotoxicity against different human tumor cell lines with IC 50 values reaching as low as 0.01 ng/mL [42,43]. The high mammalian toxicity hampered the development of myxothiazol synthetic derivatives. During the following three decades, more than 20 structurally related fungicides have been isolated from various strains of myxobacteria, including the congeneric melithiazols.
Unlike myxothiazols, all melithiazols occur only as methyl esters and lack the lipophilic heptadienyl side chain. Moreover, the greatly reduced mammalian toxicity of these metabolites with respect to the toxicity of myxothiazols makes melithiazols interesting for the development of new fungicides. Unfortunately, due to the poor amounts of melithiazols usually obtained from fermentation, it has not been possible to develop a derivatization program to investigate structure-activity relationships. However, synthetic efforts to obtain melithiazol B (56) [45] and C (57) [46] through oxidative degradation of myxothiazol A and reductive cleavage of a thiazole ring, respectively, allowed to test the biological activities of intermediates and derivatives, to obtain some structure-activity correlations.  Concerning melithiazol B (56) and its derivatives, the authors highlighted that compounds with an ester pharmacophore and a short polar side-chain (e.g., compounds 58 and 59) showed high antifungal activity (Table 7). Moreover, in general, the cytotoxicity decreased significantly by decreasing the lipophilicity of these molecules.
Structure-activity relationship studies on melithiazol C (57) derivatives showed that the amide analog (61) and the 14-hydroxy derivative lost the antifungal activity (62a), whereas the 14-ethoxy derivative (62b) showed good antifungal activity and only slightly increased cytotoxicity. Moreover, the (6Z)-isomer of melithiazol C was essentially inactive in all test systems. Melithiazols A-N contain a β-substituted β-methoxyacrylate pharmacophore and act as inhibitors of the cytochrome bc1 complex as well [44].
Unlike myxothiazols, all melithiazols occur only as methyl esters and lack the lipophilic heptadienyl side chain. Moreover, the greatly reduced mammalian toxicity of these metabolites with respect to the toxicity of myxothiazols makes melithiazols interesting for the development of new fungicides. Unfortunately, due to the poor amounts of melithiazols usually obtained from fermentation, it has not been possible to develop a derivatization program to investigate structure-activity relationships. However, synthetic efforts to obtain melithiazol B (56) [45] and C (57) [46] through oxidative degradation of myxothiazol A and reductive cleavage of a thiazole ring, respectively, allowed to test the biological activities of intermediates and derivatives, to obtain some structure-activity correlations.
Concerning melithiazol B (56) and its derivatives, the authors highlighted that compounds with an ester pharmacophore and a short polar side-chain (e.g., compounds 58 and 59) showed high antifungal activity (Table 7). Moreover, in general, the cytotoxicity decreased significantly by decreasing the lipophilicity of these molecules. Structure-activity relationship studies on melithiazol C (57) derivatives showed that the amide analog (61) and the 14-hydroxy derivative lost the antifungal activity (62a), whereas the 14-ethoxy derivative (62b) showed good antifungal activity and only slightly increased cytotoxicity. Moreover, the (6Z)-isomer of melithiazol C was essentially inactive in all test systems.
The vinyl derivative 60 and the (11E) and (11Z) methoximes 63a and 63b showed an exceptionally high antifungal activity and low cytotoxicity, comparable to that of the commercial fungicide kresoxim-methyl.
Recently, Panter and coworkers [47] elucidated the structure of two new secondary metabolites (fulvuthiacene A (64) and B (65), Figure 15) isolated from Myxococcus fulvus MCy9280 by combining the statistics-based mining of mass spectrometry approach and multidimensional NMR spectroscopy.
These compounds contain a terminal β-methoxymethyl acrylate moiety and their evaluation provided new insights into the overall structure-activity relationship picture of the βmethoxyacrylate class of bc1 complex inhibitors.
Indeed, fulvuthiacene A and B showed very poor activity against a panel of fungi and yeast. Comparing the NADH oxidation IC 50 values of known naturally occurring methoxymethacrylate type respiratory chain inhibitors, the authors observed that a wide variety of residues are tolerated in C-2 position of the β-methoxymethacrylate pharmacophore (strobilurin A IC 50 = 83 ng/mL, oudemansin IC 50 = 400 ng/mL, and cyrmenin A IC 50 = 27 ng/mL). On the contrary, for compounds bearing a C-3 extended β-methoxyacrylate warhead, the structural requirements are very narrow and a bisthiazol or thiazolinethiazol moiety is needed to produce a strong bc-1 inhibitor (melithiazol B IC 50 = 18 ng/mL, melithiazol C IC 50 = 1600 ng/mL, fulvuthiacene A no inhibition of NADH oxidation up to 64,000 ng/mL). The results reported in this study might thus serve as starting points for the development of the next generation of β-methoxymethacrylate fungicides.

Miuraenamides
Miuraenamide A-F (66-71, Figure 16) are cyclic hybrid polyketide-peptide antibiotics isolated from Paraliomixa miuraensis, a slightly halophilic myxobacterium discovered by Ojika et al. at the seashore on Miura Peninsula in Kanagawa, Japan [48,49]. The biological properties of these natural products included antimicrobial activity and inhibitory activity against NADH oxidase.  Structure-activity relationship studies on miuraenamides A-F and related derivatives were conducted on the phytopathogen Phytophthora capsici ( Table 8). The authors highlighted that the type of halogen present in the tyrosine residue in miuraenamides A-C is not important for their activity. Moreover, the presence and the geometry of the β-methoxyacrylate moiety (C12-C14, compounds 69, 70, and 74) affects the activity of these compounds. The lipophilicity of the polyketide moiety and the free phenol group on the peptide portion of the structure seem to be important for the activity of these compounds, as demonstrated by the low activity observed for miuraenamide F (71) and acetate 72. The macrocyclic structure of the miuraenamides is essential as well, with the anti-Phytophthora activity being completely lost in the case of the ring-opened derivatives 74 and 75, although these compounds contain the β-methoxyacrylate moiety.  Structure-activity relationship studies on miuraenamides A-F and related derivatives were conducted on the phytopathogen Phytophthora capsici ( Table 8). The authors highlighted that the type of halogen present in the tyrosine residue in miuraenamides A-C is not important for their activity. Moreover, the presence and the geometry of the β-methoxyacrylate moiety (C12-C14, compounds 69, 70, and 74) affects the activity of these compounds. The lipophilicity of the polyketide moiety and the free phenol group on the peptide portion of the structure seem to be important for the activity of these compounds, as demonstrated by the low activity observed for miuraenamide F (71) and acetate 72. The macrocyclic structure of the miuraenamides is essential as well, with the anti-Phytophthora activity being completely lost in the case of the ring-opened derivatives 74 and 75, although these compounds contain the β-methoxyacrylate moiety.

Crocacins
In 1994, crocacins A, B, C, and D (76-79, Figure 17) were isolated from the myxobacterium Chondromyces crocatus. The compounds crocacin A and D inhibited the electron transport chain at complex III in a beef heart mitochondrial respiration assay and inhibited the growth of several fungi in vitro. [50]. Although natural crocacins had interesting biological activity, their practical use would be severely limited by their physical properties, such as the very poor photostability and their structural complexity. However, the emergence of resistance to different pathogenic species associated with G143A mutation renewed the interest toward crocacin A, as it showed little or no cross-resistance to a strobilurin-resistant strain of Plasmopara viticola in small vine plants in the glasshouse, or against a strain of yeast engineered with the G143A mutation.

Crocacins
In 1994, crocacins A, B, C, and D (76-79, Figure 17) were isolated from the myxobacterium Chondromyces crocatus. The compounds crocacin A and D inhibited the electron transport chain at complex III in a beef heart mitochondrial respiration assay and inhibited the growth of several fungi in vitro. [50]. Although natural crocacins had interesting biological activity, their practical use would be severely limited by their physical properties, such as the very poor photostability and their structural complexity. However, the emergence of resistance to different pathogenic species associated with G143A mutation renewed the interest toward crocacin A, as it showed little or no cross-resistance to a strobilurin-resistant strain of Plasmopara viticola in small vine plants in the glasshouse, or against a strain of yeast engineered with the G143A mutation.  Crowley and coworkers [51] studied the binding of crocacins to the active site in order to design structurally simpler and more stable synthetic analogs. An inhibitor binding model to the mammalian cytochrome bc1 complex was constructed, which was consistent with X-ray crystallographic analysis of an analog bound to the chicken heart cytochrome bc1 complex. The proposed binding mode for the crocacins combined some binding features of stigmatellin A (80, Figure 17, a potent inhibitor of the quinol oxidation (Qo) site of the cytochrome bc1 complex) and some binding features of strobilurins. The differences in binding between the crocacins and methoxyacrylate stilbene could explain the apparent lack of cross-resistance of the crocacins with strobilurins.

NH
The binding site model, coupled with further molecular modeling, was used to design analogs of crocacins A (76) and D (79), with mixed results. To improve the stability of the synthetic analogs, efforts were made to replace the three double bonds, which most likely could give rise to significant instability to sunlight, with more robust groups, such as aromatic rings. This strategy had been successful in the development of stable analogs of the strobilurins.
Replacement of the side chain with simple n-alkyl chains or with side chains containing aromatic rings (e.g., n-alkoxybenzamide) gave compounds very active in the beef heart mitochondrial respiration assay but inactive as fungicides. Further modeling showed that substituting the benzamide with a 4-substituted benzyl group gave a very good overlay with the crocacin side chain. Some compounds (84-87) showed not only potent inhibition of respiration but also activity on vine downy mildew on small vine plants similar to or better than crocacins (Table 9). Crowley and coworkers [51] studied the binding of crocacins to the active site in order to design structurally simpler and more stable synthetic analogs. An inhibitor binding model to the mammalian cytochrome bc1 complex was constructed, which was consistent with X-ray crystallographic analysis of an analog bound to the chicken heart cytochrome bc1 complex. The proposed binding mode for the crocacins combined some binding features of stigmatellin A (80, Figure 17, a potent inhibitor of the quinol oxidation (Qo) site of the cytochrome bc1 complex) and some binding features of strobilurins. The differences in binding between the crocacins and methoxyacrylate stilbene could explain the apparent lack of cross-resistance of the crocacins with strobilurins. The binding site model, coupled with further molecular modeling, was used to design analogs of crocacins A (76) and D (79), with mixed results. To improve the stability of the synthetic analogs, efforts were made to replace the three double bonds, which most likely could give rise to significant instability to sunlight, with more robust groups, such as aromatic rings. This strategy had been successful in the development of stable analogs of the strobilurins.
Replacement of the side chain with simple n-alkyl chains or with side chains containing aromatic rings (e.g., n-alkoxybenzamide) gave compounds very active in the beef heart mitochondrial respiration assay but inactive as fungicides. Further modeling showed that substituting the benzamide with a 4-substituted benzyl group gave a very good overlay with the crocacin side chain. Some compounds (84-87) showed not only potent inhibition of respiration but also activity on vine downy mildew on small vine plants similar to or better than crocacins (Table 9). Table 9. Activity of crocacin analogues on NADH oxidase and vine downy mildew on plants.

Compd.
Molecules 2020, 25, 4582 19 of 28 The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies  The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 36   10 NT The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 21 >100 5

84
Molecules 2020, 25, 4582 19 of 28 The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 17 10 12

85
Molecules 2020, 25, 4582 19 of 28 The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 9 25 20 The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.

86
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 18 10 NT

87
Compd. The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies 16 10 NT The fact that some of the designed compounds were inactive despite good modeling into the active site highlights the limitations of a rather simple model based on shape fit, the complementarity of hydrogen bonding groups, and ligand conformational analysis, as the authors themselves suggest. Actually, other important factors, such as solvation and entropic effects, could be critical in binding interactions.
Nevertheless, the design strategy was successful in that analogs of the naturally occurring fungicide crocacin D (79) were synthesized. These compounds were active both in a respiration assay, on fungi, and on plants, and were significantly more stable than the natural compounds.

Neopeltolide
Neopeltolide (88, Figure 18), a marine natural product isolated in 2007, was found to inhibit the bovine heart mitochondrial cytochrome bc1 complex with an IC 50 value of 2.0 nM [52]. However, its complex structure prevented the obtainment of the compound by chemical synthesis. SAR studies showed that the carbamate-containing oxazole moiety was the key structural feature, whereas the 14-membered macrolactone moiety did not make a significant contribution to binding.
Zhu and coworkers [53] determined the binding mode of neopeltolide by integrating molecular docking, molecular dynamics (MD) simulations, and molecular mechanics Poisson-Boltzmann surface area calculations (MM/PBSA), which showed that neopeltolide is a Qo site inhibitor of the bc1 complex. Based on neopeltolide binding mode, structural modification was carried out with the aim to simplify its structure. Thus, a series of new neopeltolide derivatives with much simpler chemical structures were rationally designed and synthesized.
Compound 89 bearing a naphthylether fragment in place of the structurally complex 14-membered macrolactone moiety showed high inhibitory activity against porcine SCR, with an IC 50 value of 0.047 µM. The introduction of a bromo group on the naphthalene ring (compounds 90a and 90b) further improved the activity against SCR. Compound 90a was the most potent candidate against SCR (IC 50 = 12 nM, Table 10) [53]. Zhu and coworkers [53] determined the binding mode of neopeltolide by integrating molecular docking, molecular dynamics (MD) simulations, and molecular mechanics Poisson-Boltzmann surface area calculations (MM/PBSA), which showed that neopeltolide is a Qo site inhibitor of the bc1 complex. Based on neopeltolide binding mode, structural modification was carried out with the aim to simplify its structure. Thus, a series of new neopeltolide derivatives with much simpler chemical structures were rationally designed and synthesized.
Compound 89 bearing a naphthylether fragment in place of the structurally complex 14-membered macrolactone moiety showed high inhibitory activity against porcine SCR, with an IC50 value of 0.047 μM. The introduction of a bromo group on the naphthalene ring (compounds 90a and 90b) further improved the activity against SCR. Compound 90a was the most potent candidate against SCR (IC50 = 12 nM, Table 10) [53].
Molecular docking of the newly synthesized compounds was performed, followed by MD simulations and MM/PBSA calculations. Computational simulations revealed that most of the compounds inside the Qo site of the bc1 complex formed a hydrogen bond with Glu271 and a π-π interaction with Phe274, while the most active compound (90a) formed an additional hydrogen bond with His161.
Molecular docking of the newly synthesized compounds was performed, followed by MD simulations and MM/PBSA calculations. Computational simulations revealed that most of the compounds inside the Qo site of the bc1 complex formed a hydrogen bond with Glu271 and a π-π interaction with Phe274, while the most active compound (90a) formed an additional hydrogen bond with His161.
Recently, the same authors prepared a new set of analogs by replacing the 14-membered macrolactone ring with an indole ring linked to the carbamate moiety by an ester or an amide bond [54]. The new analogs exhibited IC 50 values ranging from 0.70 to 1.75 µM, and compound 91 showed the highest inhibitory activity against porcine SCR (IC 50 0.70 µM).

Chromanols
In the search for non-redox effects of chromanols (vitamin E-related compounds), it was noted that chromanols share structural similarity to stigmatellin (80, Figure 17), e.g., the chroman core.
For both Toc and TQ, an incomplete methyl substitution (β, γ-, and δ-congeners) increased the inhibitory potential. This affinity was further enhanced by the introduction of keto groups into the chroman ring, thus leading to chromanones. For TMC2O (6-hydroxy-4,4,7,8-tetramethylchroman-2-one, 100), it was shown that it binds to the Qo pocket of the cyt bc1 complex and delays the electron transfer from dUQH2 to cyt c1 in the complex.

Karrikinolide
Karrikinolide (3-methyl-2H-furo[2 ,3-c]-pyran-2-one, KAR1, 102) is a natural butenolide found in the smoke of burning plant material, which promotes the seed germination of a wide range of plants [56]. The molecular simplicity, structural stability, and the novel skeleton of 102 inspired the synthesis of numerous analogs in order to study their efficacy in promoting seed germination [57].
In 2016, Chen et al. exploited this scaffold with the aim to develop new cytochrome bc1 complex inhibitors and reported the synthesis of 20 karrikinolide derivatives by introducing different functional groups at C3-position ( Figure 20) [58]. The inhibitory activity of the newly prepared compounds was tested against SCR. SCR is composed of respiratory complex II (SQR) and complex III (the bc1 complex), which are believed to form a complex II-complex III supercomplex. Out of the tested derivatives, compound 103a-d exhibited limited activity against SQR at a concentration of 10 mM (Table 11). On the other hand, their inhibitory activities against SCR and the bc1 complex The authors found by enzymatic experiments that the 6-hydroxychroman structures of Toc-related compounds and the para-benzoquinone structure of TQ-related molecules can modulate the function of the mitochondrial cyt bc1 complex.
For both Toc and TQ, an incomplete methyl substitution (β, γ-, and δ-congeners) increased the inhibitory potential. This affinity was further enhanced by the introduction of keto groups into the chroman ring, thus leading to chromanones. For TMC2O (6-hydroxy-4,4,7,8-tetramethylchroman-2-one, 100), it was shown that it binds to the Qo pocket of the cyt bc1 complex and delays the electron transfer from dUQH2 to cyt c1 in the complex.
Docking experiments were performed to study the interaction of 100 with the cyt bc1 complex, which was similar but not identical to that of stigmatellin.

Karrikinolide
Karrikinolide (3-methyl-2H-furo[2,3-c]-pyran-2-one, KAR1, 102) is a natural butenolide found in the smoke of burning plant material, which promotes the seed germination of a wide range of plants [56]. The molecular simplicity, structural stability, and the novel skeleton of 102 inspired the synthesis of numerous analogs in order to study their efficacy in promoting seed germination [57].
In 2016, Chen et al. exploited this scaffold with the aim to develop new cytochrome bc1 complex inhibitors and reported the synthesis of 20 karrikinolide derivatives by introducing different functional groups at C3-position ( Figure 20) [58]. The inhibitory activity of the newly prepared compounds was tested against SCR. SCR is composed of respiratory complex II (SQR) and complex III (the bc1 complex), which are believed to form a complex II-complex III supercomplex. Out of the tested derivatives, compound 103a-d exhibited limited activity against SQR at a concentration of 10 mM (Table 11). On the other hand, their inhibitory activities against SCR and the bc1 complex support the hypothesis that these compounds act as inhibitors of the bc1 complex. In spite of the promising results, further studies will be necessary to confirm the antifungal activity of the compounds.
Molecules 2020, 25,4582 22 of 28 support the hypothesis that these compounds act as inhibitors of the bc1 complex. In spite of the promising results, further studies will be necessary to confirm the antifungal activity of the compounds.

Picolinamide UK-2A
The natural compound picolinamide UK-2A (104, Figure 21) is a pyridine carboxamide originally isolated from fermentation broths of the actinomycete Streptomyces sp. 517-02 [59]. Pyridine carboxamides inhibit mitochondrial respiration by binding to the Qi ubiquinone site of the cytochrome bc1 complex. Remarkably, no target site-based cross-resistance with Qo inhibitors was observed for these compounds. However, to date only two molecules acting as Qi site inhibitors have been commercialized: cyazofamid and amisulbrom as oomycete-specific substances.
Picolinamide UK-2A was used as a lead compound to design new macrocyclic fungicidal molecules, and recently at Dow AgroSciences LLC the novel fungicide fenpicoxamid (105, Inatreq™, Figure 21) was developed by semi-synthetic modification of the natural compound.
The acyloxymethyl ether derivative fenpicoxamid showed a broad spectrum of activity in in vitro assays and excellent efficacy on Zymoseptoria tritici (synonym, Mycosphaerella graminicola, wheat leaf blotch), the pathogen of greatest concern for wheat production in Europe.
Recently, Owen and coworkers [60][61][62] described structure-activity relationship (SAR) studies on UK-2A, analyzing the impact of modifications of the macrocycle isobutyryl ester position, ring replacement, and modifications to the macrocycle benzyl position. Moreover, the relative activities of the new analogs were rationalized, based on a homology model constructed for the Z. tritici Qi binding site.
In particular, the isobutyryl ester of UK-2A was replaced by a series of ester groups and a set of ether groups, carbonate, and carbamate moieties. The authors reported that linkages other than esters are well tolerated at the 7-position of the macrocycle, with the only exception of the carbamate analogs. The n-butyl ether (106) was the most active analog evaluated, and compound 107 (pivaloate ester) was the most active among the ester derivatives (Table 12).

Picolinamide UK-2A
The natural compound picolinamide UK-2A (104, Figure 21) is a pyridine carboxamide originally isolated from fermentation broths of the actinomycete Streptomyces sp. 517-02 [59]. Pyridine carboxamides inhibit mitochondrial respiration by binding to the Qi ubiquinone site of the cytochrome bc1 complex. Remarkably, no target site-based cross-resistance with Qo inhibitors was observed for these compounds. However, to date only two molecules acting as Qi site inhibitors have been commercialized: cyazofamid and amisulbrom as oomycete-specific substances.
Molecules 2020, 25, 4582 23 of 28 Among the analogs featuring modifications to the macrocycle benzyl position, the cyclohexyl analog (108) was the most active derivative on mitochondrial electron transport (MET) assays (IC50 1.23 nM) and inhibited the in vitro growth of Zymoseptoria tritici (EC50 2.8 ppb) and Leptosphaeria nodorum (EC50 6.2 ppb) more strongly than UK-2A (EC50 5.3 and 11.3 ppb for Z. tritici and L. nodorum, respectively). On the contrary, the introduction of heterocyclic ring systems and polar linker functionalities on this position resulted in substantial loss of activity.   Picolinamide UK-2A was used as a lead compound to design new macrocyclic fungicidal molecules, and recently at Dow AgroSciences LLC the novel fungicide fenpicoxamid (105, Inatreq™, Figure 21) was developed by semi-synthetic modification of the natural compound.
The acyloxymethyl ether derivative fenpicoxamid showed a broad spectrum of activity in in vitro assays and excellent efficacy on Zymoseptoria tritici (synonym, Mycosphaerella graminicola, wheat leaf blotch), the pathogen of greatest concern for wheat production in Europe.
Recently, Owen and coworkers [60][61][62] described structure-activity relationship (SAR) studies on UK-2A, analyzing the impact of modifications of the macrocycle isobutyryl ester position, ring replacement, and modifications to the macrocycle benzyl position. Moreover, the relative activities of the new analogs were rationalized, based on a homology model constructed for the Z. tritici Qi binding site.
In particular, the isobutyryl ester of UK-2A was replaced by a series of ester groups and a set of ether groups, carbonate, and carbamate moieties. The authors reported that linkages other than esters are well tolerated at the 7-position of the macrocycle, with the only exception of the carbamate analogs. The n-butyl ether (106) was the most active analog evaluated, and compound 107 (pivaloate ester) was the most active among the ester derivatives (Table 12). The 3-hydroxy-4-methoxypicolinic acid moiety of UK-2A can be replaced by a variety of o-hydroxy-substituted arylcarboxylic acids, while retaining strong activity against Z. tritici and other relevant fungi.
Among the analogs featuring modifications to the macrocycle benzyl position, the cyclohexyl analog (108) was the most active derivative on mitochondrial electron transport (MET) assays (IC 50 1.23 nM) and inhibited the in vitro growth of Zymoseptoria tritici (EC 50 2.8 ppb) and Leptosphaeria nodorum (EC 50 6.2 ppb) more strongly than UK-2A (EC 50 5.3 and 11.3 ppb for Z. tritici and L. nodorum, respectively). On the contrary, the introduction of heterocyclic ring systems and polar linker functionalities on this position resulted in substantial loss of activity.

Conclusions
There is an urgent need for new compounds that specifically target pathogenic fungi due to the increasing rates of resistance to the drugs currently available on the market. The cytochrome bc1 complex is one of the most important fungicidal targets. Inhibitors of the cytochrome bc1 complex have been broadly studied, in particular for controlling fungal diseases. In this context, natural products have played a crucial role. Commercial analogs of the natural compound strobilurin are among the most successful classes of agricultural fungicides, with a dominant position on the global market.
This review provides an outline of advances in the investigation of new bc1 complex inhibitors based on natural products. Emphasis has been given to studies involved in the optimization of the natural scaffold. In fact, despite their enormous potential, natural compounds are often characterized by structural complexity, toxicity, and unfavorable bioavailability, which can limit their investigation and severely hamper their development. For this reason, this review focuses on those modifications of the natural core that have expanded the potency and stability of analogs compared to the parent compounds, opening the possibility of further development.
In this context, it should be stressed that, besides the in vitro antifungal test (percentage of inhibition of fungal growth or EC 50 values), a significant number of enzymatic tests to evaluate the cyt bc1 inhibitory activity were performed using as a model system the porcine succinate cytochrome c reductase, a mixture of respiratory complex II, and bc1 complex [10,34,50,[52][53][54]56]. Despite the general lack of data around the fungal cytochrome bc1 complex, those studies were also included in the review in order to obtain a complete overview of the SAR investigations on cytochrome bc1 complex inhibitors.
The history of strobilurins clearly demonstrates the value of a detailed structure-activity investigation in the context of a strategy focused on the variation of molecular structures to optimize the biological profile of the parent molecule. From the analysis of all the reported studies, it has emerged that the knowledge of the target binding site is often crucial to define the effect of substituent changes and predict modifications for enhanced potency, safety, and circumvention of resistance.
However, a major concern in the development of new cytochrome bc1 complex inhibitors is related to the toxicity to non-target organisms., e.g., plants and animals. In some cases, the evolutionary conservation of sequence and structure of the key functional subunits of the respiratory chain complexes is reflected in a low selectivity of electron transport inhibitors in different species [6]. In fact, it has recently been demonstrated that even strobilurins, which are allegedly non-toxic to humans, can also exert their mode of action in mammalian cells [63].
Despite the improved understanding of the existing targets and their inhibitors given by crystallographic and structural biology studies [64], the design of novel inhibitors of the bc1 complex with appropriate species selectivity still remains a very challenging goal. A few papers report efforts towards this aim. Monzote et al. [65] and Rotsaert et al. [66], found a certain degree of selectivity in the inhibition of the cyt bc1 activity of different species by selected molecules.
Nevertheless, it should be stressed that the relevance of the molecular target of a fungicide may change during the life cycle of the treated organism. As a consequence, different development stages may exhibit different sensitivities to a given molecule. The ATP energy-demanding spore germination is, compared to mycelial growth, particularly sensitive to respiration inhibitors such as strobilurins. Since, among eukaryotes, spore germination occurs almost exclusively in fungi, this also contributes, on the physiological level, to selective action against fungal pathogens [6].
It should be also considered that careful optimization of the biological profile of the antifungal compounds-including modulation of their pharmacodynamic and pharmacokinetic properties-can contribute to the reduction of mammalian toxicity.
Overall, recently reported results highlight the importance of cytochrome bc1 complex as a fungicidal target and spark a renewed interest in natural products as sources for new antifungal molecules. However, a great effort to increase species selectivity and to reduce the toxicity of fungicides is still needed and remains of primary importance for future antifungal drug discoveries.