In Silico and in Vitro Evaluation of Synthetic Chalcones against Paracoccidioides brasiliensis

Abstract

Paracoccidioidomycosis (PCM) is an infectious disease caused by dimorphic fungi of the Paracoccidioides genus and causes a series of discomforts in affected patients. This work aimed to evaluate the antifungal potential of synthetic chalcones against Paracoccidioides brasiliensis (Pb) and to determine in silico possible therapeutic targets. An in silico evaluation of a database of 21 synthesized chalcones was carried out based on pharmacokinetic parameters, enzymatic inhibition, Tanimoto similarity, and the prediction of the spectrum of activity by PASS (prediction of activity spectra of substances). The most viable chalcones from the previous evaluation were selected for the measurement of minimum inhibitory concentration (MIC) against Pb and for cytotoxicity assays pre- and post-metabolization using HEPG2 cells. After in silico evaluation, the compounds 4, 11, 12, 20, and 21 were selected to carry out the molecular docking and in vitro tests. In the docking studies, multiple hydrophobic and polar intermolecular interactions were observed, such as hydrogen bonds, with emphasis on compound 20 in the active site of thioredoxin, where it made 4 hydrogen bonds with the residues Gln43, Ala36, and Thr38. In vitro testing revealed antifungal activity, with the MICs ranging from 32 to 128 μg/mL. In cytotoxicity assays, the 5 compounds exhibited reduced IC₅₀ values (5.51–14.85 μg/mL pre-metabolization and 10.48–35.4 μg/mL post-metabolization). The compounds 4, 11, 12, 20, and 21 have shown favorable predictions of pharmacokinetic characteristics and distinct actions compared to conventional medications, as well as antifungal activity with less toxicity after metabolization, making them the best candidates for further studies.

Introduction

Paracoccidioidomycosis (PCM) is a systemic disease caused by a thermodimorphic fungus of the genus Paracoccidioides spp.,¹⁾ which includes some species that are highlighted such as Paracoccidioides brasiliensis (Pb), P. lutzii, and P. americana.²⁾ The infection is prevalent in Central and South America. Among South American countries, Brazil has the highest incidence of PCM cases.³⁾ It is estimated that the incidence of this disease in Brazil reaches 80% of cases compared to South American countries.²⁾ The disease mainly affects workers who deal with soil management, such as coffee harvesting, an activity with economic relevance largely carried out by Brazilian workers.⁴⁾ But it can also affect urban populations, where a study showed a recent PCM outbreak in Rio de Janeiro linked to extensive deforestation and soil displacement during subway construction, leading to a rise in cases.^5,6)

Patients with PCM suffer from very disabling clinical manifestations such as generalized lymph node involvement, skin lesions, generalized lung involvement, and neuronal damage.^7,8) This drastically reduces the QOL of these patients and can even increase the mortality rate, leading to social and economic losses. The treatment for PCM still remains a challenge. It is very costly to the public health system, takes a long time, and causes a series of side effects in patients, such as glomerular damage, hormonal dysfunction,⁹⁾ adrenal changes, and peripheral neuropathy.¹⁰⁾ Furthermore, there are many drug interactions, which further reduce the QOL of these people.¹¹⁾ Only a few drugs are used for PCM treatment, targeting different cellular components of the fungus, like azoles.¹²⁾ Amphotericin B can also be used to treat PCM and has been shown to exhibit fungicidal activity.¹³⁾ However, despite its efficacy, its major limitation is toxicity.¹⁴⁾ Additionally, cases of PCM resistant to amphotericin B have already been reported in the past, along with isolates displaying resistance, consequently restricting therapeutic options.^15,16) Considering these facts, the search for new substances with therapeutic capacity against Pb becomes essential.

Chalcones are substances that can be obtained both synthetically and naturally. Synthetic manipulations of chalcones or the isolation from natural sources are being investigated worldwide for the development of more potent and efficient compounds for the treatment of several diseases. Previous studies have already demonstrated a range of biological activities such as antitumor,¹⁷⁾ antimicrobial,¹⁸⁾ anti-inflammatory,¹⁹⁾ antifungal,²⁰⁾ among others.

These compounds are attractive to synthetic chemists because they are easy to prepare, have a simple structure, and have a large number of replaceable hydrogens. Their synthesis is mainly carried out by a classic organic chemistry reaction—the Claisen–Schimidt condensation—which requires simple and low-cost reagents.²¹⁾ Through this reaction, it is possible to obtain a wide variety of compounds with substituted reagents, consequently resulting in a greater variety of promising biological activities.²²⁾

With recent advances in technology, such as virtual tools, in silico studies are crucial for the screening of new substances because they reduce the cost associated with drug discovery, and they may also reduce the time it takes for a drug to reach the market. This is a modern strategy to explore and elucidate possible interactions of compounds with metabolic targets.²³⁾ These aspects, added to the fact that PCM therapy has limitations, such as toxicity, drug interactions, numerous side effects, and is also a long process, demonstrate the need for the development of new compounds.²⁴⁾ Studies have concentrated efforts on the identification of new compounds using numerous tools and targeting important proteins from Pb, with an emphasis on the metabolic pathways of this pathogen.^25,26)

Metabolic enzymes such as fumarate reductase and ubiquinol-cytochrome C reductase are part of the oxidative phosphorylation of Pb and other fungi. They are basically involved in anaerobic respiration and the maintenance of fungal membrane potential.²⁷⁾ Another essential protein for Pb survival is thioredoxin, which plays a key role in the redox balance of fungi and is important for the cell’s viability. Thioredoxin also helps repair oxidative stress, modulates protein activity, and is involved in signaling.²⁸⁾ These 3 proteins are important for the pathogen’s survival, and they can be potential targets for new drugs. Chalcones, such as licochalcone A, have already demonstrated inhibitory potential for fumarate reductase, ubiquinol-cytochrome C reductase, and thioredoxin.²⁹⁾

Having taken several aspects of PCM into consideration, such as it being a relevant disease that can negatively impact the lives of affected patients via economic and social losses, in addition to being neglected by public health authorities, and that chalcones have already been demonstrated to be effective with antifungal activity in previous studies,^20,30,31) a series of 21 chalcones were tested with the aim of evaluating the in silico and in vitro antifungal potential of synthetic chalcones and predicting possible interactions with relevant molecular targets, as well as evaluating pre- and post-metabolization cytotoxicity, assuming the hypothesis of demonstrating antifungal activity so that, with further studies, they may potentially be used as antifungal medications.

Results and Discussion

In Silico 

Prediction of Pharmacokinetic Characteristics

The results of the prediction of absorption, distribution, blood–brain barrier permeabilization, partition coefficient (Log P), and water solubility (Log S) of the compounds listed in Fig. 1 are shown in Table 1.

Fig. 1. (a) Central Structure Demonstrating the Position of Aromatic Rings A′ and B′, Carbon Numbering, and the Presence of the π Bond between α and β Carbons; (b) Structure, Number, and Name of the Studied Compounds

Table 1. Results of Predicting Pharmacokinetic Characteristics with Permeabilization by Caco-2 Cells, Bioavailability, Log P, Log S, Distribution Volume, Plasma Protein Binding, and Blood–Brain Barrier Permeability

Compound	Permeabilization by Caco-2 cells (log cm/s)a)	Log P	Log S	Plasma protein binding (%)	Distribution volume (L/kg)b)	Bioavailability (%)	Blood–brain barrier permeability
1	−4.539	3.6	4.3	94.071	1.165	30% +	+
2	−4.25	3.5	4.0	94.521	0.529	30% +	+
3	−4.228	4.2	4.7	90.88	0.634	30% +	+
4	−4.176	3.5	3.8	88.027	0.844	30% +	+
5	−4.346	4.3	5.0	90.302	1.561	30% +	+
6	−4.484	3.5	4.7	91.311	1.757	30% +	+
7	−4.51	3.5	4.7	91.695	1.116	30% +	+
8	−4.241	3.5	4.0	94.296	0.963	30% +	+
9	−4.316	4.1	5.2	91.305	1.106	30% +	+
10	−4.228	4.2	4.7	90.819	0.747	30% +	+
11	−4.28	3.4	4.4	90.568	1.375	30% +	+
12	−4.235	3.5	4.1	94.056	1.056	30% +	+
13	−4.386	4.2	5.2	91.036	1.945	30% +	+
14	−4.274	3.4	4.3	89.902	1.151	30% +	+
15	−4.283	3.4	4.4	90.608	0.917	30% +	+
16	−4.499	3.5	4.7	91.301	1.233	30% +	+
17	−4.35	4.3	5.0	90.331	1.553	30% +	+
18	−4.504	3.5	4.9	94.321	1.314	30% +	–
19	−4.304	3.7	5.0	91.31	1.652	30% +	+
20	−4.221	3.8	4.1	90.149	0.885	30% +	+
21	−4.66	3.2	3.4	91.15	0.588	20–30%	+

a) Optimal: Higher than −5.15 log units. b) Volume of distribution optimal: 0.04–20 L/kg.

All compounds showed high absorption by the gastrointestinal tract and good predicted permeabilization to Caco-2 cells, which is in accordance with their nonpolar characteristics. The compounds also showed bioavailability greater than 30%, except for chalcone 21, for which the result was between 20 and 30%. According to pharmacokinetic predictions, all compounds permeabilize the blood–brain barrier, with the exception of compound 18, which can be unfavorable due to side effects, but favorable due to a greater scope of site of action. No compound was predicted to be a substrate for P-glycoprotein. Log P values were high and Log S values were low, indicating good liposolubility of all the compounds. Other studies have shown that chalcones such as phloretin also present low water solubility.³²⁾ The study by Radni et al. showed that phloretin, a flavonoid of the dihydrogen chalcone class, has low solubility in aqueous media, and solvents such as methanol, ethanol, and dimethyl sulfoxide must be used to carry out tests.³³⁾

The permeabilization of the blood–brain barrier represents a negative characteristic due to possible side effects involving the central nervous system. Serung et al. studied the ability of phytocompounds to cross the blood–brain barrier and found that 35 out of 99 compounds, including a trans-chalcone, had the ability to cross the blood–brain barrier for a potential antiangiogenic effect.³⁴⁾ The percentages of binding to plasma proteins were high, suggesting a possible influence on half-life, administration dosage, and other pharmacokinetic factors.³⁵⁾ The prediction of tissue distribution showed reduced values, demonstrating greater retention in the bloodstream and requiring the administration of a higher dosage to effectively reach the tissues.

The inhibition of enzymes involved in drug metabolism may be responsible for drug interactions, which can affect plasma levels if administered at the same time and consequently modify the intensity of desired and adverse effects.³⁶⁾ Several compounds are predicted to inhibit CYP1A2, which is responsible for metabolizing drugs such as phenacetin, for example.³⁷⁾ All compounds were candidates to inhibit CYP2C19, the enzyme responsible for the metabolism of clopidogrel³⁸⁾ (Supplementary Table S1). The CYP2D6 enzyme metabolizes, for example, tamoxifen, a drug used to treat breast cancer, and the compounds 1, 2, 8, 12, and 20 are predicted to inhibit it. The only compound predicted to inhibit the CYP3A4 enzyme was 18, which is a positive characteristic, considering that this enzyme is responsible for the metabolism of approximately 50% of the drugs currently used.³⁹⁾ There were many predictions of inhibition of enzymes 1A2, 2C19, and 2C9, but only a few inhibited enzymes 2D6 and only 18 inhibited 3A4, which represents a favorable result. Lipinski’s rules are a set of molecular characteristics that a molecule must present to be considered a good drug candidate. All tested compounds followed Lipinski’s rule of 5; such characteristics are: molar mass <500 Da, Log P < 5, donating hydrogens <5, and hydrogen acceptor groups <5.⁴⁰⁾

Tanimoto Similarity and Prediction of Activity Spectra of Chalcones

The Tanimoto similarity was assessed by comparing the compounds with reference medicines used in the treatment of PCM and with the chalcone, licochalcone A. Licochalcone A was used because it has already demonstrated inhibition of the enzymes selected as potential targets for this work, fumarate reductase, ubiquinol-cytochrome C reductase, and thioredoxin.^29,41) The compounds showed reduced Tanimoto similarity values, ranging between 0.086 (amphotericin B) and 0.343 (trimethoprim). The best result for licochalcone A was 0.777, which was already expected due to the limited structural similarity between the molecular structure of the compounds and that of the standard medicines.

Important molecular targets of Pb metabolism were defined by consulting the literature. The fumarate reductase enzyme is a heterotetrametric complex composed of a catalytic subunit, an iron–sulfur assembly, and 2 transmembrane subunits. It converts fumarate into succinate and plays an important role in anaerobic metabolism and maintenance of fungal membrane potential.⁴²⁾ Ubiquinol-cytochrome C reductase participates in the oxidative phosphorylation process and is essential for fungal survival.⁴³⁾

The fungus in its infectious form has a defense system that consists of increased expression of the enzyme thioredoxin, which is an antioxidant enzyme against reactive nitrogen and oxygen species produced by the host’s immune system.⁴⁴⁾ Table 2 shows the results of the activity spectrum of the compounds. The chalcones with the greatest predicted inhibitory activity for the fumarate reductase enzyme included chalcone 4 (probably active [Pa] 0.541), chalcone 12 (Pa 0.532), chalcone 8 (Pa 0.500), chalcone 21 (Pa 0.459), and chalcone 20 (Pa 0.453). For inhibitory activity against the enzyme ubiquinol-cytochrome C reductase, the best results were for chalcones 14 (Pa 0.897), 16 (Pa 0.893), 18 (Pa 0.891), 12 (Pa 0.888), and 20 (Pa 0.888), while for the enzyme thioredoxin, the highest scores were for chalcone 4 (Pa 0.684), chalcone 21 (Pa 0.643), chalcone 20 (Pa 0.582), chalcone 3 (Pa 0.506), and chalcone 5 (Pa 0.506). These results agree with what was observed by Mahapatra et al.,⁴⁵⁾ Tomazela et al.,⁴⁶⁾ and Zhai et al.,⁴⁷⁾ showing that the inhibition of fungal metabolism enzymes, such as glyceraldehyde-3-phosphate dehydrogenase, lactate dehydrogenase, fumarate reductase, and others, can be observed in chalcones.

Table 2. Evaluation of the Activity Spectrum of the Compounds, Highlighting the Top 3 Results for Each Enzyme

Compounds	Antifungal activity		Cell wall inhibitor		Fumarate reductase inhibitor		Shikimate 5-dehydrogenase inhibitor		Ubiquinol-cytochrome C reductase inhibitor		Thioredoxin inhibitor
	Pa	Pi	Pa	Pi	Pa	Pi	Pa	Pi	Pa	Pi	Pa	Pi
1	0.394	0.051	0.205	0.146	0.418	0.009	0.309	0.018	0.809	0.030	0.405	0.099
2	0.375	0.056	0.213	0.127	0.437	0.007	0.366	0.011	0.788	0.037	0.479	0.065
3	0.446	0.040	0.240	0.080	0.351	0.019	0.325	0.015	0.806	0.031	0.506	0.055
4	0.361	0.059	0.241	0.079	0.541	0.004	0.318	0.016	0.822	0.028	0.684	0.013
5	0.416	0.047	0.224	0.107	0.396	0.012	0.192	0.050	0.656	0.026	0.506	0.055
6	0.389	0.052	0	0	0.321	0.026	0.202	0.046	0.869	0.085	0.254	0.225
7	0.387	0.053	0	0	0.277	0.040	0.185	0.054	0.867	0.012	0.281	0.194
8	0.377	0.055	0.220	0.115	0.500	0.004	0.389	0.009	0.791	0.012	0.442	0.081
9	0.432	0.043	0.089	0.087	0.232	0.063	0.167	0.064	0.861	0.036	0.274	0.202
10	0.420	0.046	0.270	0.044	0.372	0.015	0.341	0.013	0.779	0.014	0.469	0.069
11	0.407	0.048	0.097	0.065	0.348	0.019	0.152	0.074	0.809	0.040	0.475	0.067
12	0.377	0.055	0.235	0.088	0.532	0.004	0.366	0.011	0.888	0.007	0.479	0.065
13	0.433	0.043	N.A.	0	0.253	0.051	0	0	0.817	0.027	0.296	0.179
14	0.381	0.054	0.094	0.074	0.326	0.024	0.134	0.088	0.897	0.034	0.438	0.083
15	0.407	0.048	0.097	0.065	0.348	0.019	0.152	0.074	0.754	0.005	0.475	0.067
16	0.395	0.051	0	0	0.348	0.019	0.187	0.053	0.893	0.007	0.275	0.200
17	0.429	0.044	0.209	0.137	0.372	0.015	0.171	0.062	0.609	0.007	0.469	0.069
18	0.414	0.047	0	0	0.268	0.043	0.148	0.077	0.891	0.008	0	0
19	0.423	0.045	0.089	0.088	0.293	0.034	0	0	0.887	0.006	0.361	0.127
20	0.404	0.049	0.225	0.104	0.453	0.006	0.237	0.034	0.888	0.026	0.582	0.033
21	0.454	0.039	0.246	0.071	0.459	0.006	0.410	0.007	0.820	0.026	0.643	0.020

Pa: probably active; Pi: probably inactive.

Molecular Docking

The molecular docking studies were carried out based on the results of the prediction of the activity spectrum, where the test was carried out with the 3 compounds that presented the highest Pa for the molecular targets fumarate reductase, thioredoxin, and ubiquinol-cytochrome C reductase. The compounds 4, 11, 12, 20, and 21 were chosen for molecular docking and in vitro tests due to their good pharmacokinetic parameters and the best Pa values in the PASS (prediction of activity spectra of substances) analysis for the selected targets.

The selection of protein structures from the Protein Data Bank (PDB) for docking studies was based on literature-defined molecular targets relevant to the metabolism of Pb. Due to the absence of crystallographic structures for Pb, we used homologous structures from related species, identified through BLASTP sequence comparisons.

The BLASTP server compared the amino acid sequence of the fumarate reductase protein from Saccharomyces cerevisiae and found 40.6% similarity with the amino acid sequence of Paracoccidioides nonredundant XP_010758525.1 (uncharacterized protein PADG_02592 from Pb). The comparison of the amino acid sequence of thioredoxin reductase from Aspergillus fumigatus with the amino acid sequence of XP_010756960.1 (thioredoxin reductase from Pb) found 76.47% similarity. Comparing the amino acid sequence of the ubiquinol-cytochrome C reductase protein from Candida albicans with the amino acid sequence of YP_537104.1 (apocytochrome b mitochondria from Pb) revealed 65.1% similarity (Supplementary Table S2).

The alignments visualized with Jalview software⁴⁸⁾ (Supplementary Fig. S1) demonstrated that the residues identified as ligands of chalcones (Fig. 2) are highly identical in Pb proteins, with some exceptions: in fumarate reductase, residues Ser48, Ala114, Pro132, and Glu136 correspond to Thr45, Ser111, Phe129, and Ala133, respectively, in the Pb protein; Ser and Thr are both non-charged polar residues, apparently causing no significant impact on interactions. On the other hand, Pro (apolar and rigid side chain) and Phe (apolar and large side chain); Ala (apolar side chain) and Ser (polar side chain); Glu (negatively charged) and Ala (apolar side chain) differ in their side chain structures and consequently could have an influence on potential interactions between ligands and Pb proteins.

Fig. 2. Results of the Molecular Docking

(1) Fumarate reductase: (a) Validation by redocking with the reference ligand SIN602 with pose 11 overlay in yellow, RMSD: 1.649 Å. (b) Position of the ligands in the protein, where the ideal coordinates were obtained by the reference ligand in the PDB, X = 13.949411, Y = 23.676922, and Z = 8.591432. (c) Top-ranked poses for compound 4 (score = −6.288). (d) Compound 12 (score = −6.086). (e) Compound 8 (score = −6.278). (2) Thioredoxin: (a) Validation by redocking using the reference ligand FAD401, where the overlap of pose 6 is demonstrated in yellow, RMSD: 1.443 Å. (b) Position of the ligands in the protein, where the ideal coordinates were obtained with the reference ligand in the PDB, X = 64.406189, Y = 22.537783, Z = −47.440748. (c) Docking of the top-ranked poses of compound 4 (score = −8.871). (d) compound 21 (score = −8.917). (e) compound 20 (score = −9.067). (3) Ubiquinol-cytochrome C reductase: (a) Validation by redocking using the reference ligand U10403, where the overlap of pose 10 is demonstrated in yellow (RMSD: 1.175 Å). (b) Position of the ligands in the protein, where the ideal coordinates were obtained according to the reference ligand: X = 130.348551, Y = 143.631006, and Z = 117.691850. (c) Docking of top-ranked poses of compound 14 (score = −9.325). (d) Compound 18 (score = −10.057). (e) Compound 16 (score = −9.592).

Furthermore, all residues are located within regions of the Batch CD-Search predicted conserved domains, which reinforces that the chalcone binding regions are conserved in each pair of proteins; that is, the predicted binding sites are most likely equivalent in Pb.

Studies also show that homology models based on templates with 30–40% sequence identity can provide reliable structural predictions for docking applications. Research indicates that structures within this range can yield models with moderate accuracy, sufficient for functional and interaction studies.^49,50) Additionally, comparative modeling studies suggest that models derived from low to intermediate sequence identity still retain biologically relevant binding site conformations, making them suitable for molecular docking. For example, one study analyzed the accuracy of binding sites in protein models produced by high-throughput techniques. The results indicated that the accuracy is adequate for low- to medium-resolution docking of a significant portion of known protein–protein complexes, even when the models are based on templates with sequence identities as low as 30%.⁵¹⁾

The results of the docking studies are shown in Fig. 2. The molecular docking for the compounds 4, 12, and 18 (Figs. 2-c1–2-e1, respectively) in the fumarate reductase active site shows several hydrophobic interactions between the all tested compounds and the amino acid residues in the active site of the protein, such as π −Σ interaction with the residue Thr124. The top-ranked pose of compound 18 stands out, where 2 conventional hydrogen bonds can be observed, 1 between the carbonyl of the compound and the Arg296 residue at a distance of 2.36 Å, and the other between the hydroxyl group linked to ring A′ and the Arg126 residue at a distance of 2.15 Å.

In the results of the docking studies for the compounds 4, 21, and 20 (Figs. 2-c2–2-e2, respectively) in the active site of the thioredoxin enzyme, hydrophobic intermolecular interactions can be observed in all tested compounds, for example, of the π-alkyl type. Of particular note is compound 21, which formed 4 conventional hydrogen bonds: the 1st between the carbonyl of the compounds and the Gln43 residue at a distance of 1.69 Å, the 2nd between the hydroxyl linked to the A′ ring and the Thr38 residue, and the 3rd and 4th, respectively, between the hydroxyl of the A′ ring and the Ala36 at distances of 2.57 and 2.72 Å.

For the compounds 14, 19, and 16 (Figs. 2-c3–2-e3, respectively) in the active site of the ubiquinol-cytochrome C reductase enzyme, mostly π−Σ, π-stacked, and π-alkyl type interactions occurred between the resonance of ring A′ and the residues of Val270, Pro271, and Pi–sulfur interactions between the resonance of ring B′ and the residue of Met125. Note also a hydrogen–carbon bond interaction between a hydrogen from the Gly143 residue and the oxygen from the nitro group in the compound 14 and 2 hydrogen–carbon bond interactions between the residues of Val270 and Met139 and hydrogens linked to the methoxy group in the A′ ring of compound 16.

Intermolecular interactions were observed in the substituents of aromatic rings, such as in the top-ranked pose of chalcone 20 in the active site of the thioredoxin enzyme. A meta-substituted hydroxyl group on ring A′ acted as an electron acceptor due to the high electronegativity of oxygen, forming 3 hydrogen bonds. This aligns with findings from studies by López et al. and Sá et al., indicating that the position, type, and electron-donating or electron-accepting nature of substituents impact activity.^52,53) Docking studies suggest a good interaction between the compounds and the active site of important molecular targets in the metabolism of Pb, which may explain a possible inhibition between the compounds and fungal growth. The best interaction was the topranked pose of compound 20, which presented 4 hydrogen bonds at distances of 1.69, 2.35, 2.57, and 2.72 Å with residues Gln43, Ala36, and Thr38, respectively (Fig. 2d2).

The results of the in silico tests were analyzed to select the best chalcones for in vitro studies. The parameters of administration, distribution, metabolism and excretion, enzyme inhibition, the best Tanimoto similarities, and the prediction of activity spectra by PASS for inhibition of fumarate reductase, thioredoxin, and ubiquinol-cytochrome C reductase were observed, and thus the chalcones 4, 11, 12, 20, and 21 were selected for the next assays. Compound 21 was selected due to its good results in molecular docking, despite not presenting good bioavailability, which was confirmed by the Log P and Log S values. However, studies show that subsequent modifications can be made to this type of molecule to improve bioavailability; for example, the introduction of a chlorine atom and the glycosyl moiety into its structure can increase its bioavailability.^54,55)

In Vitro 

Antifungal Activity

The approval of new pharmaceutical substances requires trials to assess their effectiveness, with each compound undergoing testing to ensure safety and efficacy. Validating the capacity to inhibit the growth of pathological agents is crucial, with in vitro testing being a fundamental step before trials in humans.⁵⁶⁾ Minimum inhibitory concentrations (MICs) represent the lowest concentration of an antimicrobial required to inhibit the visible growth of a microorganism after overnight incubation, playing a crucial role in the search for new antimicrobials.⁵⁷⁾ The MIC results of chalcones 4, 11, 12, 20, and 21 against Pb18 in comparison with amphotericin B are presented in Table 3.

Table 3. Antifungal Activity of Chalcones 4, 11, 12, 20, and 21 against Pb18 Presented as MIC (μg mL^–1)

Compounds	Pb18
	MIC (μg mL–1)
Chalcone 4	32
Chalcone 11	64
Chalcone 12	32
Chalcone 20	64
Chalcone 21	32
Amphotericin B	0.5

The best results were found for chalcones 4, 12, and 21 with an MIC of 32 μg/mL. When exploring the relationship between structure and activity, it is observed that the presence of substituents on the aromatic ring is associated with antifungal activity. Chalcone 4, considered the base, demonstrated the lowest MIC value, corroborating previous studies.⁵²⁾ Substituents such as methoxy in the para-position of the A′ ring in chalcone 12 and hydroxyl in the meta-position of the A′ ring in chalcone 21 do not seem to significantly influence the antifungal activity, maintaining MIC values similar to the base chalcone. On the other hand, the nitro substituent in the para-position of the B′ ring in chalcone 11 exerted a negative influence on the activity against Pb, resulting in the highest MIC among the chalcones tested. The methyl group at the para-position of the A′ ring in chalcone 20 also negatively impacted the antifungal activity, as reflected by the high MIC value. Previous studies indicate that fluorinated substituents enhance the antifungal effect of chalcones. Lagu et al. highlighted chalcones A3 and B3 as the most effective compounds, with MICs of 50 and 48 μg/mL, respectively, surpassing the standard fluconazole (MIC of 52 μg/mL) against C. albicans.⁵⁸⁾

Optimization of chalcones with quinoline side groups resulted in a significant improvement in antifungal activity, especially in combination with fluconazole against resistant C. albicans. The possible mechanism of action involves inhibition of hyphal growth, induction of reactive oxygen species, damage to the mitochondrial membrane, and prevention of intracellular ATP accumulation, leading to mitochondrial dysfunction. The MIC80 ranged between 6.14 and 6.90 μg/mL.⁵⁹⁾

Cytotoxicity Assay

In the development of new drug candidates, in addition to measuring the inhibition of the growth of pathogenic agents, it is important to evaluate the possible damage that these substances can cause to human cells. One way of evaluation is through cytotoxicity studies. One tool for evaluating cytotoxicity is through assays with the HepG2 cell line.

The HepG2 cell line is derived from human hepatoblastoma cells and is commonly used to evaluate the metabolism of xenobiotic compounds and cytotoxicity against liver cells. Regarding the enzymatic expression of CYP (involved in phase 1 hepatic metabolism) and the oxidation metabolism of xenobiotics, the HepG2 lineage maintains most metabolic functions in comparison to healthy hepatocytes, which provides greater reliability for in vitro studies on the involvement of specific parameters for evaluating mitochondrial toxicity and oxidative stress in hepatotoxicity.⁶⁰⁾

In vitro methods using the S9 fraction have proven to be an important tool for evaluating the biotransformation of substances. The S9 fraction can be considered the most appropriate screening method, involving phase 1 and 2 metabolizing enzymes from hepatocytes and microsomes, in addition to being a low-cost, easy-to-operate, fast, and automatable method.^61,62) To evaluate cytotoxicity, 2 tests were carried out:, the 1st with the substances screened in silico without the S9 fraction, and the 2nd using the metabolism of the S9 fraction. The results of the cytotoxicity evaluation of the compounds, in comparison with the cyclophosphamide standard, can be seen in Table 4.

Table 4. IC₅₀ of Compounds 4, 11, 12, 20, and 21 in μg mL^–1 before and after Metabolization by Fraction S9

Compounds	IC50 without S9 fraction	IC50 with S9 fraction	Decreased cytotoxicity (x)
	−S9	+S9	−S9/+S9
4	5.51 ± 0.062*	10.48 ± 2.498*,†	1.90
11	7.975 ± 0.059*	35.4 ± 1.603*,†	4.43
12	14.85 ± 0.056*	23.12 ± 5.691*,†	1.55
20	5.61 ± 0.048*	11.16 ± 0.049*,†	1.98
21	3.672 ± 0.009*	10.77 ± 1.612*,†	2.93
Cyclophosphamide	412.7 ± 0.17	65.40 ± 0.12†	−6.31

Values are expressed as mean ± standard deviation of up to 2 different experiments performed in triplicate. −S9: No liver fraction; +S9: with liver fraction. ^*p < 0.05 (ANOVA) indicates significantly different results between cyclophosphamide (+S9) and samples. ^†p < 0.05 (ANOVA) indicates significantly different results between samples in the absence (−S9) and presence of the metabolizing system (+S9).

The chalcones, along with the standard cyclophosphamide, exhibited statistically significant IC₅₀ values after exposure to the S9 metabolizing system. Notably, metabolism by the S9 fraction indicated substantial changes in IC₅₀ values, highlighting the potential relevance of hepatic phase 1 metabolism if these compounds were administered as drugs. Among the chalcones, compounds 11 and 21 were most influenced by S9 fraction metabolism, showing 4.43 and 2.93 times higher IC₅₀ values, respectively. This suggests a pronounced influence of hepatic metabolism on these specific compounds. Interestingly, none of the compounds exhibited pre- and post-metabolization products equivalent to the standard cyclophosphamide.

The observed variation in cytotoxicity before and after S9 metabolization is potentially associated with a reduction in compound liposolubility, leading to decreased cellular absorption. This observation aligns with the metabolic process’s goal of enhancing xenobiotic water solubility, facilitating subsequent elimination from the organism. The presence of aromatic rings in the compounds favors aromatic hydroxylation, initiating a process involving π-stacked complex formation with CYP450 and subsequent α-radical complex formation. In cases of ineffective detoxification, the produced radical may contribute to DNA damage and trigger oxidative stress.⁶²⁾

Recent studies on synthetic chalcone derivatives, such as those by Xu et al., Rahimzadeh et al., and Lim et al., further underscore the compounds’ high cytotoxicity. Structural modifications, such as long-chain ether and fluorine substitution, have been linked to increased cytotoxicity, emphasizing the structure–activity relationship in chalcones.^64–66) However, the observed high hepatocellular cytotoxicity of chalcones raises concerns about their applicability as antifungals. This underscores the need for further investigations into these compounds, particularly in the context of antifungal applications.

Conclusion

The study suggests that the investigated chalcones, based on pharmacokinetic predictions, would exhibit favorable gastrointestinal absorption, adequate bioavailability, the ability to cross the blood–brain barrier, potential for significant drug interactions, and likely adverse effects if used as medications against PCM. Additionally, these chalcones would act differently from conventional PCM medications, as indicated by low Tanimoto similarity, and could efficiently interact with crucial targets in fungal metabolism, as predicted by molecular docking analyses. The in silico evaluation led to the selection of compounds 4, 11, 12, 20, and 21 for in vitro testing. These compounds displayed moderate antifungal activity, ranging from 32 to 128 μg/mL. However, all evaluated compounds showed high cytotoxicity both before and after metabolism, suggesting the need for further studies before use as antifungals.

Experimental

In Silico 

Obtaining the Chalcones and Compounds Database

The chalcones were synthesized, identified, and characterized by Borlot et al.⁶⁷⁾ and donated for the development of this work. A chalcone database was created for in silico tests using the Marvin Sketch online server (https://www.chemaxon.com/products/marvin/). Figure 1 shows the central structure of the chalcones and the molecular structures of the studied compounds.

Pharmacokinetic Prediction

The absorption, distribution, metabolism, and excretion predictions were evaluated by 2 online platforms: SwissADME (http://www.swissadme.ch/) and ADMETlab (https://admetmesh.scbdd.com/).^68,69) Information such as prediction of absorption by the gastrointestinal tract, permeabilization by Caco-2 cells, bioavailability, P-glycoprotein substrate, Log P, Log S, and the predictions about inhibition of cytochrome P450 enzymes were obtained from the SwissADME server. The predictions about permeabilization through the blood–brain barrier, binding to plasma proteins, and volume of distribution of compounds were obtained from the ADMETlab server. The default parameters were used during the assessment.

Tanimoto Similarity and Prediction of Activity Spectra of Chalcones

Tanimoto similarity was performed with the BindingDB server (https://www.bindingdb.org/rwd/bind/index.jsp).⁷⁰⁾ The Tanimoto coefficients were measured between chalcones and the medications for the treatment of PCM, as well as for licochalcone A. It is known that molecules with similar pharmacophoric groups may have equivalent biological activity, and the Tanimoto coefficient measures structural similarity from 0 to 1 in an increasing manner, with values closer to 1 representing greater similarity.⁷¹⁾ The default parameters were used during both evaluations.

To predict the activity spectrum of the compounds, we used the PASS server, a free online server that predicts biological activity by classifying the molecules analyzed as Pa or probably inactive (Pi) on a scale from 0 to 1, with the following interpretation: Pa > Pi—when Pa is greater than Pi for a specific activity, it suggests that the compound is more likely to be active than inactive for that activity; Pa < Pi—if Pa is less than Pi, it indicates that the compound is more likely to be inactive than active for the given activity; Pa ≅ Pi—when Pa and Pi are approximately equal, it reflects uncertainty in the prediction, suggesting that the compound does not strongly resemble either active or inactive compounds for that activity.⁷²⁾

The terms selected to evaluate the probability of activity were antifungal activity, cell wall biosynthesis inhibitor, fumarate reductase inhibitor, shikimate-5-dehydrogenase inhibitor, ubiquinol-cytochrome C reductase inhibitor, and thioredoxin inhibitor, which are relevant enzymes in the metabolism of Pb.⁴³⁾ The targets were selected by the PASS server using the tool “select target” and also based on their importance to Pb survival. The search was carried out by inserting the SMI (SMILES) codes created in the database. A comparison was made between the previously selected targets, verifying the probability of activity values.

Crystallographic Complexes and Molecular Docking

The molecular targets with the highest Pa were selected by PASS. The crystallographic complexes were searched in the PDB⁷³⁾ (https://www.rcsb.org/), and the reference ligands were downloaded with the coordinates ideal for redocking.

The crystallographic complex of the fumarate reductase enzyme (PDB code: 5ZYN) from the species S. cerevisiae was used due to the absence of the complex from the Pb species. For the docking studies of the thioredoxin enzyme, the crystallographic complex (PDB code: 6BPY) from the species A. fumigatus was used due to the absence of the complex from the Pb species. The crystallographic complex of the enzyme ubiquinol-cytochrome C reductase (PDB code: 7RJA), corresponding to the C. albicans enzyme, was used for the same previous reason.

The protein was previously prepared and protonated to physiological pH 7.4 by the APBS server (https://server.poissonboltzmann.org/).⁷⁴⁾ The Discovery Studio 2016 program was used to create the MMFF94S force field, determine the XYZ of the active site, and elucidate intermolecular interactions.⁷⁵⁾ The ligands were prepared using the Avogadro v.1.2.0 program, where hydrogens were added at pH 7.4, the MMFF94S molecular force field was created with the standard parameters, and energy minimization was carried out with a random search for 100 conformers. Those with the lowest energy were selected based on root mean square deviation score. Redocking and docking were carried out using the DockThor server (https://www.dockthor.lncc.br/v2/).⁷⁶⁾ The box size was left at the default value of 20 for XYZ due to the active site being well defined previously. The ideal coordinates of XYZ were extracted from the ligand coupled to the active site of the crystallographic complex obtained from the PDB, which was elucidated in the experimental process.

Amino Acid Sequence Alignment and Conservation Analysis

These analyses were used to compare the percentage of similarity and the conservation between the amino acid sequences of the molecular targets found in the PDB which were used in molecular docking, and the amino acid sequences of the molecular targets from the Pb species.

The alignment and conservation assessment of fumarate reductase (PDB code: 5ZNY), thioredoxin reductase (PDB code: 6BPY), and ubiquinone cytochrome c reductase (PDB code: 7RJA) in Pb was conducted by 1st searching for similar amino acid sequences for each in the BLASTP⁷⁷⁾ server (blast.ncbi.nlm.nih.gov) using the Paracoccidioides (taxid: 38946) nonredundant sequence as the database.

The sequence alignments between each pair of proteins were performed on the Clustal Omega⁷⁸⁾ server (https://www.ebi.ac.uk/jdispatcher/msa/clustalo), and the alignment results were submitted to Jalview software,⁷⁹⁾ version 2.11.4.1, to examine and visualize the conserved amino acids.

In addition, the alignment results were also submitted to the Batch CD-Search⁸⁰⁾ server (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi?) to predict conserved protein domains.

In Vitro Methods

Cytotoxicity Assay

Cell Lineages

Human hepatoma cells (HepG2) (ATCC® HB-8065^TM) were maintained in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum (Vitrocell, São Paulo, Brazil) and incubated at 37°C with a 5% CO₂ atmosphere until reaching a confluence of about 70–90%. HepG2 cells were trypsinized and counted in a Neubauer chamber to obtain the required concentration for the cytotoxicity assay.

Basal Cytotoxicity and after Metabolization with S9 Fraction

Aliquots (0.1 mL) of the medium containing HepG2 cells at a concentration of 3 × 10⁵ cells/mL were seeded into 96-well tissue-culture plates and incubated at 37°C under 5% CO₂ for 24 h for adherence. After this time, the medium was removed, and the cells were treated with a modified medium containing various concentrations of chalcones (800–25 μg/mL), with or without the S9 system. The system consisted of 10% S9 fraction, 1% 0.4 M MgCl₂, 1% 1.65 M KCl, 0.5% 1 M d-glucose-6-phosphate disodium, 4% 0.1 M β-nicotinamide adenine dinucleotide, 54% 0.2 M phosphate buffer, and 29.5% sterile distilled water. After incubation for 24 h at 37°C under 5% CO₂, the medium was removed, and the basal cytotoxicity and metabolism-mediated cytotoxicity were evaluated using an (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide)-tetrazolium assay. Cyclophosphamide was used as the standard. The concentrations ranged from 3.125 to 50 μg/mL for cyclophosphamide and chalcones 4, 11, 12, 20, and 21. The assay was performed in triplicate and repeated at least 2 times.^81,82)

Antifungal Assays

Fungal Strain

The experiments were carried out at the Laboratory of Pathogenic Dimorphic Fungi at the Institute of Biomedical Sciences at the University of São Paulo, São Paulo, Brazil. Isolates of Pb18 were kindly provided by the Medical Research Laboratory (Mycology Sector—LIM-53) of the Institute of Tropical Medicine, São Paulo, Brazil. The isolates were maintained in semisolid Fava’s Netto medium at 37°C (0.3% protease peptone, 1% peptone, 0.5% [w/v] meat extract, 0.5% [w/v] yeast extract, 4% glucose, 0.5% NaCl).

Inoculum Preparation

For the preparation of the inoculum, the fungal cells were transferred to BHI (brain heart infusion) broth supplemented with 4% fetal bovine serum, 4% glucose, 1% penicillin/streptomycin, and 5% Pb192 filtrate. The filtrate is a product of modified McVeigh and Morton minimal medium, filtered after growth of the avirulent strain Pb192. After 4–5 d, at 36.5°C and 150 rpm of agitation, the fungus was transferred and washed with phosphate-buffered saline (PBS) under centrifugation at 4000 rpm for 8 min. Next, the fungus was transferred to the RPMI-1640 (Sigma-Aldrich, St. Louis, MO, U.S.A.) supplemented with 2% glucose and 0.165 M morpholine propanesulfonic acid, adjusted to pH 7.01. This was incubated at 36.5°C under agitation at 150 rpm for 16 h. After this adaptive phase, the cells were centrifuged, resuspended in new supplemented RPMI, and the viable cells were counted by staining with trypan blue using a hemocytometer. The final concentration was adjusted to 4 × 10⁶ cells/mL. Finally, 100 μL of this suspension was added to the plates, resulting in a final concentration of 2 × 10⁶ cells/mL after the addition of compounds.

Broth Microdilution Method

The MIC was determined by the broth microdilution method. For this analysis, synthetic chalcones were tested at different concentrations. Using a 96-well microdilution plate, 100 μL of the serial dilutions of the compounds was added to each well (concentration range of 0.5–128 μg/mL) along with 100 μL of the inoculum. The plates were then incubated at 36.5°C under agitation at 150 rpm. After 48 h, 20 μL of 0.02% resazurin solution was added and incubated subsequently for 24 h. The lowest concentration of the tested compounds with no change in the original blue color of the reagent was determined as the MIC. Three replicates were conducted for each sample concentration. Amphotericin B was tested separately as a control with a final concentration range of 0.065 to 8 μg/mL.

Statistical Analysis

Statistical analyses were conducted using GraphPad Prism 5 (GraphPad Software, Inc., San Diego, California, U.S.A). Data normality was assessed using the Shapiro–Wilk test (p > 0.05). Results for fungal growth inhibition and cytotoxicity concentration evaluation were presented as the mean of 2 separate assays performed in triplicate ± standard deviation. The dose–response curve (concentration × absorbance) for determining the IC₅₀ was derived through linear regression analysis. Two-way ANOVA with Bonferroni post hoc test was employed for result evaluation, with a significance level set at p ≤ 0.05.

Acknowledgments

The authors thank the “Fundação de Amparo à Pesquisa do Espírito Santo” (FAPES), Vitória, ES, Brazil and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Brasília, Brazil—Finance code 001—for the financial support.

Funding

This work was supported by FAPES (Grant Number: 2022-78KWB) and CAPES (Grant Number: 2280/2022), process 88881.749045/2022-01. We would also like to acknowledge Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPES-Fundação de Amparo à Pesquisa e Inovação do Espírito Santo) (2016/08730-6) for additional support.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1)  Teixeira M. M., Theodoro R. C., Nino-Vega G., Bagagli E., Felipe M. S. S., PLoS Pathog., 10, e1004397 (2014).
• 2)  Turissini D. A., Gomez O. M., Teixeira M. M., McEwen J. G., Matute D. R., Fungal Genet. Biol., 106, 9–25 (2017).
• 3)  Martinez R., J. Fungi (Basel), 3, 1–13 (2017).
• 4)  Cordova L. A., Torres J., “Paracoccidioidomycosis,” StatPearls Publishing, 2022.
• 5)  do Valle A. C. F., Marques De Macedo P., Almeida-Paes R., Romão A. R., Lazéra M. D. S., Wanke B., Emerg. Infect. Dis., 23, 1917–1919 (2017).
• 6)  Stringuini F. A., Naback P. O., Araújo L. F., Lima R. B., Martins C. J., An Bras Dermatol, 98, (2023).
• 7)  Montalvo R., Díaz A., Montalvo J., Pomazongo M., Montalvo M., Tunque E., IDCases, 29, e01507 (2022).
• 8)  Bortolini B. dos A., Santo R. B. do E., Diniz L. M., Jarske R. D., de Castro N. T., Rev. Soc. Bras. Med. Trop., 57, e00804–2024 (2024).
• 9)  Del Carmen Martínez-Martínez M., Del Rocío Hernández-Morales M., Mancilla-Hernández E., Rev. Alerg. Méx., 67, 96–101 (2020).
• 10)  Benitez L. L., Carver P. L., Drugs, 79, 833–853 (2019).
• 11)  Salazar F., Brown G. D., J. Innate Immun., 10, 373–397 (2018).
• 12)  Taborda C. P., Buccheri R., Benard G., Duarte-Neto A. N., Nosanchuk J. D., Travassos L. R., Curr. Top. Med. Chem., 18, 1333–1348 (2018).
• 13)  Stone N. R. H., Bicanic T., Salim R., Hope W., Drugs, 76, 485–500 (2016).
• 14)  Boswell G. W., Bekersky I., Buell D., Hiles R., Walsh T. J., Antimicrob. Agents Chemother., 42, 263–268 (1998).
• 15)  Dietze R., Fowler V. G., Steiner T. S., Pecanha P. M., Corey G. R., Am. J. Trop. Med. Hyg., 60, 837–839 (1999).
• 16)  McGinnis M. R., Pasarell L., Sutton D. A., Fothergill A. W., Cooper C. R. Jr., Rinaldi M. G. Antimicrob. Agents Chemother., 41, 1832–1834 (1997).
• 17)  Wang G., Peng Z., Li Y., Chem. Pharm. Bull., 67, 725–728 (2019).
• 18)  Okolo E. N., Ugwu D. I., Ezema B. E., Ndefo J. C., Eze F. U., Ezema C. G., Ezugwu J. A., Ujam O. T., Sci. Rep., 11, 21781 (2021).
• 19)  Guazelli C. F. S., Fattori V., Ferraz C. R., Borghi S. M., Casagrande R., Baracat M. M., Verri W. A. Jr, Chem. Biol. Interact., 333, 109315 (2021).
• 20)  Thebti A., Meddeb A., Ben Salem I., Bakary C., Ayari S., Rezgui F., Essafi-Benkhadir K., Boudabous A., Ouzari H. I., Antibiotics, 12, 225 (2023).
• 21)  Mishra V. K., Mishra M., Kashaw V., Kashaw S. K., Bioorg. Med. Chem., 25, 1949–1962 (2017).
• 22)  Mahapatra D. K., Bharti S. K., Asati V., Curr. Top. Med. Chem., 17, 3146–3169 (2017).
• 23)  Vijayakumar B. G., Ramesh D., Joji A., Eur. J. Pharmacol., 886, 173448–173458 (2020).
• 24)  Silva L. D. C., de Oliveira A. A., de Souza D. R., Barbosa K. L. B. E. Silva K. S. F., Carvalho Júnior M. A. B., Rocha O. B., Lima R. M., Santos T. G., Soares C. M. de A., Pereira M., J. Fungi (Basel), 6, 116–143 (2020).
• 25)  Rodrigues-Vendramini F. A. V., Faria D. R., Arita G. S., Capoci I. R. G., Sakita K. M., Caparroz-Assef S. M., Becker T. C. A., Bonfim-Mendonça P. de S., Felipe M. S., Svidzinski T. I. E., Maigret B., Kioshima É. S., PLoS Negl. Trop. Dis., 13, e0007441 (2019).
• 26)  do Carmo Silva L., Tauhata S. B. F., Baeza L. C., de Oliveira C. M. A., Kato L., Borges C. L., de Almeida Soares C. M., Pereira M., Antimicrob. Agents Chemother., 62, 1–13 (2018).
• 27)  Marcos C. M., de Oliveira H. C., da Silva J. de F., Assato P. A., Fusco-Almeida A. M., Mendes-Giannini M. J. S., Front. Microbiol., 5, 1–10 (2014).
• 28)  Piaz F., Braca A., Belisario M., De Tommasi N., Curr. Med. Chem., 17, 479–494 (2010).
• 29)  Chen M., Zhai L., Christensen S. B., Theander T. G., Kharazmi A., Antimicrob. Agents Chemother., 45, 2023–2029 (2001).
• 30)  Gupta D., Jain D. K., J. Adv. Pharm. Technol. Res., 6, 114–117 (2015).
• 31)  Vásquez-Martínez Y. A., Osorio M. E., San Martín D. A., Carvajal M. A., Vergara A. P., Sanchez E., Raimondi M., Zacchino S. A., Mascayano C., Torrent C., Cabezas F., Mejias S., Montoya M., Martín M. C. S. J. Braz. Chem. Soc., 30, 286–304 (2019).
• 32)  Zhao Y., Fan Y., Wang M., Wang J., Cheng J., Zou J., Zhang X., Shi Y., Guo D., Biomed. Pharmacother., 132, 1–9 (2020).
• 33)  Deshpande R. D., Shah D. S., Gurram S., Jha D. K., Batabyal P., Amin P. D., Sathaye S., Int. J. Pharm., 645, 123386 (2023).
• 34)  Senrung A., Tripathi T., Aggarwal N., Janjua D., Chhokar A., Yadav J., Chaudhary A., Thakur K., Singh T., Bharti A. C., Cardiovasc. Hematol. Agents Med. Chem., 22, 187–211 (2024).
• 35)  Di L., Expert Opin. Drug Discov., 12, 1453–1465 (2021).
• 36)  Lynch T., Price A., Am. Fam. Physician, 76, 391–396 (2007).
• 37)  Guo J., Zhu X., Badawy S., Ihsan A., Liu Z., Xie C., Wang X., Curr. Drug Metab., 22, 40–49 (2021).
• 38)  Wang Y., Meng X., Wang A., et al., N. Engl. J. Med., 385, 2520–2530 (2021).
• 39)  Zhou S. F., Curr. Drug. Metab., 9, 310–322 (2008).
• 40)  Roskoski R., Pharmacol. Res., 144, 19–50 (2019).
• 41)  Parente-Rocha J. A., Parente A. F. A., Baeza L. C., Bonfim S. M. R. C., Hernandez O., McEwen J. G., Bailão A. M., Taborda C. P., Borges C. L., De Almeida Soares C. M., PLOS ONE, 10, e0137619 (2015).
• 42)  Schwarz M. G. A., Antunes D., Corrêa P. R., da Silva-Gonçalves A. J., Malaga W., Caffarena E. R., Guilhot C., Mendonça-Lima L., Front. Microbiol., 11, 624121 (2021).
• 43)  Arraes F. B. M., Benoliel B., Burtet R. T., Costa P. L. N., Galdino A. S., Lima L. H. A., Marinho-Silva C., Oliveira-Pereira L., Pfrimer P., Procópio-Silva L., Reis V. C. B., Felipe M. S. S., Genet. Mol. Res., 4, 290–308 (2005).
• 44)  González A., Virulence, 8, 1108–1110 (2017).
• 45)  Mahapatra D. K., Bharti S. K., Asati V., Eur. J. Med. Chem., 101, 496–524 (2015).
• 46)  Tomazela D. M., Pupo M. T., Passador E. A., da Silva M. F., Vieira P. C., Fernandes J. B., Rodrigues Fo E., Oliva G., Pirani J. R., Phytochemistry, 55, 643–651 (2000).
• 47)  Zhai L., Chen M., Blom J., Theander T. G., Christensen S. B., Kharazmi A., J. Antimicrob. Chemother, 43, 793–803 (1999).
• 48)  Waterhouse A. M., Procter J. B., Martin D. M. A Clamp M., Barton G. J., Bioinformatics, 25, 1189–1191 (2009).
• 49)  Eswar N., Sali A., Comprehensive, Med. Chem., II, 215 (2007).
• 50)  Jaiteh M., Rodríguez-Espigares I., Selent J., Carlsson J., PLOS Comput. Biol., 16, e1007680 (2020).
• 51)  Kundrotas P. J., Vakser I. A., PLOS Comput. Biol., 6, e1000727 (2010).
• 52)  López S. N., Castelli M. V., Zacchino S. A., Domínguez J. N., Lobo G., Charris-Charris J., Cortés J. C. G., Ribas J. C., Devia C., Rodríguez A. M., Enriz R. D., Bioorg. Med. Chem., 9, 1999–2013 (2001).
• 53)  De Sá N. P., Cisalpino P. S., De Carvalho Tavares L., Espíndola L., Pizzolatti M. G., Santos P. C., De Paula T. P., Rosa C. A., De Souza D. da G., Santos D. A., Johann S., J. Antimicrob. Chemother., 70, 1–13 (2015).
• 54)  Krawczyk-Łebek A., Żarowska B., Dymarska M., Janeczko T., Kostrzewa-Susłow E., Sci. Rep., 14, 15050 (2024).
• 55)  Naumann K., J. Prakt. Chem., 341, 417–435 (1999).
• 56)  Torres-Saavedra P. A., Winter K. A., Int. J. Radiat. Oncol. Biol. Phys., 112, 22–29 (2022).
• 57)  Andrews J. M., J. Antimicrob. Chemother., 48, 5–16 (2001).
• 58)  Lagu S. B., Yejella R. P., Bhandare R. R., Shaik A. B., Pharmaceuticals (Basel), 13, 375 (2020).
• 59)  Sun A., Chai N., Zhu X., Li Y., Wang R., Zhang Y., Mao Z., Eur. J. Med. Chem., 260, 115782 (2023).
• 60)  Miranda C. A., Beretta E. M., Ferreira L. A., Silva E. S., Coimbra B. Z., Pereira P. T., Miranda R. G., Dorta D. J., Rodrigues F. T. V., Mingatto F. E., Toxicol. Rep., 10, 32–39 (2022).
• 61)  Kitagawa R. R., Vilega W., Varanda E. A., Raddi M. S. G., Bras. J. Farmacogn., 25, 395–400 (2015).
• 62)  Richardson S. J., Bai A., Kulkarni A. A., Moghaddam M. F., Drug Metab. Lett., 10, 83–90(2016).
• 63)  Martemucci G., Costagliola C., Mariano M., D’andrea L., Napolitano P., D’Alessandro A. G., Oxygen (Basel), 2, 48–78 (2022).
• 64)  Xu X., Wang J., Yan B., Leng Y., Chen Z., Pan G., Xiang C., Teng Y., Bioorg. Med. Chem. Lett., 85, 129239 (2023).
• 65)  Rahimzadeh O. S., Mirzaei S., Reza J. N. M., Hadizadeh F., Eisvand F., Mosaffa F., Ghodsi R., Bioorg. Chem., 112, 5–16 (2021).
• 66)  Lim Y. H., Oo C. W., Koh R. Y., Voon G. L., Yew M. Y., Yam M. F., Loh Y. C., Drug Dev. Res., 81, 994–1003 (2020).
• 67)  Pereira de Oliveira Borlot J. R., Schlittler dos Santos L., Schwarzt Sampaio G. J., Santos Borges A., Rodrigues R. P., de Cássia Ribeiro Gonçalves R., Bezerra dos Santos R., Kitagawa R. R., Chem. Biodivers., 20, e202301066 (2023).
• 68)  Daina A., Michielin O., Zoete V., Sci. Rep., 7, 42717 (2017).
• 69)  Dong J., Wang N. N., Yao Z. J., Zhang L., Cheng Y., Ouyang D., Lu A. P., Cao D. S., J. Cheminform., 10, 104904 (2018).
• 70)  Chen X., Lin Y., Liu M., Gilson M. K., Bioinformatics, 18, 130–139 (2002).
• 71)  Altalib M. K., Salim N., Molecules, 26, 6669 (2021).
• 72)  Filimonov D. A., Lagunin A. A., Gloriozova T. A., Rudik A. V., Druzhilovskii D. S., Pogodin P. V., Poroikov V. V., Chem. Heterocycl. Compod., 50, 444–457 (2014).
• 73)  Bernstein F. C., Koetzle T. F., Williams G. J. B., Meyer E. F. Jr., Brice M. D., Rodgers J. R., Kennard O., Shimanouchi T., Tasumi M., Arch. Biochem. Biophys., 185, 584–591 (1978).
• 74)  Dolinsky T. J., Nielsen J. E., McCammon J. A., Baker N. A., Nucleic Acids Res., 32 (Web Server), W665–W667 (2004).
• 75)  Biovia D. S., Discovery Studio Modeling Environment. Dassault Syst. Release, San Diego, 4 (2016).
• 76)  Guedes I. A., Barreto A. M. S., Marinho D., Krempser E., Kuenemann M. A., Sperandio O., Dardenne L. E., Miteva M. A., Sci. Rep., 11, 3198 (2021).
• 77)  Altschul S. F., Gish W., Miller W., Myers E. W., Lipman D. J., J. Mol. Biol., 215, 403–410 (1990).
• 78)  Sievers F., Wilm A., Dineen D., Gibson T. J., Karplus K., Li W., Higgins D. G., Molecular Systems Biology, 7, 539 (2011).
• 79)  Waterhouse A. M., Procter J. B., Martin D. M. A., Clamp M., Barton G. J., Bioinformatics, 25, 1189–1191 (2009).
• 80)  Marchler-Bauer A., Lu S., Anderson J. B., et al., Nucleic Acids Res., 39 (Database), D225–D229 (2011).
• 81)  Mosmann T., J. Immunol. Methods, 65, 55–63 (1983).
• 82)  Gonçalves R. C. R., Kitagawa R. Z., Aparecida E. V., Gonçalves S. R. M., Leite C. A., Sponchiado S. R. P., Pharm. Biol., 54 1014–1021 (2016).