Experimental
ReagentsAll reagents were purchased and used without further purification. 1H-NMR and 13C-NMR spectra were recorded on a Bruker Avance 200 MHz spectrometer using CDCl3 or dimethyl sulfoxide (DMSO)-d6 as solvent. Chemical shifts were given in ppm (δ scale) and coupling constants (J) were given in hertz (Hz). Infrared (IR) spectra were recorded on a Shimadzu IRPrestige-21 FTIR spectrometer using KBr pellets. High-resolution (HR)-MS were obtained on a Bruker micrOTOF II mass spectrometer using electrospray ionization (ESI). Melting points were measured on a Shimadzu DSC-60 thermal analyzer at a heating rate of 10°C/min, room temperature to 200°C under a nitrogen flow rate of 50 mL/min and using an aluminum standard. Analytical TLC was performed on precoated silica gel plates (aluminum sheets 60 F254, Merck) using ethyl acetate–hexane (1 : 5 v/v) as the eluent.
Preparation of Thioureas 3a–zMethod ATo a solution of isothiocyanate 1 (1.0 mmol) in dichloromethane (10 mL) was added the corresponding amine 2 (1.2 mmol), and the mixture was stirred at room temperature until isothiocyanate was consumed (TLC). After that, the organic phase was washed with 5% aqueous hydrochloric acid (3×10 mL), dried with anhydrous sodium sulfate and evaporated in a rotary evaporator to afford the pure thiourea 3, which did not require any further purification.
Method BTo a solution of isothiocyanate 1 (1.0 mmol) in tert-butanol (10 mL) was added the corresponding amine 2 (1.2 mmol), and the mixture was refluxed until isothiocyanate was consumed (TLC). Then, the solvent was evaporated, 10 mL of CH2Cl2 was added and the same work up used in method A was followed.
The following shows the characterization data of thiourea derivatives, except for thioureas 3a, 3c–d, 3f, 3j–n, 3p, 3r–w and 3z, since these data have been previously reported.8)
N-Isopropyl-N′-phenylthiourea (3b)19)Pale yellow solid; mp 99–103°C; IR (KBr): 3272, 3246, 2970, 1545, 1501, 1396, 1238, 778, 693 cm−1; 1H-NMR (CDCl3) δ: 8.25 (br s, 1H), 7.46–7.36 (m, 2H), 7.32–7.15 (m, 3H), 5.85 (br s, 1H), 4.72–4.42 (m, 1H), 1.19 (d, J=6.5 Hz, 6H); 13C-NMR (CDCl3) δ: 179.5, 136.6, 130.5, 127.3, 125.3, 47.7, 22.7; HR-MS-ESI: m/z [M+Na]+ Calcd for C10H14N2S: 217.0770. Found: 217.0770.
N-N-Bis(2-hydroxyethyl)-N′-phenylthiourea (3e)Yellow oil; IR (KBr): 3278, 2935, 2880, 1643, 1598, 1554, 1536, 1359, 1254, 1209, 1070, 1041, 758, 694 cm−1; 1H-NMR (CDCl3) δ: 9.58 (br s, 1H), 7.51–6.99 (m, 5H), 3.87 (s, 8H); 13C-NMR (50 MHz, CDCl3) δ: 184.71, 140.20, 128.68, 125.10, 124.23, 61.38, 54.07; HR-MS-ESI: m/z [M+Na]+ Calcd for C11H16N2O2S: 263.0825. Found: 263.0827.
N-N-Diphenylthiourea (3g)20)White solid; mp 153–156°C (Lit.20) 153–154°C); IR (KBr): 3207, 3037, 3005, 1600, 1549, 1495, 1450, 1348, 758, 698, 631 cm−1; 1H-NMR (CDCl3) δ: 8.13 (br s, 2H), 7.45–7.21 (m, 10H); 13C-NMR (50 MHz, CDCl3) δ: 179.7, 137.1, 129.5, 126.9, 125.2; HR-MS-ESI: m/z [M+Na]+ Calcd for C13H12N2S: 251.0613. Found: 251.0621.
N-(2,5-Dimethoxyphenyl)-N′-phenylthiourea (3h)Brown solid; mp 134–136°C; IR (KBr): 3347, 3164, 1593, 1545, 1506, 1374, 1282, 1048, 1023, 755, 726 cm−1; 1H-NMR (CDCl3) δ: 8.29–8.22 (m, 2H), 8.07 (d, J=2.8 Hz, 1H), 7.48–7.24 (m, 5H), 6.78 (d, J=8.9 Hz, 1H), 6.65 (dd, J=8.9 Hz and J=2.8 Hz, 1H), 3.78 (s, 3H), 3.72 (s, 3H); 13C-NMR (50 MHz, CDCl3) δ: 178.75, 153.80, 145.07, 137.08, 130.05, 128.10, 127.40, 125.47, 112.14, 111.11, 109.71, 56.75, 56.17; HR-MS-ESI: m/z [M+Na]+ Calcd for C15H16N2O2S: 311.0825. Found: 311.0828.
N-(2,4-Dimethoxyphenyl)-N′-phenylthiourea (3i)Gray solid; mp 125–127°C; IR (KBr): 3376, 3162, 3000, 1548, 1517, 1313, 1285, 1247, 1212, 1161, 1126, 1040, 506 cm−1; 1H-NMR (CDCl3) δ: 8.08 (br s, 1H), 7.85 (br s, 1H), 7.74 (d, J=9.0 Hz, 1H); 7.43–7.35 (m, 4H), 7.31–7.20 (m, 1H), 6.55–6.43 (m, 2H); 13C-NMR (50 MHz, CDCl3) δ: 179.8, 159.3, 153.7, 137.5, 129.6, 127.0, 126.8, 125.2, 119.7, 104.3, 99.4, 56.0, 55.7; HR-MS-ESI: m/z [M+Na]+ Calcd for C15H16N2O2S: 311.0825. Found: 311.0814.
N-Benzyl-N′-phenylthiourea (3o)21)Brown solid; mp 154–156°C (Lit.21) 153–154°C); IR (KBr): 3366, 3152, 1545, 1510, 1297, 1247, 973, 744, 695 cm−1; 1H-NMR (CDCl3) δ: 8.25 (br s, 1H), 7.47–7.18 (m, 10H), 6.33 (br s, 1H), 4.88 (d, J=5.5 Hz, 2H); 13C-NMR (CDCl3) δ: 181.0, 137.4, 136.2, 130.3, 128.9, 127.8, 127.7, 127.4, 125.4, 49.5; HR-MS-ESI: m/z [M+Na]+ Calcd for C14H14N2S: 265.0770. Found: 265.0765.
N-Benzyl-N′-N′-diphenylthiourea (3q)Brown solid; mp 176–179°C; IR (KBr): 3379, 1587, 1507, 1494, 1340, 746, 706, 692, 626 cm−1; 1H-NMR (CDCl3) δ: 7.60–7.25 (m, 15H), 6.05 (br s, 1H), 4.90 (d, J=5.3 Hz, 2H); 13C-NMR (CDCl3) δ: 183.4, 144.1, 137.7, 129.9 (2C), 128.8, 128.5 (2C), 127.7, 50.0; HR-MS-ESI: m/z [M+Na]+ Calcd for C20H18N2S: 341.1083. Found: 341.1076.
N-Cyclohexyl-N′-(3,4,5-trimethoxyphenyl)thiourea (3x)White solid; mp 139–141°C; IR (KBr): 3348, 2940, 2848, 1603, 1534, 1508, 1232, 1125, 1008, 987 cm−1; 1H-NMR (CDCl3) δ: 8.16 (br s, 1H), 6.43 (s, 2H), 5.99 (br d, J=8.3 Hz, 1H), 4.35–4.12 (m, 1H), 3.85–3.79 (m, 9H), 2.07–0.98 (m, 10H); 13C-NMR (CDCl3) δ: 179.0, 154.0, 136.8, 131.7, 102.6, 60.8, 56.1, 53.8, 32.5, 25.3, 24.6; HR-MS-ESI: m/z [M+Na]+ Calcd for C16H24O3N2S: 347.1400. Found: 347.1390.
N-N-Bis(3,4,5-trimethoxyphenyl)thiourea (3y)Pale brown solid; mp 197–199°C; IR (KBr): 3283, 3161, 2959, 2933, 1603, 1544, 1501, 1455, 1340, 1233, 1135, 1004 cm−1; 1H-NMR (CDCl3) δ: 7.97 (br s, 2H), 6.65 (s, 4H), 3.84 (s, 18H); 13C-NMR (CDCl3) δ: 179.9, 153.9, 137.2, 132.8, 103.2, 61.1, 56.5; HR-MS-ESI: m/z [M+Na]+ Calcd for C19H24O6N2S: 431.1247. Found: 431.1255.
Ethics StatementAll animal experiments were performed in strict accordance with the Brazilian animal protection law (Lei Arouca number 11.794/08) of the National Council for the Control of Animal Experimentation (CONCEA, Brazil). The protocol was approved by the Committee for Animal Use Ethics of the Universidade Federal do Rio de Janeiro (Permit Number: 128/15).
Anti-promastigote ActivityLeishmania amazonensis (WHOM/BR/75/Josefa) promastigotes were cultured at 26°C in Schneider insect medium (Sigma-Aldrich, St. Louis, MO, U.S.A.), 10% fetal calf serum (FCS, Gibco-BRL, MD, U.S.A.) and 40 µg/mL of gentamycin (Schering-Plough, Rio de Janeiro, Brazil).
The leishmanicidal properties of thioureas 3a–z were evaluated by measuring promastigote viability through the dehydrogenases activity using the XTT method (2,3 bis-[2-methoxy-4-nitro-5-sulphophenyl]-2H-tetrazolium-5-carboxanilida; Sigma-Aldrich) as described.22) Briefly, stationary-phase promastigotes were treated with the thioureas for 48 h at 26°C, and then incubated with XTT activated with phenazine methosulfate (Sigma-Aldrich) for 3 h. The reaction product was read at 450 nm. Meglumine antimoniate (MA) was used as reference compound.
Anti-amastigote ActivityMice peritoneal macrophages obtained after stimulation with 3% thioglycolate for 3 d were harvested in RPMI 1640 medium (LGC Biotec, São Paulo, Brazil) and cultured inside 24-well plates for 2 h adherence at 35°C, 5% CO2. Non-adherent cells were removed, and macrophages were incubated overnight in RPMI-10% FCS as above. Adhered macrophages were infected with Leishmania amazonensis promastigotes (stationary growth phase) at a 10 : 1 parasite/macrophage ratio during 1 h at 35°C, 5% CO2. Free parasites were washed out with 0.01 M phosphate buffered saline (PBS), and the cultures maintained for 24 h at 35°C, 5% CO2. Infected macrophages cultures were treated with different concentrations of the thiourea derivatives during 24 h at the same culture conditions mentioned above. Cultures were washed with PBS and incubated with 0.01% sodium dodecyl sulfate for 10 min followed by 1 mL of Schneider’s medium supplemented with 10% FCS, and maintained at 26°C for 2 d. The viable intracellular L. amazonensis amastigotes, that survived after treatment, were measured after promastigote transformation, using Alamar Blue (Invitrogen), as previously described.23) After 4 h of Alamar Blue incubation, the fluorescence was read at 540/610 nm excitation/emission wavelenghts in a SpectraMax Paradigm (Molecular Devices). MA was used as reference compound.
Cytotoxicity for Host MacrophagesThioglycolate-stimulated mice peritoneal macrophages adhered to 96-well plates obtained as above were treated with different concentrations of thioureas for 24 h, and cell viability was determined using 1 mg/mL XTT (2,3-Bis[2-Methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxinilide inner salt, Sigma-Aldrich), 200 µM Phenazine Methosulfate (PMS). After 3 h incubation, the reaction product was read at 450 nm. The results are expressed in percentage of viable cells compared to untreated control.24) MA was used as reference compound.
Nitric Oxide (NO) ProductionThioglycolate-stimulated mice peritoneal macrophages were obtained as above (106 cells/well in 24-well plate) and activated with 1 µg/mL lipopolysaccharide (LPS) and 100 ng/mL interferon-γ (IFN-γ) (Sigma-Aldrich) or left untreated. After incubating the cells for 24 h at 37°C, 5% CO2, they were treated with 100 µM of 3k, 3l, 3p, 3q or 3v. The nitrite concentrations in the culture supernatants were determined using the Griess method. The reaction was read at 540 nm, and the concentration of nitrite determined using a standard curve of sodium nitrite. The results are expressed as micromolar concentrations of nitrite.25)
Cell Cycle AnalysisPromastigotes were incubated in Schneider’s complete medium, with or without 100 µM 3k, 3l, 3p, 3q or 3v for 48 h. The cells were then washed with PBS, fixed in 70% (v/v) ice-cold methanol/PBS for 1 h at 4°C, washed again once with PBS and then incubated in PBS supplemented with 10 µg/mL propidium iodide (PI) and 20 µg/mL ribonuclease (RNase) at 37°C for 45 min, according to Ambit et al.26) For each sample, 10000 events were collected on a BD FACScalibur (Becton Dickinson, San Jose, CA, U.S.A.) and analyzed using CellQuest software.22)
Measurement of Mitochondrial Membrane Potential (∆Ψm)Promastigotes were treated or not with 100 µM 3k, 3l, 3p, 3q or 3v for 48 h and then incubated with JC-1 solution (5 µg/mL, Sigma-Aldrich) for 20 min at 37°C according to the manufacturer’s instructions. ∆Ψm was measured in 96-well opaque plates using 490/530 nm excitation/emission (JC-1 monomers) and 525/590 nm excitation/emission (J-aggregates) in a SpectraMax Paradigm (Molecular Devices), as described.27)
Molecular ModelingStructure–Activity Relationship (SAR) StudiesAll molecular computations were performed using SPARTAN’10 (Wavefunction Inc., Irvine. CA, U.S.A.). Briefly, the structures were optimized to a local minimum and the equilibrium geometry was obtained in a vacuum using RM1 semi empirical methods. Subsequently the structures were submitted to a single-point energy Density Functional Theory (DFT) calculation using B3LYP/6-311G* method to calculate some stereoelectronic properties and to perform SAR studies. Thus, we calculated the descriptors: highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energies, molecular weight (MW), molecular surface area, polar surface area (PSA), dipole moment and lipophilicity (c Log P).28)
In Silico Absorption, Distribution, Metabolism, Excretion, Toxicity (ADMET) StudiesThe structures of the thiourea derivatives were screened for theoretical oral bioavailability presence of Pan assay interfering compounds (PAINS) and toxicophores with FAF-DRUGS3 server using the Lead-like physicochemical filter available on the server (http://fafdrugs3.mti.univ-paris-diderot.fr/).29) Besides, the toxicity risks of the compounds were analyzed using Osiris Property Explorer (http://www.organicchemistry.org/prog/peo/). The toxicity risks are evaluated as low medium or high risk based on the Registry of Toxic Effects of Chemical Substances (RTECS) database.
Results
ChemistryA series of 26 N,N′-disubstituted thioureas was synthesized according to Chart 1. Thioureas 3a–z were prepared through the reaction of the isothiocyanate 1 and an excess of the appropriate amine 2, in a dichloromethane or tert-butanol solution (Table 1). The products were obtained in good to excellent yields (70–98%) and no further purification was required after isolation from the crude reaction mixture. The proposed structures of all prepared thioureas were confirmed by 1H-NMR, 13C-NMR, FT-IR and HR-MS.
Chart 1. Synthesis of Thiourea Derivatives 3a–z
Initially, the antileishmanial activity of the thioureas 3a–z was determined in vitro against Leishmania amazonensis promastigotes using the XTT assay. Thus, stationary-phase promastigotes were treated with 100 µM of thioureas during 48 h and parasite viability evaluated. By this initial screening, thioureas 3e, 3i, 3k, 3l, 3p, 3q, 3v, 3x, 3y and 3z were active (data not shown), and then selected for IC50 determination. As shown in Table 2, nine thioureas presented IC50 values against promastigotes under 200 µM, ranging between 21.48 and 189.10 µM.
Table 2.
In vitro Antileishmanial Activity and Cytotoxicity Results for Thiourea Derivatives
| Thiourea | Macrophages 50% cytotoxic concentration (µM) | Promastigotes IC50 (µM) | Amastigotes IC50 (µM) |
|---|
| 3e | 49.22 | 26.04±1.61 | ND |
| 3i | >200 | 127.50±12.81 | ND |
| 3k | >200 | 80.09±13.87 | 150±20 |
| 3l | >200 | 21.94±1.19 | 110±12.14 |
| 3p | >200 | 46.08±8.791 | 145±4.70 |
| 3q | >200 | 21.48±0.001 | 81.4±10.54 |
| 3v | >200 | 54.14±6.769 | 70.0±6.15 |
| 3x | >200 | 139.60±10.84 | ND |
| 3y | >200 | >200 | ND |
| 3z | >200 | 189.10±7.9 | ND |
| MA | 174±38.9 | ND | 212.30±3.7 |
ND=Not Determined. MA=Meglumine antimoniate.
Next, the safety of the ten selected thioureas on host macrophages was accessed, evaluating the dehydrogenases activity using the XTT method. Our results showed that macrophage treatment with all thioureas only affected dehydrogenase activities at higher concentrations (CC50>200 µM), except for thiourea 3e, which CC50 was 49.22 µM, and thus it was excluded from further tests (Table 2).
Following we assessed the activity of the most promising thiourea derivatives (3k, 3l, 3p, 3q and 3v) against L. amazonensis amastigotes, which is the intracellular stage of the parasite responsible for the clinical manifestations of leishmaniasis. Hence, Leishmania-infected macrophages were treated with thiourea derivatives during 24 h resulting in IC50 values ranging from 70 to 150 µM, indicating the relevance of this class of compounds in the search for new leishmanicidal drugs. Among these compounds the lowest IC50 were obtained for 3q and 3v, 81.4 and 70.0 µM, respectively, presenting greater effectiveness than MA (IC50=212.3 µM), a drug used in clinical practice (Table 2).
Thiourea derivatives tested against amastigotes were also investigated by its ability to modulate NO production, and effect on the parasite cell cycle and mitochondrial membrane potential (ΔΨm). NO is an important mediator of Leishmania death into the macrophage phagolysosomes, therefore we investigated the capacity of thiourea derivatives to modulate NO production in infected macrophages stimulated or not with IFN-γ+LPS. Results showed that leishmanicidal activity of derivatives is independent of NO production, since there was a decrease of 81, 69, 100 and 79% in NO production on unstimulated infected macrophages treated with 100 µM of 3k, 3l, 3q and 3v, respectively. Among all thiourea compounds tested, only 3p, was able to modulate NO production in IFN-γ+LPS-stimulated infected macrophages (Fig. 2).
To further investigate the leishmanicidal effect of thiourea derivatives we analyzed its effect on the parasite cell cycle. Our results demonstrated that promastigotes treated with 100 µM of 3k, 3l, 3q and 3v during 48 h, increased 4.40, 3.59, 1.67 and 6.09-fold respectively, cells in Sub-G0/G1 phase. Cells in S phase were decreased 3.00, 1.34 and 1.56-fold after treatment with 3k, 3l and 3v, respectively. Derivatives 3k and 3p decreased 1.13- and 1.24-fold, respectively, cells in G2/M phase (Table 3). In a previous work published by our group30) amphotericin B (0.1 µM) did not affect the cell cycle of the parasite.
Table 3. Effect of Thioureas on
Leishmania amazonensis Promastigotes Cell Cycle
| Treatment | % Cells in different stages of the cell cycle# |
|---|
| Sub-G0/G1 | G0/G1 | S | G2/M |
|---|
| Control | 0.866±0.1281 | 35.38±2.073 | 12.72 ±0.8517 | 28.77±1.092 |
| 3k | 3.810±0.3483** | 35.84±0.9020 | 4.23±0.3984*** | 25.42±0.4768* |
| 3l | 3.110±0.900* | 36.06±1.925 | 9.47±1.190** | 27.14±1.780 |
| 3p | 1.700±0.3408 | 38.66±1.811 | 8.01±1.626 | 23.05±0.6895* |
| 3q | 1.447±0.1988* | 27.00±0.9526 | 11.24±1.250 | 31.82±0.4719 |
| 3v | 5.280±0.5658** | 30.39±1.883 | 8.15±0.9355*** | 23.86±3.300 |
All thioureas were tested at 100 µM. # Results are the mean±S.E.M. of four independent experiments. * p<0.05, ** p<0.01, *** p<0.001, in relation to control.
The single mitochondrion of Leishmania parasites is involved in death by apoptosis, and the increase in the number of cells in the sub-G0/G1 hypodiploid DNA (Table 3) might represent promastigotes that undergo this type of death.31) Thus, we measured whether thiourea derivatives induced promastigote mitochondrial alteration by the reduction of the mitochondrial membrane potential (ΔΨm) using JC-1 assay. Our data showed that 3k and 3l reduced ΔΨm of promastigotes by 25 and 30%, respectively. Similarly, promastigotes treated with 30 µM miltefosine, which was used as a positive control, had their ΔΨm reduced by 27% (Fig. 3).
SARHerein we calculated some stereoeletronic properties of the ten most active thiourea derivatives. The descriptors HOMO and LUMO energies, lipophilicity (c Log P), molecular volume, molecular area, number of hydrogen bond donors (HBD), acceptors (HBA) groups and dipole moment were evaluated in order to correlate to the leishmanicidal activity (Table 4). The parameters related to molecular shape (Area, Volume and PSA) and weight (MW) varied within the series but we did not observe a clear correlation between these parameters and activity. Although, in case of 3x, 3y and 3z (R=3,4,5-OMe), it seems that higher MW, volume and area were correlated to the reduced activity of these compounds on promastigotes. Also, in case of 3j and 3y with 3,4,5-OMe-C6H5 substituent in R1 position, the activity was abolished. It is important to notice that changing substituents can affect the ability of the molecule to interact with the target, besides, it also changes various parameters such as its partition coefficient, electronic density, steric environment, membrane permeability and others.32) We observed that more than 5 hydrogen bond acceptors (HBA) reduced activity on promastigotes and, in general, lower PSA and higher c Log P values were better for antileishmanial activity. Thiourea 3e was an exception, with one of the lowest IC50 in promastigote (26.04 µM) and highest PSA and lower c Log P values, but as it was toxic for macrophages, we considered it as a non-promising compound for further analysis. Thiourea 3e was the only one with good activity in promastigote that present aliphatic chain in R1 and R2. We observed that higher activity was achieved by compounds with aromatic ring in R1. Derivatives bearing the rings benzo1,3)dioxole (3k, 3v and 3p), naphthalene (3l) and 1,1-diphenyl (3q) showed the best activities in promastigotes, pointing that more bulky rings in this position could improve activity. Furthermore, derivatives bearing less bulky ring (i.e., phenyl ring or hexyl) in R1 as 3f, 3g, 3h, 3j, 3o and 3u were inactive, while 3i, 3x and 3y showed some activity in promastigotes, but they were less expressive than others. Thiourea 3q (R1=R2=phenyl) revealed the importance of the second substituent for the activity and it can be shown by the comparison with the inactive 3g with phenyl only in R1.
Table 4. Stereoelectronic Properties of
N,
N′-Disubstituted Thioureas Including: Molecular Weight (MW), Area (Å
2), Volume (Å3), Polar Surface Area (PSA, Å
2), Dipole Moment (μ, Debye); HOMO and LUMO Energies (eV), Hydrogen Bond Acceptors (HBA), Hydrogen Bond Donors (HBD) and Octanol–Water Partition Coefficient (
c Log
P)
| # | MW | Volume (Å3) | Area (Å2) | PSA (Å2) | DM | EHOMO (eV) | ELUMO (eV) | HBA | HBD | c Log P |
|---|
| 3e | 240.33 | 243.46 | 264.59 | 44.36 | 3.62 | −5.36 | −0.70 | 5 | 3 | 1.34 |
| 3i | 288.37 | 291.61 | 315.02 | 32.68 | 4.91 | −5.30 | −0.95 | 5 | 2 | 3.41 |
| 3k | 272.33 | 259.97 | 272.98 | 37.83 | 6.07 | −5.51 | −1.11 | 5 | 2 | 3.44 |
| 3l | 278.38 | 285.87 | 293.93 | 20.91 | 5.81 | −5.63 | −1.67 | 3 | 2 | 4.66 |
| 3p | 286.36 | 281.62 | 303.53 | 35.49 | 5.95 | −5.57 | −0.82 | 5 | 2 | 3.51 |
| 3q | 318.44 | 338.54 | 348.59 | 11.06 | 4.88 | −5.56 | −1.06 | 3 | 1 | 5.10 |
| 3v | 300.38 | 299.61 | 320.68 | 34.24 | 5.79 | −5.52 | −0.83 | 5 | 2 | 3.79 |
| 3y | 408.48 | 397.09 | 416.37 | 58.16 | 6.77 | −5.56 | −1.17 | 9 | 2 | 2.90 |
| 3x | 324.45 | 332.41 | 358.22 | 36.89 | 6.51 | −5.50 | −0.81 | 6 | 2 | 3.16 |
| 3z | 362.41 | 341.05 | 356.02 | 56.28 | 6.83 | −5.54 | −1.17 | 8 | 2 | 3.06 |
Thiourea 3v showed the best activity in amastigotes (IC50=70 µM) and some features seemed to be important for the activity, as the presence of benzo1,3) dioxole ring as well as the length of aliphatic chain, as can be seen by the comparison with 3k (IC50=150 µM) and 3p (IC50=145 µM).
Thiourea derivatives did not show toxicity on macrophages, with CC50>200 µM, with exception of 3e (CC50=49.22 µM), which exhibited cytotoxic profile, despite its good activity against promastigotes (IC50=26.04 µM). Cytotoxic profile of 3e is probably related to the primary alcohol as substituent in R1 and/or R2. All other nontoxic thiourea derivatives presented aromatic substituents in R1 or hexyl group (e.g. 3x). In addition, it was also possible to note that R1 and R2 substitution by phenyl group resulted in compound 3q that also presented good cytotoxic profile.
In Silico ADMET StudiesThe most active thiourea derivatives (3e, 3i, 3k, 3l, 3p, 3q, 3v, 3x, 3y and 3z) were screened for an ADMET study using FAF-DRUGS3 server. The theoretical oral bioavailability was analyzed using Lipinski rule of Five (c Log P≤5, MW≤500 Da, HBD≤5, HBA≤10)33) Egan Rule (0≥tPSA≤132)34) and Veber Rule (rotatable bonds≤10 and tPSA≤140 Å).35) Drug safety profile of the derivatives was evaluated using well known rules developed by the pharmaceutical industry such as GSK 4/400 rule (higher risks of toxicity if c Log P>4 and MW>400),36) Pfizer 3/75 rule (lower toxicity when c Log P<3 and (tPSA)>75 Å2)37) and a set of 275 rules applied by Lilly to identify compounds that may interfere with biological assays.38) Furthermore, we analyzed the derivatives for the presence of toxicophores and promiscuous compounds in biochemical assays (PAINS) using FAF-DRUGS3.29)
FAF-DRUGS3 predicted a good theoretical oral bioavailability for the most active compounds using Lipinski, Veber and Egan rules. Our analysis of toxicophores and problematic groups showed an alert of low risk associated with the thiourea group. In addition, no PAINS were identified indicating that their activity is might not be a false positive. The drug safety of the most active compounds was promising with 3e, 3i, 3k, 3l, 3p, 3q, 3v, 3x, 3y and 3z passing GSK 4/400 rule. On the other hand, most of these compounds did not pass Pfizer 3/75 rule since most of them have c Log P>3 and PSA<75 Å2. Interestingly, derivatives with best activity against promastigotes, 3k, 3p, 3q and 3v, with exception of 3l, passed Lilly rules pointing that these compounds could be promising, specially 3q and 3v, the most active compounds (Table 5).
Table 5. Structural Alerts Identified with FAF-DRUGS3 and Analysis of GSK 4/400 (Higher Risks of Toxicity if Log
P> 4 and MW>400)
| # | Drug safety profile |
|---|
| GSK 4/400 | Pfizer 3/75 | Lilly rules | Alerts |
|---|
| 3e | Good | Good | Pass | Low risk thiourea |
| 3i | Good | Bad | Pass | Low risk thiourea |
| 3k | Good | Warning | Pass | Low risk benzodioxolane and thiourea |
| 3l | Good | Bad | Warning-naphthalene | Low risk thiourea and c Log P |
| 3p | Good | Warning | Pass | Low risk benzodioxolane and thiourea |
| 3q | Good | Bad | Pass | Low risk thiourea and c Log P |
| 3v | Good | Bad | Pass | Low risk benzodioxolane and thiourea |
| 3x | Good | Warning | Pass | Low risk thiourea |
| 3y | Good | Good | Pass | Low risk thiourea and H-bond acceptors |
| 3z | Good | Good | Pass | Low risk benzodioxolane and thiourea |
Pfizer 3/75 (lower toxicity when c Log P< 3 and (tPSA)>75 Å2) and Lilly rules for the most active thiourea derivatives.
The ten most active thiourea derivatives were submitted to a toxicity analysis using Osiris Property Explorer to identify structures with mutagenic, tumorigenic, irritant and potential effects on reproduction. Only 3l showed high mutagenic and tumorigenic risks, related to the naphthalene ring. Interestingly, the naphthalene group is not included on FAF-DRUGS3 list of toxicophores29) but this derivative was promptly rejected using Lilly rules for problematic compounds. The other compounds presented no risks alert in Osiris. These results suggested a satisfactory theoretical toxicity profile of the most active compounds 3k, 3p, 3q and 3v pointing them as promising lead structures for future analysis.
Discussion
Here we report the leishmanicidal activity of thiourea derivatives against L. amazonensis, a New World species that can cause cutaneous, diffuse cutaneous and visceral leishmaniasis.39)
It has been shown that different compounds affect the leishmanial cell cycle, some with characteristics suggesting the induction of incidental death, an apoptosis-like mechanism.22,30,40) Therefore, we analyzed the effect of derivatives on parasite cell cycle.
Among the 26 derivatives assayed, 3q and 3v were highly active against amastigotes with low toxicity for host macrophages. The 3v derivative showed the lowest amastigote IC50 value and both, 3q and 3v derivatives affected promastigotes cell cycle without disturb the mitochondrial membrane potential. Although their mechanism of action was not fully understood, the induction of promastigote apoptosis was evident in by the cell cycle data. Interestingly, 3k and 3l significantly increased the number of cells in sub G0/G1 phase and decreased the mitochondrial membrane potential, reflecting a pro apoptotic mechanism in promastigotes. On the other hand, these compounds exhibited lower activity than 3v and 3q against amastigotes.
Other groups have studied thiourea derivatives as antileishmanial agents. Using a phenotypic fragment-based screening, Blaazer et al. identified compounds with thiourea fragments as promising antiprotozoal candidates. However, these compounds did not show activity against Leishmania infantum amastigotes. It is interesting to point that the prototype used by this group was similar to our derivatives, bearing two substituents on nitrogens and varying the length between the thiourea group and substituents.41) Saudi et al. identified nine urea and thiourea derivatives with antileishmanial activity in vitro (L. major promastigotes), and four of them also showed in vivo activity.42) Among 31 monosubstituted thiourea derivatives which showed promastigote IC50 values above 100 µM Boechat described three derivatives which were more than 50-fold less active than pentamidine against L. amazonensis promastigotes and were not tested against amastigotes. In addition, authors proposed the enzyme arginase as a supposed target for those compounds.43)
NO production by macrophages is a potent mechanism involved in Leishmania killing, thus we tested the capacity of the thioureas derivatives to modulate NO production by infected macrophages. Although both 3q and 3v derivatives decreased NO production of unstimulated infected macrophages, they did not change NO production in LPS-IFN-γ stimulated infected macrophages, suggesting that the leishmanicidal activity of these derivatives could occur by a mechanism different from NO activation.
In this study, we used molecular modeling approaches to understand the relationship of the structures of a series of thiourea derivatives and their antileishmanial activity. The basis of SAR method is that structurally similar compounds tend to show similar biological activity and based on this fact, it is possible to found the structural characteristics and properties correlated with the activity for a series of compounds that act in the same target, even when the potential target is unknown.44)
We observed that higher activity was found for compounds with aromatic and bulky ring in R1, lower PSA and higher c Log P values were better for antileishmanial activity, while higher HBA reduced activity on promastigotes. Also, we found that cytotoxic profile of 3e is probably related to the primary alcohol as substituent. Together these features can guide the synthesis of new derivatives. It is important to highlight that the analysis of compounds’ structural features and properties should be done together as in some cases there is no significant difference in molecular properties but the compound was not able to bind to the target.
Our derivatives showed satisfactory theoretical ADMET properties, fulfilling Lipinski, Veber and Egan bioavailability guidelines, suggesting that they could be administered by oral route. We also observed that these compounds were not labeled as PAINS, indicating that the activity observed might not be result of a false positive.45) Thus, our derivatives could represent a starting point for new analogs series. However, FAF-DRUGS3 flagged the thiourea moiety as problematic, due to oxidative metabolism and formation of reactive metabolites.46) Our theoretical analysis of physicochemical properties, such as Log P, PSA and MW, showed that most derivatives did not pass Pfizer 3/75 guideline due to high lipophilicity. On the other hand, most active derivatives passed GSK 4/400 and Lilly set of guidelines, suggesting that although our derivatives do not have optimal physicochemical properties, they might be modified in order to search for better analogs. In addition, since these guidelines use different threshold for Log P, PSA and MW, thus it is expected that compounds present different results for each industrial guideline used in drug discovery.36–38) The analysis of ADMET properties late in preclinical or clinical stages results in higher costs in drug design, while in silico evaluation of these properties in preliminary stages may help the selection of the most promising compounds and elimination of problematic molecules earlier. Several rules have been developed to select molecules for optimization, they are very usefull but should not be used blindly to avoid discarding interesting molecules.29) Herein, more than developing a new prototype, we described the antileishmanial activity of thiourea derivatives and the potential of this class for further modifications to find compounds with higher activity. Overall, this study opens new possibilities of exploring thiourea derivatives to improve and/or design new antiparasitics.
