Synthesis and anti-Trypanosoma cruzi profile of the novel 4-thiazolidinone and 1,3-thiazole derivatives

Chagas disease is a serious public health problem in Brazil and world. Caused by the protozoan Trypanosoma cruzi, it is estimated 6-7 million people infected worldwide. The only drug used to treatment is Benzonidazole, but this drug is only effective in the acute phase of the disease. This paper reports the synthesis and the anti-Trypanosoma cruzi activity of 4-thiazolidinone and 1,3-thiazole derivatives based in thiosemicarbazones previously described as potent trypanocidal agent, planned through the bioisosterism strategy. Therefore, the synthesis of 28 aryl-4-thiazolidinones (2a-r and 3a-j) was achieved, which aryl ring possesses halogens atoms in meta and para positions, and the heterocyclic system had lipophilic substituents. These compounds were thus evaluated as anti-T. cruzi agents against epimastigote, trypomastigote and amastigote forms of T. cruzi. Compounds were also tested its toxicity in L929 fibroblasts. In view to investigate a bioisosteric relationship between 1,3-thiazoles and 4-thiazolidinones, eighteen 1,3thiazoles derived from trifluoromethyl thiosemicarbazone (4a-r) were also tested in the same model. It was possible to show that between the 46 tested compounds, those that possess a bromine (2a-r) atom showed better activity compared to compounds substituted by a trifluoromethyl group (3a-j) and to 1,3-thiazole derivatives (4a-r), which were inactive. In general, the 2a-r series showed low toxic profile in the cell line tested. Besides, compound 2h was the most active of then all when compared to the standard Benznidazole. *Correspondence to: Ana Cristina Lima Leite, Laboratório de Planejamento em Química Medicinal, Departamento de Ciências Farmacêuticas, Centro de Ciências da Saúde, Universidade Federal de Pernambuco, 50740-520, Recife, PE, Brazil, E-mail: acllb2003@yahoo.com.br


Introduction
Chagas disease or American Trypanosomiasis is a serious disease resulting from parasitic infection by the protozoan Trypanosoma cruzi, and triatomine insects as vectors [1,2]. About 7 million people are infected worldwide and Latin America countries are endemic areas. It is estimated that 30% of chronically infected people develop cardiac disorders and up to 10% develop digestive, neurological or mixed changes that may require special treatment [1][2][3].
Currently, the only drugs used in Chagas disease therapy are Benznidazole (BZD) and nifurtimox, which are effective in the acute phase of the disease, but less effectivity in patients in the chronic phase, presenting only palliative effect [2,[4][5][6][7]. Furthermore, BZD is not considered a suitable drug, since exhibits serious side effects such as hypersensitivity, rash, gastrointestinal disorder etc., leading to a large number of patients interruption the treatment [2,8]. Therefore, the search for new drugs and bioactive molecules more selective is pivotal.
In this context, it is well noticed that several drugs and drug candidates in clinical development possess halogen atoms in their structures [7,22]. In fact, they contribute favorably through a molecular interaction in the stabilization of the protein-ligand complex. In addition, synthesis of fluorine, chlorine, and bromine derivatives are well known in industrial scale and possess desirable stability and cost [23].
In the last few years, our research group had been used the building-block strategy in view to optimize the antiparasitical profile. Recently, the 1,3-thiazole core was obtained from 3'-bromopropiophenone-thiosemicarbazone and trifluoromethylthiosemicarbazone in view to access new lead generation of anti-T. satisfactory yield (85% for compound 1d and 94% for compound 1e). The thiazolidin-4-ones (3a-j) were prepared by cyclization of compounds 1d or 1e and respective acid/ester, under reflux for 24 hours. The precipitate was filtered in Büchner funnel, washed with cold ethanol, and then dried over SiO 2 . The yields were satisfactory (>40%). The synthesis of 4a-r series is described by [25].
In view to set up the structural planning of the 4-thiazolidinone series 2a-r and 3a-j, the aimed perspectives were based on followed points: a) the choice of the halogen (bromine or fluorine), positions to be inserted (meta or para) in the aryl ring, employing 3-bromophenyl propanone or 4-(trifluoromethyl)benzaldehyde; b) the lipophilic and steric profile of the substituents at C5 position, adopting different esters; c) attachment of a methyl or phenyl group at nitrogen N3 by applying 4-methyl or 4-phenyl-thiosemicarbazide. About the ethyl group present at R 2 position in the series 2a-r coming from the ketone used (Scheme 1). Therefore, two thiazolidinones series were synthesized: 2a-r and 3a-j. For synthesis of desired compounds (Scheme 1), it was followed procedures previously described by [10].
Regarding 1,3-thiazole series (4a-r), variations were made in the group linked to 4 position of the thiazole ring, through the insertion of aromatic rings with different substitutions in ortho, meta and para positions ( Figure 2). The synthesis of series 4a-r followed the procedures described by de Oliveira [26], and is described by 25. The chemical structures of compounds were determined by nuclear magnetic resonance (NMR, 1 H, and 13 C), infrared (I.R.) and mass spectra (HR-MS), while purity was determined by elemental analysis (EA).

Biological evaluation
The in vitro anti-T. cruzi activity of compounds for epimastigote, trypomastigote and amastigote T. cruzi forms was evaluated, and cruzi agents. In fact, most of these compounds exhibited antiparasitic activity similar to BZD [24] (Figure 1).
Bearing in mind the bioisosteric relationship between 1,3-thiazole and 4-thiazolidinone cores and by the promising results obtained by Moraes (2016) and Du (2002), in this work, our group presents two series of 4-thiazolidinones, derived by 3'-bromopropriophenonethiosemicarbazone and 3-(trifluoromethyl)thiosemicarbazone (2ar and 3a-j, respectively), in view to achieve novel and more selective candidates against T. cruzi. In this approach, 28 compounds were synthesized with variations lipophilic groups inserted at N3 and C5 positions of the 4-thiazolidinone ring. In addition, these compounds were evaluated for their trypanocidal potential in the epimastigote, trypomastigote and amastigote forms of the parasite, and their cytotoxicity was tested in L929 fibroblast cells. Analyzing the activity against the three T. cruzi forms is important in an attempt to clarify the mechanism of action of compounds. In view to make a structureactivity biological profile between 1,3-thiazole and 4-thiazolidinone nucleus, we also evaluated the anti-T. cruzi properties of eighteen 1,3-thiazoles derived from (trifluoromethyl)thiosemicarbazone (4a-r).

Chemistry
For the synthesis of the desired compounds, the procedures previously reported by 10,25 were followed. First, the thiosemicarbazones 1d and 1e were prepared by reacting commercially available thiosemicarbazide or methyl thiosemicarbazide, with 4-(trifluoromethyl)benzaldehyde (1:1 mol ratio) using ethanol, under reflux, in the presence of catalytic amount of hydrochloric acid, for 3 hours. After the final reaction, water was added for precipitation of the compounds. This reaction led to a    [10,21,23,24] their toxicity was evaluated in L929 fibroblasts. Among the tested compounds, the series 2a-r showed the best results when compared with 3a-j series. The 2a-r series possess a bromine atom at the position 3 of the aryl ring system and the 3a-j series possess a trifluoromethyl group at position 4 of aryl ring system.
In recent studies [24] a series of 1,3-thiazoles derived from 3-(bromopropiophenone) thiosemicarbazone were synthesized and their results demonstrated similar activity to BZD (T. cruzi trypomastigote form). Furthermore, compounds possessing halogens in their structure (fluorine and bromine), in general, exhibit interesting biological activities and interactions [23] in perspective of these promising outcomes, we decided to explore the anti-T. cruzi potential of bromine (series 2a-r) and trifluoromethyl group (series 3a-j) 4-thiazolidinone derivatives.
For series 3, which possesses the core thiazolidine-4-one, the compound 3d (IC 50 = 7.8 µM) was the most active of 3 series. The insertion of an group containing halogen at the para position (series 3a-j), was not beneficial for the activity against the trypomastigote form when compared to series 2a-r.
Considering the inhibitory activity of amastigote forms for the entire series, only compound 2h was active, being more potent than BZD, with an IC 50 = 2.4 μM. The notable feature of compound 2h is, besides Br atom present at the position 3 of the aryl ring, 4-thiazolidinone ring, possess a methyl group in N3 position and a methyl group in C5.
About the substitutions in N3 it was noticed that the insertion of the methyl group was responsible for the increase the activity. Also, between the compounds 2a, 2g, and 2m, it was possible to verify that the methyl group is pivotal to the reduction of IC 50 .
The compounds 1a-e, thiosemicarbazones that originated the respective 4-thiazolidinones, can be observed that intermediates 1a and 1d, which possess H in R 3 , presented the best IC 50 values for amastigote form, suggesting that the presence of hydrogen bonds donors in the structure may be of interest for improving the biological activity in the amastigote form.
When we compare the trypanocidal activity in amastigote, the cyclic 4-thiazolidinone derivatives 2g (IC 50 =12.2µM) and 2h (IC 50 =2.4µM) were about 4 and 21-folds respectively more active than intermediate compound 1b (   In general, the compounds substituted at the N3 position of the 4-thiazolidinone ring with methyl, for the sub-series 2, were active against amastigote form (2g, 2h, 2j, 2k). It is interesting to note that, in contrast to our previous work [10], the 4-thiazolidinones of the present work, with phenyl in the N3 position were not the most active, with the exception of compound 2n (IC 50trypo = 3.7 µM), which has a methyl group in C5 of the heterocyclic ring.
Described by [23], within the class of drugs that present halogen atom in their structure, those that possess a fluoride, are in greater quantity since the fluorination changes the physical, chemical and conformational parameters. This feature could eventually result in optimized pharmacological properties. The fluorine chemistry provides good opportunities for enhancing the binding affinity of potential drug candidates. These features have made the trifluoromethyl (CF 3 ) useful chemical groups in the contemporary drug design [27]. Within this observation, we can say that the molecules which present trifluoromethyl (CF 3 ) group in their structure (1d-e and 3a-j), were not the most active, as expected when compared to the series presenting the halogen bromine (1a-c and 2a-r).
In the sub-series 3, five compounds were active against trypomastigote form: 3b (IC 50 =61.9 µM), 3c ( series 2, structural characteristic is a methyl in N3 is not beneficial for series 3. The sub-series 2 had the most active compounds substituted in N3 by a methyl, generally, the most active compound being 2h (Table 1).
Regarding the 1,3-thiazole series, three compounds had lower IC 50 values for the amastigote form of T. cruzi, compounds 4b (IC 50 = 9.9μM), 4d (IC 50 = 5.3μM) and 4h (IC 50 = 8.9μM), they have very electronegative groups attached to the aromatic ring bound to the 1,3-thiazole as a characteristic, however, all compounds presented a very low selectivity index ( Table 2). Two of these compounds (4d and 4h) have in common the presence of two chlorines in the aromatic ring bound to the 1,3-thiazole. One at 2 and 4 positions (4d) and the other at 3 and 4 positions (4h), which may indicate that di-substitution with halogens in the aromatic ring may be beneficial for anti-T. cruzi activity, this data corroborates with that recently observed by 10,26.

Physicochemical properties
We also evaluated the physicochemical properties to determine if they are compliant with the Lipinski's rule [28,29]. This rule has important determinants to providing better pharmacokinetics and analyses promising future drug development. For this purpose, physicochemical and ADME properties were calculated using the SwissADME (a free web tool to evaluate pharmacokinetics, drug-  Compound obeying at least three of the four criteria are considered to adhere the Lipinski Rule [29]. All synthesized compounds are compatible with Lipinski Rule. Another interesting property is the number of rotatable bonds and the polar surface area (PSA). A large number of rotatable bonds (≥10) has been associated with poor oral bioavailability [30]. Compounds with a low PSA (≤ 140 Å2) tend to have higher oral bioavailability [30,31]. All synthesized compounds have appropriate PSA and number of rotatable bonds (Table 3 and  Table 4).

SUMMARY OF BIOLOGICAL STRUCTURE-ACTIVITY RELATIONSHIPS
As demonstrated in Table 5, the most active compounds shown variable permeability based on gastrointestinal absorption (GI), according to the BOILED-Egg predictive model (Brain Or IntestinaL EstimateD permeation method). Three compounds showed high gastrointestinal absorption (1a, 2h and 2n). With respect to oral bioavailability, it's expected 0.55 of the probability of oral bioavailability score > 10% in the rat for all compounds, similar to BZD. All these data, suggests a good in silico drug-likeness profile and great chemical stabilities for all compounds synthesized.

Conclusion
This study has synthesized, characterized and identified 46 new heterocyclic compounds, based on previous results. These compounds were assayed against epimastigote, trypomastigote, and amastigote forms of T. cruzi, as well as their toxicity in L929 fibroblasts. For epimastigote form, the most active compound was 4e. Towards trypomastigote and amastigote forms of T. cruzi, the most active heterocycle compounds identified in this study were 2n (most active against trypomastigote form) and 2h (most active against amastigote form). The findings of this research provide insights into valuable strategies of synthesis and Structure-Activity relationships (SAR) for the planning of new anti-T. cruzi drugs candidates.

Experimental General
Most the chemicals were purchased from Sigma-Aldrich (St. Louis, USA), Merck (Berlin, Germany) or Alfa-Aesar (Massachusetts, USA). Reactions in ultrasound bath were performed in a Unique EM-804 TGR instrument, with a frequency of 40 kHz and a nominal power of 180 W, without external heating. Precoated aluminum sheets (silica gel 60 F 254 , Merck) were used for thin layer chromatography (TLC) and spots were visualized under UV light. IR spectra in KBr pellets were acquired at Bruker FT-IR spectrophotometer. 1  de Siqueira LRP (2019) Synthesis and anti-Trypanosoma cruzi profile of the novel 4-thiazolidinone and 1,3-thiazole derivatives  Table 5. ADME properties of most active compounds *BBB -blood-brain barrier. **GI -Gastrointestinal absorption singlet; d, doublet; t, triplet; q, quartet; m, multiplet. Chemical shift values were given in, ppm. DEPT was employed to confirm the carbon assignment. Melting points were measured using capillaries on a Thomas device Hoover.

Synthesis of intermediate thiosemicarbazones. Example for 3-bromophenyl propylidene hydrazine carbothioamide (1a)
In a round bottom flask for 50 mL, the 3-bromophenyl propanone (0.45 g, 2.12 mmol) was dissolved in 8 mL EtOH, following by the addition of catalytic HCl. The flask was placed in an ultrasound bath (40 kHz, 180 W) and under sonication, 0.23 g (2.12 mmol) of thiosemicarbazide was added in the reaction. After 3 h, distilled water was added, and the precipitate formed was cooled at 0° C and the precipitate was filtered in a Büchner funnel with a sintered disc filter, washed with ethanol and then dried over SiO 2 . Colorless crystals were formed in a yield of approximately 70%.

Synthesis of aryl-thiazolinones (2a-2r). Example: 3-bromophenyl propylidene hydrazone -3,5-dimethylthiazolidin-4-one: (2h):
In a round bottom flask for 50 mL, was added the compound INT-2 (0.61 g, 2.12 mmol), 20 mL of ethanol and anhydrous sodium acetate (0.71 g, 8.28 mmol). After heating for about 15 minutes, methyl 2-bromopropionate was added (0.65 g, 4.24mmol) and the reaction was stirred and refluxed (120° C) for approximately 20 hours (monitored by TLC). The reaction was cooled at 0° C for two days and, subsequently, the powder was separated from the supernatant Buchner funnel with a sintered disc filter, washed with ethanol and then transferred to a desiccator under vacuum. There was obtained a yellow coloured powder with a yield of 92%.