From rational design to serendipity: Discovery of novel thiosemicarbazones as potent trypanocidal compounds

Abstract

Chagas disease is a major public health problem caused by Trypanosoma cruzi, with an estimated 6–7 million people infected and 70 million at risk of infection. T. brucei gambiense and T. brucei rhodesiense are two subspecies of related parasites that cause human African trypanosomiasis, a neglected tropical disease with also millions of people at risk of infection. Pharmacotherapy for both diseases suffers from low efficacy, side effects, or drug resistance. Recently, we reported a noncovalent competitive inhibitor of cruzain (IC₅₀ 26 μM, K_i 3 μM) and TbrCatL (IC₅₀ 50 μM), two cysteine proteases considered promising drug targets for trypanosomiasis. Here, we describe the design and synthesis of derivatives of our lead compound. The new thiosemicarbazone derivatives showed potency in the nanomolar concentration range against the two enzymes, but they were later characterized as aggregators. Nevertheless, the thiosemicarbazone derivatives showed promising antiparasitic activities against T. b. brucei (EC₅₀ 13–49.7 μM) and T. cruzi (EC₅₀ 0.027–0.59 μM) under in vitro conditions. The most active thiosemicarbazone was 200-fold more potent than the current anti-chagasic drug, benznidazole, and showed a selectivity index of 370 versus myoblast cells. We have identified an excellent candidate for further optimization and in vivo studies.

1. Introduction

Chagas disease (CD) represents a significant public health concern since its discovery in 1909. The disease threatens populations in Latin America, the area originally affected, and non-endemic countries, primarily due to migration [1]. Worldwide, approximately 6–7 million people are infected, and 70 million are at risk of infection with Trypanosoma cruzi, the etiologic agent of CD. The available pharmacotherapy is based on two drugs, benznidazole and nifurtimox, which have been the only available treatments for over 50 years [2]. Furthermore, these compounds show low efficacy in the chronic phase of the disease and cause side effects leading to low patient compliance [3,4].

Human African Trypanosomiasis (HAT) is caused by two subspecies of Trypanosoma brucei, namely T. brucei gambiense and T. brucei rhodesiense. HAT affects populations from low-income countries, mainly in sub-Saharan Africa. Although recent international initiatives have led to a significant drop in the annual number of new cases, 70 million people are still at risk of HAT infection [5]. HAT pharmacotherapy, currently composed of five drugs, is considered a challenge due to limited efficacy, drug resistance, severe side effects, and the need for medical supervision, which is problematic in endemic countries [2,5–7]. The recent approval of fexinidazole, aiming at oral therapy for both first and second-stage infection by T. b. gambiense, opens new possibilities and brings hope to the treatment after decades of pharmacotherapy stagnation [4–6,8–10]. However, like so many other parasitic diseases [11], the need to develop new drugs is constant, given the ability of T. brucei to develop drug resistance [12,13].

Both T. cruzi and T. brucei express cysteine proteases essential for successful parasitism. In T. cruzi, the multigene family of cruzipains comprises four subtypes from two protein families [14]. Over the past decades, medicinal chemistry efforts have focused on cruzain, the recombinant version of cruzipain 1, with a truncated C-terminal domain [15–20]. Cruzipains are involved in many cellular processes such as nutrition, modulation of macrophage response, and cell invasion [21,22]. In T. brucei, the major cathepsin L enzyme is TbrCatL (a.k.a. rhodesain, a nomenclature currently discouraged) [14,16,21,23]. TbrCatL is considered a HAT therapeutic target as it is crucial for penetration through the blood-brain barrier and essential for parasite survival [24–26]. Structurally, cruzain and TbrCatL are very similar, sharing 70% of their sequence and differing only by two amino acids at the catalytic site, which can translate into inhibitors active against both enzymes and parasites [27–29]. Various small molecule inhibitors have been developed as candidates for treatment of both CD and HAT, including vinyl sulfones, halomethylketones, nitriles, benzimidazoles, and thiosemicarbazones derivatives.[Reviewed in 13,14]

Recently, we reported the cruzain inhibitor 1-(5-(4-bromo-2-chlorophenyl)furan-2-yl)-N-(piperidin-4-ylmethyl)methanamine (herein MB12, following the numbering in the original publication) as a candidate for CD pharmacotherapy [30]. The trypanocidal activity of MB12 was first reported following our screening of the Malaria Box compound set [30], which was distributed worldwide by the public-private partnership initiative Medicines for Malaria Venture [31,32]. MB12 displayed activity against cruzain (IC₅₀ 26 μM, K_i 3 μM) and TbrCatL (IC₅₀ 50 μM), acting as a noncovalent competitive inhibitor. The compound did not inhibit orthologous mammalian enzymes, cathepsins S (human) and B (rat), and, furthermore, decreased the replication of T. cruzi Y strain parasites and their release from macrophages under in vitro conditions. Last, no significant cytotoxicity was demonstrated by monitoring lactate dehydrogenase release from macrophages. Therefore, MB12 was considered a promising hit for subsequent medicinal chemical optimization. Herein, we describe the design and synthesis of analogs of MB12 that inhibit the growth of T. cruzi and T. b. brucei in vitro.

2. Results

2.1. Screening of a related compound series against cruzain and TbrCatL

Krake and colleagues described the inhibitory activity of 2,5-disubstituted furans and analogs against P. falciparum [33]. Considering the structural similarity between their lead compound and MB12 (Fig. 1), we investigated this series as cruzain and TbrCatL potential inhibitors.

Fig. 1.

Molecular structures of Krake et al. lead compound against Plasmodium falciparum [33] (A) and MB12 (B).

The screening of this set allowed us to observe structural similarities, guiding subsequent modifications of the basic scaffold. Among 31 compounds evaluated, compounds 9 and 12 presented an average inhibition greater than 70% against cruzain or TbrCatL when evaluated at 100 μM (Supplementary Material, Table S1). These two compounds shared a common scaffold (Fig. 2A), i.e., bromine or chlorine substituents in the phenyl ring and the nitrogen atoms spaced by three methylene groups. Considering these characteristics, we designed new compounds by employing the bioisosteric replacement of furan with a triazole ring (Fig. 2B). This strategy has been widely used in medicinal chemistry to replace functional groups like amides, olefins, and other heterocycles, given the triazole’s polarity, rigidity, and capacity to act as a hydrogen bond acceptor [34,35]. Also, given the ease of synthesizing triazoles through click chemistry, such modification would facilitate the synthesis of an analog series.

Fig. 2.

A) Common structural features of compounds 9 and 12 from the Krake et al. set. In parenthesis, enzymatic inhibition (X ± SEM %) of cruzain and TbrCatL, respectively. Compounds were evaluated at the concentration of 100 μM without previous incubation with the enzymes. Reported values refer to the average and standard error of the mean of two experiments, each in triplicate. B) General structure of triazole derivatives designed in this study.

2.2. Design and evaluation of triazole derivatives based on arylfurans structure

From a retrosynthetic point of view, triazole derivatives can be prepared from affordable anilines and primary amines. Thus, following knowledge obtained from the first set, we selected reagents to prepare a small set of MB12 analogs, prioritizing halide substituted anilines in three different patterns (4-chloroaniline, 3,4-dichloroaniline, and 2-chloro-4-bromoaniline, as in MB12), a bulky cycloalkyl aliphatic spaced diamine (4-(aminomethyl)piperidine), and a small aminoalcohol (ethanolamine). These triazoles were evaluated against cruzain and TbrCatL at 100 μM under two conditions, with or without a 10 min pre-incubation with the enzymes. The pre-incubation procedure allows slow binding compounds to achieve equilibria before substrate addition. These compounds had low enzymatic inhibition against both enzymes. The most active compound (36) showed an IC₅₀ of 168 μM against TbrCatL (Supplementary Material, Table S2, Figs. S1 and S2), but further assays characterized it as an aggregator (Supplementary Material, Table S3). Aggregators, a common type of promiscuous inhibitors [36,37], can be experimentally detected based on their reduced inhibitory activity in assay buffers containing nonionic detergent or upon pre-incubation with BSA and are frequently characterized by steep dose-response curves [38]. Compound 36 was considered an aggregator as we observed a decrease in TbrCatL inhibition in the presence of 0.01% and 0.10% of Triton X-100 and upon pre-incubation with BSA (Supplementary Material, Table S3). These results indicated that the replacement of the furan ring was deleterious for enzymatic inhibition.

To confirm the impact of furan-triazole substitution, we synthesized new furan derivatives, including molecules containing the same substituents present in the triazole derivatives (Table 1). Two of the five synthesized compounds had IC₅₀ values between 30 and 100 μM against cruzain (37, IC₅₀ = 51 ± 27 μM, and 39, IC₅₀ = 92 ± 38 μM). and three against TbrCatL (37, IC₅₀ = 78 ± 14 μM, 39, IC₅₀ = 79 ± 40 μM, and 40, IC₅₀ = 90 ± 3 μM). Despite their low potency, compounds with the same substitution pattern of triazole derivatives showed activity against at least one of the enzymes (33 → 37 and 34 → 38), suggesting that the furan ring is important for inhibition. The incorporation of bulkier (N-ethylmorpholine) and more polar (1,3-propanediol) groups at the amine moiety of furan derivatives resulted in decreased potency (37 → 39 and 37 → 40).

Table 1.

Screening of synthesized furan derivatives against cruzain and TbrCatL.

Compound	R1	R2	Enzymatic inhibition (X ± SEM %)a				IC50 (μM)b
			Cruzain		TbrCatL		Cruzain	TbrCatL
			NI	I	NI	I
37	2-Cl, 4-Br		63 ± 6	73 ± 10	68 ± 4	67 ± 4	51 ± 27	78 ± 14
38	3,4-Cl		62 ± 4	55 ± 5	29 ± 5	40 ± 20	nd	nd
39	2-Cl, 4-Br		63 ± 6	73 ± 2	62 ± 2	60 ± 6	92 ± 38	79 ± 40
40	2-Cl, 4-Br		63 ± 6	67 ± 0	64 ± 7	63 ± 2	149 ± 21	90 ± 3
41	3-CF3, 4-Cl		66 ± 8	44 ± 5	0 ± 0	51 ± 9	133 ± 40	nd

nd: not determined.

^aCompounds were evaluated at the concentration of 100 μM with (I) and without (NI) a 10 min pre-incubation period with the enzymes. Assays were performed in triplicate in two independent experiments, and the data were expressed as means with standard error.

^bFrom at least two independent curves using at least seven compound concentrations in triplicate. Experiments were performed without pre-incubation with the enzymes.

2.3. Design of thiosemicarbazone analogs

Aiming to improve inhibitor potency and guided by molecular docking studies (Supplementary Material, Fig. S3), we next explored thiosemicarbazone derivatives using furan and triazole scaffolds. The thiosemicarbazone moiety is a privileged structure, with many reported biological activities, including trypanocidal activity by inhibiting cysteine proteases [39–42]. Their inhibition mechanism involves covalent bond formation after the nucleophilic attack of a catalytic cysteine residue on the thiosemicarbazone carbon [43,44]. Four thiosemicarbazones were obtained, varying the 5-membered ring (furan or triazole), and the aryl moiety with a 4-bromo-2-chlorine substitution (similarly to MB12) or a less bulky para-chlorine substitution.

Three of the four thiosemicarbazones, compounds 43 (IC₅₀ = 2.5 ± 0.5 μM), 44 (IC₅₀ = 0.04 ± 0 μM), and 45 (IC₅₀ = 0.05 ± 0 μM), were potent against TbrCatL, with IC₅₀ values in the low micromolar or nanomolar range (Supplementary Material, Fig. S1). Differently from the previously synthesized derivatives, we observed a time-dependent inhibitory effect by the thiosemicarbazones (Table 2), as expected for covalent inhibitors. Time-turnover curves show a great increase in the inhibitory effect of compounds 43–45 upon 5 min of incubation with TbrCatL (Supplementary Material, Fig. S4). To verify whether compounds 44 and 45 were promiscuous inhibitors, they were assayed against an unrelated enzyme, the serine protease NS2B-NS3 from Zika virus. Assays were performed at 0.78 μM, a concentration corresponding to approximately twice their IC₅₀ values against TbrCatL. Neither of these compounds inhibited NS2B-NS3 (24% ± 5 and 8% ± 2 inhibition for compounds 44 and 45, respectively), suggesting specific TbrCatL inhibition. However, since aggregation is a condition-dependent phenomenon, the formation of colloidal aggregates cannot be ruled out based on the absence of NS2B-NS3 inhibition [45]. Thus, in addition to the initial experiment performed in a buffer containing 0.01% Triton X-100, we determined the IC₅₀ values for compounds 42–45 against TbrCatL in the absence of Triton X-100 and upon pre-incubation with BSA. A comparison of the obtained curves revealed a behavior very characteristic of aggregators, with IC₅₀ values up to three orders of magnitude lower in the absence of Triton and upon pre-incubation with BSA (Fig. 3). Because compounds 44 and 45 were less affected by the removal of Triton, we also determined their IC₅₀ values with 0.1% Triton X-100, which reduced the compounds’ potencies.

Table 2.

Screening of thiosemicarbazones against cruzain and TbrCatL.

Compound	Series	R1	Enzymatic inhibition (X ± SEM %)a				IC50 (μM)b
			Cruzain		TbrCatL		Cruzain	TbrCatL
			NI	I	NI	I
42	Triazole	4-Cl	52 ± 10	72 ± 2	73 ± 15	91 ± 0	nd	16.5 ± 0.5
43		2-Cl,4-Br	16 ± 1	6 ± 6	25 ± 12	98 ± 3	nd	2.5 ± 0.5
44	Furan	4-Cl	19 ± 19	94 ± 6	14 ± 14	96 ± 4	1.0 ± 0.1	0.04 ± 0
45		2-Cl,4-Br	17 ± 5	70 ± 23	22 ± 8	90 ± 10	0.45 ± 0.25	0.05 ± 0

nd: not determined.

^aCompounds were evaluated at the concentration of 100 μM with (I) and without (NI) a 10 min pre-incubation period with the enzymes. Assays were performed in triplicate in two independent experiments, and the data were expressed as means with standard error.

^bFrom at least two independent curves using at least seven compound concentrations in triplicate. Experiments were performed after a 10 min pre-incubation with the enzymes.

Fig. 3.

IC₅₀ determination for compounds 42–45 in different conditions. The IC₅₀ values were determined in increasing concentrations of Triton X-100 (0%, 0.01%, and 0.1%, represented by blue, black, and gray curves, respectively). The experiment with 0% Triton X-100 was also performed with a 10-min pre-incubation of the compounds with 4 mg/mL BSA (red curves). At least seven different concentrations of each compound were incubated for 10 min with the enzyme prior to the substrate addition. Curves determined in the presence of 0% Triton had R² values above 0.86, in the presence of 0.01% Triton above 0.90, in the presence of 0.1% Triton above 0.98, and upon pre-incubation with BSA above 0.89.

Thus, although initial results suggested that incorporating the thiosemicarbazone moiety significantly improved activity against both parasite proteases (Table 2), careful biochemical characterization revealed these compounds as artifactual inhibitors. Intriguingly, these aggregators affect the enzymes differently, with higher IC₅₀ values observed for cruzain, despite the identical assay conditions and enzyme concentration. For a given compound (44), IC₅₀ values against cruzain were up to 25-fold higher than the corresponding values against TbrCatL (Table 2, Supplementary Figs. S1 and S2). Based on these observations, we asked whether the compounds could have a dual behavior, acting as aggregators and retrieving specific activity against cruzain, as previously reported [46]. Thus, to assess for competitive behavior, we evaluated compounds 44 and 45 against cruzain in different substrate concentrations (0.156–10 μM). For a competitive inhibitor, enzyme inhibition should be lower at higher substrate concentrations. However, we did not observe competitive behavior for compounds 44 and 45, in agreement with inhibition by colloidal aggregation (Supplementary Material, Fig. S5).

2.4. T. cruzi and T. brucei assay

Even though the synthesized thiosemicarbazones are aggregators, this compound class has been widely studied as trypanocidal, frequently without an associated molecular target [47–51]. Thus, we investigated the trypanocidal activity of the synthesized compounds using in vitro assays against bloodstream forms of T. b. brucei Lister 427 and mouse myoblasts infected with Trypanosoma cruzi CA-I/72 [52,53]. In assays against T. b. brucei, using pentamidine as a positive control, six compounds were active in the micromolar range (EC₅₀ 13–49.7 μM). In the T. cruzi assays, in which the drug benznidazole was evaluated as the positive control, compounds 39, 44, and 45 showed EC₅₀ values < 1 μM (Table 3). Thiosemicarbazone 45, with an EC₅₀ value of 27 nM against T. cruzi, was 200-fold more potent than the positive control and reference drug benznidazole. In addition, it presented a high selectivity index (>370) when tested in myoblast cells, making it a standout compound compounds tested up to 10 μM concentration. for future structural improvement and in vivo studies.

Table 3.

Evaluation of in vitro activity against T. cruzi and T. b. brucei.

Compound	EC50 (μM)
	T. b. brucei	T. cruzi (SI)a
32	>50	>10
33	>50	>10
34	>50	>10
35	>50	>10
36	49.7 ± 5.8	>10
37	13.0 ± 3.2	>10
38	>50	>10
39	24.4 ± 7.7	0.59 ± 0.0041 (>17)
40	39.5 ± 16.8	>10
41	>50	>10
42	>50	>10
43	>50	>10
44	27.0 ± 8.7	0.56 ± 0.2 (>18)
45	>50	0.027 ± 0.0004 (>370)
Pentamidine	0.11 ± 0.04	–
Benznidazole	–	5.4 ± 0.3 (>9)

^aSI: selectivity index determined as CC₅₀/EC₅₀ using mouse myoblasts, with compounds tested up to 10 μM concentration.

2.5. Synthesis

Compounds were synthesized from reactions already described in the literature, as shown in Scheme 1. The initial step consisted of converting anilines into the corresponding azides via diazotization with sodium nitrite for further coupling with propargyl alcohol [54]. This 1,3-dipolar cycloaddition catalyzed by copper salts is considered a click reaction for being fast, high yielding, and regioselective [55]. Indeed, we obtained the desired alcohols with good yields (79–96%).

Scheme 1.

Synthesis of 1,4-substituted triazoles^a

^aReagents and conditions: (a) i) 3 M HCl, 0 °C, 10min ii) NaNO₂, 0 °C, H₂O, 30 min, iii) NaN₃, H₂O, 5 °C, 1–3h; (b) propargyl alcohol, sodium ascorbate, CuSO₄.5H₂O, THF/t-butanol/distilled water (1:1:1), rt, overnight; (c) MnO₂, THF, 60 °C, overnight; (d) i) Na₂CO₃, KMnO₄, 85 °C, 1h, ii) NHS/EDC, CH₂Cl₂, rt, overnight, iii) ethanolamine, CH₂Cl₂, rt, overnight; (e) i) Na₂SO₄, CH₂Cl₂, rt, overnight ii) NaBH₄ or Na(AcO)₃BH, corresponding amine, THF/metanol, 0 °C-rt, 4h, iii) ethereal HCl; (f) thiosemicarbazide, glacial acetic acid, ethanol, reflux, 1–2h.

For the amine coupling, we used a reductive alkylation reaction. Thereby we had to convert the alcohol group to an aldehyde using freshly prepared MnO₂ as oxidant [56]. The reducing agent used was either NaBH₄ or Na(AcO)₃BH [57]. The formation of the products could be confirmed via ¹H NMR by the distinct peak of methylene protons at around 4 ppm (integral for two hydrogens) and the peaks relative to the hydrogens of corresponding aldehydes and amines used as reagents. In parallel, an amide was synthesized by oxidizing the triazole alcohol to carboxylic acid and coupling it with ethanolamine [58].

The synthetic route employed for furan derivatives is depicted in Scheme 2 and involves a two-step procedure consisting of a Meerwein reaction with 2-furancarboxaldehyde followed by reductive alkylation [30]. The final compounds were treated with an ethereal HCl solution to obtain the products as dihydrochloride salts. Thiosemicarbazones 42–45 were obtained straightforwardly by reacting the corresponding aldehyde with thiosemicarbazide [59]. ¹H NMR spectra of compound 43 revealed that it was obtained as a mixture of E/Z isomers. Analyzing the integrals of all peaks, we estimated a proportion of 70% of the E and 30% of the Z isomer. The signal at 12.06 ppm, corresponding to N–H in the Z isomer, is more deshielded, probably due to an intramolecular hydrogen bond with N3 of the triazole ring, which is only possible in the Z isomer [60].

Scheme 2.

Synthesis of 2,5-substituted furans^a

^aReagents and conditions: (a) i) HCl 15% v/v, 0 °C, 10min ii) NaNO₂, 0 °C, H₂O, 30min, iii) 2-furancarboxaldehyde, CuCl₂.2H₂O, rt, H₂O, overnight; b) i) Na₂SO₄, CH₂Cl₂, rt, overnight ii) NaBH₄, corresponding amine, THF/metanol, 0 °C-rt, 4h, iii) ethereal HCl; c) thiosemicarbazide, glacial acetic acid, ethanol, reflux, 1–2h.

3. Discussion

Here, starting from a previously characterized cruzain and TbrCatL competitive inhibitor, we designed analogs for synthesis, evaluation in biochemical assays against the target enzymes, and in vitro assays against T. cruzi and T. brucei. Two key points from this study merit detailed discussion: the characterization of the synthesized compounds in biochemical assays, which revealed that an apparent optimization of enzyme inhibition was due to aggregation, and the discovery of potent novel trypanocidal compounds with high selectivity indices.

After two rounds of synthesis and biochemical assays, without improvement of potency when compared to MB12, we synthesized thiosemicarbazones derivatives. Although the IC₅₀ values of these derivatives were up to 650 times lower than that of MB12, further assays revealed that such inhibition was an artifact due to aggregation. Previous studies of HTS hits revealed that aggregation is the most common source of false positives in biochemical assays [61–63]. Although many assays provide evidence of aggregation, several of them are inconclusive, making it sometimes challenging to identify aggregators [38]. It is also important to highlight that a given compound may cause enzyme inhibition both through aggregation and legitimate specific inhibition. Various drugs currently on the market aggregate when evaluated at high concentrations [64]. Additionally, examples of compounds presenting a competitive mechanism and aggregator properties at similar concentrations have been reported [46,65]. Furthermore, aggregation is a condition-dependent phenomenon. Follow-up assays with a set of aggregators identified in an HTS against AmpC β-lactamase indicated that while most of them showed aggregator behavior in assays with cruzain, approximately a third of these compounds failed to do so, even at high concentrations [45]. Similarly, surface plasmon resonance experiments indicated the condition dependency of aggregation [66].

In addition to the existing difficulties associated with characterizing colloidal aggregates, we further highlight the challenges involved in evaluating our synthesized compounds. It is worth noting that our lead was previously validated as a competitive cruzain inhibitor [30]. Furthermore, thiosemicarbazones have been widely described as cruzain inhibitors [39,40,42,67]. Thus, we had a very plausible foundation for the improved enzyme inhibition upon the addition of the thiosemicarbazone moiety, supported by the historical background for this compound class and the expected covalent binding mode. Moreover, our docking studies indicated high complementarity between the enzymes and the compounds synthesized (Supplementary Material Fig. S3). In addition, because aggregation is a concentration-dependent phenomenon, it is less likely to occur at nanomolar compound concentrations, although potent compounds have been reported as aggregators [65]. Finally, we did not observe some frequent characteristics of colloidal aggregators for our compounds, such as high Hill slopes and promiscuous inhibition [38,45]. In this context, the determination of full IC₅₀ curves at different Triton X-100 concentrations and in the presence of BSA, as opposed to single concentration evaluations in these assays, were decisive to confidently assign the synthesized thiosemicarbazones as aggregators.

Whereas thiosemicarbazones have been described as cruzain and TbrCatL inhibitors, compounds from this class may also have a trypanocidal effect regardless of inhibition of the protease targets. To date, at least 21 studies have reported thiosemicarbazones with trypanocidal activity against T. cruzi [40,42,47–51,68–81], among which ten evaluated cruzain inhibition as a possible mechanism for the antiparasitic activity [40,42,47–51,77,79,80]. Although five of these studies reported cruzain inhibitors [40,42,77,79,80], in the other five, the trypanocidal compounds did not cause significant cruzain inhibition, indicating that other biochemical targets must be responsible for the anti-T. cruzi activity [47–51]. We also found eight original papers describing thiosemicarbazones that were active against T. brucei [42,76,82–88], three of which included assays against TbrCatL and characterized the trypanocidal compounds as inhibitors of this protease [42,82,86]. It is also worth noting that, among the studies that included assays against cruzain and TbrCatL, several did not include Triton X-100 in the buffer [42,47,49,79,80,82,86], six employed 0.01% Triton X-100 [40,48,50,51,77,85], and only two investigated whether inhibition was different in assays with or without 0.01% Triton X-100 [48,79]. Altogether, the evidence indicates that while trypanocidal thiosemicarbazones might inhibit trypanosomal proteases, other targets must also be considered concerning the antiparasitic activities measured.

The fact that the compounds designed in our study are not specific cruzain and TbrCatL inhibitors, as initially intended, does not reduce the value of their trypanocidal properties and potential for drug development. We highlight the potent trypanocidal activity observed for the compounds reported here, including compound 45, with an EC₅₀ of 0.027 ± 0.0004 μM against T. cruzi and a very high selectivity index (SI > 370) when compared to the CC₅₀ in mouse myoblasts. This SI is well above the recommended cut-off (SI > 50) for drug development for Chagas diseases [89]. For comparison, although different methods were used to determine a SI, we found thiosemicarbazones that meet that criterion in only eight publications [47,50,69,70,76,78,81,84]. A total of nine compounds were as selective as ours (SI > 370) [47,70], eight from the same study [47], which calculated SI based on the percentage of erythrocytes that suffered lysis at 25 μM.

4. Conclusions

In this study, we synthesized and evaluated the biological activity of MB12 analogs. Bioisosteric replacement of the MB12 furan ring with a triazole decreased potency against the protease targets, motivating the synthesis of new furan derivatives. Thiosemicarbazone derivatives were, apparently, potent inhibitors, but follow-up assays demonstrated that the potency was due to colloidal aggregation. Although a molecular target has not yet been characterized for the thiosemicarbazones described here, five of them showed activity in the T. b. brucei in vitro assay and three were active against T. cruzi, including compound 45, which was 200-fold more potent than the reference drug benznidazole with a SI > 370.

5. Material and methods

5.1. Enzymatic assays

Cruzain and TbrCatL activities were measured as previously described [41]. All assays were performed in triplicate in 96-well black flat-bottom plates, using 0.1 M sodium acetate buffer (pH 5.5) with 5 mM beta-mercaptoethanol, 0.01% Triton X-100, 4 nM of the enzyme, and 2.5 μM of the fluorogenic substrate Z-Phe-Arg-aminomethylcoumarin (Z-FR-AMC). E64 at 1 μM was used as an inhibition control. These measurements used 340 nm excitation and 440 nm emission filter sets in a Biotek 87 Synergy 2 fluorimeter. Unless otherwise specified, assays were monitored for 5 min. The enzyme activity was calculated based on initial velocity rates compared with a DMSO control.

Initially, each compound was screened at 100 μM with and without a 10-min pre-incubation with the enzyme at room temperature. Compounds with a high difference in the inhibitory activity after the 10-min incubation were submitted to time-turnover assays. The compounds were incubated with the enzymes for 0, 5, 10, and 15 min and had their activity measured for 3 min. For IC₅₀ curve determination, the compounds that presented a time-dependence behavior on the screen had the experiments performed after a 10-min pre-incubation with the enzymes. The other compounds had their IC₅₀ values determined without pre-incubation with the enzymes. IC₅₀ values were calculated based on at least two independent curves. Each curve was constructed based on at least seven compound concentrations in triplicate. The data were analyzed with GraphPad Prism 6.0, employing a nonlinear regression analysis of log [inhibitor] versus response with a variable slope and four parameters. We acquired enough data to reach an upper plateau around 100% inhibition whenever possible. However, this was not possible in some cases due to limited compound solubility. Due to the lack of a well-defined upper plateau and to assure proper calculation of IC₅₀ values, most curves had their top values constrained to less than 100. No constraints were added to fit the curves of compounds 40, 44, and 45 in the presence of cruzain and of compound 40 in the presence of TbrCatL. The curve of compound 37 in the presence of TbrCatL had its top constrained to less than 100 and its bottom to over 0.

To verify the effect of Triton X-100 on enzyme inhibition, compounds at a concentration close to their respective IC₅₀ were pre-incubated with a solution containing cruzain/TbrCatL, either in the absence of Triton X-100 or in the presence of 0.01% or 0.1% of this detergent, for 10 min and assayed as described above. In addition, selected compounds were also evaluated in the presence of BSA and the absence of Triton X-100. To do so, compounds were pre-incubated with 4 mg/mL BSA for 10 min to verify if enzyme inhibition was sensitive to pre-incubation with another protein. Then, cruzain/TbrCatL was added to a final concentration of 8 nM, and the mixture was incubated for additional 10 min. After that, the substrate was added to a final concentration of 2.5 μM, and the fluorescence was monitored. Compounds were then classified as insensitive or sensitive to Triton/BSA if the inhibition observed in these conditions showed a variation of up to 20%, or above this value, respectively. In the case of compounds 42–45, instead of comparing single concentrations at the conditions above, full IC₅₀ curves were determined at different Triton X-100 concentrations and upon pre-incubation with BSA. All assays were performed in two independent experiments in triplicate.

To verify competitive behavior by compounds 44 and 45, 4 nM of cruzain was incubated with each compound at 1 and 0.25 μM for 10 min at room temperature. Then, the substrate was added at seven different concentrations: 0.156, 0.312, 0.625, 1.25, 2.5, 5, and 10 μM. The plates were read at 25 °C for 5 min. All assays were performed in triplicate.

For assays against Zika virus NS2B-NS3, compounds 44 and 45 at 0.78 μM were assayed in 10 mM Tris-HCl buffer with 20% glycerol, 1 mM CHAPS, pH 8.5 at 37 °C, in the presence of 0.2 nM NS2B-NS3 and 44 μM Bz-Nle-Lys-Lys-Arg-AMC substrate, in flat bottom 96 wells black plates. The fluorescence was monitored for 10 min using 340/440 nm filters for excitation and emission after pre-incubation of the compounds with NS2B-NS3 for 10 min at 37 °C. The inhibitory activity in the presence of the compounds was calculated in comparison to the initial velocities of the enzyme in a DMSO control. All enzyme assays were performed in a Synergy 2 plate reader (Biotek) at the Multiuser Equipment Laboratory of the Biochemistry and Immunology Department of Universidade Federal de Minas Gerais.

5.2. Literature search of thiosemicarbazones with trypanocidal activity

DataWarrior (OpenMolecules) was initially used to retrieve trypanocidal thiosemicarbazones from the ChEMBL database by a SMARTS search. In addition, a PubChem BioAssay search (https://www.ncbi.nlm.nih.gov/pcassay) using parasites’ taxonomy names and a thiosemicarbazone filter was performed to find any left-out publication. All reviews and compounds not considered trypanocidal were excluded. Structures and assays were grouped by publication. All publications were analyzed to gather assay details, including protease inhibition assays. Compounds were considered active against the enzymes using an IC₅₀ cut-off of 100 μM or a 50% activity inhibition when the IC₅₀ was not determined.

5.3. Synthesis

5.3.1. Krake et al. compound set

The synthesis and characterization of the 31 compounds from the Krake et al. set were reported in a previous publication [33]. Briefly, compounds were characterized by ¹H and proton-decoupled ¹³C NMR spectra and by high-resolution mass spectrometry (HRMS), using the electrospray ionization (ESI) technique.

5.3.2. Compounds 32–45

All reagents of analytical grade were obtained from commercial suppliers and used without further purification. The melting points were determined on a Microquímica MQAPF 301 apparatus. The FT-IR spectra were recorded using a PerkinElmer Spectrum One infrared spectrometer, and absorptions are reported as wave numbers (cm⁻¹). The NMR spectra were recorded on a Bruker AVANCE DRX400 instrument, using tetramethylsilane (TMS) as the internal standard or solvent peak(s) of CDCl₃ (7.27 and 77.1 ppm) or CD₃OD (3.31 and 49.0 ppm). Data for ¹H NMR spectra are reported as follows: chemical shift (δ ppm), multiplicity, coupling constant (Hz), and integration. Splitting patterns are designated as follows: s, singlet; d, doublet; dd, doublet of doublets; m, multiplet; br, broad. HRMS were acquired using a LCMS-Q-ORBITRAP mass spectrometer and the samples were solubilized in MeOH, following manual injection.

General procedure for preparation of azides [54].

In a round-bottom flask containing 1 equiv of the corresponding aniline 1 mL 3 M HCl solution per aniline mmol was added under an ice bath. After 10 min, 1.5 equiv of NaNO₂ in a 1 M aqueous solution was added dropwise, not allowing the temperature higher than 0 °C, stirring the reaction for additional 30 min. The NaN₃ (2 equiv–0.8 M aqueous solution) was added dropwise, keeping the temperature below 5 °C during the addition. The ice bath was removed, and the reaction stirred for 1–3 h, when it was extracted with ethyl acetate and washed with a 0.3 M HCl solution, followed by distilled water. After solvent removal, the azides obtained were immediately used in the next step without previous purification.

General procedure for preparation of triazoles alcohol intermediates. [55]

A round-bottom flask containing 1 equiv of the corresponding azide, 1.5 equiv of propargyl alcohol, 0.15 equiv of sodium ascorbate, and 0.05 CuSO₄.5H₂O in a 1:1:1 solution of THF/t-butanol/distilled water was magnetically stirred at room temperature overnight. The mixture was quenched with distilled water and extracted with ethyl acetate, followed by washing with distilled water. The organic layer was collected, dried, and the crude product was obtained after solvent removal.

General procedure for preparation of triazole aldehydes intermediates. [56]

In a round-bottom flask containing 1 equiv of the corresponding triazole alcohol in THF, 6 equiv of freshly prepared and overnight dried MnO₂ was added. The mixture was magnetically stirred at 60 °C and an additional quantity of MnO₂ was added (up to 3 equiv) if starting material was not consumed overnight. The inorganic content was filtered off using an 8 μm pore-size filter, washing with THF and methanol. The solvent mixture was evaporated under reduced pressure, affording the desired aldehydes after column chromatography purification (hexane/ethyl acetate 8:2).

General procedure for preparation of furfural derivatives. [30]

A round-bottom flask containing 1 equiv of the corresponding aniline was kept under an ice bath and 10 mL 15% v/v HCl solution was added and magnetically stirred for 10 min. Then, 1.4 equiv of NaNO₂ in distilled water was added dropwise, while keeping the reaction temperature under 0 °C. Next, 1.2 equiv of furfuraldehyde in THF, followed by 0.15 equiv of CuCl₂.2H₂O in distilled water were added. The ice bath was removed and the reaction was kept at room temperature magnetically stirring overnight. The reaction was extracted with DMC and washed with distilled water and a saturated NaHCO₃ solution.

General procedure for preparation of amines. [57]

A round-bottom flask containing 1 equiv of the corresponding aldehyde and 10 equiv of Na₂SO₄ in dichloromethane was magnetically stirred at room temperature overnight. Then, the solvent was replaced by a 1:1 THF/methanol solution and 2–4 equiv of NaBH₄ or Na(AcO)₃BH was added under an ice bath. Alternatively, 1 mL of methanol was added instead. The reaction was kept at room temperature for 2–4 h and quenched, under an ice bath, with distilled water followed by a 1 M NaOH solution until alkaline pH. The mixture was filtered, the solvent partially removed at low pressure, and the crude extracted with ethyl acetate, washing with distilled water.

2-(((1-(4-chlorophenyl)-1,2,3-triazol-4-yl)methyl)amino)ethan-1-ol hydrochloride (32):

Using 3 equiv of amine and 2 equiv of NaBH₄ in THF/methanol after 2 h reaction. The crude product was purified by neutral alumina column chromatography (dichloromethane/methanol 2% + NH₄OH) and acidified with a 3 M HCl solution, affording the hydrochloride salt as a white solid. Yield: 20.0%. MP: 240.2–243.3 °C. IR (neat, cm⁻¹): 3301, 3145, 2919, 2853, 1499, 1441, 1045. ¹H NMR (400 MHz, MeOD-d₄ + DMSO-d₆): δ 8.48 (s, 1H), 7.89 (d, J = 8.9 Hz, 2H), 7.63 (d, J = 8.9 Hz, 2H), 4.01 (s, 2H), 3.71 (t, J = 5.5 Hz, 2H), 2.83 (t, J = 5.5 Hz, 2H). ¹³C NMR (101 MHz, MeOD-d₄ + DMSO-d₆) δ 148.2, 137.1, 135.4, 131.0, 122.9, 122.4, 61.5, 51.8, 44.7. HRMS-ESI m/z: calcd for C₁₁H₁₄ClN₄O⁺ [M + H]⁺: 253.0851; found: 253.0840.

2-(((1-(4-bromo-2-chlorophenyl)-1,2,3-triazol-4-yl)methyl)amino) ethan-1-ol hydrochloride (33):

Using 3 equiv of amine and 4 equiv of Na (AcO)₃BH in THF/methanol after 48 h reaction. The crude product was purified by washing it with dichloromethane after acidification to pH 1 with a 1 M HCl solution, followed by alkalization to pH 12 of the aqueous layer with 1 M NaOH solution and extraction with dichloromethane. The resulting organic layer was dried with Na₂SO₄ and the solvent was removed under low pressure. Acidification with a 3 M HCl solution afforded the hydrochloride salt as a white solid. Yield: 36.4%. MP: 206.4–208.0 °C. IR (neat, cm⁻¹): 3375, 3168, 3046, 2924, 1579, 1490, 1048. ¹H NMR (400 MHz, MeOD-d₄): δ 8.51 (s, 1H), 7.96 (d, J = 2.0 Hz, 1H), 7.76 (dd, J = 8.5, 2.0 Hz, 1H), 7.59 (d, J = 8.5 Hz, 1H), 4.44 (s, 2H), 3.86 (d, J = 5.3 Hz, 1H), 3.20 (t, J = 5.3 Hz, 2H). ¹³C NMR (101 MHz, MeOD-d₄) δ 140.9, 134.9, 134.4, 132.8, 131.4, 130.4, 128.6, 125.7, 58.4, 50.3, 42.8. HRMS-ESI m/z: calcd for C₁₁H₁₃BrClN₄O⁺ [M + H]⁺: 330.9956; found: 330.9953.

2-(((1-(3,4-dichlorophenyl)-1,2,3-triazol-4-yl)methyl)amino)ethan-1-ol hydrochloride (34):

Using 3 equiv of amine and 4 equiv of Na(AcO)₃BH in THF/methanol after 72 h reaction. The crude product was purified by washing it with dichloromethane after acidification to pH 1 with a 1 M HCl solution, followed by alkalization to pH 12 of the aqueous layer with 1 M NaOH solution and extraction with dichloromethane. The resulting organic layer was dried with Na₂SO₄ and the solvent was removed under low pressure. Acidification with a 3 M HCl solution afforded the hydrochloride salt as a white solid. Yield: 71.0%. MP: 222.6–226.2 °C. IR (neat, cm⁻¹): 3405, 3150, 2928, 1591, 1486, 1241, 1027. ¹H NMR (400 MHz, MeOD-d₄): δ 8.75 (s, 1H), 8.13 (d, J = 2.4 Hz, 1H), 7.85 (dd, J = 8.7, 2.4 Hz, 1H), 7.78 (d, J = 8.7 Hz, 1H), 4.52 (s, 2H), 3.89 (d, J = 5.1 Hz, 1H), 3.28 (t, J = 5.1 Hz, 2H). ¹³C NMR (101 MHz, MeOD-d₄) δ 140.6, 137.1, 134.7, 134.1, 132.9, 125.3, 123.5, 121.3, 57.8, 50.0, 42.5. HRMS-ESI m/z: calcd for C₁₁H₁₃Cl₂N₄O⁺ [M + H]⁺: 287.0461; found: 287.0459.

1-(1-(4-chlorophenyl)-1,2,3-triazol-4-yl)-N-(piperidin-4-ylmethyl) methanamine dihydrochloride (36):

Using 3 equiv of amine and 4 equiv of NaBH₄ in THF/methanol after 2 h reaction. The crude product was purified by neutral alumina column chromatography (dichloromethane/methanol 2% + NH₄OH) and acidified with a 3 M HCl solution, affording the hydrochloride salt as a white solid. Yield: 32.5%. MP: 206.4–208.0 °C. IR (neat, cm⁻¹): 3380, 3140, 2924, 2853, 1627, 1594, 1503, 1450, 1402. ¹H NMR (400 MHz, MeOD-d₄): δ 8.47 (s, 1H), 7.88 (d, J = 8.8 Hz, 2H), 7.62 (d, J = 8.8, 2H), 3.78 (s, 2H), 3.66 (s, NH), 3.62 (t, J = 5.6 Hz, NH), 3.04 (d, J = 11.7 Hz, 2H), 2.70–2.60 (m, 4H), 2.17 (t, J = 11.7 Hz, 2H), 1.80 (d, J = 12.4 Hz, 2H), 1.40–1.20 (m, 1H). ¹³C NMR (101 MHz, MeOD-d₄) δ 145.6, 137.8, 135.6, 131.0, 123.6, 122.9, 60.8, 58.3, 54.0, 53.7, 47.6, 38.3, 30.5. HRMS-ESI m/z: calcd for C₁₅H₂₁ClN₅⁺ [M + H]⁺: 306.1480; found: 306.1478.

2-(((5-(4-bromo-2-chlorophenyl)furan-2-yl)methyl)amino)ethan-1-ol hydrochloride (37):

Using 10 equiv of amine and 4 equiv of NaBH₄ in dichloromethane/methanol the product was obtained as a white solid after 2.5 h and acidification with 3 M HCl followed by successive diethyl ether washings. Yield: 65%. MP: 139.0–141.0 °C. IR (neat, cm⁻¹): 3200–2000, 3152, 2951, 1582, 1549, 1466, 1086. ¹H NMR (400 MHz, MeOD-d₄): δ 7.86 (d, J = 8.6 Hz, 1H), 7.71 (d, J = 2.0 Hz, 1H), 7.57 (dd, J = 8.6, 2.0 Hz, 1H), 7.18 (d, J = 3.5 Hz, 1H), 6.82 (d, J = 3.5 Hz, 1H), 4.42 (s, 2H), 3.84 (t, J = 5.3 Hz, 2H), 3.20 (t, J = 5.3 Hz, 2H). ¹³C NMR (101 MHz, MeOD-d₄) δ 152.2, 146.5, 134.3, 132.2, 131.7, 130.6, 128.8, 122.8, 115.8, 113.6, 57.8, 50.0, 44.0. HRMS-ESI m/z: calcd for C₁₃H₁₄BrClNO₂⁺ [M + H]⁺: 329.9891; found: 329.9886.

2-(((5-(3,4-dichlorophenyl)furan-2-yl)methyl)amino)ethan-1-ol hydrochloride (38):

Using 10 equiv of amine and 4 equiv of NaBH₄ in dichloromethane/methanol the product was obtained as an amber oil after 3 h and acidification with 3 M HCl followed by successive diethyl ether washings. Yield: 31.5%. IR (neat, cm⁻¹): 3400–2400, 3344, 2952, 1583, 1530, 1462, 1027. NMR (400 MHz, MeOD-d₄): δ 7.93 (d, J = 1.8 Hz, 1H), 7.67 (dd, J = 8.4, 1.8 Hz, 1H), 7.56 (d, J = 8.4 Hz, 1H), 6.95 (d, J = 3.4 Hz, 1H), 6.77 (d, J = 3.4 Hz, 1H), 4,40 (s, 2H), 3.84 (t, J = 5.1 Hz, 2H), 3.19 (t, J = 5.1 Hz, 2H). ¹³C NMR (101 MHz, MeOD-d₄) δ 154.5, 146.8, 134.0, 132.6, 132.1, 131.6, 126.8, 124.7, 116.1, 109.1, 57.8, 49.9, 44.1. HRMS-ESI m/z: calcd for C₁₃H₁₂Cl₂NO₂⁺ [M − H]⁺: 284.0240; found: 284.0237.

N-((5-(4-bromo-2-chlorophenyl)furan-2-yl)methyl)-2-morpholinoethan-1-amine dihydrochloride (39):

Using 3 equiv of amine and 4 equiv of NaBH₄ in dichloromethane/methanol the product was obtained as a white solid after 4 h and acidification with 3 M HCl followed by successive diethyl ether washings. Yield: 60.6%. MP: 229.3–230.8 °C. IR (neat, cm⁻¹): 3100–2300, 3090, 2985, 2965, 1594, 1552, 1467, 1110. ¹H NMR (400 MHz, D₂O): δ 7.77 (m, 2H), 7.60 (d, J = 8.0 Hz, 1H), 7.22 (s, 1H), 6.88 (s, 1H), 4.54 (s, 2H), 4.03 (s, 4H), 3.66 (s, 4H), 3.48 (s, 4H). ¹³C NMR (101 MHz, CDCl₃) δ 151.5, 143.9, 133.3, 130.8, 130.6, 129.4, 127.2, 121.7, 115.5, 112.6, 64.0, 52.6, 52.1, 43.8, 40.3. HRMS-ESI m/z: calcd for C₁₇H₂₁BrClN₂O₂⁺ [M + H]⁺: 399.0469; found: 399.0465.

2-(((5-(4-bromo-2-chlorophenyl)furan-2-yl)methyl)amino)propane-1,3-diol hydrochloride (40):

Using 3 equiv of amine and 4 equiv of NaBH₄ in dichloromethane/methanol the product was obtained as a white solid after 3 h and acidification with 3 M HCl followed by successive diethyl ether washings. Yield: 40.7%. MP: 117.0–117.3 °C. IR (neat, cm⁻¹): 3400–2400, 3368, 3300, 2967, 2923, 1628, 1581, 1465, 1030. ¹H NMR (400 MHz, MeOD-d₄): δ 7.86 (d, J = 8.6 Hz, 1H), 7.70 (d, J = 1.8 Hz, 1H), 7.57 (dd, J = 8.6, 1.8 Hz, 1H), 7.18 (d, J = 3.4 Hz, 1H), 6.82 (d, J = 3.4 Hz, 1H), 4,50 (s, 2H), 3.91–3.77 (m, 4H), 3.35–3.30 (m, 1H). ¹³C NMR (101 MHz, MeOD-d₄) δ 152.1, 146.7, 134.2, 132.1, 131.6, 130.6, 128.8, 122.7, 115.8, 113.6, 61.2, 59.1, 42.2. HRMS-ESI m/z: calcd for C₁₄H₁₆BrClNO₃⁺ [M + H]⁺: 359.9997; found: 359.9991.

2-(((5-(4-chloro-3-(trifluoromethyl)phenyl)furan-2-yl)methyl)amino) propane-1,3-diol hydrochloride (41):

Using 3 equiv of amine and 4 equiv of NaBH₄ in dichloromethane/methanol the product was obtained as a white solid after 1 h and acidification with 3 M HCl followed by successive diethyl ether washings. Yield: 31.5%. MP: 80.4–81.7 °C. IR (neat, cm⁻¹): 3400–2400, 3190, 2966, 2829, 1617, 1479, 1328, 1126. ¹H NMR (400 MHz, MeOD-d₄): δ 8.13 (s, 1H), 7.98 (d, J = 8.4 Hz, 1H), 7.66 (d, J = 8.4 Hz, 1H), 7.03 (d, J = 3.4 Hz, 1H), 6.82 (d, J = 3.4 Hz, 1H), 4,52 (s, 2H), 3.92–3.79 (m, 4H), 3.36–3.30 (m, 1H). ¹³C NMR (101 MHz, MeOD-d₄) δ 154.2, 147.2, 133.4, 132.0, 130.8, 129.7, 123.9, 123.8, 116.3, 109.5, 61.3, 59.1, 42.4. HRMS-ESI m/z: calcd for C₁₅H₁₆ClF₃NO₃⁺ [M + H]⁺: 350.0765; found: 350.0761.

Procedure for preparation of amide derivative 1-(4-bromo-2-chlorophenyl)-N-(2-hydroxyethyl)-1,2,3-triazole-4-carboxamide (35).

[58] Initially, the carboxylic acid was prepared and used in the next step without purification by reacting 1 equiv of the corresponding alcohol, 0.6 equiv of Na₂CO₃ and 1.2 equiv of KMnO₄ at 85 °C for 1 h followed by acidification with HCl 3 M and filtration. The obtained carboxylic acid was added to a round-bottom flask containing 1.6 equiv of NHS and EDC in dichloromethane and magnetically stirred at room temperature overnight. Then, the reaction was quenched with distilled water, and the organic layer was separated and evaporated. To the obtained activated ester, a 5 equiv of ethanolamine in dichloromethane was added and the solution was magnetically stirred at room temperature overnight. The reaction mixture was acidified to pH 1 with HCl 3 M solution and extracted with dichloromethane. The crude solid was further purified by column chromatography (Hex/AcOEt 1:1 → AcOEt 100%). Yield: 65.0%. MP: 151.2–152.7 °C. IR (neat, cm⁻¹): 3312, 3117, 2923, 2852, 1640, 1583, 1490, 1029. ¹H NMR (400 MHz, MeOD-d₄): δ 8.69 (s, 1H), 7.89 (s, 1H), 7,70 (d, J = 8.2 Hz, 1H), 7.56 (d, J = 8.2 Hz, 1H), 3.72 (s, 2H), 3.54 (s, 2H), 3.29 (s, 1H). ¹³C NMR (101 MHz, MeOD-d₄) δ 162.1, 143.7, 134.8, 132.6, 131.3, 130.2, 129.1, 125.6, 61.5, 42.6. HRMS-ESI m/z: calcd for C₁₁H₁₁BrClN₄O₂⁺ [M + H]⁺: 344.9748; found: 344.9745.

General procedure for preparation of thiosemicarbazones. [59]

In a round-bottom flask containing 1 equiv of the corresponding aldehyde, 1.05 equiv of thiosemicarbazide and 3 drops of glacial acetic acid in ethanol were magnetically stirred under reflux for 1–2 h. The reaction was cooled down to room temperature and the solvent was partially removed at reduced pressure. The precipitate formed was filtered and washed with cold methanol.

2-((1-(4-chlorophenyl)-1,2,3-triazol-4-yl)methylene)hydrazine-1-carbothioamide (42):

Obtained as a white solid after 1 h reaction. Yield: 87.6%. MP: 224.0–225.0 °C. IR (neat, cm⁻¹): 3363, 3264, 3136, 2988, 1606, 1524, 1501, 1486, 1241, 1093. ¹H NMR (400 MHz, DMSO-d₆): δ 11.61 (s, 1H), 9.27 (s, 1H), 8.37 (s, 1H), 8.19 (s, 1H), 7.91 (d, J = 8.8 Hz, 2H), 7.70 (s, 1H), 7.50 (d, J = 8.8 Hz, 2H). ¹³C NMR (101 MHz, DMSO-d₆) δ 178.2, 144.9, 135.1, 133.3, 133.1, 129.9, 121.8, 121.3. HRMS-ESI m/z: calcd for C₁₀H₁₀ClN₆S⁺ [M + H]⁺: 281.0371; found: 281.0366.

2-((1-(4-bromo-2-chlorophenyl)-1,2,3-triazol-4-yl)methylene)hydrazine-1-carbothioamide (43):

Obtained a white solid as a isomer mixture after 2 h reaction. Yield: 83.3%. MP: 220.4–221.7 °C. IR (neat, cm⁻¹): 3372, 3281, 3268, 3190, 3112, 1640, 1629, 1522, 1479, 1302, 1041. ¹H NMR (400 MHz, DMSO-d₆): δ 12.06 + 11.60 (s, 1H), 9.04 + 9.03 (s, 1H), 8.64 + 8.33 (s, 1H), 8.22 (s, 1H), 8.17 + 8.14 (d, J = 2.0 Hz, 1H), 7.93 (s, 1H), 7.88 + 7.85 (dd, J = 8.5, 2.0 Hz, 1H), 7.78 + 7.74 (d, J = 8.5 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆) δ 178.1, 144.1, 141.4, 133.5, 133.1, 132.9, 131.6, 129.9, 129.6, 129.5, 129.1, 125.9, 125.2, 123.9. HRMS-ESI m/z: calcd for C₁₀H₉BrClN₆S⁺ [M + H]⁺: 358.9476; found: 358.9472.

2-((5-(4-chlorophenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (44):

Obtained as a white solid after 1 h reaction. Yield: 24.8%. MP: 197.7–198.7 °C. IR (neat, cm⁻¹): 3398, 3362, 3244, 3156, 1599, 1536, 1473, 1289, 1091. ¹H NMR (400 MHz, DMSO-d₆): δ 11.52 (s, 1H), 8.30 (s, 1H), 7.98 (s, 1H), 7.86 (d, J = 8.3 Hz, 2H), 7.79 (s, 1H), 7.50 (d, J = 8.3 Hz, 1H), 7.17 (d, J = 3.3 Hz, 1H), 7.08 (d, J = 3.6 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆) δ 177.7, 154.3, 149.3, 132.5, 131.7, 128.9, 128.4, 125.6, 115.3, 108.9. HRMS-ESI m/z: calcd for C₁₂H₁₁ClN₃OS⁺ [M + H]⁺: 280.0306; found: 280.0302.

2-((5-(4-bromo-2-chlorophenyl)furan-2-yl)methylene)hydrazine-1-carbothioamide (45):

Obtained as a white solid after 2 h reaction. Yield: 60.4%. MP: 215.8–217.2 °C. IR (neat, cm⁻¹): 3432, 3283, 3153, 1618, 1597, 1540, 1506, 1460, 1348, 1287. ¹H NMR (400 MHz, DMSO-d₆): δ 11.56 (s, 1H), 8.33 (s, 1H), 8.10–7.95 (m, 2H), 7.86–7.82 (m, 2H), 7.65 (dd, J = 8.6, 2.0 Hz, 1H), 7.33 (d, J = 3.6 Hz, 1H), 7.14 (d, J = 3.6 Hz, 1H). ¹³C NMR (101 MHz, DMSO-d₆) δ 177.8, 149.5, 149.4, 132.9, 131.4, 130.7, 130.0, 129.5, 126.9, 121.1, 114.7, 113.8. HRMS-ESI m/z: calcd for C₁₂H₁₀BrClN₃OS⁺ [M + H]⁺: 357.9411; found: 357.9405.

5.3.3. T. brucei assay

Cell culture:

Bloodstream forms of T. b. brucei Lister 427 were cultured in a modified HMI-9 medium [90,91] in T25 vented flasks (Thermo Fisher Scientific; Cat N°. FB012933) in a humidified atmosphere of 5% CO₂ at 37 °C. Parasites were maintained in log-phase growth (between 1 × 10⁵ and 1 × 10⁶ parasites/mL) and passaged every 48 h.

In vitro SYBR Green assay: this assay was used to measure T. b. brucei cell viability [52,92]. For the in vitro screening, compounds were serially diluted to test eight concentrations ranging from 50 μM to 82 nM or from 50 μM to 15 nM. The standard drug pentamidine was serially diluted to test eight concentrations ranging from 4 μM to 0.87 nM. All compounds were added to the cultures as solutions in DMSO (Thermo Scientific^™ 85190 (Cat N°. PI85190). Trypanosomes in log-phase growth were suspended at 2 × 10⁵ trypanosomes/mL in a modified HMI-9 medium under continuous agitation and dispensed (100 μL/well) into 96-well plates containing 1 μL of the test compound in 0.5% DMSO and 100 μL/well of fresh medium (20,000 trypanosomes/well). Plates were incubated for 72 h at 37 °C and 5% CO₂, followed by the addition of lysis solution (50 μL/well) containing 1 × SYBR Green I (10,000 × in DMSO; Invitrogen, Carlsbad, CA). Plates were incubated in the dark for at least 1 h at room temperature, and fluorescence was measured in a 2104 EnVision^® multilabel plate reader (PerkinElmer, Waltham, MA), with excitation/emission at 485/535 nm. The activity of test compounds was normalized to DMSO controls from the same plate. Half maximal effective concentration (EC₅₀) values were calculated using GraphPad Prism software, version 6.00 for the MAC OS. Data are the means of two to three biological experiments with each compound being tested in technical duplicate.

5.3.4. T. cruzi assay

Antiparasitic activity and compound cytotoxicity were determined as in Boudreau et al. [53] Mouse myoblasts cell line C2C12 (ATCCCRL-1772) were maintained in Dulbecco’s modified Eagle medium (DMEM) containing 4.5 g/L glucose, supplemented with 1% penicillin, streptomycin 10,000 U/mL and 5% fetal bovine serum at 37 °C with 5% CO₂. Trypanosoma cruzi CA-I/72 was maintained by weekly coinfection with C2C12 cells. For all activity assays, C2C12 cells were infected at a 15:1 parasite-to-cell ratio with day six. T. cruzi tissue culture-derived trypomastigotes and treated with compounds for 72 h. Compounds were evaluated at 20 concentrations (from 10 μM to 19 pM). Plates were then fixed with 4% formaldehyde for at least 2 h and stained with 0.5 μg/mL of 4′,6-diamidino-2-phenylindole (DAPI) for at least 4 h prior to reading. Plates were imaged and analyzed with the automated ImageXpress MicroXL microscope (Molecular Devices). Infection levels (parasites per host cell) were normalized to the positive control (50 μM Benznidazole) and negative control (DMSO vehicle). IC (antiparasitic activity) and CC₅₀ (host cell cytotoxicity) values were determined using GraphPad Prism software. Assays were performed in duplicates.

Abbreviations:

CD

Chagas disease

HAT

Human African Trypanosomiasis

HTS

High Throughput Screening

BSA

Bovine serum albumin

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ejmech.2022.114876.

Data availability

Data will be made available on request.