Thio- and selenosemicarbazones as antiprotozoal agents against Trypanosoma cruzi and Trichomonas vaginalis

Abstract

Herein, we report the preparation of a panel of Schiff bases analogues as antiprotozoal agents by modification of the stereoelectronic effects of the substituents on N-1 and N-4 and the nature of the chalcogen atom (S, Se). These compounds were evaluated towards Trypanosoma cruzi and Trichomonas vaginalis. Thiosemicarbazide 31 showed the best trypanocidal profile (epimastigotes), similar to benznidazole (BZ): IC₅₀ (31)=28.72 μM (CL-B5 strain) and 33.65 μM (Y strain), IC₅₀ (BZ)=25.31 μM (CL-B5) and 22.73 μM (Y); it lacked toxicity over mammalian cells (CC₅₀ > 256 µM). Thiosemicarbazones 49, 51 and 63 showed remarkable trichomonacidal effects (IC₅₀ =16.39, 14.84 and 14.89 µM) and no unspecific cytotoxicity towards Vero cells (CC₅₀ ≥ 275 µM). Selenoisosters 74 and 75 presented a slightly enhanced activity (IC₅₀=11.10 and 11.02 µM, respectively). Hydrogenosome membrane potential and structural changes were analysed to get more insight into the trichomonacidal mechanism.

Graphical Abstract

1. Introduction

Protozoa are responsible for a large number of severe parasitic diseases in humans, livestock and pets, causing important morbidity and mortality worldwide¹. This constitutes a burden for health systems in low-medium income tropical and subtropical countries from Africa, Asia and Latin America, where some of these protozooses are endemic². Despite affecting more than one billion people, almost one sixth of the human population, and causing roughly half million deaths per year, many of these diseases are largely ignored, and indeed, some of them are categorised into the so-called Neglected Tropical Diseases (NTDs) group³.

Management of diseases caused by pathogenic protozoa is not a simple task⁴. On the one hand, development of successful vaccines is a hitherto unachieved goal⁵; on the other hand, the chemotherapeutic arsenal available so far suffers from important drawbacks: most of them are old drugs that are developing chemoresistance⁶, and are endowed with severe side-effects⁷ and low efficiency⁸. Moreover, their high prices, and complex administration protocols make them unaffordable for underdeveloped countries⁹. Accordingly, the development of new antiprotozoal agents is a hot topic in current Medicinal Chemistry research^10,11.

Herein, we have focussed our attention on Chagas disease and trichomoniasis. On the one hand, Chagas disease (aka American trypanosomiasis), discovered by the Brazilian physician Carlos R.J. das Chagas in 1909, is endemic in Latin America, where it affects roughly 7 million people¹²; nevertheless, it is also being spread to USA, Canada, Europe and Australia, because of human migrations¹³, and therefore, becoming a global health problem¹⁴. Chagas disease is caused by the haemoflagellate protozoan Trypanosoma cruzi, whose transmission to humans naturally takes place by contact with faeces or urine of infected triatomine insects. The infection can also be transmitted by non-vectorial routes, such as the iatrogenic and the congenital one¹². The life-cycle of the parasite involves three stages: epimastigotes (extracellular and replicative form found in the intestine of the vector), amastigotes (intracellular and proliferative form of the vertebrate host) and trypomastigotes (extracellular and non-replicative state found in the bloodstream)^15,16. Currently, there are only two available drugs for the specific treatment of Chagas disease: benznidazole (BZ), a nitroimidazole, and nifurtimox, a nitrofurane both showing some disadvantages (e.g., low efficacy in the chronic phase, adverse effects and parasite drug resistance) that constitute one of the main drawbacks of this parasitosis¹⁷.

Cruzipain, the main papain-like cysteine peptidase in T. cruzi is currently considered as a validated therapeutic target against Chagas disease¹⁸, and thus compounds inhibiting such enzyme constitute an interesting alternative to classical antichagasic drugs. Some thio-^19‒21 and selenosemicarbazones²² have been reported to be good inhibitors of such enzymes.

On the other hand, regarding the four curable sexually-transmitted infections (STI), three of them are caused by bacteria (chlamydia, gonorrhoea and syphilis), whereas the fourth one (trichomoniasis) is caused by a flagellated protozoan. In this context, a recent report has indicated²³ that within population ranging from 15 to 49 years, 376.4 million new cases of STI were estimated to appear each year, among which, 156.0 million cases of urogenital trichomoniasis are found. Human trichomoniasis, recently categorised as a neglected parasitic infection (NPI)²⁴, not only accounts for more than 40% of curable STI, but also represents a severe health risk²⁵, as it increases the susceptibility to HIV, human papilloma virus (HPV), or herpes simplex virus (HSV) infections, as well as cervical and prostate cancers²⁶. In this sense, some authors have demonstrated the positive impact of trichomoniasis treatment on the prevention of HIV transmission²⁷. Therefore, as the development of these neoplasia seems to be associated with the inflammatory response induced by the parasite²⁶, its diagnosis and treatment could also reduce the risk of their subsequent development. Moreover, and due to the fact that trichomoniasis is not a notifiable STI, with a high number of asymptomatic patients, epidemiologic data might be underestimated²⁸. Currently, there are only two drugs available for the treatment of trichomoniasis²⁹, both of them being nitroimidazole derivatives: metronidazole (MTZ), discovered in 1959, and still the first drug of choice, and tinidazole, approved in 2004. Due to the scarce number of commercialised drugs against trichomoniasis, when drug resistance, side-effects or hipersensibility to 5-nitroimidazoles emerge, no alternative treatments are available²⁹.

Herein, we have accomplished the preparation of an ample panel of thiosemicarbazones with the aim of developing antiprotozoal agents with a different mechanism of action than that exhibited by nitroheterocyclic derivatives. This type of Schiff bases analogues has been reported to be endowed with a broad spectrum of relevant biological properties, like inhibitors of aldose reductase³⁰, tyrosinase³¹, urease³² cholinesterases³³, or β-amyloid aggregation³⁴, and also as antimicrobial³⁵, or anticancer agents^36‒38, the latter being the most widely studied. Although there are some reports on the use of thiosemicarbazones as antiprotozoal agents against Toxoplasma gondii^39‒41 or T. cruzi^42‒45, studies on T. vaginalis are very scarce and limited to nitrofurane derivatives and bisthiosemicarbazones^46–49.

2. Materials and methods

2.1. General procedures

TLCs were performed using aluminium-coated sheets (Merck 60 F₂₅₄), 0.25 mm gel thickness. Each eluant is indicated in the experimental procedures. Spots were visualised by UV light (λ = 254 nm), and by charring with 10% ethanolic vainillin containing 1% H₂SO₄, or with 5% ethanolic phosphomolibdic acid.

Column chromatography purifications were performed using silica gel stationary phase (Merck 60, particle size 40‒63 µm), eluting by gravity, or with a mild pressure. Eluants are indicated in each case.

NMR spectra were registered in the Centro de Investigación, Tecnología e Innovación de la Universidad de Sevilla (CITIUS), using Bruker Avance III 300 spectrometer (300.1 MHz for ¹H, 75.5 MHz for ¹³C), using the solvent indicated in each case. Chemical shifts (δ) are expressed in ppm, and coupling constants (J) in Hz. Residual signals from the solvent are used as internal references for the calibration. Mass spectra were registered using a QExactive™ spectrometer, using Electrospray Ionisation (ESI), and calibrated using the Pierce^TM LTQ Velos ESI Positive Ion Calibration Solution (ThermoFisher Scientific).

2.2. Biological assessments

General procedures for the biological assays accomplished herein can be found in the Supporting Information.

2.3. Chemistry

General procedures for the preparation of isothiocyanates 11‒20, thiosemicarbazides 22‒30, thiosemicarbazones 36‒65 and selenosemicarbazones 74, 75, together with spectroscopic characterisation of new compounds can be found in the Supporting Information.

3. Results and discussion

3.1. Chemistry

Considering the variety of biological properties exhibited by thiosemicarbazones, including some antiprotozoal activities, we have accomplished the preparation of an ample panel of Schiff bases analogues with the aim of targeting protozoal-mediated diseases like Chagas disease, or trichomoniasis. For this purpose, we envisioned the general structure depicted in Figure 1: an aromatic residue (phenyl, naphthyl) is located on position N-4 of the thiosemicarbazone, which in turn incorporates substituents with different stereoelectronic properties, like methyl, methoxy, halogens, or NO₂. Moreover, the second aromatic motif, on position N-1 of the Schiff base is decorated with a different number of substituents, including free and masked phenolics, and NO₂, which can modulate both the bioavailability and activity of the compounds.

Figure 1.

General structure for the ntiprotozoal thio(seleno)semicarbazones prepared herein.

We have also considered the isosteric replacement of sulphur with selenium; our group has previously prepared a plethora of organoselenium derivatives exhibiting interesting redox properties^50,51, together with strong antiproliferative activities^{37,38,52‒55}.

The synthetic approach used for accessing thiosemicarbazones is depicted in Scheme 1. Thus, treatment of commercially available anilines 1‒10 with thiophosgene as the thionating agent afforded isothiocyanates 11‒20, which in turn were transformed into target thiosemicarbazones 36‒65 upon subsequent treatment with hydrazine, followed by condensation of intermediate thiosemicarbazides 21‒31 with the corresponding aldehydes 32‒35. Final compounds precipitated in the medium; ¹³C-NMR resonances at 175‒177 ppm (C = S) and at 143‒146 ppm (C = N) confirmed their structures.

Scheme 1.

Preparation of thiosemicarbazones 36‒65.

Selenosemicarbazones 74, 75 were obtained using a similar synthetic pathway (Scheme 2); the main difference is the access to the key isoselenocyanates 70, 71. They were obtained using a methodology developed in our group^56,57 for the preparation of alkyl- and aryl isoselenocyanates, using triphosgene as a safe substitute for hazardous phosgene in the dehydration of formamides 66, 67 to furnish transient isocyanides 68, 69, which underwent addition of elemental black selenium.

Scheme 2.

Preparation of selenosemicarbazones 72, 73.

3.2. Biological assays

3.2.1. In vitro evaluation of anti-T. cruzi activity

Trypanocidal activity was evaluated by following a sequential screening procedure in which all compounds were primarily tested over T. cruzi epimastigotes of the drug-sensitive CL-B5 lacZ strain (DTU TcVI)^58,59. Then, compounds whose trypanocidal profiles were similar to that of the reference drug BZ, were assayed over such an extracellular form of the moderately drug-resistant Y strain (DTU TcII)⁶⁰. Only those compounds with selectivity indexes (SI) on epimastigotes similar or higher than that of BZ, were moved to a more specific assay against the replicative and intracellular forms of the parasite (amastigotes)^58,59. Table 1 depicts the results of trypanocidal activity shown by parent thiosemicarbazides (23‒25, 27‒31) and the corresponding thiosemicarbazones (36‒65).

Table 1.

In vitro activity against Trypanosoma cruzi CL-B5 (DTU TcVI) and Y (DTU TcII) parasites and unspecific cytotoxicity on L929 and J774 cells, expressed as IC₅₀ and CC₅₀ values, respectively. Selectivity indexes (SI) for each strain and form are also calculated.

Compound	IC50 (µM)a CL-B5 epimastigotes	CC50 (µM)a L929 cells	SIb CL-B5 epi	IC50 (µM)a Y epimastigotes	CC50 (µM)a J774 cells	SIb Y epi	IC50 (µM)a CL-B5 amastigotes	SIb CL-B5 ama
23 (R1=OMe, R2=R3=H)	53.51 ± 12.98	>256	>4.78	59.91 ± 10.43	>256	>4.27	−	−
24 (R1=F, R2=R3=H)	58.80 ± 8.77	>256	>4.35	97.26 ± 3.45	>256	>2.63	−	−
25 (R1=Cl, R2=R3=H)	44.17 ± 13.61	67.66 ± 3.16	1.53	−	−	−	−	−
27 (R1=I, R2=R3=H)	49.76 ± 14.25	50.99 ± 8.88	1.02	−	−	−	−	−
28 (R1=R2=H, R3=NO2)	44.15 ± 5.38	210.34 ± 15.34	4.76	−	−	−	−	−
29 (R1=R3=H, R2=NO2)	79.86 ± 15.40	205.76 ± 7.25	2.58	−	−	−	−	−
30 (R1=NO2, R2=R3=H)	72.49 ± 1.23	23.58 ± 2.84	0.33	−	−	−	−	−
31 (Ar = Naphthyl)	28.72 ± 4.61	>256	>8.91	33.65 ± 5.72	>256	>7.61	39.66 ± 9.91	>6.45
R1=R2= R3=H
36 R4=OH, R5=H	70.88 ± 10.96	46.29 ± 0.15	0.65	−	−	−	−	−
37 R4=OMe R5=H	39.79 ± 6,43	>256	>6.43	85.10 ± 17.15	>256	>3.01	−	−
38 R4=R5=OH	15.34 ± 2.22	45.19 ± 0.58	2.95	−	−	−	−	−
R1=Me, R2=R3=H
39 R4=OH, R5=H	39.58 ± 7.08	60.99 ± 3.72	1.54	−	−	−	−	−
40 R4=OMe R5=H	>256	>256	NDc	−	−	−	−	−
41 R4=R5=OH	118.61 ± 7.68	20.02 ± 1.87	0.17	−	−	−	−	−
R1=OMe, R2=R3=H
42 R4=OH, R5=H	49.19 ± 9.97	43.31 ± 16.61	0.88	−	−	−	−	−
43 R4=OMe R5=H	26.75 ± 2.68	>256	>9.57	55.57 ± 16.30	>256	>4.61	39.04 ± 14.33	>6.56
R1=F, R2=R3=H
44 R4=OMe R5=H	>256	>256	ND	−	−	−	−	−
45 R4=R5=OH	97.38 ± 5.33	30.47 ± 3.93	0.31	−	−	−	−	−
R1=Cl, R2=R3=H
46 R4=OMe R5=H	30.53 ± 4.10	49.72 ± 3.76	1.63	−	−	−	−	−
R1=Br, R2=R3=H
47 R4=OH, R5=H	42.79 ± 12.01	66.96 ± 5.43	1.56	−	−	−	−	−
48 R4=OMe R5=H	>256	>256	ND	−	−	−	−	−
49 R4=R5=OH	104.5 ± 15.71	16.71 ± 0.70	0.16	−	−	−	−	−
R1=I, R2=R3=H
50 R4=OH, R5=H	22.80 ± 1.61	41.44 ± 8.36	1.82	−	−	−	−	−
51 R4=OMe R5=H	>256	>256	NDc	−	−	−	−	−
52 R4=R5=OH	69.37 ± 4.73	20.82 ± 7.92	0.30	−	−	−	−	−
R1=R2=HR3=NO2
53 R4=OH, R5=H	21.77 ± 2.98	34.89 ± 8.65	1.60	−	−	−	−	−
54 R4=OMe R5=H	210.40 ± 14.54	>256	>1.22	−	−	−	−	−
55 R4=R5=OH	41.99 ± 9.06	21.40 ± 4.12	0.51	−	−	−	−	−
R1=R3=HR2=NO2
56 R4=OH, R5=H	36.85 ± 3.84	28.81 ± 3.25	0.78	−	−	−	−	−
57 R4=OMe R5=H	22.96 ± 6.33	>256	>11.15	>256	>256	ND	22.27 ± 4.20	>11.49
58 R4=R5=OH	33.79 ± 7.59	11.98 ± 1.42	0.35	−	−	−	−	−
R1=NO2 R2=R3=H
59 R4=OH, R5=H	76.37 ± 9.91	29.44 ± 2.52	0.39	−	−	−	−	−
60 R4=OMe R5=H	>256	>256	ND	−	−	−	−	−
61 R4=R5=OH	110.56 ± 14.83	5.37 ± 0.30	0.05	−	−	−	−	−
62 R4=NO2 R5=H	>256	53.50 ± 4.82	<0.21	−	−	−	−	−
Ar = Naphthyl
63 R4=OH R5=H	53.65 ± 11.28	31.14 ± 9.21	0.58	−	−	−	−	−
64 R4=OMe R5=H	>256	>256	ND	−	−	−	−	−
65 R4=R5=OH	70.23 ± 6.99	28.79 ± 7.77	0.41	−	−	−	−	−
Benznidazole	25.31 ± 1.63	>256	>10.11	22.73 ± 1.82	>256	>11.26	0.47 ± 0.09	>544.68

^aResults are expressed as the mean ± SD of three independent experiments (n = 3).

^bSI CL-B5 epi = CC₅₀ L929/IC₅₀ CL lacZ epimastigotes, SI Y epi = CC₅₀ J774/IC₅₀ Y epimastigotes and SI CL-B5 ama = CC₅₀ L929/IC₅₀ CL lacZ amastigotes.

^cND: not determined.

As it can be seen, thiosemicarbazides 23 and 31 and thiosemicarbazones 37, 43 and 57 were active on CL-B5 epimastigotes, showing selectivity indexes (SI) ranging from >4.78 to >11.15. According to these results, they were tested on Y strain epimastigotes: three of them displayed slightly lower activity against moderately drug-resistant parasites, and only 23 and 31 showed similar trypanocidal profiles over both DTU TcVI and TcII strains. As depicted in Table 1, none of these molecules triggered toxic effects on mammalian cells, either phagocytic or non-phagocytic (CC₅₀ >256 µM). In fact, previous studies introduce the capability of different series of thiosemicarbazone-based compounds to inhibit epimastigotes growth, proposing the interaction of these molecules with parasite proteases (i.e., cruzipain, one validated target for Chagas disease) as responsible of such an effect^19‒21. Improved inhibition of cruzipain (low nanomolar range) was reported²² for the isosteric replacement of sulphur with selenium, what was claimed to be responsible for the antiparasitic activity of such selenosemicarbazones (epimastigotes and intracellular amastigotes).

Regarding the activity on CL-B5 amastigotes, compounds 31, 43 and 57 were the only derivatives assayed against the replicative and intracellular form of the parasite, according to their SI on CL-B5 epimastigotes: SI 31 > 8.91, SI 43 > 9.57 and SI 57 > 11.15. Unfortunately, none of these derivatives were as active as BZ, with IC₅₀ values on amastigotes ranging from 22.27 to 39.66 µM. This reduction in the parasite burden of infected cells could also occur because of cruzipain inhibition: the cysteine protease, also present in the intracellular form of T. cruzi, has been usually proposed as putative target of thiosemicarbazones^61,62.

3.2.2. In vitro evaluation of anti-Trichomonas vaginalis activity

Activity against T. vaginalis was evaluated following the sequential flow chart protocol of drug screening in which the most active molecules against the parasite are subsequently evaluated against Vero cells to determine their selectivity indexes (SI)⁶³. The antiparasitic activity of thiosemicarbazides 23‒25, 27‒31 and thiosemicarbazones 36‒65 is shown in Table 2. Activity of all the compounds was lower than that of the reference drug MTZ (IC₅₀=2.68 µM). However, almost 37% of all the synthetic molecules exhibited from moderate to good trichomonacidal activity with an IC₅₀≤50 µM and were screened against mammalian cell lines.

Table 2.

In vitro activity against Trichomonas vaginalis (IC₅₀), unspecific cytotoxicity of the most promising molecules against Vero cells (CC₅₀) and selectivity indexes (SI).

Compound	IC50 (µM) a	CC50 (µM)	SI b
23 (R1=OMe, R2=R3=H)	255.70 [205.26–348.10]	–	–
24 (R1=F, R2=R3=H)	200.49 [134.86–366.51]	–	–
25 (R1=Cl, R2=R3=H)	49.19 [30.46–77.32]	>300	>6.10
27 (R1=I, R2=R3=H)	75.69 [57.77–102.71]	–	–
28 (R1=R2=H, R3=NO2)	>300	–	–
29 (R1=R3=H, R2=NO2)	19.14 [14.92–23.52]	>300	>15.67
30 (R1=NO2, R2=R3=H)	105.92 [63.13–236.25]	–	–
31 (Ar = Naphthyl)	30.20 [11.76–55.08]	>300	>9.93
R1=R2= R3=H
36 R4=OH, R5=H	74.01 [60.18–92.68]	–	–
37 R4=OMe R5=H	167.29 [116.44–283.25]	–	–
38 R4=R5=OH	88.59 [65.64–125.65]	–	–
R1=Me, R2=R3=H
39 R4=OH, R5=H	66.10 [43.85–103.72]	–	–
40 R4=OMe R5=H	225.96 [118.97–1084.50]	–	–
41 R4=R5=OH	80.61 [69.32–94.63]	–	–
R1=OMe, R2=R3=H
42 R4=OH, R5=H	64.79 [55.04–76.69]	–	–
43 R4=OMe R5=H	257.50 [141.94–834.94]	–	–
R1=F, R2=R3=H
44 R4=OMe R5=H	>300	–	–
45 R4=R5=OH	226.35 [192.65–275.96]	–	–
R1=Cl, R2=R3=H
46 R4=OMe R5=H	32.27 [21.71–47.99]	>300	>9.30
R1=Br, R2=R3=H
47 R4=OH, R5=H	52.36 [35.6–77.09]	–	–
48 R4=OMe R5=H	55.14 [47.61–63.92]	–	–
49 R4=R5=OH	16.39 [0.08–28.72]	>300	>18.30
R1=I, R2=R3=H
50 R4=OH, R5=H	42.25 [36.30–48.52]	>300	>7.10
51 R4=OMe R5=H	14.85 [5.94–24.24]	>300	>20.20
52 R4=R5=OH	24.89 [11.95–41.25]	>300	>12.05
R1=R2=HR3=NO2
53 R4=OH, R5=H	92.80 [81.78–103.30]	–	–
54 R4=OMe, R5=H	34.20 [29.47–39.39]	>300	>8.77
55 R4=R5=OH	132.56 [120.93–146.15]	–	–
R1=R3=HR2=NO2
56 R4=OH, R5=H	24.28 [12.49–38.45]	168.93 [127.03–239–12]	7.00
57 R4=OMe R5=H	51.91 [31.66–85.57]	–	–
58 R4=R5=OH	73.41 [50.52–110.14]	–	–
R1=NO2 R2=R3=H
59 R4=OH, R5=H	35.95 [29.98–42.63]	254.15 [170.16–725.83]	7.07
60 R4=OMe R5=H	155.17 [93.43–395.10]	–	–
61 R4=R5=OH	127.98 [77.11–294.93]	–	–
62 R4=NO2 R5=H	151.81 [112.95–209.64]	–	–
Ar = Naphthyl
63 R4=OH R5=H	14.89 [11.51–18.30]	274.75 [193.19–562.70]	18.45
64 R4=OMe R5=H	30.46 [17.60–46.35]	>300	>9.85
65 R4=R5=OH	20.14 [16.37–24.03]	174.81 [110.58–371.96]	>8.68
Metronidazole	2.68 [2.37–3.03]	>300	>111

IC₅₀ and CC₅₀ were calculated with growth inhibition values showing a standard deviation of less than 10%.

–: Not evaluated against Vero cells owing to the low trichomonacidal activity.

^aResults in brackets refer to 95% confidence interval.

^bSelectivity indexes SI = CC₅₀ Vero cells/IC₅₀ T. vaginalis.

In particular, four molecules displayed remarkable activity against T. vaginalis, namely thiosemicarbazide 29, and thiosemicarbazones 49, 51 and 63, with IC₅₀ values ranging from 14.85‒19.14 µM, and SI > 15.7.

It is important to highlight that most of the thiosemicarbazones synthetised lacked unspecific cytotoxicity at the highest concentration evaluated towards Vero cells. These results are in agreement with the phenylthiosemicarbazones evaluated by Gomes et al.⁶⁴ against T. gondii, which show cytotoxic effect on Vero cells between 300 to 700 µM. Also, the results published by Merlino et al.⁶⁵ refute the low cytotoxic profile of this scaffold showing a low non-specific cytotoxicity on human red blood cells and J-774 mouse macrophages.

In this context, due to the promising trichomonacidal profile and the absence of unspecific cytotoxicity of the most potent derivatives, these molecules can be considered as excellent candidates for further studies. According to this, two selenium compounds (i.e., 74 and 75) were prepared as the isosters of the two best trichomonacidal compounds evaluated previously (derivatives 49 and 51, respectively). The activity of these selenosemicarbazones on T. vaginalis was slightly enhanced in comparison with that of their sulphur counterparts. Curiously, both derivatives exhibit a similar trichomonacidal effect: IC₅₀ 49 = 11.10 µM and IC₅₀ 51 =11.02 µM.

Although thiosemicarbazone derivatives 49 and 51 did not exhibit unspecific cytotoxicity, their selenium analogue 74 presented a low cytotoxicity effect at the highest concentrations evaluated against Vero cells (CC₅₀ = 114.50 µM), while 75 did not present toxic effects at the maximum concentration tested. Moreover, SI observed in both compounds continues being higher than 10, which demonstrates their specific antiparasitic profile.

According to the trichomonacidal profile, only a few in vitro studies based thiosemicarbazones of 5-nitrothiophene-2-carboxaldehyde⁴⁶, or bis (thiosemicarbazone) and bis(4-methylthiosemicarbazone)⁴⁹ have been evaluated against T. vaginalis. However, the great structural difference between these molecules and the thio(seleno)semicarbazones prepared herein, makes it not possible to compare the antiparasitic effects.

3.2.2.1. Mechanistic study of the anti-trichomonas activity

In order to get more insight into the anti-trichomonas mode of action (MoA) of title thiosemicarbazones, we have analysed the alteration of the hydrogenosome membrane potential and structural changes in the protozoan; the two more promising compounds (thiosemicarbazones 51 and 63) were included in this study.

Thus, to determine if the hydrogenosome could be involved in their mode of action, alterations in the membrane potential of this organelle were studied. Membrane potential is indicative of the correct functioning of the organelle, which is essential for parasite survival and also a key aspect for the activation of 5-nitroimidazoles. For this purpose, JC-1 dye (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolocarbocyanine iodide) was used; this compound aggregates inside the mitochondria or hydrogenosomes in healthy cells, being detected by fluorescence at 590 nm. However, alterations in the membrane potential of the organelle preclude the agglutination of JC-1, which remains as a monomer in the cytoplasm with fluorescence emission at 535 nm.

Figure 2 depicts results of this assay after 4 h of incubation with derivatives 51 and 63 at 100 µM concentration; MTZ (24 µM) was included for comparison.

Figure 2.

Plots of fluorescence units at 535 nm after addition of JC-1. MTZ (metronidazole), CCCP (m-chlorocarbonylcyanide phenylhydrazone) and CC (culture control). Data indicate mean ± SD of three independent experiments (n = 3). Data that are significantly different from control experiment are marked with an asterisk (*) (ρ < 0.05).

Results demonstrate that neither the tested thiosemicarbazides nor the reference drug alter the hydrogenosome membrane potential of the parasite. Only CCCP (positive control) showed a significant difference (t-Student) with the others, suggesting a charge deregulation and thus, disturbance in the hydrogenosome.

These data are in agreement with images obtained from fluorescence microscopy. Figure 3 shows that uncoupler agent CCCP (Figure 3(B)) provokes a disturbance in the hydrogenosome membrane potential, as indicated by fluorescence distribution throughout the trophozoite body, which is much more intense and delocalised. However, for the rest of images − control (Figure 3(A)), and T. vaginalis treated with MTZ (Figure 3(C)), compound 51 (Figure 3(D)) and 63 (Figure 3(E)) − fluorescence restricted to certain regions of the parasite is observed. In other words, the good physiological condition of this organelle is demonstrated with the fluorescence results observed in Figure 2 and corroborated in the microscope. The accumulation of JC-1 in these organelles is clearly observed in growth controls and trophozoites treated with both compounds as reflects Figure 3.

Figure 3.

Fluorescence emitted by JC-1 in hydrogenosome. A. Control. B. Positive control (treatment with the uncoupler agent CCCP). C. Treatment with MTZ. D. Treatment with 51. E. Treatment with 63.

With the aim of studying any structural alteration in the parasite triggered by these thiosemicarbazones, scanning electronic microscopy (SEM) was used to observe surface disturbances upon 24 h treatment at 15 µM concentration (Figure 4).

Figure 4.

Scanning electron microscopy images of T. vaginalis cultures. A. Control. B. Treatment with compound 51. C. Treatment with compound 63.

Control experiment (Figure 4(A)) shows trophozoites with no alterations in the plasmatic membrane, with pear-shaped or even amoeboid cells. Interestingly, treatment with thiosemicarbazone 51 induced a clear disturbance in the trophozoite surface (Figure 4(B)), showing parasites with a more rounded shape and even pseudocyst forms; some trophozoites were found to exhibit invaginations and pores in the surface; structural abnormalities have therefore an apoptotic aspect. It has been reported that T. vaginalis adopts this rounded shape, with internalised flagella in endocytic vesicles upon stress situation or lack of nutrients⁶⁶. Regarding derivative 63, it induced amoeboid cells in most of the culture with smooth disturbances in the cytoplasmatic surface and showing a wrinkled shape in many trophozoites, as depicted in Figure 4(C).

Therefore, these MoA experiments show that both thiosemicarbazones provoke a trichomonacidal effect in a hydrogenosome-independent mechanism inducing the trophozoites death with a clear disruption of the parasite surface.

4. Conclusions

In the present work we have accomplished the preparation of an ample number of thio(seleno)semicarbazones as potential antiparasitic agents against T. cruzi (responsible for Chagas disease) and T. vaginalis (responsible for trichomoniasis) with a mode of action different to that of classical nitroheterocyclic compounds, the current only available drugs for treating both parasitic infections. We have carried out an extensive analysis of SARs upon modification of the stereoelectronic effects of the aromatic substituents, together with the nature of the chalcogen atom (S vs. Se). Thiosemicarbazide 31, bearing a naphthyl residue on N-4, exhibited the best trypanocidal activity at the epimastigote stage of two different parasitic strains, with similar profile to that of the reference drug benznidazole.

Moreover, halogenated thiosemicarbazones 49, 51 and naphthyl-containing 63 exhibited a remarkable trichomonacidal profile, with activities within the low micromolar range, and excellent selectivities. Their selenium isosters afforded a slightly enhancement of the activity.

Analysis of the hydrogenosome membrane potential, and structural changes through scanning electronic microscopy (51, 63) afforded valuable information concerning their mechanisms of action.

Funding Statement

The present work was financially supported by Grant PID2020-116460RB-I00, funded by MCIN/AEI/10.13039/501100011033, Junta de Andalucía (FQM-134), UCM Research Group ref. 911120 Epidemiología, Diagnóstico y Terapia antiparasitaria (PARADET) and the Mexican CONACYT (CB-2015/257465).

Disclosure statement

The authors report no conflict of interest.