INTRODUCTION
Trichomonas vaginalis is the aetiological causative agent of more than half of curable sexually transmitted infections (STI) with an estimated 256 million cases every year (WHO, 2012). Notwithstanding this high prevalence, which accounts for more cases than syphilis, chlamydiasis and gonorrhea combined, trichomonosis is still a non-reportable disease receiving low attention from public health organisms (Van der Pol, Reference Van der Pol2007). The infection has high rates of asymptomatic cases in women and especially in men (Lewis, Reference Lewis2014) and is associated with severe consequences including adverse outcomes during pregnancy (Cotch et al. Reference Cotch, Pastorek, Nugent, Hillier, Gibbs, Martin, Eschenbach, Edelman, Carey, Regan, Krohn, Klebanoff, Rao, Rhoads, Yaffe, Catz, McNellis, Berendes, Blackwelder, Kaslow, Reed, Greenberg, Williams and Rettig1997) and increased risk of acquisition of other dangerous STI as human papillomavirus (HPV) (Lazenby et al. Reference Lazenby, Taylor, Badman, McHaki, Korte, Soper and Young Pierce2014) or human immunodeficiency virus (HIV) (McClelland et al. Reference McClelland, Sangare, Hassan, Lavreys, Mandaliya, Kiarie, Ndinya-Achola, Jaoko and Baeten2007). Trichomonosis is also linked to pelvic inflammatory disease (Moodley et al. Reference Moodley, Wilkinson, Connolly, Moodley and Sturm2002) and increases the risk of cervical (El-Gayar and Rashwan, Reference El-Gayar and Rashwan2007) and prostatic neoplasia (Sutcliffe et al. Reference Sutcliffe, Neace, Magnuson, Reeves and Alderete2012) development. Furthermore, different studies have demonstrated a significant reduction of HIV-1 RNA in vaginal secretions after treatment for T. vaginalis in HIV+ patients with a corresponding reduction in vaginal shedding (Wang et al. Reference Wang, McClelland, Reilly, Overbaugh, Emery, Mandaliya, Chohan, Ndinya-Achola, Bwayo and Kreiss2001; Kissinger et al. Reference Kissinger, Amedee, Clark, Dumestre, Theall, Myers, Hagensee, Farley and Martin2009).
Metronidazole is the reference drug approved by the Food and Drug Administration for trichomonosis treatment since the 1960s (Helms et al. Reference Helms, Mosure, Secor and Workowski2008). Although this 5-nitroimidazole has supported excellent pharmacokinetic and antiparasitic profiles, clinical cases of metronidazole resistance were reported a few years after its introduction (Dunne et al. Reference Dunne, Dunn, Upcroft, O'Donoghue and Upcroft2003). Not until 2004 was an alternative drug, tinidazole, also approved for this sexually transmitted disease (Bachmann et al. Reference Bachmann, Hobbs, Seña, Sobel, Schwebke, Krieger, Scott McClelland and Workowski2011). Moreover, diverse side effects, cross-resistance between both related drugs and their possible mutagenic and teratogenic effects (Cudmore et al. Reference Cudmore, Delgaty, Hayward-McClelland, Petrin and Garber2004) evidence the necessity of discovering novel molecules with trichomonacidal properties and different structural patterns. The proposed aim of reinforcing the 5-nitroimidazole pharmacological arsenal would thus follow the guidelines of the World Health Organization, directed to accomplish diverse millennium development goals (WHO, 2007).
Accordingly, during the last decade our research group has been working on the design, synthesis and biological activity of novel potential scaffolds, especially nitroheterocycles, against trichomonosis and other protozoan diseases. In this context, trichomonacidal activity of many indazol-3-ol/indazolin-3-one derivatives has previously been reported (Arán et al. Reference Arán, Ochoa, Boiani, Buccino, Cerecetto, Gerpe, González, Montero, Nogal, Gómez-Barrio, Azqueta, López de Ceráin, Piro and Castellano2005; Marrero-Ponce et al. Reference Marrero-Ponce, Machado-Tugores, Montero Pereira, Escario, Gómez Barrio, Nogal-Ruiz, Ochoa, Arán, Martínez-Fernández, García Sánchez, Montero-Torres, Torrens and Meneses-Marcel2005, Reference Marrero-Ponce, Meneses-Marcel, Castillo-Garit, Machado-Tugores, Escario, Gómez Barrio, Montero Pereira, Nogal-Ruiz, Arán, Martínez-Fernández, Torrens, Rotondo, Ibarra-Velarde and Alvarado2006). More recently, several 1,2-disubstituted 5-nitroindazolin-3-ones (1–19; Fig. 1, series A) and 3-alkoxy-5-nitroindazoles substituted at positions 1 (20–24; Fig. 1, series B) and 2 (25–39; Fig. 1, series C), have been synthesized and tested against Trypanosoma cruzi (Vega et al. Reference Vega, Rolón, Montero-Torres, Fonseca-Berzal, Escario, Gómez-Barrio, Gálvez, Marrero-Ponce and Arán2012; Fonseca-Berzal et al. Reference Fonseca-Berzal, Escario, Arán and Gómez-Barrio2014; Muro et al. Reference Muro, Reviriego, Navarro, Marín, Ramírez-Macías, Rosales, Sánchez-Moreno and Arán2014). In order to investigate novel compounds with interesting trichomonacidal activity, the molecules mentioned above have been subjected to in vitro screening against well-established cultures of T. vaginalis following a previously validated fluorometric method (Ibáñez Escribano et al. Reference Ibáñez Escribano, Meneses Marcel, Machado Tugores, Nogal Ruiz, Arán Redó, Garcia-Trevijano and Gómez Barrio2012). To determine their non-specific cytotoxicity profiles, the most active compounds were also screened against mammalian cells according to the sequential flow-chart method used in our laboratory (Ibáñez-Escribano et al. Reference Ibáñez-Escribano, Meneses-Marcel, Marrero-Ponce, Nogal-Ruiz, Arán, Gómez-Barrio and Escario2014). Furthermore, the evaluated compounds were submitted to computational prediction of their toxicity risks including reproductive side effects and drug-like scores by using OSIRIS Property Explorer (Organic Chemistry Portal, 2014). The compounds were also tested for compliance with Lipinski's ‘rule of five’ (RO5) (Lipinski et al. Reference Lipinski, Lombardo, Dominy and Feeney1997) with the main aim of studying their bioavailability properties using the Molinspiration online software (Molinspiration Cheminformatics, 2014). The design of viable new drug candidates should consider molecules with specific activity but also with acceptable ADME (absorption, distribution, metabolism, and excretion) properties, which include a chemical structure which does not present difficulties in oral bioavailability. These in silico calculations permit the removal of molecular entities that would probably be discarded in the later stages of discovery and development due to their low oral bioavailability or toxic features.
Indazole derivatives studied in the current work.
METHODS
Chemistry: preparation of the studied compounds
Compounds 1–19 (series A), 20–24 (series B) and 25–39 (series C) were prepared by alkylation of the corresponding 2-substituted indazolin-3-ones or 1-substituted indazol-3-ols with the required alkyl halides (Vega et al. Reference Vega, Rolón, Montero-Torres, Fonseca-Berzal, Escario, Gómez-Barrio, Gálvez, Marrero-Ponce and Arán2012; Muro et al. Reference Muro, Reviriego, Navarro, Marín, Ramírez-Macías, Rosales, Sánchez-Moreno and Arán2014).
Biological assays
In vitro determination of trichomonacidal activity
Trichomonas vaginalis isolate JH31A4 from the American Culture Type Collection (ATCC) was grown in vitro in modified TYM medium at pH 6 with 10% inactivated foetal bovine serum (FBS), 100 IU penicillin mL−1 and 100 µg mL−1 streptomycin, in an humidified chamber at 37 °C and sub-cultured every 48–72 h.
In vitro trichomonacidal activity assays were carried out in glass tubes containing 105 trophozoites mL−1. After 5 h, log-phase cultures were incubated with the different nitroindazole derivatives at 100, 10 and 1 µg mL−1 for 24 h at 37 °C and 5% CO2. Stock solutions of chemical compounds were dissolved in dimethyl sulfoxide (DMSO) prior to use. Metronidazole was used as a reference drug at 4 µg mL−1 and prepared in a similar fashion to the rest of the chemical derivatives. The final concentration of DMSO in cultures never exceeded 0·2%. Afterwards, cultures were seeded in 96-well microtitre plates and washed with PBS supplemented with 0·1% glucose. The trichomonacidal effect was determined by fluorometry after 1 h of incubation with resazurin dye (stock solution 3 mm) following the methodology previously described (Ibáñez Escribano et al. Reference Ibáñez Escribano, Meneses Marcel, Machado Tugores, Nogal Ruiz, Arán Redó, Garcia-Trevijano and Gómez Barrio2012). Activity values were calculated from the percentage reduction of resazurin in treated and untreated wells. Each concentration was evaluated in triplicate and values were obtained from the average of two separate determinations.
In vitro determination of non-specific cytotoxicity in mammalian cells
Cytotoxicity against Vero cells was determined following the sequential procedure reported by our research group (Ibáñez-Escribano et al. Reference Ibáñez-Escribano, Meneses-Marcel, Marrero-Ponce, Nogal-Ruiz, Arán, Gómez-Barrio and Escario2014). The monkey kidney epithelial cell line Vero CCL-81 (from ATCC) was grown in RPMI-1640 medium (Sigma) supplemented with 10% heat-inactivated FBS and antibiotics in a humidified atmosphere of 95% air and 5% CO2 at 37 °C. Cells were seeded at a density of 5 × 104 cells/well in 96-well flat-bottom microplates (Nunc). After cell attachment, the drugs dissolved in RPMI medium were added at the same concentrations as for the evaluation against T. vaginalis. The plates were incubated for 24 h at 37 °C in 5% CO2 and the cytotoxic activity was revealed by adding 20 µL of 1 mm resazurin solution. After 3 h in contact with the redox dye, the fluorescence readings were carried out in a fluorometer (Infinite 200, TECAN) as described previously (Ibáñez-Escribano et al. Reference Ibáñez-Escribano, Reviriego, Nogal-Ruiz, Meneses-Marcel, Gómez-Barrio, Escario and Arán2015) with minor modifications. The selectivity index (SI) was determined as the ratio between the concentration of the compound needed to reduce cell viability to 50% (CC50) and the concentration that showed a 50% growth inhibitory effect (GI50) on the parasitic culture. Only those compounds showing an antiparasitic profile, defined as having a significant trichomonacidal activity without non-specific toxicity, were tested again at six different serial 2-fold dilutions and their SI was determined.
Computational analysis: molecular properties, bioavailability parameters and prediction of potential biological risks
The molecular properties derived from the chemical structure were calculated using online bioinformatics tools. The different properties related to molecular permeability such as physicochemical features involved in the Lipinski's RO5 (Lipinski et al. Reference Lipinski, Lombardo, Dominy and Feeney1997) as well as topological polar surface area (TPSA), rotatable bonds and molecular volume were computed using free online software (Molinspiration Cheminformatics, 2014). The percentage absorption was estimated following the criteria of Zhao et al. (Reference Zhao, Abraham, Le, Hersey, Luscombe, Beck, Sherborne and Cooper2002). On the other hand, the online program OSIRIS (Organic Chemistry Portal, 2014) was employed for the prediction of potential risks including mutagenic, tumorigenic, irritative and reproductive side effects by comparing the chemical structures of the studied compounds with those of commercial drugs.
RESULTS AND DISCUSSION
Compounds of series A, B and C (Fig. 1) as well as the reference drug metronidazole were analysed for their in vitro antiparasitic activity against a well-established isolate of T. vaginalis following a previously reported procedure (Ibáñez Escribano et al. Reference Ibáñez Escribano, Meneses Marcel, Machado Tugores, Nogal Ruiz, Arán Redó, Garcia-Trevijano and Gómez Barrio2012). As shown in Table 1, 16 of the synthesized compounds exhibited a significant trichomonacidal profile with more than 85% growth inhibition after 24 h in contact with the trophozoites. When compared with the reference drug, none of the compounds reached a minimum inhibitory concentration (MIC100) lower than that of metronidazole (MIC100−MTZ = 4 µg mL−1); nevertheless, compounds 23, 24, 25 and 27 from series B and C, inhibited parasite growth by more than 55% at 10 µg mL−1 (30–48 µ m) and, in addition, the first two compounds (23, 24) still showed a noticeable activity at the lowest dose tested (1 µg mL−1, ca. 3 µ m), inhibiting parasite growth by nearly 40%.
In vitro trichomonacidal effect of 1,2-disubstituted indazolinones 1–19 (series A), 1-substituted 3-alkoxyindazoles 20–24 (series B) and 2-substituted 3-alkoxyindazoles 25–39 (series C)
a Micromolar concentration (μ m) corresponding to the highest tested concentration (100 µg mL−1) is indicated for each compound.
b Results are expressed as a percentage. All molecules were assayed in triplicate in at least two independent experiments. All the results displayed a standard deviation of less than 10%.
From the observed results we can conclude that 1,2-disubstituted 5-nitroindazolin-3-ones (1–19, series A), which were very efficient against T. cruzi (Vega et al. Reference Vega, Rolón, Montero-Torres, Fonseca-Berzal, Escario, Gómez-Barrio, Gálvez, Marrero-Ponce and Arán2012; Fonseca-Berzal et al. Reference Fonseca-Berzal, Escario, Arán and Gómez-Barrio2014) show very low trichomonacidal activity, while 3-alkoxy-5-nitroindazoles containing simple substituents at positions 1 (20–24, series B) or 2 (25–29, series C) display moderate activity. We have not been able to establish a general structure-activity relationship; however, it is clear that among compounds of series B, the best compounds (23, 24) show bulky lipophilic substituents (phenethoxy and 2-naphthylmethoxy, respectively) at position 3 of the indazole ring. Conversely, for series C the best compounds (25, 27) support small substituents (methoxy and isopropoxy, respectively) at the same position. Accordingly, series B and C scaffolds could be promising for further chemical modifications. In fact, we have previously reported that for compounds of series B, activity increases considerably after the introduction of complex ω-(dialkylamino)alkyl chains at position 1 (Arán et al. Reference Arán, Ochoa, Boiani, Buccino, Cerecetto, Gerpe, González, Montero, Nogal, Gómez-Barrio, Azqueta, López de Ceráin, Piro and Castellano2005).
On the other hand, a comparison of the activities of compound 9 and its denitro analogue 10 at 100 µg mL−1 shows that the 5-nitro group of the indazole ring plays a relevant role in trichomonacidal activity. In fact, a similar effect has previously been noticed for 1-substituted indazol-3-ols (Marrero-Ponce et al. Reference Marrero-Ponce, Meneses-Marcel, Castillo-Garit, Machado-Tugores, Escario, Gómez Barrio, Montero Pereira, Nogal-Ruiz, Arán, Martínez-Fernández, Torrens, Rotondo, Ibarra-Velarde and Alvarado2006).
With the results gathered in the present article, we complete the exploration of trichomonacidal activity of indazol-3-ol/indazolin-3-one tautomeric system derivatives substituted at the pyrazole ring. In fact, previous studies carried out for monosubstituted derivatives have shown that 1-substituted 5-nitroindazol-3-ols are very efficient (Marrero-Ponce et al. Reference Marrero-Ponce, Machado-Tugores, Montero Pereira, Escario, Gómez Barrio, Nogal-Ruiz, Ochoa, Arán, Martínez-Fernández, García Sánchez, Montero-Torres, Torrens and Meneses-Marcel2005, Reference Marrero-Ponce, Meneses-Marcel, Castillo-Garit, Machado-Tugores, Escario, Gómez Barrio, Montero Pereira, Nogal-Ruiz, Arán, Martínez-Fernández, Torrens, Rotondo, Ibarra-Velarde and Alvarado2006), while 2-substituted 5-nitroindazolin-3-ones and 3-alkoxy-5-nitroindazoles, although there are few studied cases, seem to display very low activity (Marrero-Ponce et al. Reference Marrero-Ponce, Machado-Tugores, Montero Pereira, Escario, Gómez Barrio, Nogal-Ruiz, Ochoa, Arán, Martínez-Fernández, García Sánchez, Montero-Torres, Torrens and Meneses-Marcel2005).
The metronidazole and tinidazole mode of action is based on the nitro group located at position 5 of the imidazole ring. These compounds are in fact prodrugs, which enter into the parasite by passive diffusion and are activated inside the hydrogenosome by redox parasitic enzymes. It is usually accepted that reduction of the nitro group generates cytotoxic radicals and intermediates that are highly reactive towards DNA and proteins, leading to the formation of covalent adducts with essential parasite biomolecules (Leitsch et al. Reference Leitsch, Kolarich, Binder, Stadlmann, Altmann and Duchêne2009). A similar mode of action associated with the generation of NO2 radicals inducing oxidative stress could explain the importance of the nitro group in a remarkable number of nitro derivatives showing trichomonacidal activity (Adagu et al. Reference Adagu, Nolder, Warhurst and Rossignol2002; Navarrete-Vázquez et al. Reference Navarrete-Vázquez, Rojano-Vilchis, Yépez-Mulia, Meléndez, Gerena, Hernández-Campos, Castillo and Hernández-Luis2006; Hernández-Núñez et al. Reference Hernández-Núñez, Tlahuext, Moo-Puc, Torres-Gómez, Reyes-Martínez, Cedillo-Rivera, Nava-Zuazo and Navarrete-Vazquez2009; Kumar et al. Reference Kumar, Sarswat, Lal, Sharma, Jain, Kumar, Verma, Maikhuri, Kumar, Shukla and Gupta2010, Reference Kumar, Sarswat, Lal, Jain, Kumar, Kumar, Maikhuri, Pandey, Shukla, Gupta and Sharma2011, Reference Kumar, Jain, Lal, Sarswat, Jangir, Kumar, Singh, Shah, Jain, Maikhuri, Siddiqi, Gupta and Sharma2012). Keeping in mind, however, the recent advances on the reductive metabolism and mechanism of action of other nitroheterocyclic prodrugs such as the antichagasic drugs nifurtimox (Hall et al. Reference Hall, Bot and Wilkinson2011) and benznidazole (Hall and Wilkinson, Reference Hall and Wilkinson2012; Trochine et al. Reference Trochine, Creek, Faral-Tello, Barrett and Robello2014), related mechanistic studies in the field of trichomonacidal nitroheterocycles are required.
With the main aim of studying the specific antiprotozoal activity of the synthetic compounds, the four most active molecules 23, 24, 25 and 27, were evaluated again against the parasite at six different concentrations and a non-specific cytotoxicity test on VERO cells was simultaneously performed to determine the SI following a previously described method (Ibáñez-Escribano et al. Reference Ibáñez-Escribano, Meneses-Marcel, Marrero-Ponce, Nogal-Ruiz, Arán, Gómez-Barrio and Escario2014). Two of the four compounds (23 and 25) displayed a remarkable GI50 (T. vaginalis 50% growth inhibition) calculated from log-probit analyses using linear regression (SPSS, IBM v.22). Compound 25 exhibited a GI50 of 6·69 µg mL−1 (18·51 µ m) while 23 showed a GI50 of 11·63 µg mL−1 (39·12 μ m).
The screening assays revealed no non-specific cytotoxic effects against Vero cells at the assayed concentrations (1–100 µg mL−1), displaying a reduction of cellular growth of <10% at the highest concentration studied. Moreover, no difference was observed between cells grown with 100 µg mL−1 of compounds 23 and 24 during 24 h and growth controls. Only a slight percentage reduction of 6·68 ± 4·97 and 9·44 ± 3·22 was detected for 25 and 27, respectively. Based on this in vitro screening profile, compounds 23, 24, 25 and 27 show a remarkable SI of more than 7·5. The absence of non-specific cytotoxic activity against mammalian cells is in agreement with previous studies conducted by our research group with other 3-alkoxy-5-nitroindazole derivatives (Arán et al. Reference Arán, Ochoa, Boiani, Buccino, Cerecetto, Gerpe, González, Montero, Nogal, Gómez-Barrio, Azqueta, López de Ceráin, Piro and Castellano2005; Vega et al. Reference Vega, Rolón, Montero-Torres, Fonseca-Berzal, Escario, Gómez-Barrio, Gálvez, Marrero-Ponce and Arán2012; Muro et al. Reference Muro, Reviriego, Navarro, Marín, Ramírez-Macías, Rosales, Sánchez-Moreno and Arán2014).
Furthermore, these in vitro non-specific cytotoxicity results are in agreement with the virtual prediction of several risks made by a fragment-based method using the OSIRIS software (Supplementary data, Table S1). It was found that all the synthesized 1,2-disubstituted 5-nitroindazolin-3-one derivatives (1–19) exhibit non-toxic features in terms of mutagenic (Mut.), tumorigenic (Tum.), irritative (Irrit.) or reproductive (Reprod.) risks. These results increase the interest in compounds of series A, taking in account the fact that some of its members, especially 11–13, have shown to be very effective against T. cruzi (Vega et al. Reference Vega, Rolón, Montero-Torres, Fonseca-Berzal, Escario, Gómez-Barrio, Gálvez, Marrero-Ponce and Arán2012; Fonseca-Berzal et al. Reference Fonseca-Berzal, Escario, Arán and Gómez-Barrio2014). On the other hand, the results shown in Supplementary Table S1 suggest that substitution at position 2 of the 3-alkoxy-5-nitroindazole moiety (25–39) induces tumorigenic and mutagenic effects, but no risks were predicted for molecules with substituents at position 1 (20–24). Fortunately no irritative or detrimental effects on mammalian reproduction were predicted for the evaluated molecules with minor exceptions. In fact, only compound 34 (2-benzyl-3-butoxy-5-nitro-2H-indazole) was classified as potentially irritative also having the lowest calculated drug-score (0·08; see below). The risk of detrimental reproductive effects from metronidazole was also corroborated by the virtual screening as summarized in Supplementary Table S1.
The drug-score gives information related to the overall potential of a compound to be classified as a drug, taking into consideration different parameters related to physicochemical, risk and drug likeness calculations. Thus, its estimation allows compounds that will probably be poor drugs to be discarded at an early stage, according to the balance of their overall structure and molecular properties. In our case, the values calculated by the OSIRIS software for compounds of series A and B reflect the suitability of these structures for biological studies. As shown in Supplementary Table S1, 5-nitroindazolin-3-ones (1–9 and 11–19, series A) and 1-substituted 3-alkoxy-5-nitroindazoles (20 – 24, series B) display remarkable drug-scores in the range of 0·26–0·51, close in many cases to that of metronidazole (0·51). In the case of 2-substituted 3-alkoxy-5-nitroindazoles (25 – 39, series C), drug-scores are, except for compound 28, very low (0·08–0·19).
In relation to Lipinski's ‘rule of five’ (RO5) (Lipinski et al. Reference Lipinski, Lombardo, Dominy and Feeney1997) and further studies conducted by other research groups (Ertl et al. Reference Ertl, Rohde and Selzer2000; Veber et al. Reference Veber, Johnson, Cheng, Smith, Ward and Kopple2002), several physicochemical properties such as molecular weight (MW), hydrogen-bond acceptors (Ha) and donors (Hd), volume, number of atoms, calculated coefficient of partition (CLogP), TPSA and rotatable bonds (Rotb) have been empirically inferred (Supplementary data, Tables S2 and S3). Neither indazolin-3-one derivatives (1–19) nor 3-alkoxy-5-nitroindazoles (20–39) violate Lipinski's rule. The rule states that compounds with more than 5 Hd, 10 Ha, a MW > 500 and a CLogP > 5 are more likely to exhibit poor absorption and permeation (Lipinski et al. 1997). Regarding the rest of the physicochemical properties, all the evaluated compounds were adequately sized between 15 and 27 atoms, and had less than 8 Rotb which correlates with a suitable molecular flexibility (Veber et al. Reference Veber, Johnson, Cheng, Smith, Ward and Kopple2002). Their TPSA values were 72·763 and 72·884 for series A and series B/C, respectively. These values predict good oral bioavailability since they can be described as excellent oral absorption parameters (Palm et al. Reference Palm, Stenberg, Luthman and Artursson1997), defined as a suitable percentage of absorption >84% for the 39 studied compounds, calculated according to the method of Zhao et al. (Reference Zhao, Abraham, Le, Hersey, Luscombe, Beck, Sherborne and Cooper2002). The TPSA values (c. 73 Å2) and the number of H-bond acceptors (Ha = 6) classified these entities as inadequate for trespassing the blood-brain barrier by passive-diffusion (Lipinski, Reference Lipinski2004). In agreement with these results, no undesirable side effects associated with the central nervous system should be expected.
In conclusion, the in silico and in vitro results obtained in the current study suggest 2-substituted (series C) and, especially, 1-substituted 3-alkoxy-5-nitroindazoles (series B) as promising lead-like scaffolds for further chemical modifications. The evaluated compounds of series B presented interesting trichomonacidal activity and specificity as well as remarkable bioavailability and safety profiles. This is the case, in particular, for 3-phenethoxy- (23) and 3-(2-naphthylmethoxy) (24) indazole derivatives. Novel structures should be developed, however, with the aim of identifying new highly selective trichomonacidal drugs.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S0031182015001419.
ACKNOWLEDGMENTS
The Research Group ‘Antiparasitic Therapy’ (Ref. 911120) of the Universidad Complutense de Madrid and CEI Moncloa (cluster of i-Medicine) is gratefully acknowledged by all the authors.
FINANCIAL SUPPORT
A. Ibáñez-Escribano thanks Moncloa Campus of International Excellence (UCM-UPM & CSIC) for a PICATA predoctoral fellowship.
