Efficacy of Novel Pyrazolone Phosphodiesterase Inhibitors in Experimental Mouse Models of Trypanosoma cruzi

Pyrazolones are heterocyclic compounds with interesting biological properties. Some derivatives inhibit phosphodiesterases (PDEs) and thereby increase the cellular concentration of cyclic AMP (cAMP), which plays a vital role in the control of metabolism in eukaryotic cells, including the protozoan Trypanosoma cruzi, the etiological agent of Chagas disease (CD), a major neglected tropical disease. In vitro phenotypic screening identified a 4-bromophenyl-dihydropyrazole dimer as an anti-T. cruzi hit and 17 novel pyrazolone analogues with variations on the phenyl ring were investigated in a panel of phenotypic laboratory models.

ABSTRACT

Pyrazolones are heterocyclic compounds with interesting biological properties. Some derivatives inhibit phosphodiesterases (PDEs) and thereby increase the cellular concentration of cyclic AMP (cAMP), which plays a vital role in the control of metabolism in eukaryotic cells, including the protozoan Trypanosoma cruzi, the etiological agent of Chagas disease (CD), a major neglected tropical disease. In vitro phenotypic screening identified a 4-bromophenyl-dihydropyrazole dimer as an anti-T. cruzi hit and 17 novel pyrazolone analogues with variations on the phenyl ring were investigated in a panel of phenotypic laboratory models. Potent activity against the intracellular forms (Tulahuen and Y strains) was obtained with 50% effective concentration (EC₅₀) values within the 0.17 to 3.3 μM range. Although most were not active against bloodstream trypomastigotes, an altered morphology and loss of infectivity were observed. Pretreatment of the mammalian host cells with pyrazolones did not interfere with infection and proliferation, showing that the drug activity was not the result of changes to host cell metabolism. The pyrazolone NPD-227 increased the intracellular cAMP levels and was able to sterilize T. cruzi-infected cell cultures. Thus, due to its high potency and selectivity in vitro, and its additive interaction with benznidazole (Bz), NPD-227 was next assessed in the acute mouse model. Oral dosing for 5 days of NPD-227 at 10 mg/kg + Bz at 10 mg/kg not only reduced parasitemia (>87%) but also protected against mortality (>83% survival), hence demonstrating superiority to the monotherapy schemes. These data support these pyrazolone molecules as potential novel therapeutic alternatives for Chagas disease.

INTRODUCTION

Chagas disease (CD) belongs to the group of 20 neglected tropical diseases (NTDs) and affects around 8 million people worldwide (1). The currently available drugs, benznidazole (Bz) and nifurtimox, were introduced in clinical practice more than 50 years ago but present limited efficacy, severe side effects, and some parasite strains display a natural resistance against both and are thus untreatable (2). To overcome these issues, highly recurrent among NTDs, collaborative public-private partnerships involving industry, academia, biotechnology companies, and regulatory agencies have been formed to promote research and development of novel antiparasitic drug candidates (3). Within a 4-year, EU-funded project, the consortium PDE4NPD (https://cordis.europa.eu/project/id/602666/reporting) focused on new therapies for CD, sleeping sickness, leishmaniasis, and schistosomiasis by targeting parasite phosphodiesterases. Phosphodiesterases (PDEs) are a group of hydrolases that catalyze the hydrolysis of cyclic nucleotides (cAMP and cGMP) to their inactive 5′-AMP and 5′-GMP forms, respectively (4). These second messengers play a role in eukaryotic cell functions such as signal transduction and gene expression, cell proliferation and differentiation cycles, inflammation state, and cell death processes like apoptosis (5).

Signaling cascades promoted by cAMP are present in almost all organisms from humans to bacteria, including the Kinetoplastida parasites (6), and among the most conserved elements are the PDEs (7). PDEs structurally related to human isoforms have been reported in all trypanosomatids (PDEA, PDEB1/B2, PDEC, and PDED) (8). In Trypanosoma cruzi, the causative agent of CD, PDEC is localized into the contractile vacuole complex (CVC), which is essential for parasite fitness, and its inhibition causes dysregulation of the osmotic control, indicating the potential use of PDEs inhibitors as drug targets in this parasite (9, 10). The fact that control of cyclic nucleotide metabolism is essential in kinetoplastid parasites, and the catalytic domains of their PDEs align closely with human homologues (10), justifies the potential repurposing of PDE inhibitor classes as potential antiparasitic agents (11), since such inhibitors have been successfully used for a large spectrum of clinical conditions, such as intermittent claudication, chronic obstructive pulmonary disease, erectile dysfunction (12), and psoriatic arthritis (13).

Nitrogen heterocyclic compounds have been extensively used as pharmacological scaffolds in myriad conditions (14, 15). Among them, derivatives containing the pyrazolone nucleus were shown to be active on bacteria and fungi (16), on the nervous system (17, 18), and as anticancer agents (19). The pyrazolone NPD-227 is a potent in vitro anti-T. cruzi agent with 50% effective concentration (EC₅₀) values ranging between 0.1 and 0.5 μM against intracellular forms belonging to different discrete typing units (DTU-II for the Y strain and DTU-I for the Colombiana strain) (20).

In this context, the efficacy and selectivity of 17 pyrazolones (Fig. 1 and Table 1) were phenotypically evaluated in vitro and validated through cAMP measurements. NPD-227, the most potent of these and able to sterilize parasite cultures, was further tested on acute mouse models of T. cruzi infection, under monotherapy as well as under coadministration schemes, and displayed promising results.

FIG 1.

Chemical structure of pyrazolone derivatives.

TABLE 1.

Chemical characterization of NPD-126, NPD-134, and NPD-977

Compound	Analytical data
NPD-126	1H NMR (500 MHz, Chloroform-d) δ 8.78 (d, J = 2.3 Hz, 1H, Ar-H), 8.60 (dd, J = 4.8, 1.7 Hz, 1H, Ar-H), 7.96 – 7.86 (m, 3H, Ar-H), 7.43 – 7.35 (m, 1H, Ar-H), 7.05 (d, J = 8.5 Hz, 1H, Ar-H), 5.23 (hept, J = 6.6 Hz, 1H, C-H i-Pr), 3.88 (s, 3H, MeO), 1.57 (s, 6H, 2× CH3), 1.42 (d, J = 6.6 Hz, 6H, CH3, i-Pr). 13C NMR (126 MHz, Chloroform-d) δ 201.9, 169.3, 158.5, 150.1, 148.4, 137.1, 133.6, 129.6, 128.6, 127.8, 123.5, 123.2, 111.3, 62.0, 55.9, 49.7, 26.9, 20.1. LC-MS (ESI) m/z found: 354 [M + H]+; retention time: 4.97 minutes. HRMS-ESI [M + H]+ calculated for C20H24N3OS: 354.1635, found: 354.1628.
NPD-134	1H NMR (500 MHz, Chloroform-d) δ 8.75 (s, 1H, Ar-H), 8.65 (d, J = 4.5 Hz, 1H, Ar-H), 7.90 (d, J = 2.1 Hz, 1H, Ar-H), 7.87 (d, J = 7.9 Hz, 1H, Ar-H), 7.81 (dd, J = 8.6, 1.9 Hz, 1H, Ar-H), 7.42 (dd, J = 7.8, 4.9 Hz, 1H, Ar-H), 7.32 (d, J = 8.6 Hz, 1H, Ar-H), 6.45 (t, J = 72.9 Hz, 1H, O-CF2H), 4.51 (hept, J = 6.7 Hz, 1H, C-H, i-Pr), 1.48 (s, 6H, 2× CH3), 1.36 (d, J = 6.7 Hz, 6H, CH3, i-Pr). 13C NMR (126 MHz, Chloroform-d) δ 177.70, 160.04, 149.60, 148.88, 148.86, 148.84, 148.79, 137.07, 132.61, 130.92, 129.07, 129.02, 127.25, 123.31, 119.88, 117.61, 115.53, 113.44, 48.71, 45.40, 22.55, 20.84. LC-MS (ESI) m/z found: 374 [M + H]+; retention time: 4.15 minutes. HRMS-ESI [M + H]+ calculated for C20H22F2N3O2: 374.1675, found: 354.1660.
NPD-977	1H NMR: (600 MHz, Chloroform-d) δ 8.81 (s, 1H, Ar-H), 8.65 (dd, J = 4.8, 1.5 Hz, 1H, Ar-H), 8.02 (dd, J = 2.2, 0.9 Hz, 1H, Ar-H), 7.95 (dd, J = 7.8, 2.1 Hz, 1H, Ar-H), 7.72 (dd, J = 2.2, 1.0 Hz, 1H, Ar-H), 7.44 – 7.38 (m, 1H, Ar-H), 4.50 (hept, J = 6.7 Hz, 1H, C-H, i-Pr), 3.51 (s, 3H, OMe), 1.47 (s, 6H, 2× CH3), 1.35 (d, J = 6.7 Hz, 6H, CH3, i-Pr). 13C NMR: (151 MHz, Chloroform-d) 177.64, 159.38, 155.64, 149.36, 149.04, 136.66, 133.53, 133.06, 130.98, 129.02, 127.63, 123.42, 118.83, 76.79, 60.83, 48.67, 45.43, 22.54, 20.84. LC-MS (ESI) m/z found: 368 [M + H]+; retention time: 4.50 minutes. HRMS-ESI [M + H]+ calculated for C21H26N3O3: 368.1969, found 368.1958.

RESULTS

Following a previously standardized flow chart (21), the compounds were first assayed against intracellular amastigotes of the Tulahuen strain (DTU VI) transfected with the β-galactosidase gene. As presented in Table 2, 14 out of the 17 compounds revealed higher or similar activity (EC₅₀ ranging from 0.17 to 3.3 μM) compared to benznidazole (Bz) (2.85 μM) and 10 compounds also demonstrated greater selectivity (51 to 599). Cardiotoxicity was assessed by 24 h of incubation of primary cardiac cells (CC) and most compounds did not exert toxic effects up to 200 μM. However, longer incubations (>96 h) using L929 cell cultures revealed moderate toxicity, with NPD-122, NPD-373, and NPD-616 displaying 50% lethal concentration (LC₅₀) values of 56, 30, and 36 μM, respectively.

TABLE 2.

Activity (EC₅₀ ± SD) and selectivity (SI) of pyrazolone derivatives against intracellular amastigotes of Tulahuen strain and bloodstream trypomastigotes of Y strain^a

Compound	Intracellular amastigotes Tulahuen strain, 96 h		Bloodstream trypomastigotes Y strain, 24 h
	EC50 in μM ± SD	SI	EC50 in μM ± SD	SI
NPD-095	0.69 ± 0.3	171	>50	ND
NPD-101	0.44 ± 0.06	224	29.1 ± 9.7	>7
NPD-104	0.61 ± 0.09	123	>50	ND
NPD-122	3.3 ± 1.6	17	>50	ND
NPD-126	0.53 ± 0.28	116	29.9 ± 10.9	3
NPD-134	2.3 ± 0.4	51	>50	1.6
NPD-185	2 ± 0.4	77	>50	ND
NPD-193	1 ± 0.17	>200	>50	ND
NPD-206	2.5 ± 0.5	>80	40.4 ± 0.0	4.3
NPD-224	>10	ND	>50	ND
NPD-225	>10	ND	>50	ND
NPD-227	0.36 ± 0.32	278	>50b	ND
NPD-228	0.17 ± 0.1	599	>50	ND
NPD-373	1.12 ± 0.8	27	>50	ND
NPD-616	3.3 ± 2.1	11	>50	ND
NPD-845	2.62 ± 1.9	38	40.7 ± 3.2	1.2
NPD-977	>10	ND	>50	ND
Benznidazole	2.85 ± 0.84	51	12.9 ± 1.9	77

^aData are the mean ± SD of 3 experiments. SI, selectivity index: EC₅₀ (CC cells)/EC₅₀ (T. cruzi), determined after 96 h and 48 h of incubation with L929 and CC cells, respectively; ND, not defined.

^bSource: Sijm et al., 2019 (20).

To investigate the effect on the other relevant parasite form for mammalian infection, bloodstream trypomastigotes (BT) of the Y strain (DTU II) were incubated for 24 h at increasing compound concentrations. None presented higher potency than Bz (EC₅₀ = 12.9 ± 1.9 μM), with the most active pyrazolones NPD-101 and NPD-126 showing EC₅₀ values of 29.1 ± 9.7 and 29.9 ± 10.9 μM, respectively (Table 2). As their different potency on the parasite forms could be related to the different strains (Tulahuen for amastigotes and Y strain for BT), the most active compounds (NPD-227, NPD-228, NPD-616, and NPD-845) were screened against intracellular forms of the Y strain inside cardiac cells. The compounds showed similar activity against intracellular forms regardless of the parasite strain, with the EC₅₀ upon the Y strain ranging from 0.1 to 1 μM, strongly suggesting that the differential effect must be related to the parasite form. Although not being able to kill BT, the pyrazolones induced profound alterations in parasite morphology, triggering cellular rounding, as observed in the fluorescence analysis (Fig. 2A to D) by labeling parasite surface and their DNA with ConA-FITC (fluorescein isothiocyanate) and DAPI (4′,6-diamidino-2-phenylindole), respectively. To explore whether these altered BT maintained their ability to invade mammalian cells, additional assays were run using NPD-227. After exposure at 60 and 200 μM NPD-227 for 24 h, surviving and rounded parasites were used to infect CC at the same parasite:host cell ratio as untreated ones. Compared to the untreated control, NPD-227 significantly reduced (by ∼50%) the percentage of infection, the number of parasites per cell, and the corresponding endocytic index (Fig. 2E to G).

FIG 2.

Treatment with NPD-227 alters morphology and infective potential of bloodstream trypomastigotes. As evidenced by ConA-FITC (A and C) and DAPI (B and D), NPD-227 causes cellular rounding of the treated parasites (C, arrow), while the characteristic long shape forms are noticed in untreated samples (A, asterisks). Treatment with this compound prior to infection of cardiac cultures (10:1 parasite:cell ratio) also seems to impair their infective capacity, as observed by light microscope images of untreated control (E), 60 μM (F), and 200 μM (G) of NPD-227 showing infected cells (arrows).

Pretreatment of the host cells with the pyrazolones was performed to rule out that the different effect of the PDE inhibitors could be due to their impact on host cell physiology, thereby modulating their microbicidal ability against the intracellular forms. Briefly, uninfected L929 cultures were treated for 24 h with the PDE inhibitors at the concentration of their respective EC₅₀ values and, after washing, infected with trypomastigotes (Tulahuen strain) for 96 h in drug-free medium. Positive controls were performed in the presence of the diamidine DB569. Pretreatment with the pyrazolones prior to infection caused a maximum reduction of 10% in the infection index (for NPD-227 and NPD-616), while DB569 resulted in 85% decrease. The lack of drug interference on the host cell physiology was further corroborated by the assays using extracellular amastigotes (Y strain) obtained from the supernatant of infected CC (50:1 parasite:cell ratio). NPD-227 sustained the same activity as observed for intracellular forms (EC₅₀ value of 2 ± 1 μM).

To evaluate whether NPD-227 was able to cause irreversible damage on parasite viability and how fast it was acting, “wash-out assays” were performed. L929 cultures infected with the Tulahuen strain were treated with the PDE inhibitors and, after 6, 12, and 24 h, the infected cultures were washed with fresh culture medium and further incubated for 72 h in drug-free conditions. The positive control was the bis-amidine DB745, which is considered a fast killer (22). By comparing the EC₅₀ values in this model with the data from 96 h of treatment, it was possible to observe that NPD-227 already presents a high potency after 24 h of exposure (EC₅₀ = 1.9 ± 1.7). Another important point was related to its ability to induce sterility in T. cruzi-infected cultures. In these assays, infected L929 cultures were treated with NPD-227 for 96 h, washed, and incubated an additional 168 h in the absence of drugs. The analysis under the light microscope for the presence of the parasite in the supernatant revealed that, although infected and untreated controls exhibited a time-dependent release of parasites, the samples treated with this pyrazolone did not reveal the presence of the T. cruzi, neither in the supernatant nor in the cytoplasm of host cells (data not shown).

To validate the parasite target, the cAMP levels of extracellular amastigotes from T. cruzi (Y strain) was measured in untreated and NPD-227-treated parasites (Fig. 3). NPD-227 induced a significant increase in the intracellular cAMP content at 2× and 5× EC₅₀ after 2.5 h of incubation (P values of 0.0096 and 0.0197, respectively).

FIG 3.

Effect of NPD-227 on cAMP levels of extracellular amastigotes from T. cruzi Y strain. Graph expresses the measurement of cAMP in the intracellular milieu of untreated and PDE inhibitors-treated parasites incubated with 2× and 5× EC₅₀ values for 2.5 h of incubation time. *, P < 0.05; **, P < 0.01.

Since combined therapy is considered a successful strategy, the activity on intracellular amastigotes of the Tulahuen strain using NPD-227 associated with Bz in a fixed-ratio method was investigated, revealing a mean sum of fractional inhibitory concentration indexes (xΣFICI) of 1.01, conferring a status of no interaction.

As a proof-of-concept, NPD-227 was next evaluated in a mouse model for T. cruzi acute infection using ABT (an inhibitor of cytochrome P₄₅₀ enzymes) once a day, while NPD-227 was given every 6 h (four times a day). For the efficacy studies, ABT and NPD were administered alone and under a combined therapeutic scheme using a suboptimal dose of Bz (10 mg/kg). NPD-227 alone given in association or not with ABT did not reduce parasitemia nor mortality rates (Fig. 4A and B). On the other hand, when infected mice were treated for 5 days with the combination of Bz plus NPD-227, we found a decrease of ∼90% in the parasitemia peak, leading to 100% animal survival, while Bz (at 10 mg/kg) alone led to a reduction of 72% in the parasitemia peak and 40% mortality (Fig. 4A and B).

FIG 4.

In vivo efficacy of monotherapy and combined therapy NPD-227 and benznidazole. Parasitemia (A and C) and mortality (B and D) curves of two independent assays. Use of 1-aminobenzotriazole (ABT) is assessed in graphs (A) and (B). *, P < 0.05; **, P < 0.01; #, P < 0.05 compared to Bz 10 mg in monotherapy.

Considering these encouraging results in vivo, a second set of assays was conducted using Bz (10 mg/kg) given in association with NPD-227 (10 mg/kg) twice daily for 5 days, adopting two different schemes: starting the drug administration at either (i) the 1st dpi or (ii) the 5th dpi, which distinguishes between abortive treatment and treatment of an established infection. Controls using both Bz (10 and 100 mg/kg) and NPD alone (10 mg/kg) were run in parallel. The combined scheme with NPD-227 at 10 mg/kg plus Bz at 10 mg/kg started at 5 dpi resulted in 86% reduction of parasitemia at 8 dpi, while Bz alone at 10 mg/kg led to a decrease of 64%, and there was no effect by the pyrazolone alone (Fig. 4C). A survival rate of 85% was observed for the combination of Bz + NPD-227 starting at 5 dpi, while other treatment approaches did not have any impact for both parasitemia and mortality rates (Fig. 4D).

DISCUSSION

In this study, 17 new pyrazolone derivatives were profiled for activity against T. cruzi forms belonging to distinct DTUs. All had a greater effect against intracellular amastigotes (Tulahuen strain DTU VI) compared to BT (Y strain DTU II), leading us to question if this differential response was related to the strain or to the parasite stage. To answer that, some of the most active compounds on amastigotes were evaluated on the Y strain, whereby no significant variation was noted, hence arguing against variability in drug susceptibility according to parasite strain. The present findings corroborate previous studies (20) showing that the pyrazolone NPD-227 has quite similar potency upon intracellular amastigotes regardless of the studied strain (Y and Colombiana, with EC₅₀ values of 0.1 and 0.5 μM, respectively). Prior exposure of mammalian cells to NPD-227 before infection by T. cruzi did not impact clearance or reduction of parasite intracellular load, while pretreatment using the aromatic diamidine DB569 induced a 85% reduction of parasite burdens. The data using DB569 corroborate previous studies using another diamidine, DB750, showing a significant decrease in Neospora caninum proliferation after prior treatment of the host cells for 6, 12, or 24 h (23). The authors offered several possibilities to explain their findings: (i) that these compounds were still available at the intracellular content after rinsing the host cell cultures; (ii) that the drug exposure induced some microbicidal arsenal in the host cell to control parasite multiplication; or (iii) the aromatic compound impaired some cellular function or structure essential for parasite proliferation. Additional evidence for the direct effect of the pyrazolone toward the parasite viability is the fact that NPD-227 sustained antiparasitic activity on free amastigotes, displaying a similar potency as that observed against intracellular amastigotes. Another important characteristic of the studied pyrazolones is the ability to induce sterilization (i.e., cure) of in vitro infection, suggesting a trypanocidal rather than growth inhibitory (trypanostatic) effect, as incubation with the EC₉₀ concentration irreversibly sterilized cultures even after removal of the compound and a continued culture for 168 h.

The high selectivity associated with the trypanocidal potential and the ability to induce sterility in T. cruzi-infected-cultures are both highly desirable characteristics for a novel candidate for a CD drug (24). Although NPD-227 was not directly active on bloodstream forms (no induction of lysis), this compound altered the parasite morphology (leading to a rounding aspect) and strongly impaired the ability of the pathogen to invade mammalian cells. The rounding of parasites induced by NPD-227 suggests an impact on the osmoregulation system, already described as an outcome of PDEC inhibition in T. cruzi (9). Indeed, we have previously shown that exposure to other classes of PDE inhibitors, specifically imidazoles (25) and phthalazinones (26), causes profound ultrastructural damage in the Golgi, flagellar pocket, and plasma membrane of T. cruzi. These observations are suggestive of osmotic stress and may be related to increased levels of cAMP. In this context, we investigated the possibility of a PDE target for NPD-227 and observed that it induced a significant increase in the cellular cAMP concentration, confirming the targeting of at least one PDE parasite isoform, consistent with reports on other PDE inhibitor chemotypes (25, 26).

Since the currently available drugs (Bz and nifurtimox) present important limitations, especially in the later chronic phase of CD, there is an urgent need to explore new therapeutic options (27). Combined therapy has been proposed as an ideal approach, considering its potential to enhance efficacy and delay drug resistance by acting upon different cellular targets (28, 29). This strategy represents an important tool to optimize the potential of PDE inhibitors within combined approaches, particularly in view of their incomplete effects on trypomastigotes, and our successful experience with some combination approaches (26, 30–32). In the in vivo efficacy proof of concept, coadministration of Bz and NPD-227 resulted in improved efficacy in acute mouse models of T. cruzi infection, leading to ∼90% reduction in parasitemia and lower mortality rates (0 to 17%) than seen with either drug alone (Bz and NPD monotherapy: 40 to 63% and 40 to 50%, respectively). This improved outcome was consistent with the additive effects of NPD-227 and Bz observed in vitro. Another interesting point is the absence of effect when the combination therapy was started at the 1st dpi, which may be due to better potency of NPD-227 toward the highly multiplicative intracellular parasites, since these stages are found at the 5th dpi when the infection is already well established and parasites are disseminated into different tissues and organs.

In summary, these in vitro and in vivo assays demonstrate the potential use of PDE inhibitors for drug development of novel candidates for therapy of Chagas disease, especially in combination approaches, and aim to identify safer and more effective options for the millions of Chagasic patients.

MATERIALS AND METHODS

Compounds.

Twenty millimolar stock solutions (in 100% dimethyl sulfoxide [DMSO], maximum final concentration of 1% in assays) were prepared using the set of 17 pyrazolone test compounds (Fig. 1) synthetized as reported (20), except NPD-977, NPD-126, and NPD-134, which were prepared by analogous methods and their chemical characterization is given in Table 1. As a reference drug, benznidazole (N-benzyl-2-[2-nitroimidazol-1-yl]acetamide2-nitroimidazole) (Laboratório Farmacêutico do Estado de Pernambuco, Brazil) was used in parallel (25). For in vivo assays, the cytochrome P450 (CYP₄₅₀) inhibitor 1-aminobenzotriazole (ABT) (Sigma-Aldrich, MO, USA) was diluted in distilled sterile water (33). NPD-227 was prepared in 10% DMSO in distilled water while Bz was formulated using 3% Tween₈₀ in distilled water (30). The synthesis of the arylimidamide DB745 and the aromatic diamidine DB569 were performed as reported (34, 35). Negative controls were run in parallel using DMSO at 0.6% and 10% for in vitro and in vivo assays, respectively.

Mammalian cells.

Primary cardiac cells (CC) cultures were obtained from mouse embryos and plated onto coverslips in 96- or 24-well plates coated with 0.01% gelatin (36). L929 fibroblasts lineages were grown as described (21).

Parasites.

T. cruzi bloodstream trypomastigotes (BT) of the Y strain were obtained by cardiac puncture of infected Swiss Webster mice on the parasitemia peak (36, 37). Trypomastigotes of the Tulahuen strain expressing β-galactosidase were collected from the supernatant of infected L929 cultures (21). Extracellular amastigotes of the Y strain were obtained from the supernatant of infected cultures of CC upon high rate infection (50:1 parasite:cell) and nutritional stress (38).

Cytotoxicity assays.

Noninfected CC/L929 cultures were incubated at 37°C for 24 h/96 h with increasing compound concentrations (up to 200 μM) diluted in supplemented RPMI/DMEM (Dulbecco’s modified Eagle medium). Morphology was evaluated by light microscopy and cellular viability was determined upon PrestoBlue (CC) and alamarBlue (L929) addition. The results are expressed as the difference in reduction between treated and nontreated cells adopting the manufacturer’s instructions. The LC₅₀ (minimum concentration that reduces 50% of cellular viability) was determined (25).

Trypanocidal activity

• i.Bloodstream trypomastigotes. BT of Y strain (5 × 10⁶/mL) were incubated for either 2 or 24 h at 37°C in RPMI in the presence or absence of 1:3 serial dilutions of the compounds (0 to 50 μM) for determination of parasite death rates through the direct quantification of live parasites by light microscopy, allowing the EC₅₀ (minimum compound concentration that reduces 50% of the number of parasites) to be calculated (25).
• ii.Intracellular amastigotes. T. cruzi-infected L929 cultures of the Tulahuen strain (2 h of interaction with trypomastigotes following 48 h of infection) were incubated for 96 h at 37°C with 2-fold serial dilutions (0 to 10 μM) in RPMI. After incubation, 500 μM chlorophenol red glycoside (CPRG) in 0.5% Nonidet P40 was added and the plate was incubated for 18 h at 37°C. The absorbance was measured spectrophotometrically at 570 and 600 nm. Uninfected cells and infected cells treated only with vehicle and/or with Bz were run in parallel as controls. Results are expressed as percent parasite growth inhibition by comparing optical density data from treated and control cell cultures. Next, selected compounds were evaluated against intracellular amastigotes (Y strain). After 24 h of infection (10:1 parasite:cell ratio), the cultures were exposed to the test compounds that were predetermined to be nontoxic to the host cells for 48 h at 37°C. The CC were then rinsed with saline, fixed with Bouin for 5 min, stained with Giemsa and evaluated by light microscopy. The percentage of infected host cells and the number of parasites per cell were determined for the calculation of the infection index and the EC₅₀ values determined from a sigmoid plot of infection index against test compound concentration.
• iii.Extracellular amastigotes. For the analyses on extracellular amastigotes (Y strain), the parasites were incubated (5 × 10⁶/mL) with increasing concentrations of the selected compounds for 24, 48, and 96 h and the parasite death rates were determined by direct quantification of the parasites by light microscopy for EC₅₀ calculation. Selectivity index (SI) is expressed as the ratio between LC₅₀ (toxicity for mammalian cells) and the EC₅₀ (activity on the parasite).

Reversibility assays.

Tulahuen strain-infected L929 cultures were incubated in the presence of increasing concentrations of selected compounds for 6 to 24 h and then washed and incubated with fresh culture medium without any compound for another 72 h. The absorbance was evaluated as described above. As positive control, the bis-arylimidamide DB745 was used because previous reports demonstrated this amidine compound to be a fast and irreversible killer for T. cruzi (22).

Pretreatment effect.

Noninfected L929 cultures were treated with the inhibitors at their respective EC₅₀ concentrations for 24 h, rinsed to remove the drugs, and infected for 2 h with trypomastigotes of the Tulahuen strain. The cultures were then rinsed again with saline to remove noninternalized parasites and the infection followed for 96 h (21). Furamidine phenyl-substituted analog DB569 was used as positive control as previous studies demonstrated the ability of aromatic diamines to induce a memory effect in vitro during the infection of host cells with Neospora caninum and Toxoplasma gondii (23, 39).

Sterility evaluation.

L929 cells infected with the Tulahuen strain were treated for 96 h with 2 μM NPD-227 (equivalent to its EC₉₀), then washed and incubated for 168 h with fresh drug-free medium. The cultures were monitored by light microscopy and the number of parasites released into the supernatant was quantified. At the end of the assay, the intracellular parasite load was also determined following the protocol described above (21).

Morphology analysis by fluorescence microscopy.

Aliquots of 10 × 10⁶ bloodstream forms (5 × 10⁶ parasites/ml) were incubated for 4 h with 200 μM NPD-227, rinsed twice, and incubated for a further 30 min with 0.5 μg/ml of Concanavalin A-FITC (ConA-FITC; Sigma-Aldrich, MO, USA) at 37°C. Parasites were then washed with saline, fixed for 1 h with 4% paraformaldehyde at room temperature, adhered to coverslips previously coated with poly-l-lysine, and further stained with 1 μg/ml of 4,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich, MO, USA). The samples were mounted with 2.5% DABCO and analyzed by fluorescence microscopy (37).

Infective parasite profile.

BT were incubated (5 × 10⁶ parasites/ml) for 24 h at 37°C/5% CO₂ with a concentration of NPD-227 that induced altered morphology (>60 μM), washed, and quantified through light microscopy, and then used for the infection of CC in 24-well plates (10:1 parasite:cell ratio). Following 48 h of interaction, the infected cultures were rinsed, fixed with Bouin for 5 min, stained with Giemsa, and evaluated by light microscopy. The percentage of infected host cells and the number of parasites per cell were determined for the calculation of the infection index as reported (26).

In vitro combination therapy.

A fixed-ratio method (40) was used to investigate drug interactions against intracellular amastigotes of the Tulahuen strain by combining the selected compound with Bz. EC₅₀ values were used to determine the top concentrations of the individual ratios so that the midpoint of a seven-point 2-fold dilution series corresponded to the EC₅₀. The fixed ratios of 5:0, 4:1, 3:2, 2:3, 1:4, and 0:5 were used, as previously reported (41).

Determination of FICI, classification of interaction and isobologram construction.

Fractional inhibitory concentration index (FICIs) and the sum of the FICs (ΣFICs) were calculated using the formula: FICI of NPD-227 = EC₅₀ of NPD-227 in combination/EC₅₀ of NPD-227 alone. The same equation was applied to Bz. ΣFICIs were calculated as the FICI of NPD-227 plus the FICI of Bz. An overall mean ΣFICI (xΣFICI) was calculated for each combination and used to classify the nature of the interaction, as follows: “synergy” for xΣFICI of ≤0.5; “antagonism” for xΣFICI of >4.0; and “no interaction” for xΣFICI of 0.5 to 4.0 (42). Isobolograms were constructed by plotting the EC₅₀ of NPD-227 against the EC₅₀ of Bz. Maximum concentration of NPD-227 was 2.88 μM and 21.6 μM for Bz (41).

cAMP measurement.

Extracellular amastigotes of Y strain (20 × 10⁶ parasites/ml) were treated with NPD-227 at 2× and 5× the EC₅₀ for 2.5 h at 37°C. After the incubation times, the tubes were centrifuged and the supernatant was collected. The pellet was resuspended in 100 μl of HCl 0.1 M and incubated at 4°C for 20 min, followed by centrifugation and collection of the supernatant. Samples were analyzed using a cAMP enzyme-linked immunosorbent assay (ELISA) kit (Cayman Chemicals, MI, USA), accordingly to the manufacturer instructions (25, 26, 43).

In vivo assays and ethical statement.

Male Swiss Webster mice (18 to 23 g) were obtained from the Instituto de Ciência e Tecnologia em Biomodelos (ICTB/Fiocruz; Rio de Janeiro, Brazil). A maximal of six mice were housed per cage, kept in a conventional room at 20 to 24°C under 12 h/12 h light/dark cycle. Sterilized water and chow were provided ad libitum. All procedures followed the guidelines compliant with the Fiocruz Committee of Ethics for the Use of Animals (CEUA LW16/14) (31).

In vivo efficacy.

Male mice (18 to 20 g, n = 6 per group) were infected i.p. with 10⁴ BT for two protocols as follows. (i) NPD-227 was given at 10 mg/kg i.p. 4 times a day in association or not with ABT (100 mg/kg once a day, p.o.). Combination therapy was assessed using Bz (10 mg/kg p.o.) and NPD-227 (10 mg/kg i.p.) 4 times a day. (ii) NPD-227 (10 mg/kg twice a day, i.p.) was given in combination with Bz (10 mg/kg twice a day, p.o.) starting at day 1 and day 5 postinfection (dpi). Only animals with positive parasitemia were included and monotherapy of ABT, NPD-227, and Bz at the tested doses was run in parallel. Bz was also given at the optimal dose (100 mg/kg p.o.) in combination with the same scheme for NPD-227. All treatments followed a 5-day dosing regimen. Uninfected and T. cruzi-infected mice treated only with distilled sterile water and 10% DMSO (vehicle) served as controls (31).