Nitrotriazole-based acetamides and propanamides with broad spectrum antitrypanosomal activity.

3-Nitro-1H-1,2,4-triazole-based acetamides bearing a biphenyl- or a phenoxyphenyl moiety have shown remarkable antichagasic activity both in vitro and in an acute murine model, as well as substantial in vitro antileishmanial activity but lacked activity against human African trypanosomiasis. We have shown now that by inserting a methylene group in the linkage to obtain the corresponding propanamides, both antichagasic and in particular anti-human African trypanosomiasis potency was increased. Therefore, IC50 values at low nM concentrations against both T. cruzi and T. b. rhodesiense, along with huge selectivity indices were obtained. Although several propanamides were active against Leishmania donovani, they were slightly less potent than their corresponding acetamides. There was a good correlation between lipophilicity (clogP value) and trypanocidal activity, for all new compounds. Type I nitroreductase, an enzyme absent from the human host, played a role in the activation of the new compounds, which may function as prodrugs. Antichagasic activity in vivo was also demonstrated with representative propanamides.

Introduction

American trypanosomiasis (Chagas disease), human African trypanosomiasis (HAT disease or sleeping sickness) and Leismaniasis are parasitic infections. They are considered neglected tropical diseases (NTD) because they constitute a major health problem in particularly poor countries around the world [1]. HAT disease (caused by Trypanosoma brucei rhodesiense and T. b. gambiense) is endemic throughout sub-Saharan Africa while Chagas disease (caused by T. cruzi) affects populations in South and Central America. In contrast, leishmaniasis (caused by Leishmania species) is prevalent in many sub-tropical and tropical regions of the world, recently expanding in non-tropical regions as HIV/AIDS co-infection [2]. These insect transmitted diseases affect more than 20 million people and are responsible for more than 110,000 deaths per year [3]. HAT disease ranks high on the list of NTD because it is fatal if untreated and the treatment options are limited. The incidence of T. cruzi infection has significantly declined recently, due to implementation of vector control initiatives, however, the number of cases in non-endemic sites (United States, Australia, Europe and Japan) is rising, primarily due to human and vector migration and contaminated blood transfusions [4], [5], [6].

The treatment of neglected diseases is based on drugs with serious limitations. Thus, nifurtimox (Nfx) and benznidazole (Bnz), the two currently used medications for Chagas disease (Fig. 1) are associated with limited efficacy, severe toxicity and long treatment requirements [7], [8]. Similarly, drugs used to treat HAT and leishmaniasis are highly toxic (e.g. melarsoprol, antimonials), or require i.v. administration (e.g. melarsoprol, suramin, DFMO, antimonials) resulting in severe side effects, or are of high cost (e.g. DFMO, liposomal amphotericin B, miltefosine and paromomycin) [9], [10], [11]. Therefore, new effective, safe and affordable drugs are urgently needed for the treatment of these neglected diseases.

Fig. 1

The chemical structure of Bnz, Nfx and the general structure of representative classes of 3-nitrotriazole-based trypanocidal compounds (A: piperazides, B & C: amides, and D: sulfonamides).

We have demonstrated that various chemical classes of 3-nitro-1H-1,2,4-triazole-based compounds, including aliphatic/aromatic amines, amides, sulfonamides, carbinols, piperazines and piperazides (some of them shown in Fig. 1) exhibit excellent antichagasic activity both in vitro and in vivo, with several analogs also showing appreciable anti-T. b. rhodesiense activity in vitro. [12], [13], [14], [15], [16], [17], [18] Futhermore, 3-nitrotriazole-based compounds are significantly more potent and less toxic than their 2-nitroimidazole-based counterparts [12], [13], [14], [15], [16], [17], [18], [19]. Nfx, Bnz and other nitroheterocyclics work as prodrugs, needing enzymatic activation to exert their trypanocidal activity [20], [21], [22], [23]. We have previously shown that 3-nitrotriazole-based compounds are excellent substrates of a type I nitroreductase (NTR), an oxygen-insensitive nitroreductase present in the mitochondrion of trypanosomatids and absent from most other eukaryotes [20], [21], [22], [23] and that part of the trypanocidal activity of these compounds depends on the parasite's expression of type I NTR [12], [13], [15], [16], [17], [18], [24].

Despite the failure of the antifungal drug posaconazole to treat chronic Chagas disease in clinical trials [25], there is still a great interest in developing more specific inhibitors of T. cruzi CYP51 (TcCYP51), the orthologous enzyme of the fungal sterol 14α-demethylase enzyme (CYP51) [26], [27], [28]. Sterol 14α-demethylase is crucial for the formation of viable membranes and the regulation of metabolic processes such as cell growth and division, not only in fungi but also in trypanosomatids [29], [30], [31], [32]. Since the triazole ring plays a significant role in CYP51 inhibition [31], we have previously evaluated 3-nitrotriazole-based amides with a linear, rigid core, as well as 3-nitrotriazole-based carbinols (fluconazole analogs) as bifunctional agents; such compounds can act as substrates for type I NTR in addition to being inhibitors of TcCYP51 [17]. These bifunctional compounds demonstrated remarkable antichagasic activity both in vitro and in an acute murine model [17]. A subclass of such bifunctional antitrypanosomal agents was 3-nitrotriazole-based aryloxyphenylacetamides, in which the 3-nitrotriazole ring is separated from the amidic carbonyl by one methylene-group [24]. 3-Nitrotriazole-based aryloxyphenylacetamides, besides being very potent antichagasic agents in vitro and in vivo, demonstrated also remarkable in vitro activity against L. donovani axenic amastigotes, something that was not seen before with other 3-nitrotriazole-based derivatives [24]. However, these acetamides were only moderately active against T. b. rhodesiense, with poor selectivity for this parasite, despite the fact that they were excellent substrates of TbNTR [24].

Therefore, in the present work we tried to further optimize the class of 3-nitrotriazole-based aryloxyphenylamides with the goal to increase their anti-HAT activity. This was obtained via linkage elongation between the nitrotriazole ring and the amidic carbonyl by inserting one additional methylene group. Thus, novel 3-nitrotriazole-based propanamides were synthesized and screened for antitrypanosomal activity. The novel propanamides were then compared side by side with the corresponding acetamides.

Results and discussion

Chemistry

The structure of the twelve novel compounds is shown in Table 1, together with the structure of previously synthesized analogs (in orange), for comparison purposes. The synthesis of the new compounds in Table 1 is straightforward and based on well-established chemistry, outlined in Scheme 1. Thus, acetamides 2, 3, 13 and 15 were obtained by nucleophilic substitution of the appropriate chloroacetamide 1a-c with the potassium salt of 3-nitro-1,2,4-triazole (and in the case of 3 the potassium salt of 2-nitroimidazole) under refluxing conditions. Similarly, propanamides 6–12 and 14 were obtained by nucleophilic substitution of the appropriate bromopropanamide 4a-h with the potassium salt of 3-nitro-1,2,4-triazole under refluxing conditions. During the synthesis of bromopropanamides 4a-h, the acrylamides 5a-h were also formed as β-elimination byproducts, which however were not isolated due to a similar Rf value they share on TLC with bromopropanamides 4a-h. Fortunately, the acrylamides 5a-h not only did not prevent the next step but in fact they furnished as starting materials for the formation of the final propanamides through Michael addition. The final compounds were obtained in 45–78% yield.

Table 1

In vitro antiparasitic activity, host toxicity and physical properties of tested compounds.

Scheme 1

Synthesis of compounds on Table 1.

Chloroacetamides 1a-c and bromopropanamides 4a-h were prepared from appropriate, commercially available arylamines and chloroacetyl chloride or 3-bromopropanoyl chloride, respectively, in the presence of triethylamine, at room temperature.

Biological evaluation

Antiparasitic activity

Compounds in Table 1 were screened for anti-parasitic activity against three trypanosomatids: T. cruzi, T. b. rhodesiense and Leishmania donovani and compared with previously made analogs shown in orange and designated with a. The concentration of compound that inhibits parasite growth by 50% (IC50) was calculated from dose response curves for each parasite (Table 1). In addition, compounds were tested for toxicity in L6 rat skeletal myoblasts, the host cells for T. cruzi amastigotes, in order to calculate a selectivity index for each parasite [SI = IC50L6/IC50parasite] (Table 1). The TDR (Special Programme for Research and Training in Tropical Diseases, World Health Organization) criteria were adopted to interpret antiparasitic activity and selectivity [33].

According to the TDR criteria, all 3-nitrotriazole-based analogs were selectively ‘active’ antichagasic agents (IC50 of <4.0 μM, SI of ≥50), whereas the 2-nitroimidazole-based analog 3 was ‘moderately active’ against T. cruzi but with an unacceptable SI value (IC50 between 4.0 and 60 μM, SI < 50). The 3-nitrotriazole-based propanamides (6–12) were similar to or more potent antichagasic agents than the corresponding 3-nitrotriazole-based acetamides (6a-11a and 13), most of the latter being previously evaluated (highlighted in orange and designated with a).

All 3-nitrotriazole-based propanamides were ‘active’ (IC50 < 0.5 μM) (6–11) or ‘moderately active’ (IC50 between 0.5 and 6.0 μM) (12 and 14) anti-HAT agents that displayed acceptable selectivity (SI of ≥100) towards T. b. rhodesiense (Table 1). Interestingly, these compounds were up to 214-fold more potent against this parasites than the corresponding acetamides (6a-11a, 13 and 15), with the latter structures being deemed ‘inactive’ (IC50 > 6.0 μM) or ‘moderately active’ at best. The increased anti-HAT activity displayed by the propanamides relative to the acetamides was always accompanied with a favorable shift in selectivity (Table 1).

The 3-nitrotriazole-based acetamide 2 was inactive against T. b. rhodesiense, but still more potent than the corresponding 2-nitroimidazole-based analog 3, whereas compound 2a of equal molecular weight with 2 was 5.8-fold more potent against T. b. rhodesiense than 2, suggesting the importance of the amidic proton.

Previously we have found that 3-nitrotriazole-based aryloxy-phenylamides are selectively ‘active’ against L. donovani amastigotes in vitro (IC50 < 1 μM, SI ≥ 20). Increasing the distance between the nitrotriazole ring and the amidic carbonyl in compounds 6–11 and 14 sustained antileishmanial activity, although a shift to ‘moderate activity’ (IC50 between 1.0 and 6.0 μM) was observed for propanamides 6–8 in comparison with their acetamide-analogs 6a-8a (Table 1).

SAR analysis of antichagasic activity

The 3-nitrotriazole-based acetamide 2, having an N-methyl substituent, was synthesized to be compared with the previously evaluated acetamide 2a, which is unsubstituted on the amidic nitrogen but of equal molecular weight with 2. This N-substitution in 2 resulted in a 1.6-fold decrease in antichagasic activity compared to 2a, but also in a 2-fold decrease in toxicity against L6 cells, despite the fact that 2 was slightly more lipophilic than 2a (clogP 1.99 vs 1.83). Thus, the SI of 2 was also slightly higher than that of 2a. However, both compounds were slightly more potent antichagasic agents than Bnz (Table 1).

Acetamide 2 was also compared as antichagasic agent with its 2-nitroimidazole-based counterpart 3. In keeping with our previous findings the former structure was shown to be 32-fold more potent towards Trypanosoma cruzi amastigotes than the latter. In addition, compound 3 was shown to display a higher degree of toxicity towards L6 cells compared to 2, presumably due to its higher lipophility (Table 1). Together these initial phenotypic screens demonstrate that 3 had an unfavorable SI score (∼4) relative to that determined for 2 (∼165) with these results suggesting once again that 3-nitrotriazole-based compounds may be better antichagasic agents than 2-nitroimidazole-based analogs, at least in vitro.

Evaluation of the antichagasic properties of the nitrotriazole-based phenyloxyphenyl propanamides (6–8) with their acetamide counterparts (6a-8a) revealed that for each compound pairing similar potencies towards the parasite were observed (Table 1). However, due to their greater lipophilicity, propanamides displayed an increased toxicity towards L6 cells. Despite this issue, the nitrotriazole-based phenyloxyphenyl propanamides exhibited very encouraging SI values for T. cruzi (398–3343). The nitrotriazole-based p-chlorophenyloxyphenyl propanamide 6 was an extremely potent antichagasic agent with an IC50 of 5 nM against T. cruzi amastigotes, slightly more potent than the acetamide analog 6a (IC50 of 8 nM). This was not the case with the fluorinated analogs 7 and 7a and the unsubstituted pairs 8 and 8a. Both acetamides 7a and 8a appeared slightly better as antichagasic agents than their propanamide-analogs 7 and 8 (Table 1). Antichagasic activity (as well as toxicity in L6 cells) decreased from the chloro-to the fluoro-, to the non-substituted phenyloxyphenyl-propanamide or acetamide, presumably as a result of decreasing lipophilicity rather than electronegativity of the substituent, since fluorine is more electronegative than chlorine (Table 1).

The growth inhibitory effect of nitrotriazole-based biphenylamides (9, 9a, 10 and 10a) towards T. cruzi and L6 cells was also assessed (Table 1). There was about 1.5-fold increase in potency of p-cyanobiphenyl-propanamide 9 against T. cruzi parasites compared to acetamide analog 9a. In addition, propanamide 9 was less toxic to L6 cells compared to 9a, despite its slightly higher lipophilicity. This reduced toxicity towards L6 cells coupled with the increased potency towards T. cruzi parasites resulted in a better SI of 9 compared to that of 9a (Table 1). A greater clogP value in m-biphenyl-propanamide 10 resulted in a 2.2-fold better antichagasic activity, with slightly increased toxicity to L6 cells compared to its acetamide analog 10a, without, however, compromising selectivity. A drastic reduction in mammalian cell toxicity was noted for the two cyano-containing biphenylamides 9 and 9a, being >6- and >3.5-fold less toxic towards L6 cells, respectively, relative to unsubstituted compounds 10 and 10a. This reduction in mammalian cell toxicity, which may be related to both the electron withdrawing effect of the p-cyano group (inductive and resonance effects) as well as the para-position of the biphenyls in compounds 9 and 9a, resulted in superior SI values compared to those obtained for compounds 10 and 10a (Table 1).

Antichagasic activity and selectivity was improved also in the m-benzyloxyphenyl propanamide 11, compared to its acetamide analog 11a. The slightly greater lipophilicity of 11 resulted in a slightly increased toxicity towards L6 cells, without, however, compromising selectivity towards the parasite.

Finally, four novel 3-nitrotriazole based phenylisoxazole-acetamides/propanamides (12–15) were also evaluated for antichagasic activity in vitro. The p-fluorophenyl-isoxazole propanamide 12, in which the p-fluorophenyl group is attached in the 5-position of the isoxazole ring was similarly active as an antichagasic agent with the acetamide analog 13. Despite being more lipophilic than 13, propanamide 12 was less toxic towards L6 cells, thus yielding a greater SI compared to 13. In contrast, the p-chlorophenylisoxazole-propanamide 14, in which the p-chlorophenyl group is attached in the 3-position of the isoxazole ring, was 6-fold more active as an antichagasic agent than its acetamide analog 15. Moreover, propanamide 14 was about 10-fold more selective towards this parasite than acetamide 15. The slightly greater clogP value of 14 may have played a role in its increased antichagasic activity but did not explain its decreased cytotoxicity towards L6 cells compared to 15.

Plotting logIC50 values against T. cruzi amastigotes versus clogP values for all novel compounds in Table 1, we obtained a good correlation between antichagasic activity and lipophilicity (Fig. 2). Correlation between lipophilicity and cytotoxicity towards L6 cells lead to an R2 of only 0.62, indicating that lipophilicity alone did not account for toxicity.

Fig. 2

Correlation between lipophilicity (clogP values) and activity against T. cruzi (logIC50 values) of all novel compounds in Table 1.

The amidic pKa value does not seem to play a role in the antichagasic activity of propanamides versus acetamides although more acidic amides seem to be less potent antichagasic agents (Table 1).

All novel compounds in Table 1 were 1.4- to 441-fold better antichagasic agents than Bnz (Table 1).

SAR analysis of anti-HAT activity

Evaluation of the growth inhibitory effects of the 3-nitrotriazole-based acetamides 2 & 2a and their 2-nitroimidazole-based analog 3 against T. b. rhodesiense revealed the following: firstly, all three were characterized ‘inactive’ anti-HAT agents according to the TDA criteria. Secondly, N-methylation in 2 resulted in a decreased anti-HAT activity compared to 2a of an equal molecular weight, demonstrating the importance of the amidic hydrogen for trypanocidal activity. This decrease in the anti-HAT activity of 2 was even greater than the corresponding decrease in its antichagasic activity compared to 2a. Finally, compound 3 was also significantly less potent as anti-HAT agent compared to both 2 and 2a, showing yet again the inferior trypanocidal properties of 2-nitroimidazole-based compounds relative to their 3-nitrotriazole-based counterparts.

The phenotypic screens were then extended to evaluate the susceptibility of T. b. rhodesiense to 3-nitrotriazole-based propanamides (6–12, 14) relative to their acetamide-analogs (6a-11a, 13, 15). For each propanamide/acetamide pairing it was clear that the former compounds were significantly more potent towards this parasite than the latter with the propanamide derivatives deemed ‘active’ (6–11) or ‘moderately active’ (12, 14) anti-HAT agents relative to acetamides that were largely classed as ‘moderately active’ (6a-8a, 11a), or ‘inactive’ (10a, 13, 15). The increased potency of the propanamides towards T. b. rhodesiense consistently translated to improved SI values.

Closer examination of the susceptibility data revealed that for the phenoxyphenyl-containing structures the p-chloro-substituted propanamide 6 was 32-fold more potent towards T. b. rhodesiense than its acetamide analog 6a, the p-fluoro-substituted propanamide 7 was 12-fold more potent than 7a whereas the unsubstituted propanamide 8 was 17.4-fold more potent than 8a. A smaller shift towards lower IC50 values against T. b. rhodesiense was observed in the pair 9 and 9a, since acetamide 9a was already an active anti-HAT agent. Therefore the p-cyanobiphenyl propanamide 9 was only 3.7 fold more potent as an anti-HAT agent than the corresponding acetamide 9a (Table 1).

An enormous shift to anti-HAT activity was observed in propanamide 10 compared to acetamide 10a. Thus, m-biphenyl propanamide 10 was 214-fold more potent against T. b. rhodesiense compared to the corresponding acetamide 10a, which was inactive against this parasite. This increase in the anti-HAT activity of 10 cannot be explained only by its greater lipophilicity compared to 10a, since a similar difference in lipophilicity in the pair 6 and 6a did not have the same effect in anti-HAT activity.

A shift from ‘moderately active’ to ‘active’ anti-HAT agent occurred in the case of the m-benzyloxyphenyl-based compounds with the propanamide 11 being 22-fold more active against T. b. rhodesiense than the corresponding acetamide 11a (Table 1). A similar shift in anti-HAT activity was noted for the phenylisoxazole pairs 12 & 13 and 14 & 15. Therefore, from the inactive p-fluorophenylisoxazole acetamide 13 and p-chlorophenylisoxazole acetamide 15 we obtained the ‘moderately active’ propanamides 12 and 14, respectively. Propanamide 12 was 5.6-fold more active than acetamide 13 and propanamide 14 was 23-fold more active than acetamide 15 (Table 1).

As in the case of antichagasic activity, there was an even better correlation between lipophilicity (clogP values) and anti-HAT activity (logIC50 values against T. b. rhodesiense), with an R2 of 0.85 (Fig. 3). However, lipophilicity alone cannot explain the huge increase observed in the anti-HAT activity of propanamides vs acetamides. Interestingly, all propanamides demonstrated a larger pKa value than acetamides for the amidic ionization, which means that more acidic amides are less potent anti-HAT agents (Table 1). However, it is not clear if these in vitro active propanamides will exert anti-HAT activity in vivo, because they demonstrate relatively high PSA values (115–132), which may not permit penetration of the blood-brain barrier and thus targeting of infected tissue in the central nervous system.

Fig. 3

Correlation between lipophilicity (clogP values) and activity against T. b. rhodesiense (logIC50 values) of all novel compounds in Table 1.

SAR analysis of antileishmanial activity

Linkage elongation between the nitrotriazole ring and the amidic carbonyl by one methylene group resulted in an ∼6- to 34-fold decreased antileishmanial activity in propanamides 6–8 compared to acetamides 6a-8a, designating propanamides 6–8 ‘moderately active’ against this parasite. In addition, this shift to larger IC50 values resulted in unacceptable SI values (<20) for compounds 6 and 8. (Table 1). A shift to decreased antileishmanial activity and an unacceptable SI value was also observed in propanamide 10 compared to acetamide-analog 10a. Propanamide 9 was moderately active against L. donovani and with a good selectivity but it could not be compared to acetamide 9a because the antileishmanial activity of the latter was not determined. In contrast, an increase in antileishmanial activity and selectivity was observed in propanamides 11, 12 and 14 compared to acetamide analogs 11a, 13 and 15. Although the benzyloxyphenyl-propanamide 11 demonstrated moderate activity against L. donovani according to the criteria set, it was ca. 3-fold more potent than the acetamide analog 11a, with both exhibiting acceptable SI values (Table 1).

In the case of the p-fluorophenylisoxazoles 12 and 13, we can observe that the IC50 value of propanamide 12 was slightly lower than that of acetamide 13, although both were deemed ‘inactive’ against L. donovani. In contrast, the p-chlorophenylisoxazole-propanamide 14 was ‘moderately active’ against L. donovani and with an acceptable selectivity, whereas the corresponding acetamide 15 was ‘inactive’ against this parasite (Table 1).

Once again, there was a good correlation between lipophilicity and antileishmanial activity (R2 = 0.78) for the novel compounds (data not shown) but in this case increased antileishmanial activity was associated with increased amidic acidity in pairs 6-6a, 7-7a, 8-8a, and 10–10a while decreased antileishmanial activity was associated with increased acidity in pairs 11–11a, 12–13 and 14–15 (Table 1).

Involvement of type I nitroreductase

To elucidate the mechanism of action of the novel compounds in Table 1, representative 3-nitrotriazole-based propanamides (6–8, 10, 11, 14) and acetamides (13, 15) were evaluated as substrates of purified, recombinant TbNTR and compared to benznidazole (Fig. 4). Enzyme specific activity was measured as oxidized NADH (nmol) per min per mg of protein. Compounds 9 and 12 were also analyzed but these precipitated in the assay buffer and no activity could be determined. All remaining compounds were shown to be substrates of TbNTR, with compounds 6, 11 and 13 yielding specific enzyme activity values up to ∼4-fold greater than benznidazole. In the only propanamide/acetamide pairing tested (14, 15), TbNTR metabolized both compounds at equivalent rates despite the former exhibiting greater anti-HAT activity [24]. As we have reported before with other classes of 3-nitrotriazole-based antitrypanosomal compounds, no correlation between anti-HAT activity and TbNTR specific activity was observed [12], [13], [15], [16], [17], [18].

Fig. 4

Activity of TbNTR toward different 3-nitrotriazole-based compounds. The TbNTR specific activity expressed as nmol NADH oxidized min−1mg−1 of purified his-tagged protein was assessed using representative 3-nitrotriazoles as substrates and the values shown are the means from three experiments ± standard deviation. The activity obtained when using benznidazole (BNZ) as substrate is also shown.

When the above biochemical tests were extended to phenotypic screens, BSF T. b. brucei parasites engineered to express elevated levels of TbNTR were shown to be more susceptible to the compounds under study than controls, thus indicating that within the parasite the nitrotriazoles are functioning as TbNTR activated prodrugs (Table 2). Intriguingly, the ratio of the IC50 values between the recombinant line and wild type did vary from agent to agent. For some compounds (e.g. 6 and 7) the fold difference between the two lines was low (∼2-fold) with these structures generally exhibiting a moderate anti-T. b. brucei activity against wild type parasites (IC50 values of ∼0.65 μM). In contrast, the TbNTR overexpressing line was shown to be 10- to 25-fold more susceptible to other agents (e.g. 10, 11, and 15) that exhibit a lower potency against control lines to start with (IC50 values of 4–8 μM). Together this indicates that for compounds which are less effective against wild type T. b. brucei it is the activation step that appears to limit their trypanocidal potential. When the rate limiting reaction is alleviated, in this case through the over expression of the trypanosomal type I NTR activity, then most of the nitrotriazoles tested exhibit equivalent potencies towards the parasite at which point other downstream factors may now be affecting the anti-T. b. brucei activity.

Table 2

Susceptibility of bloodstream form T.b. brucei with altered levels of NTR towards selected nitrotriazoles.

Compound	IC50 value (μM) T.b. brucei			Ratio –tet/+tet
	Wild type	TbNTR (-tet)	TbNTR (+tet)
nfx	3.98 ± 0.15	6.359 ± 0.119	0.869 ± 0.046	7.3
6	0.64 ± 0.02	0.62 ± 0.03	0.34 ± 0.00	1.8
7	0.57 ± 0.05	0.68 ± 0.07	0.33 ± 0.03	2.1
8	0.14 ± 0.01	1.47 ± 0.03	0.41 ± 0.01	3.6
9	0.12 ± 0.01	0.55 ± 0.04	0.12 ± 0.01	4.7
10	6.66 ± 0.92	5.58 ± 0.62	0.23 ± 0.02	24.6
11	4.25 ± 0.16	3.70 ± 0.58	0.36 ± 0.02	10.4
12	0.76 ± 0.10	1.61 ± 0.10	0.40 ± 0.00	4.1
13	26.61 ± 6.02	–	–
14	0.27 ± 0.05	0.91 ± 0.03	0.29 ± 0.02	4.2
15	8.65 ± 0.49	13.27 ± 1.40	1.23 ± 0.06	10.8

Growth-inhibitory effect as judged by IC50 values (in μM) of selected 3-nitroriazoles on T.b. brucei wild type and recombinant parasites expressing wild type [TbNTR (-tet)] or elevated [TbNTR (+tet)] levels of TbNTR. Data are means from 4 experiments ± standard deviation. Nifurtimox (nfx) was used as control. Only compounds exhibiting an IC50 < 10 μM in wild type parasites were screened against the recombinant line.

In vivo evaluation

Compounds 6, 7 and 9 were chosen for in vivo evaluation, based on their high in vitro potency and selectivity against Trypanosoma cruzi amastigotes (Table 1). A fast luminescence assay was used in an acute infected murine model, in which transgenic T. cruzi parasites expressing firefly luciferase were injected in 5 week-old Balb/c mice. Mice were injected with D-luciferin before imaging. Groups of 5 mice/group were treated i.p. for 10 consecutive days with each compound, at 13 or 15 mg/kg/day. In our previous work with 3-nitrotriazole-based rigid amides we have seen very good in vivo antichagasic activity at daily doses of 15 mg/kg [17], therefore we decided to use a slightly lower dose for compounds 6 and 7 which demonstrated IC50 values against T. cruzi amastigotes at low nM concentrations (Table 1). Bnz was used in parallel as a positive control at 15 mg/kg/day (i.p.). The mean ratio of parasite levels was calculated after 5 and 10 days of treatment. The data are summarized in Fig. 5. All tested compounds, including Bnz, reduced the parasite load to undetectable levels after 10 days of treatment with statistical significance (p < 0.0001). Statistically significant reduction in parasitic level (compared to untreated control) was also observed after 5-day treatment with each compound. However, mice in the group treated with compound 6 did not look very healthy, although no deaths occurred. Compound 6 demonstrates the lowest IC50 value in L6 cells among all analogs tested (Table 1), and perhaps a smaller dose should have been used in mice. No toxicity symptoms were observed with compounds 7 and 9.

Fig. 5

In vivo evaluation of the antichagasic efficacy of compounds 6, 7, 9 and Bnz in the acute murine model. Compounds 6 and 7 were administered (i.p.) at 13 mg/kg/day while compound 9 and Bnz were administered at 15 mg/kg/day (i.p.) up to 10 consecutive days. Parasite ratios were calculated on day 5 and 10. The P values between each treated group and control group for 5-day treatment are shown on the graph. The P value was 0.0001 between each treated group and control group for 10-day treatment. Groups of 5 mice/group were used.

Conclusions

3-Nitrotriazole-based propanamides with a broad spectrum of antitrypanosomal activity were synthesized. The in vitro antichagasic activity of propanamides was excellent and comparable to that of the corresponding acetamides. However, their in vitro anti-HAT activity was 4- to 214-fold greater than that of the corresponding acetamide analogs. With regard to the antileishmanial activity, a decrease was observed for some propanamides whereas a comparable or greater activity was observed for others in comparison with their acetamide-analogs. Antitrypanosomal activity was in good correlation with lipophilicity. Decreased amidic acidity tended to favor anti-HAT activity whereas increased amidic acidity tended to favor antileishmanial activity in most of the compounds. 3-Nitrotriazole-based propanamides are good substrates for type I NTR, but no better than the corresponding acetamides. There was no correlation between enzymatic specific activity and trypanocidal activity, presumably due to the existence of additional targets in the trypanosome, permeability issues with regard to mitochondrion, compound stability and pharmacokinetic factors in general. Such lack of correlation was observed and discussed before with other classes of 3-nitrotriazole-based antitrypanosomal agents [12], [13], [15], [16], [17], [18]. With regard to potential additional targets, it was demonstrated that in addition to type I NTR activation, Bnz toxicity against Trypanosoma cruzi is mediated by various other mechanisms including thiol depletion (by Bnz reduction products) [34], interferon-γ mediated activation of the immune system [35], and NADH-fumarate inhibition [36]. Therefore, further investigation is needed to elucidate if additional enzymes are involved in the mechanism of action of 3-nitrotriazole-based propanamides as well as the role of CYP51.

Representative 3-nitrotriazole-based propanamides demonstrated very good antichagasic activity in an acute T. cruzi murine model, but further investigation is required to see if such compounds can provide cures in the chronic model, taking into account also toxicity issues.

Experimental

General

All starting materials and solvents were purchased from Sigma-Aldrich (Milwaukee, WI), were of research-grade quality and used without further purification. Solvents used were anhydrous and the reactions were carried out under a nitrogen atmosphere and exclusion of moisture. Melting points were determined by using a Mel-Temp II Laboratory Devices apparatus (Holliston, MA) and are uncorrected. Proton NMR spectra were obtained on a Varian Inova-500 or an Agilent Hg-400 spectrometer at 500 or 400 MHz, respectively, and are referenced to Me4Si or to the corresponding solvent, if the solvent was not CDCl3. High-resolution electrospray ionization (HRESIMS) mass spectra were obtained on an Agilent 6210 LC-TOF mass spectrometer at 11,000 resolution. Thin-layer chromatography was carried out on aluminum oxide N/UV254 or polygram silica gel G/UV254 coated plates (0.2 mm, Analtech, Newark, DE). Chromatography was carried out on preparative TLC alumina GF (1000 μm) or silica gel GF (1500 μm) plates (Analtech). All final compounds were purified by preparative TLC chromatography on silica gel plates and also checked by HPLC (Agilent 1100) coupled to mass spectrometry (MS) and ultraviolet (UV) absorbance detectors. Purity was determined by comparing the area under the curve of the test compound peak to the combined areas of all peaks in the chromatogram, excluding DMSO and any peaks appearing in blank runs. Priority reporting was given to the UV traces (since mass spectrometer response can vary widely depending on a given compound's ease of ionization). Purity of compounds lacking UV activity was reported using mass spectrometry detection. All final compounds were 100% pure, according to HPLC-UV analysis.

Synthesis of precursors 1a-c and (4 + 5)a-h

Compounds 1a-c and (4 + 5)a-h were synthesized as follows: In a dichloromethane solution (3 mL) of chloroacetylchloride (1.1 eq) or 3-bromopropanoyl chloride (1.1 eq), a dichloromethane solution (8–10 mL) of the appropriate amine (1 eq, usually 200–300 mg) and triethylamine (1.1 eq) was added dropwise and the reaction mixture was stirred overnight at room temperature under a nitrogen atmosphere. The reaction mixture was evaporated and redissolved in ethyl acetate. The inorganic salts were filtered away and the filtrate was evaporated to give the desire product in 80–85% yield, which however, in most cases contained the corresponding acrylamide (5a-h) as a β-elimination byproduct. In certain cases, chromatography was applied to separate the bromides from the acrylamides for identification purposes, however in most cases the unpurified mixture was used for the next step, since both precursors lead to the desired final product. The spectroscopic data of compounds 1a-c and 4a-h (as well as of some acrylamide byproducts) are provided in the Supplementary Material.

General synthesis of nitro(triazolyl/imidazolyl)acetamides and propanamides 2, 3, 6–15

The potassium salt of 3-nitro-1,2,4-triazole or 2-nitroimidazole (1 eq, usually 100 mg) was formed in CH3CN (6–10 mL), by refluxing with KOH (1.2 eq) for 30 min. To this suspension 1a-c or (4 + 5)a-h (1.1 eq) was added and the reaction mixture was refluxed under a nitrogen atmosphere for 10 h. The reaction mixture was checked by TLC for completion of the reaction and the solvent was evaporated. The residue was dissolved in ethyl acetate and the inorganic salts were filtered away. Upon preparative TLC (silica gel; ethyl acetate-petroleum ether), the desired product was obtained usually as a powder. Purity was checked also by HPLC and it was 100%.

N-methyl-2-(3-nitro-1H-1,2,4-triazol-1-yl)-N-(4-(trifluoromethyl)phenyl)acetamide (2)

White microcrystals (216 mg, 64%): mp 133–134 °C; 1H NMR (400 MHz, CDCl3) δ: 8.35 (s, 1H, triazolic), 7.82 (d, J = 8.0 Hz, 2H, phenylic), 7.47 (d, J = 8.4 Hz, 2H, phenylic), 4.83 (s, 2H, COCH2-), 3.37 (s, 3H, CH3). HRESIMS calcd for C12H11F3N5O3 and C12H10F3N5NaO3 m/z [M+H]+ and [M+Na]+ 330.0809 and 352.0628, found 330.0806 and 352.0619.

N-methyl-2-(2-nitro-1H-imidazol-1-yl)-N-(4-(trifluoromethyl)phenyl)acetamide (3)

Thick oil that solidified to yellowish crystals (180 mg, 62%): mp 45–47 °C; 1H NMR (400 MHz, CDCl3) δ: 7.82 (d, J = 8.0 Hz, 2H, phenylic), 7.55 (d, J = 8.4 Hz, 2H, phenylic), 7.18 (s, 1H, imidazolic), 7.02 (s, 1H, imidazolic), 4.86 (s, 2H, COCH2-), 3.36 (s, 3H, CH3). HRESIMS calcd for C13H12F3N4O3 and C13H11F3N4NaO3 m/z [M+H]+ and [M+Na]+ 329.0856 and 351.0675, found 329.0849 and 351.0669.

N-(4-(4-chlorophenoxy)phenyl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (6)

Off white powder (126 mg, 48%): mp 138–140 °C; 1H NMR (400 MHz, CD3COCD3) δ: 9.37 (br s, 1H, amidic), 8.66 (s, 1H, triazolic), 7.64 (d, J = 9.2 Hz, 2H, phenylic), 7.37 (d, J = 9.2 Hz, 2H, phenylic), 6.98 (dd, J = 9.2, 6.4 Hz, 4H, phenylic), 4.74 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.13 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C17H15ClN5O4 and C17H14ClN5NaO4 m/z [M+H]+ and [M+Na]+ 388.0807, 390.0784 and 410.0627, 412.0604, found 388.0806, 390.0786 and 410.0628, 412.0605.

N-(4-(4-fluorophenoxy)phenyl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (7)

Off white microcrystals (174 mg, 58%): mp 124–125 °C; 1H NMR (400 MHz, CD3COCD3) δ: 9.34 (br s, 1H, amidic), 8.66 (s, 1H, triazolic), 7.61 (d, J = 9.2 Hz, 2H, phenylic), 7.13 (t, J = 9.2 Hz, 2H, phenylic), 7.024–6.99 (m, 2H, phenylic), 6.94 (d, J = 9.2 Hz, 2H, phenylic), 4.74 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.12 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C17H15FN5O4 and C17H14FN5NaO4 m/z [M+H]+ and [M+Na]+ 372.1103 and 394.0922, found 372.1104 and 394.0924.

3-(3-Nitro-1H-1,2,4-triazol-1-yl)-N-(4-phenoxyphenyl)propanamide (8)

Light yellow microcrystals (172 mg, 57%): mp 108–110 °C; 1H NMR (400 MHz, CD3COCD3) δ: 9.35 (br s, 1H, amidic), 8.66 (s, 1H, triazolic), 7.62 (d, J = 8.8 Hz, 2H, phenolic), 7.35 (dt, J = 7.6, 1.2 Hz, 2H, phenolic), 7.09 (t, J = 7.6 Hz, 1H, phenolic), 6.96 (dd, J = 9.6, 2.0 Hz, 4H, phenolic), 4.74 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.13 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C17H16N5O4 and C17H15N5NaO4 m/z [M+H]+ and [M+Na]+ 354.1197 and 376.1016, found 354.120 and 376.1018.

N-(4′-cyano-[1,1′-biphenyl]-4-yl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (9)

Off white microcrystals (120 mg, 48%): mp 200–204 °C (dec.); 1H NMR (400 MHz, CD3COCD3) δ: 9.50 (br s, 1H, amidic), 8.69 (s, 1H, triazolic), 7.87–7.69 (m, 8H, phenylic), 4.76 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.18 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C18H15N6O3 and C18H14N6NaO3 m/z [M+H]+ and [M+Na]+ 363.120 and 385.102, found 363.1205 and 385.1017.

N-([1,1′-biphenyl]-3-yl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (10)

White powder (121 mg, 65%): mp 174–176 °C; 1H NMR (400 MHz, CD3COCD3) δ: 9.42 (br s, 1H, amidic, 8.68 (s, 1H, triazolic), 7.96 (d, J = 2.0 Hz, 1H, phenylic), 7.62–7.34 (m, 8H, phenylic), 4.76 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.17 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C17H16N5O3 and C17H15N5NaO3 m/z [M+H]+ and [M+Na]+ 338.1248 and 360.1067, found 338.1251 and 360.1071.

N-(3-(benzyloxy)phenyl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (11)

White microcrystals (81 mg, 45%): mp 119–120 °C; 1H NMR (400 MHz, CD3COCD3) δ: 9.30 (br s, 1H, amidic), 8.66 (s, 1H, triazolic), 7.48–7.33 (m, 6H, phenylic), 7.19 (t, J = 8.4 Hz, 1H, phenylic), 7.09 (d, J = 8.8 Hz, 1H, phenylic), 6.72 (dd, J = 8.0, 1.6 Hz, 1H, phenylic), 5.08 (s, 2H, benzylic), 4.73 (t, J = 6.8 Hz, 2H, COCH2CH2-), 3.12 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C18H18N5O4 and C18H17N5NaO4 m/z [M+H]+ and [M+Na]+ 368.1353 and 390.1173, found 368.1356 and 390.1178.

N-(5-(4-fluorophenyl)isoxazol-3-yl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (12)

White powder (197 mg, 75%): mp 215 °C (dec); 1H NMR (400 MHz, CD3COCD3) δ: 10.23 (br s, 1H, amidic), 8.69 (s, 1H, triazolic), 7.95 (dd, J = 9.2, 5.6 Hz, 2H, phenylic), 7.32 (t, J = 9.2 Hz, 2H, phenylic), 7.27 (s, 1H, isoxazolic), 4.79 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.29 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C14H12FN6O4 and C14H11FN6NaO4 m/z [M+H]+ and [M+Na]+ 347.0899 and 369.0718 found 347.0906 and 369.0724.

N-(5-(4-fluorophenyl)isoxazol-3-yl)-2-(3-nitro-1H-1,2,4-triazol-1-yl)acetamide (13)

White powder (120 mg, 46%): mp > 220 °C; 1H NMR (400 MHz, CD3COCD3) δ: 10.67 (br s, 1H, amidic), 8.78 (s, 1H, triazolic), 7.97 (dd, J = 9.2, 5.6 Hz, 2H, phenylic), 7.33 (t, J = 8.8 Hz, 2H, phenylic), 7.28 (s, 1H, isoxazolic), 5.59 (s, 2H, COCH2-). HRESIMS calcd for C13H10FN6O4 and C13H9FN6NaO4 m/z [M+H]+ and [M+Na]+ 333.0742 and 355.0562, found 333.0743 and 355.0564.

N-(3-(4-chlorophenyl)isoxazol-5-yl)-3-(3-nitro-1H-1,2,4-triazol-1-yl)propanamide (14)

Off white microcrystals (220 mg, 78%): mp 207–209 °C; 1H NMR (400 MHz, CD3COCD3) δ: 10.85 (br s, 1H, amidic), 8.69 (s, 1H, triazolic), 7.90 (d, J = 8.4 Hz, 2H, phenylic), 7.54 (d, J = 8.4 Hz, 2H, phenylic), 6.72 (s, 1H, isoxazolic), 4.81 (t, J = 6.4 Hz, 2H, COCH2CH2-), 3.32 (t, J = 6.4 Hz, 2H, COCH2CH2-). HRESIMS calcd for C14H12ClN6O4 and C14H11ClN6NaO4 m/z [M+H]+ and [M+Na]+ 363.0603, 365.0579 and 385.0423, 387.0398 found 363.0606, 365.0581 and 385.0422, 387.0392.

N-(3-(4-chlorophenyl)isoxazol-5-yl)-2-(3-nitro-1H-1,2,4-triazol-1-yl)acetamide (15)

Pinkish powder (130 mg, 71%): mp 215–217 °C; 1H NMR (400 MHz, CD3COCD3) δ: 11.26 (br s, 0.5H, amidic), 8.78 (s, 1H, triazolic), 7.91 (d, J = 8.8 Hz, 2H, phenylic), 7.54 (d, J = 8.8 Hz, 2H, phenylic), 6.78 (s, 1H, isoxazolic), 5.62 (s, 2H, COCH2-). HRESIMS calcd for C13H10ClN6O4 and C13H9ClN6NaO4 m/z [M+H]+ and [M+Na]+ 349.0447, 351.0422 and 371.0266, 373.0241 found 349.0448, 351.0422 and 371.0264, 373.0239.

In vitro screening

In vitro activity against T. cruzi, T. b. rhodesiense, L. donovani and cytotoxicity assessment using L6 cells (rat skeletal myoblasts) was determined using a 96-well plate format as previously described [37]. Data were analyzed with the graphic program Softmax Pro (Molecular Devices, Sunnyvale, CA, USA), which calculated IC50 values by linear regression from the sigmoidal dose inhibition curves.

In vitro T. brucei antiproliferating assays and susceptibility studies

The BSF of T. brucei parasites were seeded at 1 × 103 mL−1 in 200 μL of growth medium containing different concentrations of nitroheterocycle. Where appropriate, induction of the TbNTR was carried out by adding tetracycline (1 μg/mL). After incubation for 3 days at 37 °C, 20 μL of Alamar blue was added to each well and the plates incubated for a further 16 h. The cell density of each culture was determined as described before [20] and the IC50 established.

Enzymatic activity studies with type I TbNTR

Recombinant TbNTR was prepared and assayed as previously described [38], [39]. The specific activity of purified his-tagged TbNTR was assessed spectrophotometrically at 340 nm using various nitrotriazole substrates (100 μM) and NADH (100 μM) and expressed as nmol NADH oxidized min−1 mg−1 of enzyme.

In vivo antichagasic activity assessment of selected compounds

For in vivo studies, a Brazilian strain trypomastigotes from transgenic T. cruzi parasites expressing firefly luciferase were used as described before [15]. Briefly, parasites (105 trypomastigotes) were injected in 5 week-old Balb/c mice (Taconic) and three days later mice were anesthesized by inhalation of isofluorane, followed by an injection with 150 mg/kg of D-luciferin potassium-salt in PBS. Mice were imaged 5–10 min after injection of luciferin with an IVIS 100 (Xenogen, Alameda, CA) and the data acquisition and analysis were performed with the software LivingImage (Xenogen) as described before [40]. Treatment with test compounds started 4 days post infection at 13 or 15 mg/kg/day × 10 days, given i.p. The vehicle control was 2% methylcellulose +0.5% Tween 80 and groups of 5 mice/group were used. Mice were imaged after 5 and 10 days of treatment. The ratio of parasite levels was calculated for each animal dividing the luciferase signal after treatment by the luciferase signal on the first imaging (before treatment). Mean values of all animals in each group ± SD were used for plotting.