Optimization of Orally Bioavailable Antileishmanial 2,4,5-Trisubstituted Benzamides

Abstract

Leishmaniasis, a neglected tropical disease caused by Leishmania species parasites, annually affects over 1 million individuals worldwide. Treatment options for leishmaniasis are limited due to high cost, severe adverse effects, poor efficacy, difficulty of use, and emerging drug resistance to all approved therapies. We discovered 2,4,5-trisubstituted benzamides (4) that possess potent antileishmanial activity but poor aqueous solubility. Herein, we disclose our optimization of the physicochemical and metabolic properties of 2,4,5-trisubstituted benzamide that retains potency. Extensive structure–activity and structure–property relationship studies allowed selection of early leads with suitable potency, microsomal stability, and improved solubility for progression. Early lead 79 exhibited an 80% oral bioavailability and potently blocked proliferation of Leishmania in murine models. These benzamide early leads are suitable for development as orally available antileishmanial drugs.

Introduction

Leishmaniasis is a neglected disease endemic in tropical and subtropical regions of Asia, Africa, southern Europe, and the Americas. Leishmaniasis is caused by Leishmania species parasites that are vectored between humans by the bites of infected phlebotomine sandflies.¹ Depending upon the immune response of the host and species of Leishmania parasite, leishmaniasis is classified into one of three forms: (i) cutaneous (most common and affecting skin), (ii) visceral (severe and affecting multiple organ systems), and (iii) mucocutaneous (rare and affecting both skin and mucous membranes).¹ Roughly 1 billion people are at high risk of infection, and there is an annual morbidity of 1 million new cases of cutaneous leishmaniasis and 30,000 new cases of visceral leishmaniasis.² Treatment options for leishmaniasis are very limited due to historically low efforts devoted to developing new drugs.^3,4 The mainstay antileishmanial drugs sodium stibogluconate, meglumine antimonate, and pentamidine were introduced in the middle of the last century and remain widely used despite their toxicity and widespread resistance.⁵⁻⁸ Many newer drugs such as paromomycin and amphotericin B carry severe adverse effects, poor efficacy, difficulty of use, high cost, and emerging drug resistance.^3,9,10 Of the approved antileishmanial drugs, only miltefosine (1, Figure 1) is orally bioavailable; unfortunately, it also has severe toxicity and high cost.^3,4,11 Therefore, novel orally bioavailable antileishmanial drugs with a strong safety profile, low cost, and high efficacy are urgently needed.

Figure 1.

Structures of miltefosine and other potential antileishmanial agents.

One major issue hindering antileishmanial drug discovery is the paucity of well-validated Leishmania targets. We and others have utilized phenotypic screening of small molecules to identify early leads for leishmaniasis.¹²⁻¹⁷ The Drugs for Neglected Diseases initiative (DNDi) used this approach to identify the nitroimidazooxazoles, including DNDI-0690 (2b, Figure 1),¹⁴ nitroimidazooxazines,^15,18 and benzoxaborole DNDI-6148 (2a, Figure 1)¹⁹ as promising preclinical candidates for leishmaniasis. Other nitro-containing drugs also possess potent activity against leishmaniasis.^20,21 The Genomics Institute of the Novartis Research Foundation (GNF) reported oxazole-carboxamide derivatives (3a, GNF6702; and 3b, LXE408) as leads against multiple kinetoplastid infections including leishmaniasis.^22,23 LXE408 is currently in phase II clinical trial (ClinicalTrials.gov identifier NCT05593666). Our group has reported a phenotypic screen of ∼600,000 molecules against Leishmania mexicana promastigotes²⁴ that identified multiple leads, including a chloronitrobenzamide that we had previously explored for sleeping sickness (4, Figure 1). Compound 4 exhibited a reasonable pharmacokinetic profile when administered orally in a murine model of cutaneous leishmaniasis but was less efficacious than the positive control, miltefosine. We hypothesized that this might be due to solubility-limited maximal plasma concentrations and that modifications of scaffold 4 might improve physicochemical properties and provide bioavailable early leads. We also believed that we might find an opportunity to improve potency and reduce present toxicophores while maintaining in vivo activity. Herein, we report the results of this effort.

Strategy

Lipophilic efficiency (LipE) is an important parameter during the optimization of early leads,²⁵ and structural modifications to improve LipE (= pIC₅₀ – cLogD) can significantly improve lead quality by improving pharmaceutics and reducing off-target interactions.²⁶ Therefore, we focused on improving aqueous solubility and reducing metabolism while maintaining potency by careful control of cLogD (Figure 2, Supporting Information Table S4). A key secondary goal was exploring whether the potential toxicophore chloronitrobenzamide was critical for activity (Figure 2). In the first round of optimization, we designed, synthesized, and tested phenylbenzothiazole-containing compounds (7–41) incorporating varying functionalities at C-2, -4, and/or -5 positions of ring A (Figure 2). After the initial SAR analysis of compounds 7 to 41, we explored replacing rings B and/or C with substituted aromatic or heterocyclic moieties in compounds 44 to 97. We purified all compounds to greater than 95% purity and evaluated their ability to inhibit the proliferation of L. mexicana amastigotes cultured within macrophages using our standard assay.^24,27 Initially, we tested all the compounds at fixed concentrations of 10 μM to determine if they affected the proliferation of luciferase-expressing L. mexicana intracellular amastigotes cultured in immortalized J774A.1 macrophages. Compounds with >70% inhibition activity at 10 μM were tested in dose–response experiments to establish the potency of proliferation inhibition (EC₅₀) for L. mexicana intracellular amastigotes. Compounds showing EC₅₀ < 1 μM were considered potent. For comparison, miltefosine, a current front-line antileishmanial drug has EC₅₀ = 2.8 μM. Select potent compounds were tested to establish selectivity relative to mammalian fibroblasts (BJ) and host macrophages (J774A.1) and to establish kinetic aqueous solubility (Supporting Information Table S1–S3). We note that the macrophage proliferation assay is a poor marker for actual in vivo white blood cell toxicity as multiple compounds in this series have been dosed in vivo for weeks without causing white blood cell suppression. This is likely due to the transformed macrophage line being sensitive to the chloronitrobenzamides while nontransformed macrophages in vivo are not. Compounds with good aqueous solubility and selectivity for parasites were tested in mouse microsomal stability studies to estimate the potential for oxidative metabolism and clearance. This work afforded a series of early leads with significantly improved physiochemical properties, significantly reduced oxidative metabolism, and retained potency. We examined in vivo pharmacokinetics and efficacy of one selected analogue (79) in the mouse, validating this series as a starting point for the development of a pivotal treatment against leishmaniasis.

Figure 2.

Design strategy for optimizing 2,4,5-trisubstituted benzamides.

Results

Synthesis

The previously reported route (Scheme 1) afforded benzothiazole-containing compounds 7–30 .^17,28 Condensation of 3-amino-4-methyl benzoic acid (5) with 2-aminothiophenol in phosphoric acid at 160 °C for 14 h provided the key intermediate 2-(4-aminophenyl)benzothiazole (6). Coupling of compound 6 with substituted acids or acid chlorides then afforded compounds 7–30 in adequate to good yields (17–82%). Subsequent treatment of compound 7 with cyclic or acyclic aliphatic amines in the presence of diisopropylethylamine (DIPEA) at room temperature for 14 h produced compounds 31–39 in adequate yields (36–62%). N-Boc deprotection of compounds 31 and 36 using trifluoroacetic acid (TFA) afforded compounds 40 and 41, respectively (Scheme 2).

Scheme 1. Synthesis of Compounds 7–30.

Reagents and reaction conditions: (a) polyphosphoric acid, 12 h, 170 °C, 63% yield; (b) substituted benzoic acid or nicotinic acid, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (1.2 equiv), N,N-diisopropylethylamine (1.2 equiv), DMF, 14 h, room temperature, 17–82% yield; (c) substituted benzoyl chloride or nicotinoyl chloride, N,N-diisopropylethylamine (1.2 equiv), dichloromethane, 14 h, room temperature, 49–76% yield.

Scheme 2. Synthesis of Compounds 31–41.

Reagents and reaction conditions: (a) cyclic or acyclic aliphatic amines (1.2 equiv), N,N-diisopropylethylamine (1.3 equiv), dioxane, 14 h, room temperature, 36–62% yield; (b) TFA, dichloromethane, 1 h, 0 °C, room temperature, 48–59% yield.

Synthesis of acid chloride (43) from acid (42) by treatment with oxalyl chloride and a catalytic amount of dimethylformamide (DMF) followed by reaction with substituted aromatic amines produced compounds 44–77 which possessed modifications of the B ring and/or C ring of compound 7 (Scheme 3). Subsequent treatment of the aryl fluorides with either piperazine or methyl piperazine under basic conditions produced compounds 78–97.

Scheme 3. Synthesis of Compounds 44–97.

Reagents and reaction conditions: (a) oxalyl chloride (1.2 equiv), DMF (0.02 equiv), dichloromethane, 1–4 h, refluxed at 50 °C, 61–85% yield; (b) appropriately substituted aromatic amines, N,N-diisopropylethylamine (1.2 equiv), dichloromethane, 14 h, room temperature, 43–68% yield; (c) piperazine or methyl piperazine (1.1 equiv), N,N-diisopropylethylamine (1.1 equiv), dioxane, 14 h, room temperature, 47–56% yield.

All compounds presented herein were purified by silica column chromatography to a purity of at least 95% as confirmed by ultraperformance liquid chromatography–mass spectrometry (UPLC-MS). The structures of compounds were confirmed by NMR and MS. The detailed characterization of each compound is given in the Supporting Information.

Structure–Activity Relationships of Analogues of 4

Modifications at Ring A

First, we systematically modified ring A of compound 4 to establish the relationships between structures of benzamide derivatives and their growth inhibitory potency (EC₅₀) for intramacrophage L. mexicana amastigotes (Tables 1 and 2, compounds 7-41). Our first goal was to understand whether a halogen at C-2 and/or C-4 of the A ring was a critical pharmacophore element, as was the case for the activity of the series against Trypanosoma spp.(17) Compounds 7–11 all possessed reasonable potency (EC₅₀ = 0.6–4.3 μM) when either the C-2 position (R₁) or C-4 (R₂, methyl piperazine group) was substituted with F or Cl (Table 1). Both halogens gave a similar performance, with the potencies being within 3-fold of one another (compounds 7–11). There was some divergence in the SAR, depending upon the identity of the halogen. If the R₂ was F, then compounds containing a proton at R₁ were active, whereas if the R₂ was Cl, compounds containing a proton at R₁ were inactive (compound 10vs12). In addition, if the R₂ was a proton, then no compounds containing a halogen or hydroxyl at R₁ were active (compounds 13–16). Interestingly, deletion of the nitro group led to significantly lower potency (compounds 27–30) regardless of the presence of halogens at R₁ and R₂, suggesting its essentiality for antileishmanial activity. This pattern suggests that the halogen does not act as a leaving group for a nucleophilic attack but rather fulfills both steric and electrostatic roles in optimal interaction with the target.

Table 1. Structures and In Vitro Antileishmanial Activities of Compounds 7–30 Modified at Ring A.

^aL. mexicana EC₅₀.

^bBJ Cells EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the fibroblast (BJ) cells, respectively. ND indicates that values were not determined.

Table 2. Structures and In Vitro Antileishmanial Activities of Compounds 31–41 Modified at the C-4 Position of Ring A.

^aL. mexicana EC₅₀.

^bBJ Cells EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the mammalian fibroblast (BJ) cells, respectively. ND indicates that values were not determined.

There were also some key differences in the SAR regarding target species. The A ring could be replaced with a substituted pyridine for Trypanosoma species with retention of potency.¹⁶ However, such analogues (compounds 17–26) were generally 3- to 5-fold less potent than their phenyl analogues for Leishmania species. This may point to a different target in each species, although it could also have to do with cellular partitioning differences between the extracellular (trypanosomes) versus intracellular (Leishmania amastigotes) species being studied.

Next, we explored the effects of varying the amino substituent at C-4 of ring A (compounds 31–41, Table 2). Maintaining potency clearly requires the presence of a basic amine (all compounds with EC₅₀ < 1 μM). The presence of a bulky alkyl group imposes a requirement for a terminal secondary amine (41vs35 and 36) to maintain reasonable potency. Because one of the major goals of this study was improving aqueous solubility while maintaining potency, we measured the solubility of potent compounds 31–34, 37, 40, and 41 to determine if the basic amines might improve solubility. Unfortunately, all showed poor aqueous solubility (Supporting Information Table S1), relative to the equivalent potency compound 8, which has fluorine substituents at both the C-2 and C-4 positions and possesses good water solubility (∼22 μM at pH 7.4). Therefore, we centered subsequent work on modifications of the B and C rings while keeping the halogen groups at the C-2 and/or C-4 position of ring A.

Modifications at Rings B and C

Subsequent studies were carried out in two phases. First, we explored modifications of rings B and C while fixing 2-chloro-4-fluoro-5-nitro benzamide (compounds 44 to 66, Table 3). Our initial foci were as follows: (1) was the methyl substituent required on the B ring? (2) Was ring C required? Neither of these elements proved to be an essential part of the pharmacophore. Most of the compounds with simply substituted B rings, and/or lacking the methyl group, exhibited comparable antileishmanial activity (EC₅₀ = ∼1 μM, Table 3, compounds 44–56). Interestingly, compound 57, with an unsubstituted 3-pyridyl B ring, showed reduced activity (EC₅₀ = 4.5 μM), while compound 58, with the 4,5-dichloro substituted 3-pyridyl B ring, exhibited good potency (EC₅₀ = 0.5 μM), suggesting that the chlorines at C-4 and C-5 contribute to improving activity. Compounds 59–66 also exhibited EC₅₀ = ∼1 μM with not much variation in the SAR. Both a phenyl or 3-pyridyl B ring are tolerated when a C ring pyridine (59 and 60), thiazole (61), oxazole (62), benzoxazole (63-65), or benzothiazole (66) is present. The results presented in Table 4 reinforce that ring C is not necessary for an antileishmanial activity for compounds containing fluorine at both C-2 and C-4 positions of ring A (67–69). 3-Pyridine can replace ring B (70 and 74). In addition, both monocycles (71 and 72) and benzoxazoles (73 and 74) can replace ring C. Overall, the di-fluoro nitrobenzamides tolerate a wide range of hydrophobic substituents. However, removal of the halogen at C-2 of ring A (75–77) decreases the potency 5- to 10-fold regardless of the substitution pattern of ring B. Although most compounds with modifications of rings B and C showed good potency (EC₅₀ = 0.2–0.9 μM) and improved LipE (up to ∼4) as compared to compound 4, there was no significant improvement in aqueous solubility for these compounds (44, 50, 64–72, and 74) (Supporting Information Tables S2 and S4). Only compound 71 showed reasonable water solubility (∼9 μM at pH 7.4). Thus, improving LipE by increasing potency and decreasing cLogD did not functionally improve aqueous solubility, the key physiochemical parameter.

Table 3. Structures and In Vitro Antileishmanial Activities of Compounds 44–66 Modified at B and C Rings.

^aL. mexicana EC₅₀.

^bBJ Cells’ EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the mammalian fibroblast (BJ) cells, respectively. ND: not determined.

Table 4. Structures and In Vitro Antileishmanial Activities of Compounds 67–77 Modified at A, B, and C Rings.

^aL. mexicana EC₅₀.

^bBJ Cells EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the mammalian fibroblast (BJ) cells, respectively. ND: not determined.

Based on the initial SARs established for rings A, B, and C, we aimed to further optimize the series by improving aqueous solubility without reducing potency. We synthesized compounds 78–97, replacing the C-4 fluorine with piperazine or methyl piperazine on ring A and simplifying rings B and C. Our strategy was to improve LipE by decreasing cLogD. As expected, most of the compounds exhibited good potency against L. mexicana amastigotes (Table 5) and improved LipE (= 2.3–4.3) as compared to parent compound 4 (Supporting Information Table S4). It is interesting to note that unlike the SAR established for compounds listed in Table 1, compounds 78–97, with a piperazine or methyl piperazine at the C-4 position of ring A, exhibited good potency regardless of the presence or absence of the 2-chloro or 5-nitro group. All the tested compounds showed improved aqueous solubility (3.6–188 μM kinetic solubilities, Supporting Information Table S3) in comparison to parent compound 4. This set of compounds revealed the consistent observation that the ring A chloride or nitro groups can be removed without significantly reducing potency (78, 79, and 87). This opens the possibility of moving forward with a late lead lacking the potential thiol-reactive p-chloronitro substitution pattern.

Table 5. Structures and In Vitro Antileishmanial Activities of Compounds 78–97 Modified at A, B, and C Rings.

^aL. mexicana EC₅₀.

^bBJ Cells’ EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the mammalian fibroblast (BJ) cells, respectively. ND: not determined.

In Vitro Metabolic Stability

To further understand the development potential of the second-generation compounds, we tested the metabolic stability of selected compounds (8, 71, 78, 79, 86–88, 92, and 94–96) in mouse liver microsomal models. Among these compounds, 79, 86, and 94–96 possessed good metabolic stability (t_1/2 > 110 min). Contrariwise, compound 71 lacked metabolic stability (t_1/2 = 0.6 min) and the other compounds possessed moderate stability (t_1/2 = 27–67 min). These studies demonstrated that removal of electron-rich substituents on the A ring, replacement of the electron-rich B ring phenyl with pyridine, and removal of the methyl group from either the A or B rings reduce oxidative metabolism.

Of the tested analogues, compound 79 possessed the best-balanced properties: good potency (0.66 μM), reasonable aqueous solubility (46 μM), and good oxidative metabolic stability [t_1/2 = 115 min; CL_int = 12 mL/min/kg (Table 6)]. Therefore, we carried out in vivo murine pharmacokinetic studies with compound 79 to confirm improvements in oral bioavailability and other pharmacokinetic parameters. This improvement in the potency, aqueous solubility, and in vitro metabolic stability data is probably due to increased LipE from compound 4 (LipE 1.69) to the optimized compound 79 (LipE 3.72) (Supporting Information Table S5).

Table 6. Mouse Microsomal Stability and Aqueous Solubility of Selected Compounds.

^aL. mexicana EC₅₀.

^bBJ Cells’ EC₅₀ values represent the mean ± standard deviation for at least three experiments calculated from dose–response curves against intracellular amastigotes of L. mexicana and the mammalian fibroblast (BJ) cells, respectively.

^ct_1/2.

^dCLint values represent for the half-life and intrinsic clearance determined using mouse liver microsomes.

^eAqueous solubility. ND: not determined.

In Vivo Pharmacokinetics

We performed single-dose intravenous (IV) and oral (PO) pharmacokinetic studies of compound 79 in mice to establish an in vitro to in vivo correlation (IVIC). Following an IV dose (3 mg/kg) to mice, compound 79 reached a peak plasma concentration (C_max) of 1.95 μM, had an elimination half-life (t_1/2) of 3.6 h, an AUC of 2.85 μM·h, a clearance (CL) of 2.57 L/h/kg, and a volume of distribution (V_d) of 9.32 L/kg (Table 7). The plasma concentration remained above its EC₅₀ for inhibition of Leishmania proliferation in macrophages (0.66 μM) for approximately 1.5 h (Figure 3).

Table 7. Summary of Pharmacokinetic Studies on Compound 79 Based on Intravenous (3 mg/kg) and Oral (10 mg/kg) Administration in Mice^a.

route	t1/2 (h)	Cmax (μM)	Tmax (h)	AUC (μM·h)	CL (L/h/kg)	Vd (L/kg)	F (%)
IV	3.4	1.95	NA	2.85	2.57	9.3	NA
PO	3.6	1.0	2	7.82	3.21	6.8	80

^at_1/2 is the compound half-life in plasma; C_max is the maximum concentration; T_max is the time the compound takes to achieve the maximum plasma concentration; AUC is the area under the curve; CL is the clearance; V_d is the volume of distribution at the steady state for IV and apparent volume of distribution for other routes; F is the bioavailability.

Figure 3.

In vivo pharmacokinetic profiling of compound 79 in mice administered intravenously at 3 mg/kg and orally at 10 mg/kg.

Following a single oral dose to mice (10 mg/kg), using a highly solubilizing formulation [20% PEG 400/80% (20%SBE-beta-CD in water)], compound 79 exhibited a C_max of 1.0 μM, a t_1/2 of 3.6 h, a T_max of 2 h, and an AUC of 7.82 μM·h (Table 7), giving an apparent high oral bioavailability (F = 80%). Thus, a single PO dose of 79 to mice at 10 mg/kg affords plasma concentration above an efficacious dose (EC₅₀) for about 5.5 h. These results strongly suggest that compound 79 is a promising late lead for an orally bioavailable antileishmanial. Finally, we evaluated the in vivo efficacy of compound 79. We treated compound 79 with hydrochloric acid to afford more soluble salt (Supporting Information) for in vivo efficacy.

In Vivo Efficacy

To assess the efficacy of compound 79, we employed a murine model of cutaneous leishmaniasis, BALB/c mice were infected by footpad injection, and vehicle or compound was administered by oral gavage for 10 consecutive days beginning at day 18 postinfection, and lesion size was monitored up to 5 weeks postinfection (Figure 4). Compound 79 strongly delayed lesion development, with an ∼10-fold reduction in lesion size at 5 weeks postinfection. Control of the disease was not complete as smaller lesions began to develop following treatment with compound 79.

Figure 4.

In vivo efficacy study of compound 79 in mice. BALB/c mice (BALB/c strain, 5 per group) were infected with L. mexicana and treated with either the vehicle or compound 79 (PO, 50mg/kg/day) for 10 consecutive days beginning on day 18 postinfection. Lesion sizes were measured by caliper and are plotted as mean ± SD. Compound-treated animals had significantly smaller lesions at weeks 4 and 5 (asterisks, p < 0.0001, Mann–Whitney). Vehicle control animals were sacrificed in the following week 5 due to the large lesion size.

Discussion and Conclusions

We and other groups previously reported the synthesis and SARs of substituted carboxamides as antiparasitic leads for Trypanosoma brucei.^17,29−32 Our extensive SAR studies showed that potent antitrypanosomal activity by our chemotype required an intact ring A chloronitrobenzamide, or the analogous 4-chloropyridine, and a ring B benzothiazole or benzoxazole group at the C-5 position.¹⁶ In contrast, the similar chemotype reported by Gelb and co-workers allowed replacement of the 4-chloropyridine or 2-chloro-5-nitrophenyl ring with furan or pyrrolidine.²⁹ Furthermore, another study reported antitrypanosomal activities of pyrrolidine analogues containing benzothiazole, benzimidazole, and benzoxazole fused rings at region B and a phenyl ring at region C.³² Therefore, despite the apparent need for an electrophilic A ring shown in our work, there was not a consensus pharmacophore covering all the superficially similar series—a finding that may point to multiple targets being addressed by structurally similar inhibitors.

In the present study, we expand upon the observation that our series also possessed activity against Leishmania parasites by exploring SAR-defining activity against L. mexicana amastigotes. In contrast to the observations for T. brucei, 2-chloro-5-nitrophenyl is not essential for antileishmanial activity. However, while the canonical electrophile is not required, antileishmanial activity is not maintained when the A ring is replaced with a 4-chloropyridine. This represents a critical difference in the driving SAR for Leishmania relative trypanosomes. Furthermore, our SAR study (Figure 5) revealed that cyclic or acyclic diamines at C-4 of the nitrophenyl ring contribute to potency, that ring C is not essential, and that the B ring substitution pattern can be simplified.

Figure 5.

SAR summary of antileishmanial 2,4,5-trisubstituted benzamide derivatives.

We also examined the metabolic stability of a subset of compounds selected for their potent growth inhibition (EC₅₀) of L. mexicana amastigotes, selectivity relative to mammalian fibroblast cells and host macrophages, and aqueous solubility. This revealed that removing electron-rich substituents on the A ring and a methyl group on the B ring afforded compounds stable to incubation with microsomal mixtures. Pharmacokinetic experiments with compound 79, chosen for the combination of good potency, good solubility, and low microsomal clearance, revealed good oral bioavailability (F = 80%) and sustained plasma concentrations above an efficacious dose (EC₅₀). The improved LipE value of compound 79 to that of compound 4 indicated that adequate control of lipophilicity probably contributed to the substantial improvement in solubility and pharmacokinetic properties. Daily dosing of compound 79 strongly decreased formation of footpad lesions in the murine L. mexicana model.

In conclusion, we expanded the SAR and SPR for substituted benzamide analogues to enhance their antileishmanial features. This delivered promising antileishmanial agents with good potency, aqueous solubility, and in vitro microsomal stability data. Compound 79 provided a proof of concept of the viability of the series with good in vivo pharmacokinetic properties and promising potential for the development of new orally bioavailable antileishmanial drugs.

Experimental Section

Chemistry

¹H and ¹³C NMR spectra were recorded at room temperature in CDCl₃ or DMSO-d₆ on a 400 or 500 MHz Agilent or 600 MHz Bruker spectrometer. Chemical shifts (δ) were recorded in parts per million (ppm) calibrating with internal TMS (δ 0.0 ppm for ¹H and ¹³C) or internal CHCl₃ (δ 7.26 ppm for ¹H and 39.5 ppm for ¹³C), or internal DMSO (δ 2.50 ppm for ¹H and 39.5 ppm for ¹³C) as the reference. ¹H NMR data are reported as follows: chemical shift, multiplicity [s = singlet, bs = broad singlet, d = doublet, t = triplet, q = quartet, p = pentet, sext = sextet, sep = septet, m = multiplet, dd = double of doublets, dt = doublet of triplets, td = triplet of doublets, qd = quartet of doublets, and coupling constants (J) in hertz (Hz), and integration]. Flash column chromatography was performed using Biotage Isolera One and Biotage KP-SIL SNAP cartridges. All the tested compounds were characterized using 1H NMR and LCMS. Compounds that proved critical to our chemical analysis were further characterized using ¹³C NMR. All compounds were confirmed to be ≥ 95% pure prior to testing. Purity was assessed using UPLC-MS (Waters, MA) equipped with a PDA detector and a single quadruple detector. A BEH-C18 column (1.7 μm, 2.1 × 50 mm²) was used. The flow rate was 0.7 mL/min, and the gradient started with 90% A (0.1% formic acid in H₂O), changed to 95% B (0.1% formic acid in acetonitrile), and then returned to 90% A. The mass spectrometer was operated in the positive-ion mode with electrospray ionization. Integration was performed using Masslynx software 4.2. Starting materials and reagents were obtained from Sigma-Aldrich, TCI, or Alfa Aesar and used without further purification. Thin-layer chromatography and column chromatography were performed using Kieselgel 60 F₂₅₄ (Merck) and silica gel (Kieselgel 60, 230–400 mesh, Merck), respectively.

Pharmacokinetics

All animal experiments were carried out under a protocol approved by the UK IACUC.

General Procedure for the Synthesis of the Phenyl Benzothiazole Intermediate (6)

To a solution of 3-amino-4-methylbenzoic acid 5 (1 g, 6.62 mmol) in polyphosphoric acid (7 mL) was added 2-aminothiophenol (0.83 g, 6.62 mmol), and the mixture was stirred for 12 h at 170 °C, cooled to room temperature, and pH adjusted with 2 N aqueous sodium hydroxide to pH 10. Addition of dichloromethane (200 mL) in the solution resulted in formation of a precipitate, which was filtered, and the resulting filtrate was concentrated under reduced pressure to yield pure intermediate 6 (63%) as an off-white solid.

General Method for the Synthesis of Compounds 7–30

Compounds 7–30 were synthesized using method A or B with an overall yield of 29–82% as described below.

Method A

To a solution of substituted benzoic acid or nicotinic acid (0.22 mmol) in DMF (1 mL) were added intermediate 6 (0.23 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (0.28 mmol), and diisopropylethylamine (0.28 mmol) under a nitrogen atmosphere, and the mixture was stirred at room temperature overnight. Then, the mixture was diluted with water, extracted with ethyl acetate, dried over magnesium sulfate, filtered, concentrated under reduced pressure, and purified by flash column chromatography (ethyl acetate/hexanes, 5–40%, v/v) to yield the desired product.

Method B

To a solution of each substituted benzoyl chloride or nicotinoyl chloride (0.17 mmol) in dichloromethane (2 mL) were added intermediate 6 (0.18 mmol) and diisopropylethylamine (0.2 mmol) under a nitrogen atmosphere, and the resulting mixture was stirred at room temperature overnight. Then, the mixture was diluted with water, extracted with ethyl acetate, dried over magnesium sulfate, filtered, concentrated under reduced pressure, and purified by flash column chromatography (ethyl acetate/hexanes, 5–40%, v/v) to yield the desired product.

General Method for the Synthesis of Compounds 31–39

To a solution of a cyclic or acyclic aliphatic amine (0.13 mmol) and diisopropylethylamine was added a solution of compound 7 (0.1 mmol) in dioxane (3 mL) under a nitrogen atmosphere. The mixture was stirred at room temperature for 12 h and then diluted with water, extracted with ethyl acetate, dried over magnesium sulfate, filtered, concentrated under reduced pressure, and purified by flash column chromatography (methanol/dichloromethane, 2–12%, v/v) to afford compounds 31–39 in a 36–62% yield.

General Method for the Synthesis of Compounds 40 and 41

TFA (1 mL) was added to a solution of compounds 31 and 36 (0.1 mmol each) in dichloromethane (3 mL) at 0 °C under a nitrogen atmosphere. The mixture was stirred at room temperature overnight and then neutralized with a saturated aqueous solution of sodium bicarbonate, filtered, washed with water and dichloromethane, concentrated under reduced pressure, and purified by flash column chromatography (methanol/dichloromethane, 2–12%, v/v) to afford compounds 40 and 41 in 45 and 59% yields, respectively.

General Method for the Synthesis of Benzoyl Chloride Intermediates (43)

To a mixture of a substituted benzoic acid (42, 4.6 mmol) and 2-dimethylformamide (0.09 mmol) in dichloromethane (10 mL) was added oxalyl chloride (6.9 mmol) dropwise. The resulting mixture was heated at 50 °C for 1–4 h. Then, the reaction mixture was cooled to room temperature and concentrated under reduced pressure to afford corresponding substituted benzoyl chloride intermediates 43 in a 61–85% yield which were subsequently used for the next step without purification.

General Method for the Synthesis of Compounds 44–77

To a solution of a substituted aromatic amine (0.15 mmol) in dichloromethane (2 mL) was added a benzoyl chloride (43, 0.18 mmol), followed by diisopropylethylamine (0.18 mmol) under a nitrogen atmosphere. The mixture was stirred at room temperature for 10–14 h, diluted with water, extracted with ethyl acetate, dried over magnesium sulfate, filtered, concentrated under reduced pressure, and purified by flash column chromatography (ethyl acetate/hexanes, 10–50%, v/v) to afford compounds 44-77 in a 43–68% yield.

General Method for the Synthesis of Compounds 78–97

To a solution of substituted compounds 44–77 (0.13 mmol) in dioxane (3 mL) were added piperazine or methyl piperazine (0.14 mmol) and diisopropylethylamine (0.14 mmol) under a nitrogen atmosphere. The mixture was stirred at room temperature for 1 h, then diluted with water, extracted with ethyl acetate, dried over magnesium sulfate, filtered, concentrated under reduced pressure, and purified by flash column chromatography (methanol/dichloromethane, 5–15%, v/v) to afford compounds 78–97 in a 47–56% yield.

Glossary

Abbreviations

CL

cutaneous leishmaniasis

DIPEA

N,N-diisopropylethylamine

DMF

N,N-dimethylformamide

DMSO

dimethyl sulfoxide

EC50

half-maximal proliferation inhibitory concentration

EDC

1-ethyl-3-(3-dimethylaminopropyl)carbodiimide

LipE

lipophilic efficiency

L

Leishmania

SAR

structure–activity relationship

SI

selectivity index

SPR

structure–property relationships

UPLC

ultraperformance liquid chromatography

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jmedchem.3c00056.

• Molecular formula strings and growth inhibition data for L. mexicana intracellular amastigotes, J774A.1 macrophages, and normal fibroblasts (BJ) (CSV)

• NMR (¹H and ¹³C) characterization, mass spectra, and experimental details regarding kinetic solubility, microsomal stability, and in vivo PK and efficacy studies (PDF)

Author Present Address

^⊥ Center for Scientific Review, National Institutes of Health, 6701 Rockledge Drive, Bethesda, Maryland 20892, United States

Author Contributions

^# H.S.K., D.O., and T.M.K. contributed equally to this work. H.S.K.—designed compound series, made and purified compounds. D.O.—tested antileishmanial activity. T.M.K.—synthesized large-scale batches, drafted the initial manuscript, and contributed to the formulation. C.M.F.—provided technical support for cell biology and performed animal experiments. J.T.H.—contributed to the design and project management. Y.C.—pharmacokinetics experiments. A.L.R.—mammalian cell line activity. K.L.B.—synthesized large-scale batches, G.S.—replicated antileishmanial activity testing. W.P.—microsomal stability experiments. P.A.Y.—contributed to data analysis. M.S.—provided technical support for cell biology. S.M.L.—funded the project, oversaw the project, and wrote the manuscript. R.K.G.—funded the project, oversaw the project, and writing of the manuscript. All authors have given approval to the final version of the manuscript.

NIH/NIAID 5R33AI127591.

The authors declare no competing financial interest.