Identification of N-Benzyl 3,5-Dinitrobenzamides Derived from PBTZ169 as Antitubercular Agents

Abstract

A series of benzamide scaffolds were designed and synthesized by the thiazinone ring opening of PBTZ169, and N-benzyl 3,5-dinitrobenzamides were finally identified as anti-TB agents in this work. 3,5-Dinitrobenzamides D5, 6, 7, and 12 exhibit excellent in vitro activity against the drug susceptive Mycobacterium tuberculosis H37Rv strain (MIC: 0.0625 μg/mL) and two clinically isolated multidrug-resistant strains (MIC < 0.016–0.125 μg/mL). Compound D6 displays acceptable safety and better pharmacokinetic profiles than PBTZ169, suggesting its promising potential to be a lead compound for future antitubercular drug discovery.

Tuberculosis (TB) is an airborne infectious disease mainly caused by Mycobacterium tuberculosis (MTB).¹ The World Health Organization (WHO) reported that MTB still caused an estimated approximately 10.4 million infections and 1.3 million deaths in 2016.² Recently, the increase of multidrug-resistant (MDR) TB and even the emergence of extensively drug-resistant (XDR) TB, together with coinfection with Human Immunodeficiency Virus (HIV), have challenged us seriously.³⁻⁵ Two drugs with new mechanisms of action, Bedaquiline⁶ and Delamanid (Figure 1),⁷ were approved (in 2012 and 2014, respectively) for the treatment of MDR-TB, but some adverse events have been noted.³ Therefore, it is urgently needed to develop novel anti-TB agents active against MDR- and XDR-TB.⁸

Figure 1.

Structures of selected dinitrobenzene derivatives, Q203, and Delamanid.

Recently, 8-nitro-6-(trifluoromethyl)-1,3-benzothiazin-4-ones (BTZs), a novel class of TB agents targeting DprE1,⁹⁻¹¹ have been extensively studied, and candidate PBTZ169 (Figure 2) is in Phase II clinical trials at present.² Meanwhile, some nitroaromatic compounds that contain an electron-withdrawing substituent (NO₂, CF₃) in the position meta to the nitro group were also identified as DprE1 inhibitors, such as dinitrobenzamide DNB1¹² and its analogue CT319,¹³ dinitrobenzene derivatives (here abbreviated as DNDs),¹⁴ and xanthone derivative 1-methyl-2,4,7-trinitroxanthone (MTX) (Figure 1).^13,15 Moreover, 3,5-dinitrophenylmethanethiol derivatives DNPT1–4 (Figure 1)¹⁶⁻¹⁸ targeting the synthesis of mycobacterial nucleic acids were found to have considerable antimycobacterial activity.

Figure 2.

Design of the new molecules.

In our previous studies, many BTZs containing various cyclic ketoxime, spiro-heterocycle, and piperidine moieties were identified as potential antitubercular agents.¹⁹⁻²¹ Moreover, replacement of the N-benzyl group on the amide linker of Q203 (Figure 1) with a N-(2-phenylamino)ethyl or N-(2-phenoxy)ethyl moiety maintains potent activity.^22,23 Given that above compounds belong to aromatic carboxamides, we intended to simplify the structure of PBTZ169 to benzamide scaffolds by the thiazinone ring opening in this work. Thus, a series of 3-nitro-5-(trifluoromethyl)benzamides bearing N-oxyethyl (A1–7), N-aminoethyl (B1–4), and N-benzyl (C1), as well as N-benzyl-3,5-dinitro-benzamide (D1), were first designed to identify the optimal benzamide cores (Figure 2), and then the effect of substituents on the amide linker was further investigated. Our primary objective was to find optimized benzamides with potent antimycobacterial activity. A preliminary structure–activity relationship (SAR) study was also explored.

Detailed synthetic pathways to side chains 3, 5, and 12 and target compounds A–D are outlined in Schemes 1 and 2, respectively. Coupling of different phenols (RH) with N-(2-hydroxyethyl)phthalimide 1 in the presence of diethyl azodicarboxylate (DEAD) and PPh₃ (2a–g) followed by treatment with hydrazine hydrate in ethanol at 50 °C yielded desired amines 3a–g. Treatment of anilines (R₁H) with 2-bromoethanamine hydrobromide 4 in toluene under reflux condition gave the corresponding 1,2-diamines 5a,b. Nucleophilic substitution of 1-(cyclohexylmethyl)piperazine 6 and 1-phenylpiperazine 7 with 2-(2-bromoethyl)isoindoline-1,3-dione in acetonitrile in the presence of potassium carbonate gave intermediate compounds 8 and 9, which were treated with hydrazine hydrate in ethanol yielded side chains 5c and 5d, respectively. 4-Fluorobenzonitrile 10 was treated with various nitrogen heterocyclic amines (YH) in DMSO in the presence of K₂CO₃, and the resulting intermediates 11a–l were subsequently reduced with LiAlH₄ in THF to produce the desired benzylamines 12a–l (Scheme 1).

Scheme 1. Synthesis of Side Chain Compounds 3, 5, and 12.

Reagents and conditions: (i) DEAD, PPh₃, THF, 0–5 °C; (ii) hydrazine hydrate, EtOH, 50 °C; (iii) toluene, reflux, then NaOH, rt. (iv) K₂CO₃, MeCN, reflux; (v) K₂CO₃, DMSO; (vi) LiAlH₄, THF.

Scheme 2. Synthesis of Target Compounds A–D (See Tables 1 and 2).

Reagents and conditions: (i) BOP-Cl, TEA, CH₂Cl₂

Target compounds A–D were easily obtained by amidation of acids 13a,b with the above side chain compounds 3a–g, 5a–d, and 12a–l in the presence of triethylamine (TEA) and condensation agent bis(2-oxo-3-oxazolidinyl)phosphonic chloride (BOP-Cl) (Scheme 2).

The target compounds A1–7, B1–4, and C1 bearing different kinds of substituents on the amide linker to ensure side chain flexibility and structure diversity and N-benzyl 3,5-dinitro-benzamide (D1) were first synthesized. They were preliminarily examined for in vitro activity against drug sensitive MTB strain (H37Rv ATCC27294), using the Microplate Alamar Blue Assay (MABA).²⁴ The minimum inhibitory concentration (MIC) is defined as the lowest concentration effecting a reduction in fluorescence of >90% relative to the mean of replicate bacterium-only controls. The MIC values of the compounds along with PBTZ169, isoniazid (INH), and rifampicin (RFP) as references were obtained in Table 1.

Table 1. Structures and Activity of Compounds A–D against MTB H37Rv.

^aMIC determined in μg/mL; Cpds., compounds; INH, isoniazid; RFP, rifampicin.

Synthesized compound A1 exhibits a MIC value of 1 μg/mL against the H37Rv strain of MTB, much less than that (75 ng/mL)¹² of DNB1 (a bioisostere of A1), which reveals that replacement of the nitro group of DNB1 with trifluoromethyl is detrimental. Except for N-(2-pyridin-4-yloxy)ethyl compound A7 (MIC: 8 μg/mL), activity of the other N-phenoxyethyl analogues (A2–6) is, as expected, comparable to A1. However, compound B1, wherein a 4-(cyclohexylmethyl)piperazine ring (the same side chain of PBTZ169) is directly attached to the amide through an ethylene, displays obviously decreased activity (MIC: 4 μg/mL) compared to PBTZ169 (MIC: 0.0625 μg/mL), but introduction of a benzene ring on the piperazine instead of the alkyl group is beneficial to the activity (B1 vs B2). Furthermore, N-phenoxyethyl group is preferred over N-phenylaminoethyl (A3 vs B3).

Interestingly enough, compound C1, having the same side chain as Q203, shows better activity (MIC: 0.5 μg/mL) than A1–7 and B1–4. Replacement of the trifluoromethyl group of C1 with another electron-withdrawing one (NO₂) in compound D1 leads to a MIC of 0.0625 μg/mL (Table 1). These results indicate that the 3,5-dinitrobenzamide core and N-benzyl group are preferred over the corresponding 3-nitro-5-(trifluoromethyl)benzamide and N-phenoxyethyl or N-aminoethyl, respectively.

Encouraged by the above SAR, N-benzyl 3-nitro-5-(trifluoromethyl)/3,5-dinitro benzamides with various groups at the para-position of the benzene ring were further designed and synthesized. As shown in Table 2, 12 compounds show considerable activity against MTB H37Rv strain (MIC: < 1 μg/mL). Among them, six compounds C6 and D1, 5–7, and 12 (MIC: 0.0625 μg/mL) are more active than INH/RFP (MIC: 0.0781 μg/mL) and comparable to PBTZ169. Overall, the data reveal that with a few exceptions (C3, C4, C6, and C11), 3-nitro-5-(trifluoromethyl) benzamides are less active than the corresponding 3,5-dinitrobenzamides. Based on this, we here only discuss the effect of the substituent (Y) on the benzene ring of 3,5-dinitrobenzamide series. The nature of the substituents greatly influences activity. Compound D2 having the same side chain as Delamanid, namely, insertion of an oxygen atom between the piperidine and 4-trifluoromethoxybenzene rings of D1, displays less activity (MIC: 0.5 μg/mL) than D1. Removal of 4-trifluoromethoxybenzene ring of D1 was found to be detrimental (D1 vs D3). Replacement of the piperidine ring of D3 (MIC: 2 μg/mL) with morpholine in compound D4 leads to a complete loss of activity, but thiomomorpholine in D5 or introduction of an electron-withdrawing group (CF₃) in D6 leads to obviously increased activity (MIC: 0.0625 μg/mL). Interestingly, octahydro-1H-isoindole analogue D7 also demonstrates the most potent MIC value of 0.0625 μg/mL. However, among the different para-groups on the piperazine ring, compounds with alkyl groups, D8 (methyl) and D9 (isopropyl), or heteroaromatic groups, D10 (4-pyridyl) and D11 (2-pyridyl), display comparatively less activity (MIC: 1–4 μg/mL), and the only exception is compound D12 with an aromatic group (4-fluorophenyl) (MIC: 0.0625 μg/mL).

Table 2. Structures and Activity of N-Benzylated Analogues C1–12 and D1–12^a.

^aCpds., compounds; INH, isoniazid; RFP, rifampicin; MDR-MTB1, MDR-MTB12525; MDR-MTB2, MDR-MTB14231.

Considering their strong potency against the drug sensitive MTB H37Rv strain (MIC: 0.0625 μg/mL), compounds C6 and D1, 5–7, and 12 were further evaluated against two clinical isolated MTB-MDR (12525 and14231) strains resistant to both INH and RFP. Except for C6 and D1, the other four compounds exhibit better activity (MIC < 0.016–0.125 μg/mL) than PBTZ169 (Table 2), suggesting their promising potential for both drug-sensitive and resistant MTB strains (Tables 1, 2).

Based on the measured activity levels against all of the tested strains, compounds D1, 5–7, and 12 were tested for in vivo tolerability by recording the number of survivors after a single oral dose in mice of 500 mg/kg, followed by a 7-day observation. All of them display the low oral acute lethal toxicity (Table 3). The in vivo PK profiles of these compounds were further evaluated in mice after a single oral administration of 50 mg/kg. As shown in Table 3, compared to PBTZ169, compound D1 has a relatively longer T_1/2 of 5.52 h, but less T_max, C_max, and AUC_0-∞. Absorption of compounds D5, 7, and 12 in plasma is very poor, or not detectable. Compound D6, with a 4-trifluoro-methylpiperidine moiety, displays better PK properties, with T_1/2 of 3.63 h, T_max of 0.92 h, C_max of 1719 ng/mL, and AUC_0-∞ of 5605 h·ng/mL.

Table 3. Acute Toxicity and PK Profiles of Selected Compounds.

		PKb
Cpds.	acute toxicitya	T1/2 (h)	Tmax (h)	Cmax (ng/mL)	AUC0-∞ (h·ng/mL)
D1	5/5	5.52	0.83	366	3262
D5	5/5	3.30	2.75	19.0	41.1
D6	5/5	3.63	0.92	1719	5605
D7	5/5	NA	NA	NA	NA
D12	5/5	13.6	1.08	15.0	219
PBTZ169	NT	2.87	0.83	1300	5478

^aNumber of animals that survived/total.

^bDosed orally in mice at 50 mg/kg (n = 3); NT, not tested; NA, unavailable; Cpds., compounds.

In conclusion, a series of various benzamide scaffolds were designed and synthesized by the thiazinone ring opening of PBTZ169, and N-benzyl 3,5-dinitrobenzamides were finally identified as anti-TB agents in this work. Four N-benzyl 3,5-dinitrobenzamides D5, 6, 7, and 12 exhibit excellent in vitro inhibitory activity against both drug-sensitive MTB strain H37Rv (MIC: 0.0625 μg/mL) and drug-resistant clinical isolates (MIC < 0.016–0.125 μg/mL). Moreover, compound D6 displays acceptable safety and better PK properties than PBTZ169, and it may serve as a promising lead compound for further antitubercular drug discovery. Studies to determine the in vivo efficacy of D6 are currently underway.

Glossary

ABBREVIATIONS

MTB

Mycobacterium tuberculosis

MDR-TB

multidrug-resistant tuberculosis

XDR-TB

extensively drug-resistant tuberculosis

WHO

World Health Organization

HIV

human immunodeficiency virus

BTZs

8-nitro-6-(trifluoromethyl)-1,3-benzothiazin-4-ones

DprE1

decaprenyl phosphoryl-β-d-ribose 2-epimerase

DEAD

diethyl azodicarboxylate

DMSO

dimethyl sulfoxide

MIC

minimum inhibitory concentration

MABA

Microplate Alamar Blue Assay

C_max

maximum concentration

T_max

time to maximum concentration

AUC_0-∞

area under curve from time zero to infinity

t_1/2

plasma elimination half-life

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsmedchemlett.8b00177.

• Experimental procedures and analytical data of the synthesized compounds (PDF)

Author Contributions

^‡ These authors contributed equally. M.L.L. and Y.L. conceived and designed the project. L.H., K.L., Z.T., A.W., H.W., and Y.G. synthesized the compounds. Y.Y. and J.S. analyzed checked the data. B.W. and Y.L. evaluated the antimycobacterial activity. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

This work was financially supported by National Mega-project for Innovative Drugs (2018ZX09721001-004-007, 2018ZX09711001-007-002, 2015ZX09102007-008, 2015ZX09102007-015, 2015ZX09304006-016), CAMS Initiative for Innovative Medicine (CAMS-2016-I2M-1-010, CAMS-2017-I2M-1-011), PUMC Youth Fund (2017350011).

The authors declare no competing financial interest.