Synthesis and Biological Evaluation of Novel 2-Methoxypyridylamino-Substituted Riminophenazine Derivatives as Antituberculosis Agents

Clofazimine, a member of the riminophenazine class, is one of the few antibiotics that are still active against multidrug-resistant Mycobacterium tuberculosis (M. tuberculosis). However, the clinical utility of this agent is limited by its undesirable physicochemical properties and skin pigmentation potential. With the goal of maintaining potent antituberculosis activity while improving physicochemical properties and lowering skin pigmentation potential, a series of novel riminophenazine derivatives containing a 2-methoxypyridylamino substituent at the C-2 position of the phenazine nucleus were designed and synthesized. These compounds were evaluated for antituberculosis activity against M. tuberculosis H37Rv and screened for cytotoxicity. Riminophenazines bearing a 3-halogen- or 3,4-dihalogen-substituted phenyl group at the N-5 position exhibited potent antituberculosis activity, with MICs ranging from 0.25~0.01 μg/mL. The 3,4-dihalogen- substituted compounds displayed low cytotoxicity, with IC50 values greater than 64 μg/mL. Among these riminophenazines, compound 15 exhibited equivalent in vivo efficacy against M. tuberculosis infection and reduced skin discoloration potential in an experimental mouse infection model as compared to clofazimine. Compound 15, as compared to clofazimine, also demonstrated improved physicochemical properties and pharmacokinetic profiles with a short half-life and less drug tissue accumulation. This compound is being evaluated as a potential drug candidate for the treatment of multidrug resistant tuberculosis.


Introduction
Tuberculosis (TB) has become a global public health emergency due to the rapid development of multidrug-resistant tuberculosis (MDR-TB). According to the Global Tuberculosis Report released by World Health Organization in 2013, an estimated 8.6 million people developed TB and 1.3 million people died from TB in 2012, including 320,000 deaths among HIV-positive people [1]. Some progress has been made to develop new TB drugs. Currently, the global TB drug pipeline consists of about 10 new or repurposed drugs in Phase II or Phase III clinical trials. Bedaquiline, a new TB drug with a novel mechanism of action has become the first drug approved by the US Food and Drug Administration (FDA) for the treatment MDR-TB in more than 40 years [1][2][3]. Despite the progress, there is still an urgent need for new and more effective therapy for the treatment MDR-TB and repurposing or optimizing existing antibiotics has been proved an effective approach in identifying new TB drugs [2,4].
Clofazimine (CFZ, Figure 1) was first introduced in 1960s for the treatment of leprosy [5]. This agent has demonstrated excellent activity against MDR-TB both in vitro and in vivo, and has been used in clinical trials for the treatment of MDR-TB [6]. We have conducted a systematic structural modification on CFZ to address the skin discoloration problem, which is one of the major obstacles for its clinical use [7][8][9]. We have identified a novel series of riminophenazine derivatives containing a 2-methoxypyridylamino substituent at the C-2 position of the phenazine nucleus. This series of compounds, exemplified by TBI-1004 (Figure 1), displayed potent antituberculosis activity, reduced lipophilicity, and improved pharmacokinetic profiles as compared to CFZ [8]. We also concluded that an electron-withdrawing group at the para-position of the phenyl ring at the N-5 position was beneficial to antituberculosis activity [9]. Previously, O'Sullivan and coworkers found that compounds bearing a 3,4-dichlorophenyl or 3,4,5-trichlorophenyl group at the N-5 position and a tetramethylpiperidyl group at the C-3 position, such as compound B4100 (Figure 1), possessed improved antituberculosis activity as compared to CFZ [10,11]. These work prompted us to further investigate the effect of the halogen atoms on the phenyl ring at the N-5 position for antituberculosis activity based on our 2-methoxypyridylamino-substituted riminophenazine system. The present work aimed to identify novel riminophenazine derivatives with potent antituberculosis activity, improved physicochemical property and pharmacokinetic profiles and low skin pigmentation potential. Hence, we kept the phenazine nucleus with the 2-methoxypyridylamino group at the C-2 position intact. Our previous work indicated that the 2-methoxypyridylamino group is a privileged moiety [9]. We focused our investigation on the halogen effect of the N-5 phenyl group. In addition, we introduced cyclic alkyl group including O-containing heterocyclic group to the imino nitrogen of CFZ and evaluated their impact on antituberculosis activity and lipophilicity (ClogP).

Chemistry
The synthetic route for target compounds 8-25 is illustrated in Scheme 1. The most compounds were prepared based on the previously published protocols [9,12]. Nitro compounds 2a-d were synthesized from commercially available 2-fluoronitrobenzene (1) and aryl amines via aromatic nucleophilic substitution. The nitro group in compounds 2a-d was reduced to give amine compounds 3a-d, and followed by substitution with 1,5-difluoro-2,4-dinitrobenzene (DFDNB) to afford compounds 4a-d. The formation of the key intermediates 5a-d was achieved by aromatic nucleophilic displacement in good yields. Reduction of nitro groups in 5a-d was carried out using different methods depending on the structure of the substrate. For examples, compounds 5a and 5c were reduced by using catalytic hydrogenation, while compounds 5b and 5d were reduced by using zinc powder and glacial acetic acid. Compounds 6a-d underwent spontaneous cyclization to afford riminophenazines 7a-d. Final target compounds 8-25 were prepared by replacing the imines compounds 7a-d with different amine side chains in the presence of glacial acetic acid. Table 1 summarizes the structure and biological data for 18 new riminophenazines. CFZ and TBI-1004 were included as reference compounds. All new compounds were screened for in vitro activity against M. tuberculosis H37Rv using the Microplate Alamar Blue Assay (MABA) in 96-well plate format. The antibacterial activity of the compounds is indicated by minimum inhibitory concentration (MIC) values. The target compounds were also tested for cytotoxicity using Vero cells measured as a concentration inhibiting 50% growth (IC 50 ) as compared to a no treatment control [8]. The lipophilicity is estimated by ClogP, calculated by ChemOffice 2004 software.
Based on our systematic evaluation of lipophilicity, antituberculosis activity and cytotoxicity, two compounds 15 and 22 were selected for a multiple doses in vivo toxicity study in mice. Table 2 summarizes the tolerability data by recording the number of mice which survived after an oral administration at a dose of 500 mg/kg, once daily for 4 days, followed by a 7-day observation. As shown in Table 2 Table 3. Compound 15 demonstrated excellent efficacy in mice, the bacterial burden in the lungs was reduced by 3.8 logs colony forming units (CFU) as compared to the untreated control group and equivalent to CFZ. Furthermore, during the in vivo study, we also observed that the discoloration of the ears, internal organs and fat tissues of the mice treated with compound 15 appeared less intense as compared to CFZ. However, compound 22 failed to display efficacy in mouse model. The inconsistency between the in vivo and in vitro results might be explained by poor pharmacokinetics [16]. Finally, preliminary PK in mice was evaluated for compound 15. As shown in Table 4, compound 15 demonstrated a shorter plasma half-life (t 1/2 ), higher Cmax and AUC as compared to CFZ, which could explain its excellent efficacy and reduced skin discoloration potential compared to CFZ.   All reagents and solvents were purchased from commercial sources unless otherwise indicated. Melting points were determined on Yanaco MP-J3 melting point apparatus. Thin-layer chromatography was performed with fluorescent silica gel plates GF254, which were checked under UV (254 nm) light. 1 H-NMR spectra were recorded on Varian mercury-300 MHz or Varian-400 MHz NMR spectrometer in CDCl 3 or DMSO-d 6 . 13 C-NMR spectra were obtained on Varian-400 at 100 MHz in CDCl 3 . High-resolution mass spectra were measured on an Agilent 1100 series LC/MSD trap mass spectrometer (ESI-TOF).

Minimum Inhibitory Concentration and Cytotoxicity Assays
These were carried out according to our published protocols [8].

In vivo Acute M. tuberculosis H37Rv Infection Assay and Mouse Pharmacokinetic Study
These were carried out according to our published protocols [8].

5-(3-Fluorophenyl)-3-imino-2-(2-methoxy-3-pyridyl)amino-3,5-dihydrophenazine (7a).
A mixture of 5a (2.99 g, 6.1 mmol) and 10% Pd/C (0.3 g) in methanol and THF was shaken at room temperature under a hydrogen atmosphere (40 psi) for 15 h. After filtration, the filtrate was concentrated under vacuum and the residue 6a was dissolved in methanol. The result solution stirred at room temperature under air for 18 h. The mixture was filtered to give 1.72 g crude 7a as a red solid in 69% yield. Compound 7a was taken to the next step without further purification.

5-(3-Chlorophenyl)-3-imino-2-(2-methoxy-3-pyridyl)amino-3,5-dihydrophenazine (7b). Zinc powder
(1.63 g, 25 mmol) was added portionwise into a mixture of 5b (2.53 g, 5 mmol) and glacial acetic acid (0.86 mL, 15 mmol) in CH 2 Cl 2 (300 mL) at room temperature. The mixture was stirred until the color turned to light green and then filtered and washed with CH 2 Cl 2 . The filtrate was concentrated, and the residue was treated with water and made alkaline with ammonia. The solid was filtered, washed with water, and then dissolved in anhydrous methanol. The solution was stirred under air overnight. The solid formed was filtered to give 2.01 g crude 7b in 94% yield, which was taken to the next step without further purification.

General Procedure for Preparation of Compounds 8-25
To a solution of 7a-d (1.0 equiv.) and amine (2.0 equiv.) in dioxane was added glacial acetic acid (0.2 equiv.), then the mixture was stirred and heated at 110 °C in a sealed tube for 10-20 h. The mixture was concentrated under vacuum, and the residue was purified by column chromatography on silica gel, using petroleum ether/ethyl acetate to afford target products 8-25.   (10 (11 (12 (13 (14). Red solid. Yield 308 mg (86%  (15).  (16). Although the number, type and position of the halogen atoms on the phenyl ring at the N-5 position have no significant impact on the antituberculosis activity, the halogen substitution pattern appears important for cytotoxicity. The 3,4-dihalogen substitution pattern displayed significantly lower cytotoxicity. Compound 15 exhibited good pharmacokinetic properties, low skin pigmentation potential and excellent efficacy in vivo and warrants further evaluation. Our observations indicate that novel riminophenazine derivatives with a 2-methoxypyridylamino moiety as the promising pharmacophore have the potential to deliver a new drug candidate for the treatment of TB and MDR-TB.