Synthesis of new 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxaline and 5-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine derivatives, as potential bacterial multidrug resistance pump inhibitors

Abstract

The synthesis of new 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxaline derivatives 1a-l is described in five or six steps starting from various substituted nitroanilines 2a-e. The bioisostere 5-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine 1m was also prepared. The new derivatives were evaluated as efflux pump inhibitors (EPIs) in a model targeting the NorA system of Staphylococcus aureus. The antibiotic susceptibility of two strains overproducing NorA, SA-1199B and SA-1, was determined alone and in combination with the neo-synthesised compounds by the agar diffusion method and MIC determination, in comparison with reserpine and omeprazole taken as reference EPIs. A preliminary structure-activity relationship study firstly allowed to clarify the influence of the substituents at positions 7 and/or 8 of the pyrrolo[1,2-a]quinoxaline nucleus. Methoxy substituted compounds, 1b and 1g, were more potent EPIs than the unsubstituted compounds (1a and 1f), followed by chlorinated derivatives (1c-d and 1h). Moreover, the replacement of the N,N-diethylamino group (compounds 1a-e) by a bioisostere such as pyrrolidine (compounds 1f-h) enhanced the EPI activity, in contrast with the replacement by a piperidine moiety (compounds 1i-k). Finally, the pyrrolo[1,2-a]thieno[3,2-e]pyrazine compound 1m exhibited a higher EPI activity than its pyrrolo[1,2-a]quinoxaline analogue 1a, opening the way to further pharmacomodulation.

Keywords::

• MDR
• 4-[2-(alkylamino)ethylthio]pyrrolo[1
• 2-a]quinoxaline
• Staphylococcus aureus
• NorA efflux pump inhibitors

Introduction

Active efflux is a mechanism of cell detoxification, reported in an increasing variety of organisms, including bacteria 1-3. Probably all bacterial species possess an array of intrinsic efflux systems capable to export a wide range of structurally unrelated antibiotics and biocides, the so-called “Multi-Drug Resistant” (MDR) efflux pumps, resulting in a reduced intracellular accumulation and thus drug insensitivity [4]. Staphylococcus aureus is a major human pathogen, responsible for skin and soft tissues infections, and septicaemia. Among the MDR efflux pumps present in this species, NorA, which belongs to the Major Facilitator Superfamily (MFS) is considered both as the most efficient and as representative of the MDR systems in the Gram positive bacteria. In particular, NorA promotes the active efflux of fluoroquinolones, an important class of broad-spectrum antimicrobials with a potent activity 5-7. Furthermore, this system protects the cell against chloramphenicol, a wide range of organic compounds like ethidium bromide, rhodamine, and amphipathic cations such as benzalkonium chloride and cetrimide [8,9].

Strategies to combat efflux-mediated resistance are based on the search for either new antibiotics by-passing the efflux systems or efflux pump inhibitors (EPIs). The latter option is attractive since a single EPI active against MDR pumps and used as adjunct therapy should (i) decrease intrinsic resistance, (ii) reverse acquired resistance and (iii) reduce the emergence of highly resistant mutants towards a series of existing antibiotics. It has been previously shown that drug efflux can be inhibited by the toxic protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP), but also by drugs which are in clinical use for other indications such as the competitive pump blocker reserpine 10-1620, omeprazole [10], verapamil [10,17], paroxetine [13,18] and chlorpromazine [12]. On the other hand, the screening of natural or chemical compounds has allowed to identify compounds endowed with an EPI activity. In a library of synthetic compounds, thioquinolines and quinolines derivatives have been found to behave as EPIs on several prokaryotic and eukaryotic systems 19-21.

The pyrrolo[1,2-a]quinoxaline nucleus, previously developed as a template for the design of new compounds active on many biological targets, is an analogue of the quinoline or quinoxaline moiety 23-25. In this context and as part of a programme on the development of antibacterial agents, the aim of the present study was to synthesise new 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxaline derivatives, which are structural analogues of the thioquinolines, and to assess their in vitro activity as EPIs in a model targeting the NorA efflux pump of S. aureus (Figure 1) 19-22.

Figure 1 Structure of reserpine, omeprazole, thioquinoline derivatives and new compounds 1a-m.

Materials and methods

Chemistry

Instrumentation

Melting points were determined with an SM-LUX-POL Leitz hot-stage microscope and reported uncorrected. Infrared (IR) spectra were determined in KBr discs on a BRUKER IFS-25 spectrometer. NMR spectra were recorded on a BRUKER AVANCE 300 spectrometer (300 MHz). Chemical shifts refer to tetramethylsilane which was used as an internal standard. Elemental analyses were conducted by CNRS, Vernaison, France and the results were within ± 0.3% of their calculated values.

1-(5-Phenyl-2-nitrophenyl)pyrrole (3e)

A mixture of 5-chloro-2-nitroaniline 2e (0.018 mol) and 2,5-dimethoxytetrahydrofuran (0.018 mol) in acetic acid (40 mL) was refluxed for 1 h with vigorous stirring. After cooling, the reaction mixture was poured into water (100 mL). The precipitate was filtered, washed with water and dissolved in diethyl ether (110 mL), dried over magnesium sulfate and evaporated to dryness under reduced pressure to give a crude oil which was extracted with petroleum ether to yield 3e. Yield: 65%, orange oil; ¹H NMR δ (300 MHz, CDCl₃) 6.43 (dd, J 2.20 and 2.20 Hz, 2H, H-β), 6.89 (dd, J 2.20 and 2.20 Hz, 2H, H-α), 7.51-7.55 (m, 3H, H-4, H-3' and H-5'), 7.63-7.70 (m, 4H, H-2', H-6', H-4' and H-6), 7.99 (d, J 8.10 Hz, 1H, H-3). Anal. Calcd for C₁₆H₁₂N₂O₂: C, 72.71; H, 4.58; N, 10.60. Found: C, 72.89; H, 4.67; N, 10.53.

1-(5-Phenyl-2-aminophenyl)pyrrole (4e)

To a solution of 1-(5-chloro-2-nitrophenyl)pyrrole 3e (0.012 mol) in ethanol (45 mL) was added BiCl₃ (0.018 mol). Sodium borohydride (0.10 mol) was added portion-wise at 0°C to the reaction mixture which was then stirred at room temperature for 2 h. The reaction mixture was then poured into an aqueous hydrochloric acid solution (1M, 45 mL) and stirred for 1 h. Ethanol was evaporated under reduced pressure. The residue was made alkaline with 30% aqueous ammonium hydroxide solution and then extracted with ethyl acetate. The organic layer was dried over MgSO₄ and evaporated to dryness under reduced pressure. The oily residue was then extracted with petroleum ether to give 4e. Yield: 37%, pale-orange oil; ¹H NMR δ (300 MHz, CDCl₃) 3.81 (bs, 2H, NH₂), 6.40 (dd, J 2.15 and 2.15 Hz, 2H, H-β), 6.93 (dd, J 2.15 and 2.15 Hz, 2H, H-α), 7.28-7.47 (m, 6H, H-3, H-4, H-6, H-3', H-4' and H-5'), 7.55-7.59 (m, 2H, H-2' and H-6'). Anal. Calcd for C₁₆H₁₄N₂: C, 82.02; H, 6.02; N, 11.96. Found: C, 81.87; H, 5.81; N, 12.13.

8-Phenyl-5H-pyrrolo[1,2-a]quinoxalin-4-one (5e)

To a solution of compound 4e (4.5 mmol) in toluene (20 mL) was added triphosgene (1.5 mmol). The reaction mixture was refluxed for 4 h, and nitrogen was bubbled in to drive off excess of phosgene. The solution was then set aside for 30 min. The heavy crystalline precipitate was filtered off and washed with diethyl ether to give 5e. Yield: 36%, beige crystals, mp > 260°C; IR ν_max (KBr)/cm^{− 1} 3300 (NH), 1685 (CO). ¹H NMR δ (300 MHz, d_6-DMSO) 6.71 (dd, J 3.80 and 2.80 Hz, 1H, H-2), 7.06 (dd, J 3.80 and 1.40 Hz, 1H, H-3), 7.39 (t, J 7.45, 2H, H-3' and H-5'), 7.48 (d, J 8.40 Hz, 1H, H-6), 7.51 (t, J 7.45 Hz, 1H, H-4'), 7.62 (dd, J 8.40 and 1.80 Hz, 1H, H-7), 7.80 (d, J 7.45 Hz, 2H, H-2' and H-6'), 8.35 (d, J 1.80 Hz, 1H, H-9), 8.41 (dd, J 2.80 and 1.40 Hz, 1H, H-1), 11.32 (s, 1H, NH). Anal. Calcd for C₁₇H₁₂N₂O: C, 78.44; H, 4.65; N, 10.76. Found: C, 78.65; H, 4.78; N, 10.97.

4-Chloro-8-phenylpyrrolo[1,2-a]quinoxaline (6e)

A solution of 8-chloro-5H-pyrrolo[1,2-a]quinoxalin-4-one 5e (1.35 mmol) in POCl₃ (6 mL) was refluxed for 4 h. After removing excess of reactive under vacuum, the residue was carefully dissolved in water at 0°C and the resulting solution was made basic with 30% aqueous ammonium hydroxide solution. The precipitate was filtered, dried and recrystallized from ethyl acetate to give 6e. Yield: 72%, beige crystals, mp = 128°C; ¹H NMR δ (300 MHz, CDCl₃) 6.93 (dd, J 4.00 and 2.75 Hz, 1H, H-2), 7.09 (dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.45 (t, J 7.35 Hz, 1H, H-4'), 7.53 (t, J 7.35 Hz, 2H, H-3' and H-5'), 7.68 (dd, J 8.40 and 1.90 Hz, 1H, H-7), 7.70 (d, J 7.35 Hz, 2H, H-2' and H-6'), 7.96 (d, J 8.40 Hz, 1H, H-6), 8.00 (d, J 1.90 Hz, 1H, H-9), 8.04 (dd, J 2.75 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₇H₁₁ClN₂: C, 73.25; H, 3.98; N, 10.05. Found: C, 73.20; H, 4.11; N, 10.25.

Synthesis of Pyrrolo[1,2-a]quinoxaline-4-thiols (8a-c)

A mixture of 1-(2-aminophenyl)pyrrole 4a-c (25.3 mmol) in ethanol (120 mL) and CS₂ (89 mmol) and sodium hydroxide (50.6 mmol) in water (6 mL) was refluxed for 48 h. The product obtained after removal of ethanol was purified by dissolving in a hot 5% aqueous sodium hydroxide solution, filtering and acidifying the filtrate with acetic acid. Repeated washings with water, then with ethanol, and crystallization from ethanol gave pyrrolo[1,2-a]quinoxaline-4-thiols 8a-c.

Pyrrolo[1,2-a]quinoxaline-4-thiol (8a)

Yield: 56%, beige crystals, mp = 274°C [26]; IR ν_max (KBr)/cm^{− 1} 2380 (SH). ¹H NMR δ (300 MHz, d₆-DMSO) 6.79 (dd, J 4.00 and 2.70 Hz, 1H, H-2), 7.29 (dd, J 4.00 and 1.40 Hz, 1H, H-3), 7.34 (m, 2H, H-7 and H-8), 7.55 (dd, J 8.00 and 1.25 Hz, 1H, H-6), 8.11 (dd, J 8.00 and 1.25 Hz, 1H, H-9), 8.32 (dd, J 2.70 and 1.40 Hz, 1H, H-1), 12.90 (s, 1H, SH).

7-Methoxypyrrolo[1,2-a]quinoxaline-4-thiol (8b)

Yield: 61%, yellow crystals, mp = 245°C; IR (KBr), cm^{− 1}: 2390 (SH); ¹H NMR δ (300 MHz, d₆-DMSO) 3.81 (s, 3H, CH₃O), 6.75 (dd, J 4.05 and 2.75 Hz, 1H, H-2), 6.98 (dd, J 9.05 and 2.75 Hz, 1H, H-8), 7.12 (d, J 2.75 Hz, 1H, H-6), 7.26 (dd, J 4.05 and 1.50 Hz, 1H, H-3), 8.08 (d, J 9.05 Hz, 1H, H-9), 8.26 (dd, J 2.75 and 1.50 Hz, 1H, H-?), 8.32 (dd, J 2.70 and 1.40 Hz, 1H, H-1), 12.80 (s, 1H, SH). Anal. Calcd for C₁₂H₁₀N₂OS: C, 62.59; H, 4.38; N, 12.17. Found: C, 62.72; H, 4.37; N, 12.25.

8-Chloropyrrolo[1,2-a]quinoxaline-4-thiol (8c)

Yield: 28%, beige crystals, mp = 293°C; IR (KBr), cm^{− 1}: 2360 (SH); ¹H NMR δ (300 MHz, d₆-DMSO) 6.82 (dd, J 4.00 and 2.75 Hz, 1H, H-2), 7.32 (dd, J 4.00 and 1.45 Hz, 1H, H-3), 7.44 (dd, J 8.70 and 2.10 Hz, 1H, H-7), 7.56 (d, J 8.70 Hz, H-6), 8.36 (d, J 2.10 Hz, 1H, H-9), 8.41 (dd, J 2.75 and 1.45 Hz, 1H, H-1), 13.00 (s, 1H, SH). Anal. Calcd for C₁₁H₇ClN₂S: C, 56.29; H, 3.01; N, 11.94. Found: C, 56.49; H, 2.95; N, 12.04.

Synthesis of 4-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]quinoxalines (7a-d), and 5-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine (10)

Method A: To the sodium salt of N,N-diethylaminoethylthiol prepared from the thiol (3 mmol) and sodium hydride (4.5 mmol of 60% suspension in mineral oil) in dioxane (40 mL) in 1 h was added 4-chloropyrrolo[1,2-a]quinoxaline 6a-d or 4-chloropyrrolo[1,2-a]thieno[3,2-e]pyrazine 9 (3 mmol). The mixture was heated at 100°C for 4 h. Dioxane was evaporated under reduced pressure, and the oily residue was treaturated with water then extracted with diethyl ether. The organic layer was washed with water (100 mL), dried over sodium sulfate and evaporated to dryness. The residue was flash chromatographed on silica gel, eluting with 90% chloroform/10% methanol to give 7a-d and 10. Method B: To a mixture of CuI (0.12 mmol), potassium carbonate (45.3 mmol) and 4-chloropyrrolo[1,2-a]quinoxaline 6a (2.46 mmol) under nitrogen were added 2-propanol (5 mL), ethylene glycol (4.93 mmol) and the N,N-diethylaminoethylthiol (2.46 mmol). The reaction mixture was then heated to 80°C and stirred for 24 h. After cooling, ethyl acetate (15 mL) was added. The reaction mixture was then filtered and concentrated under reduced pressure. The crude product was purified by flash chromatography on silica gel, eluting with 90% chloroform / 10% methanol to afford the desired thioether 7a. Method C: To a mixture of t-BuONa (3.69 mmol), CuI (0.25 mmol), and neocuproine (0.25 mmol) under nitrogen were added the N,N-diethylaminoethylthiol (2.71 mmol), the 4-chloropyrrolo[1,2-a]quinoxaline 6a (2.46 mmol) and toluene (15 mL). The reaction mixture was then stirred at 110°C for 24 h. The mixture was then cooled to room temperature and filtered to remove any insoluble residues. The filtrate was concentrated in vacuo, the residue was then purified by flash chromatography on silica gel, eluting with 90% chloroform/10% methanol, to obtain the pure product 7a. Method D: A solution of 6a (2.46 mmol), N,N-diethylaminoethylthiol (2.46 mmol), t-BuOK (4.92 mmol), and catalytic amount of Pd (PPh₃)₄ (0.1 mmol) in n-butanol (25 mL) was heated at 110°C under a nitrogen stream with stirring for 5 h. The mixture was then cooled to room temperature and filtered. The filtrate was evaporated under reduced pressure, the residue was then purified by flash chromatography on silica gel, eluting with 90% chloroform / 10% methanol, to obtain the pure product 7a.

4-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline (7a)

Yield: 42% (method A), 60% (method B), 80% (method C), 66%(method D), yellow oil; ¹H NMR δ (300 MHz, CDCl₃) 0.98 (t, J 7.10 Hz, 6H, 2CH₃), 2.55 (q, J 7.10 Hz, 4H, 2CH₂), 2.72 (t, J 7.20 Hz, 2H, CH₂), 3.37 (t, J 7.20 Hz, 2H, CH₂), 6.85 (m, 2H, H-2 and H-3), 7.43 (t, J 7.50 Hz, 1H, H-7), 7.45 (t, J 7.50 Hz, 1H, H-8), 7.70 (d, J 7.50 Hz, 1H, H-6), 8.20 (d, J 7.50 Hz, 1H, H-9), 8.38 (dd, J 2.70 and 1.45 Hz, 1H, H-1). Anal. Calcd for C₁₇H₂₁N₃S: C, 68.19; H, 7.07; N, 14.03. Found: C, 68.28; H, 6.95; N, 14.21.

7-Methoxy-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline (7b)

Yield: 44% (method A), orange oil; ¹H NMR δ (300 MHz, CDCl₃) 1.01 (t, J 7.15 Hz, 6H, 2CH₃), 2.60 (q, J 7.15 Hz, 4H, 2CH₂), 2.77 (t, J 7.40 Hz, 2H, CH₂), 3.41 (t, J 7.40 Hz, 2H, CH₂), 3.85 (s, 3H, CH₃O), 6.81 (dd, J 4.15 and 2.45 Hz, 1H, H-2), 6.83 (dd, J 4.15 and 1.30 Hz, 1H, H-3), 7.10 (dd, J 8.95 and 2.70 Hz, 1H, H-8), 7.20 (d, J 2.70 Hz, 1H, H-6), 8.13 (d, J 8.95 Hz, 1H, H-9), 8.33 (dd, J 2.45 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₈H₂₃N₃SO: C, 65.62; H, 7.03; N, 12.76. Found: C, 65.75; H, 7.17; N, 12.79.

8-Chloro-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline (7c)

Yield: 45% (method A), yellow oil; ¹H NMR δ (300 MHz, d₆-DMSO) 1.00 (t, J 7.20 Hz, 6H, 2CH₃), 2.54 (q, J 7.20 Hz, 4H, 2CH₂), 2.73 (t, J 7.35 Hz, 2H, CH₂), 3.37 (t, J 7.35 Hz, 2H, CH₂), 6.86 (dd, J 3.90 and 2.85 Hz, 1H, H-2), 6.88 (dd, J 3.90 and 1.30 Hz, 1H, H-3), 7.43 (dd, J 8.65 and 2.30 Hz, 1H, H-7), 7.68 (d, J 8.65 Hz, 1H, H-6), 8.40 (d, J 2.30 Hz, 1H, H-9), 8.45 (dd, J 2.85 and 1.30 Hz, 1H, H-1); ¹³C NMR δ (100 MHz, d₆-DMSO) 12.0 (CH₃), 25.7 (CH₂), 46.4 (CH₂), 51.6 (CH₂), 105.9 (C-2), 114.0 (C-3), 114.7 (C-1), 116.9 (C-9), 123.3 (C-3a), 125.5 (C-8), 127.0 (C-9a), 129.1 (C-7), 130.5 (C-6), 133.7 (C-5a), 153.9 (C-4). Anal. Calcd for C₁₇H₂₀ClN₃S: C, 61.15; H, 6.04; N, 12.59. Found: C, 61.02; H, 6.15; N, 12.74.

7-Chloro-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline (7d)

Yield: 26% (method A), orange oil; ¹H NMR δ (300 MHz, CDCl₃) 1.18 (t, J 7.15 Hz, 6H, 2CH₃), 2.77 (q, J 7.15 Hz, 4H, 2CH₂), 2.95 (t, J 7.55 Hz, 2H, CH₂), 3.52 (t, J 7.55 Hz, 2H, CH₂), 6.81 (dd, J 4.00 and 2.75 Hz, 1H, H-2), 6.93 (dd, J 4.00 and 1.25 Hz, 1H, H-3), 7.38 (dd, J 8.70 and 2.35 Hz, 1H, H-8), 7.73 (d, J 8.70 Hz, 1H, H-9), 7.79 (d, J 2.35 Hz, 1H, H-6), 7.84 (dd, J 2.75 and 1.25 Hz, 1H, H-1). Anal. Calcd for C₁₇H₂₀ClN₃S: C, 61.15; H, 6.04; N, 12.59. Found: C, 61.28; H, 6.14; N, 12.49.

8-Phenyl-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline (7e)

Yield: 38% (method A), yellow oil; ¹H NMR δ (300 MHz, CDCl₃) 1.19 (t, J 7.15 Hz, 6H, 2CH₃), 2.76 (q, J 7.15 Hz, 4H, 2CH₂), 2.95 (t, J 7.45 Hz, 2H, CH₂), 3.55 (t, J 7.45 Hz, 2H, CH₂), 6.84 (dd, J 4.00 and 2.75 Hz, 1H, H-2), 6.94 (dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.44 (t, J 7.40 Hz, 1H, H-4'), 7.52 (t, J 7.40 Hz, 2H, H-3' and H-5'), 7.70 (dd, J 8.40 and 1.90 Hz, 1H, H-7), 7.73 (d, J 7.40 Hz, 2H, H-2' and H-6'), 7.89 (d, J 8.40 Hz, 1H, H-6), 7.95 (dd, J 2.75 and 1.30 Hz, 1H, H-1), 7.99 (d, J 1.90 Hz, 1H, H-9). Anal. Calcd for C₂₃H₂₅N₃S: C, 73.56; H, 6.71; N, 11.19. Found: C, 73.75; H, 6.85; N, 11.36.

5-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine (10)

Yield: 33% (method A), yellow oil; ¹H NMR δ (300 MHz, CDCl₃) 1.18 (t, J = 7.05, 6H, 2CH₃), 2.75 (q, J = 7.05, 4H, 2CH₂), 2.95 (t, J 7.40 Hz, 2H, CH₂), 3.51 (t, J 7.40 Hz, 2H, CH₂), 6.86 (dd, J 4.15 and 2.65 Hz, 1H, H-7), 6.92 (dd, J 4.15 and 1.30 Hz, 1H, H-6), 7.08 (d, J 5.65 Hz, 1H, H-3), 7.34 (d, J 5.65 Hz, 1H, H-2), 7.45 (dd, J 2.60 and 1.30 Hz, 1H, H-8). Anal. Calcd for C₁₅H₁₉N₃S₂: C, 58.98; H, 6.27; N, 13.76. Found: C, 58.85; H, 6.35; N, 13.89.

Synthesis of 4-[2-(Pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxalines (7f-h), 4-[2-(Piperidino)ethylthio]pyrrolo[1,2-a]quinoxalines (7i-k), and 4-Butylthio]pyrrolo[1,2-a]quinoxaline (7l)

To a solution of the pyrrolo[1,2-a]quinoxaline-4-thiol 8a-c (5 mmol) in DMF (25 mL) was added sodium hydride (7.5 mmol of 60% suspension in mineral oil). The mixture was stirred at room temperature for 2 h, and then treated with 2-pyrrolidinoethyl chloride, or 2-piperidinoethyl chloride or butyl chloride (5.5 mmol) in DMF (8 mL). The temperature was raised to 90°C and the mixture kept stirred for 18 h. The solvent was then removed under reduced pressure. After treatment of the residue by an excess of HCl (2 M), the solution was filtered. The filtrate was extracted with diethyl ether and the aqueous phase made alkaline with 30% aqueous ammonium hydroxide solution, then extracted with diethyl ether (100 mL). The organic layer was washed with water (80 mL), dried over sodium sulfate and evaporated to dryness. The residue was flash chromatographed on silica gel, eluting with 95% dichloromethane / 5% methanol to give 7f-l.

4-[2-(Pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7f)

Yield: 80%, orange oil; ¹H NMR δ (300 MHz, CDCl₃) 1.87 (m, 4H, 2CH₂ pyrrolidine), 2.73 (m, 4H, 2CH₂ pyrrolidine), 2.92 (t, J 7.40 Hz, 2H, CH₂), 3.63 (t, J 7.40 Hz, 2H, CH₂), 6.84 (dd, J 4.00 and 2.75 Hz, 1H, H-2), 6.94 dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.45 (m, 2H, H-7 and H-8), 7.82 (d, J 7.55 Hz, 1H, H-6), 7.87 (d, J 7.55 Hz, 1H, H-9), 7.89 (dd, J 2.75 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₇H₁₉N₃S: C, 68.65; H, 6.44; N, 14.13. Found: C, 68.68; H, 6.61; N, 14.23.

7-Methoxy-4-[2-(pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7 g)

Yield: 57%, orange oil; ¹H NMR δ (300 MHz, CDCl₃) 1.92 (m, 4H, 2CH₂ pyrrolidine), 2.79 (m, 4H, 2CH₂ pyrrolidine), 2.99 (t, J 7.45 Hz, 2H, CH₂), 3.66 (t, J 7.45 Hz, 2H, CH₂), 3.92 (s, 3H, CH₃O), 6.77 (dd, J 4.05 and 2.70 Hz, 1H, H-2), 6.90 (dd, J 4.05 and 1.30 Hz, 1H, H-3), 7.04 (dd, J 8.95 and 2.85 Hz, 1H, H-8), 7.33 (d, J 2.85 Hz, 1H, H-6), 7.72 (d, J 8.95 Hz, 1H, H-9), 7.82 (dd, J 2.70 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₈H₂₁N₃SO: C, 66.02; H, 6.46; N, 12.83. Found: C, 65.85; H, 6.56; N, 12.98.

8-Chloro-4-[2-(pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7 h)

Yield: 79%, orange crystals, mp = 37°C; ¹H NMR δ (300 MHz, CDCl₃) 1.87 (m, 4H, 2CH₂ pyrrolidine), 2.77 (m, 4H, 2CH₂ pyrrolidine), 2.96 (t, J 7.40 Hz, 2H, CH₂), 3.62 (t, J 7.40 Hz, 2H, CH₂), 6.82 (dd, J 4.00 and 2.80 Hz, 1H, H-2), 6.93 (dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.37 (dd, J 8.65 and 2.20 Hz, 1H, H-7), 7.76 (d, J 8.65 Hz, 1H, H-6), 7.79 (d, J 2.20 Hz, 1H, H-9), 7.81 (dd, J 2.80 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₇H₁₈ClN₃S: C, 61.52; H, 5.47; N, 12.66. Found: C, 61.68; H, 5.64; N, 12.77.

4-[2-(Piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7i)

Yield: 64%, yellow oil; ¹H NMR δ (300 MHz, CDCl₃) 1.35 (m, 2H, CH₂ piperidine), 1.48 (m, 4H, 2CH₂ piperidine), 2.43 (m, 4H, 2CH₂ piperidine), 2.59 (t, J 7.30 Hz, 2H, CH₂), 3.46 (t, J 7.30 Hz, 2H, CH₂), 6.86 (m, 2H, H-2 and H-3), 7.46 (m, 2H, H-7 and H-8), 7.74 (d, J 7.45 Hz, 1H, H-6), 8.18 (d, J 7.45 Hz, 1H, H-9), 8.38 (dd, J 2.70 and 1.35 Hz, 1H, H-1). Anal. Calcd for C₁₈H₂₁N₃S: C, 69.41; H, 6.79; N, 13.49. Found: C, 69.59; H, 6.60; N, 13.77.

7-Methoxy-4-[2-(piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7j)

Yield: 61%, orange oil; ¹H NMR δ (300 MHz, d₆-DMSO) 1.50 (m, 2H, CH₂ piperidine), 1.70 (m, 4H, 2CH₂ piperidine), 2.60 (m, 4H, 2CH₂ piperidine), 2.80 (t, J 7.60 Hz, 2H, CH₂), 3.60 (t, J 7.60 Hz, 2H, CH₂), 6.77 (dd, J 4.00 and 2.70 Hz, 1H, H-2), 6.90 (dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.05 (dd, J 8.95 and 2.80 Hz, 1H, H-8), 7.35 (d, J 2.80 Hz, 1H, H-6), 7.72 (d, J 8.95 hz, 1H, H-9), 7.80 (dd, J 2.70 and 1.30 Hz, 1H, H-1). Anal. Calcd for C₁₉H₂₃N₃SO: C, 66.83; H, 6.79; N, 12.31. Found: C, 66.97; H, 6.58; N, 12.47.

8-Chloro-4-[2-(piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline (7k)

Yield: 37%, orange crystals, mp = 66°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.45 (m, 2H, CH₂ piperidine), 1.65 (m, 4H, 2CH₂ piperidine), 2.60 (m, 4H, 2CH₂ piperidine), 2.75 (t, J 7.60 Hz, 2H, CH₂), 3.60 (t, J 7.60 Hz, 2H, CH₂), 6.80 (dd, J 3.95 and 2.80 Hz, 1H, H-2), 6.90 (dd, J 3.95 and 1.20 Hz, 1H, H-3), 7.38 (dd, J 8.65 and 2.15 Hz, 1H, H-7), 7.72 (d, J 8.65 Hz, 1H, H-6), 7.78 (m, 2H, H-1 and H-9). Anal. Calcd for C₁₈H₂₀ClN₃S: C, 62.50; H, 5.83; N, 12.15. Found: C, 62.74; H, 5.97; N, 12.32.

4-(Butylthio)pyrrolo[1,2-a]quinoxaline (7l)

Yield: 48%, yellow crystals, mp = 110°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.00 (t, 3H, J = 7.30, CH₃), 1.58 (sextuplet, J 7.30 Hz, 2H, CH₂), 1.84 (qt, J 7.30 Hz, 2H, CH₂), 4.23 (t, J 7.30 Hz, 2H, CH₂), 6.96 (m, 1H, H-2), 7.25 (m, 1H, H-3), 7.46 (m, 1H, H-8), 7.53 (m, 1H, H-7), 7.86 (d, J 7.25 Hz, 1H, H-6), 8.08 (m, 1H, H-1), 9.24 (d, J 8.85 Hz, 1H, H-9). Anal. Calcd for C₁₅H₁₆N₂S: C, 70.27; H, 6.29; N, 10.93. Found: C, 70.36; H, 6.41; N, 11.05.

Synthesis of 4-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]quinoxalines, 4-[2-(Pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxalines, 4-[2-(Piperidino)ethylthio]pyrrolo[1,2-a]quinoxalines, 4-(Butylthio)pyrrolo[1,2-a]quinoxalines and 5-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine Hydrochloride (1a-m)

To solution of amines 7a-l and 10 (20 mmol) in diethyl ether (60 mL) was bubbled with an hydrochloric acid gas flow. The precipitate was filtered, washed with diethyl ether and dried to give 1a-m as crystals.

4-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1a)

Yield: 69%, beige crystals, mp = 204°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.37 (t, J 7.20 Hz, 6H, 2CH₃), 3.25 (m, 4H, 2CH₂), 3.39 (m, 2H, CH₂), 3.71 (m, 2H, CH₂), 4.04 (bs, 1H, NH⁺), 6.91 (dd, J 3.95 and 2.80 Hz, 1H, H-2), 6.94 (dd, J 3.95 and 1.25 Hz, 1H, H-3), 7.49 (t, J 7.70 Hz, 1H, H-7), 7.56 (t, J 7.70 Hz, 1H, H-8), 7.82 (d, J 7.70 Hz, 1H, H-6), 8.28 (d, J 7.70 Hz, 1H, H-9), 8.49 (dd, J 2.80 and 1.25 Hz, 1H, H-1), 10.75 (bs, 1H, NH⁺). Anal. Calcd for C₁₇H₂₃Cl₂N₃S: C, 54.83; H, 6.22; N, 11.28. Found: C, 55.11; H, 6.55; N, 11.46.

7-Methoxy-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1b)

Yield: 73%, beige crystals, mp = 172°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.32 (t, J 7.20 Hz, 6H, 2CH₃), 3.22 (m, 4H, 2CH₂), 3.51 (m, 2H, CH₂), 3.65 (m, 2H, CH₂), 3.86 (s, 3H, CH₃O), 4.10 (bs, 1H, NH⁺), 6.87 (dd, J 4.10 and 2.85 Hz, 1H, H-2), 6.91 (dd, J 4.10 and 1.10 Hz, 1H, H-3), 7.19 (dd, J 9.05 and 2.80 Hz, 1H, H-8), 7.23 (d, J 2.80 Hz, 1H, H-6), 8.23 (d, J 9.05 Hz, 1H, H-9), 8.44 (dd, J 2.85 and 1.10 Hz, 1H, H-1), 9.93 (bs, 1H, NH⁺). Anal. Calcd for C₁₈H₂₅Cl₂N₃SO: C, 53.73; H, 6.26; N, 10.44. Found: C, 53.98; H, 6.10; N, 10.65.

8-Chloro-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1c)

Yield: 77%, beige crystals, mp = 206°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.32 (t, J 7.15 Hz, 6H, 2CH₃), 3.25 (m, 4H, 2CH₂), 3.38 (m, 2H, CH₂), 3.70 (m, 2H, CH₂), 4.06 (bs, 1H, NH⁺), 6.93 (m, 1H, H-2), 6.97 (m, 1H, H-3), 7.52 (dd, J 8.65 and 2.15 Hz, 1H, H-7), 7.81 (d, J 8.65 Hz, 1H, H-6), 8.48 (d, J 2.15 Hz, 1H, H-9), 8.55 (m, 1H, H-1), 10.78 (bs, 1H, NH⁺). Anal. Calcd for C₁₇H₂₂Cl₃N₃S: C, 50.19; H, 5.45; N, 10.33. Found: C, 50.35; H, 5.68; N, 10.52.

7-Chloro-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1d)

Yield: 63%, beige crystals, mp = 221°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.22 (t, J 7.25 Hz, 6H, 2CH₃), 3.24 (m, 4H, 2CH₂), 3.35 (m, 2H, CH₂), 3.80 (m, 2H, CH₂), 4.03 (bs, 1H, NH⁺), 6.93 (m, 1H, H-2), 6.98 (m, 1H, H-3), 7.61 (dd, J 8.85 and 2.20 Hz, 1H, H-8), 7.85 (d, J 2.20 Hz, 1H, H-6), 8.33 (d, J 8.85 Hz, 1H, H-9), 8.52 (m, 1H, H-1), 10.32 (bs, 1H, NH⁺). Anal. Calcd for C₁₇H₂₂Cl₃N₃S: C, 50.19; H, 5.45; N, 10.33. Found: C, 50.11; H, 5.61; N, 10.20.

8-Phenyl-4-[2-(N,N-diethylamino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1e)

Yield: 83%, yellow crystals, mp = 61°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.35 (t, J 7.15 Hz, 6H, 2CH₃), 3.28 (q, J 7.15 Hz, 4H, 2CH₂), 3.42 (t, J 7.50 Hz, 2H, CH₂), 3.75 (t, J 7.50 Hz, 2H, CH₂), 4.43 (bs, 1H, NH⁺), 6.93 (m, 1H, H-2), 6.97 (m, 1H, H-3), 7.43 (t, J 7.45 Hz, 1H, H-4'), 7.54 (t, J 7.45 Hz, 2H, H-3' and H-5'), 7.83 (dd, J 8.30 and 1.95 Hz, 1H, H-7), 7.90 (m, 3H, H-6, H-2' and H-6'), 8.59 (d, J 1.95 Hz, 1H, H-9), 8.72 (m, 1H, H-1), 10.79 (bs, 1H, NH⁺). Anal. Calcd for C₂₃H₂₇Cl₂N₃S: C, 61.60; H, 6.07; N, 9.37. Found: C, 61.69; H, 5.86; N, 9.52.

4-[2-(Pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1f)

Yield: 69%, white crystals, mp = 170°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.93 (m, 2H, CH₂ pyrrolidine), 2.01 (m, 2H, CH₂ pyrrolidine), 3.15 (m, 2H, CH₂ pyrrolidine), 3.49 (m, 2H, CH₂), 3.65 (m, 2H, CH₂ pyrrolidine), 3.72 (m, 2H, CH₂), 4.49 ((bs, 1H, NH⁺), 6.92 (m, 2H, H-2 and H-3), 7.49 (t, J 7.80 Hz, 1H, H-7), 7.55 (t, J 7.80 Hz, 1H, H-8), 7.90 (d, J 7.80 Hz, 1H, H-6), 8.26 (d, J 7.80 Hz, 1H, H-9), 8.49 (dd, J 2.50 and 1.25 Hz, 1H, H-1), 11.21 (bs, 1H, NH⁺). Anal. Calcd for C₁₇H₂₁Cl₂N₃S: C, 55.13; H, 5.71; N, 11.35. Found: C, 55.36; H, 5.98; N, 11.14.

7-Methoxy-4-[2-(pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1g)

Yield: 82%, white crystals, mp = 168°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.93 (m, 2H, CH₂ pyrrolidine), 2.03 (m, 2H, CH₂ pyrrolidine), 3.14 (m, 2H, CH₂ pyrrolidine), 3.50 (m, 2H, CH₂), 3.69 (m, 4H, CH₂ and CH₂ pyrrolidine), 3.87 (s, 3H, CH₃O), 4.58 (bs, 1H, NH⁺), 6.85 (dd, J 4.05 and 2.60 Hz, 1H, H-2), 6.89 (dd, J 4.05 and 1.30 Hz, 1H, H-3), 7.17 (dd, J 9.00 and 2.85 Hz, 1H, H-8), 7.45 (d, J 2.85 Hz, 1H, H-6), 8.20 (d, J 9.00 Hz, 1H, H-9), 8.42 (dd, J 2.60 and 1.30 Hz, 1H, H-1), 11.14 (bs, 1H, NH⁺). Anal. Calcd for C₁₈H₂₃Cl₂N₃SO: C, 54.00; H, 5.79; N, 10.50. Found: C, 53.84; H, 5.53; N, 10.81.

8-Chloro-4-[2-(pyrrolidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1h)

Yield: 65%, orange crystals, mp = 199°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.89 (m, 2H, CH₂ pyrrolidine), 2.04 (m, 2H, CH₂ pyrrolidine), 3.15 (m, 2H, CH₂ pyrrolidine), 3.53 (m, 2H, CH₂), 3.64 (m, 2H, CH₂ pyrrolidine), 3.71 (m, 2H, CH₂), 4.60 (bs, 1H, NH⁺), 6.93 (dd, J 3.90 and 2.85 Hz, 1H, H-2), 6.98 (dd, J 3.90 and 1.15 Hz, 1H, H-3), 7.53 (dd, J 8.55 and 2.05 Hz, 1H, H-7), 7.90 (d, J 8.55 Hz, 1H, H-6), 8.49 (d, J 2.05 Hz, 1H, H-9), 8.55 (dd, J 2.85 and 1.15 Hz, 1H, H-1), 10.69 (bs, 1H, NH⁺). Anal. Calcd for C₁₇H₂₀Cl₃N₃S: C, 50.44; H, 4.98; N, 10.38. Found: C, 50.72; H, 5.15; N, 10.59.

4-[2-(Piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1i)

Yield: 85%, white crystals, mp = 204°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.43 (m, 1H, CH₂ piperidine), 1.74 (m, 5H, CH₂ piperidine), 3.01 (m, 2H, CH₂), 3.39 (m, 2H, CH₂ piperidine), 3.58 (m, 2H, CH₂ piperidine), 3.75 (m, 2H, CH₂), 3.90 (bs, 1H, NH⁺), 6.91 (dd, J 4.00 and 2.70 Hz, 1H, H-2), 6.94 (dd, J 4.00 and 1.30 Hz, 1H, H-3), 7.49 (t, J 8.00 Hz, 1H, H-7), 7.56 (t, J 8.00 Hz, 1H, H-8), 7.89 (d, J 8.00 Hz, 1H, H-6), 8.27 (d, J 8.00 Hz, 1H, H-9), 8.49 (dd, J 2.70 and 1.30 Hz, 1H, H-1), 10.56 (bs, 1H, NH⁺). Anal. Calcd for C₁₈H₂₃Cl₂N₃S: C, 56.25; H, 6.03; N, 10.93. Found: C, 56.39; H, 5.84; N, 11.17.

7-Methoxy-4-[2-(piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1j)

Yield: 71%, beige crystals, mp = 214°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.42 (m, 1H, CH₂ piperidine), 1.71 (m, 3H, CH₂ piperidine), 1.84 (m, 2H, CH₂ piperidine), 3.05 (m, 2H, CH₂ piperidine), 3.34 (m, 2H, CH₂), 3.51 (m, 2H, CH₂ piperidine), 3.70 (m, 2H, CH₂), 3.88 (s, 3H, CH₃O), 3.95 (bs, 1H, NH⁺), 6.87 (dd, J 3.95 and 2.65 Hz, 1H, H-2), 6.91 (dd, J 3.95 and 1.25 Hz, 1H, H-3), 7.19 (dd, J 8.95 and 2.75 Hz, 1H, H-8), 7.35 (d, J 2.75 Hz, 1H, H-6), 8.22 (d, J 8.95 Hz, 1H, H-9), 8.42 (dd, J 2.65 and 1.20 Hz, 1H, H-1), 9.60 (bs, 1H, NH⁺). Anal. Calcd for C₁₉H₂₅Cl₂N₃SO: C, 55.07; H, 6.08; N, 10.14. Found: C, 54.85; H, 6.25; N, 10.28.

8-Chloro-4-[2-(piperidino)ethylthio]pyrrolo[1,2-a]quinoxaline hydrochloride (1k)

Yield: 72%, beige crystals, mp = 229°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.45 (m, 1H, CH₂ piperidine), 1.78 (m, 5H, CH₂ piperidine), 3.00 (m, 2H, CH₂ piperidine), 3.24-3.39 (m, 4H, CH₂ piperidine and CH₂), 3.70 (t, J 7.60 Hz, 2H, CH₂), 3.78 (bs, 1H, NH⁺), 6.94 (dd, J 3.75 and 2.85 Hz, 1H, H-2), 6.98 (dd, J 3.75 and 1.20 Hz, 1H, H-3), 7.53 (dd, J 8.95 and 2.00 Hz, 1H, H-7), 7.87 (d, J 8.95 Hz, 1H, H-6), 8.49 (d, J 2.00 Hz, 1H, H-9), 8.55 (dd, J 2.85 and 1.20 Hz, 1H, H-1), 9.83 (bs, 1H, NH⁺). Anal. Calcd for C₁₈H₂₂Cl₃N₃S: C, 51.62; H, 5.29; N, 10.03. Found: C, 51.50; H, 5.36; N, 9.85.

4-(Butylthio)pyrrolo[1,2-a]quinoxaline hydrochloride (1l)

Yield: 73%, pale-yellow crystals, mp = 156°C; ¹H NMR δ (300 MHz, d₆-DMSO) 0.94 (t, 3H, J 7.35 Hz, CH₃), 1.47 (sextuplet, J 7.35 Hz, 2H, CH₂), 1.72 (qt, J 7.35 Hz, 2H, CH₂), 3.38 (t, J 7.35 Hz, 2H, CH₂), 4.23 (bs, 1H, NH⁺), 6.89 (m, 2H, H-2 and H-3), 7.46 (t, J 7.90 Hz, 1H, H-8), 7.52 (t, J 7.90 Hz, 1H, H-7), 7.78 (d, J 7.90 Hz, 1H, H-6), 8.25 (d, J 7.90 Hz, 1H, H-9), 8.44 (dd, J 2.65 and 1.40 Hz, 1H, H-1). Anal. Calcd for C₁₅H₁₇ClN₂S: C, 61.52; H, 5.85; N, 9.57. Found: C, 61.68; H, 6.02; N, 9.53.

5-[2-(N,N-Diethylamino)ethylthio]pyrrolo[1,2-a]thieno[3,2-e]pyrazine hydrochloride (1m)

Yield: 68%, pale-yellow crystals, mp = 220°C; ¹H NMR δ (300 MHz, d₆-DMSO) 1.28 (t, J 6.40 Hz, 6H, 2CH₃), 3.23 (m, 4H, 2CH₂), 3.32 (m, 2H, CH₂), 3.61 (t, J 6.85 Hz, 2H, CH₂), 3.82 (bs, 1H, NH⁺), 6.96 (m, 2H, H-6 and H-7), 7.39 (d, J 5.65 Hz, 1H, H-3), 7.53 (d, J 5.65 Hz, 1H, H-?), 8.02 (m, 1H, H-8), 9.90 (bs, 1H, NH⁺). Anal. Calcd for C₁₅H₂₁Cl₂N₃S₂: C, 47.61; H, 5.59; N, 11.10. Found: C, 47.74; H, 5.78; N, 11.26.

Pharmacology

Bacterial strains and media

The strains of S. aureus used in this study included SA 1199 (a fluoroquinolone susceptible clinical isolate) and SA 1199B (its overproducing NorA mutant), generously provided as gifts from G.W. Kaatz (University of Michigan, USA) [6,14,15] and S. aureus ATCC 25923, a reference strain and its overproducing NorA mutant SA-1, selected in our laboratory [27]. These strains were routinely cultured on Mueller Hinton agar (MH) (BIO-RAD, Marnes-la-Coquette, France) and/or on broth adjusted to contain 20 μg/ml of Ca^{2 +} and 10 μg/ml of Mg^{2 +} (AES, Bruz, France), at 37°C. They were stored in 30% glycerol Brain Heart broth at − 80°C.

Antibiotic and chemicals

Norfloxacin, reserpine, and omeprazole were purchased from Sigma-Aldrich (Saint Quentin Fallavier, France). Solutions of reserpine, omeprazole, and the candidate inhibitors 1a-m were extemporaneously prepared in 100% dimethylsulfoxide (DMSO), but the highest final concentration of DMSO present in all assays (4%, v/v) caused no inhibition of bacterial growth (data not shown).

Antibiotic susceptibility test

The antibiotic susceptibility of the strains was determined by the disk diffusion method in MH medium alone or supplemented with the EPIs at 1/4 MIC [28]. The two hydrophilic fluoroquinolones, norfloxacin (5 μg) and ciprofloxacin (5 μg), which are the main substrates of NorA efflux pump were tested. After overnight incubation at 37°C, the inhibition zone diameters were measured. All tests were done at least in triplicate and the mean values were retained. Then, results were expressed as the percentages of NorA efflux pump inhibition, calculated as follows: [(diameter for the resistant strain in presence of EPI) minus (diameter for the resistant strain without EPI)] divided by [(diameter for the susceptible strain without EPI) minus (diameter for the resistant strain without EPI)] and multiplied by 100.

MIC determinations

The minimum inhibitory concentrations (MICs) of norfloxacin, EPIs and their combinations were determined by an agar dilution method in MH medium [28]. Serial two-fold dilutions of the stock solutions were prepared extemporaneously and incorporated to agar plates. The bacterial inocula were obtained by suspending fresh cultures in peptone water, to an optical density equivalent to a 0.5 Mac Farland ( ≈ 1.5 × 10⁸ CFU/mL). Then, these suspensions were applied to the surface of the drug-containing media by means of a Steers replicator, yielding a final concentration of approximately 10⁴ colony forming units (CFU) per spot. After incubation for 24 h in air at 37°C, the MIC was defined as the lowest concentration that inhibited any visible growth. All tests were done at least in triplicate and the mode values were retained.

Results and discussion

Chemistry

The synthesis of the new alkylthiopyrrolo[1,2-a]quinoxalines 1a-l has been achieved starting from various substituted 2-nitroanilines 2a-e in five or six steps (Scheme 1). Not commercially available 5-phenyl-2-nitroaniline 2e was prepared via the Suzuki-Miyaura reaction by coupling 5-chloro-2-nitroaniline 2c with phenylboronic acid using Pd(OAc)₂ as catalyst in the presence of the 2-(biphenyl)-di-tert-butylphosphine ligand [29]. The Clauson-Kaas reaction [30] of anilines 2a-e with 2,5-dimethoxytetrahydrofuran (DMTHF) in acetic acid gave the pyrrolic derivatives 3a-e, which were reduced using a BiCl₃-NaBH₄ treatment to provide the attempted 1-(2-aminophenyl)pyrroles 4a-e [31,32]. The cyclization was then possible between the NH₂ and the C-α of the pyrrole ring by reacting compounds 4a-e with triphosgene in toluene to give the lactams 5a-e, which were subsequently chlorodehydroxylated with phosphorous oxychloride to obtain the chloroquinoxalines 6a-e. Finally a nucleophilic substitution of the chlorine atom in compounds 6a-e with an appropriate dialkylaminoethylthiol in the presence of NaH as a base furnished the pyrrolo[1,2-a]quinoxalines 7a-e.

Scheme 1 Synthesis of compounds 1a-l. Reagents: (i) DMTHF, CH₃COOH, Δ; (ii) BiCl₃ / NaBH₄, EtOH; (iii) CO(OCCl₃)₂, toluene, Δ; (iv) POCl₃, Δ; (v) Method A: NaH, HS-CH₂-CH₂-R₃, dioxane; Method B: K₂CO₃, HS-CH₂-CH₂-R₃, CuI, HOCH₂CH₂OH, 2-propanol; Method C: t-BuONa, HS-CH₂-CH₂-R₃, CuI, neocuproine, toluene; Method D: t-BuOK, HS-CH₂-CH₂-R₃, Pd(PPh₃)₄, n-BuOH (vi) CS₂, NaOH, EtOH; (vii) NaH, Cl-CH₂-CH₂-R₃, DMF; (viii) HCl, Et₂O.

Due to the moderate yields (26–45%) obtained by this first method in the preparation of 7a-e (method A), we decided to investigate metal-catalyzed nucleophilic substitutions of 4-chloropyrrolo[1,2-a]quinoxaline 6a by the N,N-diethylaminoethylthiol (Table I). First, a copper-catalyzed carbon-sulfur bond formation reaction was developed (method B). Thus, optimised reaction conditions using 5 mol % CuI, K₂CO₃ (2 eq.), and ethylene glycol (2 eq.) in reagent-grade 2-propanol at 80°C under nitrogen led to 7a (yield: 60%) [33,34]. Then, a third new method (method C) used a mild palladium-free synthetic protocol for the cross-coupling reaction of the 4-chloropyrrolo[1,2-a]quinoxaline 6a with N,N-diethylaminoethylthiol in presence of 10 mol % CuI, 10 mol % neocuproine, and t-BuONa as the base, in toluene at 110°C 33-35. Using this protocol, the arylsulfide 7a was synthesised (yield: 80%). The reaction of 6a with the thiolate anion, prepared by the action of t-BuOK on the N,N-diethylaminoethylthiol, in the presence of a catalytic amount of tetrakis(triphenylphosphine)palladium in n-butanol (method D) was also found to be useful in the preparation of the sulfide 7a (66%) [36,37].

Table I.  Optimized reaction conditions for coupling of 6a and N,N-diethylaminoethylthiol.

Method	Base	Catalyst	Ligand	Reaction conditions	% Yield of 7a
A	NaH	–	–	dioxane, 100°C, 4 h	42
B	K2CO3	CuI	HOCH2CH2OH	2-propanol, 80°C, 24 h	60
C	t-BuONa	CuI	Neocuproine	toluene, reflux, 24 h	80
D	t-BuOK	Pd(PPh3)4	–	n-BuOH, reflux, 5 h	66

A second investigated route to 7f-l first involved the treatment of aminophenylpyrroles 4a-c with CS₂ in alcoholic alkali to afford the expected pyrrolo[1,2-a]quinoxaline-4-thiols 8a-c [26]. It was then followed by the reaction of 8a-c with aminoalkyl halides to give the S-alkylated products 7f-l. All alkylthiopyrrolo[1,2-a]quinoxalines 7a-l were converted into their hydrochlorides 1a-l by treatment with hydrochloric acid gas in diethyl ether.

A similar nucleophilic substitution with the N,N-diethylaminoethylthiol and 5-chloropyrrolo[1,2-a]thieno[3,2-e]pyrazine 9 [38] furnished the pyrrolo[1,2-a]thieno[3,2-e]pyrazine 10, converted in its hydrochloride 1 m (Scheme 2).

Scheme 2 Synthesis of compound 1m. Reagents: (i) Method A: NaH, HS-CH₂-CH₂-N(C₂H₅)₂, dioxane; (ii) HCl, Et₂O.

Pharmacology

In this study, a series of twelve new pyrrolo[1,2-a]quinoxaline derivatives 1a-l and one pyrrolo[1,2-a]thieno[3,2-e]pyrazine 1m were synthesised from various commercially nitroanilines in order to obtain new compounds endowed with a higher bacterial EPI activity. These target derivatives were evaluated in a model targeting a typical MFS transporter of S. aureus, involved in the fluoroquinolone resistance of this bacterial species [5,14,17,39]. This model used the S. aureus strains SA 1199B and SA-1, which overexpress the NorA efflux pump. In preliminary experiments, the MIC of the reference and putative EPIs were determined. Reserpine and omeprazole were found to lack any intrinsic antibacterial activity (MIC > 512 μg/mL), but the derivatives 1a-m exhibited various MICs, ranging between 16 μg/mL (compound 1e) and >512 μg/mL (compounds 1l-m) (data not shown). All tested EPIs were subsequently used at concentrations devoid of bacterial toxicity (25% of their MIC) in order to observe the most important EPI effect. Then, EPI activity of all newly synthesised compounds was evaluated by the high throughput disk diffusion method and by MIC determination. All results were in agreement with the literature [5,10,15,39] and are reported in Table II.

Table II.  EPI activities of reserpine, omeprazole, 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxaline derivatives 1a-l and pyrrolo[1,2-a]thieno[3,2-e]pyrazine 1m on the S. aureus strains, SA 1199B and SA-1.

		SA 1199B			SA-1
		% inhibition^b		MIC(μg/mL)^c	% inhibition^b		MIC(μg/mL)^c
EPI	Concentration^a (μg/mL)	Nor^d	Cip^d	Nor^d	Nor^d	Cip^d	Nor^d
–	–	–	–	64	–	–	16
Reserpine	128	33	69	8	100	100	2
Omeprazole	128	38	69	8	100	100	2
1a	64	28	62	16	79	83	4
1b	64	33	62	8	86	100	2
1c	16	0	15	64	21	17	16
1d	32	28	62	16	71	83	2
1e	4	28	62	16	71	75	2
1f	64	43	77	8	86	92	1
1 g	16	52	85	4	100	100	1
1 h	16	14	38	16	57	67	4
1i	32	14	46	16	50	58	4
1j	16	0	23	32	43	50	8
1k	16	0	31	32	43	50	8
1 l	128	0	23	64	29	25	16
1m	64	48	85	4	100	100	1

^aIn all experiments, the EPIs were used at 0.25 × MIC_(EPI).

^b% inhibition were calculated with the following equation: % = [Ø_R(AB + EPI) - Ø_R(AB)] / [Ø_S(AB) - Ø_R(AB)] × 100 Ø = diameter of the inhibition zone, R = Resistant strain, S = fluoroquinolone susceptible strain, AB = antibiotic, +EPI = in presence of EPI [40].

^cMIC is the minimum inhibitory concentration of norfloxacin in absence or in presence of EPI.

^dNor: norfloxacin, Cip: ciprofloxacin.

The antibiotic resistance and inhibition profiles of SA 1199B and SA-1 reflected the overexpression of NorA efflux pump, which preferentially exports the hydrophilic fluoroquinolones [14,39]. Biological results showed that SA 1199B exhibited a high-level fluoroquinolone resistance that was not completely reversed by the EPIs, as previously described [10]. In fact, SA 1199B, commonly used to evaluate NorA expression and inhibition, is a double target (A116E in GrlA)/efflux mutant [15] and a lesser affinity of the drug for its modified target probably leads to a diminished entry by passive diffusion into the bacterial cell, that cannot be amended by EPIs. In contrast, the fluoroquinolone susceptibility of the single NorA mutant, SA-1, was totally restored by reserpine and omeprazole [27] (Table II). Thus, SA-1 is probably a better test strain for investigating NorA efflux pump inhibition. In a first step, the agar diffusion method showed that the percentage of NorA inhibition by reserpine, omeprazole and the newly synthesised compounds reached 29–100% for norfloxacin and 25–100% for ciprofloxacin, in SA-1. Two new derivatives, 1 g and 1m, appeared to be more active than the reference EPIs. In a second step, MIC determination of norfloxacin confirmed these results. Indeed, the addition of 1 g and 1m resulted in a similar or higher decrease of the norfloxacin MIC (32-fold for SA-1; 16-fold for SA-1199B) than the combination with omeprazole and reserpine (32-fold for SA-1; 8-fold for SA-1199B).

Consequently, all of these preliminary biological results allowed clarification of the influence of the substituents at positions 7 and/or 8 of the pyrrolo[1,2-a]quinoxaline nucleus and of the 4-alkylamino-ethylthio chain on the EPI activity. Indeed, with regard to the unsubstituted 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxalines 1a, 1f and 1i, their 7-methoxy analogues 1b, 1 g and 1j were more active, whereas the chlorosubstituted compounds 1c-d, 1 h and 1k, and the 8-phenyl substituted compound 1e provided the lowest EPI effects. Moreover, the EPI activity was significantly affected by replacement of the N,N-diethylamino group by bioisostere groups, such as pyrrolidine or piperidine in the di(alkylamino)ethylthio chain. Indeed, the substitution of the diethylamino group by its isomer with restricted conformation such as pyrrolidine moiety enhanced the EPI activity, i.e. compound 1a versus 1f, and 1b versus 1 g. Replacement by a piperidine ring leading to a greater flexibility in the saturated heterocycle (compounds 1i-k) decreased the EPI efficiency. However, the introduction of a n-butylthio function at position 4 of the pyrrolo[1,2-a]quinoxaline moiety in compound 1 l was not found beneficial, suggesting the importance of a potential protonable nitrogen atom in the alkyl chain. Replacement of the pyrroloquinoxaline moiety by a pyrrolothienopyrazine moiety (compound 1m), bioisostere of the pyrroloquinoxaline nucleus, showed the most efficient EPI activity, suggesting that an electron-rich (or electronegative) atom, such as a sulfur atom, potentialized this activity. Finally, four pyrroloquinoxaline derivatives 1b, 1f-g and 1m were found more efficient than reserpine, taken as the reference EPI.

Conclusion

In the present paper, we have described the synthesis and the efflux pump inhibition activity of twelve new 4-[2-(alkylamino)ethylthio]pyrrolo[1,2-a]quinoxalines 1a-l and one pyrrolo[1,2-a]thieno[3,2-e]pyrazine compound 1m. Through this study, it was observed that few pyrroloquinoxaline derivatives 1 showed a higher EPI activity than reserpine or omeprazole. Substitution of the pyrroloquinoxaline nucleus by a methoxy group increased the EPI activity, whereas introduction of a chlorine atom decreased it. The best results were observed with a pyrrolidine moiety fixed to the 4-substituted thioethyl chain. Based on these preliminary structure-activity results, it could be possible to further identify new pyrrolo[1,2-a]quinoxaline or pyrrolo[1,2-a]thieno[3,2-e]pyrazine derivatives developed as EPIs on NorA efflux pump. Combinations of all pyrrolo[1,2-a]quinoxaline SAR studies might allowed to identify pharmacophore moiety and potential inhibition mechanism in order to obtain therapeutically useful derivatives.