Heterocycles [h]-Fused Onto 4-Oxoquinoline-3-Carboxylic Acid, Part VIII [1]. Convenient Synthesis and Antimicrobial Properties of Substituted Hexahydro[1,4]diazepino[2,3-h]quinoline-9-carboxylic acid and Its Tetrahydroquino[7,8-b]benzodiazepine Analog

[1,4]Diazepino[2,3-h]quinolone carboxylic acid 3 and its benzo-homolog tetrahydroquino[7,8-b]benzodiazepine-3-carboxylic acid 5 were prepared via PPA-catalyzed thermal lactamization of the respective 8-amino-7-substituted-1,4-dihydro-quinoline-3-carboxylic acid derivatives 8, 10. The latter compounds were obtained by reduction of their 8-nitro-7-substituted-1,4-dihydroquinoline-3-carboxylic acid precursors 7, 9 which, in turn, were prepared by reaction of 7-chloro-1-cyclopropyl-6-fluoro-8-nitro-1,4-dihydroquinoline-3-carboxylic acid (6) with each of β-alanine and anthranilic acid. All intermediates and target compounds were characterized using elemental analysis, NMR, IR and MS spectral data. The prepared targets and the intermediates have shown interesting antibacterial activity mainly against Gram positive strains. In particular, compound 8 showed good activity against S. aureus (MIC = 0.39 µg/mL) and B. subtilis (MIC = 0.78 µg/mL). Compounds 5a and 9 have also displayed good antifungal activity against C. albicans (MIC = 1.56 µg/mL and 0.78 µg/mL, respectively). None of the compounds tested showed any anticancer activity against solid breast cancer cell line MCF-7 cells or a human breast adenocarcinoma cell line.

Owing to the potential biological interest in these heterocyclic compounds, the present research addresses the synthesis and characterization of new heterocyclic system incorporating 4-oxopyridine nucleus condensed either to 1,5-benzodiazpinone to form the target compound 3 (Figure 1, Scheme 1) or to the analogous dibenzo[b,e] [1,4]diazepinone to form compound 5a ( Figure 2, Scheme 2). Such hybrid tri-and tetracyclic systems (3, 5a,b) might exhibit interesting bio-properties such as antimicrobial and/or antitumor activity.
The identification of the prepared intermediates and target compounds was based on elemental analysis, IR, MS, 1 H-and 13 C-NMR spectral data, given in the Experimental. These spectral data were all consistent with the proposed structures. Signal assignments to the various proton and carbons were mostly determined following DEPT and 2D (COSY, HMQC and HMBC) experiments. It was clearly apparent that H-7 in 3 and H-5 in 5, 7-10, which resonate at around 8.0 ppm (d, 3 J H-F ≈ 13 Hz), showed consistent splitting patterns in all compounds due to coupling with the vicinal fluorine atom. It was revealed from the new broad signal at around 7 to 8 ppm, assigned for the NH at C-7, that the primary amine constituent was introduced in 7, 9. The same proton was down field shifted upon reduction of these compounds indicating the formation of the 8-amino derivative 8, 10. In case of the target compounds 3, 5a,b a singlet peak for the amide -NH was observed at around 10 ppm indicating that lactamization has taken place. For compound 3, long-range correlations are observed between H-10 and each of C-8, C-11a and CO 2 H. Corresponding long-range correlations are also observed between H-7 and its neighbor carbons C-8, C-11a and C-5a. Similar pattern of long-range correlations were observed for 5a,b. The skeletal carbons of the fused benzene ring (B) are recognizable by their signal splitting arising from coupling with fluorine atom (different value of J for each carbon) and from longrange coupling with neighboring protons.

Antimicrobial activity
The in vitro antibacterial activity of all intermediates and targeted products was evaluated against an assortment of Gram positive and Gram negative bacterial strains using the minimum inhibitory concentration (MIC) approach. The prepared targets and the intermediates have shown interesting antibacterial activity mainly against Gram positive strains (Table 1), while none have shown any activity against Gram negative bacteria. The activity ranged from week to strong against both S. aureus (with MIC range 12.5-0.39 µg/mL) and B. subtilis (with MIC range 6.25-0.78 µg/mL). In particular, compound 8 showed good activity against S. aureus (with MIC 0.39 µg/mL) and B. subtilis (with MIC 0.78 µg/mL). It is generally assumed that the more lipophilic quinolones can penetrate better the lipophilic cell membrane of Gram positive bacteria, while less lipophilic compounds are more liable to penetrate the cell wall of Gram negative bacteria [52][53]. The activities of the target compounds (3 and 5) and intermediates prepared in this work (7)(8)(9)(10) are in correlation with this theory since they are lipophilic. On the other hand, the anthranilic acid derivatives 5a and 9 have also displayed excellent antifungal activity against Candida albicans with MIC values of 1.56 µg/mL and 0.78 µg/mL, respectively. Table 1. MICs (µg/mL) for compounds 3, 5 and 7-10 against Gram positive bacterial strains and Candida albicans.

Compound
No.

Cytotoxicity towards cancerous epithelial cells
Preliminary cytotoxicity studies were carried out for four candidate compounds (3, 8, 5a, 5b) with MCF-7 cells, a human breast adenocarcinoma cell line, to test whether these compounds are toxic to epithelial cells, or they would have a potential as anticancerous agents. Cells were trypsinized, seeded in 96 well plates and incubated for 24 h. The substances were first dissolved in DMSO and then diluted with RPMI 1640 cell culture media, added to the cells, and incubated at a concentration range of 0.001 to 1.0 µg/mL. The cells were incubated with the compounds for 48 h, and sulphrodamine B assay was run afterwards. All tests were performed in triplicates and repeated twice using two different passages. All compounds did not change the proliferation rate of the cells as compared to controls (cells incubated with media only, with the same ratio of DMSO). This would suggest that these compounds are not toxic to epithelial cells. Further evaluation should be considered for exact determination of the IC 50 .
High resolution mass spectra (HRMS) were measured in negative ion mode by electrospray ionization (ESI) technique on a Bruker APEX-2 instrument. The samples were dissolved in acetone, diluted in spray solution (methanol + water + ammonia, in the ratio 1:1:1, v/v/v) and infused using a syringe pump with a flow rate of 2 mm 3 /min. External calibration was conducted using arginine cluster in a mass range m/z = 175-871.

benzodiazepine-3-carboxylic acid (5b)
A stirred solution of compound 10 (0.2 g, 0.5 mmol) and conc. sulphuric acid (8 mL) was heated under reflux conditions (100 °C) for 3 h. The resulting mixture was poured onto water (60 mL) with vigorous stirring. The precipitated solid product was collected by suction filtration, washed with water (2 x 10 mL) and dried to furnish green-yellowish solid product.  Bacterial suspensions were prepared in sterilized distilled water, in a concentration around 1x10 7 cfu/mL, which was standardized according to 0.5 McFarland suspension as described by the Clinical and Laboratories Standards Institute (CLSI) 2007. The minimum inhibitory concentrations (MICs, µg/mL) of test compounds were determined by broth dilution method, screening different concentrations in the range 100-0.097 µg/mL. The MIC is defined as the lowest concentration of the tested compound showing no growth. A stock solution of each tested compound was prepared in DMSO (100 µg/mL). The MIC test was performed in 96 flat bottom microtiter plates, 100 µL of previously prepared and sterilized broth (prepared by dissolving 1.3 g of dry preparation in 100 mL of distilled water) was added in each well, with an exception to the first well where 100 µL of double strength, sterilized broth was added (prepared by dissolving 1.3 g of dry preparation in 50 mL of distilled water) in order to maintain the consistency of the broth along the plate after the addition of the tested compound. An equivalent volume of 50 µg/mL of each compound was added to the first well, mixed with the broth, followed by two fold serial dilution onto successive wells across the plate to end up with 11 successive two fold dilutions for each of the tested compounds. Then 10 µL of bacterial suspension was used to inoculate each well. Control tests for each experiment were performed. Positive growth control was performed by adding one drop of each microorganism suspensions to four wells in each plate of the culture medium without the test compound. Negative growth control was also performed using four un-inoculated wells of medium without the test compound. Plates were incubated at 37 °C for 24 h, and were checked for turbidity. Two fold serial dilutions were carried out in a similar manner for DMSO (20% v/v in water) to test its antibacterial activity. Ciprofloxacin standard was tested also as reference compound. The turbidity was determined visually and using microplate reader.