Synthesis of New Fused Heterocyclic 2-Quinolones and 3-Alkanonyl-4-Hydroxy-2-Quinolones

Herein, we report the synthesis of 5,12-dihydropyrazino[2,3-c:5,6-c′]difuro[2,3-c:4,5-c′]-diquinoline-6,14(5H,12H)diones, 2-(4-hydroxy-2-oxo-1,2-dihydroquinolin-3-yl)-1,4-diphenyl- butane-1,4-diones and 4-(benzo-[d]oxazol-2-yl)-3-hydroxy-1H-[4,5]oxazolo[3,2-a]pyridine-1-one. The new candidates were synthesized and identified by different spectroscopic techniques, and X-ray crystallography.


Instruction
Furoquinolones are an interesting class of 2-quinolones and representative examples of them as naturally occurring compounds are shown in compounds such as Skimmianine and γ-Fagarine (Figure 1), both compounds that have been found to have anti-cancer activity [1,2]. It was reported that furopyrazine scaffold was functionalized with an amino-and a carboxy-terminus resulting in a conformationally restricted dipeptidomimetic scaffold [3].
Alkyl quinolones (AQs) are a species-specific class of quorum-sensing molecule that have been described in P. aeruginosa [4,5] and related bacteria including P. putida and Burkholderia spp. [6]. More than 55 distinct AQs (i.e., an example is shown in Figure 1 and assigned as PQS) are produced through the PqsABCDE (Figure 1) biosynthetic pathway in P. aeruginosa, with the majority of the diversity arising from unsaturation, different alkyl chain lengths, and modification of the ring-substituted nitrogen [6,7]. An insight into the evolutionary basis of AQ diversity has emerged from Burkholderia thailandensis where AQ analogues (i.e., two examples assigned as HHQ and HQNP and are shown in Figure 1) are shown to act synergistically to inhibit bacterial growth [8,9].
The aforementioned interesting pharmaceutical and biological activities of 4-hydroxy-2-quinolones make them valuable in drug research and development. Hence, many publications have recently dealt with their synthetic analogous and the synthesis of their heteroannelated derivatives. Consequently, we have found that it is of importance to shed new light on these interesting heterocycles. Accordingly, the reactivity of 1,6-disubstituted-4-hydroxy-quinolinones 1a-f was tested towards 2,3-dichloropyrazine (2) and (E)-dibenzoylethene (4).

Results and Discussion
Upon mixing equivalent amounts of 2,3-dichloropyrazine (2) and 6,7-disubstituted-4-hydroxyquinolin-2-ones 1a-f, followed by refluxing in dimethylformamide (DMF) and catalyzed by triethylamine (Et3N), 3a-f were obtained as single products (Scheme 1). Structure elucidation of compounds 3a-f was carried out by infrared (IR), 1 H-nuclear magnetic resonance (NMR), 13 C-NMR and mass spectrometry, as well as elemental analyses. The reaction products were identified as pyrazino[2′,3′:4,5]furo[3,2-c]quinolin-6(5H)-ones 3a-f. The IR spectra showed absorption for C=N at ν = 1630-1600 cm −1 . Besides the NH stretching appeared as broad peaks at ν = 3320-3260 cm −1 . As for example, the elemental analysis and mass The reaction products were identified as pyrazino [2 ,3 :4,5]furo[3,2-c]quinolin-6(5H)-ones 3a-f. The IR spectra showed absorption for C=N at ν = 1630-1600 cm −1 . Besides the NH stretching appeared as broad peaks at ν = 3320-3260 cm −1 . As for example, the elemental analysis and mass spectra proved its molecular formula of 3a as C 22 H 10 N 4 O 4, which indicated addition of two moles of 1a to two moles of 2 accompanied with elimination of four moles of HCl. The expected structure 7 was ruled out, since 1 H-NMR spectrum did not show the expected azomethine protons ( Figure 3). Either the syn-structure spectra proved its molecular formula of 3a as C22H10N4O4, which indicated addition of two moles of 1a to two moles of 2 accompanied with elimination of four moles of HCl. The expected structure 7 was ruled out, since 1 H-NMR spectrum did not show the expected azomethine protons ( Figure 3). Either the syn-structure 3a′ or the anti-form has to show symmetric carbon signals in 13 C-NMR spectrum. Most assigned carbons are the two carbonyl and C=N carbons which appeared at δ = 167.1, 165.2 and 160.8, 160.2 ppm. In Spartan 18: geometries program [21] optimized at the 6-31G* level with B3LYP the energy difference between calculated Anti-3a: and Syn-3a' shows that Anti-3a is more stable by 2.029 kcal/mol. Compounds 3 are the products of reaction between one molecule of 2,3-dichloropyrazpine (2) and two molecules of a 4-hydroxy-2-quinolinone (1); substituents on 3 correspond to those on 1 in the obvious way. The reaction results in replacement of both chlorines and both hydrogens of 2, by either an α-carbon or a pseudo-phenolic oxygen of 1. Both positions involved on 1 are formally nucleophilic, although the α-carbon normally reacts first (cf. reaction of 1 with 4 to give 6). We are unaware of any direct reaction of 2 with four nucleophiles; in systems where the chlorines and hydrogens of 2 are all replaced, the reaction of C-5 and C-6 (the hydrogen-bearing carbons) involves reaction with an oxidant (e.g., a molecular halogen), sometimes followed by an organometallic coupling [22,23]. On the other hand, pyrazines bearing leaving groups readily undergo displacement of those leaving groups, by either SNAr or Addition of the Nucleophile, Ring Opening, and Ring Closure (ANRORC) mechanisms [24][25][26]. If the pyrazine is activated by an electrophile, the nucleophile can be as weak as o-phenylenediamine. [27]. If 2 does not undergo four-fold nucleophilic substitution, the likeliest scenario would seem to be two-fold displacement followed by two-fold oxidative cyclization, presumably by air. If the chlorides are the leaving groups, only one can undergo ipso substitution: the other nucleophile must attack the other side of the ring. The SNAr process can give ipso substitution only, but the ANRORC process can proceed at the pseudo-meta position (Scheme 2). We are therefore led to propose the mechanism for formation of 3, shown in Scheme 3. Here "Nu: -" is 1, attacking via its α-carbon. If one substitution gave ipso displacement and the other gave pseudo-meta displacement, one would expect the observed anti regiochemistry: the order of the two substitutions would not matter.  Compounds 3 are the products of reaction between one molecule of 2,3-dichloropyrazpine (2) and two molecules of a 4-hydroxy-2-quinolinone (1); substituents on 3 correspond to those on 1 in the obvious way. The reaction results in replacement of both chlorines and both hydrogens of 2, by either an α-carbon or a pseudo-phenolic oxygen of 1. Both positions involved on 1 are formally nucleophilic, although the α-carbon normally reacts first (cf. reaction of 1 with 4 to give 6). We are unaware of any direct reaction of 2 with four nucleophiles; in systems where the chlorines and hydrogens of 2 are all replaced, the reaction of C-5 and C-6 (the hydrogen-bearing carbons) involves reaction with an oxidant (e.g., a molecular halogen), sometimes followed by an organometallic coupling [22,23]. On the other hand, pyrazines bearing leaving groups readily undergo displacement of those leaving groups, by either S N Ar or Addition of the Nucleophile, Ring Opening, and Ring Closure (ANRORC) mechanisms [24][25][26]. If the pyrazine is activated by an electrophile, the nucleophile can be as weak as o-phenylenediamine [27]. If 2 does not undergo four-fold nucleophilic substitution, the likeliest scenario would seem to be two-fold displacement followed by two-fold oxidative cyclization, presumably by air. If the chlorides are the leaving groups, only one can undergo ipso substitution: the other nucleophile must attack the other side of the ring. The S N Ar process can give ipso substitution only, but the ANRORC process can proceed at the pseudo-meta position (Scheme 2). We are therefore led to propose the mechanism for formation of 3, shown in Scheme 3. Here "Nu: -" is 1, attacking via its α-carbon. If one substitution gave ipso displacement and the other gave pseudo-meta displacement, one would expect the observed anti regiochemistry: the order of the two substitutions would not matter. spectra proved its molecular formula of 3a as C22H10N4O4, which indicated addition of two moles of 1a to two moles of 2 accompanied with elimination of four moles of HCl. The expected structure 7 was ruled out, since 1 H-NMR spectrum did not show the expected azomethine protons (Figure 3). Either the syn-structure 3a′ or the anti-form has to show symmetric carbon signals in 13 C-NMR spectrum. Most assigned carbons are the two carbonyl and C=N carbons which appeared at δ = 167.1, 165.2 and 160.8, 160.2 ppm. In Spartan 18: geometries program [21] optimized at the 6-31G* level with B3LYP the energy difference between calculated Anti-3a: and Syn-3a' shows that Anti-3a is more stable by 2.029 kcal/mol. Compounds 3 are the products of reaction between one molecule of 2,3-dichloropyrazpine (2) and two molecules of a 4-hydroxy-2-quinolinone (1); substituents on 3 correspond to those on 1 in the obvious way. The reaction results in replacement of both chlorines and both hydrogens of 2, by either an α-carbon or a pseudo-phenolic oxygen of 1. Both positions involved on 1 are formally nucleophilic, although the α-carbon normally reacts first (cf. reaction of 1 with 4 to give 6). We are unaware of any direct reaction of 2 with four nucleophiles; in systems where the chlorines and hydrogens of 2 are all replaced, the reaction of C-5 and C-6 (the hydrogen-bearing carbons) involves reaction with an oxidant (e.g., a molecular halogen), sometimes followed by an organometallic coupling [22,23]. On the other hand, pyrazines bearing leaving groups readily undergo displacement of those leaving groups, by either SNAr or Addition of the Nucleophile, Ring Opening, and Ring Closure (ANRORC) mechanisms [24][25][26]. If the pyrazine is activated by an electrophile, the nucleophile can be as weak as o-phenylenediamine. [27]. If 2 does not undergo four-fold nucleophilic substitution, the likeliest scenario would seem to be two-fold displacement followed by two-fold oxidative cyclization, presumably by air. If the chlorides are the leaving groups, only one can undergo ipso substitution: the other nucleophile must attack the other side of the ring. The SNAr process can give ipso substitution only, but the ANRORC process can proceed at the pseudo-meta position (Scheme 2). We are therefore led to propose the mechanism for formation of 3, shown in Scheme 3. Here "Nu: -" is 1, attacking via its α-carbon. If one substitution gave ipso displacement and the other gave pseudo-meta displacement, one would expect the observed anti regiochemistry: the order of the two substitutions would not matter. Afterward, we investigated the reactions of (E)-dibenzoylethene (4) and 1a-f under the same conditions mentioned above in pyridine/Et3N (Scheme 1). The structure elucidation depends intensively on NMR spectra. For example, the 1 H spectrum of 6f consists of a 1H singlet at δH = 11.35 and a broad signal at δH = 10.77; in the aromatic region, two sets of resonances from monosubstituted phenyl rings and a three-spin system from the quinoline; and a three-spin system and a 3H methyl singlet upfield. The integrals require that there be two phenyl rings and one quinolone and, consequently, they rule out the alternative structures 8f′ and 8f′′ (Figure 4). The NMR correlations of 6f are detailed in full in Table S1 (see Supplementary Material). The structure of compound 6f was finally proved by X-ray structure analysis as shown in Figure 5.  Surprisingly, on attempting to prepare 4-hydroxy-2-quinolone 1g from 2-hydroxyaniline (9) and diethyl malonate (10) in polyphosphoric acid (PPA) according to the procedure described in Afterward, we investigated the reactions of (E)-dibenzoylethene (4) and 1a-f under the same conditions mentioned above in pyridine/Et 3 N (Scheme 1). The structure elucidation depends intensively on NMR spectra. For example, the 1 H spectrum of 6f consists of a 1H singlet at δ H = 11.35 and a broad signal at δ H = 10.77; in the aromatic region, two sets of resonances from monosubstituted phenyl rings and a three-spin system from the quinoline; and a three-spin system and a 3H methyl singlet upfield. The integrals require that there be two phenyl rings and one quinolone and, consequently, they rule out the alternative structures 8f and 8f (Figure 4). The NMR correlations of 6f are detailed in full in Table S1 (see Supplementary Material). The structure of compound 6f was finally proved by X-ray structure analysis as shown in Figure 5. Afterward, we investigated the reactions of (E)-dibenzoylethene (4) and 1a-f under the same conditions mentioned above in pyridine/Et3N (Scheme 1). The structure elucidation depends intensively on NMR spectra. For example, the 1 H spectrum of 6f consists of a 1H singlet at δH = 11.35 and a broad signal at δH = 10.77; in the aromatic region, two sets of resonances from monosubstituted phenyl rings and a three-spin system from the quinoline; and a three-spin system and a 3H methyl singlet upfield. The integrals require that there be two phenyl rings and one quinolone and, consequently, they rule out the alternative structures 8f′ and 8f′′ (Figure 4). The NMR correlations of 6f are detailed in full in Table S1 (see Supplementary Material). The structure of compound 6f was finally proved by X-ray structure analysis as shown in Figure 5.  Surprisingly, on attempting to prepare 4-hydroxy-2-quinolone 1g from 2-hydroxyaniline (9) and diethyl malonate (10) in polyphosphoric acid (PPA) according to the procedure described in Afterward, we investigated the reactions of (E)-dibenzoylethene (4) and 1a-f under the same conditions mentioned above in pyridine/Et3N (Scheme 1). The structure elucidation depends intensively on NMR spectra. For example, the 1 H spectrum of 6f consists of a 1H singlet at δH = 11.35 and a broad signal at δH = 10.77; in the aromatic region, two sets of resonances from monosubstituted phenyl rings and a three-spin system from the quinoline; and a three-spin system and a 3H methyl singlet upfield. The integrals require that there be two phenyl rings and one quinolone and, consequently, they rule out the alternative structures 8f′ and 8f′′ (Figure 4). The NMR correlations of 6f are detailed in full in Table S1 (see Supplementary Material). The structure of compound 6f was finally proved by X-ray structure analysis as shown in Figure 5.  Surprisingly, on attempting to prepare 4-hydroxy-2-quinolone 1g from 2-hydroxyaniline (9) and diethyl malonate (10) in polyphosphoric acid (PPA) according to the procedure described in Surprisingly, on attempting to prepare 4-hydroxy-2-quinolone 1g from 2-hydroxyaniline (9) and diethyl malonate (10) in polyphosphoric acid (PPA) according to the procedure described in reference [28], compound 11 was obtained in 80% yield (Scheme 4). Similarly, reaction of 11 with 4 under the same conditions produced compound 12 in 85% yield. The NMR spectroscopic data of compound 12 (detailed in full in Table S2;  gives Heteronuclear Multiple Bond Correlation (HMBC) correlation with a carbon at δ = 156.92 ppm, which could be either C-1 or C-3, and is assigned based on chemical shift as C-1. Correlations within the benzoyl groups are straightforward. The structure of compound 12 was confirmed by X-ray crystallography ( Figure 6). The NMR spectroscopic data of compound 12 (detailed in full in Table S2; see Supplementary Material) show a broad OH, two sets of phenyl signals, and eight other proton signals, although most of the benzoxazole signals cannot be solved fully. The protons on sp 3 carbons are distinctive at δ = 5.55, 4.36, and 2.85 ppm; the latter two are assigned as H-2b based on their geminal coupling constant of 17 Hz. The attached carbons appear at δ = 39.5 (C-2a) and 37.35 ppm (C-2b). The C-2a gives Heteronuclear Multiple Bond Correlation (HMBC) correlation with a carbon at δ = 156.92 ppm, which could be either C-1 or C-3, and is assigned based on chemical shift as C-1. Correlations within the benzoyl groups are straightforward. The structure of compound 12 was confirmed by X-ray crystallography ( Figure 6). The NMR spectroscopic data of compound 12 (detailed in full in Table S2; see Supplementary Materials) show a broad OH, two sets of phenyl signals, and eight other proton signals, although most of the benzoxazole signals cannot be solved fully. The protons on sp 3 carbons are distinctive at δ = 5.55, 4.36, and 2.85 ppm; the latter two are assigned as H-2b based on their geminal coupling constant of 17 Hz. The attached carbons appear at δ = 39.5 (C-2a) and 37.35 ppm (C-2b). The C-2a gives Heteronuclear Multiple Bond Correlation (HMBC) correlation with a carbon at δ = 156.92 ppm, which could be either C-1 or C-3, and is assigned based on chemical shift as C-1. Correlations within the benzoyl groups are straightforward. The structure of compound 12 was confirmed by X-ray crystallography ( Figure 6). 7 Figure 6. Molecular structure of compound 12 assigned as 2-(4-hydroxy-6-methyl-2-oxo-1,2-dihydroquinolin-3-yl)-1,4-diphenylbutane-1,4-dione.

Material and Methods
Melting points were taken in open capillaries on a Gallenkamp melting point apparatus (Weiss-Gallenkamp, Loughborough, UK) and are uncorrected. The IR spectra were recorded from potassium bromide disks with a Fourier Transform Infrared (FT-IR) device (Mettler-Toledo GmbH, Giessen, Germany), Minia University. Elemental analyses were carried out at the Perkin-Elmer out with Elementar 306 (Perkin-Elmer, Walluf, Germany). NMR data were recorded on Bruker AM 400 or AV400 spectrometers (Bruker, Karlsruhe, Germany), at 400 MHz for 1 H and 100 MHz for 13 C. Chemical shifts were reported in ppm from tetramethylsilane using solvent resonance in DMSO-d6 solutions as the internal standard. Coupling constants are stated in Hz. Correlations were established using 1 H-1 H COSY, and 1 H-13 C and 1 H-15 N heteronuclear single quantum coherence (HSQC) and HMBC experiments. Mass spectra were recorded on a Finnigan MAT 312 instrument Fab 70 eV (Thermo Fisher, Bremen, Germany), Institute of Organic Chemistry, Karlsruhe University, Karlsruhe, Germany. Thin Layer Chromatography (TLC) was performed on analytical Merck 9385 silica aluminum sheets (Kieselgel 60) with Pf254 indicator; TLCs were viewed at λ max = 254 nm. Elemental analyses for C, H, N were carried out with Elementar 306.

Material and Methods
Melting points were taken in open capillaries on a Gallenkamp melting point apparatus (Weiss-Gallenkamp, Loughborough, UK) and are uncorrected. The IR spectra were recorded from potassium bromide disks with a Fourier Transform Infrared (FT-IR) device (Mettler-Toledo GmbH, Giessen, Germany), Minia University. Elemental analyses were carried out at the Perkin-Elmer out with Elementar 306 (Perkin-Elmer, Walluf, Germany). NMR data were recorded on Bruker AM 400 or AV400 spectrometers (Bruker, Karlsruhe, Germany), at 400 MHz for 1 H and 100 MHz for 13 C. Chemical shifts were reported in ppm from tetramethylsilane using solvent resonance in DMSO-d 6 solutions as the internal standard. Coupling constants are stated in Hz. Correlations were established using 1 H-1 H COSY, and 1 H- 13

Reaction of 1a-f with 2,3-Dichloropyrazine (2)
A suspension of 1,6-disubstituted quinoline-2,4-(1H,3H)-diones 1a-f (2 mmol) in 10 mL dimethylformamide (DMF) was added to a solution of (E)-1,4-diphenylbut-2-ene-1,4-dione (2, 0.148 g, 1 mmol) in 15 mL DMF and 0.5 mL of triethylamine. The reaction mixture was gently refluxed for 20-25 h, until the reactants had disappeared (monitored by TLC). The resulting precipitates of 3a-f which were obtained cold were filtered off and dried. The precipitates were recrystallized from the stated solvents. A mixture of 1a-f (1 mmol) and (E)-1,2-dibenzoylethene (4) (0.246 g, 1 mmol) in pyridine (50 mL) and 0.5 mL of triethylamine, was gently refluxed for 10-15 h, until the reactants had disappeared (monitored by TLC). The resulting precipitates of 6a-f, which obtained on cold were filtered off and dried. The precipitates were recrystallized from the stated solvents. Funding: The authors thank Egyptian Mission, Ministry of higher education, Egypt for its financial support to Mrs Lamiaa E. Abd El-Haleem during her accommodation in Institute für Technology, Karlsruhe, Germany. The authors also thank the DFG Foundation for providing Prof Ashraf A. Aly, one-month fellowship enabling him to carry out the compounds analysis in Karlsruhe Institute of Technology, Karlsruhe, Germany in July-August 2019.