Hydrogen Bond Assisted Three-Component Tandem Reactions to Access N-Alkyl-4-Quinolones

Hydrogen-bonding catalytic reactions have gained great interest. Herein, a hydrogen-bond-assisted three-component tandem reaction for the efficient synthesis of N-alkyl-4-quinolones is described. This novel strategy features the first proof of polyphosphate ester (PPE) as a dual hydrogen-bonding catalyst and the use of readily available starting materials for the preparation of N-alkyl-4-quinolones. The method provides a diversity of N-alkyl-4-quinolones in moderate to good yields. The compound 4h demonstrated good neuroprotective activity against N-methyl-ᴅ-aspartate (NMDA)-induced excitotoxicity in PC12 cells.

Investigations of CPA in the field of hydrogen-bonding catalytic chemical reactions made clear that the P = O and O-H serve as hydrogen bond acceptor and donor, respectively [49][50][51][52][53][54][55][56][57]. Previous reports show that polyphosphate ester (PPE) could activate the nitrogen functional group toward nucleophilic attack, and serve as alkylating reagents [58]. Recently, our group reported polyphosphoric acid (PPA) promoted tandem reactions for the synthesis of heterocycles, which revealed that PPA is an effective condensation reagent [59,60]. On the basis of the previous work, we envisioned that N-alkyl-4-quinolones could be synthesized via the three-component tandem reactions of Investigations of CPA in the field of hydrogen-bonding catalytic chemical reactions made clear that the P=O and O-H serve as hydrogen bond acceptor and donor, respectively [49][50][51][52][53][54][55][56][57]. Previous reports show that polyphosphate ester (PPE) could activate the nitrogen functional group toward nucleophilic attack, and serve as alkylating reagents [58]. Recently, our group reported polyphosphoric acid (PPA) promoted tandem reactions for the synthesis of heterocycles, which revealed that PPA is an effective condensation reagent [59,60]. On the basis of the previous work, we envisioned that N-alkyl-4-quinolones could be synthesized via the three-component tandem reactions of readily accessible 2aminoacetophenones, aldehydes and alcohols using PPA (Scheme 1b). Specifically, PPE 6 and condensation product 7 were formed in the presence of PPA, and subsequent formation of PPE-7 complex 8 via hydrogen-bonding interactions, which are the driving force for the formation of 9. Immediately, 9 undergoes alkylation to produce intermediate 10 followed by a tautomerization to deliver the 4-quinolone 4y.

Results and Discussion
To support this hypothesis, several control experiments and density functional theory (DFT) calculations were performed (Scheme 2). Fortunately, the desired product 4y could be obtained directly from 1a, 2a and 3c in 37% yield (Scheme 2a). As depicted in Scheme 2b(1), condensation product 7 was afforded in the 32% isolated yield. Subsequently, 7 was also employed as substrate and the 4-quinolone 4y was obtained in 35% yield (Scheme 2b (2)). Furthermore, PPE 6 was detected by 31 P NMR (Scheme 2b(3), see Supplementary Materials for details). As shown in Scheme 2c, diphosphoric acid ethyl ester 11 (a model PPE) reacts with 7 to produce complex 12 with a computed free energy of 1.6 kcal/mol. Subsequently, the intermediate 13 was generated via transition state TS1, which then undergoes N-alkylation to yield product 4y. The DFT calculations suggest that this process is thermodynamically feasible. Additionally, the generation of such a PPE-7 complex 8 was further verified by 31  readily accessible 2-aminoacetophenones, aldehydes and alcohols using PPA (Scheme 1b). Specifically, PPE 6 and condensation product 7 were formed in the presence of PPA, and subsequent formation of PPE-7 complex 8 via hydrogen-bonding interactions, which are the driving force for the formation of 9. Immediately, 9 undergoes alkylation to produce intermediate 10 followed by a tautomerization to deliver the 4-quinolone 4y.

Results and Discussion
To support this hypothesis, several control experiments and density functional theory (DFT) calculations were performed (Scheme 2). Fortunately, the desired product 4y could be obtained directly from 1a, 2a and 3c in 37% yield (Scheme 2a). As depicted in Scheme 2b(1), condensation product 7 was afforded in the 32% isolated yield. Subsequently, 7 was also employed as substrate and the 4-quinolone 4y was obtained in 35% yield (Scheme 2b(2)). Furthermore, PPE 6 was detected by 31 P NMR (Scheme 2b(3), see Supplementary Materials for details). As shown in Scheme 2c, diphosphoric acid ethyl ester 11 (a model PPE) reacts with 7 to produce complex 12 with a computed free energy of 1.6 kcal/mol. Subsequently, the intermediate 13 was generated via transition state TS1, which then undergoes N-alkylation to yield product 4y. The DFT calculations suggest that this process is thermodynamically feasible. Additionally, the generation of such a PPE-7 complex 8 was further verified by 31    Next, we commenced our studies using 2-aminoacetophenone 1a, benzaldehyde 2a and 1-octanol 3a as model substrates to optimize the reaction conditions. A mixture of 1a, 2a, 3a and PPA (5.0 equiv.) in DMF refluxed for 1 h under nitrogen atmosphere to give the target compound 4a in 38% yield (Table S1, entry 1). PPA amount, time and temperature screening revealed that 1.0 equiv., 3 h and reflux was the best choice (see Table S1 for more details). We further screened the additives including P 2 O 5 and PPA/P 2 O 5 ( Table 1, entries 1-2). According to the screening results, PPA was identified as the best. The effect of solvent was also surveyed, yet the results were inferior to those of DMF (Table 1, entries 3-8 vs. Table S1, entry 7). Subsequently, reaction was performed in the presence of 4 Å MS indicated that 4 Å MS had no effect on this transformation (Table 1, entry 9). Finally, the optimized reaction conditions were determined as follows: 1a (1 equiv.), 2a (1.2 equiv.), 3a (1 mL) and PPA (1.0 equiv.) in DMF refluxed for 3 h under a nitrogen atmosphere (Table S1, entry 7). Under the optimized conditions, a gram scale reaction was conducted and provided 4a in 70% yield (Table 1, entry 10). Next, we commenced our studies using 2-aminoacetophenone 1a, benzaldehyde 2a and 1-octanol 3a as model substrates to optimize the reaction conditions. A mixture of 1a, 2a, 3a and PPA (5.0 equiv.) in DMF refluxed for 1 h under nitrogen atmosphere to give the target compound 4a in 38% yield (Table S1, entry 1). PPA amount, time and temperature screening revealed that 1.0 equiv., 3 h and reflux was the best choice (see Table S1 for more details). We further screened the additives including P2O5 and PPA/P2O5 (Table  1, entries 1-2). According to the screening results, PPA was identified as the best. The effect of solvent was also surveyed, yet the results were inferior to those of DMF (Table 1, entries 3-8 vs. Table S1, entry 7). Subsequently, reaction was performed in the presence of 4 Å MS indicated that 4 Å MS had no effect on this transformation (Table 1, entry 9). Finally, the optimized reaction conditions were determined as follows: 1a (1 equiv.), 2a (1.2 equiv.), 3a (1 mL) and PPA (1.0 equiv.) in DMF refluxed for 3 h under a nitrogen atmosphere (Table S1, entry 7). Under the optimized conditions, a gram scale reaction was conducted and provided 4a in 70% yield (Table 1, entry 10). With the optimized reaction conditions in hand, we then explored the scope and limitations of this reaction (Table 2). Gratifyingly, diverse 2-aminoacetophenones, aldehydes and alcohols were compatible with the reaction, producing the target 4-quinolones 4a-4ao in satisfactory yields. As can be seen, the electron-donating group (methyl or ethyl) on benzaldehyde led to higher yields, while the electron-withdrawing group (F, Cl or Br) gave lower yields (4a, 4e and 4f vs. 4b-4d). It is noteworthy that naphthaldehyde, thenaldehyde and cyclohexanecarboxaldehyde also underwent smooth transformation to give corresponding 4-quinolones (4g-4i). Then, the scope of the reaction was evaluated regarding various aminoacetophenones with different substituents, and it was found that electron-withdrawing groups, such as F, Cl and Br, provided lower yield (4j-4l and 4n-4p vs. 4m and 4q). We further examined the scope of alcohol substrates. Different alcohols worked well with aminoacetophenones and aldehydes yielding diverse N-alkyl-substituted 4-quinolones 4r-4ao in moderate to good yields (48-84%). It was also observed that this reaction was sensitive to the steric hindrance of the alcohols. The yield was decreased with increasing the steric hindrance (e.g., 4r vs. 4y vs. 4af vs. 4am vs. 4an vs. 4ao). Next, we carried out a reaction between 2-aminoacetophenone, benzaldehyde With the optimized reaction conditions in hand, we then explored the scope and limitations of this reaction (Table 2). Gratifyingly, diverse 2-aminoacetophenones, aldehydes and alcohols were compatible with the reaction, producing the target 4-quinolones 4a-4ao in satisfactory yields. As can be seen, the electron-donating group (methyl or ethyl) on benzaldehyde led to higher yields, while the electron-withdrawing group (F, Cl or Br) gave lower yields (4a, 4e and 4f vs. 4b-4d). It is noteworthy that naphthaldehyde, thenaldehyde and cyclohexanecarboxaldehyde also underwent smooth transformation to give corresponding 4-quinolones (4g-4i). Then, the scope of the reaction was evaluated regarding various aminoacetophenones with different substituents, and it was found that electron-withdrawing groups, such as F, Cl and Br, provided lower yield (4j-4l and 4n-4p vs. 4m and 4q). We further examined the scope of alcohol substrates. Different alcohols worked well with aminoacetophenones and aldehydes yielding diverse N-alkyl-substituted 4-quinolones 4r-4ao in moderate to good yields (48-84%). It was also observed that this reaction was sensitive to the steric hindrance of the alcohols. The yield was decreased with increasing the steric hindrance (e.g., 4r vs. 4y vs. 4af vs. 4am vs. 4an vs. 4ao). Next, we carried out a reaction between 2-aminoacetophenone, benzaldehyde and phenol. Unfortunately, phenol was not tolerated in this case, and the target product 4ap was not detected. and phenol. Unfortunately, phenol was not tolerated in this case, and the target product 4ap was not detected.  Interestingly, phenols were compatible with the reaction in the presence of Pd/C ( Table 3, optimization study see Table S2 for more details). A π-alkene-palladium complex might be formed [61], which would promote the cyclization of intermediate 8.
However, the use of palladium catalysts proved to be less effective in the case of alcohols (Table S1, entries 12-14 and Table S2, entry 18). and phenol. Unfortunately, phenol was not tolerated in this case, and the target product 4ap was not detected.  Interestingly, phenols were compatible with the reaction in the presence of Pd/C ( Table 3, optimization study see Table S2 for more details). A π-alkene-palladium complex might be formed [61], which would promote the cyclization of intermediate 8.
However, the use of palladium catalysts proved to be less effective in the case of alcohols (Table S1, entries 12-14 and Table S2, entry 18). Interestingly, phenols were compatible with the reaction in the presence of Pd/C ( Table 3, optimization study see Table S2 for more details). A π-alkene-palladium complex might be formed [61], which would promote the cyclization of intermediate 8. However, the use of palladium catalysts proved to be less effective in the case of alcohols (Table S1,  entries 12-14 and Table S2, entry 18).  In order to gain deeper insight into this reaction, phosphates (14 and 15) were subjected to the reaction, giving the target products in 29% and 36% yields, respectively (Scheme 3a). The electron-withdrawing F substituent has long been proposed to act as a nonclassical hydrogen bond acceptor on the aryl ring [62,63]. We then introduced a hydrogen-bonding acceptor to 2-aminochalcone; 16, featuring the electron-withdrawing F, was tested. The 16 presented product 4u with a better yield then that of 4r (Scheme 3b vs. 3c). N-methyl substituted chalcone 17 provided the desired product 4r in 2% yield, probably owing to the fact that hydrogen-bonding interactions were weakened by the methyl (Scheme 3d). These observations collectively suggested that the hydrogen bonds between PPE and 2-aminochalcones might be formed.  In order to gain deeper insight into this reaction, phosphates (14 and 15) were subjected to the reaction, giving the target products in 29% and 36% yields, respectively (Scheme 3a). The electron-withdrawing F substituent has long been proposed to act as a nonclassical hydrogen bond acceptor on the aryl ring [62,63]. We then introduced a hydrogen-bonding acceptor to 2-aminochalcone; 16, featuring the electron-withdrawing F, was tested. The 16 presented product 4u with a better yield then that of 4r (Scheme 3b vs. 3c). N-methyl substituted chalcone 17 provided the desired product 4r in 2% yield, probably owing to the fact that hydrogen-bonding interactions were weakened by the methyl (Scheme 3d). These observations collectively suggested that the hydrogen bonds between PPE and 2-aminochalcones might be formed. In order to gain deeper insight into this reaction, phosphates (14 and 15) were subjected to the reaction, giving the target products in 29% and 36% yields, respectively (Scheme 3a). The electron-withdrawing F substituent has long been proposed to act as a nonclassical hydrogen bond acceptor on the aryl ring [62,63]. We then introduced a hydrogen-bonding acceptor to 2-aminochalcone; 16, featuring the electron-withdrawing F, was tested. The 16 presented product 4u with a better yield then that of 4r (Scheme 3b vs. Scheme 3c). Nmethyl substituted chalcone 17 provided the desired product 4r in 2% yield, probably owing to the fact that hydrogen-bonding interactions were weakened by the methyl (Scheme 3d). These observations collectively suggested that the hydrogen bonds between PPE and 2-aminochalcones might be formed.  In order to gain deeper insight into this reaction, phosphates (14 and 15) were subjected to the reaction, giving the target products in 29% and 36% yields, respectively (Scheme 3a). The electron-withdrawing F substituent has long been proposed to act as a nonclassical hydrogen bond acceptor on the aryl ring [62,63]. We then introduced a hydrogen-bonding acceptor to 2-aminochalcone; 16, featuring the electron-withdrawing F, was tested. The 16 presented product 4u with a better yield then that of 4r (Scheme 3b vs. 3c). N-methyl substituted chalcone 17 provided the desired product 4r in 2% yield, probably owing to the fact that hydrogen-bonding interactions were weakened by the methyl (Scheme 3d). These observations collectively suggested that the hydrogen bonds between PPE and 2-aminochalcones might be formed.

Scheme 3. Experimental mechanistic investigations. Reactivity of phosphates (a) and chalcones (b-d).
Previous studies show that quinolones were identified as potent NMDA/glycine antagonists with neuroprotective properties [58,64,65]. Thus, all the 4-quinolones were evaluated for their neuroprotective activity against NMDA-induced injury in PC12 cells. The substituent on the nitrogen of 4-quinolone had effect on the activity. N-octyl -substituted compounds exhibited better activity than other targeted compounds ( Figure S2). Among the compounds, the most active is compound 4h, exhibiting neuroprotective potency similar to the MK-801 at 20 µM (Figure 2). To investigate the possible binding mode for this series, 4h was docked into the NMDA receptor. In the binding mode, 4h fitted well in the active pocket. The binding energies of 4h and DCKA were −7.46 kcal/mol and −7.65 kcal/mol, respectively (see Figure S3 for more details). Previous studies show that quinolones were identified as potent NMDA/glycine antagonists with neuroprotective properties [58,64,65]. Thus, all the 4-quinolones were evaluated for their neuroprotective activity against NMDA-induced injury in PC12 cells. The substituent on the nitrogen of 4-quinolone had effect on the activity. N-octyl -substituted compounds exhibited better activity than other targeted compounds ( Figure  S2). Among the compounds, the most active is compound 4h, exhibiting neuroprotective potency similar to the MK-801 at 20 μM (Figure 2). To investigate the possible binding mode for this series, 4h was docked into the NMDA receptor. In the binding mode, 4h fitted well in the active pocket. The binding energies of 4h and DCKA were −7.46 kcal/mol and −7.65 kcal/mol, respectively (see Figure S3 for more details).

General Information
Fetal bovine serum, Dulbecco's modified Eagle's medium, penicillin and streptomycin were obtained from Gibco (Gibco, Paisley, UK). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) and NMDA were purchased from Sigma-Aldrich (Saint Louis, MO, USA). MK-801 was from MCE (Shanghai, China). PC12 cells from Institute of Cell Biology (Chinese Academy of Sciences, Shanghai, China). Unless otherwise noted, all reagents, catalysts and solvents were purchased from commercial suppliers and used without further purification unless otherwise noted. Column Chromatography was performed with silica gel (200-300 mesh). Melting points were determined using an X-4 melting point apparatus with microscope. The IR spectra were recorded with a Mattson FTIR spectrometer 5000. Absorption maxima were measured in cm −1 . 1 H NMR (600 MHz) and 13 C NMR (151 MHz) spectra were achieved in CDCl3 on a Bruker AVANCE 600 MHz spectrometer. High-resolution mass spectra were measured on a ThermoFish QE Focus facility.

Synthesis of Compounds 4a-4ao
General procedure. To a solution of DMF (0.5 mL) was added 2-aminoacetophenones 1 (0.37 mmol), benzaldehydes 2 (0.44 mmol), alcohols 3 (1 mL) and PPA (0.37 mmol) in a 50 mL round bottom flask. The reaction mixture was refluxed for 3 h. The solution was quenched with water and the organic layer was dried over Na2SO4, filtered and evaporated. The resulting crude compound was purified by silica gel column chromatography using hexane/ethyl acetate mixtures to afford the corresponding products 4a-4ao.

General Information
Fetal bovine serum, Dulbecco's modified Eagle's medium, penicillin and streptomycin were obtained from Gibco (Gibco, Paisley, UK). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and NMDA were purchased from Sigma-Aldrich (Saint Louis, MO, USA). MK-801 was from MCE (Shanghai, China). PC12 cells from Institute of Cell Biology (Chinese Academy of Sciences, Shanghai, China). Unless otherwise noted, all reagents, catalysts and solvents were purchased from commercial suppliers and used without further purification unless otherwise noted. Column Chromatography was performed with silica gel (200-300 mesh). Melting points were determined using an X-4 melting point apparatus with microscope. The IR spectra were recorded with a Mattson FTIR spectrometer 5000. Absorption maxima were measured in cm −1 . 1 H NMR (600 MHz) and 13 C NMR (151 MHz) spectra were achieved in CDCl 3 on a Bruker AVANCE 600 MHz spectrometer. High-resolution mass spectra were measured on a ThermoFish QE Focus facility (Waltham, MA, USA).

Synthesis of Compounds 4a-4ao
General procedure. To a solution of DMF (0.5 mL) was added 2-aminoacetophenones 1 (0.37 mmol), benzaldehydes 2 (0.44 mmol), alcohols 3 (1 mL) and PPA (0.37 mmol) in a 50 mL round bottom flask. The reaction mixture was refluxed for 3 h. The solution was quenched with water and the organic layer was dried over Na 2 SO 4 , filtered and evaporated. The resulting crude compound was purified by silica gel column chromatography using hexane/ethyl acetate mixtures to afford the corresponding products 4a-4ao.

Statistical Analysis
Data were expressed as mean ± SD. Multiple group differences were evaluated using one-way analysis of variance (ANOVA) followed by the post hoc LSD test. p < 0.05 were considered statistically significant.

Conclusions
In conclusion, we have developed a novel strategy for the synthesis of N-alkyl-4quinolones via mechanistically intriguing three-component tandem reactions of 2-aminoacetophenones, aldehydes and alcohols. Our work provided the first proof of PPE as a dual hydrogen-bonding catalyst. Notable advantages of this protocol include readily accessible reagents, a one-pot procedure, broad substrate scope and being transition-metal free. The neuroprotective activity evaluation showed that compound 4h demonstrated good protective potency.