Synthesis of highly substituted tetrahydroquinolines using ethyl cyanoacetate via aza-Michael–Michael addition

A three-component cascade reaction involving 2-alkenyl aniline, aldehydes, and ethyl cyanoacetate in the presence of DBU to synthesize highly substituted 1,2,3,4-tetrahydroquinolines is reported. The reaction proceeded through the Knoevenagel condensation of ethyl cyanoacetate with aldehydes followed by the aza-Michael–Michael addition with 2-alkenyl anilines to prepare the tetrahydroquinoline scaffolds.


Results and discussion
In this paper, it reports the simple one-pot economical preparation of highly substituted tetrahydroquinolines by using 2alkenyl substituted aniline, aromatic aldehydes, and ethyl cyanoacetate; this method saves time during the workup procedure and purication of intermediates and yields minimal reagent waste.
The DBU plays a dual role in the cascade conversion of the Knoevenagel condensation intermediate as well as in the aza-Michael-Michael addition to prepare 1,2,3,4-tetrahydroquinolines.Thus, the overall conversion was integrated irrespective of the Michael acceptors attached to aniline, and resulted in high diastereoselectivity up to 93 : 7. Initial reaction conditions were tested with tert-butyl 2-alkenyl substituted imines (1) and ethyl cyanoacetate with bases including TEA, DIPEA, DABCO, and DBN (Table 1, entries 1-4) in DCM; however no characteristic reactions occurred. 27K 2 CO 3 as a base in DMF and DMSO demonstrated reasonable conversion (Table 1, entries 5 and 6), and it was found that DBU in DCM enabled excellent conversion of (E)-tert-butyl-3-(2-((E)-4-nitrobenzylideneamino)phenyl)acrylate into tetrahydroquinolines 3a/4a at room temperature (Table 1, entries 10 and 11, 95%, racemate).DBU was deemed superior to the other bases.
The 3a/4a isomers were separated through column chromatography, were recrystallized in DCM, and underwent X-ray analysis (Fig. 1) to facilitate understanding of the relative conguration of the diastereomers.The groups of 1,3-cis-tetrahydroquinoline 3a (major isomer) with the distorted chair conguration of 4-NO 2 Ph, -CH 2 CO 2 -t Bu preferred the same side of the ring; the opposite was observed for 4a (alternative, both hydrogens were 1,3 cis in the major diastereomer and in its opposite).To further evaluate diastereoselectivity, we measured the reaction temperature and catalyst loading; however, the results revealed a poor yield and no evident improvement in diastereoselectivity (Table 1, entries 10 and 12).
Further investigations were conducted using solvents such as MeOH, THF, and DMF (Table 1, entries 7-9) with DBU as a base, but no signicant improvements in diastereomeric ratio (dr) or yield were observed.
The combination of DCM and DBU was preferable to the other solvents.The mixture of diastereomers was inevitable, and it further experimented with the versatility of the reaction through the cascade addition.Methyl 2-alkenyl-substituted imine (Table 1, entries 15 and 16) yielded products with improved diastereoselectivity.
Tetrahydroquinolines obtained from 1-naphthaldehyde demonstrated improve yield and diastereoselectivity compared with those obtained from 2-naphthadehyde (5b and 5c  Scheme 1 One pot-three component cascade reaction. Scheme 1).Unexpectedly, when DBU was used as the base, the synthesis of 3f (Scheme 1), was unsuccessful aer the corresponding imine reacted with ethyl cyanoacetate (Table 1, entry 17).In addition, the synthesis of 5a (Scheme 1) produced a low yield, and we managed to isolate the intermediate 5a1, which altered our understanding regarding the mechanistic pathway of the cascade reaction and veried the formation of 1,2,3,4tetrahydroquinolines through a Knoevenagel-condensation intermediate.
Aer the initial formation of enol intermediate 2a, the intermediate reacted with aldehyde to produce an aldol product that subsequently endured base-induced elimination to form 7a (Fig. 2).Reactions between Schiff base's and enol intermediate 2a (Mannich reaction) had failed in earlier experiments (Table 1, entries 13, 14, and 17) because the imines were mostly inert, and thus unable to react with ethyl cyanoacetate.It understand from the crystal structures 3a/4a (Fig. 1) that the initial aza-Michael addition to a Knoevenagel intermediate considerably increased the diastereoselectivity whereas subsequent Michael addition to a,b-unsaturated esters yielded a diastereomeric mixture.Thus, for the synthesis of 1,2,3,4-tetrahydroquinoline, it propose a plausible mechanism with a Knoevenagel intermediate that favours cascade transition through the aza-Michael-Michael addition. 30o determine the effective substrate scope of the reaction, it was reviewed systematic studies performed under optimized conditions (Table 2).In this study, 2-alkenyl-4-chloroanilines were efficiently converted to their corresponding tetrahydroquinolines 9a-9g (Table 2, entries 1-7).Regardless of the groups (X ¼ Cl, H or CO 2 Me) present at 2-alkenylaniline, the yields of the tetrahydroquinolines primarily varied according to the reactivity of the aldehydes.Heteroaromatic aldehydes underwent one-pot conversion into 1,2,3,4-tetrahydroquinolines (9e-9g) with moderate yields (Table 2, entries 5-7).Aromatic aldehydes under the same conditions produced 9a, 9j, and 9n (Table 2, entries 1, 10, and 14) and demonstrated excellent yields compared with the other heteroaromatic Scheme 2 Two component approach via Knoevenagel intermediate.aldehydes (Table 2, entries 5-7).In the synthesis of tetrahydroquinoline 9a up to 93 : 7, o-anisaldehyde exhibited the highest diastereoselectivity (Table 2, entry 1).Naphthaldehydes (Table 2, entries 3 and 4) were converted into 1,2,3,4-tetrahydroquinolines (9c and 9d) under optimized conditions, and 5-methoxycarbonylaniline analogues were converted into their corresponding tetrahydroquinolines (9h-9k) with good to moderate yields (Table 2, entries 8-11).In addition to examining the versatility of the reaction toward Michael acceptor a,b-unsaturated esters (Schemes 1 and 2), it was also examined that of the reaction toward a,b-unsaturated phenyl ketones in tetrahydroquinoline synthesis; the results demonstrated high efficiency.In one-pot, 2-amino substituted chalcones were converted into 1,2,3,4-tetrahydroquinolines with good to moderate yield; all yields were superior to those of the other analogues (Table 2, entries 12-18).No major differences in diastereoselectivity were caused by a,b-unsaturated phenyl ketones (Table 2, entries 13 and 14); however, better yields were obtained with high diastereoselectivity upto 90 : 10.The stronger electron-withdrawing phenyl ketone group accelerated cascade conversion more easily than the other a,bunsaturated esters.Separation of the diastereomers through column chromatography and preparative TLC failed in most cases; therefore, they were able to triturate the 1,3-cis isomer (major) separately from the mixture of diastereomers by using methanol.

Conclusions
In summary, it was developed a simple DBU mediated cascade process to effectively synthesize a new class of highly substituted 1,2,3,4-tetarhydroquinolines by using ethyl cyanoacetate in one pot.The reaction mechanism was investigated through control experiments, namely three reactions involving Knoevenagel condensation followed by aza-Michael-Michael addition efficiently conducted at the room temperature with simple practicability.

General methods
Melting points were recorded using a Yanagimoto Micro Melting Point Apparatus Model-S3 capillary melting point apparatus and are uncorrected.TLC analysis was carried out on silica gel 60 F254 precoated glass sheets and detected under UV light. 1 H NMR spectra were obtained at 300, 400 or 500 MHz (as indicated), and 13 C NMR spectra were obtained at 75.5, 100 or 125.6 MHz, using a Bruker NMR spectrometer.Chemical shis (d) are reported in parts per million (ppm) relative to CDCl 3 (7.26 and 77.0 ppm), the coupling constants are reported in hertz (Hz) and the multiplicities are indicated as b ¼ broad, s ¼ singlet, d ¼ doublet, dd ¼ doublet of doublet, dt ¼ doublet of triplet, t ¼ triplet, m ¼ multiplet.In each case proton NMR showed the presence of indicated solvent(s).Infrared spectra were recorded using PerkinElmer FT/IR spectrometer.Mass spectra were recorded on a Micromass Platform II or Finnigan/ Thermo Quest MAT 95XL spectrometer.All reactions were carried out in anhydrous solvents.CH 2 Cl 2 , DMF, DMSO were distilled from Molecular Sieves.MeOH was distilled from Mg cake.All chemicals and solvents were purchased from Aldrich Chemical Co.

Table 1
Ethyl cyanoacetate as nucleophile 2 Ph (3d/4d) TEA (50) DCM 12 --19 Me 4-NO 2 Ph (3d/4d) DIPEA (100) DCM 12 -a All reactions were performed in 30 to 50 mg scale.b Yield of isolated product is a mixture of diastereomers aer column chromatography.c Determined by 1 H NMR analysis of crude reaction mixture.d Reactions were completed at À10 C to rt, 10 h.

Table 2
Substrate scope a All reactions were performed in 50 mg scale at room temperature.b Yield of isolated product was a mixture of diastereomers aer column chromatography.c Determined by 1 H NMR analysis of the crude reaction mixture.