New Organocatalytic Asymmetric Synthesis of Highly Substituted Chiral 2-Oxospiro-[indole-3,4′-(1′,4′-dihydropyridine)] Derivatives

Herein, we report our preliminary results concerning the first promising asymmetric synthesis of highly functionalized 2-oxospiro-[indole-3,4′-(1′,4′-dihydropyridine)] via the reaction of an enamine with isatylidene malononitrile derivatives in the presence of a chiral base organocatalyst. The moderate, but promising, enantioselectivity observed (30%–58% ee (enantiomeric excess)) opens the door to a new area of research for the asymmetric construction of these appealing spirooxindole skeletons, whose enantioselective syntheses are still very limited.


Introduction
In the last few years, the development of new synthetic methods leading to spirooxindoles has aroused remarkable interest [1][2][3][4][5][6]. This structural motif can be found in natural and non-natural products, and its relevance is in part due to its challenging architecture, but also because many of these scaffolds exhibit interesting biological activity (Figure 1) [7][8][9][10][11][12]. Moreover, the potential of isatins to act both as an electrophile and as a nucleophile in many reactions and their easy availability have made them valuable building blocks in organic synthesis, attracting the attention of many scientists [13][14][15]. Moreover, 1,4-dihydropyridine derivatives are a significant class of heterocyclic compounds frequently found in natural products, and many of them also exhibit pharmacological properties [16][17][18][19][20]. As in other drugs, the role of the stereochemistry at C-4 can disclose both qualitative and quantitative differences in the biological activity. Thus, the control of the stereoselectivity in these chiral centers becomes an inspiring task of research, and therefore, there is growing interest for the development of enantioselective methods. Additionally, the generation of quaternary carbon centers is a very active and challenging area of investigation [21][22][23][24][25][26][27][28].

Hypothesis of Work
As part of our ongoing research program about the synthesis of new chiral isatin derivatives, we focused our attention on 2-oxospiro-[indole-3,4′-(1′,4′-dihydropyridine)], in particular on its racemic synthesis reported by Perumal [35] and Yan [36]. Both groups used the same multicomponent strategy employing Et3N under reflux of EtOH (Scheme 1a). However, it is remarkable that Dabiri and Bazgir's group obtained pyrano-fused spirooxindoles [37], instead of dihydropyridine-fused spirooxindoles, under very similar conditions [38,39]. The authors invoked a mechanism wherein a basic medium, isatin and malononitrile, condenses to give an intermediate that reacts with the enamine formed from the acetylenedicarboxylate and the amine (Scheme 1a). Based on this previous work and our experience with Brønsted bases as organocatalysts [40][41][42], we envisioned that a chiral organic base could promote this appealing and controversial reaction, starting directly from the preformed intermediates, enamines 1 and isatylidene malononitriles 2, to give enantioenriched spirooxindoles 3 (Scheme 1b).

Synthesis of Starting Materials: Enamines 1 and Isatylidene Malononitriles 2
For this purpose, we firstly synthesized four different enamines 1a-d in one synthetic step, as described in Scheme 2 [43,44].
The synthesis of three differently-protected isatylidene malononitriles 2a-c was also accomplished (Scheme 3), since the protection of the isatins has been found to be important in the reactivity and enantioselectivity of different processes [13][14][15].
The syntheses are performed in two steps: first, with the protection of the isatin 6 and, then, a Knoevenagel condensation with malononitrile, affording very good yields for each step, after column chromatography.

Screening
To test the viability of our hypothesis, we studied the efficiency of different organocatalysts I-VIII ( Figure 2) in a model reaction between enamine 1a and isatylidene malononitrile 2a (Table 1).  The results suggest a clear influence of the electronic effects of the enamine ring 1 in both the reactivity and the enantioselectivity of the process, although the pattern of correlation is not clear at this point. Thus, while better reactivity was afforded with enamine 1c (71% yield), the best enantioselectivity was reached with the dimethoxy substituted enamine 1a (42% ee). In addition, untreatable reaction crude was observed with enamine 1d.
Moreover, since the protecting group of the isatin scaffold can be relevant in the process, two additional protecting groups (allyl and ethyl) were tested in the reaction with enamine 1b (Scheme 5).

Scheme 5.
Influence of the protecting group on isatylidene malononitrile 2.
Although the enantioselectivity of the process was similar in the three cases, the reactivity was slightly higher with the firstly used, i.e., benzyl-protected isatylidene malononitrile 2a.
Interestingly, an X-ray diffraction structure of compound 3bb was obtained, and it is shown as evidence of the high complexity and functionalization of final target products ( Figure 3) [51]. This structure is in agreement with the kind of molecules obtained by groups of Perumal and Yan through their multicomponent approaches [35,36]. Taking into account the above-mentioned results, summarized in Figure 4, we continued testing different key parameters, such as catalyst loading, concentration and solvent, with enamine 1a and isatylidene malononitrile 2a, which afford the best value of enantioselectivity in the final product (42% ee) using 30 mol % of catalyst VII (Table 2).  We firstly analyzed the effect of the catalyst loading (Entries 1-3), and no improvement was found lowering the amount of catalyst to 20 and 10 mol %. Then, we concentrated the reaction medium, getting slightly improved results (Entry 4). The exploration of the solvents (Entries 4-10), which was performed at the latter concentration, showed MeCN to be the best solvent in this case (Entry 4). As a conclusion, the best reaction conditions of those explored in this work were found to be 30 mol % of catalyst VII and 0.5 mL of MeCN as the solvent (Entry 4).

Scope of the Reaction
With the aim of exploring the generality of this reaction, various isatylidene malononitrile derivatives 2aa′-ad′ were studied under the optimized reaction conditions [52]  The final adducts 3aaa′-aad′ were obtained with moderate to good yield (40%-82%) and with moderate enantioselectivity (30%-58% ee). The results suggest the dependence of the reactivity of the process with the electronic properties of the aromatic ring of the isatin, since derivative 2ab′ (40% yield), with two methyl groups in its structure, was less reactive than those bearing an electron-withdrawing group in their structures (2aa′, 2ac′ and 2ad′ (65%-82% yield)) or the one without substituent (2a (55% yield)). In contrast, the enantioselectivity of the process seems to be independent of the electronic environment in the isatin skeleton.

Mechanism of the Reaction
Based on the previous reported mechanism for the non-asymmetric version of this reaction [35,36] and our experimental results, we propose the tentative mechanism depicted in Scheme 7. Although at this stage, we cannot ensure the bifunctional role for the catalyst in this system [53,54], the experimental results suggest that the presence of both moieties in the skeleton, thiourea and Brønsted base seems to be crucial for the success of this process. Additional studies are actually ongoing in our lab in order to shed light on the mechanism and with the aim of improving the enantiomeric excess values obtained so far.

General Experimental Methods
Purification of reaction products was carried out by flash chromatography using silica gel (0.063-0.200 mm). Analytical thin layer chromatography was performed on 0.25 mm silica gel 60-F plates. 1 H-NMR spectra were recorded at 300 and 400 MHz; 13 C-APT-NMR spectra were recorded at 75 and 100 MHz; CDCl3 as the solvent. Chemical shifts were reported in the δ scale relative to residual CHCl3 (7.26 ppm) for 1 H-NMR and to the central line of CDCl3 (77.0 ppm) for 13 C-APT-NMR.

Materials
All commercially available solvents and reagents were used as received. Catalyst I, II and III were synthesized following our reported protocol [55], and the NMR spectra ( 1 H-NMR and 13 C-APT-NMR) for them are consistent with values previously reported in the literature: I [56], II [57] and III [56]. (3) 3.

Synthesis of E-Enamines 1a-d
To a mixture of diethyl 2-butynedioate 5 (5 mmol) in 40 mL of CH2Cl2, the appropriated aniline 4a-d (10 mmol) was added at room temperature. The reaction vessel was covered with foil in order to prevent the decomposition of 5. The reaction mixture was stirred 24 h at room temperature. Then, the solvent was evaporated under vacuum, and the reaction crude was purified by column chromatography (SiO2, Hex:EtOAc 85:15) (see Scheme 2). Diethyl 2-(4-Methoxyphenylamino)fumarate (1b): Following the general procedure, Compound 1b was obtained as a yellow oil in a 60% yield [58].   (6) To a mixture of the protecting reagent RBr (0.2 mmol) and K2CO3 (0.138 g) in MeCN (10 mL), isatin (6) was added (0.147 g) at room temperature. After that, the reaction mixture was stirred 24 h at reflux. Then, the solvent was evaporated under vacuum, and the reaction crude was purified by column chromatography (SiO2, Hex:EtOAc 8:2), giving rise to the corresponding product 7 (see Schemes 3 and 8).

Knoevenagel Condensation
To a mixture of 7 (1 mmol) in EtOH (10 mL), malononitrile (66 mg) was added. After that, the mixture was heated 24 h at reflux. Then, the solvent was evaporated under vacuum, and the reaction crude was purified by column chromatography (SiO2, Hex:EtOAc 8:2), giving rise to the corresponding product 2 (see Schemes 3 and 8).

General Procedure for the Synthesis of Spirooxindoles 3
To a mixture of catalyst VII (30 mol %, 12.4 mg) and enamine 1a (0.2 mmol, 65 mg), in MeCN (0.5 mL), the isatin derivative 2 (0.1 mmol) was added. The reaction mixture was stirred 5 days at the indicated temperature. Then, the solvent was evaporated under vacuum, and the reaction crude was purified by column chromatography (SiO2, Hex:EtOAc 85:15), giving rise to the corresponding final adduct 3 (see Figure 4, Table 2 (Entry 4), and Scheme 6).

Conclusions
In summary, we have developed an organocatalytic approach for the chiral formation of 2-oxospiro-[indole-3,4′-(1′,4′-dihydropyridine)] derivatives under mild conditions and operational simplicity. Final adducts were reached with promising results of enantioselectivity for the first time. Further mechanistic studies are required in order to understand and to prove the role of the used catalyst in this process. Moreover, additional studies with the aim of improving the enantioselectivity of the method are actively on-going in our laboratory.