Synthesis of Polycyclic Spiro Pyrrolidone Derivatives via DBU‐Catalyzed Diastereoselective Vinylogous Aldol‐Michael Cascade Reaction

A new DBU‐catalyzed vinylogous aldol‐Michael cascade strategy was developed for highly efficient construction of polycyclic heterocycles bearing a spiro pyrrolidone unit. Readily available o‐succinimide‐substituted benzaldehydes and 2(5H)‐furanone were utilized as bifunctional starting materials, both of which possess two potential reactive sites. Good to high yields were achieved under mild metal‐free conditions, along with the excellent diastereoselective generation of four adjacent stereogenic centers. This newly developed methodology provided an efficient and complementary synthetic approach to access spiro pyrrolidone derivatives.

Keywords: metal-free; DBU; vinylogous aldol reaction; cascade reaction; spiro pyrrolidone

Vinylogous aldol reaction: A variety of spiro pyrrolidone derivatives were readily constructed via an unprecedented organocatalytic vinylogous aldol‐Michael cascade reaction of o‐succinimide‐substituted benzaldehydes with 2(5H)‐furanone. Good functional group tolerance along with excellent diastereoselectivities for four adjacent stereogenic centers was achieved under mild reaction conditions.

Polycyclic oxygen heterocycles, such as tetrahydronaphtho[2,3‐b]furanones, were found wide existence in various natural and artificial organic compounds, which generally displayed a wide range of biological activities (Figure a). Whereas, the spiro pyrrolidone subunit proved to be one of the most privileged structural motifs for diverse important natural products, pharmaceutical and agrochemical molecules (Figure b). The combination of the aforementioned two unique frameworks might contribute generation of novel pharmacophores.

Given their ubiquitous occurrence and important bioactivities, several synthetic methods have been pursued for the direct assembly of these compounds.[, [18], [21]] Recently, Stephenson[18] and Chemler group independently developed Pd and Cu‐catalyzed oxidative cascades to construct tetrahydronaphtho[2,3‐b]furanones and tetrahydronaphtho[2,3‐b]furans respectively (Scheme a). Later, Peng and coworkers reported a Ni‐mediated tandem reductive cyclization for the direct access to tetrahydronaphtho[2,3‐b]furanones (Scheme b).[21] However, the above reported examples for preparation of these complex polycyclic frameworks generally limited in intramolecular reaction fashion and required metal catalyst and stoichiometric additives. Meanwhile, they also suffered from either tedious long synthetic route for difficultly accessible starting materials or unavoidable byproducts resulted from excessive oxidant or reductant. Thus, it is of great importance to develop a novel and efficient methodology to construct such valuable skeletons with high step and atom economy under mild conditions. The recently emerged organocatalysis turned out to be a sustainable and ecologically friendly strategy for pharmacy industry, because it can completely avoid metal residue issues.[22] For last decade, cascade reaction has drawn tremendous attention from organic chemists in both academic and industrial area, which can be attributed to its powerful capability to form several C−C or C−X bonds at the same time in one‐pot manipulation, omitting the time‐consuming and costly purification process for possible intermediates.[28] Moreover, high levels of stereoselectivity can be precisely controlled or adjusted in some cases.

Recently, our group also implemented several organocatalytic cascade reactions to construct a variety of useful heterocycles through rational design of activation models.[, ] We previously depicted that o‐succinimide‐substituted benzaldehyde was a versatile bifunctional platform molecule for rapid preparation of various spiro heterocycles. Nevertheless, all the previous reaction model initiated from the nucleophilic addition of benzylic position of o‐succinimide‐substituted benzaldehyde to diverse electrophiles, then the subsequent intramolecular addition to aldehyde group fulfilled the cyclization process to construct bicyclic spiro pyrrolidone skeletons (Scheme c). From the point view of diversity‐oriented synthesis,[41] the investigation of new reaction model is in high demand for the development of new cascade reactions. Meanwhile, we noticed that 2(5H)‐furanone, as a useful synthon, could participate in a variety of reactions, such as Aldol,[44] Michael,[50] Mannich,[57] MBH reaction,[62] arylation and alkylation reactions. However, to the best of our knowledge, direct cascade reaction of 2(5H)‐furanone forming multi carbon‐carbon bonds has been much less explored. We envisaged that o‐succinimide‐substituted benzaldehyde as a versatile bifunctional reactant could participate the cascade reaction with 2(5H)‐furanone to build fused polycyclic heterocycles. As part of our continuous efforts towards the development of novel cascade reaction for efficient construction of useful complex heterocycles, herein we demonstrate a DBU‐catalyzed diastereoselective vinylogous aldol‐Michael cascade reaction of o‐succinimide‐substituted benzaldehydes with 2(5H)‐furanone for the direct access to tetrahydronaphtho[2,3‐b]furanone compounds bearing spiro pyrrolidone unit (Scheme d).

To verify our hypothesis, o‐succinimide‐substituted benzaldehyde 1 a and 2(5H)‐furanone 2 were selected as model substrates for reaction conditions optimization, and the results are summarized in Table . In light of our previous experience in cascade reactions, a series of inorganic and organic bases was initially evaluated (Table , entries 1–3). To our delight, the cascade reaction could forward in presence of catalytic amount base at room temperature to provide expected product 3 a in moderate yield. More importantly, the reaction occurred in high diastereoselective manner and only one isomer could be detected on crude NMR analysis. The structure of 3 a was unambiguously confirmed by X‐ray crystallographic diffraction.[67] DBU proved to be superior to inorganic bases and gave higher yields. Then catalyst loading was examined, and the yield decreased a bit along with decreasing DBU amount (Table , entries 4 and 5). Next, solvent influence was assessed (Table , entries 6–10). The cascade reaction performed well in most tested solvents to generate the desired product with good yields except for THF (Table , entry 8). DMSO turned out to be optimal reaction medium giving highest efficiency (Table , entry 10). Subsequent temperature investigation revealed that 60 °C was suitable, while higher temperature could not improve the efficiency anymore (Table , entries 11 and 12). The reaction could also scale up to 1 mmol scale without any notable difference for model reaction (Table , entry 11).

Optimization of reaction conditions. [a]

1 [a] The reactions were carried out on a 0.2 mmol scale of 1 a with 0.3 mmol of 2 (1.5 equiv) in the presence of x mol % base catalyst, in 2 mL solvent at room temperature for 24 hours. Isolated yield. A single diastereomer was detected. [b] No reaction. [c] Reaction time was shortened to 4 hours. [d] 1 mmol scale reaction, yield in parentheses.

With the optimized reaction conditions in hand, the substrate scope for this cascade reaction was investigated. The results are demonstrated in Scheme . In general, most of the tested substrates reacted smoothly with 2(5H)‐furanone 2 to provide the corresponding tetrahydronaphtho[2,3‐b]furanone products 3 in moderate to good yields along with excellent diastereoselectivities (>20 : 1 dr) for the four continuous stereocenters. Only four products could detect the minor isomers (3 b, 3 d, 3 i and 3 l). Either the electronic property or steric hindrance exhibited evident influence on the cascade process. A broad range of substituents on different positions of the benzene ring of aldehydes were well tolerated to deliver satisfactory outcomes. A series of halogen atoms, including fluorine, chloride and bromide (3 b–3 d, 3 g–3 i, 3 k–3 m) were compatible under standard conditions, which permit potential opportunities for further cross couplings. Substrates bearing strong electron withdrawing groups, such as nitro, ester and methylsulfonyl group could react well to provide expected products in acceptable yields (3 j, 3 n and 3 o). Eventually, different alkyl substituents on pyrrolidone were assayed and they did not influence the transformation evidently (3 q–3 s).

Based on previously literature results and X‐ray crystallographic analysis of 3 a, a plausible catalytic cycle was proposed in Scheme . Firstly, 2(5H)‐furanone 2 was deprotonated by DBU to deliver enolate anion 4 which is active nucleophile. Subsequent nucleophilic addition of 4 to benzaldehyde 1 formed the first carbon‐carbon bond providing alcohol anion intermediate 5. Starting from here, there are two possible pathways to fulfill the catalytic cycle. In path a, DBU acted as proton shuttle to protonate 5 to generate 6. Then 6 was deprotonated by DBU again on benzylic position to provide 7. Subsequent intramolecular Michael addition forwarded 8. On the other hand, there is another possible path b, in which the intramolecular 1,5‐H shift and concomitant Michael addition of benzylic position to cyclic α,β‐unsaturated lactone constructed the second carbon‐carbon bond resulting intermediate 8. Final protonation delivered expected product 3 along with regeneration of catalyst DBU for next catalytic cycle.

In conclusion, an unprecedented organocatalytic vinylogous aldol‐Michael cascade reaction of o‐succinimide‐substituted benzaldehydes with 2(5H)‐furanone was successfully developed for the facile assembly of diverse polycyclic heterocycles bearing tetrahydronaphtho[2,3‐b]furanone and spiro pyrrolidone core structures. DBU was employed as sole proton shuttle catalyst. Excellent diastereoselectivities for four adjacent stereogenic centers was achieved for a broad substrate scope under simple and mild metal‐free conditions, which proved the practical usage of this method. A plausible catalytic cycle was proposed to clarify the cascade reaction mechanism. Further development of novel cascade reaction for highly efficient construction of complex polycyclic heterocycles are underway in our laboratory.

A 10 ml reaction tube was charged with o‐succinimide‐substituted benzaldehyde 1 (0.2 mmol, 1.0 equiv), 2(5H)‐furanone 2 (25.2 mg, 0.3 mmol, 1.5 equiv) and DBU (6.1 mg, 0.04 mmol, 20 mol%). Then 2 mL DMSO was added by syringe. The reaction tube was sealed and the resulting mixture was stirred at 60 °C for 4 hours. The mixture was cooled down to room temperature and 10 mL brine was added. Then the mixture was abstracted by 10 mL ethyl acetate for 4 times. Combined organic phase was washed by brine and dried over Na2SO4. The solvent was removed with rotary evaporator. The residue was purified with column chromatography on silica gel, eluting with petroleum ether and ethyl acetate to afford the corresponding product 3.

This work was financially supported by grants from the National Natural Science Foundation of China (no. 21602089), the Natural Science Foundation of Jiangxi Province (no. 20181BAB203004), and the Fundamental Research Funds for the Central Universities (no. 2020kfyXJJS044), and the National Natural Science Foundation of China‐Henan Provincial People's Government Joint Fund (no. U1804102).

The authors declare no conflict of interest.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

GRAPH: Supplementary