Introduction

Benzene rings represent the most frequently encountered structural motifs in pharmacopeia, serving either as primary bioactive cores or as topological anchors for pharmacophore alignment. Despite their ubiquity, recent studies have highlighted that the indiscriminate incorporation of phenyl rings often imparts suboptimal physicochemical properties to advanced molecules, ultimately limiting their development into effective therapeutic agents1. To overcome these drawbacks, the concept of “escaping from flatland” has emerged, emphasizing the design of F(sp³)-rich scaffolds. Increased carbon saturation enhances molecular solubility and three-dimensionality, thereby improving receptor binding selectivity, reducing off-target interactions, and raising the probability of clinical success2,3. For example, Stepan and Mykhailiuk independently demonstrated that substituting phenyl rings with saturated three-dimensional bioisosteres can significantly improve drug-like properties4,5,6,7. Among such bioisosteres, bicyclo[2.1.1]hexanes (BCHs) have emerged as privileged replacements for aromatic rings (Fig. 1a). They offer unique opportunities to simultaneously optimize pharmacological performance and expand intellectual property space, which has spurred strong interest in developing efficient synthetic routes8,9,10,11,12. One of the most powerful strategies relies on the [2π + 2σ] cycloaddition of bicyclo[1.1.0]butanes (BCBs) with π-systems, a transformation that delivers BCHs with excellent atom economy and synthetic efficiency13,14,15,16,17.

Fig. 1: Synthetic approaches to BCHs and key strategies for enantioselective BCHs synthesis.
Fig. 1: Synthetic approaches to BCHs and key strategies for enantioselective BCHs synthesis.
Full size image

a BCHs as 3D benzene bioisosteres. b Previous works focused on construction of BCHs. c Previous work: Enantioselective [2π + 2σ] cycloaddition via metal or photocatalysis. d This work: Organocatalytic enantioselective [2π + 2σ] cycloaddition.

Significant progress has been made toward the synthesis of racemic BCHs, notably via Lewis acid catalysis18,19,20,21,22,23,24,25,26,27,28,29, radical-mediated strategies30,31,32,33,34,35,36,37,38, and triplet-energy-transfer-enabled [2π + 2σ] cycloadditions (Fig. 1b)39,40,41,42. However, studies by Baran and co-workers have shown that enantiomerically distinct BCH analogs can exhibit dramatically different biological activities when serving as phenyl bioisosteres43. This underscores the critical importance of developing stereoselective methods for BCH synthesis. Early contributions include Bach’s pioneering work, which achieved enantioselective additions of BCBs to quinolones using a chiral template (two equivalents) combined with triplet-energy transfer under 366 nm irradiation at –60 °C44. Subsequently, Jiang’s group reported a bifunctional chiral phosphoric acid–catalyzed process between BCBs and vinylazaarenes under blue LED light at –40 °C45. More recently, Hong and Liu independently realized asymmetric [2π + 2σ] cycloadditions of BCBs with α,β-unsaturated ketones and coumarins using Lewis acid catalysis with chiral ligands (Fig. 1c)46,47. In parallel, Zheng and Zi independently reported palladium-catalyzed enantioselective additions of vinyl–carbonyl–BCBs to alkenes48,49. Photocatalytic asymmetric cycloadditions of BCBs with α,β-unsaturated ketones, vinyl azides, and naphthalenes have also been disclosed by Yoon, Zheng, and You using chiral Lewis or Brønsted acid systems (see Supplementary Fig. S1)50,51,52.

While these elegant studies provide viable strategies for accessing enantioenriched BCHs, they generally suffer from several drawbacks: (i) reliance on cryogenic conditions under inert atmospheres, (ii) requirement for specialized photochemical setups, and (iii) use of potentially toxic transition metals, which diminishes their suitability for pharmaceutical applications. To overcome these limitations, we envisioned a mechanistically distinct approach—leveraging secondary amine catalysis under ambient air, at room temperature, and under operationally simple conditions. This strategy not only expands the synthetic toolbox for BCHs but also addresses key challenges in developing practical, sustainable, and pharmaceutically compatible methods.

Results

Condition optimization

To evaluate the feasibility of our design, we first examined the reaction between cinnamaldehyde (1a) and pyrazole amide–substituted bicyclo[1.1.0]butane (2a) as model substrates (Table 1). A range of secondary amines were screened. With imidazolinone 4a as the catalyst, the desired product 3a was obtained in 75% yield but with only 8% ee. Seeking improved enantioselectivity, we tested sterically hindered catalyst 4b; however, the outcome was unsatisfactory. Seeking improved enantioselectivity, we tested sterically hindered catalyst 4b; however, the outcome was unsatisfactory. In contrast, the use of diarylprolinol silyl ether 4c enhanced both yield and enantioselectivity. Encouraged by this improvement, we further tested the more reactive and sterically demanding catalyst 4d. Although the yield decreased slightly, product 3a was obtained with very high enantioselectivity. We next screened solvents and found acetone to be the optimal medium. To further enhance efficiency, we incorporated hydrogen-bonding catalyst 5 based on our previously established supramolecular iminium ion catalysis concept, followed by comprehensive screening and optimization (see Supplementary Table S1)53,54,55,56. Ultimately, the use of auxiliary catalyst 5a afforded 3a in 82% yield and 99% ee within 24 h.

Table 1 Optimization of the reaction conditionsa
Full size table

Substrate scope

With optimal conditions established, we explored the substrate scope of aldehydes (Fig. 2). Substituents such as methyl (3b), methoxy (3c), or halides (3d–3f) at the para-position of cinnamaldehyde maintained excellent enantioselectivity, though yields were slightly reduced. Similar trends were observed for meta-substitution (3m–3r), highlighting sensitivity to electronic effects. Aldehydes bearing a bulky tert-butyl group (3i) also gave excellent enantioselectivity. Importantly, substrates featuring synthetically versatile groups—including nitro (3g), cyano (3h), ester (3j), trifluoromethylthio, and biphenyl moieties (3k, 3l)—afforded the desired products in moderate yields with consistently high enantioselectivity. Beyond substituted phenyl groups, fused aryl and heteroaryl aldehydes proved compatible, delivering products 3s–3y with excellent enantioselectivity. However, ortho-substituted substrates (3za–3zc) suffered from steric hindrance that impeded reactivity. Interestingly, employing a naphthyl-substituted aldehyde not only improved reaction yield but also facilitated product isolation. Thus, substrate 1s was selected as the representative aldehyde for exploring the scope of BCBs. Substituted BCBs exhibited varied reactivity. Halogen or electron-donating groups at the para- or meta-positions of the phenyl ring afforded good results (3ze–3zi). However, para-electron-donating substituents led to competitive self-ring-opening under acidic conditions, reducing the yield to 42% (3zd). In contrast, strongly electron-withdrawing groups such as para-trifluoromethyl suppressed BCB activation, producing only trace product (3zj). Furthermore, BCBs bearing groups other than acyl pyrazole proved unstable under the reaction conditions, leading to low yields and facile decomposition57. By comparison, the acyl-pyrazole–substituted BCB exhibited superior stability, underscoring its unique suitability for this transformation (see Supplementary Fig. S4 for unsuccessful substrates).

Fig. 2: Scope of substrates.
Fig. 2: Scope of substrates.
Full size image

Standard reaction conditions: aldehyde (0.1 mmol), BCB (0.15 mmol), cat. 4 d (20 mol%), cat. 5a (20 mol%), TFA (40 mol%) and acetone (2 mL).

Product derivatization

To assess the practicality of this protocol, we carried out a gram-scale synthesis of product 3s (3 mmol). The reaction maintained excellent enantioselectivity, with only a slight decrease in yield (–4%). The exposed aldehyde group of 3s proved highly versatile for downstream transformations. Under NaH conditions, it underwent a Wittig reaction to afford alkene 6 in 65% yield with complete retention of enantioselectivity. Reduction with NaBH4 produced alcohol 7 in 86% yield and 98% ee. The amide–pyrazole group could be smoothly converted into a methyl ester in 99% yield under DBU/MeOH conditions, while aldehyde protection with ethylene glycol afforded acetal 9, enabling single-crystal X-ray analysis (CCDC 2441435). Moreover, the aldehyde functionality was transformed into a terminal alkyne in 96% yield using the Bestmann–Ohira reagent. This alkyne intermediate served as a versatile handle for further diversification: it participated in a Yamanaka–Sakamoto–Sonogashira indole synthesis to furnish indole 11 (92% yield, 94% ee), underwent Sonogashira coupling with aryl iodides to provide 12 (79% yield, 93% ee), and engaged in CuAAC “click” chemistry to yield triazole 13 (85% yield, 91% ee) (Fig. 3).

Fig. 3: Applications.
Fig. 3: Applications.
Full size image

Scale-up synthesis and post-synthetic modifications.

We further demonstrated the utility of 3s by oxidizing the aldehyde group to carboxylic acid 14 via Pinnick oxidation (Fig. 4). This transformation greatly expanded opportunities for late-stage functionalization with bioactive molecules. Through amidation or esterification, we successfully conjugated 14 with amine-containing compounds—including tryptamine (16), rivaroxaban (17), fluvoxamine (18), oseltamivir (20), dehydroabietylamine (21), and linagliptin (25)—as well as alcohol-containing molecules such as geraniol (15), cholesterol (19), diacetone-D-glucose (22), pregnenolone (23), L-menthol (24), and testosterone (26). These reactions furnished a diverse library of biologically relevant BCH analogs in good to excellent yields (58–98%).

Fig. 4: Late-stage modification of drugs.
Fig. 4: Late-stage modification of drugs.
Full size image

Late-stage functionalization of the BCH scaffold for bioactive molecule diversification.

Mechanistic investigation

To gain insight into the reaction pathway, we performed a series of control experiments (Fig. 5a). The reaction did not proceed in the absence of either the secondary amine catalyst or the acid, demonstrating that both components are essential. To clarify the role of the acid, we systematically evaluated a range of acids. Neither weakly acidic benzoic acid (BA) nor strongly acidic p-toluenesulfonic acid (TsOH) delivered the target product (Fig. 5b). 1H NMR studies (Supplementary Fig. S5) confirmed that BA failed to promote condensation between the aldehyde and secondary amine, thereby preventing iminium ion formation and halting the reaction. In contrast, TsOH successfully promoted iminium ion generation but its excessive acidity induced rapid ring-opening of the BCB substrate, leading to nearly quantitative formation of ring-opened byproducts and preventing product formation. When 1a was treated with acid alone, we isolated product 27, confirming that BCB ring-opening can occur under acidic conditions in the absence of the amine catalyst.

Fig. 5: Proposed mechanism.
Fig. 5: Proposed mechanism.
Full size image

a Control experiments. b Reaction with different acid. c Proposed catalytic cycle.

Based on these results, we propose the mechanism outlined in Fig. 5c. The secondary amine catalyst 4d and hydrogen-bonding catalyst 5a undergo acid-promoted condensation with aldehyde 1a to form supramolecular iminium intermediate A. The acid-activated BCB then attacks the methylene carbon of A, generating carbocation intermediate D, which subsequently undergoes cyclization with the enamine. The tandem enamine–imine catalytic cycle ultimately regenerates the secondary amine catalyst, affording the target product 3.

DFT calculations

To further support this proposal, density functional theory (DFT) calculations were carried out (Fig. 6). The transition state ts-a-II, formed from 2a in the presence of TFA, was stabilized by 5.6 kcal/mol compared with the TFA-free system ts-a, with subsequent steps also exhibiting reduced energy barriers. These results corroborate the role of TFA in activating the BCB moiety. Additionally, we evaluated the effect of catalyst 5a on the iminium ion pair. Prior to its introduction, the N–O distance measured 2.92 Å. Incorporation of 5a led to a 1.6 kcal/mol stabilization and increased the N–O distance to 3.20 Å. This indicates that 5a weakens electrostatic interactions between the trifluoroacetate anion and the iminium cation, thereby stabilizing the ion pair and improving reactivity. Notably, catalyst 5a also enhances the electrophilicity of the iminium double bond, particularly at the methylene carbon, facilitating nucleophilic attack by the BCB-derived carbanion. These computational insights strongly support the cooperative catalytic role of 5a in the reaction.

Fig. 6: DFT calculations.
Fig. 6: DFT calculations.
Full size image

a Investigated whether TFA assists in the ring-opening of BCBs through DFT calculations. b Investigated the influence of hydrogen-bonding catalysts on iminium ion pairs through DFT calculations.

Discussion

In summary, we have developed a secondary amine–catalyzed enantioselective [2π + 2σ] cycloaddition of BCBs with α,β-unsaturated aldehydes. The reaction proceeds efficiently under ambient conditions without the need for an inert atmosphere, displaying broad functional group tolerance and delivering products with outstanding enantioselectivity (up to 99% ee). Beyond the direct transformation, the resulting products proved highly versatile, undergoing a variety of downstream modifications to furnish structurally and functionally diverse derivatives. Mechanistic investigations, supported by control experiments and DFT studies, confirmed an acid-promoted secondary amine catalytic pathway. Overall, this work establishes a robust and practical platform for accessing chiral BCH scaffolds and offers a valuable complement to the limited existing methods for asymmetric BCB cycloadditions.

Methods

General procedure for enantioselective cycloaddition of and α,β-unsaturated aldehydes and bicyclo[1.1.0]butanes

To a 10 mL Schlenk tube equipped with a magnetic stir bar was added cat. 4d (0.02 mmol, 0.2 eq.), cat. 5a (0.02 mmol, 0.2 eq.), aldehydes 1 (0.1 mmol, 1.0 eq.), BCBs 2 (0.15 mmol, 1.5 eq.) and TFA (0.04 mmol, 0.4 eq.), then acetone (2 mL) was added. The resulting mixture was stirred at 10 °C for 24 h. Upon completion of the reaction, the reaction mixture was concentrated under reduced pressure, and the resulting crude mixture was purified by silica gel column chromatography to afford the pure product 3.