Efficient access to structurally diverse molecular scaffolds from readily available starting materials remains a central goal of organic synthesis. One strategy is to use a functional group to facilitate scaffold assembly and then remove it once its synthetic role is complete. Classical carbonyl chemistry exemplifies this “delete-after-use” approach through α-functionalization followed by decarbonylation or decarboxylation. When functional group deletion generates a stereogenic center, achieving both selective bond cleavage and stereochemical control presents a key challenge. Developing stereoselective methods for functional group deletion would therefore extend this strategy to the synthesis of diverse chiral molecules.

Malononitrile, an inexpensive and readily available building block, offers a practical starting point for exploring this concept. Its two cyano groups facilitate carbon−carbon bond formation, enabling the assembly of diverse molecular structures. Subsequent stereoselective removal of one cyano group would provide diverse chiral nitriles, with the remaining cyano group serving as a versatile handle for further transformations.

To achieve efficient C–CN bond activation with stereocontrol, the Liu group developed a photoenzymatic decyanation platform for the modular synthesis of chiral nitriles. The system couples the reducing power of a photoexcited flavin cofactor with the chiral environment of an enzyme active site. The work was published in the Journal of the American Chemical Society under the title “Molecular Diversity Generation via Selective Enzymatic C–CN Bond Scission.”

Figure 1. Modular access to chiral nitriles through selective photoenzymatic cyano group deletion.

Enzyme screening and directed evolution

The researchers hypothesized that photoexcitation of fully reduced flavin mononucleotide (FMN_hq*) could generate a sufficiently strong reductant to initiate direct C–CN bond cleavage through single-electron transfer (SET). Screening a panel of old yellow enzymes (OYEs) identified OYE1 from Saccharomyces pastorianus as the optimal catalyst for the model transformation. After optimization of the reaction conditions, wild-type OYE1 afforded the desired product in 78% yield and 13:87 e.r.

To further enhance the yield and stereoselectivity, the team targeted residues within the enzyme active site using focused rational iterative site-specific mutagenesis (FRISM) and site-saturation mutagenesis (SSM). Across three rounds of directed evolution, the team surveyed 24 positions and screened over 400 mutants. The resulting optimal variant, OYE1_G3, carried the A85N, V117A, and N251A substitutions. It furnished the model chiral nitrile in 87% yield and 5:95 e.r.

Figure 2. Enzyme screening, reaction optimization, and directed evolution of OYE1.

Substrate scope and modular synthesis

The optimal variant OYE1_G3 accepted a broad range of geminal dinitriles, including aryl-alkyl, dialkyl-, and mono-aryl-substituted compounds. Notably, unprotected phenol and aniline, as well as alkene, alkyne, and ester functionalities, were well tolerated. Several transformations were demonstrated on a preparative scale, supporting the synthetic utility of the platform.

The team also developed a one-pot cascade that combines Michael addition with photoenzymatic decyanation. Aryl-substituted malononitriles first undergo a spontaneous, nonenzymatic Michael addition to electron-deficient alkenes in the reaction buffer. The resulting geminal dinitrile adducts then undergo enzyme-catalyzed cyano group deletion to furnish chiral nitriles. This formal decyanative coupling offers a convenient way to generate structural diversity by combining different donors and acceptors.

To demonstrate modular synthesis, the researchers used malononitrile as a common starting material and introduced different substituents through cross-coupling and nucleophilic substitution reactions. Photoenzymatic decyanation then established the desired stereochemistry. The retained nitrile served as a handle for subsequent conversion into carboxylic acids or amines, enabling access to bioactive molecules including (R)-naproxen and phenibut. These examples connect the scaffold assembly of malononitrile with stereocontrolled functional group deletion in a single synthetic workflow.

Figure 3. Modular synthesis from malononitrile, downstream nitrile derivatization, and applications to (R)-naproxen and phenibut.

Mechanistic investigations

To investigate the proposed mechanism and identify the predominant reaction pathway, the team conducted a series of mechanistic studies. TEMPO-trapping experiments and room-temperature electron paramagnetic resonance (EPR) spectroscopy supported the formation of a benzylic radical intermediate (Int1), consistent with SET-initiated C–CN bond cleavage. Deuterium incorporation experiments then probed how this radical is converted into the product. Experimental results support solvent-derived protonation as the source of the hydrogen, rather than direct hydrogen atom transfer (HAT) from the flavin cofactor.

To further support the involvement of a carbanion intermediate, the team designed probe substrates capable of β-elimination or intramolecular nucleophilic substitution. Products consistent with these pathways provided evidence for carbanion formation. Together, these experiments support a predominant pathway in which initial SET-induced C–CN bond scission generates Int1, a second SET produces the carbanion Int2, and asymmetric protonation furnishes the chiral nitrile. This sequential SET–PT pathway expands the mechanistic understanding of radical termination in flavin-dependent photoenzymatic catalysis.

Figure 4. Proposed catalytic cycle, mechanistic experiments, and enzymatic cyanide degradation.

Cyanide waste degradation

To address concerns regarding cyanide (CN⁻) waste generated during decyanation, the researchers demonstrated its enzymatic degradation using the cyanide hydratase NitAn1 from Aspergillus niger K10. More than 95% of the cyanide generated by the photoenzymatic reaction was degraded. This result enhances the environmental compatibility of the system.

The study establishes a versatile photoenzymatic platform for stereocontrolled C–CN cleavage of malononitriles, enabling the preparation of diverse chiral nitriles and their derivatives. Combining this transformation with scaffold assembly from malononitrile demonstrates the “delete-after-use” principle in modular synthesis and provides a basis for developing further enzyme-catalyzed carbon–carbon bond cleavage reactions.

Authors

Dr. Zhen Liu and Dr. Lizhi Tao are the corresponding authors. Qing-Qing Zeng is the first author. Other authors include Xinchen Yang, Baixu Ma, Fenshuai Sun, Minghao Xu, Dr. Jian Tang and Dr. Marc Garcia-Borràs.

Research support

This work was supported by the “U35 Pei Miao Funding Program” of Changping District, Beijing (Grant No. CHPU35202501004 to Z.L.), the Ministry of Science and Technology of China (Grant No. 2023YFF1204400 to Z.L.), the Tsinghua Institute of Multidisciplinary Biomedical Research, Tsinghua University, and the Open Project of State Key Laboratory of Synergistic Chem-Bio Synthesis (Grant No. sklscbs202506 to Z.L.). This work was also supported by Shenzhen Science and Technology Program (ZDSYS20230626090803004 to L.T.), the Spanish Ministry of Science and Innovation MICINN (PID2022-141676NB-I00 and TED2021-130173B-C42 projects, and RYC 2020-028628-I grant to M.G.B), and the Generalitat de Catalunya (2021SGR00623 project to M.G.B.).

Original article

Zeng, Q.-Q.; Yang, X.; Ma, B.; Xu, M.; Sun, F.; Garcia-Borràs, M.; Tang, J.; Tao, L.; Liu, Z. Molecular Diversity Generation via Selective Enzymatic C–CN Bond Scission. J. Am. Chem. Soc. 2026, DOI: 10.1021/jacs.6c17512.

http://doi.org/10.1021/jacs.6c17512