Nature: oxidative cross-coupling builds a linkage aromatic backbone, and so can enzymes

This article is from the WeChat public number: X-MOLNews

The dianionic compounds are not only the core backbone of many chiral ligands and chiral catalysts, but also widely found in natural products, pharmaceutical active molecules, and materials (Figure 1a). Currently, chemists have developed several methods to construct the bialyl backbone (Nat. Chem.. , 2017,9, 558, click to read details), such as transition metal-catalyzed cross-coupling (Figure 1b), central to axial chiral transfer, desymmetrization and cycloaddition of linked aryl compounds. However, transition metal-catalyzed cross-coupling reactions are hampered by factors such as substrate prefunctionalization, multi-step synthesis, and harsh conditions, which hinder their selectivity and connectivity, resulting in the construction of four-neighbor-substituted diaryl skeletons that remain a major challenge in organic chemistry. In addition, oxidative coupling provides an alternative strategy for the construction of aryl backbones by converting two C-H bonds into C-C bonds (Figure 1b), but oxidative coupling is severely hampered by the inherent spatial and electronic properties of the substrate, which severely hinder its chemoselectivity, site-selectivity and blocking selectivity (Figure 1c), and this strategy is only applicable to electron-rich phenolic substrates, while oxidative coupling of electron-deficient phenols is still very challenging . On the other hand, nature has evolved numerous oxidative enzymes, such as laccase and cytochrome P450 enzymes, that can mediate the dimerization of phenolic substrates to build associative aromatic natural products. However, the application of laccase in selective catalysis is relatively limited due to the defect of autogenous selectivity control. In contrast, an increasing number of P450 enzymes have been found to mediate highly selective oxidation reactions, and a small fraction of these enzymes can also catalyze site-selective and blocking trans-selective dimerization reactions. This fraction of wild-type P450 enzymes with oxidative coupling activity, however, has not yet been used to catalyze unnatural cross-coupling reactions.

Recently, the group of Prof. Alison R. H. Narayan at the University of Michigan (UMich) has successfully achieved oxidative cross-coupling reactions of phenolic substrates using a cytochrome P450 enzyme-mediated biocatalytic strategy (Figure 1d) to construct a series of co-aromatic compounds. Through directed evolution, the authors obtained a P450 enzyme with high reactivity, high site-selectivity and blocking trans-selectivity. This biocatalytic strategy for constructing spatially site-blocked linkage aromatic bonds makes a programmable molecular assembly platform with catalyst-controlled reactivity and selectivity a reality. The related results were published inNatureUp.

Nature: oxidative cross-coupling builds a linkage aromatic backbone, and so can enzymes

Figure 1. Oxidative cross-coupling to construct a linker backbone. Image credits.Nature

First, the authors performed the biotransformation of coumarin 4 using brewer's yeast heterologously expressing KtnC (a cytochrome P450 enzyme), and dimerization product 7 formation could be observed, but brewer's yeast was not suitable for preparative scale reactions because of substrate loading limitations. To develop a whole-cell biocatalytic platform in yeast that could be used for the scale-up reaction, the authors integrated the gene encoding KtnC into baker's yeast, which resulted in a more than threefold increase in the overall yield of the bialkyl product and enabled a preparative scale reaction. With an optimized biocatalytic platform, the authors attempted to shift the dimerization reaction to unnatural oxidative cross-coupling chemistry (Figure 2a). They concluded that the enzymatic active site could facilitate the binding of two non-equivalent phenolic substrates and overcome the traditional problems of spatial and electronic limitations in organic small molecule-mediated oxidative cross-coupling reactions. Indeed, the addition of coumarins 4 and 16 to KtnC-producing cultures resulted in the formation of unnatural cross-coupling products. Further screening showed that adjusting the stoichiometry of the cross-coupling substrates resulted in cross-coupling-dominated products. On this basis, the authors examined a set of coumarin substrates using a KtnC-catalyzed cross-coupling reaction strategy (Figure 2b). The initial cross-coupling reaction between 4 and 9-12 showed that maintaining a similar substitution pattern to 4 at the C4 and C5 positions was beneficial for the reaction, but did not fundamentally affect the activity. In addition, KtnC is tolerant to a range of coumarins with electron-rich and electron-deficient substituents (13-16), as well as to coumarins with bulky C4 and C5 positions (17-22). It is important to note that most cross-coupling reactions maintain their site-selectivity and blocking trans-selectivity in dimerization reactions (Figure 2c). For example, the cross-coupling reaction between 4 and 18 yields the 8,8′-linked product with excellent blocking transfer selectivity (P)-23, while a small number of rare 6,6'-linked products were also observed (P)-24. Cross-coupling with butyl ester (22) then gave the 8,8'-product (25, minor) and 6,6'-product (26, major) in a 2:5 site-selective manner, whereas the conventional approach was unable to construct a 6,6'-linked backbone. In addition, the authors used diverse phenolic substrates as couplers for coumarin 4 (Figure 2D) and observed excellent reactivity of phenolic substrates similar to 4; whereas, as the substrate structures became more different from the natural coumarin backbone, their activity and site selectivity were significantly reduced.

Nature: oxidative cross-coupling builds a linkage aromatic backbone, and so can enzymes

Figure 2. Range of substrates for wild-type KtnC-catalyzed oxidative cross-coupling. Image credits.Nature

Second, since wild-type KtnC showed low activity and site selectivity (<3% total yield of the four cross-coupled products), the authors attempted to obtain a more efficient cross-coupled biocatalyst (LxC) through a directed evolutionary strategy, thus transforming the low-yielding, non-selective reaction into a promising synthetic method (Figure 3a). Specifically, the authors first constructed protein libraries to generate thousands of variants by semirational mutagenesis that have one or more amino acid residue substitutions within the 12 Å range of the wild-type KtnC active site. Coumarin 10 and 2-naphthol (31) were then used as model reactions for high-throughput screening, and the yields of cross-coupling products were analyzed using Rapid-Fire mass spectrometry (MS), and samples with higher yields were then subjected to LC-MS analysis to determine their cross-coupling site selectivity. After multiple iterations of directed evolution, the total activity and site selectivity of the cross-coupling reactions gradually improved (Figure 3c). The first two rounds of protein engineering resulted in a 19-fold increase in the total cross-coupling product yield and maintained a low coumarin dimerization. However, the site-selectivity of the reaction was not ideal, with the C-O bonded cross-coupling product (35) accounting for nearly half of all cross-coupling products synthesized catalyzed by LxC2. Subsequent directed evolution was aimed at improving the site-selectivity of the engineered enzyme. After five iterations of directed evolution, a biocatalyst LxC5 with a 92-fold increase in activity was obtained, and preparative-scale biotransformation also yielded the target product 32 in 50% isolated yield, much higher than conventional synthetic methods. Although the problems of yield and site selectivity were solved, unfortunately, the blocking selectivity of the reaction was reduced from 80:20 er (wild-type enzyme) to 52:48 er (Lcx5). For this reason, the authors performed two additional rounds of directed evolution to improve the blocking selectivity (Figure 3d) and successfully obtained the ideal biocatalyst LxC7, which gave the target product 32 at 77:23 er. These results indicate that the activity, site selectivity and blocking selectivity of this biocatalytic platform can be modulated by protein engineering, thus providing a programmable platform for catalyst-controlled oxidative cross-coupling reactions. provides a programmable platform.

Nature: oxidative cross-coupling builds a linkage aromatic backbone, and so can enzymes

Figure 3. Improving the activity and selectivity of P450 biocatalysts by directed evolution. Image credits.Nature

Along with the use of protein engineering strategies, the authors concluded that other relevant wild-type enzymes exist that could serve as complementary catalysts for oxidative cross-coupling reactions. To this end, the authors used a bioinformatics approach to screen and analyze the sequences of enzymes that can mediate intermolecular oxidative dimerization reactions and obtained enzymes that may catalyze oxidative cross-coupling reactions with naphthol. The artificial fusion enzyme P450-RhFRed was constructed by ligating the P450RhF reductase structural domain (RhFRed) to the carboxyl terminus of each P450 to reconstitute the electron transfer required for catalytic activity. Screening using the P450-RhFRed library, the authors identified fusion enzymes that enable the formation of cross-coupling products of multiple phenolic substrates with naphthol in a 1:1 ratio (Figure 4b). Although the yields of the different cross-coupling reactions differed (43-46), these transformations favored the generation of cross-coupling products rather than competing dimers, thus reducing the requirement for a substantial excess of one substrate over another and improving substrate chemoselectivity and site selectivity. In conclusion, different P450 catalysts can achieve complementary reactivities, providing new biocatalytic pathways for the construction of different linker backbones.

Nature: oxidative cross-coupling builds a linkage aromatic backbone, and so can enzymes

Figure 4. Exploring biocatalysts with complementary reactivity. Image credits.Nature.


Professor Narayan used cytochrome P450 enzymes to catalyze oxidative cross-coupling reactions of phenolic substrates to construct aryl backbones and reverse the unfavorable situation of low reactivity and selectivity of wild-type enzymes through directed evolution. The method is not limited by the spatial and electronic properties of the substrate, overcoming the barrier of substrate reactivity in traditional methods and providing new opportunities for catalyst-controlled site selectivity and blocking trans-selectivity. The method not only provides an efficient and environmentally friendly method for the construction of co-aryl compounds, but also provides a platform for programmable molecular assembly with catalyst-controlled reactivity and selectivity.


Lara E. Zetzsche, Jessica A. Yazarians, Suman Chakrabarty, Meagan E. Hinze, Lauren A. M. Murray, April L. Lukowski, Leo A. Joyce, Alison R. H. Narayan

Nature, 2022,603, 79-85, DOI: 10.1038/s41586-021-04365-7

(This article was contributed by Pyridoxine)