(R)-Citronellal is one of the key chiral intermediates in the synthesis of the isomer (−)-menthol, one of the most commercialised terpenoid flavours worldwide. Enzymatic approaches could represent a less energy-demanding alternative for its synthesis, such as a previously reported bienzymatic cascade starting from inexpensive, commercially available geraniol. A copper radical oxidase (CgrAlcOx) followed by a flavin-dependent ene reductase (OYE2) were used to obtain (R)-citronellal. Here, we used a metal-affinity immobilisation strategy on the His-tagged enzymes for the cascade and studied enzyme recovery and reusability as well as increased solvent tolerance. After screening a panel of resins for enzyme immobilisation and water-immiscible co-solvents, we successfully obtained 95% conversion to (R)-citronellal with 96.9% enantiomeric excess (ee) in a concurrent cascade after 7 h of reaction time, starting from 10 mM of geraniol.

Enzymes are attractive catalysts due to their high chemo-, regio-, and enantioselectivity. In recent years, the application of enzymes in organic synthesis has expanded dramatically, especially for the synthesis of chiral alcohols and amines, two very important functional groups found in many active pharmaceutical ingredients (APIs). Indeed, many elegant routes employing such compounds have been described by industry. Yet, for the synthesis of chiral thiols and thioethers, likewise found in APIs albeit less ubiquitous, only very few biocatalytic syntheses have been reported, and stereocontrol has proved challenging. Here, we apply ene-reductases (EREDs), whose ability to initiate and control chemically challenging radical chemistries has recently emerged, to the synthesis of chiral thioethers from α-bromoacetophenones and pro-chiral vinyl sulfides, without requiring light. Depending on the choice of ERED either enantiomer of the product could be accessed. The highest conversion and selectivity were achieved with GluER T36A using fluorinated substrates, reaching up to 82% conversion and >99.5% ee. With α-bromoacetophenone and α-(methylthio)styrene, the reaction could be performed on a 100 mg scale, affording the product in a 46% isolated yield with a 93% ee. Finally, mechanistic studies were carried out using stopped-flow spectroscopy and protein mass spectrometry, providing insight into the preference of the enzyme for the intermolecular reaction. This work paves the way for new routes for the synthesis of thioether-containing compounds.

The unmatched chemo-, regio-, and stereoselectivity of enzymes renders them powerful catalysts in the synthesis of chiral active pharmaceutical ingredients (APIs). Inspired by the discovery route toward the LPA₁-antagonist BMS-986278, access to the API building block (1S,3R)-3-hydroxycyclohexanecarbonitrile was envisaged using an ene reductase (ER) and alcohol dehydrogenase (ADH) to set both stereocenters. Starting from the commercially available cyclohexene-1-nitrile, a C-H oxyfunctionalization step was required to introduce the ketone functional group, yet several chemical allylic oxidation strategies proved unsuccessful. Enzymatic strategies for allylic oxidation are underdeveloped, with few examples on selected substrates with cytochrome P450s and unspecific peroxygenases (UPOs). In this case, UPOs were found to catalyze the desired allylic oxidation with high chemo- and regioselectivity, at substrate loadings of up to 200 mM, without the addition of organic cosolvents, thus enabling the subsequent ER and ADH steps in a three-step one-pot cascade. UPOs even displayed unreported enantioselective oxyfunctionalization and overoxidation of the substituted cyclohexene. After screening of enzyme panels, the final product was obtained at titers of 85% with 97% ee and 99% de, with a substrate loading of 50 mM, the ER being the limiting step. This synthetic approach provides the first example of a three-step, one-pot UPO-ER-ADH cascade and highlights the potential for UPOs to catalyze diverse enantioselective allylic hydroxylations and oxidations that are otherwise difficult to achieve.

Ene reductases (EREDs) catalyze asymmetric reduction with exquisite chemo-, stereo-, and regioselectivity. Recent discoveries led to unlocking other types of reactivities toward oxime reduction and reductive C–C bond formation. Exploring nontypical reactions can further expand the biocatalytic knowledgebase, and evidence alludes to yet another variant reaction where flavin mononucleotide (FMN)-bound ERs from the old yellow enzyme family (OYE) have unconventional activity with α,β-dicarbonyl substrates. In this study, we demonstrate the nonconventional stereoselective monoreduction of α,β-dicarbonyl to the corresponding chiral hydroxycarbonyl, which are valuable building blocks for asymmetric synthesis. We explored ten α,β-dicarbonyl aliphatic, cyclic, or aromatic compounds and tested their reduction with five OYEs and one nonflavin-dependent double bond reductase (DBR). Only GluER reduced aliphatic α,β-dicarbonyls, with up to 19% conversion of 2,3-hexanedione to 2-hydroxyhexan-3-one with an R-selectivity of 83% ee. The best substrate was the aromatic α,β-dicarbonyl 1-phenyl-1,2-propanedione, with 91% conversion to phenylacetylcarbinol using OYE3 with R-selectivity >99.9% ee. Michaelis–Menten kinetics for 1-phenyl-1,2-propanedione with OYE3 gave a turnover kcat of 0.71 ± 0.03 s–1 and a Km of 2.46 ± 0.25 mM. Twenty-four EREDs from multiple classes of OYEs and DBRs were further screened on 1-phenyl-1,2-propanedione, showing that class II OYEs (OYE3-like) have the best overall selectivity and conversion. EPR studies detected no radical signal, whereas NMR studies with deuterium labeling indicate proton incorporation at the benzylic carbonyl carbon from the solvent and not the FMN hydride. A crystal structure of OYE2 with 1.5 Å resolution was obtained, and docking studies showed a productive pose with the substrate.

Chiral N-heterocycles are a common motif in many active pharmaceutical ingredients; however, their synthesis often relies on the use of heavy metals. In recent years, several biocatalytic approaches have emerged to reach enantiopurity. Here, we describe the asymmetric synthesis of 2-substituted pyrrolidines and piperidines, starting from commercially available ω-chloroketones by using transaminases, which has not yet been comprehensively studied. Analytical yields of up to 90% and enantiomeric excesses of up to >99.5% for each enantiomer were achieved, which has not previously been shown for bulky substituents. This biocatalytic approach was applied to synthesize (R)-2-(p-chlorophenyl)pyrrolidine on a 300 mg scale, affording 84% isolated yield, with >99.5% ee.

Noncanonical redox cofactors are attractive low-cost alternatives to nicotinamide adenine dinucleotide (phosphate) (NAD(P)⁺) in biotransformation. However, engineering enzymes to utilize them is challenging. Here, we present a high-throughput directed evolution platform which couples cell growth to the in vivo cycling of a noncanonical cofactor, nicotinamide mononucleotide (NMN⁺). We achieve this by engineering the life-essential glutathione reductase in Escherichia coli to exclusively rely on the reduced NMN⁺ (NMNH). Using this system, we develop a phosphite dehydrogenase (PTDH) to cycle NMN⁺ with ~147-fold improved catalytic efficiency, which translates to an industrially viable total turnover number of ~45,000 in cell-free biotransformation without requiring high cofactor concentrations. Moreover, the PTDH variants also exhibit improved activity with another structurally deviant noncanonical cofactor, 1-benzylnicotinamide (BNA⁺), showcasing their broad applications. Structural modeling prediction reveals a general design principle where the mutations and the smaller, noncanonical cofactors together mimic the steric interactions of the larger, natural cofactors NAD(P)⁺.