Biocatalytic Synthesis of Isoxazolines Enabled by Cryptic Nitrile Oxide Formation by a Vanadium‐Dependent Chloroperoxidase

ABSTRACT

Isoxazolines are an important class of heterocycles with a broad range of biological activities. One of the most prevalent synthetic strategies to access isoxazolines is through the oxidative [3+2] cycloaddition between nitrile oxides and alkenes initiated by halogenation of a starting aldoxime, but current methods rely on toxic oxidants that produce significant quantities of waste and are largely bioincompatible. We have recently discovered that the vanadium‐dependent haloperoxidase (VHPO) class of enzymes are efficient catalysts for the in situ generation of nitrile oxides. Herein, we have developed a chemoenzymatic protocol for the conversion of aldehydes to nitrile oxides that features a sequential condensation of hydroxylamine with a starting aldehyde, followed by oxidative [3+2] cycloaddition with alkenes enabled by VHPO‐catalyzed halogenation of aldoximes on a broad range of structurally diverse substrates in high yield and excellent chemoselectivity. The protocol is conducted on a gram scale, demonstrated using whole cells and cell lysate, and extended to isoxazole synthesis. Finally, this process is coupled to lipase‐mediated conversion of amines to oximes to generate isoxazolines.

The vanadium‐dependent haloperoxidase (VHPO) class of enzymes has been shown to catalyze cryptic nitrile oxide generation in an oxidative [3+2] cycloaddition reaction between aldoximes and alkenes to generate isoxazolines. This biotechnology has been applied to a chemoenzymatic sequence starting from aldehydes through in situ condensation with hydroxylamine and subsequent oxidative [3+2] cycloaddition.

1. Introduction

Isoxazolines are an important class of nitrogen‐ and oxygen‐containing heterocycles in biologically active pharmaceuticals, agrochemicals, and natural products (Scheme 1a) [1, 2, 3]. One of the most prevalent synthetic strategies for accessing isoxazolines is through oxidative [3+2] cycloaddition of nitrile oxides and alkenes, enabled by the oxidation of aldoximes for nitrile oxide generation (Scheme 1b) [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. Despite the reported synthetic utility of this transformation in chemical synthesis, these methods rely on reagents that require independent preparation, produce a stoichiometric quantity of toxic byproducts, and operate primarily in organic solvents, limiting their application in chemoenzymatic synthesis. Moreover, these protocols depend on the synthesis of the starting aldoxime, leaving an operationally simple protocol starting from readily available aldehydes elusive.

SCHEME 1.

(a) Biologically active isoxazolines. (b) Traditional methods for nitrile oxide generation from aldoximes. (c) Proposed biocatalytic synthesis of isoxazolines through cryptic nitrile oxide formation.

Enzymes are an attractive option for nitrile oxide generation because of their efficiency and selectivity parameters [21, 22, 23, 24]. Despite the promise of using enzymes for halogenation‐induced nitrile oxide synthesis, nature is void of enzymes that have demonstrated aldoxime halogenation. The discovery of enzymes that perform this reaction type would provide a useful and biocompatible catalyst system for isoxazoline synthesis and allow for the selective generation of nitrile oxides for new applications in chemoenzymatic synthesis. Among the host of halogenating enzymes in nature [25, 26, 27, 28], the vanadium‐dependent haloperoxidase (VHPO) class of enzymes has emerged as a powerful platform for chemical synthesis because of their exceptional operational parameters, including their tolerance to organic solvents and lack of requirement for a separate turnover system [29, 30, 31, 32]. As a result, they have attracted significant attention from organic chemists for performing reactions outside of their native substrate scope [33, 34, 35, 36, 37, 38]. We have recently been interested in using VHPOs for performing cryptic halogenation reactions [39], whereby the enzyme is responsible for the repetitive oxidation of a catalytic quantity of halide to generate hypohalous acid as the oxidizing agent using hydrogen peroxide (H₂O₂) as the terminal oxidant—a mechanism we refer to as enzymatic halide recycling (EHR) [40, 41, 42, 43, 44]. Herein, we report that VHPOs are a viable catalyst system for isoxazoline synthesis through cryptic nitrile oxide formation (Scheme 1c).

2. Experimental Section

Our studies began by screening a structurally diverse set of VHPOs in performing the oxidative [3+2] cycloaddition between benzaldehyde oxime (1) and tert‐butyl acrylate (2) to give tert‐butyl 4,5‐dihydroisoxazole‐5‐carboxylate (3). The enzyme panel included the chloroperoxidase from Curvularia inaequalis (CiVCPO) [45] and vanadium bromoperoxidases from Corallina officinalis (CoVBPO) [46], Corallina pilulifera (CpVBPO) [47], and Acaryochloris marina (AmVBPO) [31]. Using 0.0125 mol% enzyme loading, sodium orthovanadate (Na₃VO₄, 1 mM), and 1.1 equiv. each of sodium chloride (NaCl) and H₂O₂ in acetate buffer (25 mM, pH 5) with MeCN (20% v/v) as cosolvent, CiVCPO catalyzed the desired transformation in 90% yield after 4 h (Scheme 2, Entry 1). No detectable product formation was observed for any of the bromoperoxidases examined (Scheme 2, Entries 2–4). The reaction could be performed under EHR conditions with CiVCPO by using a NaCl loading of only 0.2 equiv., affording 3 in 89% yield (Scheme 2, Entry 5). Increasing the H₂O₂ loading to 3.0 equiv. under these conditions, further improved the yield to 95% (Scheme 2, Entry 6). To confirm the necessity of each reaction component, control reactions were conducted in which CiVCPO, Na₃VO₄, NaCl, and H₂O₂ were individually excluded from the reaction mixture (Scheme 2, Entries 7–10). Notably, NaCl loadings between 0.2 and 5.0 equiv. were well tolerated, with higher chloride concentrations providing no additional improvement in the yield (Figure S2). The reaction was also compatible with higher H₂O₂ loadings up to 8.0 equiv. (Figure S3). Additionally, both buffer composition and pH exerted a significant influence on activity, with an ideal pH of 5 and acetate concentrations of 25–50 mM (Figures S4 and S5). The reaction maintained excellent performance across a wide range of cosolvents (Figure S6), performing best with 20% v/v MeCN (Figure S7).

SCHEME 2.

Optimization experiments for biocatalytic oxidative [3 + 2] cycloaddition (1). Reaction conditions: 1 (4.0 µmol, 0.49 mg), 2 (4.0 µmol, 0.52 mg), VHPO (0.0125 mol%), Na₃VO₄ (1 mM), NaCl (0.2–1.1 equiv.), H₂O₂ (1.1–3.0 equiv.), acetate buffer (25 mM, pH = 5, 50 µL), MeCN (200 µL),1 mL total reaction volume, 4 h, rt. Yields determined by HPLC based on a calibration curve. See the Supporting Information for details.

With optimized conditions identified, we next investigated the substrate scope of CiVCPO‐catalyzed oxidative [3 + 2] cycloaddition. The reaction could be conducted in a one‐pot sequence combining aldoxime formation and cycloaddition step, through in situ oxime generation via base‐mediated condensation of the starting aldehyde with hydroxylamine hydrochloride (NH₂OH•HCl), followed by oxidative [3+2] cycloaddition under the standard conditions developed. Gratifyingly, no additional NaCl was required in this sequence, as the neutralization of NH₂OH•HCl by sodium carbonate (Na₂CO₃) provides the corresponding chloride anion in situ, enabling catalytic oxidation of halide by CiVCPO. The developed chemoenzymatic protocol accommodates a broad range of aldehyde substrates in oxidative [3+2] cycloaddition with tert‐butyl acrylate. Electron‐donating substituents, including methyl‐, methoxy‐, and tert‐butyl at the para position of benzaldehyde, are well tolerated, affording the corresponding isoxazolines in 75%–90% yield (Scheme 3, 4–6). Similarly, para‐substituted electron‐withdrawing groups, including bromo‐, chloro‐, fluoro‐, and nitro substituents, furnished products in 66%–82% yield (Scheme 3, 7–10). Aldehydes with meta and ortho methoxy‐ and methyl‐substitution are tolerated, providing the corresponding products in 85%–91% yield (Scheme 3, 11–12) and 90%–92% yield (Scheme 3, 13–14), respectively. Moreover, the optimized reaction conditions are compatible with heteroaromatic aldehydes, including furan and thiophene, in 72%–83% yield (Scheme 3, 15–16). The transformation is readily applicable to the aliphatic substrate, isobutyraldehyde, furnishing the desired isoxazoline in 90% yield (Scheme 3, 17). Finally, the protocol demonstrates excellent compatibility with a diverse set of olefin partners, including styrene, α‐methylstyrene, methyl acrylate, methyl methacrylate, and allyl alcohol, generating the corresponding isoxazolines in 78%–85% yield (Scheme 3, 18–22).

SCHEME 3.

Substrate scope for chemoenzymatic Isoxazoline Synthesis. Reaction conditions: aldehyde substrate (0.4 mmol, 1.0 equiv.), NH₂OH‐HCl (0.44 mmol, 1.1 equiv.), Na₂CO₃ (0.44 mmol, 1.1 equiv.), EtOH (300 uL), H₂O (600 µL), rt, 18 h then olefin substrate (0.44 mmol, 1.1 equiv.), CiVCPO (0.0125 mol%), Na₃VO₄ (1 mM final concentration), H₂O₂ (3.0 equiv.), acetate buffer (25 mM, pH = 5), MeCN (20%), 4 h, rt. used. *reaction time = 8 h. Yields determined by isolation. See the Supporting Information for more details.

Over the course of our investigation, we were interested in understanding the mechanistic features and reaction profile of the VHPO‐mediated oxidative [3+2] cycloaddition between aldoximes and alkenes. To gain insight into the behavior of reactive intermediates in solution, we independently synthesized hydroximoyl chloride 23 using known methods [48]. Subjection of 23 to standard reaction conditions and controls (no enzyme, Na₃VO₄, H₂O₂, NaCl) immediately led to the formation of isoxazoline 3 in near quantitative yield, providing insight that (1) confirms intermediacy of hydroximoyl chloride 23 and (2) suggests that nitrile oxide formation occurs spontaneously in situ after formation of 23, enabling the ensuing [3+2] cycloaddition (Scheme 4a). To gain further insight into the steady state concentration of 23 in solution, a time course study was conducted. Over the course of the reaction, an expected steady consumption of aldoxime 1 and an increase in isoxazoline 3 were observed. Interestingly, a low steady‐state concentration of ∼15% was observed for hydroximoyl chloride 23, suggesting that a critical feature of running the reaction under biocatalytic and primarily aqueous reaction conditions is to ensure that there is enough alkene coupling partner (2) to prevent undesired reactivity (Scheme 4b).

SCHEME 4.

Mechanistic and time course experiments for the conversion of aldoxime 1 to isoxazoline 3.

A proposed mechanism for VHPO‐mediated oxidative [3+2] cycloaddition is outlined in Scheme 5. Our developed one‐pot procedure begins with base‐mediated condensation between NH₂OH•HCl and the starting aldehyde to generate the corresponding aldoxime and chloride anion (Cl⁻) that can be used by the VHPO (Scheme 5, Step 1). Consistent with previously proposed mechanisms [25, 26, 27, 28], VHPOs are dependent on a vanadate cofactor bound to a histidine residue in the enzyme active site (I). When exposed to H₂O₂, two water molecules are displaced, leading to the generation of the corresponding peroxovanadium intermediate (II) that is primed for nucleophilic attack by chloride, generating a vanadium‐bound hypohalite (III) that can either participate directly in a halogenation event or is released from the coordination sphere to generate hypohalous acid (HOX) as the halogenating agent. We propose that one of these halogenation pathways is responsible for the chlorination of the in situ‐generated aldoxime to give the corresponding hydroximoyl chloride (IV) (Scheme 5, Step 2). In analogy to proposed mechanisms for aldoxime chlorination [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20], IV will convert to a nitrile oxide (V) in situ to initiate the ensuing [3+2] cycloaddition reaction and generate the corresponding isoxazoline.

SCHEME 5.

Proposed mechanism for VHPO‐catalyzed oxidative cyclization of oximes with alkenes.

Upon completion of reaction development, we next evaluated the synthetic utility of the developed VHPO‐catalyzed oxidative [3+2] cycloaddition. The optimized one‐pot, two‐step sequence was readily applied to the gram‐scale synthesis 3 from benzaldehyde (24) (Scheme 6a). The reaction could also be performed with cell lysate and whole cells in 93% and 91% yield, starting from aldoxime 1, respectively. Gratifyingly, this process could also be performed starting from 24 in 82% over two steps (Scheme 6b). To further showcase the versatility of cryptic nitrile generation, the methodology was extended to the synthesis of an isoxazole using dimethyl but‐2‐ynedioate as the dipolarophile, affording the desired product in 88% yield (Scheme 6c, 25). Moreover, the cycloaddition protocol was effectively merged with lipase‐catalyzed oxime generation from amine oxidation [49], thereby establishing an integrated one‐pot, two‐enzyme tandem process for the synthesis of isoxazoline derivatives. This tandem system was well tolerated by benzylamine and para‐substituted benzylamines bearing methyl‐ and fluoro‐ groups, furnishing the corresponding isoxazolines in 79%–89% yield (Scheme 6d, 3, 4, 9). Finally, furfurylamine was well tolerated under these conditions, yielding the corresponding isoxazoline in 81% yield (Scheme 6d, 15).

SCHEME 6.

Synthetic utility of VHPO‐catalyzed oxidative [3+2] cycloaddition. (a) Gram‐Scale Synthesis of tert‐butyl 4,5‐dihydroisoxazole‐5‐carboxylate. Reaction conditions: 1 (4.8 mmol, 1.0 equiv.), NH₂OH‐HCl (5.28 mmol, 1.1 equiv.), Na₂CO₃ (5.28 mmol, 1.1 equiv.), EtOH (3.6 mL), H₂O (7.2 mL), rt, 18 h then 2  (5.28 mmol, 1.1 equiv.), CiVCPO (0.0125 mol%), Na₃VO₄ (1 mM final concentration), H₂O₂ (3.0 equiv.), acetate buffer (25 mM, pH = 5), MeCN (20%),  4 h, rt. used. (b) Oxidative cycloaddition with lysate and whole cells. Reaction conditions: 1 (4.0 µmol, 0.49 mg), 2 (4.0 µmol, 0.52 mg), wet lysate/whole cells (10 µL), Na₃VO₄ (1 mM), NaCl (1.1 equiv.), H₂O₂ (3.0 equiv.), acetate buffer (50 mM, pH = 5, 50 µL), MeCN (200 µL),1 mL total reaction volume, 4 h, rt. (c) Extension to isoxazole synthesis. Reaction conditions: 24 (0.4 mmol, 1.0 equiv.), NH₂OH‐HCl (0.44 mmol, 1.1 equiv.), Na₂CO₃ (0.44 mmol, 1.1 equiv.), EtOH (300 uL), H₂O (600 µL), rt, 18 h then dimethyl but‐2‐ynedioate (0.44 mmol, 1.1 equiv.), CiVCPO (0.0125 mol%), Na₃VO₄ (1 mM final concentration), H₂O₂ (3.0 equiv.), acetate buffer (25 mM, pH = 5), MeCN (20%), 8 h, rt. (d) Dual enzymatic cascade integrating oxime formation and oxidative cycloaddition. Reaction conditions: amine substrate (0.4 mmol, 1.0 equiv.), Candida antarctica lipase B (CalB immo Plus, 10.0 mg), H₂O₂ (2.0 equiv), EtOAc (0.8 mL), rt, 1 h then 2 (0.4 mmol, 1.1 equiv.), CiVCPO (0.0125 mol%), NaCl (1.1 equiv.), Na₃VO₄ (1 mM final concentration), H₂O₂ (3.0 equiv.), acetate buffer (25 mM, pH = 5), MeCN (20%), 4 h, rt.

3. Conclusion

In conclusion, we have discovered that the VHPO class of enzymes is a viable biocatalyst platform for oxidative [3+2] cycloaddition through cryptic nitrile oxide generation. This catalyst system is effective for the synthesis of isoxazolines in high yield and chemoselectivity. This chemoenzymatic platform is scalable, can be performed with wet lysate and whole cells, can be used to generate isoxazoles, and can be integrated into a multi‐enzyme cascade starting from amines. It also serves as the first example of a cross‐selective reaction performed using an EHR mechanism. These studies expand the synthetic repertoire of VHPOs in organic chemistry and provide a biocompatible platform for nitrile oxide formation in chemoenzymatic synthesis.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

The authors have cited additional references within the Supporting Information [50–61]. Supporting File: anie72152‐sup‐0001‐SuppMat.Pdf.

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.