Chemoenzymatic Halocyclization of γ,δ‐Unsaturated Carboxylic Acids and Alcohols

Abstract A chemoenzymatic method for the halocyclization of unsaturated alcohols and acids by using the robust V‐dependent chloroperoxidase from Curvularia inaequalis (CiVCPO) as catalyst has been developed for the in situ generation of hypohalites. A broad range of halolactones and cyclic haloethers are formed with excellent performance of the biocatalyst.

Ac hemoenzymatic method for the halocyclization of unsaturated alcoholsa nd acids by using the robust V-dependent chloroperoxidase from Curvularia inaequalis (CiVCPO)a sc atalyst has been developed for the in situ generation of hypohalites. Ab road range of halolactones and cyclic haloethers are formed with excellent performance of the biocatalyst.
First, we evaluated the influence of several reaction parameters, such as pH and reagent concentration, on the efficiency of the bromolactonization of 4-pentenoic acid. In accordance with our previous findings, [10,11,13] the reactionp roceeded optimally at pH 5( with more than 80 %a ctivity at both pH 7a nd pH 4; Table 1). Although this behavior can most likely be attributed to the pH-dependency of the biocatalyst,t he protonation stage of the carboxylate group may also playarole here. Reactions in non-buffered media were lesse fficient, most probably owing to the alkalization of the reaction mediumi nt he course of the reaction. The concentrations of bromide and H 2 O 2 both directly influenced the conversion of the reaction. Performing the reaction in the absence of the biocatalyst did not result in any significant conversion within the timeframe of the experiment.
At ypical time course of the chemoenzymatic bromolactonization is shown in Figure 1. Very pleasingly, CiVCPO performed more than 5c atalytic cycles per second and at least 325 000 catalytic cycles.
Next, we furthere valuated the product scope of the chemoenzymatic halolactonization reaction ( Table 2). Pleasingly,a ll startingm aterials were converted with good to excellent conversions into the corresponding halolactones. In particular,t he cyclohexene-derived (enantiomerically pure) products may serve as building blocks for ar ange of natural products. [14] The selectivity of the reactionw as generally satisfactory with the correspondinghydroxylactone as the sole byproduct. [15] The relative configuration for product 10 a was established based on coupling constantsa nd NOE experiments. The NOE correlation between H-5 (m, dH 4.51-4.48) and H-6b (dd, dH 3.60, J = 12.7, 5.0 Hz, 1H)s uggested the same orientation of H-5 and H-6b. The NOE correlations between H-5a nd the H-4b (m, dH 1.97-1.89), as well as the methyl group at 1.25 ppm indicated protonsl ocated in the same orientation (see the Supporting Information, Figures S7 and S8).
One apparent drawback of the current chemoenzymatic halolactonizationr eaction lies with the nonselective chemical step producing racemic lactones. We therefore envisioned complementing the halolactonization reaction with ah ydrolase-catalyzed kinetic resolution step (Scheme 2). In total, 9 commercial and self-made hydrolasesw eres creened. However, none of the enzymes exhibited an enantioselectivity high enoughf or efficient kinetic resolution ( Figures S56 and S57). Currently,p rotein engineering of the lipase Streptomyces sp. [16] is ongoing to obtainamore enantioselectivea nd hence, practical catalyst.   Finally,w ei nvestigated the possibility of performingh aloetherification reactions in the currents etup. Assuming the intermediate halonium ion is sufficiently stable under the aqueous conditions, we reasoned that intramolecular etherifications should be feasible (Scheme 3).
The proof-of-concept reaction proceeded smoothly to full conversion ( Figure 2). Overall 36 mm of 2-(bromomethyl)tetrahydro-2H-pyran were obtained within 24 h, corresponding to a turnovern umber of more than 360 000 for the biocatalyst.
Indeed,w ith all commerciallya vailablea lkenols tested, we found significant formation of the expected cyclic ethers (Scheme 4). As in case of the lactonizationr eactions, the sole byproducts observed in these reactions were the hydroxyethers (X = OH). The relative configuration of compound 19 a was determined depending upon NOEc orrelations.B ased on the structure of the startingm aterial (À)-carveol, the NOE correlation of CH 3 -2 and H-3 indicated the positioning of thesef unctional groups on the same side( Figure S27).
Also, the relative configuration of compound 20 a was determined based on NOE correlations. Based on the structure of the starting material (+ +)-citronellol and of the methyl group at position3 (CH 3 -3), the NOE correlation of CH 3 -5 and H-2 elucidated the b orientation of Me-2 and H-2 (dH 1.13;F igures S30 and S31).
In the current contribution, we have expanded the scope of CiVCPO as ab iocatalyst for organic synthesis. As emiquantitative comparison [17] of the proposed chemoenzymatic halolacto-Scheme2.Envisionedkinetic resolution of the racemic lactones obtainedf rom the chemoenzymatic bromolactonization reaction.
Scheme3.Envisionedchemoenzymatich aloetherification reaction.  Table 3). The mass intensities of the chemical and chemoenzymatic reactions are comparable. However,t he quality of the reagents and waste products differs significantly.I nt he case of chemicals ynthesis, methylenec hloride as solventi sq uestionable, especiallyc ompared to simple citric acid buffer.F urthermore, stoichiometric amountso fs uccinimide,t he recycling of which necessitates furtherd own-stream processing (DSP) steps, is formed as ab yproduct in the chemical process, whereas the chemoenzymatic process yields water (and unreactedb romide) as byproduct. Finally,t he catalyst consumption of both processes also differs significantly.
Following the established method, the presentp rocedure entailed extraction of the products with dichloromethane, which obviously is questionable from an environmental pointof-view.T herefore, future efforts will concentrate on the substitution of CH 2 Cl 2 with more acceptable alternatives, such as ethyl acetate. [18] Ap articular focus will lie on the intensification of the reaction, that is, increasing the substrate loading (and consequently also the product concentration). This will reduce the relatively large E-factor contribution of the solvent.
Overall, we are convinced that the proposed chemoenzymatic method forh alocyclization represents ap romisinga lternative to establishedc hemical procedures. Further upscaling and characterization of the reaction is currently ongoing in our laboratory.

Experimental Section
Ad etailed description of the biocatalyst preparation and purification as well as ac omplete description of the experimental and analytical procedures can be found in the Supporting information.

Halocyclization of g,d-unsaturated carboxylic acids and alcohols
The halocyclization reactions were performed by using 1mLg lass vials containing 40 mm unsaturated acids, and/or alcohols in 0.1 m citrate buffer (pH 5) with 160 mm KBr and 100 nm CiVCPO. Reactions were started by the addition of 100 mm of H 2 O 2 and stirred by am agnetic bar at 500 rpm for 24 h. The reaction mixtures were extracted with ethyl acetate (1 mL;containing 5mm acetophenone as an internal standard), dried over anhydrous MgSO 4 ,a nd analyzed by GC (Shimadzu;see Table S1).

Preparative-scale chloro-a nd bromolactonization reactions
The reaction was performed in a1 00 mL Erlenmeyer flask at room temperature with stirring. The reaction medium consisted of 0.1 m citrate buffer (pH 5, final volume of 50 mL) with 160 mm of KBr or KCl, 4-pentenoic acid or 2-methyl-4-pentenoic acid (10 mmol), and 100 nm CiVCPO. The reaction was started by the addition of 100 mm of H 2 O 2 .A fter 24 ht he reaction mixture was acidified, extracted with dichloromethane (3 100 mL), and dried over anhydrous Na 2 SO 4 .T he combined organic layers were concentrated under reduced pressure. The chloro-and bromolactone products were purified by flash column chromatography on silica gel (EtOAc/hexanes, 1:2 v/v); 0.914, 1.4, and 1.15 go fc hloro-and bromolactone products were isolated with 70, 80, and 60 %y ield, respectively,a sw ell as hydroxylactone in 30 %y ield in the case of bromolactonization of 2-methyl-4-pentenoic acid and analyzed by NMR spectroscopy.