Kinetic resolution of 2-hydroxymethyl-1,4-benzodioxanes by Pseudomonas fluorescens

Pseudomonas fluorescens lipase-catalyzed transesterification of 2-hydroxymethyl-1,4-benzodioxanes of different substitution patterns were studied in the presence of vinyl acetate in organic solvents. The influence of structural features on the conversion and enantioselectivity is discussed and it is shown that the presence of a C-3 methyl or aryl substituent significantly hinders the binding of the substrate to the active site of the enzyme. Thus, with a bulky C-3 substituent no transesterification of the 2-hydroxymethyl group could be observed


Results and Discussion
According to our earlier results, 13 the lipase-catalyzed transesterification of (±)-3 and -4 using vinyl acetate (VA) as an irreversible acyl donor has been carried out in dry dioxane at room temperature in the presence of Pseudomonas fluorescens (PsfL) (Scheme 2).The progress of the reaction was monitored by TLC, which surprisingly showed that no transformation occurred even after 168 hours (Table 1, entries 1 and 2).The use of different solvents of various dielectric constants and dipole moments (entries 3-6) did not help the situation either.Similarly, the 1,4benzodioxane derivative 5, possessing no formyl group on the aromatic ring, did not show transformation at all (entry 7).Moreover, it should be also noted that the lipase from Pseudomonas cepacia (PCL) did not catalyze the transesterification of these compounds [(±)-3-5, entries 8 and 10] either, although this enzyme could be used successfully for the kinetic resolution of hindered 3-hydroxymethyl-2-aryl-2,3-dihydrobenzo[b]furans. 24Since 2hydroxymethyl-1,4-benzodioxane [(±)-6] could be resolved efficiently by PsfL (entry 11), 13 the above results clearly indicate that it is the presence of the bulky C-3 aryl group that hinders the binding of the substrates (±)-3-5 to the active site of the enzyme.

Scheme 2
The role of this steric hindrance inflicted by the C-3 substituent was also proved by transesterification of (±)-7 possessing a readily accessible C-7 hydroxymethyl group besides the sterically hindered C-2 one.In accordance with our expectation, its enzyme-catalyzed transesterification took place exclusively at the C-7 hydroxymethyl group and after 154 h, it afforded the monoacetate derivative (±)-8 with 100% conversion which could also be obtained by a simple acetolysis of (±)-7 carried out at the boiling point of acetic acid.This result also clearly indicated that the presence of a bulky group at C-3, such as an aryl one, makes the C-2 hydroxymethyl group unable to fit the active site of the enzyme, and thus the readily available C-7 hydroxymethyl group is acylated preferably, which process is certainly not enantioselective.When the C-3 aryl group was replaced by a smaller one such as a methyl group [(±)-9], the transesterification took place sluggishly with low enantioselectivity (entry 12; 168 h, 25% conversion and 23% ee), as the C-2 hydroxymethyl group became more available for the active site of the enzyme.25. e Conversion degrees were calculated from the yields of the isolated products obtained by preparative TLC.The reaction was carried out at room temperature.f The enantiomeric excess was determined by measuring the optical rotation in chloroform (c = 0.2-1 g/100mL) at sodium D-line and compared with those published in the literature or by HPLC using Chiracel OJ column as chiral stationary phase.g Absolute configuration of compound was determined by CD measurements based on our chiroptical rule 27 and chemical correlation as well.h Calculated according to Ref. 26.
As a summary, we have shown that Pseudomonas fluorescens lipase-catalyzed acetylation 2hydroxymethyl-1,4-benzodioxanes possessing no substituent at C-3 using vinyl acetate as acylating agent provides the corresponding levorotatory alcohols of (2S) absolute configuration in an acceptable chemical yield (40-50%) and enantiomeric purity (ee% > 65), since the (R) enantiomer reacted faster in all cases.Moreover it has been also shown that the formation of the enzyme-substrate complex (ES) is not considerably influenced by the substituents of the aromatic ring, having different steric effect and polarity.However, the replacement of the 3-H to a methyl group hinders the formation of the enzyme-substrate complex, resulting in low conversion and enantioselectivity.Interestingly, with a C-3 substituent larger than a methyl group, the ES complex does not form at all.Interpretation of our experimental results is in progress by using computer-assisted molecular design (CAMD) and the X-ray structure of PfsL. 30perimental Section General Procedures.Column chromatography was performed on silica gel (Merck 60, 70-230 mesh).The thin layer chromatography was performed on aluminum backed TLC plates of silica gel 60 F254 (Merck, 0.2 mm) with the indicated eluent.The HPLC was performed on Chiralcel OD column (250x4.6 mm, 10 µm, analytical column, Daicel Chemical Industries, LTD.) with Jasco type HPLC system: Jasco PU-980 HPLC Pump, Jasco MD-910 Multiwavelength detector.NMR spectra were recorded on a Bruker AM 360 (360.13MHz for 1 H, 90.03 MHz for 13 C) spectrometer.Chemical shifts (δ) are given from internal CHCl 3 (7.26)for 1 H NMR and (77.00) for 13 C NMR. Coupling constants (J in Hz) are accurate to ±0.2 Hz for 1 H. Elemental analyses (C, H) were conducted using the Carlo Erba 1106 EA instrument.IR spectra were recorded on an Perkin Elmer 16 PC FTIR spectrometer and absorption bands presented in cm -1 .CD spectra were obtained with a Jasco-810 spectropolarimeter in Sharlau Spectosolv grade solvents at room temperature.The tested racemic compounds 3-6, 22 9-13, 19, 27 14 28 and 15 29 were prepared as described in the literature.The compounds characterized below have not been reported yet in the literature even in a racemic form.

General procedure for the lipase-catalyzed resolution of racemic alcohols
A solution of racemic 2-hydroxymethyl-1,4-benzodioxane (0.5 mmol) in dry dioxane (dichloromethane or tetrahydrofuran) was treated with vinyl acetate (1 mL, 22 mmol) and the lipase Pseudomonas fluorescens (10 mg).The suspension was stirred at room temperature.The reaction was stopped by filtration of the enzyme, the filtrate was evaporated to dryness at reduced pressure and the residue was separated by means of preparative TLC to give the corresponding (-)-S-alcohol and (+)-R acetate.

Table 1 .
Enzymatic resolution of 2-hydroxymethyl-1,4-benzodioxanes a Enzymes were commercially available (Fluka).b In units of enzyme per mmol of substrate.c All the solvents were anhydrous and contained 44 mmol vinyl acetate/mmol substrate.DCM: dichloromethane.d Dielectric constant and dipole moment (Debye) values were taken from Ref.