Lipase catalysed resolution and microbial reduction for obtaining enantiopure 1-(2-thienyl)alkanols

Efficient methods have been developed for resolving 1-(2-thienyl)alkanols with lipase B from Candida antarctica as catalyst. Fermenting cells and cell-free reduction systems of Geotrichum candidum (IFO 4597) have also been tested in asymmetric reductions of 1-(2-thienyl)alkan-1-ones.


Introduction
Kinetic resolution is an efficient method for providing both enantiomers of chiral compounds.The success of a process depends profoundly on the ability of the enzyme to select between the two enantiomers as expressed by the enantiomeric ratio E, which is the relative rate of reaction with the two enantiomers. 1Compared to asymmetric synthesis, kinetic resolution has the benefit that a high enantiomeric excess can be obtained even with a moderate E-value.One of the short comings of lipase catalysed resolutions is that the maximum yield is limited to 50 %.This can be overcome for instance by performing a dynamic kinetic resolution in which the unreacted enantiomer is continuously racemised.

Stereo preference of resolutions
3][4][5] Since no optical rotation has been reported for enantiopure 1-(2-thienyl)butanol (3b) or the corresponding butanoate, the ester was hydrolysed and treated with Raney-Ni to give (R)-4-octanol10 with = +0.25°This shows that the (R)-butanoate also was formed in this case.This way of proving the configuration is not applicable for 1-(2-thienyl)pentanol (4b) since the formed secondary alcohol is achiral.However, based on knowledge of the stereo preference of CALB and and the elution order of enantiomeric alcohols and esters (Table 2), it is reasonable to assume that the (R)-ester and the (S)-alcohol are the products also in the resolution of 4b 11 .
In order to further prove the stereopreference in the transesterifications, we subjected the remaining ester [(R)-1c] to enzyme catalysed methanolysis.The produced alcohol gave [α] 20 D = +21.66°, almost exactly opposite the reported values for the (S)-alcohol.

Microbial asymmetric reduction
The ketones 1a-4a were subject either to fermenting cells or a cell free acetone powder12 of Geotrichum candidum (IFO 4597).Only 1a was reduced to give (S)-1b.This is in agreement with earlier reported substrate requirements for the oxido-reductases of Geotrichum candidum.12Fermenting cells gave an ee of 74 %, but high ee´s (> 99 %) were obtained using acetone powder in buffer.However, the yields were not optimal, being 54 % in MES buffer and 78 % in phosphate buffer after 24 h.We tried to improve this by performing the reduction in organic solvents using G. candidum cells immobilized on Celite in hexane.Unfortunately this failed to give any measurable reduced product for any of the substrates 1a-4a.A different system consisting of fermenting cells of G. candidum, immobilised on the water-absorbing polymer LP-100, also failed to give reductions for the substrates 2a-4a.However, 1a was reduced in 7 % yield after 4 days.This was considered too low to be of practical importance and further attempts were not carried out.

Experimental Section
General Procedures.Immobilized CALB (Novozyme 435, Novozymes ) had an activity of 7000 PLU/g, and a water content of 1-2 % w/w.Solvents were dried over molecular sieves.The cultivation of Geotrichum candidum (IFO 4597) and preparations of its acetone-powder (AGP) are described elsewhere. 12Column chromatography was performed using silica gel 60 from Fluka.Enzymatic reactions were performed in a shaker incubator (New Brunswick, Edison, NJ, USA).1-(2-Thienyl)ethan-1-one (1a) was purchased from Fluka and the water absorbing polymer LP-100 via Sigma, Norway.Analyses Optical rotations were determined using an Optical Activity Ltd.AA-10 automatic polarimeter, concentrations are given in g/100 mL.NMR spectra were recorded in CDCl 3 solutions, using Bruker DPX 300 and 400 instruments, operating at 300 and 400 MHz for 1 H and 75 and 100 MHz for 13 C, respectively.Chemical shifts are in ppm relative to TMS and coupling constants in Hz.Enantiomeric ratios, E were calculated on the basis of measurement of enantiomeric excess of both substrate (ees) and product (eep) at several degrees of conversion, using the computer program E & K calculator version 2.03 13 .Mass spectra were recorded on a Fison 8065/Fison TRIO 1000 GC-MS (EI, 70eV) system with a quadrupole mass filter, equipped with a DB-1701 column from J&W Scientific (25 m, 0.25 mm ID, 0.25 µm film thickness).Helium was used as carrier gas.Chiral analyses were performed using a Varian 3400 or 3800 gas chromatograph from Varian Instrument Group, Walnut Creek, California, USA.The columns were either Chiraldex G-TA, 10 m, 0.25 mm ID, 0.25 µm film thickness from Astec, Whippany, N.J., USA or CP-Chirasil-Dex CB, 25 m, 0.25 mm ID, 0.25 µm film thickness from Chrompack, Chrompack Norge A.S., Sandvika, Norway.The gas chromatographs were pressure regulated, carrier gas was hydrogen (Hydrogen 5.0, purity: 99.999%) with an outlet pressure of 3 bar (7 psi).Injection was performed in split mode at 200 o C, split ratios 60.Detection was done using FID detectors at 200 o C with air (300 mL/min) and hydrogen (30 mL/min) as flame gases.Integration was performed with Varian Star 4.0/4.5.The chromatographic details are given in Table 2.

Fermenting cells-LP-100
Geotrichum candidum was grown in 1L standard medium and filtered after 36 h to give 25g of wet cells. 12The cells were re-suspended in distilled water (100 mL) and the water-absorbing polymer (LP-100, 15 g) was added.Each reaction was performed in hexane (6 mL) by adding the immobilised cells (4 g) and substrate (20 mg).Small scale transesterifications Substrate alcohol (2.2 x 10 -4 mole) was dissolved in solvent (3 mL), vinyl butanoate (5 equivalents) was added and the reaction was started by adding immobilised CALB (10 mg) to the reaction mixture at 30 o C. Chiral GLC analysis gave the enantiomeric excess of substrate (ees) and product (eep) from which conversion, c, was calculated, c = ees / ( ees + eep).In control experiments without enzyme, no acylation was observed using vinyl butanoate as acyl donor.

Enzymatic reduction in water (small scale)
The ketones 1a-4a (0.013 mmole), NAD+ (7 µmole) and cyclopentanol or 2-propanol (100 µL) were added to a suspension of AGP (20 mg) in MES buffer, (0.1M, pH 7.0, 3 mL) or phosphate buffer (0.1M, pH 7.0, 3 mL).The mixture was shaken at 180 rpm at 30 °C.Small aliquots were withdrawn and extracted with Et 2 O for GLC analyses.When GLC showed maximum yield, the biomass was filtered off and filtrate saturated with NaCl, before extracting with Et 2 O.The combined extracts were dried with MgSO 4 and evaporated.In control experiments without cofactor, no reduction was observed.

Reduction by fermenting cells
Geotrichum candidum was grown for 24 hours in 250 mL standard medium as described elswehere.12The subtrates (100 mg) was added to 1 mL of ethanol, added to the cultures and incubation continued using the same conditions.Progress of the reactions was measured by TLC.Workup and analysis was equal to the reduction by AGP in water.
The racemic secondary alcohols 1b-4b were synthesised from the ketones 1a-4a by NaBH 4reduction under standard conditions, quenching the reaction with NH 4 Cl 14 .