Optical resolution of RS-(±)-mandelic acid by ... - Springer Link

1 downloads 0 Views 63KB Size Report
Byung-Yong Kim, Ki-Chul Hwang, Hee-Sang Song, Namhyun Chung & Won-Gi Bang. ∗. Department of Agricultural Chemistry, Korea University, Seoul, South ...
Biotechnology Letters 22: 1871–1875, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.

1871

Optical resolution of RS-(±)-mandelic acid by Pseudomonas sp. Byung-Yong Kim, Ki-Chul Hwang, Hee-Sang Song, Namhyun Chung & Won-Gi Bang∗ Department of Agricultural Chemistry, Korea University, Seoul, South Korea 136-701 ∗ Author for correspondence (Fax: +82-2-923-8183; E-mail: [email protected]) Received 31 August 2000; Accepted 21 September 2000

Key words: fed-batch fermentation, mandelic acid, optical resolution, Pseudomonas sp.

Abstract For the optical resolution of R-(−)-mandelic acid from (±)-mandelic acid, Pseudomonas sp. MA02, which assimilated S-(+)-mandelic acid as carbon and energy source, was isolated from soil. Using the fed-batch culture under optimal condition, R-(−)-mandelic acid was accumulated up to the maximum theoretical yield of 50% (30 g l−1 ) and entiomeric excess of 99.4%.

Introduction R-(−)-Mandelic acid is used as a precursor for semisynthetic penicillin and cephalosporin (Fulenmeier et al. 1976, Yamazaki & Kajiwara 1988) plus many other pharmaceuticals (Mills et al. 1983, Yamazaki & Kajiwara 1988). There have been a number of studies to prepare R-(−)-mandelic acid (Oda et al. 1992, Ohta & Miyamoto 1992, Shimao et al. 1996, Takahashi et al. 1995, Yamamoto et al. 1991, Yamazaki & Kajiwara 1988). However, the practical production of R-(−)-mandelic acid was often hampered by low optical purity and/or yield, complex processes et cetera. The growing realization of the different properties of stereoisomers in biological systems has led to a shift in the life science industry towards stereochemically pure compounds. One of the most used techniques for the development of chiral compounds involves kinetic (biocatalytic) resolutions, and more recently dynamic kinetic resolutions (Gihani & Williams 1999). However, classical fermentation may be economical under certain circumstances. We have looked for a simpler method for the large scale preparation of R(−)-mandelic acid. In this study, we have isolated a strain that selectively assimilates S-(+)-mandelic acid from the racemate (±)-mandelic acid that has been produced by chemical synthesis, thus leaving R-(−)mandelic acid in culture broth. The method has the

theoretical yield of 50% like classical kinetic resolution (Gihani & Williams 1999). However, the method may have several advantages over other processes mentioned above: the biomass can be readily removed or the racemate added sequentially to improve the final concentration of R-(−)-mandelic acid. Additionally, we have achieved the maximum theoretical yield of 50% with enantiomeric excess of 99.4%.

Materials and methods Chemicals (±)-, R-(−)-, and S-(+)-mandelic acid were obtained from Sigma Chemical Co. (St. Louis, USA). All other reagents were of analytical reagent grade. Strain selection A portion of soil was sampled from a creek located in Ssangmun, Seoul. The soil was suspended in sterile 0.9% NaCl solution and filtered through Whatman No. 2. Bacterial strains were isolated by enrichment culture using an inorganic salts solution amended with 1% (±)-mandelic acid and inoculated with the filtrate. Four serial transfers were made for 4 days. Serial dilutions were made from the final culture, and 0.2 ml portions of the appropriate dilution were smeared on agar plates containing 1% (±)-mandelic acid, 0.2%

1872 (NH4 )2 SO4 , 0.1% KH2 PO4 , 0.1% MgSO4 · 7H2 O and 0.02% yeast extract. The plates were incubated at 30 ◦ C for 48 h. Each bacterial colony was picked and streaked on agar plates containing 0.5% R-(−)- or S-(+)-mandelic acid plus the salts mentioned above. A number of developed single colonies were cultivated in the medium containing 0.5% R-(−)- or S-(+)-mandelic acid for a day. Optical resolution of (±)-mandelic acid by cultivation The ability of strains to use R-(−)- or S-(+)-mandelic acid as sole carbon source was tested by measuring both the growth turbidometrically (at 600 nm) of cultivates and concentration of R-(−)- or S-(+)-mandelic acid, using HPLC. For the removal of S-(+)-mandelic acid from the racemate by assimilation of isolated strain, a media containing 2% (±)-mandelic acid, 0.1% KH2 PO4 , 0.16% NH4 Cl, 0.1% MgSO4 · 7H2 O, and 0.1% corn steep liquor (pH 7.5) was employed unless otherwise stated. With S-(+)-mandelic acid being assimilated by isolated strain, the enantiomeric excess (e.e.) (%) [ = (R − S)/(R + S) × 100] increased. Other experimental conditions are described in the figure legends and tables. Analysis of mandelic acid Each cultivate was centrifuged and passed through 0.22 µm syringe filters (Micron Separations, Westboro, MA, USA) to remove particulate matter prior to HPLC analysis. The samples were analysed using HPLC (Spectra System P1000, Thermo Separation Products, San Jose, CA, USA) fitted with a Chirex Dpenicillamine column (300 × 4.6 mm; Phenomenex, Torrance, CA, USA). The solvent system was 2 mM CuSO4 /acetonitrile (85:15 v/v). The flow rate and column temperature were 1 ml and 40 ◦ C, respectively. The compounds (benzylformic acid, R-(−)- or S-(+)mandelic acid) were detected by absorbance at 254 nm using variable wavelength UV detector (Spectra 100, Thermo Separation Products). The residence times of benzylformic acid, R-(−)-, and S-(+)- mandelic acid were 29.2, 60.5 and 80 min, respectively. Characterization of isolated strain Isolated strain was grown on Luria-Bertani agar plate and subjected to Gram staining. Morphology of isolated strain was observed with transmittance electron microscope (ZEISS EM109, FGR) after staining with 2% (w/v) phosphotungstic acid (pH 6.8). Isolated

strain was also identified by several morphological, biochemical, and cultural methods (Holtz et al. 1994).

Results and discussion Strain selection According to the ability to grow on R-(−)- or S-(+)mandelic acid, 17 isolates were chosen. Some of the strains grew equally well on both enantiomers or others a little better on S form than R form, suggesting that most strains can use both enantiomers almost equally well as carbon and energy source. After 24 h cultivation at 30 ◦ C, optical density ranged from 0.28 to 0.86 except strain MA02. The strain MA02 showed relatively much better growth on S form (O.D.600 nm = 0.22) than R form (O.D.600 nm = 0.003). Strain MA02 was chosen as a microbe to selectively remove S form from the racemate. Strain MA02 had a rod shape and flagella and was Gram-negative. That bacterium was identified by the findings above and other tests as Pseudomonas sp. Effect of substrate concentration on the resolution of DL-mandelic acid In the course of metabolic removal of S-(+)-mandelic acid, the change in concentration of R-(−)-, S-(+)mandelic acid, or other intermediates was investigated. After 12 h cultivation, the concentration of S form decreased and that of R form stayed the same. However, benzylformic acid appeared at 12 h and then significantly decreased at 48 h with further decrease in the concentration of S-(+)-mandelic acid (data not shown), suggesting that benzylformic acid was a major intermediate during the assimilation of S-(+)-mandelic acid. The benzylformic acid was not detectable with more cultivation. This result also indicates that the isolated strain MA02 may not have mandelic acid racemase but S-(+)-mandelic acid dehydrogenase in the mandelate pathway (Hegeman et al. 1970). Using transposon mutagenesis, Shimao et al. (1996) have constructed a strain of Pseudomonas that is lacking mandelic acid racemase in the mandelate pathway. The strain had a capability to selectively assimilate S-(+)-mandelic acid. Thus, the isolation of Pseudomonas sp. MA02 in the present study may show the diversity of microbial flora in nature. The concentrations of the racemate in the culture media were varied from 1 to 4% to observe the effect of substrate concentration on the metabolic removal of

1873

Fig. 1. Effect of substrate concentration on the resolution of (±)-mandelic acid by strain MA02. Cultivations were carried out for 5 days at 30 ◦ C in the medium containing 0.2% (NH4 )2 SO4 , 0.1% KH2 PO4 , 0.1% MgSO4 · 7H2 O and 0.02% yeast extract (pH 7.0), and various concentrations of (±)-mandelic acid.

Fig. 2. Effect of carbon/nitrogen ratio on the resolution of (±)-mandelic acid by strain MA02. Cultivations were carried out for 5 days at 30 ◦ C in the medium containing 2% (±)-mandelic acid, 0.1% KH2 PO4 , 0.1% MgSO4 · 7H2 O, 0.02% yeast extract, and various concentrations of NH4 Cl at the indicated carbon/nitrogen ratio (pH 7.0).

Table 1. Effect of inorganic nitrogens on the resolution of (±)-mandelic acid by strain MA02a . Nitrogen source (C/N = 30)

Growth (O.D.600 nm )

Acid e.e. (%)b

NH4 Cl (NH4 )2 SO4 (NH4 )2 HPO4 NH4 NO3 NaNO3 KNO2

3.56 3.51 3.12 3.11 2.88 0.18

95 94 84 80 34