Gas chromatography was performed using GL Science

GC-353 gas chromatographs (GL Science Inc., Tokyo, Japan) equipped with capillary columns (DB-Wax, Agilent Technologies, Santa Clara, CA, USA, 0.25 μm, 0.25 mm × 30 m; TC-1, GL Science Inc., 0.25 μm, 0.25 mm × 30 m; CP-Chirasil-DEX CB, Varian Inc., Lake Forest, CA, USA, 0.25 μm, 0.25 mm × 25 m; Gamma DEX 225, Sigma-Aldrich Co., St. Louise, MO, USA, 0.25 μm, 0.25 mm × 30 m). Ethyl pyruvate (Figure 1, 1a), diatomaceous earth (granular), polypepton, L-alanine, and Daigo’s artificial seawater SP were purchased from Wako Pure Chemical Industries Ltd., Osaka, Japan. Difco^™ soluble starch and Bacto^™ yeast extract were purchased from Becton, Dickinson and Co., Franklin Lakes, NJ, USA. Ethyl lactate (2a), ethyl 3-methyl- 2-oxobutyrate (1f), ethyl 2-oxo-4-phenylbutyrate (1h), ethyl 2-hydroxy-4-phenylbutyrate (2h), and beef extract were purchased from Sigma-Aldrich. Ethyl benzoylformate (1g) and ethyl mandelate (2g) were obtained from Tokyo Chemical Industry, Co. Ltd., Tokyo, Japan. Ethyl 2-oxobutanoate (1b), ethyl 2-oxopentanoate (1c), ethyl 2-oxohexanoate (1d), ethyl 2-oxoheptanoate (1e), and α-hydroxy esters (2b-f) were prepared according to procedures in the literature [24].

Micromonospora sp. NBRC107092, 107093, 107094, 107095, 107096, 107097, 107098, 107099 (from marine sponge), Streptomyces hygroscopicus subsp. Hygroscopicus NBRC100580 (from marine fish), Streptomyces marinus NBRC105047 (from marine sponge), Streptomyces tateyamensis NBRC105048 (from marine sponge), Streptomyces haliclonae NBRC105049 (from marine sponge), and Streptomyces sp. NBRC105896 (from marine sponge) were purchased from the National Institute of Technology and Evaluation, Biological Resource Center, Japan (NBRC). These marine-derived actinomycetes were each maintained in NBRC-recommended medium (227, 228, 1076, and 1077) solidified with 1.5% agar.

The 227 medium (ISP medium No. 2) comprised 4.0 g of Bacto^™ yeast extract, 10.0 g of Bacto^™ malt extract, and 4.0 g of D-glucose per 1 liter of distilled water (pH 7.3). The 228 medium comprised 1.0 g of Bacto^™ yeast extract, 1.0 g of beef extract, 2.0 g of NZ amine, type A, and 10.0 g of D-glucose per 1 liter of distilled water (pH 7.3). The 325 medium (75% marine water) is comprised of 10.0 g of polypepton, 2.0 g of Bacto^™ yeast extract, 0.5 g of MgSO₄·7H₂O, and 27.0 g of Daigo’s artificial seawater SP per 1 liter of distilled water (pH 7.3). The 1076 medium (ISP No. 2, 50% marine water) comprised 4.0 g of Bacto^™ yeast extract, 10.0 g of Bacto^™ malt extract, 4.0 g of D-glucose, and 18.0 g of Daigo’s artificial seawater SP per 1 liter of distilled water (pH 7.2). The 1077 medium (50% marine water) comprised 2.0 g of Bacto^™ yeast extract, 10.0 g of Difco^™ soluble starch, and 18.0 g of Daigo’s artificial seawater SP per 1 liter of distilled water (pH 7.2). The 1076-25% medium (ISP No. 2, 25% marine water) comprised 4.0 g of Bacto^™ yeast extract, 10.0 g of Bacto^™ malt extract, 4.0 g of D-glucose, and 9.0 g Daigo’s artificial seawater SP per 1 liter of distilled water (pH 7.3). The P-1076-25% medium comprised 10.0 g of polypepton, 4.0 g of Bacto^™ yeast extract, 10.0 g of Bacto^™ malt extract, 4.0 g of D-glucose, and 9.0 g of Daigo’s artificial seawater SP per 1 liter of distilled water (pH 7.3).

Micromonospora strains were grown in 227, 1076, 1076-25%, and P-1076-25% media (500 mL) for 3 days at 25˚C with aerobic shaking in baffled 2-L flasks in the dark. The Streptomyces strains were grown in 325, 1076, 1076-25%, and P-1076-25% media (500 mL) for 5 days at 25˚C with aerobic shaking in baffled 2-L flasks in the dark. The marine actinomycete cells were harvested by filtration on filter paper (Whatman, No. 4) in vacuo and washed with saline (0.85% NaCl aq.).

Saline-washed wet cells (0.5 g, dry weight approximately 0.2 g) were resuspended in a large test tube (f 30 mm × 200 mm) containing 20 mL of saline. The substrate (0.15 mmol; corresponding substrate concentration was 7.5 mM) and additive (5.0 mmol) were added, and the reaction mixture was incubated aerobically (reciprocating shaking at 120 min^–1) at 25˚C. A portion (0.5 mL) of the mixture was filtered using a short diatomaceous earth column (f 10 mm × 30 mm), extracted with diethyl ether (5.0 mL), and then concentrated under reduced pressure.

Conversions of the produced alcohols (Figure 1, 2a-h) was measured using a GLC with a DB-WAX capillary column (100 kPa He, 110˚C; 1a, 3.78 min; 2a, 4.75 min; 1b, 4.73 min; 2b, 5.92 min; 1f, 4.54 min; 2f, 6.41 min; 120˚C; 1c, 4.84 min; 2c, 6.45 min; 150˚C, 1d, 3.83 min; 2d, 4.68 min; 1e, 4.78 min; 2e, 6.07 min; 180˚C, 1g, 9.01 min; and 2g, 12.08 min) or a TC-1 capillary column (100 kPa He, 140˚C; 1h, 10.02 min; and 2h, 10.96 min).

The enantiomeric excess (e.e.) of the product was measured using a GLC equipped with an optically active CP-Chirasil-DEX CB (2a-e, 2g-h) or Gamma DEX 225 capillary column (2f). The e.e. was calculated according to the following formula: e.e. (%) = {(R – S)/(R + S)} × 100. These R and S are the respective peak areas on GLC analyses. The absolute configurations of α-hydroxy esters (2a-h) were identified by comparing their retention times from the GLC analyses with those of authentic samples [24].

To search for a suitable medium for the liquid culture, the amount of wet cells obtained by cultivating of marine actinomycetes in various culture media was measured. The NBRC-recommended medium for 8 Micromonospora strains tested in this study was the 227 medium. Micromonospora strains cultured in the 227 medium, with the exception of NBRC 107094 and 107095, produced 5.0 g or less of the wet cells (Table 1). To improve the culture rate in the 227 medium, 3 new culture media (1076, 1076-25%, and P-1076-25% medium) containing marine water were tested for efficacy (1076-25% and P-1076-25% media were newly designed in this study).

In the cultivation of these 8 strains using the 1076 medium including marine water (50% concentration), although over 6.0 g of wet cells of the NBRC107096 strain were obtained, 2.0 g or less of wet cells were obtained for the NBRC107093, 107095, 107097 and 107098 strains. On the other hand, we obtained over 7.0 g of wetcells when the NBRC107096 and 107097 strains were cultured in the 1076-25% medium (25% marine water). Furthermore, to obtain more wet cells, Micromonospora strains were cultured in P-1076-25% medium including polypepton (25% marine water); however, 5.0 g of wet cells was not obtained.

In a similar manner, a suitable medium for the liquid culture of 5 marine-derived Streptomyces strains was determined (Table 2). We found that the P-1076-25% medium is most suitable for the liquid culture of the Streptomyces strains; in particular, the culture of Streptomyces sp. and S. tateyamensis yielded over 5.0 g of wet cells. In contrast, Streptomyces marinus did not show good growth in any culture medium (less than 1.0 g of wet cells was obtained).

Therefore, we investigated the possibility that 2 Micromonospora strains (NBRC107096 and 107097) and 2 Streptomyces strains (Streptomyces tateyamensis NBRC105048 and Streptomyces sp. NBRC105896) can act as biocatalysts for the asymmetric reduction of carbonyl compounds.

3.2. Reduction of α-Keto Esters by Marine-Derived Micromonospora Strains Two Micromonospora strains (NBRC107096 and 107097) were tested for their ability to reduce α-keto esters (Figure 1). The results of the α-keto ester reductions are summarized in Table 3. We found that both Micromonospora strains reduced aliphatic and aromatic α-keto esters (1a-h) to the corresponding alcohols (2a-h). The reduction of α-keto esters by the NBRC107097 strain exhibited a higher conversion ratio than the reduction by the NBRC107096 strain; in particular, the NBRC107097

Table 1. The cultivation of marine-derived Micromonospora strains in various culture media.

Table 2. The cultivation of marine-derived Streptomyces strains in various culture media.

Table 3. The reduction of α-keto esters (1a-h) to the corresponding alcohols (2a-h) by 4 marine-derived actinomycetes.

¹Substrate (0.15 mmol) and 0.85% NaCl aq. (20 ml) were added to the wet cells (0.5 g) cultured in 1076-25% medium, and the reaction mixture was incubated aerobically (reciprocating shaking at 120 min^–1) at 25˚C for 48 hrs. ²Substrate (0.15 mmol) and 0.85% NaCl aq. (20 ml) were added to the wet cells (0.5 g) cultured in P-1076-25% medium, and the reaction mixture was incubated aerobically (reciprocating shaking at 120 min⁻¹) at 25˚C for 48 hrs. ³Conversion was measured by a GLC analysis. ⁴Enantiomeric excess (e.e.) and absolute configuration (R/S) were determined by GLC analyses with optically active capillary columns.

strain had high reducing activity for substrates that had a short alkyl chain.

However, except for ethyl pyruvate (1a), the stereoselectivity of the produced alcohols was not high (13% - 80% e.e.). In the microbial reduction of carbonyl compounds using bakers’ yeast or filamentous fungi (eukaryote), it is well known that the introduction of small organic molecules or metal ions will increase the stereoselectivity of the produced alcohols [25-27]. In contrast, in the reduction using actinomycetes (prokaryote), there are several reports that the addition of amino acids or sugars is effective in improving the conversion rate and stereoselectivity of products [12-14,23].

Therefore, the effect of additives on the reduction of keto esters using marine actinomycetes was investigated (see Table 4). Among various additives (e.g., sugars such as glucose, fructose, sucrose, and maltose or amino acids such as L-alanine, L-glycine, L-glutamate, and Laspartate), the introduction of L-glutamate or sucrose remarkably increased the conversion ratio of the reduction. In particular, the reduction by the NBRC107096 strain in the presence of L-glutamate gave the corresponding α-hydroxy ester with >99% conversion ratio for all substrates. It appears that the increase in reduced nicotinamide-adenine dinucleotide (possibly NADPH) through the oxidative degradation of L-glutamate accelerates the reduction of α-keto esters to the corresponding alcohols.

Two Streptomyces strains (NBRC105048 and 105896) were tested for their ability to reduce α-keto esters. As shown in Table 3, we found that α-keto esters (1a-h) were reduced by both Streptomyces strains to the corresponding hydroxy esters (2a-h). The NBRC105,896 strain had high reducing activity for α-keto esters when compared with the NBRC105,048 strain and showed excellent conversion ratios (>99% in 2a-d and 2f-h). Fur-

Table 4. Effects of additives on the reduction of α-keto esters (1a-h) with marine-derived Micromonospora strains¹.

¹Substrate (0.15 mmol), additive (5.0 mmol), and 0.85% NaCl aq. (20 ml) were added to the wet cells (0.5 g) cultured in 1076-25% medium, and the reaction mixture was incubated aerobically (reciprocating shaking at 120 min^–1) at 25˚C for 48 hrs. ²Conversion was measured by a GLC analysis. ³Enantiomeric excess (e.e.) and absolute configuration (R/S) were determined by GLC analyses with optically active capillary columns.

Table 5. Effects of additives on the reduction of α-keto esters (1a-h) with marine-derived Streptomyces strains¹.

¹Substrate (0.15 mmol), additive (5.0 mmol), and 0.85% NaCl aq. (20 ml) were added to the wet cells (0.5 g) cultured in P-1076-25% medium, and the reaction mixture was incubated aerobically (reciprocating shaking at 120 min^–1) at 25˚C for 48 hrs. ²Conversion was measured by a GLC analysis. ³Enantiomeric excess (e.e.) and absolute configuration (R/S) were determined by GLC analyses with optically active capillary columns.

thermore, the reduction of 1a-c using the NBRC105,896 strain gave the corresponding (S)-hydroxy esters exclusively (>99% e.e.).

An additive was introduced to the reaction mixture to improve the conversion ratio and stereoselectivity (see Table 5). As a result, the reduction of substrates using the NBRC105,048 strain in the presence of L-alanine or sucrose produced the corresponding alcohols with high conversion ratios (L-alanine, >99%, in 2a-d, 2f; sucrose, >99%, in 2a-g). In particular, the introduction of Lalanine improved not only the conversion rate but also the stereoselectivity of the products (>99% e.e. in 2a-f). The effects of other additives (L-glycine, L-glutamateL-aspartate glucose, fructose, and maltose) were tested; however, the stereoselectivity of the produced alcohols did not increase (data not shown). In contrast, in the reduction by the NBRC105,896 strain, the conversion ratio was decreased by the introduction of the additive. This decrease in the conversion ratio was not expected.