Efficient Synthesis of Methyl 3-Acetoxypropionate by a Newly Identified Baeyer-Villiger Monooxygenase

BVMOs are emerging as a green alternative to traditional oxidants in the BV oxidation of ketones. Although many BVMOs are discovered and used in organic synthesis, few are really applied in industry, especially in the case of aliphatic ketones. Herein, a highly soluble and relatively stable monooxygenase from Rhodococcus pyridinivorans (BVMO_Rp) was identified with high activity and excellent regioselectivity toward most aliphatic ketones. BVMO_Rp possesses unusually high substrate loading during the catalysis of the oxidation of biobased methyl levulinate to 3-hydroxypropionic acid derivatives. This study indicates that the synthesis of 3-hydroxypropionate from readily available biobased levulinate by BVMO_Rp-catalyzed oxidation holds great promise to replace traditional fermentation.

ABSTRACT

Baeyer-Villiger monooxygenases (BVMOs) are an emerging class of promising biocatalysts for the oxidation of ketones to prepare corresponding esters or lactones. Although many BVMOs have been reported, the development of highly efficient enzymes for use in industrial applications is desirable. In this work, we identified a BVMO from Rhodococcus pyridinivorans (BVMO_Rp) with a high affinity toward aliphatic methyl ketones (K_m < 3.0 μM). The enzyme was highly soluble and relatively stable, with a half-life of 23 h at 30°C and pH 7.5. The most effective substrate discovered so far is 2-hexanone (k_cat = 2.1 s⁻¹; K_m = 1.5 μM). Furthermore, BVMO_Rp exhibited excellent regioselectivity toward most aliphatic ketones, preferentially forming typical (i.e., normal) products. Using the newly identified BVMO_Rp as the catalyst, a high concentration (26.0 g/liter; 200 mM) of methyl levulinate was completely converted to methyl 3-acetoxypropionate after 4 h, with a space-time yield of 5.4 g liter⁻¹ h⁻¹. Thus, BVMO_Rp is a promising biocatalyst for the synthesis of 3-hydroxypropionate from readily available biobased levulinate to replace the conventional fermentation.

IMPORTANCE BVMOs are emerging as a green alternative to traditional oxidants in the BV oxidation of ketones. Although many BVMOs are discovered and used in organic synthesis, few are really applied in industry, especially in the case of aliphatic ketones. Herein, a highly soluble and relatively stable monooxygenase from Rhodococcus pyridinivorans (BVMO_Rp) was identified with high activity and excellent regioselectivity toward most aliphatic ketones. BVMO_Rp possesses unusually high substrate loading during the catalysis of the oxidation of biobased methyl levulinate to 3-hydroxypropionic acid derivatives. This study indicates that the synthesis of 3-hydroxypropionate from readily available biobased levulinate by BVMO_Rp-catalyzed oxidation holds great promise to replace traditional fermentation.

INTRODUCTION

The Baeyer-Villiger (BV) reaction, inserting an oxygen atom between the carbon atom of a carbonyl and the adjacent carbon atom to form an ester or lactone, is one of the most widely used oxidation reactions in organic synthesis. Although BV oxidation was reported by Baeyer and Villiger more than 100 years ago, it is still under continual development. BV monooxygenases (BVMOs) are emerging as a green alternative to using traditional oxidants in the BV oxidation of ketones (1–3). BVMOs can typically be categorized into three types, I, II, and O (3). Among them, the NADPH-dependent and flavin adenine dinucleotide (FAD)-dependent type I enzymes are the most frequently reported, which have two conserved sequence motifs (FxGxxxHxxxWP/D and G/AGxWxxxxF/YPG/MxxxD) (4, 5).

Due to their excellent chemo-, regio-, and enantioselectivity, as well as great potential for the synthesis of value-added chemicals, BVMOs are considered alternative green oxidative biocatalysts. The cyclohexanone monooxygenase from Acinetobacter calcoaceticus NCIMB 9871 (CHMO_Acineto) was the first reported BVMO, and this enzyme is able to selectively oxidize a wide range of cyclic ketones to the corresponding lactones (6). Numerous naturally occurring BVMOs with various substrate specificities have since been identified, some of which can catalyze the conversion of aromatic ketones to aromatic esters, such as the thermostable phenylacetone monooxygenase (PAMO) from Thermobifida fusca and 4-hydroxyacetophenone monooxygenases from Pseudomonas fluorescens ACB and Pseudomonas putida JD1 (7–9). Some enzymes can oxidize aliphatic ketones to aliphatic esters, such as the BVMO from Pseudomonas fluorescens DSM 50106 (BVMO_Pf), the BVMO from Pseudomonas putida KT2440 (BVMO_Pp), and the BVMO from Aspergillus flavus NRRL3357 (BVMO_AFL838) (10–12). Also, several stable BVMOs have been identified (13, 14), and protein engineering has been employed to improve the properties of BVMOs (15), including expanding the substrate scope (16–19), enhancing regioselectivity (20–22) or enantioselectivity (23–25), stimulating sulfoxidation activity (26–28), changing the cofactor dependency (29–31), and increasing stability (32–34).

Recently, using the developed BVMOs as an attractive alternative to traditional chemical oxidants and/or in combination with other enzymes, several useful chemicals have been successfully synthesized from readily available starting materials. Park and coworkers constructed an elegant multienzymatic cascade reaction by combining BVMO_Pp or BVMO_Pf with other enzymes to cleave renewable fatty acids or plant oils into α,ω-dicarboxylic acids and ω-hydroxycarboxylic acids of varied carbon chain lengths. In this case, varying the position of the double bond in unsaturated fatty acids and the regioselectivity of hydratases and BVMOs resulted in the formation of diverse monomeric chemicals (35, 36). Fraaije and coworkers first identified several BVMOs that catalyze the partial monooxygenation of 2-butanone into the abnormal product methyl propanoate, an important precursor for the synthesis of polymethyl methacrylates (37). The research groups of Kroutil and Bornscheuer also developed two multienzymatic cascades for the synthesis of bulk chemicals 6-aminohexanoic acid and poly-ε-caprolactone, respectively, starting from readily available cyclohexanol (38, 39). In the first two steps, cyclohexanol (200 mM) is oxidized by an alcohol dehydrogenase to cyclohexanone, followed by BV oxidation by a variant of CHMO_Acineto to ε-caprolactone with almost 100% yield. Fink and Mihovilovic synthesized 3-hydroxypropionic acid (3-HP) derivatives from inexpensive levulinic acid derivatives with a space-time yield (STY) of 2.4 g liter⁻¹ day⁻¹ using a CPMO variant from Comamonas sp. strain NCIMB 9872 (40). Oppermann’s group employed BVMO_AFL838 to convert 50 mM octyl aldehyde for the synthesis of alkyl formates that are widely used in the flavor and fragrance industries (41). Mihovilovic’s group utilized a biocatalytic multistep cascade involving CHMO_Acineto to produce chiral carvolactone starting from limonene (42).

Although numerous BVMOs have been reported and used in organic synthesis, new enzymes with potential for industrial applications are in high demand. In the present study, a highly soluble BVMO from Rhodococcus pyridinivorans DSM 44555 (BVMO_Rp; NCBI RefSeq accession number WP_060655096) was cloned and overexpressed in Escherichia coli BL21(DE3). BVMO_Rp displayed excellent regioselectivity for linear aliphatic ketones and their derivatives. Using crude cell extract containing BVMO_Rp, 200 mM methyl levulinate was completely converted into 3-hydroxypropionate with an STY of 5.4 g liter⁻¹ h⁻¹, demonstrating great potential for the industrial production of 3-HP from renewable biobased levulinic acid.

RESULTS

Screening of BVMOs.

Seven putative BVMO genes from R. pyridinivorans DSM 44555 were cloned and heterologously expressed in E. coli BL21(DE3). In order to identify the best candidate enzyme, these enzymes were tested for their ability to oxidize several aliphatic ketones. The results showed that BVMO_Rp (NCBI RefSeq accession number WP_060655096) was the most effective catalyst for the conversion of all substrates (Table 1) and was therefore chosen for further study.

TABLE 1.

Conversion of ketones by various BVMOs^a

Enzyme	NCBI RefSeq or GenBank accession no.	Conversion (%) for:
		2-Octanone (substrate 3)	3-Octanone (substrate 4)	Methyl levulinate (substrate 12)
BVMORp	WP_060655096	>99	98	>99
BVMO1	WP_064061488	95	70	66
BVMO2	WP_006550557	99	41	32
BVMO3	WP_064060773	0.3	0.3	0.4
BVMO4	WP_064061141	2.0	3.0	0.6
BVMO5	WP_006554318	6.0	5.0	<0.1
BVMO6	SEB82458	0.6	0.3	0.3

^aReaction conditions were 100 mM substrate, 150 mM glucose, 0.5 mg glucose dehydrogenase, 0.5 mM NADP⁺, 0.1 mM FAD, cell extract from 10 mg cells (wet weight), and 400 mM potassium phosphate buffer (KPB [pH 7.5]), with 0.5 ml total volume at 30°C and 1,000 rpm for 4 h.

Characterization of BVMO_Rp.

SDS-PAGE showed that BVMO_Rp was present almost exclusively in the soluble fraction of the induced cell lysate, yielding a prominent band at 60 kDa (Fig. 1). BVMO_Rp was active between pH 6.0 and 10.3, retaining over 40% of maximal activity at pH 8.0 to 9.0, with an optimal pH of 8.0 in potassium phosphate buffer (Fig. 2a). The optimum pH for stability was 7.0 to 7.5, and ∼85% of the initial activity was retained after incubation for 12 h at 30°C, compared with only 55% at pH 8.0 (Fig. 2b). Although the optimal temperature for activity in the 10-min assay was between 35°C and 40°C (Fig. 2c), BVMO_Rp was relatively stable at 30°C, with a half-life of 23 h, compared with 10 h at 35°C and 6.1 min at 40°C (Fig. 2d). Therefore, pH 7.5 and 30°C were chosen for use in the subsequent BV oxidation catalyzed by BVMO_Rp.

FIG 1.

SDS-PAGE analysis of purified recombinant BVMO_Rp. Lane 1, protein molecular weight markers; lane 2, supernatant of induced cells; lane 3, flowthrough; lanes 4 to 8, elution fractions.

FIG 2.

Effects of pH and temperature on the activity and stability of BVMO_Rp. (a) Effect of pH on activity determined by assaying the activity at 30°C in 100 mM NaAc-HAc buffer (pH 5.5 to 6.0 [●]), potassium phosphate buffer (pH 6.0 to 8.0 [▲]), and Gly-NaOH (pH 8.0 to 11.0 [■]). (b) Effect of pH on stability determined by assaying the residual activity of BVMO_Rp after 12 h at 30°C and different pH values in 100 mM potassium phosphate buffer (▲), and Gly-NaOH (■). (c) Effect of temperature on activity. (d) Effect of temperature (30°C, ▲; 35°C, ■; 40°C, ●) on stability. Residual activity is expressed as a percentage of activity measured at 0 h.

Specific activities of BVMO_Rp toward various ketones.

A variety of aliphatic ketones and their derivatives were used for exploration of the substrate scope of BVMO_Rp (Fig. 3). Activity toward methyl aliphatic ketones (substrates 1, 3, and 6) was the highest among linear aliphatic ketones of equal chain length, and there was a slight decrease with increasing carbon chain length (Fig. 4).

FIG 3.

Selected ketone substrates used for BVMO_Rp oxidation.

FIG 4.

Specific activity of BVMO_Rp toward different ketone substrates.

According to previous literature, the position of the carbonyl group does not significantly affect the reactivity of BVMOs. For instance, BVMO_Pp, BVMO_Pf, and MekA display similar efficiencies toward 2-, 3-, and 4-decanones (10, 11, 43). For octanones, BVMO_Rp also displayed similar specific activities toward 2-, 3-, and 4-octanone (substrates 3 to 5, respectively). Interestingly, for decanones, this enzyme exhibited significantly lower activity toward 3-decanone (substrate 7) than toward other 2-, 4-, and 5-decanones (substrates 6, 8, and 9, respectively) (Fig. 4).

In addition, enzyme activity was also observed toward methyl ketones (substrates 10 to 12) with different substituents at the end of the open chain (Fig. 4). Compared with nonpolar ester groups, polar substituents with carboxyl and hydroxyl groups were less efficient substrates for BVMO_Rp, possibly due to incompatibility with the predominately hydrophobic cavity of the enzyme.

Kinetic parameters for BVMO_Rp with various ketone substrates.

In order to explore how carbonyl group position and carbon chain length influence BVMO_Rp performance, the kinetic parameters K_m and k_cat were determined (Table 2). The K_m for BVMO_Rp was <3.0 μM with aliphatic methyl ketones 1, 3, and 6, comparable to the value observed for MekA (6.0 μM) toward 2-butanone (43).

TABLE 2.

Kinetic parameters for BVMO_Rp with various ketone substrates

Substrate	Km (mM)	kcat (s–1)	kcat/Km (s–1 mM–1)
1	1.5 × 10−3	2.06 ± 0.05	1,372
2	1.4 ± 0.1	1.70 ± 0.04	1.3
3	1.8 × 10−3	1.99 ± 0.02	1,120
4	0.15 ± 0.01	1.72 ± 0.02	11.2
5	0.23 ± 0.01	1.85 ± 0.03	7.9
6	2.3 × 10−3	1.80 ± 0.04	776
7	0.48 ± 0.06	0.045 ± 0.001	0.093
8	0.17 ± 0.02	0.52 ± 0.02	3.1
9	0.16 ± 0.01	0.31 ± 0.01	1.9
12	0.35 ± 0.02	1.52 ± 0.02	4.3

For hexanones and octanones with carbonyl groups at different positions, BVMO_Rp exhibited a similar k_cat but dramatically decreased affinity for 3-ketone substrates 2 and 4 compared with 2-ketone substrates 1 and 3 (Table 2). A similar phenomenon was also observed previously for BVMO_AFL838 (44). Because the enzyme was saturated by substrates at 2.0 mM, it displayed similar specific activities toward 2-, 3-, and 4-octanones (Fig. 4). For decanones, the position of the carbonyl group also had an influence on both K_m and k_cat values. Among all tested decanones (substrates 6 to 9), the K_m value for 3-decanone (substrate 7) was the highest, but the k_cat value was the smallest, which explains the worst reactivity toward 3-decanone.

Regioselectivity of BVMO_Rp with various ketone substrates.

According to typical BV reactions, the so-called “normal” product is produced by introducing an oxygen atom next to the carbonyl group on the most substituted side (products 1a to 9a) (Fig. 3). However, “abnormal” products can be identified during the conversion of several chiral compounds and long aliphatic ketones. Biotransformation products of linear aliphatic ketones obtained using purified BVMO_Rp were analyzed by gas chromatography (GC) or GC-mass spectrometry (GC-MS). As shown in Table 3, all bioconversion results were consistent with the specific activities. All methyl aliphatic ketones (substrates 1, 3, 6, and 12) were completely converted to normal acetates, as expected. For 3-hexanone (substrate 2), 4-ketone (substrates 5 and 8), and 5-decanone (substrate 9), normal products generally accounted for >90% of the products. Furthermore, even for 3-octanone (substrate 4) and 3-decanone (substrate 7), BVMO_Rp displayed moderate regioselectivity (Table 3). These results clearly indicate that BVMO_Rp identified in this work is a highly regioselective enzyme.

TABLE 3.

Regioselective oxidation of aliphatic ketones by purified BVMO_Rp

Substrate	Enzyme loading (mg ml–1)	Conversion (%)	Ratio of products (a:b)
1	0.04	100	100:0
2	0.20	84	97:3
3	0.04	100	100:0
4	0.04	100	83:17
5	0.04	96	95:5
6	0.04	100	100:0
7	0.20	24	73:26
8	0.20	97	99:1
9	0.20	75	92:8
12	0.20	100	100:0

Preparative biosynthesis of methyl 3-acetoxypropionate by BVMO_Rp.

The BV oxidation of levulinate using traditional peroxyacid as a catalyst is not efficient because of the low conversion and undesirable by-product formation. On the contrary, the results in Tables 1 and 3 show that substrate 12 is a moderately good substrate for BVMO_Rp, with a relatively high activity and excellent regioselectivity. A 200 mM concentration of substrate 12, which is a relatively high substrate concentration for BVMOs, was completely converted within 4 h, affording product 12a at 74% yield (2.15 g) as a colorless liquid (Fig. 5). ¹H nuclear magnetic resonance (NMR) (400 MHz, CDCl₃): δ = 4.32 to 4.22 (m, ²H), 3.65 (dd, J = 2.6, 0.9 Hz, ³H), 2.64 to 2.54 (m, ²H), 1.98 (dd, J = 2.7, 1.0 Hz, ³H). The results of ¹H NMR were consistent with those from a previous report (40). The corresponding STY of 5.4 g liter⁻¹ h⁻¹ was based on the isolated product.

FIG 5.

Preparative oxidation of methyl levulinate to 3-hydroxypropionate by BVMO_Rp.

DISCUSSION

In the present study, we cloned and expressed seven putative BVMOs from R. pyridinivorans DSM 44555. These enzyme genes could also be found in the genome of a methyl-ethyl-ketone-degrading bacterium Rhodococcus pyridinivorans SB3094 (45). Among them, BVMO_Rp, BVMO₁, and BVMO₂ exhibited good conversion toward linear aliphatic ketones, and the BVMO_Rp enzyme provided the highest conversion toward all the tested substrates (Table 1). On the basis of kinetic parameters and regioselectivity shown in Tables 2 and 3, we found that 3-ketones are a class of special substrates for BVMO_Rp. BVMO_Rp showed the worst affinity for 3-hexanone (substrate 2) and the lowest k_cat for 3-decanone (substrate 7). The regioselectivities of BVMO_Rp toward 4-octanone (substrate 5) and 4-decanone (substrate 8) were even higher than those toward 3-octanone (substrate 4) and 3-decanone (substrate 7). More investigations, such as molecular dynamics simulation, are required to explore the potential reasons. In addition, the k_cat values of all substrates listed in Table 2 are within the range of 0.31 to 2.06 s⁻¹ (except for 3-decanone), probably due to the limited release rate of NADP⁺, as originally observed for CHMO_Acineto (46) and since reported for 4-hydroxyacetophenone monooxygenase (HAPMO) and PAMO (7, 8).

Compared to most reported BVMOs, BVMO_Rp has the advantage of high-level expression of soluble protein. A considerable amount of BVMOs showed low functional expression in E. coli (10, 11, 43, 47). Many efforts have been devoted to the improvement of protein solubility, such as optimization of the gene expression systems and codon sequences and introduction of the molecular chaperones and fusion tags (10, 11, 47–49). However, seldom have strategies been successfully used for increasing the expression level to a large extent. Luckily, high-level functional expression of BVMO_Rp was achieved without extra optimization. However, the imbalance of high-level functional expression of BVMO_Rp and insufficient intracellular FAD in recombinant E. coli resulted in the existence of inactive enzyme molecules, because the activity of cell extract of BVMO_Rp increased significantly with the addition of 10 μM FAD to the standard assay mixture. Already reported methods in which genes of key enzymes involved in FAD synthesis were overexpressed can be adopted to increase intracellular FAD concentration. Also, BVMO_Rp was relatively stable and has a half-life of 23 h at 30°C. With the exception of several BVMOs from thermophiles, BVMO_Rp was found to be more stable than most reported BVMOs. Therefore, BVMO_Rp is a promising candidate for further engineering efforts to develop a practical BVMO.

In previous studies, the most studied biological approach for 3-HP production is fermentation employing engineered Klebsiella pneumoniae or E. coli (50–53). For example, 3-HP has been produced in the highest titer of 83.8 g/liter and an STY of 28 g liter⁻¹ day⁻¹ in K. pneumoniae (51). Fink and Mihovilovic (40) proposed an alternative approach for the production of 3-hydroxypropionates, with an STY of 2.4 g liter⁻¹ day⁻¹, starting from inexpensive renewable levulinic acid derivatives by Baeyer-Villiger oxidation. More efficiently, methyl 3-acetoxypropionate herein was synthesized by BVMO_Rp-catalyzed oxidation of 200 mM methyl levulinate with an STY of 5.4 g liter⁻¹ h⁻¹. This represents the highest substrate loading reported to date for the BVMO_Rp-catalyzed oxidation of methyl levulinate. Moreover, this substrate, methyl levulinate, is a derivative of renewable levulinic acid from biomass, and the product can be hydrolyzed to 3-HP, a well-known platform compound. Therefore, BVMO_Rp-catalyzed oxidation holds great promise for industrial-scale production of 3-hydroxypropionate from readily available biobased levulinic acid to replace the traditional fermentation.

MATERIALS AND METHODS

Chemicals and materials.

All chemicals and reagents were purchased from authentic suppliers, were of reagent grade or higher, and were used without further purification, unless otherwise specified. Tryptone and yeast extract were obtained from Oxoid (Hampshire, UK). Primers were synthesized by Shanghai Generay Biotech Co., and DNA polymerases, restriction enzymes, and T4 DNA ligase were from TaKaRa (Dalian, China). The 5-ml HisTrap FF crude column was supplied by GE Healthcare. ¹H NMR analysis was conducted on a Bruker Avance 400 MHz spectrometer, and chemical shifts (δ) are reported in parts per million (ppm), while coupling constants (J) are reported in Hz.

Gene cloning, expression, and protein purification.

The gene encoding BVMO_Rp was amplified by PCR using genomic DNA isolated from R. pyridinivorans as the template and primers 5′-CCCAAGCTTGCATGACATCAACCCATCTCCCTGAGCACGC-3′ (forward) and 5′-CCGCTCGAGTCACGATACCCGCAGTCCGTCGGTGTATC-3′ (reverse). The PCR product was purified using a PCR product purification kit, followed by digestion with the HindIII and XhoI restriction enzymes. The restriction fragment was inserted into the pET-28a(+) expression vector digested by the same enzymes, resulting in the pBVMO_Rp plasmid. After verification by DNA sequencing, the pBVMO_Rp plasmid was transformed into E. coli BL21(DE3) cells prior to protein expression experiments.

Recombinant E. coli BL21(DE3) cells were cultured at 37°C with shaking at 180 rpm in Luria-Bertani (LB) medium containing 50 mg/liter kanamycin until the absorbance at 600 nm (OD₆₀₀) reached 0.6 to 0.8. Protein expression was induced by the addition of 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG), and culturing was continued at 16°C and 180 rpm for 24 h. Cells were harvested by centrifugation (4°C, 7,000 × g, 10 min), resuspended in binding buffer (20 mM phosphate, 500 mM NaCl, 10 mM imidazole [pH 7.4]), and disrupted by ultrasonication. The cell lysate was centrifuged for 30 min at 4°C at 18,000 × g. The supernatant was loaded onto a 5-ml HisTrap FF column preequilibrated with binding buffer and washed with buffer containing 20 mM phosphate and 500 mM NaCl (pH 7.4), accompanied by stepwise addition of imidazole at 10 mM, 60 mM, 160 mM, 310 mM, and 500 mM. After analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), elution fractions containing high-purity BVMO_Rp were collected and buffer exchanged into 20 mM phosphate (pH 7.4) using a 30-kDa-cutoff centrifugal concentrator (Millipore, USA). The purified enzyme was stored with 20% (vol/vol) glycerol and 10 mM FAD at −80°C.

Enzyme assay.

Enzyme activity was assayed spectrophotometrically at 30°C by monitoring the decrease in absorbance of NADPH at 340 nm (ε₃₄₀ = 6.22 cm⁻¹mM⁻¹). The standard assay mixture (1 ml) contained 100 mM potassium phosphate buffer (pH 8.0), 2.0 mM substrate, 0.1 mM NADPH, and an appropriate amount of enzyme. One unit of BVMO_Rp was defined as the amount of enzyme catalyzing the oxidation of 1 μmol NADPH per min. The uncoupling NADPH was taken into consideration.

The kinetic parameters for BVMO_Rp with different ketone substrates were determined by standard activity assays at 30°C at various substrate concentrations and a fixed NADPH concentration of 0.2 mM. Apparent K_m and k_cat values were calculated by nonlinear regression fitting of data to the Michaelis-Menten equation using the Origin software. Protein concentration was determined using a modified Bradford protein assay kit (Sangon Biotech, China).

Effect of pH and temperature on activity and stability.

Cyclohexanone was used as a model substrate for investigating the effects of pH and temperature on activity and stability. The effect of pH on the activity of BVMO_Rp was determined by enzyme activity assays at 30°C in buffers of different pH (5.5 to 11.0). The effect of pH on the stability of BVMO_Rp was examined by measuring the residual activity after incubating purified BVMO_Rp (0.5 mg/ml) at 30°C for 12 h in buffers of different pH (6.0 to 10.0). The buffers were as follows: 100 mM NaAc-HAc (pH 5.5 to 6.0), 100 mM potassium phosphate (pH 6.0 to 8.0), and 100 mM Gly-NaOH (pH 8.0 to 11.0).

The effect of temperature on the activity of BVMO_Rp was investigated by assaying enzyme activity in potassium phosphate buffer (100 mM [pH 7.5]) at different temperatures (20 to 50°C). Thermostability was examined by incubating purified BVMO_Rp (0.75 mg/ml) in potassium phosphate buffer (100 mM [pH 7.5]) at 30, 35, and 40°C and measuring the residual activity for the designated time periods.

Biotransformation of aliphatic ketones by BVMO_Rp.

Reaction mixtures (200 μl) for biotransformations were composed of potassium phosphate buffer (100 mM [pH 8.0]), 2 mM substrate, ethanol (5% [vol/vol]), 3 mM NADPH, and various quantities of purified BVMO_Rp. Reaction mixtures were incubated in sealed bottles (2 ml) at 30°C with shaking at 1,000 rpm for 3 h and then extracted with an equivalent volume of ethyl acetate. Extracted samples were analyzed by gas chromatography (GC) or GC-mass spectrometry (MS) to measure the conversion of substrate and the ratio of the two ester products.

Analytical methods.

The conversion was analyzed by a Shimadzu GC-2014 instrument equipped with an Rxi-5Sil MS column (25 m by 0.25 mm by 0.25 μm; Restek). The two distinct products obtained from the oxidation of 2-hexanone, 3-hexanone, 2-octanone, 3-octanone, 2-decanone, and 3-decanone substrates were analyzed by GC (see Table S1 in the supplemental material).

The major and minor products formed from 4-octanone (substrate 5), 4-decanone (substrate 8), and 5-decanone (substrate 9) were analyzed using a Shimadzu GCMS-QP2010 instrument coupled to an InertCap 5MS/NP column (30 m by 0.25 mm by 0.25 μm; GL Science) using helium as the carrier gas. Mass spectra were collected using electrospray ionization. As the products of 5-decanone (substrate 9), pentyl pentanoate (product 9a) generates the characteristic ion at m/z 103, while butyl hexanoate (product 9b) produces the characteristic ion at m/z 99, and the proportion of product 9a in the product mixture was found to be linearly correlated with the ratio of the peak area of ion at m/z 103 to the sum of the peak areas of ions at m/z 103 and m/z 99 (Fig. S2). The regioselectivity of the enzyme was calculated using a calibration curve. Similarly for the products of 4-decanone (substrate 8), hexyl butyrate (product 8a) forms characteristic ions at m/z 71 and 89, whereas propyl heptanoate (product 8b) produces characteristic ions at m/z 113 and m/z 131. Meanwhile, as the products of 4-octanone (substrate 5), butyl butyrate (product 5a) generates characteristic ions at m/z 71 and 89, whereas propyl pentanoate (product 5b) has characteristic ions at m/z 85 and m/z 103.

Preparative synthesis of methyl 3-acetoxypropionate by BVMO_Rp.

Methyl levulinate (2.6 g, 20 mmol) was added to 100 ml potassium phosphate buffer (400 mM [pH 7.5]) containing cell lysate from E. coli expressing BVMO_Rp (2.0 g cells [wet weight]), 100 mg lyophilized crude glucose dehydrogenase (GDH; 2,500 U), d-glucose (5.4 g, 30 mmol), 0.5 mM NADP⁺, and 0.1 mM FAD. The reaction was performed in a sealed 1-liter reactor using pure oxygen instead of air at 30°C under magnetic stirring. After the conversion reached 99%, the reaction mixture was saturated with NaCl, extracted three times with ethyl acetate, and centrifuged (18,000 × g for 5 min) for phase separation. The combined ethyl acetate was washed twice with saturated NaCl solution, dried over anhydrous Na₂SO₄, and evaporated under reduced pressure. The crude product was purified by column chromatography on silica with a mixture of petroleum ether and ethyl acetate as the eluent (10:1 [vol/vol]).