Expression and Substrate Range of Streptomyces Vanillate Demethylase

Abstract

Vanillate is converted to protocatechuate by an O-demethylase consisting of VanA and VanB in Streptomyces sp. NL15-2K. In this study, vanillate demethylase from this strain was functionally expressed in Escherichia coli, and its substrate range for vanillate analogs was determined by an in vivo assay using recombinant whole cells. Among aromatic methyl ethers, vanillate, syringate, m-anisate, and veratrate were good substrates, whereas ferulate, vanillin, and guaiacol were not recognized by Streptomyces vanillate demethylase. After vanillate, 4-hydroxy-3-methylbenzoate was a better substrate than m-anisate and veratrate, and the 3-methyl group was efficiently oxidized to a hydroxymethyl group. These observations suggest that the combination of a carboxyl group on the benzene ring and a hydroxyl group in the para-position relative to the carboxyl group may be preferable for substrate recognition by the enzyme. ¹H-NMR analysis showed that the demethylation product of veratrate was isovanillate rather than vanillate. Therefore, it was concluded that O-demethylation of veratrate by Streptomyces vanillate demethylase occurred only at the meta-position relative to the carboxyl group.

Lignin is a major component of plant biomass, and partial decay of lignin yields numerous aromatic compounds, such as vanillin and catechol, that have many applications in the cosmetics, food, pharmaceutical, and chemical industries. To develop a bioconversion system for lignin-related aromatic compounds, we have focused on the enzymology and genetics of their metabolic pathways in bacteria. In many bacteria, most lignin-related aromatic compounds are degraded to protocatechuate and further broken down via specific ring cleavage pathways. We previously isolated Streptomyces sp. NL15-2K, which degrades various lignin-related aromatic compounds, from forest soil, and have studied the metabolic pathways in this strain.^1–4)

Vanillate demethylase is responsible for the conversion of vanillate to protocatechuate in the degradation of lignin-related aromatic compounds in bacteria. This enzyme from Streptomyces sp. NL15-2K is a two-component monooxygenase (class I) consisting of oxygenase and reductase components encoded by vanA and vanB, respectively,²⁾ and is assumed to convert vanillate to protocatechuate via hydroxylation of the O-methyl group, thereby forming an unstable hemiacetal that spontaneously decomposes into protocatechuate and formaldehyde.^5,6) The Streptomyces vanA (1071 bp) and vanB (936 bp) genes are organized in a cluster and co-transcribed from a putative promoter that is located approximately 61 bp upstream of the initiation codon ATG in vanA.²⁾ The enzymatic characteristics of vanillate demethylases from Pseudomonas spp. and Acinetobacter sp. have been determined in studies using cell-free extracts or recombinant Escherichia coli cells transformed with the vanillate demethylase gene.^7–9) Vanillate demethylase from P. testosteroni has a broad substrate range and shows activity towards m- and p-anisate in addition to vanillate,⁷⁾ whereas the enzyme from P. fluorescens is inactive toward p-anisate.⁸⁾ Similarly, although vanillate demethylase from Acinetobacter sp. has a wide substrate range, including m-anisate, veratrate, and 3,4,5-trimethoxybenzoate, it does not attack the 4-methoxy group relative to the carboxyl group.⁹⁾ Additionally, all of these enzymes have not been purified because their demethylation activity is very sensitive to dilution and to oxidation by air.^5,7,9) In a preliminary experiment, when a lysate was prepared from Streptomyces sp. NL15-2K cells exhibiting vanillate demethylase activity, activity of the enzyme was not detected in the lysate regardless of supplementation with cofactors such as reduced nicotinamide adenine dinucleotide (NADH) and reduced nicotinamide adenine dinucleotide phosphate (NADPH). Therefore, Streptomyces vanillate demethylase was also assumed to be unstable. The other characteristics of this enzyme have not been elucidated. Here, we describe the construction of a plasmid for expression of the Streptomyces vanillate demethylase gene in E. coli and present data on the substrate range, which was determined using an in vivo assay with recombinant E. coli cells.

MATERIALS AND METHODS

Bacterial Strains, Vector, Cultivation Media, and Chemicals

Streptomyces sp. NL15-2K was used for isolation of chromosomal DNA, which was extracted according to Hopwood et al.¹⁰⁾ E. coli DH5α was used as the host strain for recombinant plasmid preparation. E. coli BL21(DE3) and pET-28a(+) (Novagen, U.S.A.) were used for protein expression. E. coli strains were routinely grown in LB broth or on LB agar. M9 medium was used for an in vivo assay with recombinant E. coli cells. When necessary, kanamycin (30 µg/mL) was added to the media. All chemicals were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan) or Sigma-Aldrich Co. (St. Louis, MO, U.S.A.).

Plasmid Construction

The Streptomyces vanA and vanB genes (GenBank accession no. AB252870) were separately amplified to replace their own Shine-Dalgarno (SD) sequences with the pET-28a vector-derived SD sequence and inserted into pET-28a. Polymerase chain reaction (PCR) amplification was performed with PrimeSTAR®GXL DNA polymerase (TaKaRa Bio, Shiga, Japan). The primers used in this study are listed in Table 1. The coding sequence of the vanA gene was amplified by PCR using the vanA-F1 and vanA-R primers. The vanA-F1 primer contains an NcoI-compatible PciI site, in which an ATG initiation codon of the vanA gene was introduced. The vanA-R primer contains a SacI site introduced immediately downstream of a TGA stop codon of the vanA gene. The amplified vanA gene was digested with PciI and SacI, purified by agarose-gel electrophoresis, and inserted into the NcoI and SacI sites of pET-28a. The resulting plasmid pVDM-A was transformed into E. coli DH5α. Small-scale preparation of pVDM-A from the transformed cells was performed with a QIAGEN Plasmid Mini Kit (Qiagen, U.S.A.). To insert the vector-derived SD sequence just upstream of the ATG codon of the vanA gene, PCR for the vanA gene was performed again with pVDM-A as a template and with the primers vanA-F2 and vanA-R. The vanA-F2 primer is complementary to a region upstream of the SD sequence of the vector, and an NheI site was introduced into the primer sequence. The amplified product was digested with NheI and SacI, purified by electrophoresis, and inserted into the NheI and SacI sites of the pET vector. The resulting plasmid pVDM-sdA was transformed into E. coli DH5α for small-scale plasmid preparation. The coding sequence of the vanB gene was amplified with the vanB-F and vanB-R primers. The vanB-F primer contains an NcoI-compatible BspHI site, in which an ATG initiation codon of the vanB gene was introduced, and the vanB-R primer contains an NdeI site introduced just downstream of a TGA stop codon of the vanB gene. The amplified product was digested with BspHI and NdeI, purified by electrophoresis, and inserted into the NcoI and NdeI sites of the pVDM-sdA plasmid. The resulting plasmid was designated pVDM-B/A (Fig. 1A).

Table 1. List of Primers

Primer	Sequence (5′ to 3′)	Restriction site	Specific use
vanA-F1	CCCACATGTCCCACATGACCGCC	PciI	PCR for vanA
vanA-F2	CCCCGCTAGCAATAATTTTGTTTAAC	NheI	PCR for vanA with pET-derived SD sequence
vanA-R	ACGGAGCTCGTCACTGGACCGTCTCC	SacI	PCR for vanA with pET-derived SD sequence
vanB-F	AGCTCATGACCGTGTACGAAGCC	BspHI	PCR for vanB
vanB-R	TTCATATGACCTCAGGATCACAGATCCAGC	NdeI	PCR for vanB

Underlined sequences indicate restriction sites.

Fig. 1. Expression of Streptomyces Vanillate Demethylase Gene in E. coli

(A) Structure of the expression plasmid pVDM-B/A. All arrows in the plasmid indicate the direction of the genes. (B) SDS-PAGE analyses of the expressed VanAB proteins. Fractionation of total cell lysates by centrifugation was performed at 20000×g for 10 min. The resulting insoluble pellet was solubilized with EzApply (ATTO). Lane 1, molecular mass markers (sizes indicated); lanes 2 and 3, total cell lysates from E. coli harboring pET-28a or pVDM-B/A, respectively; lanes 4 and 5, supernatant and insoluble pellet, respectively, of a cell lysate from E. coli harboring pVDM-B/A. The positions of the VanA and VanB proteins are indicated with arrows.

Expression of the Vanillate Demethylase Gene

E. coli BL21(DE3) strains harboring pVDM-B/A or pET-28a were grown in LB broth containing 1% glucose and kanamycin at 37°C. The overnight cultures were diluted 1 : 100 into fresh LB broth (10 mL in a 50-mL flask) supplemented with kanamycin and incubated at 30°C with shaking at 120 rpm. When the cell density (optical density at 600 nm [OD₆₀₀]) reached 0.8, isopropyl-β-D-thiogalactopyranoside (IPTG) was added to the culture medium at 1 mM to induce gene expression. After 4 h, 3.6 mM vanillate was added to the culture, and incubation was continued for 16 h. The culture fluid was fractionated by centrifugation at 8000×g for 5 min. The resulting pellet and supernatant were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and HPLC on a Kinetex 5 u C18 100 A column (4.6×250 mm; Phenomenex), respectively. Elution was performed with 25% methanol containing 0.1% phosphoric acid at a flow rate of 1.0 mL/min at 45°C and monitored by absorption at 220 nm. A 10-µL sample was injected into the HPLC system.

SDS-PAGE

Bacterial pellets were suspended with water and disrupted using a sonicator (Handy Sonic model UR-20P; TOMY Seiko). The resulting lysate was solubilized with EzApply (ATTO, Japan) and subjected to SDS-PAGE on a pre-cast 12.5% polyacrylamide slab gel (e-PAGEL; ATTO) under reducing conditions. Proteins were stained with Bio-Safe Coomassie Stain (Bio-Rad, U.S.A.).

NMR Analyses of Metabolites

Conversion of veratrate, syringate, and 4-hydroxy-3-methylbenzoate was also analyzed by ¹H-NMR spectroscopy. Each substrate was incubated with E. coli cells harboring pVDM-B/A according to the in vivo assay technique described later, except that the cell concentration, substrate concentration, and incubation time were changed from 1.6×10⁹ cells/mL to 9.6×10⁹ cells/mL, 0.36 mM to 3.6 mM, and 60 min to 21 h, respectively. After incubation, the bacterial pellet was removed by centrifugation and the supernatant was acidified with dilute HCl aqueous solution. The mixture of residual substrate and metabolites was extracted with ethyl acetate and concentrated under reduced pressure to dryness. The ¹H-NMR spectra (CD₃OD) of the residue were measured at 25°C on a Bruker Avance III 600 (600 MHz; Bruker Daltonics) with methanol (3.30 ppm) as an internal standard.

In Vivo Assay for Recombinant Vanillate Demethylase Activity

Overnight cultures of E. coli BL21(DE3) harboring pVDM-B/A or pET-28a were diluted 1 : 100 into LB broth (25 mL in a 125-mL baffled flask) supplemented with kanamycin, and incubated at 30°C. When the OD₆₀₀ reached 0.8, IPTG was added at 0.5 mM and the incubation was continued. After 4 h of incubation, the culture fluid was centrifuged at 8000×g for 5 min. The bacterial pellet was recovered and washed with M9 medium supplemented with 0.5% glucose, 0.5 mM IPTG, and kanamycin. The pellet was then suspended (1.6×10⁹ cells/mL) in the same medium supplemented with 0.36 mM vanillate or other substrates and incubated at 30°C for 60 min. The residual substrates and their products in the culture supernatant were quantified by HPLC under the same conditions described above, except that 35% methanol–0.1% phosphoric acid was used for elution of m-anisate, m-hydroxybenzoate, m-toluate, and 3,5-dimethoxybenzoate. Vanillate, veratrate, vanillin, ferulate, m-anisate, 4-hydroxy-3-methylbenzoate, m-toluate, guaiacol, 3,5-dimethoxybenzoate, and syringate were used as substrates for the in vivo assay. Compounds were identified by comparison with the retention times of authentic standards.

RESULTS AND DISCUSSION

To express the Streptomyces vanillate demethylase gene in E. coli, we constructed the plasmid pVDM-B/A, in which vanA and vanB were positioned to be co-transcribed from the T7lac promoter, and the pET-derived SD sequence was inserted just upstream of each gene (Fig. 1A). To evaluate the plasmid, E. coli cells harboring pVDM-B/A or pET-28a were incubated with 1 mM IPTG and 3.6 mM vanillate for 16 h. Two polypeptides of 40 kDa and 33 kDa, which were assumed to be vanA and vanB products, respectively, were observed in the lysate of E. coli harboring pVDM-B/A (Fig. 1B). However, when the amounts of vanillate and protocatechuate in the culture supernatant were measured by HPLC, the percentage of vanillate converted to protocatechuate was 32.4% (Fig. 2B), which seemed to be slightly low relative to the amount of the expressed VanAB proteins. The lysate was thus further fractionated by centrifugation, and the fractions were subjected to SDS-PAGE. Most of the VanA and VanB proteins were detected in the supernatant fraction and in the insoluble fraction, respectively (Fig. 1B). Therefore, VanA is expressed as a soluble protein, whereas VanB is expressed as an inclusion body, and the low level of vanillate demethylase activity was confirmed to result from inclusion body formation of the VanB protein. Similar characteristics have been reported for a Streptomyces coniferyl alcohol dehydrogenase expressed in E. coli,⁴⁾ where inclusion body formation was controlled by exposing the cells to heat shock immediately before induction by IPTG. Application of this heat-shock treatment to VanB protein expression will be evaluated in the next stage of our study.

Veratrate has two methoxy groups at the 3- and 4-positions of its benzoate and is demethylated by vanillate demethylase from strain NL15-2K. To identify the position where demethylation occurs, ¹H-NMR analysis was performed. After incubation of E. coli BL21(DE3) harboring pVDM-B/A with 3.6 mM veratrate, the veratrate metabolites in the culture supernatant were extracted and analyzed by ¹H-NMR spectroscopy. The extract included the residual veratrate and a single product (44 : 56 molar ratio). The spectral data for veratrate were δ 3.86 (3H, s, OCH₃), 3.89 (3H, s, OCH₃), 7.01 (1H, d, J=8.4 Hz, 5-H), 7.55 (1H, d, J=1.8 Hz, 2-H), 7.62 (1H, dd, J=8.4, 1.8 Hz, 6-H). The data for the product were δ 3.91 (3H, s, 4-OCH₃), 6.97 (1H, d, J=8.4 Hz, 5-H), 7.43 (1H, d, J=1.8 Hz, 2-H), 7.54 (1H, dd, J=8.4, 1.8 Hz, 6-H). The spectral data were comparable with those of authentic veratrate and isovanillate. Because vanillate was not detected, demethylation by Streptomyces vanillate demethylase was concluded to occur only at the 3-position of veratrate and not at the 4-position. Similarly, the products formed from syringate and 4-hydroxy-3-methylbenzoate were analyzed by ¹H-NMR. In the extract of the culture supplemented with syringate, three molecular species, namely residual syringate, 3,4-dihydroxy-5-methoxybenzoate, and gallate, were identified; the respective molar ratio was determined to be approximately 50 : 39 : 11 from the spectra (Fig. 3). The ¹H-NMR spectral data for syringate were δ 3.88 (6H, s, 3-OCH₃ and 5-OCH₃), 7.32 (2H, s, 2-H and 6-H). The data for 3,4-dihydroxy-5-methoxybenzoate were δ 3.87 (3H, s, 5-OCH₃), 7.17 (1H, d, J=1.8 Hz, 2-H or 6-H), 7.18 (1H, d, J=1.8 Hz, 2-H or 6-H). The data for gallate were δ 7.05 (2H, s, 2-H and 6-H). We confirmed that the spectral data of syringate and the two demethylated products were comparable with those of authentic samples. This observation indicates that one of the two methoxy groups in the meta-position of syringate was demethylated first, and then the second methoxy group was demethylated at a slower rate. In contrast, the reaction with 4-hydroxy-3-methylbenzoate gave an oxidized product at the 3-position: 4-hydroxy-3-(hydroxymethyl)benzoate. The product structure was identified by ¹H-NMR analysis: δ 4.65 (2H, s, CH₂OH), 6.80 (1H, d, J=8.4 Hz, 5-H), 7.79 (1H, dd, J=8.4, 2.4 Hz, 6-H), 8.01 (1H, d, J=2.4 Hz, 2-H). Residual substrate and contaminants were not observed. Thus, Streptomyces vanillate demethylase is confirmed to also oxidize the 3-methyl group, yielding a hydroxymethyl group.

Fig. 2. HPLC Analyses of Vanillate Demethylation by Recombinant E. coli Harboring pET-28a (A) or pVDM-B/A (B)

Conversion of vanillate to protocatechuate by vanillate demethylase expressed in recombinant E. coli strains was monitored by HPLC. The weak peak at 4.6 min in chromatogram A was derived from the protocatechuate present as a trace impurity in the vanillate used in the experiment.

Fig. 3. ¹H-NMR Spectra of the Extract in the Reaction with Syringate (600 MHz, CD₃OD)

Next, we investigated the substrate range of vanillate demethylase for vanillate and its analogs using an in vivo assay. Vanillate was the best substrate among 10 aromatic compounds examined; the amount of vanillate decreased by 94% over 60 min (Fig. 4). After vanillate, 4-hydroxy-3-methylbenzoate was a better substrate than veratrate and m-anisate, suggesting that the combination of a carboxyl group on the benzene ring and a hydroxyl group in the para-position relative to the carboxyl group may be preferable for substrate recognition. This speculation is supported by the lack of detectable activity for guaiacol, vanillin, and ferulate, which do not have carboxyl groups on their benzene rings. Although syringate was a good substrate, it was demethylated at a lower rate than vanillate, indicating that the two methoxy groups in the meta-position may interfere with access to the enzyme active site. Furthermore, trace activity was detected towards 3,5-dimethoxybenzoate (2%) and m-toluate (1%). These results indicate that the vanillate demethylase from Streptomyces sp. NL15-2K is different from homologous enzymes from P. fluorescens and Acinetobacter sp. with regard to substrate preference.

Fig. 4. Substrate Range of Streptomyces Vanillate Demethylase for Vanillate and Its Analogs

Conversion rates were calculated as percentages by comparing the peak areas of the substrate before and after incubation with E. coli cells harboring pVDM-B/A. There were no detectable changes in the peak areas of substrates in the control experiment with E. coli cells harboring pET-28a.

In this study, we expressed vanillate demethylase from Streptomyces in E. coli and identified its substrate range for vanillate analogs using recombinant E. coli whole cells. Because E. coli is generally a more convenient and effective host for protein expression than Streptomyces, the recombinant strain prepared in this study may also be potentially valuable as a whole-cell biocatalyst for bioconversion of aromatic methyl ethers or methyl compounds. However, for this purpose, it is indispensable to improve the expression of recombinant vanillate demethylase in E. coli cells. Control of inclusion body formation by heat-shock treatment is now under investigation in our laboratory.

Acknowledgment

This research was supported by the Science Foundation of Yasuda Women’s University (No. 102118) and in part by Japan Society for the Promotion of Science KAKENHI Grant Number 24790122.

REFERENCES

• 1) Nishimura M, Ooi O, Davies J. Isolation and characterization of Streptomyces sp. NL15-2K capable of degrading lignin-related aromatic compounds. J. Biosci. Bioeng., 102, 124–127 (2006).
• 2) Nishimura M, Ishiyama D, Davies J. Molecular cloning of Streptomyces genes encoding vanillate demethylase. Biosci. Biotechnol. Biochem., 70, 2316–2319 (2006).
• 3) Nishimura M, Kohno K, Nishimura Y, Inagaki M, Davies J. Characterization of two isozymes of coniferyl alcohol dehydrogenase from Streptomyces sp. NL15-2K. Biosci. Biotechnol. Biochem., 75, 1770–1777 (2011).
• 4) Nishimura M. Molecular cloning and expression of the Streptomyces coniferyl alcohol dehydrogenase gene in Escherichia coli. Protein Expr. Purif., 89, 109–115 (2013).
• 5) Brunel F, Davison J. Cloning and sequencing of Pseudomonas genes encoding vanillate demethylase. J. Bacteriol., 170, 4924–4930 (1988).
• 6) Priefert H, Rabenhorst J, Steinbüchel A. Molecular characterization of genes of Pseudomonas sp. strain HR199 involved in bioconversion of vanillin to protocatechuate. J. Bacteriol., 179, 2595–2607 (1997).
• 7) Ribbons DW. Requirement of two protein fractions for O-demethylase activity in Pseudomonas testosteroni. FEBS Lett., 12, 161–165 (1971).
• 8) Cartwright NJ, Smith ARW. Bacterial attack on phenolic ethers: An enzyme system demethylating vanillic acid. Biochem. J., 102, 826–841 (1967).
• 9) Morawski B, Segura A, Ornston LN. Substrate range and genetic analysis of Acinetobacter vanillate demethylase. J. Bacteriol., 182, 1383–1389 (2000).
• 10) Hopwood DA, Bibb MJ, Chater KF, Kieser T, Bruton CJ, Kieser HM, Lydiate DJ, Smith CP, Ward JM, Schrempf H. Genetic Manipulation of Streptomyces: A Laboratory Manual, The John Innes Foundation, Norwich, pp. 69–80 (1985).