De novo biosynthesis of pterostilbene in an Escherichia coli strain using a new resveratrol O-methyltransferase from Arabidopsis
© The Author(s) 2017
Received: 2 November 2016
Accepted: 8 February 2017
Published: 15 February 2017
Pterostilbene, a structural analog of resveratrol, has higher oral bioavailability and bioactivity than that of the parent compound; but is far less abundant in natural sources. Thus, to efficiently obtain this bioactive resveratrol analog, it is necessary to develop new bioproduction systems.
We identified a resveratrol O-methyltransferase (ROMT) function from a multifunctional caffeic acid O-methyltransferase (COMT) originating from Arabidopsis, which catalyzes the transfer of a methyl group to resveratrol resulting in pterostilbene production. In addition, we constructed a biological platform to produce pterostilbene with this ROMT gene. Pterostilbene can be synthesized from intracellular l-tyrosine, which requires the activities of four enzymes: tyrosine ammonia lyase (TAL), p-coumarate:CoA ligase (CCL), stilbene synthase (STS) and resveratrol O-methyltransferase (ROMT). For the efficient production of pterostilbene in E. coli, we used an engineered E. coli strain to increase the intracellular pool of l-tyrosine, which is the initial precursor of pterostilbene. Next, we tried to produce pterostilbene in the engineered E. coli strain using l-methionine containing media, which is used to increase the intracellular pool of S-adenosyl-l-methionine (SAM). According to this result, pterostilbene production as high as 33.6 ± 4.1 mg/L was achieved, which was about 3.6-fold higher compared with that in the parental E. coli strain harboring a plasmid for pterostilbene biosynthesis.
As a potential phytonutrient, pterostilbene was successfully produced in E. coli from a glucose medium using a single vector system, and its production titer was also significantly increased using a l-methionine containing medium in combination with a strain that had an engineered metabolic pathway for l-tyrosine. Additionally, we provide insights into the dual functions of COMT from A. thaliana which was characterized as a ROMT enzyme.
Pterostilbene (3,5-dimethoxy-4′-hydroxy-trans-stilbene) is a naturally derived compound found primarily in blueberries and in the heartwood of red sandalwood (Pterocarpus santalinus) . Pterostilbene has numerous preventive and therapeutic properties for a vast range of human diseases that include neurological, cardiovascular, metabolic, and hematologic disorders [2, 3]. Further benefits of pterostilbene have been reported in preclinical trials, in which pterostilbene was shown to be a potent anticancer agent in several malignancies [3, 4]. Pterostilbene is structurally similar to resveratrol; however, the substitution of the hydroxyl with methoxy groups increases the lipophilicity of pterostilbene over resveratrol, which results in a high bioavailability [5, 6]. These differences in the pharmacokinetics might explain the higher biological activity of pterostilbene over its parental compound resveratrol [3, 7].
Resveratrol is most commonly found in grapes and the wine made from those grapes; however, pterostilbene is found primarily in some grapes and blueberries, which is at levels of 99–520 ng/g dry berry samples compared with the highest content of 5 mg/g of resveratrol . Therefore, pterostilbene has become an attractive target for bioengineering; however, only a few attempts to produce of pterostilbene have been tried in microbes and plants thus far [10–14]. Katsuyama et al.  reported on the production of pterostilbene in recombinant E. coli using the pinosylvin methyltransferase gene from Oryza sativa. Rimando et al.  reported on the accumulation of pterostilbene in tobacco and Arabidopsis by co-expression of O-methyltransferase and stilbene synthase. Jeong et al.  reported on the production of a trace amount of pterostilbene from resveratrol in E. coli by the expression of O-methyltransferase from Sorghum bicolor. Wang et al.  recently reported on the production of pterostilbene from p-coumaric acid in both E. coli and yeast. Meanwhile, the O-methyltransferase (OMT) family is responsible for catalyzing the transfer of a methyl group from S-adenosyl-l-methionine (SAM) to a wide range of secondary metabolites . It is generally thought that some OMTs use a diverse range of substrates, including phenylpropanoids, alkaloids, and flavonoids [16, 17].
In the present study, we describe the production of pterostilbene with the recombinant E. coli that harbors a new artificial biosynthetic pathway. In addition, this system was achieved with the novel resveratrol O-methyltransferase (ROMT) function of COMT gene from Arabidopsis which was originally reported to convert caffeic acid to ferulic acid.
Results and discussion
Characterization of the ROMT function of the Arabidopsis COMT gene
Kinetic analysis of the purified recombinant COMT
K m [μM]
K cat [s−1]
Catalytic efficiency K cat /K m
40.5 ± 6.6
(35.3 ± 5.7) × 10−3
44.9 ± 3.2
(12.8 ± 1.4) × 10−3
Construction of a de novo artificial biosynthetic pathway in E. coli to produce pterostilbene
Although the production of pterostilbene has been reported by some previously bio-engineering methods [9, 11, 13, 14], we tried a de novo synthesis in E. coli by engineering an artificial biosynthetic pathway using newly identified ROMT. This could be a useful approach for economic production by one-pot fermentation without a precursor feeding process.
Production of pinostilbene and pterostilbene by engineered E. coli strains
10.7 ± 1.9
9.3 ± 2.8
27.5 ± 4.4
19.4 ± 5.3
17.4 ± 6.8
33.6 ± 4.1
Improved production of pterostilbene in a l-tyrosine overproducing E. coli strain
To improve for the production yield, we used the pterostilbene production system in a l-tyrosine overproducing E. coli strain. p-Coumaric acid is the pivotal intermediate of the resveratrol and pterostilbene pathway starting from the deamination of tyrosine or the hydroxylation of cinnamic acid. Our strategy to harness microorganisms for the production of pterostilbene was to design and express artificial pathways with bacterial TAL. Thus, l-tyrosine serves as an initial endogenous precursor to the pterostilbene biosynthesis pathway. Recently, we reported on an engineered l-tyrosine overproducing E. coli ΔCOS1 strain through deregulation of the aromatic amino acid biosynthesis pathway . The genome engineered l-tyrosine producer, E. coli ΔCOS1, showed a substantial capacity for p-coumaric acid production through tal gene expression . Therefore, it is a suitable platform strain for the production of resveratrol and pterostilbene, using l-tyrosine as a common precursor.
Using the same experimental conditions described above, the tyrosine-overproducing E. coli ΔCOS1 strain (P2) harboring the pET-opT4CvS vector produced 19.4 ± 5.3 mg/L of pterostilbene after 72 h culture (Table 2). This productivity shows 1.8-fold improvement over the titers of the original producer (P1). At the same time, a significant amount of accumulated pinostilbene (27.5 ± 4.4 mg/L) was also detected, and there was trace accumulation of the intermediate p-coumaric acid and resveratrol (Fig. 3c). The amount of accumulated pinostilbene was slightly reduced and pterostilbene increased after 96 h culture (Fig. 4). This result means that the extra pinostilbene are not well converted to pterostilbene and accumulated in the cell. Therefore, methylation in the metabolic flow of resveratrol to pterostilbene may act as a bottleneck because of a shortage of methyl donors in the l-tyrosine overproducing cells.
Effects of l-methionine addition on pterostilbene production
As previously stated, this COMT enzyme uses SAM as the methyl donor. SAM production is improved when supplemented with excessive l-methionine in a l-methionine S-adenosyltransferase (MAT) overexpressing strain . In addition, the production of methylated compounds is remarkably improved by the feeding of SAM and l-methionine [22, 24].
In this study, we successfully demonstrated the de novo synthesis of pterostilbene, a di-methylated resveratrol, which is a more stable compound in vivo and potentially has more effective biological activities than that of resveratrol. This system was achieved with the novel resveratrol O-methyltransferase (ROMT) function of COMT gene from A. thaliana which originally was reported to convert caffeic acid to ferulic acid. The purified recombinant COMT showed significant ROMT activity catalyzing the conversion of resveratrol to pterostilbene. Furthermore, the de novo production of pterostilbene from a tyrosine overproducing strain (P2) with the addition of l-methionine to the medium was determined to be 3.6-fold over the E. coli P1 strain. The titers of the pterostilbene reached up to 33.6 ± 4.1 mg/L after 72 h of culturing in a minimal medium containing 1 mM l-methionine.
Bacterial strains, plasmids, and chemicals
The strains and plasmids used in this study are listed in Additional file 1: Table S1. Antibiotics were added to the medium as required at the following concentrations: ampicillin, 100 mg/L; kanamycin, 50 mg/L. T-blunt vector (Solgent, Korea) was used in the polymerase chain reaction (PCR) cloning. pET-22b(+) and pET-28a(+) were purchased from Novagen (USA). Caffeic acid, ferulic acid, resveratrol, and pterostilbene were purchased from Sigma-Aldrich (USA), pinostilbene was purchased from Tokyo Chemical Industry, Co (Japan) as a standard for compound identification by HPLC.
The restriction enzymes (NEB, USA; Takara, japan), KOD-plus-DNA polymerase (TOYOBO, japan), an AccuPower Ligation kit (Bioneer, Korea) and DNA ligation kit (Takara, japan) were used according to the manufacturers’ instructions. The optimized tyrosine ammonia lyase gene (optal) from Saccharothrix espanaensis [GenBank: DQ357071], cinnamate/p-coumarate: CoA ligase gene (4cl2nt) from Nicotiana tabacum [GenBank: AAB18638], stilbene synthase gene (vvsts) from Vitis vinifera [GenBank: NM_001281005.1] and caffeic acid-O-methyltransferase gene (comt) from Arabidopsis thaliana [GenBank: AY062837.1; Arabidopsis gene number: AT5G54160.1] were synthesized.
After cultivation of E. coli BL21(DE3) harboring the pET-COM , cells were harvested by centrifugation at 4000 rpm for 20 min at 4 °C. Harvested cells were re-suspended in lysis buffer (50 mM Tris–HCl, 300 mM NaCl, 10 mM imidazole, pH 7.4), and disrupted by sonication while chilled on ice for 30 min. Cell lysates were centrifuged at 15,000 rpm for 10 min at 4 °C, and the soluble fractions were collected from the supernatant. Soluble fractions were mixed with the His-Hyper Agarose resin (Lugen Sci. Co. Korea.) in poly-prep chromatography columns (Bio-Rad). After the binding of the enzyme, which was tagged with polyhistidine, the resin was washed with 50 mL of wash buffer (equilibration buffer containing 50 mM imidazole), and enzymes were eluted using 6 mL of elution buffer (equilibration buffer containing 250 mM imidazole). Protein samples were analyzed using 10% (w/v) SDS–polyacrylamide gel electrophoresis (SDS–PAGE). After gel electrophoresis, gels were stained with the gel staining solution (LPS solution, Korea).
Quantification of enzyme activity
The purified recombinant COMT protein were incubated in a total volume of 100 μL of 100 mM potassium phosphate buffer (pH 7.8) containing 0.5 mM S-adenosyl-l-methionine, 4 μg COMT protein and 6.25–400 μM substrate at 37 °C for 30 min. After reaction, 20 μL aliquot was subjected to HPLC as described above. The protein concentration was determined by the Bradford method using a protein assay dye (Bio-Rad, Hercules, CA, USA). The methylation activity of COMT were examined using caffeic acid and resveratrol to ferulic acid and pinostilbene, respectively. The substrate affinity (K m ) and turnover number (K cat ) values were calculated from Lineweaver–Burk plots. The analysis was performed in triplicate.
Construction of pterostilbene expression vector
The four genes (optal, 4cl2nt, comt and vvsts) were independently cloned into pET-28a(+) vectors [19–21, 23]. Using the optal, 4cl2nt, comt, and vvsts genes as templates, four DNA fragments were amplified by PCR with the appropriate pairs of primers. In order to assemble the pET-opT4CvS vector, the TAL coding region was amplified using pET-opTAL as a template with the primer opTAL-F (5′-CAT ATGACCCAGGTGGTTGAACGCC-3′) and Cpac (the sequence is located downstream of the T7 terminator region of the pET vector and contains the designed PacI site: 5′-TTAATTAATGCGCCGCTACAGGGCGCGTCC-3′), the CCL coding region was amplified using pET-4cl2nt as a template with the primer Npac (the sequence is located upstream of the T7 promoter region of the pET vector and contains the designed PacI site: 5′-TTAATTAATCGCCGCGACAATTTGCGACGG-3′) and Cspe (the sequence is located downstream of the T7 terminator region of the pET vector and the sequence contains the designed SpeI site 5′-ACTAGT TCCTCCTTTCAGCAAAAAACCCCTC-3′), the STS coding region was amplified using pET-STS as a template with the primer Nspe (the sequence is located upstream of the T7 promoter region of the pET vector and contains the designed SpeI site 5′-ACTAGTAGGTTGAGGCCGTTGAGCACCGCC-3′) and STS-R(5′-CTCGAGTTAGTTG GTGACCATCGGG-3′). Each of the amplified fragments were digested with corresponding sites and cloned between the NdeI and XhoI digested pET-28a(+) via ligation, which resulted in pET-opT4vS. The COMT coding region was amplified using pET-COM as a template with the primer Nspe and Cspe. A 1.6 kb SpeI fragment containing the comt gene was cloned between the SpeI digested pET-opT4vS which resulted in pET-opT4CvS (Fig. 3).
Culture conditions for production
Recombinant E. coli strains harboring plasmids were grown at 37 °C in a Luria–Bertani (LB) medium containing 50 μg/mL kanamycin. The overnight culture was inoculated into fresh LB medium supplemented with the same concentration of kanamycin. The culture was grown at 37 °C to an optical density at 600 nm (OD600) of 0.6, and IPTG was added to the final concentration of 1 mM, and the culture was incubated for 6 h at 26 °C. The cells were harvested by centrifugation, suspended, and incubated at 26 °C in a modified M9 minimal medium (M9C; Na2HPO4·7H2O 12.8 g/L, KH2PO4 3 g/L, NaCl 0.5 g/L, NH4Cl2 1 g/L, yeast extract 0.25 g/L, 2 mM MgSO4, 0.1 mM CaCl2, CaCO3 25 g/L, Glucose 15 g/L, 1 mM IPTG and appropriate antibiotics) and addition 1 mM l-methionine to M9C medium (M9M). For the feeding experiments, the cultures were supplemented with caffeic acid and resveratrol (final concentration: 0.2 mM), respectively, and allowed to grow for an additional 60 h.
Detection and quantification of the products
Ten milliliters of culture was extracted with an equal volume of ethyl acetate. The ethyl acetate was dried and resuspended in 800 μL of methanol. Twenty microliters of the extract was applied to a J’sphere ODS-H80 column (4.6 × 150 mm i.d., 5 μm; YMC, Japan) using a high performance liquid chromatography (HPLC) system [CH3CN–H2O (0.05% trifluoroacetic acid) 20–100% acetonitrile (CH3CN) for 20 min, 100% CH3CN for 5 min, at flow rate of 1 mL/min; Thermo Scientific, USA] equipped with a photodiode array detector. Samples was dissolved in a methanol and analyzed by electrospray ionization (ESI) mass spectrometry in the positive ion mode using a Thermo U3000-LTQ XL (Thermo Scientific, USA) system coupled to the ion trap mass spectrometer with a ESI (Electrospray ionization) source operating in the positive mode. Two microliters of the sample was injected into a HSS T3 C18 column (2.1 × 150 mm; 2.5 μm particle size) from Waters; the mobile phase used for gradient elution consisted of CH3CN (0.1% formic acid) as system A and H2O (0.1% formic acid) as system B. The flow rate was 0.3 mL/min. The gradient elution program started with 5% A, raised A to 100% in the following 15 min, remained at 100% A for 5 min, and returned to the initial condition (5% A) within the following 5 min. The data-dependent mass spectrometry experiments were controlled using the menu driven software provide with the Xcalibur system (version 2.2 SP1.48; Thermo Scientific). Quantification of the three above-mentioned compounds was based on the peak areas of absorbance at 320 nm. For the quantification of ferulic acid, pinostilbene, and pterostilbene, the HPLC analysis was followed as our previously described methods [12, 21]. The data shown in this study were generated from triplicate independent experiments.
KTH and SYK performed the experiments and wrote the manuscript. KTH and SYK contributed equally to this work. YSH designed all the experiments and wrote the manuscript. All authors read and approved the final manuscript.
We thank Dr. JH Jang and Mr. JK Lee (KRIBB) for technical assistance with LC/MS measurements.
The authors declare that they have no competing interests.
This work was supported in part by the KRIBB Research Initiative Program and by the Next-Generation BioGreen 21 Program (SSAC, PJ001108401) funded by the RDA, Republic of Korea.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Rimando AM, Kalt W, Magee JB, Dewey J, Ballington JR. Resveratrol, pterostilbene, and piceatannol in vaccinium berries. J Agric Food Chem. 2004;52:4713–9.View ArticleGoogle Scholar
- McCormack D, McFadden D. A review of pterostilbene antioxidant activity and disease modification. Oxid Med Cell Longev. 2013;2013:575482.View ArticleGoogle Scholar
- Kosuru R, Rai U, Prakash S, Singh A, Singh S. Promising therapeutic potential of pterostilbene and its mechanistic insight based on preclinical evidence. Eur J Pharmacol. 2016;789:229–43.View ArticleGoogle Scholar
- Fulda S. Resveratrol and derivatives for the prevention and treatment of cancer. Drug Discov Today. 2010;15:757–65.View ArticleGoogle Scholar
- Roupe KA, Remsberg CM, Yanez JA, Davies NM. Pharmacometrics of stilbenes: seguing towards the clinic. Curr Clin Pharmacol. 2006;1:81–101.View ArticleGoogle Scholar
- Shao X, Chen X, Badmaev V, Ho CT, Sang S. Structural identification of mouse urinary metabolites of pterostilbene using liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom. 2010;24:1770–8.View ArticleGoogle Scholar
- Remsberg CM, Yanez JA, Ohgami Y, Vega-Villa KR, Rimando AM, Davies NM. Pharmacometrics of pterostilbene: preclinical pharmacokinetics and metabolism, anticancer, antiinflammatory, antioxidant and analgesic activity. Phytother Res. 2008;22:169–79.View ArticleGoogle Scholar
- Schmidlin L, Poutaraud A, Claudel P, Mestre P, Prado E, Santos-Rosa M, Wiedemann-Merdinoglu S, Karst F, Merdinoglu D, Hugueney P. A stress-inducible resveratrol O-methyltransferase involved in the biosynthesis of pterostilbene in grapevine. Plant Physiol. 2008;148:1630–9.View ArticleGoogle Scholar
- Rimando AM, Pan Z, Polashock JJ, Dayan FE, Mizuno CS, Snook ME, Liu CJ, Baerson SR. In planta production of the highly potent resveratrol analogue pterostilbene via stilbene synthase and O-methyltransferase co-expression. Plant Biotechnol J. 2012;10:269–83.View ArticleGoogle Scholar
- Katsuyama Y, Funa N, Horinouchi S. Precursor-directed biosynthesis of stilbene methyl ethers in Escherichia coli. Biotechnol J. 2007;2:1286–93.View ArticleGoogle Scholar
- Jeong YJ, An CH, Woo SG, Jeong HJ, Kim YM, Park SJ, Yoon BD, Kim CY. Production of pinostilbene compounds by the expression of resveratrol O-methyltransferase genes in Escherichia coli. Enzyme Microb Technol. 2014;54:8–14.View ArticleGoogle Scholar
- Kang SY, Lee JK, Choi O, Kim CY, Jang JH, Hwang BY, Hong YS. Biosynthesis of methylated resveratrol analogs through the construction of an artificial biosynthetic pathway in E. coli. BMC Biotechnol. 2014;14:67.View ArticleGoogle Scholar
- Wang Y, Bhuiya MW, Zhou R, Yu O. Pterostilbene production by microorganisms expressing resveratrol O-methyltransferase. Ann Microbiol. 2015;65:10.Google Scholar
- Martinez-Marquez A, Morante-Carriel JA, Ramirez-Estrada K, Cusido RM, Palazon J, Bru-Martinez R. Production of highly bioactive resveratrol analogues pterostilbene and piceatannol in metabolically engineered grapevine cell cultures. Plant Biotechnol J. 2016;14:1813–25.View ArticleGoogle Scholar
- Lam KC, Ibrahim RK, Behdad B, Dayanandan S. Structure, function, and evolution of plant O-methyltransferases. Genome. 2007;50:1001–13.View ArticleGoogle Scholar
- Do CT, Pollet B, Thevenin J, Sibout R, Denoue D, Barriere Y, Lapierre C, Jouanin L. Both caffeoyl coenzyme A 3-O-methyltransferase 1 and caffeic acid O-methyltransferase 1 are involved in redundant functions for lignin, flavonoids and sinapoyl malate biosynthesis in Arabidopsis. Planta. 2007;226:1117–29.View ArticleGoogle Scholar
- Byeon Y, Lee HY, Lee K, Back K. Caffeic acid O-methyltransferase is involved in the synthesis of melatonin by methylating N-acetylserotonin in Arabidopsis. J Pineal Res. 2014;57:219–27.View ArticleGoogle Scholar
- Bugos RC, Chiang VL, Campbell WH. cDNA cloning, sequence analysis and seasonal expression of lignin-bispecific caffeic acid/5-hydroxyferulic acid O-methyltransferase of aspen. Plant Mol Biol. 1991;17:1203–15.View ArticleGoogle Scholar
- Choi O, Lee JK, Kang SY, Pandey RP, Sohng JK, Ahn JS, Hong YS. Construction of artificial biosynthetic pathways for resveratrol glucoside derivatives. J Microbiol Biotechnol. 2014;24:614–8.View ArticleGoogle Scholar
- Choi O, Wu CZ, Kang SY, Ahn JS, Uhm TB, Hong YS. Biosynthesis of plant-specific phenylpropanoids by construction of an artificial biosynthetic pathway in Escherichia coli. J Ind Microbiol Biotechnol. 2011;38:1657–65.View ArticleGoogle Scholar
- Kang SY, Choi O, Lee JK, Hwang BY, Uhm TB, Hong YS. Artificial biosynthesis of phenylpropanoic acids in a tyrosine overproducing Escherichia coli strain. Microb Cell Fact. 2012;11:153.View ArticleGoogle Scholar
- Kang SY, Lee JK, Jang JH, Hwang BY, Hong YS. Production of phenylacetyl-homoserine lactone analogs by artificial biosynthetic pathway in Escherichia coli. Microb Cell Fact. 2015;14:191.View ArticleGoogle Scholar
- Kang SY, Choi O, Lee JK, Ahn JO, Ahn JS, Hwang BY, Hong YS. Artificial de novo biosynthesis of hydroxystyrene derivatives in a tyrosine overproducing Escherichia coli strain. Microb Cell Fact. 2015;14:78.View ArticleGoogle Scholar
- Hu H, Qian J, Chu J, Wang Y, Zhuang Y, Zhang S. Optimization of l-methionine feeding strategy for improving S-adenosyl-l-methionine production by methionine adenosyltransferase overexpressed Pichia pastoris. Appl Microbiol Biotechnol. 2009;83:1105–14.View ArticleGoogle Scholar