Article pubs.acs.org/JAFC

Microbial Synthesis of Myrcene by Metabolically Engineered Escherichia coli Eun-Mi Kim,† Jin-Hee Eom,†,‡ Youngsoon Um,†,∥ Yunje Kim,† and Han Min Woo*,†,§,∥ †

Clean Energy Research Center, Korea Institute of Science and Technology, Hwarang-ro 14-gil 5, Seongbuk-gu, Seoul 136-791, Republic of Korea ‡ Department of Chemistry, and §Green School (Graduate School of Energy and Environment), Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 136-701, Republic of Korea ∥ Department of Clean Energy and Chemical Engineering, Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 305-350, Republic of Korea ABSTRACT: Myrcene, a monoterpene (C10), has gathered attention as a starting material for high-value compounds, such as geraniol/linalool and (−)-menthol. Metabolic engineering has been successfully applied to produce monoterpenes, such as pinene and limonene, at high levels in microbial hosts. However, microbial synthesis of myrcene has not yet been reported. Thus, we metabolically engineered Escherichia coli for production of myrcene by introducing a heterologous mevalonate pathway and overexpressing tailoring enzymes, such as geranyl diphosphate synthase (GPPS) and myrcene synthase (MS). Although MSs have broad ranges of functionality for producing various monoterpenes, our engineered E. coli strains harboring MS from Quercus ilex L. produced only myrcene (1.67 ± 0.029 mg/L). Subsequent engineering resulted in higher production of myrcene by optimizing the levels of GPPS in amino-acid-enriched (EZ-rich) defined medium, where glycerol as a carbon source was used. The production level of myrcene (58.19 ± 12.13 mg/L) was enhanced by 34-fold using in situ two-phase extraction to eliminate cellular toxicity and the evaporation of myrcene. KEYWORDS: mevalonate pathway, metabolic engineering, monoterpene, myrcene



INTRODUCTION Monoterpenes [C10 terpenes, consisting of two isoprene (C5) units] are an important and large class of natural products that are widely used in the cosmetic and pharmaceutical industries.1,2 Either the mevalonate (MVA) pathway3 in eukaryotes and some bacteria or the methylerythritol phosphate (MEP) pathway4 in plastids, bacteria, protozoa, and algae can produce isopentenyl diphosphate (IPP) and dimethylallyl disphosphate (DMAPP). These are catalyzed by prenyltransferase to form geranyl diphosphate (GPP, a precursor to monoterpene), which is further converted to monoterpene by various monoterpene synthases. Over 1000 naturally occurring monoterpenes (C10; acylic-, mono-, and bicyclic monoterpene isomers) have been identified among the volatile organic compounds found in the resins of many woody plants, notably in citrus plants.5 Among the monoterpenes, myrcene (β-myrcene, the only naturally occurring isomer) has been considered as a starting material for complex compounds, including heteroatom-linked chemicals (nerol/geraniol/linalool) and carbon-rearranged chemicals [(−)-menthol], which can be used as commercial olefinic scents.1 Myrcene synthase (MS) encoded by cDNA from flowers,6 grand fir,7 holm oak,8 and labiatae leaves9 was isolated and functionally characterized. Although myrcene can be extracted from essential oils or derived from β-pinene by pyrolysis, economic large-scale production has not yet been commercialized because of the high volatility of the substance and the inefficiency of the harvesting technology. Thus, development of engineered microbial platforms for large-scale industrial production is needed to provide myrcene from sugars.10,11 © XXXX American Chemical Society

Metabolic engineering of microorganisms has enabled production of monoterpenes, such as α/β-pinene,12−14 limonene,15,16 and sabinene.17 The monoterpenes were biosynthesized in Escherichia coli by introducing a heterologous MVA pathway for re-directing carbon flux from acetyl-coA to IPP/DMAPP and in a yeast by engineering the native MVA pathway. However, the levels of monoterpenes are lower than the levels of sesquiterpene18,19 (C15) or diterpene20 (C20) from the same microbial hosts, suggesting that geranyl pyrophosphate synthase (GPPS) and monoterpene synthase are still bottlenecks to high-level productions. It is also important that monoterpenes, including myrcene, are toxic to microbial hosts (minimum inhibitory concentrations: myrcene, 2.12 mM; β-pinene, 1.45 mM; and limonene, 0.44 mM, for yeast).12,15,21,22 Here, we report metabolic engineering of E. coli for the production of myrcene using heterologous expression of the MVA pathway and engineering at prenyl phosphate nodes with overexpression of GPPS and MS. In addition, the toxicity of myrcene was investigated and relieved in E. coli using in situ twophase extraction. Further biological engineering to optimize highcell-density fermentation will be the next challenge for achieving high-level myrcene production for industrial applications.



MATERIALS AND METHODS

Construction of Plasmids and Strains. E. coli DH5α (RBC Bioscience, Songnam, South Korea) was used as a host for cloning. Received: March 4, 2015 Revised: April 23, 2015 Accepted: April 24, 2015

A

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry Table 1. Bacteria Strains and Plasmids Used in This Study strain or plasmid E. coli HIT-DH5α E. coli DH1 myrcene producing E. coli DH1 pBbA5c-RFP pBbE1a-RFP pBbA5c-MevT(co)-MBIS(co) pM1 pM2 pM3 pM4 pM(Ag) pM(Pf) pM(Qi) pGM(Ag) pGM(Pf) pGM(Qi)

relevant characteristics

reference

Strains F−(80d lacZ M15) (lacZYA-argF) U169 hsdR17(r− m+) recA1 endA1 relA1 deoR96 F- endA1 recA1 gyrA96 thi-1 glnV44 relA1 hsdR17(rK-mK+) λDH1, co-transformation of MevT-MBI plasmid and myrcene production plasmid Plasmids p15A, Cmr, PlacUV5, rfp ColE1, Ampr, Ptrc, rf p p15A, Cmr, PlacUV5, MVA pathway genes p15A, Cmr, PlacUV5, pBbA5c-MevT(co)-MBI(co) p15A, Cmr, PlacUV5, pBbA5c-MevT(co)-MBI(co)-tGPPS2(co.Ag) p15A, Cmr, PlacUV5, Ptrc, pBbA5c-MevT(co)-T1-MBI(co) p15A, Cmr, PlacUV5, Ptrc, pBbA5c-MevT(co)-T1-MBI(co)-tGPPS2(co.Ag) ColE1, Ampr, Ptrc, tMS from A. grandis, pBbE1a-tMS(co.Ag) ColE1, Ampr, Ptrc, tMS from P. frutescens, pBbE1a-tMS(co.Pf) ColE1, Ampr, Ptrc, tMS from Q. ilex L., pBbE1a-tMS(co.Qi) ColE1, Ampr, Ptrc, tGPPS2 and tMS from A. grandis, pBbE1a-tGPPS2(co.Ag)-tMS(co.Ag) ColE1, Ampr, Ptrc, tGPPS2 from A. grandis and tMS from P. fruntescens, pBbE1a-tGPPS2(co.Ag)-tMS(co.Pf) ColE1, Ampr, Ptrc, tGPPS2 from A. grandis and tMS from Q. ilex L., pBbE1a-tGPPS2(co.Ag)-tMS(co.Qi)

RBC Bioscience CGSC this work 21 21 22 this work this work this work this work this work this work this work this work this work this work

Figure 1. Scheme of myrcene production in engineered E. coli and the cellular toxicity of myrcene. (A) Construction of the myrcene biosynthetic pathway in E. coli. The heterologous MVA pathway in E. coli consists of six enzymes for conversion from acetyl-CoA to IPP. To convert IPP to myrcene, GPPS and MS were expressed. Abbreviations of the corresponding genes to the MVA pathway were described in the Materials and Methods. (B) Plasmid construction for metabolic engineering of E. coli for myrcene production. A two-plasmid system was used. Plasmid 1 contains genes corresponding to the MVA pathway or MVA pathway with truncated AgGPPS2 on a medium copy number vector and medium strength promoter. Plasmid 2 harbors genes encoding for tMS or tGPPS2-MS on a high copy number vector and strong promoter. Three different kinds of MSs were introduced. (C) Cellular toxicity test of myrcene to E. coli. Growth of E. coli under various concentrations of myrcene [0, 0.01, 0.03, 0.06, and 0.12% (v/v)] was tested. The arrow indicates the myrcene addition to the culture medium at OD600 of 1. All plasmids and strains used in this study are listed in Table 1. All plasmids were prepared from the BglBrick expression plasmid library using the BglBrick standard cloning method.23,24 For construction of the heterologous MVA pathway in E. coli, the top portion (MevT) of the MVA pathway (Figure 1A) contains three genes needed for the conversion of acetyl-CoA to MVA.18 These are the genes encoding for acetyl-CoA thiolase (AtoB from E. coli), 3-hydroxyl-3methyl-glutaryl-coA synthase, and reductase (HMGS and HMGR from Saccharomyces cerevisiae). The bottom portion (MBI) of the MVA pathway contains four genes for the conversion of MVA to IPP and DMAPP.25 These are the genes encoding mevalonate kinase (MK from S. cerevisiae), phosphomevalonate kinase (PMK from S. cerevisiae), mevalonate diphosphate decarboxylase (PMD from S. cerevisiae), and isopentenyl diphosphate isomerase (Idi from E. coli). For construction

of the MVA pathway plasmids (plasmid 1 in Figure 1B), the gene encoding for farnesyl disphodphate synthase from E. coli (ispA) was removed from pBbA5c-MevT(co)-MBIS(co, ispA)25 using polymerase chain reaction (PCR) amplification of MevT(co)-MBI(co) (forward primer, 5′-ATT GAA TTC AAA AGA TCT AAA GGA GGC C-3′; reverse primer, 5′-TAT CTC GAG TTT GGA TCC TTA TTT AAG CTG GGT AAA TGC AG-3′, where underlines describe restriction enzyme sites), which was inserted into the pBbA5c plasmid. This resulted in a pM1 plasmid (Table 1). An additional pTrc promoter was inserted into a pM1, to construct a single operon of MevT and MBI, resulting in a pM3 plasmid. The DNA sequences were confirmed by DNA sequencing. To convert IPP/DMAPP to myrcene, a myrcene production plasmid containing genes encoding for MS and GPPS was constructed. The B

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry N-terminal plastid-targeting sequence of MSs and GPPS2 was truncated, and four codon-optimized genes (three MSs and one GPPS2) were synthesized (GenScript, Inc., Piscataway, NJ). Using the SignalP 4.1 server, amino acid sequences for signal peptide cleavage were predicted as followings: MS(Ag), 1−62; MS(Pf), 1−57; MS(Qi), 1−56; and GPPS2(Ag), 1−84, in the amino acid sequence. The genes encoding for MS were cloned in a pBbE1a, resulting in pM(Ag), pM(Pf), and pM(Qi). Subsequently, the gene encoding for GPPS2 was cloned into MevT-MBI plasmids (pM1 and pM3), resulting in pM2 and pM4. In addition, the GPPS2 gene was cloned into the myrcene production plasmids, yielding the series pGM(Ag), pGM(Pf), and pGM(Qi). The gene accession numbers of MSs and GPPS2 are MS(Ag) from Abies grandis (AAB71084), MS(Pf) from Perilla frutescens (AAF76186), MS(Qi) from Quercus ilex L. (CAC41012), and GPPS2 from A. grandis (AF513112). For myrcene production, the MevT-MBI plasmids and the myrcene production plasmids were combinatorially co-transformed into a E. coli DH1 strain, yielding different myrcene production strains. Growth Conditions for Myrcene Toxicity Test. E. coli DH1 was used as a host for the myrcene toxicity test and was cultured in an aminoacid-enriched (EZ-rich) defined medium (Teknova, Hollister, CA) in an Erlenmeyer flask (250 mL) at 37 °C and 200 rpm. When cell densities reached an OD600 of 0.8−1.0, different concentrations of myrcene [0, 0.01, 0.03, 0.06, 0.12% (v/v)] were added for the myrcene toxicity test in the absence of dodecane overlay. Then, cells were cultivated at 30 °C and 200 rpm for optical density (OD) measurements. Growth Conditions for Myrcene Production. E. coli DH1 was used as a host for myrcene production and was pre-cultured in Luria− Bertani (LB) medium. Myrcene production strains were grown in the three different media: LB medium, EZ-rich defined medium (Teknova, Hollister, CA), or M9-MOPS medium [M9 salt, 75 mM 3-(N-morpholino)propanesulfonic acid (MOPS), 2 mM MgSO4, 0.01 mM CaCl2, 1 mg/L thiamine·HCl, and 2.78 mg/L FeSO4 at pH 7]. The micronutrients provided were 3 nM ammonium molybdate, 0.4 μM boric acid, 30 nM cobalt chloride, 23 nM cupric sulfate, 80 nM manganese chloride, and 10 nM zinc sulfate. These were supplemented with either 1% glucose or 1% glycerol. Chloramphenicol (Cm) and ampicillin (Amp) were provided for plasmid maintenance at final concentrations of 30 and 100 μg/mL, respectively. Pre-cultured production strains were added to inoculate the media (1:100 dilution). For the cultivation in M9-MOPS medium, cells were adapted to M9-MOPS medium. The cultures (50 mL) were grown in an Erlenmeyer flask (250 mL) at 37 °C and 200 rpm. When cell densities reached at OD600 of 0.8−1.0, 100 μM isopropyl-β-D-1-thiogalactopyranoside (IPTG) was added for induction and 20% (v/v) dodecane overlay was added for in situ extraction. Cells were cultivated at 30 °C and 200 rpm for production of myrcene. Samples were collected from the dodecane layer at 24, 48, and 72 h for quantification of myrcene. Myrcene Analysis and Quantification. After the centrifugation (4400g for 10 min), the dodecane layers were collected. The collected samples (20 μL) were diluted with ethyl acetate (980 μL) containing 5 μg/mL (−)-trans-caryophyllene as an internal standard. To quantify myrcene, the diluted samples were analyzed using gas chromatography− mass spectrometry [GC−MS, Aglilent 6890N series GC/TOF-MS (LECO), under these conditions: injector temperature, 250 °C; flow rate, 1.2 mL/min; split ratio, 2:1; oven initially at 60 °C for 5 min, followed by 4 °C/min, and increase to 240 °C; carrier gas, He; and HPUltra2 column (25 m long, 0.2 mm diameter, and 0.11 μm film thick)]. Myrcene (CAS Registry Number 123-35-3) was used as an authentic standard for quantitative analysis, and the concentrations were normalized using an internal standard. All chemicals used in this study were purchased from Sigma-Aldrich.

heterologous MVA pathway (see the Materials and Methods). As shown by the authors of the previous studies,15,18,26 balanced gene expression was crucial for high-level production of terpenes. Because a pBbA5c vector containing a medium copy origin and a medium strength promoter has worked for production of various terpenes, including mono- and sesquiterpenes,12,18,27 we also chose the pBbA5c vector for the expression of MVA pathway genes. In combination with the MevT-MBI plasmid, we chose the pBbE1a vector (the second plasmid), which has a high copy origin and a strong pTrc promoter. Thus, MS was overexpressed in the pBbE1a vector (plasmid 2 in Figure 1B) to achieve myrcene production based on the “push and pull strategy” of metabolic engineering. Also, this two-plasmid system allowed screening for more efficient MS versions, from diverse candidates. Another important strategy for myrcene production is providing sufficient intermediates. Biosynthesis of monoterpene in an engineered E. coli was previously limited because of insufficient GPP and the low activity of monoterpene synthase.12,14 Combinational expression of tGPPS2 from A. grandis and pinene synthase Pt30 from Pinus taeda increased α-pinene production significantly up to 5.44 mg/L in a shaking flask and 0.95 g/L in fed-batch fermentation.14 Thus, we used tGPPS2 from A. grandis, specific geranyl diphosphate synthase, which has been shown to achieve high production results in monoterpene synthesis.12,15 Truncated AgGPPS2 was expressed in different plasmids to optimize gene expressions in (1) MevT-MBI plasmids (pM2/pM4), (2) myrcene production plasmids (pGM), and (3) both plasmids. AgGPPS2 expressions from different gene expression platforms could produce different levels of GPP, which are used as a substrate for MS. Also, higher levels of GPP might inhibit the activity of monoterpene synthase.28 Therefore, optimal gene expression of GPP is to be examined for higher production of myrcene, and we constructed the Ec-pM1/pGM, Ec-pM2/pM, and Ec-pM2/pGM strains. The last strategy for the development of the production strain was balancing gene expression in the MVA pathway. As shown previously using targeted proteomics, low expression of MK and PMK was identified as potential bottlenecks to sesquiterpene production.26 To overcome low expression, a strong pTrc promoter was inserted to increase gene expressions in the bottom portion of MVA, resulting in high production of bisabolene25 and limonene.15 Thus, we employed this strategy, yielding Ec-pM3/ pGM, Ec-pM4/pM, and Ec-pM4/pGM strains. Monoterpene, including pinene and limonene, has been shown to exhibit microbial toxicity because of phase toxicity caused by extraction during cell−solvent contact and molecular level of toxicity.13,21,22,29 To alleviate monoterpene toxicity, twophase extraction fermentation was widely applied to restore the growth to the level of the control. Interestingly, myrcene showed relatively low toxicity to microbial cells. The minimum inhibitory concentration (MIC, the amount of terpene for 50% reduction in growth) of myrcene for S. cerevisiae is 2.12 mM (288 mg/L), higher compared to other monoterpenes (limonene, MIC of 0.4 mM; β-pinene, MIC of 1.45 mM).21 Thus, we determined the myrcene toxicity in E. coli by adding different amounts of myrcene up to 0.12% (v/v) in the absence of dodecane when the cells reached an OD600 of 1 (Figure 1C). As a result, E. coli spiked with myrcene showed almost no growth inhibition. However, because myrcene has very low solubility (0.043 mM) and high volatility, we applied in situ extraction of myrcene from cell cultures using dodecane overlay. Selection of Myrcene Synthases for Myrcene-Specific Production. Monoterpene synthases are capable of producing a



RESULTS AND DISCUSSION Development of the E. coli Strain for Myrcene Production. For microbial production of myrcene, a two-plasmid system was employed to modulate the MVA pathway and terpene biosynthesis as previously studied.18,26 The first plasmid, MevTMBI (plasmid 1 in Figure 1B), consisted of seven genes of the C

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 2. GC−MS analysis of dodecane fractions from the engineered E. coli strains: (A) standard myrcene and (B) myrcene produced by a production strain [Ec-pM2/pM(Qi)]. Caryophyllene was used as the internal standard.

2 mg/L). Thus, we changed the medium to the EZ-rich defined and M9-MOPS media with 1% glucose. The growth profiles were similar, regardless of the media and strains, resulting in the fact that the pattern of the myrcene concentration (mg/L) is almost identical to that of specific production of myrcene (mg/L/OD600). Higher levels of myrcene (over 15 mg/L) were produced using the M9-MOPS medium compared to production when using the LB or EZ-rich media (Figure 3). Although each expression of AgGPPS2 in the three production strains was not quantitatively analyzed, different gene expressions of AgGPPS2 could be tested for the optimization based on the combination of the number of plasmid copies and the promoter strength of the BglBrick vectors.24 Thus, we expected that the relative levels of AgGPPS2 expression would differ in Ec-pM2/pM(Qi), Ec-pM1/ pGM(Qi), and Ec-pM2/pGM(Qi) and that the levels of expression would be low, intermediate, and high, respectively. Interestingly, Ec-pM1/pGM(Qi) showed the highest level of myrcene (15.08 ± 0.63 mg/L) with 1% glucose, which is a 3-fold increase from the lowest level of production. This result suggests that the Ec-pM1/pGM(Qi) strain, with the intermediate level of AgGPPS2, is optimal for myrcene production. As a third approach, balancing the gene expression of the top and bottom portions by dividing two-different operons [Ec-pM3/ pGM(Qi), pM4/pM(Qi), and Ec-pM4/pGM(Qi)] did not enhance myrcene production (5−8 mg/L). This could be due to poor GPPS and MS activity in converting IPP/DMAPP to myrcene; however, molecular analysis of MS has not yet been extensively studied. Besides glucose, glycerol was used as a renewable carbon source, a byproduct of the biodiesel synthesis.30,31 Because glycerol is a more reduced carbon source than glucose, it could enhance the production of myrcene in the NADPH-consuming MVA pathway. Thus, the Ec-pM2/pM(Qi), Ec-pM1/pGM(Qi), and Ec-pM2/pGM(Qi) strains were cultivated with 1% glycerol in either EZ-rich or M9-MOPS medium (panels A and B of Figure 4). As a result, the Ec-pM2/pM(Qi) strain in EZ-rich medium showed the highest level of myrcene production (58.19 ± 12.13 mg/L) from 1% glycerol, which is a 34-fold increase,

variety of monoterpenes from GPP. Monoterpene synthase catalyzes divalent metal ion-dependent ionization and isomerization of GPP to enzyme-bound linalyl diphosphate, which further converts to various monoterpenes.7 After isomerization and cyclic steps of the α-terpinyl cation, monocylic [(−)-limonene] and bicyclic [(−)-α/β-pinene and β-phellandrene] products were synthesized, whereas deprotonation of the carbocatation leads to an acyclic product (myrcene). Thus, it is important to express a myrcene-specific synthase in E. coli to achieve microbial synthesis of myrcene. From the literature of identified and characterized MSs, we selected three promising MSs (from A. grandis,7 P. frutescens,9 and Q. ilex L.8). Using a two-plasmid system, we cloned each MS encoding gene, yielding pM(Ag), pM(Pf), and pM(Qi), respectively. Subsequently, each plasmid was co-transformed with the MevT-MBI plasmid in E. coli DH1. Thus, we constructed three production strains: Ec-pM1/pM(Ag), Ec-pM1/ pM(Pf), and Ec-pM1/pM(Qi). None of these strains produced myrcene (data not shown). To express heterologous GPPS in E. coli, truncated AgGPPS2 was cloned into three MS-containing plasmids. As a result, Ec-pM2/pM(Ag), Ec-pM2/pM(Pf), and Ec-pM2/pM(Qi) were constructed. LB medium with 1% glucose was used for the production test. Only strains harboring MS(Qi) produced amounts of myrcene detectable by GC−MS (Figure 2A, 1.67 ± 0.029 mg/L myrcene was produced; limit of detection of 0.030 mg/L). As reported previously,8 we also confirmed that MS(Qi)-expressing E. coli produced only myrcene and other side products were not detected. Therefore, we further engineered the E. coli strain harboring the MS(Qi)-encoding gene [Ec-pM2/ pM(Qi)] to enhance myrcene production. Microbial Synthesis of Myrcene. Metabolic engineering of the E. coli strain harboring the MS(Qi)-encoding gene was applied to increase myrcene production. On the basis of the “push and pull” strategy, we decided to overexpress AgGPPS2 in addition to MS. The Ec-pM2/pM(Qi), Ec-pM1/pGM(Qi), and Ec-pM2/pGM(Qi) strains were cultivated in LB medium with 1% glucose (Figure 3A). Over the production times (24, 48, and 72 h), the three strains produced low levels of myrcene (less than D

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 3. Myrcene production from 1% glucose using engineered E. coli: (A) LB, (B) EZ-rich, and (C) M9-MOPS media (green bar, 24 h; red bar, 48 h; and blue bar, 72 h). Myrcene concentrations (mg/L) and specific production of myrcene (mg/L/OD600) were shown in the left and right panels, respectively. All data are the mean ± standard deviation (SD) from triplicate cultures.

Figure 4. Myrcene production from 1% glycerol using engineered E. coli: (A) EZ-rich and (B) M9-MOPS media (green bar, 24 h; red bar, 48 h; and blue bar, 72 h). Myrcene concentrations (mg/L) and specific production of myrcene (mg/L/OD600) were shown in the left and right panels, respectively. All data are the mean ± SD from triplicate cultures. E

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

intracellular myrcene. Therefore, a two-phase culture system with in situ extraction using dodecane is needed to eliminate cellular toxicity and the evaporation of myrcene. For the first time, we have demonstrated metabolic engineering of E. coli for the production of myrcene. Combinatorial gene expression of tGPPS2(Ag) and tMS(Qi) was a key success for myrcene production in E. coli harboring the heterologous MVA pathway. For enhanced production, system biology could be applied to understand the cellular metabolism of the myrceneproducing E. coli to identify bottlenecks. Furthermore, bioprocess engineering is needed to achieve commercially relevant product concentrations and space-time yields of myrcene from engineered E. coli.

compared to production by the same strain in LB medium with 1% glucose. Unlike for myrcene production from glucose, a relatively low expression of AgGPPS2 was preferred in E. coli for production of myrcene from glycerol. Therefore, the combinatorial expressions of GPPS and MS must be regarded as a critical task for myrcene production in E. coli. Targeted proteomic analysis18,26 could be informative to understand the quantitatively correlation with GPPS/MS expressions and myrcene production depending upon the carbon sources (glucose and glycerol). Nonetheless, the level of myrcene production (58.19 ± 12.13 mg/L) from our engineered strain showed the highest production level published thus far. Further protein engineering of AgGPPS2 and MS based on directed evolution through random mutagenesis or protein structures could be necessary for enhancing activities of GPPS and MS and improving myrcene production. In Situ Extraction of Volatile Myrcene from E. coli. Because of high volatility of myrcene, in situ extraction of myrcene from cell cultures was conducted using dodecane overlay.12,15,18 The dodecane layer can be used as an efficient product sink. Without the dodecane overlay, we observed the complete loss of the myrcene product from the Ec-pM2/pM(Qi) strain using 1% glycerol (Figure 5A). From this result, we



AUTHOR INFORMATION

Corresponding Author

*Telephone: +82-2-958-5249. Fax: +82-2-958-5209. E-mail: [email protected]. Funding

This work was financially supported by the Research and Development (R&D) Convergence Program of the National Research Council of Science and Technology (NST) of the Republic of Korea and the Korea Institute of Science and Technology (KIST, 2E25402) and the Korea Carbon Capture and Sequestration (CCS) R&D Center (KCRC) Grant 2014M1A8A1049277, funded by the Korean Government [Ministry of Science, Information and Communications Technology (ICT) and Future Planning]. Notes

The authors declare no competing financial interest.



REFERENCES

(1) Behr, A.; Johnen, L. Myrcene as a natural base chemical in sustainable chemistry: A critical review. ChemSusChem 2009, 2, 1072− 1095. (2) Davis, E. M.; Ringer, K. L.; McConkey, M. E.; Croteau, R. Monoterpene metabolism. Cloning, expression, and characterization of menthone reductases from peppermint. Plant Physiol. 2005, 137, 873− 881. (3) Bloch, K.; Chaykin, S.; Phillips, A. H.; De Waard, A. Mevalonic acid pyrophosphate and isopentenylpyrophosphate. J. Biol. Chem. 1959, 234, 2595−2604. (4) Fellermeier, M.; Raschke, M.; Sagner, S.; Wungsintaweekul, J.; Schuhr, C. A.; Hecht, S.; Kis, K.; Radykewicz, T.; Adam, P.; Rohdich, F.; Eisenreich, W.; Bacher, A.; Arigoni, D.; Zenk, M. H. Studies on the nonmevalonate pathway of terpene biosynthesis. The role of 2C-methylD-erythritol 2,4-cyclodiphosphate in plants. Eur. J. Biochem. 2001, 268, 6302−6310. (5) Marmulla, R.; Harder, J. Microbial monoterpene transformationsA review. Front. Microbiol. 2014, 5, 346. (6) Aros, D.; Gonzalez, V.; Allemann, R. K.; Muller, C. T.; Rosati, C.; Rogers, H. J. Volatile emissions of scented Alstroemeria genotypes are dominated by terpenes, and a myrcene synthase gene is highly expressed in scented Alstroemeria flowers. J. Exp. Bot. 2012, 63, 2739−2752. (7) Bohlmann, J.; Steele, C. L.; Croteau, R. Monoterpene synthases from grand fir (Abies grandis). cDNA isolation, characterization, and functional expression of myrcene synthase, (−)-(4S)-limonene synthase, and (−)-(1S,5S)-pinene synthase. J. Biol. Chem. 1997, 272, 21784−21792. (8) Fischbach, R. J.; Zimmer, W.; Schnitzler, J. P. Isolation and functional analysis of a cDNA encoding a myrcene synthase from holm oak (Quercus ilex L.). Eur. J. Biochem. 2001, 268, 5633−5638. (9) Hosoi, M.; Ito, M.; Yagura, T.; Adams, R. P.; Honda, G. cDNA isolation and functional expression of myrcene synthase from Perilla f rutescens. Biol. Pharm. Bull. 2004, 27, 1979−1985.

Figure 5. In situ extraction of myrcene using a dodecane overlay. (A) Concentration of myrcene produced by the Ec-pM2/M(Qi) strain from 1% glycerol in the EZ-rich medium either with (black square) or without (red circle) dodecane over the culture medium. (B) Cell growth of the Ec-pM2/M(Qi) strain either with (black bar) or without (red bar) dodecane over the culture. OD600 was measured at 24, 48, and 72 h. Error bars present the SD from triplicate cultures.

confirmed that myrcene rapidly evaporates after microbial synthesis. The cellular growth was also inhibited (Figure 5B), although the level of microbial synthesis of myrcene in E. coli was far less than exogenous myrcene tested for cellular toxicity in this study. This could be due to molecular level toxicity of F

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry (10) Reiling, K. K.; Yoshikuni, Y.; Martin, V. J.; Newman, J.; Bohlmann, J.; Keasling, J. D. Mono and diterpene production in Escherichia coli. Biotechnol. Bioeng. 2004, 87, 200−212. (11) Kirby, J.; Keasling, J. D. Metabolic engineering of microorganisms for isoprenoid production. Nat. Prod. Rep. 2008, 25, 656−661. (12) Sarria, S.; Wong, B.; Garcia Martin, H.; Keasling, J. D.; PeraltaYahya, P. Microbial synthesis of pinene. ACS Synth. Biol. 2014, 3, 466− 475. (13) Kang, M. K.; Eom, J. H.; Kim, Y.; Um, Y.; Woo, H. M. Biosynthesis of pinene from glucose using metabolically-engineered Corynebacterium glutamicum. Biotechnol. Lett. 2014, 36, 2069−2077. (14) Yang, J.; Nie, Q.; Ren, M.; Feng, H.; Jiang, X.; Zheng, Y.; Liu, M.; Zhang, H.; Xian, M. Metabolic engineering of Escherichia coli for the biosynthesis of α-pinene. Biotechnol. Biofuels 2013, 6, 60. (15) Alonso-Gutierrez, J.; Chan, R.; Batth, T. S.; Adams, P. D.; Keasling, J. D.; Petzold, C. J.; Lee, T. S. Metabolic engineering of Escherichia coli for limonene and perillyl alcohol production. Metab. Eng. 2013, 19, 33−41. (16) Willrodt, C.; David, C.; Cornelissen, S.; Buhler, B.; Julsing, M. K.; Schmid, A. Engineering the productivity of recombinant Escherichia coli for limonene formation from glycerol in minimal media. Biotechnol. J. 2014, 9, 1000−1012. (17) Zhang, H.; Liu, Q.; Cao, Y.; Feng, X.; Zheng, Y.; Zou, H.; Liu, H.; Yang, J.; Xian, M. Microbial production of sabineneA new terpenebased precursor of advanced biofuel. Microb. Cell Fact. 2014, 13, 20. (18) Woo, H. M.; Murray, G. W.; Batth, T. S.; Prasad, N.; Adams, P. D.; Keasling, J. D.; Petzold, C. J.; Lee, T. S. Application of targeted proteomics and biological parts assembly in E. coli to optimize the biosynthesis of an anti-malarial drug precursor, amorpha-4,11-diene. Chem. Eng. Sci. 2013, 103, 21−28. (19) Paddon, C. J.; Westfall, P. J.; Pitera, D. J.; Benjamin, K.; Fisher, K.; McPhee, D.; Leavell, M. D.; Tai, A.; Main, A.; Eng, D.; Polichuk, D. R.; Teoh, K. H.; Reed, D. W.; Treynor, T.; Lenihan, J.; Fleck, M.; Bajad, S.; Dang, G.; Dengrove, D.; Diola, D.; Dorin, G.; Ellens, K. W.; Fickes, S.; Galazzo, J.; Gaucher, S. P.; Geistlinger, T.; Henry, R.; Hepp, M.; Horning, T.; Iqbal, T.; Jiang, H.; Kizer, L.; Lieu, B.; Melis, D.; Moss, N.; Regentin, R.; Secrest, S.; Tsuruta, H.; Vazquez, R.; Westblade, L. F.; Xu, L.; Yu, M.; Zhang, Y.; Zhao, L.; Lievense, J.; Covello, P. S.; Keasling, J. D.; Reiling, K. K.; Renninger, N. S.; Newman, J. D. High-level semisynthetic production of the potent antimalarial artemisinin. Nature 2013, 496, 528−532. (20) Ajikumar, P. K.; Xiao, W. H.; Tyo, K. E.; Wang, Y.; Simeon, F.; Leonard, E.; Mucha, O.; Phon, T. H.; Pfeifer, B.; Stephanopoulos, G. Isoprenoid pathway optimization for Taxol precursor overproduction in Escherichia coli. Science 2010, 330, 70−74. (21) Brennan, T. C.; Turner, C. D.; Krömer, J. O.; Nielsen, L. K. Alleviating monoterpene toxicity using a two-phase extractive fermentation for the bioproduction of jet fuel mixtures in Saccharomyces cerevisiae. Biotechnol. Bioeng. 2012, 109, 2513−2522. (22) Dunlop, M. J.; Dossani, Z. Y.; Szmidt, H. L.; Chu, H. C.; Lee, T. S.; Keasling, J. D.; Hadi, M. Z.; Mukhopadhyay, A. Engineering microbial biofuel tolerance and export using efflux pumps. Mol. Syst. Biol. 2011, 7, 487. (23) Anderson, J. C.; Dueber, J. E.; Leguia, M.; Wu, G. C.; Goler, J. A.; Arkin, A. P.; Keasling, J. D. BglBricks: A flexible standard for biological part assembly. J. Biol. Eng. 2010, 4, 1. (24) Lee, T. S.; Krupa, R. A.; Zhang, F.; Hajimorad, M.; Holtz, W. J.; Prasad, N.; Lee, S. K.; Keasling, J. D. BglBrick vectors and datasheets: A synthetic biology platform for gene expression. J. Biol. Eng. 2011, 5, 12. (25) Peralta-Yahya, P. P.; Ouellet, M.; Chan, R.; Mukhopadhyay, A.; Keasling, J. D.; Lee, T. S. Identification and microbial production of a terpene-based advanced biofuel. Nat. Commun. 2011, 2, 483. (26) Redding-Johanson, A. M.; Batth, T. S.; Chan, R.; Krupa, R.; Szmidt, H. L.; Adams, P. D.; Keasling, J. D.; Lee, T. S.; Mukhopadhyay, A.; Petzold, C. J. Targeted proteomics for metabolic pathway optimization: Application to terpene production. Metab. Eng. 2011, 13, 194−203. (27) Alonso-Gutierrez, J.; Kim, E. M.; Batth, T. S.; Cho, N.; Hu, Q.; Chan, L. J.; Petzold, C. J.; Hillson, N. J.; Adams, P. D.; Keasling, J. D.;

Garcia Martin, H.; Lee, T. S. Principal component analysis of proteomics (PCAP) as a tool to direct metabolic engineering. Metab. Eng. 2014, 28C, 123−133. (28) Phillips, M. A.; Savage, T. J.; Croteau, R. Monoterpene synthases of loblolly pine (Pinus taeda) produce pinene isomers and enantiomers. Arch. Biochem. Biophys. 1999, 372, 197−204. (29) Sikkema, J.; Weber, F. J.; Heipieper, H. J.; Debont, J. A. M. Cellular toxicity of lipophilic compoundsMechanisms, implications, and adaptations. Biocatal. Biotransform. 1994, 10, 113−122. (30) da Silva, G. P.; Mack, M.; Contiero, J. Glycerol: a promising and abundant carbon source for industrial microbiology. Biotechnol. Adv. 2009, 27, 30−39. (31) Yang, F.; Hanna, M. A.; Sun, R. Value-added uses for crude glycerolA byproduct of biodiesel production. Biotechnol. Biofuels 2012, 5, 13.

G

DOI: 10.1021/acs.jafc.5b01334 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Microbial Synthesis of Myrcene by Metabolically Engineered Escherichia coli.

Myrcene, a monoterpene (C10), has gathered attention as a starting material for high-value compounds, such as geraniol/linalool and (-)-menthol. Metab...
2MB Sizes 0 Downloads 14 Views