Viridiflorol production in auxotrophic Escherichia coli.

Abstract

To identify methods of viridiflorol synthesis that can produce higher yields in a host cell.SOLUTION: A method comprises the steps of: a) providing a host cell that produces one or more terpenoids, said host cell comprising a vector comprising a polynucleotide sequence encoding a terpene synthase enzyme; b) modifying the vector to: i. introduce an inducible promoter operably linked to the polynucleotide sequence encoding the terpene synthase enzyme; and ii. introduce a polynucleotide sequence encoding a ribosomal binding site (RBS) that increases translation initiation rate of the terpenoid compared to a wild type ribosomal binding site; c) determining the dosage of an inducer capable of inducing the inducible promoter; d) culturing the host cell in a culture medium in the presence of the inducer at the dosage determined from step c); and e) isolating the terpenoid from the culture medium.SELECTED DRAWING: None

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority to Singapore Application No. 10201907354Y, filed August 8, 2019, the contents of which are incorporated herein by reference in their entirety for all purposes.

[0002] Technical Field The present invention relates to the field of biotechnology. In particular, the present invention relates to a method for producing terpenoids and its optimization. [Background technology]

[0003] Terpenoids, also known as isoprenoids, have a great deal of regiochemical and stereochemical diversity, ranging from C5 (hemiterpenes) to C40 (tetraterpenes) to polyterpenes (e.g., natural rubber). This diversity of terpenoids is primarily due to terpene synthases (TPSs), which convert acyclic prenyl diphosphate precursors into a multitude of cyclic and acyclic terpene skeletons. Corresponding to their structural diversity, the functions of terpenoids range from mediating mutualistic or antagonistic interactions between organisms to contributing to electron transfer, protein prenylation, or membrane fluidity. The structural and functional diversity of terpenoids has led to their widespread use in a variety of applications, including medicines, nutraceuticals, agriculture, flavors, fragrances, and biofuels.

[0004] Terpenes and terpenoids are major constituents of many plant essential oils and are major contributors to the aroma of these oils. One terpenoid of particular interest is viridiflorol, a known fragrance molecule isolated from Niaouli (Melaleuca quinquenervia) essential oil. Due to its antibacterial, anti-inflammatory and antioxidant properties, viridiflorol is currently used in several cosmetic and personal care products. In addition, viridiflorol is a strong feeding deterrent against Melaleuca leaf weevils and a potent inhibitor of bacterial biofilm formation. Therefore, viridiflorol has potential applications in the agricultural and food industries to replace a wide range of current toxic pesticides and harmful food preservatives. However, the high cost and supply fluctuations of plant extracts due to political and meteorological conditions have limited the production and use of viridiflorol. Chemical synthesis of viridiflorol has low enantioselectivity. The presence of six chiral centers, or 64 different stereoisomers, has posed a great challenge to the chemical synthesis of viridiflorol, resulting in low yields and very high costs. Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, there is a need to identify methods for the synthesis of viridiflorol that can produce higher yields. [Means for solving the problem]

[0006] In one embodiment, there is provided a method for increasing terpenoid production in a host cell that produces one or more terpenoids, comprising: a) providing a host cell that produces one or more terpenoids, the host cell comprising a vector comprising a polynucleotide sequence encoding a terpene synthase enzyme; b) the vector, i. introducing an inducible promoter operably linked to the polynucleotide sequence encoding the terpene synthase enzyme; and ii. to introduce a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of said terpenoid compared to a wild-type ribosome binding site; and modifying the c) determining the dosage of an inducer capable of inducing said inducible promoter; d) culturing said host cell in a medium in the presence of said inducer at said dosage determined from step c); e) isolating the terpenoids from the medium. A method is provided comprising:

[0007] In another aspect, a host cell comprising at least one vector, said at least one vector comprising: a) a polynucleotide sequence encoding one or more genes of the mevalonate pathway operably linked to an inducible promoter; b) a polynucleotide sequence encoding the ispA gene; and c) a polynucleotide sequence encoding a terpene synthase enzyme operably linked to an inducible promoter; d) a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of the terpene synthase enzyme compared to a wild-type ribosome binding site; A host cell comprising the

[0008] In another embodiment, there is provided an E. coli cell deleted for genes aroA, aroB, and aroC, comprising: a. a first vector, i. a polynucleotide sequence encoding a viridiflorol synthase enzyme operably linked to a TM1 promoter, said viridiflorol synthase enzyme being mutated, said mutations being G227C, V314Y and a deletion of amino acids at positions 2-85 of SEQ ID NO:24; ii. a polynucleotide sequence encoding a RBS ribosome binding site (RBS) that increases the rate of translation initiation of a terpenoid compared to a wild-type ribosome binding site, said polynucleotide sequence being located upstream of said polynucleotide sequence encoding a viridiflorol synthase enzyme, said polynucleotide sequence encoding said RBS being a polynucleotide sequence selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17; iii. a polynucleotide sequence encoding an ispA gene operably linked to a promoter; and iv. A polynucleotide sequence encoding an aroA gene operably linked to a promoter A first vector comprising: b. a second vector, i. a polynucleotide sequence encoding the genes atoB, hmgS and a truncated hmgR operably linked to the TM2 promoter; and ii. A polynucleotide sequence encoding the aroC gene operably linked to a promoter and a second vector comprising c. a third vector, i. a polynucleotide sequence encoding the genes mevK, pmK, pmd, and idi operably linked to the TM3 promoter; and ii. A polynucleotide sequence encoding an aroB gene operably linked to a promoter and a third vector comprising An E. coli cell comprising:

[0009] In another aspect, an engineered viridiflorol synthase enzyme is provided comprising one or more amino acid substitutions at positions selected from the group consisting of positions 227, 249, 267, 314, 318, 326 of SEQ ID NO:24, deletions of amino acids at positions 2-85, and combinations thereof.

[0010] In another aspect, there is provided a method for producing viridiflorol, the method comprising culturing a host cell described herein, or an E. coli cell described herein, in a medium containing about 0.001-0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) or about 0.1-50 mM lactose.

[0011] In another aspect, a kit for producing viridiflorol is provided comprising a host cell as described herein, or an E. coli cell as described herein, together with instructions for use.

[0012] definition As used herein, the term "terpene" refers to a class of organic compounds produced by plants, bacteria, fungi, and insects. Terpenes are derived from isoprene, which has the molecular formula C5H8. Terpenes may be classified by the number of isoprene units in the molecule.

[0013] As used herein, the term "terpenoid" refers to a large and diverse class of organic compounds derived from terpenes, including terpenes.Terpenes are a subset of terpenoids.The basic unit of terpenes and terpenoids has a 5-carbon isoprene unit and contains additional functional groups, usually oxygen-containing functional groups.Terpenoids may be derived from chemical modifications of terpenes, such as oxidation of terpenes or recombination of the carbon skeleton of terpenes.

[0014] As used herein, the term "terpene synthase" refers to an enzyme that enzymatically modifies the terpene precursors isopentenyl pyrophosphate (IPP), dimethylallyl pyrophosphate (DMAPP) or polyprenyl pyrophosphate to produce terpene or terpenoid compounds. Terpene synthase refers to an enzyme that catalyzes complex carbocation-induced cyclization, rearrangement, and elimination reactions that allow the conversion of several acyclic prenyl diphosphate substrates into a large chemical library of terpenes or terpenoids.

[0015] The term "inducer" as used herein refers to a factor or substance that acts to stimulate the rate of transcription from an inducible promoter. An inducer may directly stimulate transcription from a promoter by physically interacting with the promoter or with a DNA sequence in the vicinity of the promoter. Alternatively, an inducer may indirectly stimulate transcription from a promoter by interacting with a molecule that physically interacts with the promoter or with a DNA sequence in the vicinity of the promoter. An inducer may be a biomolecule or a factor such as light and temperature. Examples of chemical inducers include lactose, arabinose, tryptophan, allolactose, and isopropyl β-D-1-thiogalactopyranoside (IPTG).

[0016] As used herein, the term "promoter" refers to a region of DNA that initiates transcription of a gene. A promoter may be a major promoter, a minor promoter, or a selective promoter. A major promoter is the promoter most frequently used for transcription of a gene. A promoter may be a constitutive promoter or an inducible promoter. A constitutive promoter is a promoter that is always active. As used herein, the term "inducible promoter" refers to a promoter that can be controlled in the presence of an inducer, which may include a certain biomolecule. Examples of inducible promoter systems in E. coli include the Tet-on system, the Tet-off system, the T7 system, the Trp system, the Tac system, and the Lac system.

[0017] In the context of this application, the term "ribosome binding site" refers to a site in an mRNA molecule to which a ribosome can bind and direct the ribosome to select the correct start codon for initiating translation. Ribosome binding sites are particularly important for controlling protein translation in prokaryotes. In some prokaryotes, this polynucleotide sequence in the mRNA is called the Shine-Dalgarno sequence, which base pairs with the 16S RNA of the ribosome. [Brief description of the drawings]

[0018] The invention will be better understood with reference to the detailed description when considered in conjunction with the non-limiting examples and the accompanying drawings.

[0019] [Figure 1A] Figure 1A shows the biosynthetic pathway of viridiflorol / amorphadiene and its production using lactose titration. Figure 1A shows the biosynthetic pathway: module 1 AHT contains atoB, hmgS and thmgR, module 2 MPPI contains mevk, pmk, pmd and idi, and module 3 VI contains vs or ads and ispA (or fpps). Abbreviations for compounds: HMG-CoA, 3-hydroxy-3-methyl-glutaryl-coenzyme A; MVA, mevalonate; MVAP, phosphomevalonate; MVAPP, diphosphomevalonate; IPP, isopentenyl pyrophosphate; DMAPP, dimethylallyl pyrophosphate; FPP, farnesyl pyrophosphate. Dashed arrows indicate multiple enzymatic steps. The expressed genes encode the following enzymes: atoB, acetoacetyl-CoA thiolase; hmgS, HMG-CoA synthase; thmgR, truncated HMG-CoA reductase; mevk, mevalonate kinase; pmk, phosphomevalonate kinase; pmd, mevalonate pyrophosphate decarboxylase; idi, IPP isomerase; fpps, FPP synthase; vs, viridiflorol synthase; ads, amorpha-4,11-diene synthase. [Figure 1B] Figure 1B shows the biosynthetic pathway of viridiflorol / amorphadiene and its production using lactose titration. Figure 1B shows the specific viridiflorol yield (mg / g dry cell weight (DCW)) against different inducer concentrations for the six strains. [Figure 1C] Figure 1C shows the biosynthetic pathway of viridiflorol / amorphadiene and its production using lactose titration. Figure 1C shows the titer (mg / L) against different inducer concentrations for the six strains. [Figure 1D]Figure 1D shows the biosynthetic pathway of viridiflorol / amorphadiene and its production using lactose titration. Figure 1D shows mathematical modeling of the optimal range of inducer dosage based on Sum x M3, where "Sum" is M1 + M2 + M3, and M1, M2, M3 referred to the PS of modules 1, 2, and 3, respectively. [Figure 1E] Figure 1E shows the biosynthetic pathway of viridiflorol / amorphadiene and its production using lactose titration. Figure 1E shows mRNA data for different modules using inducer titration. Error bars, mean ± standard deviation, n=6. [Figure 2A] Figure 2A shows the transcription optimization and modeling. Figure 2A shows the specific yield (mg / g DCW) for 27 transcriptionally regulated strains. Here, the top three strains (#12, #21, and #24) are highlighted. [Figure 2B] Figure 2B shows the optimization and modeling of transcription. Figure 2B shows the model predicted yield vs. experimental yield. The regression model used a polynomial. [Figure 2C] Figure 2C shows the optimization and modeling of transcription. Figure 2C shows the two-factor interaction plot. [Figure 2D] Figure 2D shows the optimization and modeling of transcription. Figure 2D shows the correlation between predicted yield and M1, M2 or M3. [Figure 2E] Figure 2E shows the optimization and modeling of transcription. Figure 2E is a triangular plot of viridiflorol (mg / g DCW) versus M1, M2 and M3. [Figure 3A] Figure 3A shows translation optimization and correlation. Figure 3A shows the correlation between viridiflorol specific yield (mg / g DCW) and computationally predicted ribosome binding site (RBS) strength or translation initiation rate (TIR). [Figure 3B] Figure 3B shows the translation optimization and correlation. Figure 3B shows the correlation between titer (mg / L) and TIR. In this figure, the control (#12), the top five (#38, #39, #40, #41 and #46) and the bottom three producers (#34, #35 and #31) are highlighted. [Figure 3C] Figure 3C shows the translation optimization and correlation. Figure 3C shows RBS Library I and their nucleic acid sequences. Design 1 (SEQ ID NO: 60) refers to the design of RBS sequences in RBS Library I. The first base "W" of Design 1 can be adenine or thymine. The second base "D" of Design 1 can be adenine, guanine or thymine. [Figure 3D] FIG. 3D shows the translation optimization and correlation. FIG. 3D shows the RBS library II and their nucleic acid sequences. Strains no. 12 and no. 31-no. 52 had different upstream sequences. The upstream nucleic acid sequences of no. 12 and no. 31-no. 52 were aacaggaggaattaaccacgacggaagcgtcaaagcacactaaatagact (SEQ ID NO: 58) and actataggggaattgtgagcggataacaattcccctccactaaatagact (SEQ ID NO: 59), respectively. Design 2 (SEQ ID NO: 61) refers to the design of the RBS sequences in RBS library II. The first base "N" of Design 2 can be any base. The second base "K" of Design 2 can be guanine or thymine. The fifth base "R" of Design 2 can be adenine or guanine. [Figure 4A] Figure 4A shows viridiflorol synthase (VS) mutations for improved activity. Figure 4A shows the relative activity of 48 single mutations selected based on the phylogeny of fungal terpene synthases. [Figure 4B] Figure 4B shows viridiflorol synthase (VS) mutations for improved activity. Figure 4B shows viridiflorol titers of wild type, single mutant, double mutant and triple mutant. Error bars, mean ± standard deviation, n=3. [Figure 4C] Figure 4C shows viridiflorol synthase (VS) mutations for improved activity. Figure 4C shows the specific yield of wild type, single mutant, double mutant and triple mutant. Error bars, mean ± standard deviation, n=3. [Figure 4D]Figure 4D shows viridiflorol synthase (VS) mutations for improved activity. The VS homolog model and six selected amino acid mutation sites (G227C, L249F, E267S, V314Y, M318V, and G326A) in the model are highlighted. The ligand pyrophosphate and magnesium are labeled. [Figure 4E] Figure 4E shows the viridiflorol synthase (VS) mutations for improved activity. Figure 4E shows the distance (in Angstroms) between the pyrophosphate of the ligand and three amino acids (G227, E267, and V314). [Figure 5A] Figure 5A shows the combination of truncations and mutations of VS with auxotrophic design. Figure 5A shows the primary structure of VS. Amino acids 2-85 and within the active site are labeled. [Figure 5B] Figure 5B shows the combination of VS truncations and mutations with auxotrophic design. Figure 5B shows viridiflorol titers in ZYM and synthetic media. Error bars, mean ± standard deviation, n = 3. Statistical significance of viridiflorol yields was expressed as *P < 0.05 (two-tailed Student's t-test). [Figure 5C] Figure 5C shows the combination of VS truncations and mutations with auxotrophic design. Figure 5C shows the specific viridiflorol yield in ZYM and synthetic media. Error bars, mean ± standard deviation, n = 3. Statistical significance of viridiflorol yield was expressed as *P < 0.05 (two-tailed Student's t-test). [Figure 5D] Figure 5D shows the combination of truncations and mutations of VS with auxotrophic design. Figure 5D shows SDS-PAGE gel and Western blot of wild-type (WT, 48 kDa) and truncated (del2-85, 39.2 kDa) VS. An empty plasmid was used as a control. IspA was expressed in the same operon as VS. [Figure 6A] Figure 6A shows fed-batch fermentation of viridiflorol and amorphadiene. Figure 6A shows the time course profiles of dry cell weight (DCW) and viridiflorol titer (mg / L). [Figure 6B] Figure 6B shows fed-batch fermentation of viridiflorol and amorphadiene. Figure 6B shows a summary of all the methods used in this study and their viridiflorol titers achieved. [Figure 6C] Figure 6C shows fed-batch fermentation of viridiflorol and amorphadiene. Figure 6C shows the time course profiles of dry cell weight (DCW) and amorpha-4,11-diene titer (mg / L). The dashed arrows indicate the induction time. Error bars, mean ± standard deviation, n=2. [Figure 7(1)] Figure 7 shows the GC chromatograms and mass spectra of viridiflorol and amorphadiene. β-caryophyllene was added to the medium as an internal control. The retention index and mass spectrum were used to confirm the viridiflorol product. [Figure 7(2)] This is a continuation of Figure 7(1). [Figure 7(3)] This is a continuation of Figure 7(2). [Figure 8] Figure 8 shows the data fitting of viridiflorol yield (mg / g DCW) and lactose concentration for six strains. Two equations were used for the data fitting. The relative strength of the promoter for each module was listed in the figure. EQ1 was used to simulate strains no. 1, no. 7, no. 9, no. 12 and no. 24, and EQ2 was used to simulate strain no. 27. [Figure 9] Figure 9 shows the correlation of optimal inducer dose with different parameters (M1, M2, M3, Sum and Sum x M2). M1, M2, M3 referred to the PS of modules 1, 2 and 3, respectively. Sum referred to the sum of M1, M2 and M3. Optimal inducer dose correlated better with Sum x M2 than other parameters. "■" refers to the upper limit of optimal inducer dose. "●" refers to the lower limit of optimal dose. [Figure 10A] Figure 10A shows the validation of the inducer model. Figure 10A shows the sum x VI values ​​for the 27 strains. [Figure 10B]Figure 10B shows validation of the inducer model. Figure 10B shows yields across different inducer doses for six strains (#2, #5, #13, #17, #20 and #25). [Figure 10C] Figure 10C shows the validation of the inducer model. The filled (black) data points were the data of the six strains for validation. "●" indicates the upper limit of the optimal inducer dose. "■" indicates the lower limit of the optimal inducer dose. The open circles and squares are the initial randomly selected strains (#1, #7, #9, #12, #24, and #27 in Figure 8). The filled circles and squares are the validation strains (#2, #5, #13, #17, #20, and #25). [Figure 11] Figure 11 shows the 4D and triangular plots of viridiflorol titer versus three modules. M1, M2, M3 referred to the promoter strength (PS) of modules 1, 2 and 3, respectively. The shade and size of the circle indicate the viridiflorol titer. Shade from dark to light: low to high titer. Size from small to large: low to high titer. Both figures represent the effect of the promoter strength of each module on viridiflorol titer. [Figure 12] Figure 12 shows an SDS-PAGE gel for viridiflorol synthase (VS) with RBS of different strengths. IspA was co-expressed in the same operon as VS. VS, 48 kDa; IspA, 33 kDa. [Figure 13] Figure 13 shows the homology model and predicted active site of VS. The ligand pyrophosphate is shown in salmon and magnesium in brick. Hydrogen bonds between the substrate, cofactor and active site are shown as dashed yellow lines. [Figure 14(1)] Figure 14 shows the transmembrane topology and signal peptide prediction for VS using Phobius and the Signal P4.1 server. Both of these web tools predicted the presence of a signal peptide region for VS. The S score is the signal peptide score, where amino acids above a threshold are predicted to be in the signal peptide region. [Figure 14(2)]This is a continuation of Figure 14(1). [Figure 15A] Figure 15A shows inducer titration for viridiflorol production of strain del2-85 (truncated VS). Figure 15A shows titers in mg / L. Strains were induced with lactose or IPTG in synthetic medium. [Figure 15B] Figure 15B shows inducer titration for viridiflorol production of strain del2-85 (truncated VS). Figure 15B shows the specific yield in mg / g DCW. Strains were induced with lactose or IPTG in synthetic medium. [Figure 15C] Figure 15C shows inducer titration for viridiflorol production of strain del2-85 (truncated VS). Figure 15C shows optical density OD. Strains were induced by lactose or IPTG in synthetic medium. [Figure 16] Figure 16 shows the results of the plasmid stability check. Wild-type and auxotrophic strains were grown in synthetic medium with or without antibiotics. After about 20 generations, the plasmid copy number was measured by qPCR and compared with the plasmid copy number of the first generation. Here, p15A-cam-mevK-pmk-pmd-idi(L2-5) was chosen because it was found to be the most unstable. The data in this figure showed that the auxotrophic strain had better stability than the wild-type strain. The qPCR primers for this plasmid were rt-mk-890-f and rt-mk-985-r. The primers for genomic DNA were cysG-f / r (Table 1). The plasmid copy number was calculated as the ratio of the copy number of the plasmid gene and the genomic gene. [Figure 17A] Figure 17A shows the plasmid map of the vector used. Figure 17A shows the plasmid map of the vector containing the polynucleotide sequences encoding the atoB, hmgS and truncated hmgR genes. [Figure 17B]Figure 17B shows the plasmid map of the vector used. Figure 17B shows the plasmid map of the vector containing the polynucleotide sequence encoding the genes mevK, pmK, pmd and idi. [Figure 17C] Figure 17C shows the plasmid map of the vector used, which contains two genes (ads, the amorphadiene synthase gene and ispA). [Figure 17D] Figure 17D shows the plasmid map of the vector used. Figure 17D shows the plasmid map of the vector containing the polynucleotide sequence encoding two genes (vs, the viridiflorol synthase gene and ispA). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] In a first aspect, the present invention provides a method for increasing terpenoid production in a host cell producing one or more terpenoids, comprising: a) providing a host cell that produces one or more terpenoids, the host cell comprising a vector comprising a polynucleotide sequence encoding a terpene synthase enzyme; b) the vector, i. introducing an inducible promoter operably linked to the terpene synthase enzyme; and ii. to introduce a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of said terpenoid compared to a wild-type ribosome binding site; and modifying the c) determining the dosage of an inducer capable of inducing said inducible promoter; d) culturing said host cell in a medium in the presence of said inducer at said dosage determined from step c); e) isolating the terpenoids from the medium. The present invention relates to a method for providing the same.

[0021] In one embodiment, the vector further comprises a polynucleotide sequence encoding the ispA gene.

[0022] The host cell may include one or more additional vectors. In one embodiment, the one or more additional vectors include a polynucleotide sequence encoding one or more genes of the mevalonate pathway operably linked to an inducible promoter. The genes of the mevalonate pathway may include, but are not limited to, atoB, hmgS, truncated hmgR, mevK, pmK, pmd, and idi. The gene truncated hmgR produces a protein having a deletion of amino acids at positions 2-553 of hmgR (SEQ ID NO: 26). The genes of the mevalonate pathway may be present in one module or may be divided into multiple metabolic modules, and each module may be expressed in a plasmid vector.

[0023] In another embodiment, the host cell further comprises two additional vectors, each vector comprising a polynucleotide sequence encoding a gene in the module, in yet another embodiment, a first additional vector comprises a polynucleotide sequence encoding the mevalonate pathway genes atoB, hmgS and a truncated hmgR, and a second additional vector comprises a polynucleotide sequence encoding the mevalonate pathway genes mevK, pmK, pmd and idi.

[0024] In one embodiment, the inducible promoter operably linked to the polynucleotide sequence encoding the terpene synthase enzyme may be selected from the group consisting of a T7 promoter, a Lac promoter, a Trp promoter, a Tac promoter, a tetracycline (Tet) inducible promoter, a LasR quorum sensing promoter, and an EL222 light inducible promoter.

[0025] In one embodiment, the inducible promoter is a T7 promoter. In another embodiment, the T7 promoter is a variant of the wild-type T7 promoter. A variant of the T7 promoter refers to a T7 promoter having a polynucleotide sequence that differs from the wild-type T7 promoter by one or more nucleotides. A variant of the T7 promoter is generated using mutations. A variant of the T7 promoter may be selected from the group consisting of TM1, TM2, and TM3. In one embodiment, TM1, TM2, and TM3 have the polynucleotide sequences of SEQ ID NO:28, SEQ ID NO:29, and SEQ ID NO:30, respectively.

[0026] In one embodiment, the inducible promoter in each vector is the same T7 promoter variant. In another embodiment, the inducible promoter in each vector is a different T7 promoter variant.

[0027] In one embodiment, the inducible promoter in the vector comprising the polynucleotide sequence encoding the terpene synthase enzyme is TM1, the inducible promoter in the first further vector comprising the atoB, hmgS and truncated hmgR genes is TM2, and the inducible promoter in the second further vector comprising the mevK, pmK, pmd and idi genes is TM3.

[0028] In one embodiment, in the vector comprising a polynucleotide sequence encoding the terpene synthase enzyme, the polynucleotide sequence encoding the RBS is located upstream of the polynucleotide sequence encoding the terpene synthase enzyme.

[0029] In another embodiment, the polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:13, SEQ ID NO:14, SEQ ID NO:15, SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22 and SEQ ID NO:23. In yet another embodiment, the polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17.

[0030] The inducer capable of inducing the inducible promoter operably linked to the terpene synthase enzyme may be selected from the group consisting of lactose, isopropyl β-D-1-thiogalactopyranoside (IPTG), tryptophan, arabinose, tetracycline, or light. In one embodiment, the inducer is lactose or IPTG.

[0031] The inducer may be added to the medium from the beginning and may be present in the medium throughout the growth of the host cells, or the inducer may be added to the medium after the host cells have grown to optimal density.

[0032] In one embodiment, inducer dosage is determined using a regression model of promoter strength and terpenoid yield. The regression is based on experimental sampling, where inducer dosage is adjusted and the corresponding terpenoid yield is measured. In one embodiment, an optimal range for a selected strain is identified and different mathematical equations are used to fit the correlation with promoter strength of different modules.

[0033] In another embodiment, the inducer dosage is determined using the product of the total promoter strength and the strength of the promoters operably linked to the mevK, pmK, pmd and idi genes. In one embodiment, the inducer dosage is determined by multiplying M2 by the sum of M1, M2 and M3, where M1 refers to the promoter strength of the promoters operably linked to the atoB, hmgS and truncated hmgR genes, M2 refers to the promoter strength of the promoters operably linked to the mevK, pmK, pmd and idi genes, and M3 refers to the promoter strength of the promoters operably linked to the viridiflorol synthase and ispA genes.

[0034] In one embodiment, the inducer dosage is about 0.001 to 0.5 mM IPTG or about 0.1 to 50 mM lactose, in another embodiment, the inducer dosage is about 0.01 to 0.2 mM IPTG or about 1 to 25 mM lactose.

[0035] In the methods of increasing terpenoid production in a host cell described herein, the host cell may be a bacterial host cell. In one embodiment, the bacterial host cell is an E. coli cell.

[0036] The E. coli strain may be selected from the group consisting of BL21, BL21(DE3), K-12(RV308), K-12(HMS174), K-12(MG1655), W strain (ATCC9637, and JM109(DE3). In one embodiment, the E. coli is a MG1655 strain. In another embodiment, the E. coli is a BL21(DE3) strain.

[0037] In one embodiment, the E. coli cell comprises a polynucleotide encoding a T7 RNA polymerase integrated into its genome, hi another embodiment, the E. coli is a MG1655 strain comprising a polynucleotide encoding a T7 RNA polymerase integrated into its genome.

[0038] In one embodiment, the host cell is defective in at least one gene involved in amino acid synthesis.

[0039] In one embodiment, the vector comprising the polynucleotide sequence encoding a terpene synthase enzyme and / or the one or more further vectors further comprise a polynucleotide sequence encoding the one or more genes for which the host cell is deficient. The polynucleotide sequence encoding the one or more genes for which the host cell is deficient may be located on the same or a different vector as the vector comprising the polynucleotide sequence encoding the terpene synthase enzyme. In one embodiment, the one or more genes for which the host cell is deficient are present on only one vector. The one or more genes involved in amino acid synthesis for which the host cell is deficient encode one or more enzymes that produce one or more amino acids.

[0040] The one or more genes that the host cell is defective may be selected from the group consisting of aroA, aroB, and aroC.

[0041] In one embodiment, the vector comprising a polynucleotide sequence encoding a terpene synthase enzyme further comprises a polynucleotide sequence encoding the gene aroA. In another embodiment, a first further vector comprising polynucleotide sequences encoding the genes atoB, hmgS and truncated hmgR further comprises a polynucleotide sequence encoding the gene aroC, and a second further vector comprising polynucleotide sequences encoding the genes mevK, pmK, pmd and idi further comprises a polynucleotide sequence encoding the gene aroB.

[0042] In one embodiment, the host cell is deficient in aroA, aroB and aroC, and the host cell comprises: a) a vector comprising a polynucleotide sequence encoding a terpene synthase enzyme, a polynucleotide sequence encoding the gene ispA and a polynucleotide sequence encoding the gene aroA; b) a first further vector comprising a polynucleotide sequence encoding the genes atoB, hmgS, a truncated hmgR and aroC, and c) a second further vector comprising a polynucleotide sequence encoding the genes mevK, pmK, pmd, idi and aroB Includes.

[0043] In one embodiment, the terpene synthase enzyme may be selected from the group consisting of viridiflorol synthase and amorphadiene synthase, hi one embodiment, the terpene synthase enzyme is viridiflorol synthase.

[0044] In one embodiment, the viridiflorol synthase is isolated from Agrocybe aegerita (willow matsutake mushroom). In another embodiment, the amorphadiene synthase is isolated from Artemisia annua (Sweet wormwood).

[0045] In one embodiment, the viridiflorol synthase is mutated at one or more amino acid positions. The mutation may be selected from the group consisting of substitution, insertion, deletion, truncation, and combinations thereof. The mutation may be introduced by targeted mutation, random mutation, or combinations thereof. The term "truncated" when used in reference to viridiflorol synthase refers to a viridiflorol synthase sequence having one or more amino acid deletions.

[0046] In one embodiment, the mutation is a substitution of one or more amino acids at a position selected from the group consisting of positions 227, 249, 267, 314, 318, 326 of SEQ ID NO:24. In another embodiment, the mutation is selected from the group consisting of G227C, L249F, E267S, V314Y, M318V, G326A, and combinations thereof. In another embodiment, the mutation is a deletion of an amino acid at positions 2-85 of SEQ ID NO:24. In another embodiment, the mutation is selected from the group consisting of a deletion of an amino acid at positions 2-40, 2-60, or 2-85 of SEQ ID NO:24. In another embodiment, the mutation is selected from the group consisting of G227C, L249F, E267S, V314Y, M318V, G326A, and combinations thereof, and a deletion of an amino acid at positions 2-85 of SEQ ID NO:24. In yet another embodiment, the mutations are G227C, V314Y and deletion of amino acids at positions 2-85 of SEQ ID NO:24.

[0047] In one embodiment, in the methods of increasing terpenoid production described herein, the host cells are cultured in a batch, fed-batch or continuous fermentation culture. In another embodiment, the host cells are cultured in a fed-batch fermentation culture.

[0048] In another embodiment, the host cells are cultured in media that does not contain antibiotics.

[0049] In one embodiment, in the methods of increasing terpenoid production described herein, terpenoids are produced at a rate of 0.3 g / L / h to 0.5 g / L / h, in another embodiment, terpenoid production is increased by at least 2000 fold compared to a host cell producing one or more terpenoids comprising a vector comprising a polynucleotide sequence encoding an unmodified terpene synthase enzyme.

[0050] In one embodiment, a host cell comprising at least one vector, said at least one vector comprising: a) a polynucleotide sequence encoding one or more genes of the mevalonate pathway operably linked to an inducible promoter; b) a polynucleotide sequence encoding the ispA gene; c) a polynucleotide sequence encoding a terpene synthase enzyme operably linked to an inducible promoter; and d) a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of a terpene synthase enzyme compared to a wild-type ribosome binding site. A host cell comprising the

[0051] In one embodiment, the polynucleotide sequences encoding the terpene synthase enzyme and the ispA gene are located on the same vector, and the polynucleotide sequences encoding one or more genes of the mevalonate pathway are located on a different vector. The mevalonate pathway genes may include, but are not limited to, atoB, hmgS, truncated hmgR, mevK, pmK, pmd, and idi. In yet another embodiment, the polynucleotide sequences encoding the terpene synthase enzyme and the ispA gene are located on the same vector, and the polynucleotide sequences encoding one or more genes of the mevalonate pathway are located on one or two different vectors.

[0052] In one embodiment, polynucleotide sequences encoding the mevalonate pathway genes atoB, hmgS, and truncated hmgR are located on a first vector, and polynucleotide sequences encoding the mevalonate pathway genes mevK, pmK, pmd, and idi are located on a second vector.

[0053] In one embodiment, the inducible promoter operably linked to the polynucleotide sequence encoding one or more genes of the mevalonate pathway or to the polynucleotide sequence encoding a terpene synthase enzyme may be selected from the group consisting of a T7 promoter, a Lac promoter, a Trp promoter, a Tac promoter, a tetracycline (Tet)-inducible promoter, a LasR quorum sensing promoter, and an EL222 light-inducible promoter.

[0054] In one embodiment, the inducible promoter is a T7 promoter. In another embodiment, the T7 promoter is a variant of the wild-type T7 promoter. The T7 promoter variant may be selected from the group consisting of TM1, TM2 and TM3. In one embodiment, TM1, TM2 and TM3 have the polynucleotide sequences of SEQ ID NO:28, SEQ ID NO:29 and SEQ ID NO:30, respectively.

[0055] In one embodiment, the inducible promoter in each vector is the same T7 promoter variant. In another embodiment, the inducible promoter in each vector is a different T7 promoter variant.

[0056] In one embodiment, the inducible promoter in the above vector comprising a polynucleotide sequence encoding a terpene synthase enzyme is TM1, the inducible promoter in the first vector comprising atoB, hmgS and a truncated hmgR gene is M2, and the inducible promoter in the second vector comprising mevK, pmK, pmd and idi genes is TM3.

[0057] In one embodiment, in the above vector comprising a polynucleotide sequence encoding a terpene synthase enzyme, the polynucleotide sequence encoding an RBS is located upstream of the polynucleotide sequence encoding the terpene synthase enzyme.

[0058] In another embodiment, the polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11, SEQ ID NO:12, SEQ ID NO:13, SEQ ID NO:14, SEQ ID NO:15, SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22 and SEQ ID NO:23. In yet another embodiment, the polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17.

[0059] In one embodiment, the host cells described herein are defective in at least one gene involved in amino acid synthesis.

[0060] In one embodiment, the at least one vector further comprises a polynucleotide sequence encoding one or more genes for which the host cell is deficient, the one or more genes involved in amino acid synthesis for which the host cell is deficient encoding one or more enzymes that produce one or more amino acids.

[0061] The one or more genes that the host cell is defective may be selected from the group consisting of aroA, aroB, and aroC.

[0062] In one embodiment, the vector comprising a polynucleotide sequence encoding the terpene synthase enzyme further comprises a polynucleotide sequence encoding the gene aroA. In another embodiment, the first vector comprising a polynucleotide sequence encoding the genes atoB, hmgS and a truncated hmgR further comprises a polynucleotide sequence encoding the gene aroC, and the second vector comprising a polynucleotide sequence encoding the genes mevK, pmK, pmd and idi further comprises a polynucleotide sequence encoding the gene aroB.

[0063] In one embodiment, the terpene synthase enzyme may be selected from the group consisting of viridiflorol synthase and amorphadiene synthase, hi one embodiment, the terpene synthase enzyme is viridiflorol synthase.

[0064] In one embodiment, the viridiflorol synthase is isolated from Agrocybe aegerita (willow matsutake mushroom). In another embodiment, the amorphadiene synthase is isolated from Artemisia annua (Sweet wormwood).

[0065] In one embodiment, the viridiflorol synthase is mutated at one or more amino acid positions. The mutation may be selected from the group consisting of substitution, insertion, deletion, truncation, and combinations thereof. The mutation may be introduced by targeted mutation, random mutation, or combinations thereof.

[0066] In one embodiment, the mutation is a substitution of one or more amino acids at positions selected from the group consisting of positions 227, 249, 267, 314, 318, 326 of SEQ ID NO:24. In another embodiment, the mutation is selected from the group consisting of G227C, L249F, E267S, V314Y, M318V, G326A, and combinations thereof. In another embodiment, the mutation is a deletion of an amino acid at positions 2-85 of SEQ ID NO:24. In another embodiment, the mutation is selected from the group consisting of G227C, L249F, E267S, V314Y, M318V, G326A, and combinations thereof, and a deletion of an amino acid at positions 2-85 of SEQ ID NO:24. In yet another embodiment, the mutation is G227C, V314Y, and a deletion of an amino acid at positions 2-85 of SEQ ID NO:24.

[0067] The host cell described herein may be a bacterial host cell. In one embodiment, the bacterial host cell is an E. coli cell.

[0068] The E. coli strain may be selected from the group consisting of BL21, BL21(DE3), K-12(RV308), K-12(HMS174), K-12(MG1655), W strain (ATCC9637) and JM109(DE3). In one embodiment, the E. coli is a MG1655 strain. In another embodiment, the E. coli is a BL21(DE3) strain.

[0069] In another embodiment, there is provided an E. coli cell deleted for genes aroA, aroB, and aroC, comprising: a. a first vector, i. a polynucleotide sequence encoding a viridiflorol synthase enzyme operably linked to a TM1 promoter, said viridiflorol synthase enzyme being mutated, said mutations being G227C, V314Y and a deletion of amino acids at positions 2-85 of SEQ ID NO:24; ii. a polynucleotide sequence encoding a RBS ribosome binding site (RBS) that increases the rate of translation initiation of a terpenoid compared to a wild-type ribosome binding site, said polynucleotide sequence being located upstream of a polynucleotide sequence encoding a viridiflorol synthase enzyme, wherein the polynucleotide sequence encoding the RBS is a polynucleotide sequence selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17; iii. a polynucleotide sequence encoding an ispA gene operably linked to a promoter; iv. A polynucleotide sequence encoding an aroA gene operably linked to a promoter A first vector comprising: b. a second vector, i. a polynucleotide sequence encoding the genes atoB, hmgS and a truncated hmgR operably linked to the TM2 promoter; and ii. A polynucleotide sequence encoding the aroC gene operably linked to a promoter and a second vector comprising c. a third vector, i. a polynucleotide sequence encoding the genes mevK, pmK, pmd, and idi operably linked to the TM3 promoter; and ii. A polynucleotide sequence encoding an aroB gene operably linked to a promoter and a third vector comprising An E. coli cell comprising:

[0070] In another aspect, an engineered viridiflorol synthase enzyme is provided comprising one or more amino acid substitutions at positions selected from the group consisting of positions 227, 249, 267, 314, 318, 326 of SEQ ID NO:24, deletions of amino acids at positions 2-85, and combinations thereof.

[0071] In one embodiment, the engineered viridiflorol synthase enzyme comprises the mutations G227C, V314Y and a deletion of amino acids at positions 2-85 of SEQ ID NO:24.

[0072] In another aspect, there is provided a method for producing viridiflorol, the method comprising culturing a host cell described herein, or an E. coli cell described herein, in a medium containing about 0.001 to 0.5 mM IPTG or about 0.1 to 50 mM lactose.

[0073] In one embodiment, there is provided a method for producing amorpha-4,11-diene, comprising culturing a host cell described herein in a medium comprising about 0.001-0.5 mM IPTG or about 0.1-50 mM lactose, in one embodiment, the medium does not contain an antibiotic.

[0074] In another aspect, a kit for producing viridiflorol is provided comprising a host cell as described herein, or an E. coli cell as described herein, together with instructions for use.

[0075] In one embodiment, a kit for producing amorpha-4,11-diene is provided that comprises a host cell as described herein together with instructions for use. EXAMPLES

[0076] The present invention will now be further described in more detail with reference to the following non-limiting examples and comparative examples, which should not be construed as limiting the scope of the present invention in any way.

[0077] Materials and Methods Strains and plasmid construction E. coli K12 MG1655 ΔrecAΔendA DE3 and MG1655 ΔrecAΔendAΔaroAΔaroBΔaroC DE3 were used for terpenoid production. Auxotrophic E. coli strains were obtained by sequential knockout of the three genes aroABC using primers aroA-F&R, aroB-F&R and aroC-F&R (Table 1) and λ red recombinase. Plasmids p15A-spec-hmgS-atoB-hmgR (L2-8) and p15A-cam-mevK-pmk-pmd-idi (L2-5) were also used. The operon crtEBI in plasmid p15A-kan-crtEBI-ispA was replaced with the codon-optimized viridiflorol synthase (VS) from Agrocybe aegerita, resulting in plasmid p15A-kan-vs-ispA. Amorpha-4,11-diene synthase (ADS) was amplified from plasmid pACM-T7-dxs-T7-idi-ADS-ispA and cloned together with ispA into the p15A-kan vector. Then, genes aroA, aroB and aroC were incorporated into plasmids p15A-kan, p15A-cam and p15A-spec, respectively. The promoters of the plasmids were mutated using an in-house cloning method modified based on the Agilent QuikChange II method. The strains and plasmids used in this study are summarized in Table 2.

[0078] [Table 1]

[0079] [Table 2]

[0080] Building an RBS library Two RBS libraries were designed for viridiflorol synthase (VS) using degenerate primers OL(24)-F&R (SEQ ID NO:37-SEQ ID NO:38) and TS-20-OL(24)-F&R (SEQ ID NO:39-SEQ ID NO:40) (Table 1), respectively. The cloning method used was the Cross-Lapping In Vitro Assembly (CLIVA) method. RBS design was aided by the RBS Calculator, version 2.0. The predicted translation efficiencies of the RBS libraries are listed in Figure 3C and Figure 3D.

[0081] statistical analysis Linear regression models were used to study the relationship between the response (viridiflorol yield) and variables (e.g. promoter strength of different modules, RBS strength or TIR of VS genes). For transcriptional optimization, our strain design consisted of three 3 The full factorial design [three factors or variables (M1, M2, and M3) and three levels for each factor (TM1=0.92, TM2=0.37, TM3=0.16)] was used. As in response surface methodology, a polynomial model (Figure 2B) was used to describe the relationship between the response and the variables. In the present invention, the model includes linear terms (e.g., M1), cross-product terms (e.g., M1×M2), and quadratic terms (e.g., M1×M2) for each factor. 2 ) The parameters (a, b, …, j) were calculated by multiple linear regression. The coefficient of determination (R 2 ) was calculated using a polynomial model and used to compare model fitting.

[0082] Enzyme Engineering Targeted point mutations were introduced into the VS using an in-house cloning method modified based on the Agilent QuikChange II method. The primers used (1F&R-52F&R) are listed in Table 1. Random mutations were introduced using the GeneMorph II Random Mutagenesis Kit (Agilent, Stratagene). Double and triple mutations were introduced sequentially using a single VS mutant as a template. For VS truncation, the same cloning method was used with the primer pair 4879_DEL_F&R. All enzyme mutants were then sequenced to confirm the mutations.

[0083] Homolog model construction A homology model of VS was constructed on the structure of epi-isodizene synthase (PDB ID: 4ltz). Sequence alignment (Figure 14) was performed using Clustal Omega (https: / / www.ebi.ac.uk / Tools / msa / clustalo / ) with default settings. A binding pocket consisting of 15 residues within 6 angstroms of the substrate was determined by PyMOL software v2.1.1. All protein structures in this study were prepared by PyMOL software.

[0084] Media used in this study ZYM medium was prepared using 1% tryptone, 0.5% yeast extract, 25 mM Na2HPO4, 25 mM KH2PO4, 50 mM NH4Cl, 5 mM Na2SO4, 2 mM MgSO4, 0.5% glycerol, 0.05% glucose and 1-30 mM α-lactose. The synthetic medium contained 10 g / L glucose, 2 g / L (NH4)2SO4, 4.2 g / L KH2PO4, 11.24 g / L K2HPO4, 1.7 g / L citric acid, 0.5 g / L MgSO4 and 10 ml / L trace element solution, with a pH of 7.0. The trace element solution (100X) contained 0.25 g / L CoCl2·6H2O, 1.5 g / L MnSO4·4H2O, 0.15 g / L CuSO4·2H2O, 0.3 g / L H3BO3, 0.25 g / L Na2MoO4·2H2O, 0.8 g / L Zn(CH3COO)2, 5 g / L Fe(III) citrate, and 0.84 g / L EDTA, with a pH of 8.0.

[0085] Test tube culture of E. coli strains For strain optimization, cells were grown in 1 mL ZYM medium in 14 ml BD Falcon™ tubes at 28°C / 250 rpm for 2-3 days. In addition, 200 μL dodecane (containing 500 mg / L β-caryophyllene as an internal control) was used to extract viridiflorol during cell culture. When synthetic medium was used, cells were initially grown at 37°C / 250 rpm until OD600 reached approximately 0.8, induced with 0.01-0.15 mM IPTG (or 1-30 mM lactose), and then grown at 28°C for 2 days. For wild-type strains, antibiotics (34 μg / ml chloramphenicol, 50 μg / ml kanamycin, and 50 μg / ml spectinomycin) were added to the medium to maintain the three plasmids.

[0086] Bioreactor fermentation for viridiflorol production Two 250 ml Mini Bioreactors (Applikon Biotechnology) were used in this study with a working volume of 100–200 mL. Cells (-80°C stock) were grown in 10 ml of synthetic medium at 37°C for 48 h. Five milliliters of the grown cell suspension (OD ∼5) was harvested, washed, and inoculated into 100 ml of synthetic medium in the bioreactor. When the OD reached ∼5, the feed solution (500 g / L glucose and 5 g / L MgSO4) was added to the bioreactor at a rate of 0.6–1.8 mL / h for ∼3 h, and another feed cycle was initiated by a DO trigger or an offline check of residual glucose (e.g., DO suddenly increased or glucose concentration dropped below 3 g / L). This feed cycle was repeated until induction. When the OD reached ∼30–40 (16–18 h after inoculation), the cells were induced with 0.1 mM IPTG. After induction, constant feeding rates of 5 g / L / h glucose and 0.05 g / L / h MgSO4 were maintained. The culture temperature was adjusted to 30°C, and 20% (v / v) dodecane containing 1 g / L caryophyllene was fed to the bioreactor. During fermentation, the dissolved oxygen level was maintained at 30% (800-2000 rpm) by supplying filtered air at a gas rate of 1.5 vvm. The pH of the culture was controlled at 7.0 using 28% ammonia solution. Fed-batch experiments were carried out in synthetic medium without antibiotics.

[0087] Theoretical yields of viridiflorol and amorphadiene It is known that the production of isoprene or isopentenyl pyrophosphate (IPP) via the mevalonate pathway under aerobic fermentation requires three acetyl-coenzyme A (AcCoA), three ATP, and two NAD(P)H. (1) 1.5 glucose + 2 O2 → 3 AcCoA + 3 ATP + 3 CO2 + 6 NAD(P)H (glycolysis) (2) 3AcCoA + 2 NAD(P)H → MVA (3) MVA + 3ATP → IPP + CO2 (4) 3IPP → Viridiflorol / Amorphadiene Overall 4.5 Glucose + 9O2 → Viridiflorol / Amorphadiene + 12CO2+ 15H2O and the mass yields of viridiflorol and amorphadiene relative to glucose are 27.4% and 25.2%, respectively.

[0088] Terpenoid Quantification Terpenoid samples were prepared by diluting 10-20 μl of the organic layer into 1000 μl hexane. These samples were analyzed on an Agilent 7890 gas chromatograph equipped with an Agilent 5977B MSD. Samples were injected onto an Agilent VF-WAXms column at 240°C with a split ratio of 40:1. The oven program started at 100°C for 1 min, ramped to 150°C at 50°C / min, then ramped to 240°C at 15°C / min and held at 240°C for an additional 2 min. Compound concentrations were calculated by interpolation using a standard curve prepared with authentic viridiflorol standards (Santa Cruz Biotechnology, USA) or relative to the internal standard β-caryophyllene. The mass spectrometer was operated in EI mode with full scan analysis (m / z 30–300, 2 spectra / s).

[0089] RNA purification and quantitative PCR assay Total RNA samples from E. coli cells were prepared using the PureLink® RNA Mini Kit (Thermo Fisher Scientific, USA) according to the manufacturer's instructions. RNA samples were collected 1 hour after induction (approximately OD of 1.5). 600The strains were collected in biological triplicates (at 0.05°C, 0.05°F, ...

[0090] result Example 1: Transcription optimization and statistical analysis The mevalonate pathway genes of Saccharomyces cerevisiae were grafted into the E. coli strain MG1655. Briefly, the mevalonate pathway genes were divided into two metabolic modules, module 1 or AHT, containing genes atoB, hmgS and a truncated hmgR, and module 2 or MPPI, containing genes mevK, pmK, pmd and idi (Figure 1). Farnesyl diphosphate (FPP) synthase (fpps, or ispA) was on module 3 or VI along with the VS genes. These three modules were expressed on three plasmids (Table 2). It has been previously shown that such a modular structure works for both carotenoids and sesquiterpenes. And this naturally modular approach by dividing genes into modules according to their location in the metabolic pathway (e.g., upstream, middle and downstream) allows for easier diagnosis of pathway bottlenecks. Notably, fungal VS produced viridiflorol as a single product (Figure 7, where β-caryophyllene was the internal control). It has been shown that the above three T7 promoter variants (TM1, TM2 and TM3, whose strengths were 92%, 37% and 16% of the strength of the T7 promoter, respectively) were sufficient to cover the initial search resolution. Thus, full permutation (three promoters and three modules, 3 3 Twenty-seven different strains (strain no. 1 to no. 27, Table 3) covering a range of genera (generally 100 or 2000) were constructed. Using different promoters, modules were transcriptionally integrated to prevent the accumulation of toxic intermediates and thus maximize product yield.

[0091] [Table 3]

[0092] The inducer concentration is very important to achieve maximum yield of protein or metabolite. In general, strains with stronger promoters require less amount of inducer than strains with weaker promoters. However, little is known about the absolute inducer dosage for each strain, and it is tedious to test many strains with many different concentrations of inducer. Therefore, a study was carried out to investigate the mathematical relationship between inducer concentration and viridiflorol yield through autoinduction in ZYM medium with lactose as inducer. This is because autoinduction medium is more convenient and can avoid human error and missed induction timing, especially when the growth rate of different strains varies. The experiment was started with six randomly selected strains (no. 1, no. 7, no. 9, no. 12, no. 24 and no. 27). First, the data in Fig. 1B and Fig. 1C reaffirm that inducer dosage is very important for viridiflorol production, and some strains have a clear optimum inducer dosage. Second, we measured the relationship with an R of 0.85 to 0.95. 2 Two equations were sufficient to simulate the values ​​(Figure 8). Moreover, the optimal range (upper and lower limits) of inducer dosage determined by the data in Figure 8 correlated with M1, M2, M3, the sum, and the product of the sum and M1. Here, "M1", "M2" and "M3" refer to the promoter strengths (PS) of modules 1, 2 and 3 of each strain, respectively, and "sum" refers to the sum of M1, M2 and M3 (e.g., for strain number 7, M1, M2 and M3 were 0.92, 0.92 and 0.16, respectively, the sum was 0.92+0.92+0.16=2, and the sum x M3 was 2 x 0.16=0.32). The optimal inducer dosage was found to be better (inversely) correlated with the product of the sum and M1 (Figure 9).

[0093] A linear regression model was used to predict the optimal inducer range for all 27 strains (Figure 1D and Figure 10A). Based on this model, the optimal range would gradually shift from 5-15 mM to 25-30 mM for the 27 strains, with one of the two concentrations (15 and 25 mM) falling within the optimal range for all 27 strains. To validate this model, another six strains (no. 2, no. 5, no. 13, no. 17, no. 20, and no. 25) were tested, and the experimental data were consistent with our predictions (Figures 10B and 10C). Furthermore, to better understand the relationship between viridiflorol production and inducer dosage, mRNA data was measured. For most of the above genes, the transcription levels were found to plateau at 10-15 mM lactose (Figure 1E), while for the others, they were found to plateau at 25-30 mM, consistent with our inducer model in Figure 1D.

[0094] Therefore, the 27 strains were tested at their own optimal inducer dose (15 or 25 mM). Among them, the top three producers were strains no. 12, no. 21 and no. 24, with no. 12 producing the highest amount of viridiflorol (specific yield 174±8 mg / g dry cell weight, or DCW, titer 275±12 mg / L, FIG. 2A and FIG. 11). A polynomial model was used to simulate and predict the response (viridiflorol yield) over a range of input parameter values ​​(M1, M2 and M3). The model was fitted to the experimental data (R 2 =0.78, Fig. 2B). Based on this model, viridiflorol production was positively correlated with M3 but negatively correlated with M2. The predicted optimum values ​​for M1, M2 and M3 were 0.4±0.1, 0.1±0.1 and 0.8±0.1, respectively (Fig. 2C and D). Among the 27 strains, strain no. 12 (whose M1, M2 and M3 were 0.37, 0.16 and 0.92, respectively) was closest to the optimum. In fact, no. 12 had the highest production, about 180 mg / g DCW or 283 mg / L (Fig. 2A and Fig. 11).

[0095] Example 2: Translational optimization of viridiflorol Through statistical analysis (Figure 2), it was observed that viridiflorol yield was positively correlated with the transcription of module 3. This led to the hypothesis that viridiflorol yield may still be limited by insufficient enzyme activity for module 3. Module 3 contains two enzymes, VS and ispA. IspA is a native E. coli enzyme with high expression and good activity, and is therefore unlikely to be the limiting step. In contrast, as a fungal terpene synthase, VS was not detectable in sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) gels when co-expressed with ispA in the same operon (Figure 12). Thus, production was more likely limited by insufficient VS activity. To test this hypothesis, we designed two ribosome binding site (RBS) libraries for VS, covering a wide range of translation initiation rates (TIR) ​​from 4000 to 18,206 au. (Strains no. 31-52, Figure 3, where the TIR was not experimental but computationally predicted by the RBS calculator, https: / / salislab.net / software / ). The top producer among the 27 strains, strain no. 12 (whose TIR was 11,608 au), was used as the parent strain for translation optimization (strains no. 31-52). As hypothesized above, adjustment of the VS RBS had a significant effect on viridiflorol production (Figures 3A and 3B). Among all strains, strain no. 38 produced the highest amount of viridiflorol (262 ± 28 mg / g DCW, 511 ± 37 mg / L), which was about 50% higher than its parent strain no. 12. A weak but positive correlation between viridiflorol yield and predicted TIR was observed (R for specific yield and titer, respectively). 2= 0.5 and 0.55). There are two possible explanations for this weak correlation. First, some of the predicted TIRs were not accurate. Second, other unknown post-transcriptional effects may have led to the difference between the experimental and predicted VS. To test the first hypothesis, three representative groups of RBSs were selected: no. 34 and no. 35 with low viridiflorol production and low TIR; no. 52 with high viridiflorol production and low TIR; no. 38 and no. 39 with high viridiflorol yield and high TIR. For any of them, the VS band (48 kDa) was not observed on the SDS-PAGE gel, even though the ispA (polycistronic gene of VS) band (33 kDa) was very clear (Figure 12).

[0096] Example 3: Enzyme engineering of viridiflorol synthase After transcriptional and translational optimization studies, it was observed that insufficient VS activity limited viridiflorol production. The synthetic biology toolbox used in this regard could not resolve this flux limitation beyond what was already available. To overcome this, enzyme engineering approaches were further explored. A homology model of fungal VS was constructed based on the structure of epi-isodizene synthase (PDB ID: 4ltz). Pyrophosphate (POP) and magnesium (Mg) were used as ligands, and the amino acids in the binding pocket are shown in Figure 13. Using VS as a template, BLAST searches were performed in UniProt (https: / / www.uniprot.org / ) and the Joint Genome Institute fungal database (JGI, https: / / jgi.doe.gov / ). Fourteen putative fungal terpene synthases sharing high similarity with VS or AAE3_13291 were identified (Figure 14). Based on this homology model and the alignment of the 15 enzymes, a small library was designed to target 16 sites located near the active site or in conserved regions (such as the magnesium-binding motif). Instead of using saturation mutagenesis, in the absence of suitable high-throughput screening, phylogenetic-based mutagenesis using the 15 enzymes (Figure 4A) was chosen. However, all 48 selected single mutations were less active than the wild type. Specifically, L174 is crucial for VS activity, and all six mutations at that position (L174M, L174Y, L174V, L174I, L174T, and L174F) reduced viridiflorol production by more than 95%. L174 is located near the "DEYTD" (magnesium-binding motif), and therefore its mutations could affect cofactor binding and thus reduce activity. As the targeted mutations were unsuccessful, random mutations were introduced into VS by error-prone PCR. By screening about 400 colonies, one VS mutant was found to have higher viridiflorol production. Sequencing results showed that this mutant had two mutations, G227C and V314Y.Although the single mutation of V314Y reduced viridiflorol production by more than 50%, V314Y and G227C together improved viridiflorol titer by more than 38% (Figure 4B). Unlike V314, which is located in the substrate-binding pocket, G227 is distal to the active site (Figures 4D and 4E). On top of this double mutant, other mutations, L249F, E267S, M318V and G326A, which are the best four mutations in Figure 4A, were further introduced. Among these four triple mutants, V314Y-G227C-E267S increased the specific viridiflorol yield by 52% over the wild type, despite the decreased final cell density (about 17% lower than the wild type: Figures 4B and 4C). However, the mutant V314Y-G227C-M318V significantly reduced viridiflorol yield by more than 90%.

[0097] In the alignment of VS with closely related fungal terpene synthases, it was noted that VS has about 80 additional amino acids at the N-terminus than the others (Figure 14). Further studies with the signal peptide prediction programs SignalP (http: / / www.cbs.dtu.dk / services / SignalP / ) and Phobius (http: / / phobius.binf.ku.dk / ) confirmed the possibility of an N-terminal signal peptide in VS (Figure 15). This N-terminal signal peptide region would direct VS to the E. coli membrane, which could negatively affect expression and reduce contact with its substrates in the cytosol. The latter hypothesis could also explain why wild-type VS is not visible in protein gels, since membrane proteins are known to be less accessible. Therefore, we hypothesized that a shortening (truncating) of this region could be beneficial for viridiflorol production. The 2nd to 85th amino acids were then removed from the N-terminus of the wild-type and double mutant G227C-V314Y (Figure 5A). As a result, this truncation significantly increased the viridiflorol titer and specific yield by approximately 200% and 100% for wild-type and mutant VS, respectively (Figures 5B and 5C). In addition, the combination of truncation and enzyme mutation had a synergistic effect on viridiflorol production in ZYM-rich medium. Consistent with the hypothesis that wild-type VS is not visible in SDG-PAGE gels due to the N-terminal signal peptide sequence, the truncated version del2-85 (39.2 kDa) was clearly detected and further confirmed by Western blot using anti-6x his-tag antibody (Figure 5D).

[0098] Example 4: Fed-batch fermentation of auxotrophic E. coli and viridiflorol Natural media are often subject to batch-to-batch variability and are expensive. Therefore, synthetic media are preferred for large-scale industrial fermentation. More importantly, it was observed that biological replicates sometimes had very different results and that this was likely due to plasmid instability, especially for the plasmid carrying the module MPPI based on the results of restriction enzyme mapping. To solve this, an auxotrophic E. coli strain was created by deleting three aromatic amino acid synthesis genes (aroA, aroB and aroC) from the genome. These genes were then grafted onto three plasmids, respectively. Thus, since no amino acids are present in the synthetic medium, the engineered strain was forced to maintain the plasmids. Indeed, good stability was achieved for the auxotrophic strain even in the absence of antibiotics (Figure 17).

[0099] All four strains (WT, G227C-V314Y, del2-85, and the combination of del2-85 and G228C-V314Y) had higher production in the synthetic medium compared to the rich medium. However, the synergistic effect of truncation and targeted mutations was no longer observed. The underlying reason is unclear and deserves further investigation. A possible reason is that the achieved yield was no longer limited by VS activity but by other factors (e.g., availability of cofactors), and therefore the synergistic benefit could not be visualized. Strain del2-85 had a slightly higher viridiflorol yield than the strain del2-85 with G227-V314Y (Figure 5b and Figure 5c). Titration results showed that the specific yield gradually increased with increasing concentration of IPTG (Figure 16). When induced with 0.1 mM IPTG, the cells produced 1.4 g / L viridiflorol and 1.4 g / L dry cells from 10 g / L glucose.

[0100] Example 5: Fed-batch fermentation The performance of the del2-85 strain in fed-batch fermentation was tested using the synthetic medium described above. The cells were grown at mid-stage (OD 600was induced with 35). Within 62 h, the strain produced 25.7 g / L viridiflorol and 41 g / L DCW. Thus, the productivity was about 0.42 g / L / h. The total glucose used in the whole process was 151 g / L, and the process consisted of two stages, namely a biomass stage and a production stage (Figure 6A). In the production stage, only viridiflorol was produced, and in this stage, the glucose consumed was 115 g / L, and therefore the carbon yield was 0.22 g / g glucose, about 81% of its theoretical yield of the mevalonate pathway (27.4%, 1 mole of viridiflorol from 4.5 moles of glucose). It needs to be emphasized that although the theoretical yield was calculated based on zero growth conditions, this theoretical yield cannot be achieved in practice in fed-batch fermentation, where the cells are still growing. Combined with transcription, translation, enzyme, strain and process optimization, the viridiflorol titer increased 2860-fold from 0.009 g / L to 25.7 g / L (Figure 6B).

[0101] Example 6: Applying the findings to other isoprenoids To test whether the transcriptional model in Figure 2 can be used to optimize another terpenoid, amorpha-4,11-diene, a precursor to the potent antimalarial drug artemisinin, was chosen. Until recently, malaria still causes about 440,000 deaths annually. Both amorpha-4,11-diene and viridiflorol are sesquiterpenes, which share the same metabolic pathway, except for terpene synthase (Figure 1). Therefore, amorpha-4,11-diene is an ideal example to test the predictive properties of the model and the performance of the optimized strain. A strain producing 1.2 g / L of amorpha-4,11-diene in batch fermentation was obtained without the need to test many combinations of pathway designs and inducer conditions. Using the same fermentation medium and process as for viridiflorol, this strain produced 30 g / L of amorpha-4,11-diene in 80 hours with a carbon yield of 0.19 g / g glucose (FIG. 6C). The achieved carbon yield is about 75% of its theoretical yield (25.2%, 4.5 moles of glucose to 1 mole of amorpha-4,11-diene). Although this fermentation process was not fully optimized, the titers, production rates, and yields ("TRY") here for both viridiflorol and amorpha-4,11-diene were comparable to Amyris Inc's excellent results for artemisinic acid (25 g / L in 140 hours) and higher than most previously reported "TRY" data for terpenoid production in engineered microorganisms (1.5 g / L sclareol in 60 hours, 323 mg / L nerolidol in 48 hours) and plants.

[0102] [Table 4(1)]

[0103] [Table 4(2)]

[0104] Equal The above examples are presented for the purpose of illustrating the present invention and should not be construed as imposing any limitations on the scope of the present invention. It will be readily apparent that many variations and modifications may be made to the specific embodiments of the present invention described above and illustrated in the examples without departing from the principles underlying the invention. All such variations and modifications are intended to be encompassed by this application.

Claims

1. 1. A method for increasing viridiflorol production in an E. coli host cell producing one or more terpenoids, comprising: a) providing an E. coli host cell that produces one or more terpenoids, said E. coli host cell comprising a vector comprising a polynucleotide sequence encoding a terpene synthase enzyme; the terpene synthase enzyme is viridiflorol synthase isolated from Agrosybe aegerita; The vector is i. the polynucleotide sequence encoding the viridiflorol synthase enzyme operably linked to an inducible promoter, and ii. a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of the viridiflorol synthase enzyme compared to a wild-type ribosome binding site, wherein the polynucleotide sequence encoding the RBS is located upstream of the polynucleotide sequence encoding the viridiflorol synthase enzyme; the polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17, b) determining the dosage of an inducer capable of inducing said inducible promoter; c) culturing said E. coli host cells in a medium in the presence of said inducer at the dosage determined from step b); d) isolating the terpenoids from the medium; A method for providing the above.

2. 2. The method of claim 1, wherein the E. coli host cell comprises one or more additional vectors, the one or more additional vectors comprising a polynucleotide sequence encoding one or more genes of the mevalonate pathway operably linked to an inducible promoter.

3. 3. The method of claim 2, wherein the E. coli host cell further comprises two additional vectors, a first additional vector comprises polynucleotide sequences encoding the mevalonate pathway genes atoB, hmgS, and a truncated hmgR, and a second additional vector comprises polynucleotide sequences encoding the mevalonate pathway genes mevK, pmK, pmd, and idi.

4. The method of any one of claims 1 to 3, wherein the inducible promoter is a T7 promoter variant selected from the group consisting of TM1, TM2 and TM3, wherein TM1, TM2 and TM3 have the polynucleotide sequences of SEQ ID NO:28, SEQ ID NO:29 and SEQ ID NO:30, respectively.

5. The method of claim 4, wherein the inducible promoter in the vector comprising a polynucleotide sequence encoding the viridiflorol synthase enzyme is TM1, the inducible promoter in the first further vector comprising the atoB, hmgS and truncated hmgR genes is TM2, and the inducible promoter in the second further vector comprising the mevK, pmK, pmd and idi genes is TM3.

6. The method according to any one of claims 1 to 5, wherein the inducer is lactose or isopropyl β-D-1-thiogalactopyranoside (IPTG).

7. 7. The method of any one of claims 1 to 6, wherein the inducer dosage is determined using the product of total promoter strength and the strength of the promoters operably linked to the mevK, pmK, pmd and idi genes.

8. 8. The method of any one of claims 1 to 7, wherein the inducer dosage is about 0.001 to 0.5 mM IPTG or about 0.1 to 50 mM lactose.

9. 2. The method of claim 1, wherein the E. coli is a DE3 strain encoding T7 RNA polymerase integrated into its genome.

10. 10. The method of any one of claims 1 to 9, wherein the host cell is defective in at least one gene involved in amino acid synthesis.

11. A method according to any one of claims 1 to 10, wherein the vector comprising a polynucleotide sequence encoding the viridiflorol synthase enzyme and the one or more further vectors further comprise a polynucleotide sequence encoding the one or more genes for which the E. coli host cell is defective.

12. 12. The method of claim 10 or claim 11, wherein the at least one gene is selected from the group consisting of aroA, aroB and aroC.

13. 13. The method according to any one of claims 1 to 12, wherein the vector comprising a polynucleotide sequence encoding the viridiflorol synthase enzyme further comprises a polynucleotide sequence encoding the gene aroA.

14. 14. The method of any one of claims 3 to 13, wherein the first further vector comprising polynucleotide sequences encoding the genes atoB, hmgS and truncated hmgR further comprises a polynucleotide sequence encoding the gene aroC, and the second further vector comprising polynucleotide sequences encoding the genes mevK, pmK, pmd and idi further comprises a polynucleotide sequence encoding the gene aroB.

15. The method of claim 1, wherein the viridiflorol synthase is mutated at one or more amino acid positions.

16. 16. The method of claim 15, wherein the mutations are G227C, V314Y and deletion of amino acids at positions 2 to 85 of SEQ ID NO:

24.

17. 17. The method of any one of claims 1 to 16, wherein viridiflorol production is increased by at least 2000-fold compared to an E. coli host cell producing one or more terpenoids comprising a vector comprising a polynucleotide sequence encoding an unmodified terpene synthase enzyme.

18. 1. An E. coli host cell comprising at least one vector, said at least one vector comprising: a) a polynucleotide sequence encoding one or more genes of the mevalonate pathway operably linked to an inducible promoter; b) a polynucleotide sequence encoding the ispA gene; and c) a polynucleotide sequence encoding a viridiflorol synthase enzyme operably linked to an inducible promoter; d) a polynucleotide sequence encoding a ribosome binding site (RBS) that increases the rate of translation initiation of the viridiflorol synthase enzyme compared to a wild-type ribosome binding site, wherein the polynucleotide sequence encoding the RBS is located upstream of the polynucleotide sequence encoding the viridiflorol synthase enzyme; The polynucleotide sequence encoding the RBS is selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:

17. and 1. An Escherichia coli host cell comprising:

19. An Escherichia coli cell lacking the genes aroA, aroB, and aroC, a. a first vector, i. a polynucleotide sequence encoding a viridiflorol synthase enzyme operably linked to a TM1 promoter, said viridiflorol synthase enzyme being mutated, said mutations being G227C, V314Y and a deletion of amino acids at positions 2-85 of SEQ ID NO:24; ii. a polynucleotide sequence encoding a RBS ribosome binding site (RBS) that increases the rate of translation initiation of the viridiflorol synthase enzyme compared to a wild-type ribosome binding site, said polynucleotide sequence being located upstream of said polynucleotide sequence encoding the viridiflorol synthase enzyme, said polynucleotide sequence encoding said RBS being selected from the group consisting of SEQ ID NO:9, SEQ ID NO:10, SEQ ID NO:11 and SEQ ID NO:17; iii. a polynucleotide sequence encoding an ispA gene operably linked to a promoter; and iv. A polynucleotide sequence encoding the aroA gene operably linked to a promoter A first vector comprising: b. A second vector, i. a polynucleotide sequence encoding the genes atoB, hmgS and a truncated hmgR operably linked to the TM2 promoter; and ii. A polynucleotide sequence encoding the aroC gene operably linked to a promoter and a second vector comprising c. A third vector, i. a polynucleotide sequence encoding the genes mevK, pmK, pmd and idi operably linked to the TM3 promoter; and ii. A polynucleotide sequence encoding the aroB gene operably linked to a promoter and a third vector comprising coli cells containing Claim 20: i. The amino acid substitutions G227C and V314Y of SEQ ID NO:24; ii. G227C, E267S and V314Y amino acid substitutions, or iii. Deletion of amino acids at positions 2-85 2. An engineered viridiflorol synthase enzyme comprising a mutation selected from the group consisting of:

21. 21. The engineered viridiflorol synthase enzyme of claim 20, comprising the mutations G227C, V314Y and a deletion of amino acids at positions 2-85 of SEQ ID NO:

24.

22. 20. A method for producing viridiflorol, comprising culturing an E. coli host cell according to claim 18 or an E. coli cell according to claim 19 in a medium comprising about 0.001-0.5 mM IPTG or about 0.1-50 mM lactose.

23. 20. A kit for producing viridiflorol, comprising the E. coli host cell of claim 18 or the E. coli cell of claim 19 together with instructions for use.

Images