ERG20-dlGAS fusion enzyme, construction method and application

Abstract

The application discloses an ERG20-dlGAS fusion enzyme, a construction method and application, relates to the technical field of enzyme engineering, and belongs to the field of enzyme engineering. The fusion enzyme is a recombinant fusion protein formed by fusing ERG20 and an M119N mutant through a flexible connecting peptide GGGGS. The application further provides a construction method of the fusion enzyme, a recombinant host cell containing the fusion enzyme and application of the fusion enzyme in preparation of germacrene A. Experimental results show that the GA yield of the M119N-GGGGS-ERG20 fusion enzyme reaches 953.4 mg / L, is increased by about 28% compared with the M119N mutant alone, and is increased by about 2.41 times compared with the wild-type dlGAS. The application realizes high-efficiency conversion of FPP to GA through a substrate channeling effect, and provides excellent enzyme elements and a technical scheme for high-efficiency microbial synthesis of a sesquiterpene natural product.

Description

Technical Field

[0001] This invention belongs to the field of enzyme engineering technology, and particularly relates to an ERG20-dlGAS fusion enzyme, its construction method, and its application. Background Technology

[0002] Germacrene A (GA) is a biosynthetic precursor for various sesquiterpene lactones and medicinal natural products (such as artemisinin and β-elemene). The thermal rearrangement product of GA, β-elemene, exhibits broad-spectrum antitumor activity, low toxicity, and good blood-brain barrier penetration, showing promising application prospects in the treatment of lung cancer, liver cancer, and glioma. However, currently, β-elemene mainly relies on turmeric (Curcuma longa) for its synthesis. Curcuma wenyujin Extracted from plants such as , the production cost is high due to the long growth cycle of plants, the complex extraction process, and the mixed composition of the products, which seriously restricts its promotion and application.

[0003] Germacrene A synthase (GAS) is a key enzyme catalyzing the production of GA from farnesyl pyrophosphate (FPP). In Saccharomyces cerevisiae (Saccharomyces cerevisiae)... Saccharomyces cerevisiae In this process, FPP is generated by the condensation of isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP) catalyzed by farnesyl pyrophosphate synthase (ERG20 / FPS). ERG20 is a core precursor supply enzyme in sesquiterpene biosynthesis, and its catalytic efficiency and substrate transfer efficiency with downstream sesquiterpene synthases directly affect the yield of the target product.

[0004] In traditional multi-enzyme expression systems, the FPP generated by ERG20 needs to diffuse from the ERG20 active site to the downstream GAS active pocket. This diffusion process easily leads to the consumption of FPP by other competing metabolic pathways (such as sterol synthesis), resulting in the loss of intermediate metabolites and low product accumulation efficiency. Fusion enzyme technology, by linking two functionally related enzymes with a linker to form a single multifunctional protein complex, can achieve substrate channeling, significantly shortening the diffusion distance of intermediate metabolites, increasing local substrate concentration, and reducing the consumption of intermediate products by competing metabolic pathways.

[0005] By using genetic engineering techniques to fuse ERG20 and GAS for expression, the efficient transfer of FPP from ERG20 to GAS can be achieved, which has important application value for improving the microbial synthesis efficiency of GA. Summary of the Invention

[0006] The purpose of this invention is to address the technical problems of low efficiency in FPP to GAS transfer and easy consumption of intermediate metabolites by competing pathways in the prior art, and to provide an ERG20-dlGAS fusion enzyme, its construction method and application.

[0007] To achieve the above objectives, in a first aspect, the present invention provides an ERG20-dlGAS fusion enzyme, wherein the fusion enzyme is a recombinant fusion protein formed by fusing Saccharomyces cerevisiae farnesyl pyrophosphate synthase ERG20 with a mutant of Charae decidua gemmaeene A synthase M119N via a flexible linker peptide; wherein the M119N mutant is formed by mutating methionine at position 119 to asparagine based on the amino acid sequence of wild-type dlGAS.

[0008] The amino acid sequence of the wild-type dlGAS is shown in SEQ ID NO.1, the amino acid sequence of the M119N mutant is shown in SEQ ID NO.2, and the amino acid sequence of ERG20 is shown in SEQ ID NO.30.

[0009] Preferably, the flexible linker peptide is GGGGS.

[0010] Preferably, the fusion enzyme is M119N-GGGGS-ERG20, that is, the M119N mutant is located upstream and ERG20 is located downstream, and the two are linked by the GGGGS linker peptide.

[0011] Secondly, the present invention provides a nucleic acid molecule encoding the above-mentioned fusion enzyme.

[0012] Thirdly, the present invention provides a recombinant expression vector comprising the above-mentioned nucleic acid molecules.

[0013] Preferably, the recombinant expression vector is PESC-LEU.

[0014] Fourthly, the present invention provides a recombinant host cell comprising the above-described recombinant expression vector.

[0015] Preferably, the recombinant host cell is Saccharomyces cerevisiae GQ01.

[0016] Fifthly, the present invention provides a method for constructing the above-mentioned fusion enzyme, comprising the following steps: (1) Obtain the gene fragment encoding ERG20 and the gene fragment encoding the dlGAS M119N mutant, respectively; (2) The ERG20 gene fragment and the M119N gene fragment were fused by overlapping extension PCR through the coding sequence of a flexible linker peptide to obtain the fusion gene; (3) The fusion gene is cloned into an expression vector to obtain a recombinant expression vector; (4) Transform the recombinant expression vector into the host cell and express the fusion enzyme.

[0017] In a sixth aspect, the present invention provides the use of the above-mentioned fusion enzyme or the above-mentioned recombinant host cell in the preparation of gemmaene A.

[0018] In a seventh aspect, the present invention provides a method for producing gemmaene A, comprising the following steps: (1) Culture the above-mentioned recombinant host cells; (2) Fermentation culture is carried out under suitable fermentation conditions; (3) D-galactose was added during fermentation to induce expression; (4) Collect the fermentation products and extract gemmaene A.

[0019] Preferably, the fermentation conditions include: fermentation temperature of 28-30℃, oscillation rate of 200-220rpm, and fermentation time of 72-96h; the concentration of D-galactose added is 0.25-5mM.

[0020] Preferably, the concentration of D-galactose added is 1 mM.

[0021] Compared with the prior art, the present invention has the following advantages and technical effects: (1) This invention is the first to fuse the M119N mutant with the upstream precursor supply enzyme ERG20 via a flexible GGGGS linker peptide to construct the ERG20-dlGAS fusion enzyme. This fusion enzyme, through substrate channelization, enables the efficient delivery of FPP generated by ERG20 to the dlGAS-M119N active pocket, reducing intermediate diffusion loss and competitive pathway consumption. Experimental results show that the GA yield of the M119N-GGGGS-ERG20 fusion enzyme reaches 953.4 mg / L, which is approximately 28% higher than that of the M119N mutant alone (744 mg / L) and approximately 2.41 times higher than that of wild-type dlGAS (393 mg / L).

[0022] (2) The present invention systematically compared the effects of different linking peptides and different fusion directions on catalytic efficiency and determined the optimal fusion strategy—forward fusion using GGGGS flexible linking peptides, which provides an important reference for the design of fusion enzymes of sesquiterpene synthases.

[0023] (3) Through the optimization of fermentation conditions, the GA yield of the fusion enzyme strain can reach about 952 mg / L under the condition of 1 mM D-galactose induction for 72 h, providing excellent enzyme elements and engineered strains for the efficient microbial production of GA and β-elemene. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 Three-dimensional structure of dlGAS protein obtained by the Swiss-Model method for homology modeling; Figure 2 The results of the reliability assessment of the stereoconformation of the dlGAS protein; Figure 3 The diagram shows the six conformations with the lowest binding energy for molecular docking FPP-dlGAS, where a is the conformation of the first binding site, b is the conformation of the second binding site, c is the conformation of the third binding site, d is the conformation of the fourth binding site, e is the conformation of the fifth binding site, and f is the conformation of the sixth binding site. Figure 4 The three-dimensional complex structure of dlGAS and substrate FPP and the spatial distribution characteristics of the active pockets are shown. Figure 5 The results of multiple sequence alignment and conservation analysis of dlGAS and its homologous sesquiterpene synthases are shown; where a is the multiple sequence alignment results of dlGAS protein and its homologous sesquiterpene synthases from different species, and b is the sequence conservation diagram generated by the WebLogo tool. Figure 6 This is a schematic diagram of the GAS phylogenetic tree; Figure 7 For containing dlGAS Schematic diagram of the recombinant gene expression vector PESC-LEU-dlGAS; Figure 8 The results of agarose gel electrophoresis of plasmid DNA extracted and purified from recombinant Saccharomyces cerevisiae; Figure 9 The yeast growth curves of the blank group, empty vector group, and experimental group are compared; where 'a' represents the OD of the blank group, empty vector group, and experimental group. 600 The change is shown in b, which represents the linear fit of the logarithmic growth phase of the blank group, the empty vector plasmid group, and the experimental group. Figure 10 The results of total ion chromatography and mass spectrometry analysis in GC-MS determination are shown; where a is the total ion chromatogram of the engineered yeast GQ01-PESC-WT expression system, and b is the mass spectrum of the corresponding product. Figure 11The comparison shows the GA production of wild-type dlGAS and its 12 mutant strains; where a is a comparison of the extracellular GA production of wild-type dlGAS and the 12 mutant strains after 72 h of shake-flask fermentation in engineered yeast GQ01; b is the heatmap corresponding to a. Figure 12 The results show the molecular docking of wild-type dlGAS and M119N mutant; where a is a schematic diagram of molecular docking between wild-type dlGAS and substrate FPP, b is a schematic diagram of molecular docking between M119N mutant and substrate FPP, c is a schematic diagram of non-covalent interaction between wild-type dlGAS and d is a schematic diagram of non-covalent interaction between M119N mutant. Figure 13 Visualizations of the active pocket structures of wild-type dlGAS and the M119N mutant are shown below; where a is a visualization of the active pocket structure of wild-type dlGAS and b is a visualization of the active pocket structure of the M119N mutant. Figure 14 A comparison of GA yields for M119N-ERG20 fusion enzymes constructed using different fusion methods is shown in Figure a. Figure a shows the GA yields of M119N and four fusion enzyme strains: ERG20-GGGGS-dlGAS, ERG20-GSG-dlGAS, M119N-GGGGS-ERG20, and M119N-GSG-ERG20. Figure b shows the corresponding heatmap. Figure 15 The effect of fermentation time on GA synthesis by the M119N-GGGGS-ERG20 fusion enzyme strain is shown in Figure a. a) Comparison of GA yield of the M119N-GGGGS-ERG20 fusion enzyme strain at different fermentation times; b) Time-GA yield change curve. Figure 16 The graph shows the comparison of GA production of the M119N-GGGGS-ERG20 fusion enzyme strain under different D-galactose concentrations. In the graph, a shows the effect of D-galactose concentration on GA production of the M119N-GGGGS-ERG20 fusion enzyme strain, and b shows the corresponding heatmap. Figure 17 The effects of glycerol concentration on the GA yield and reaction kinetics of the M119N-GGGGS-ERG20 fusion enzyme were investigated. Among them, a) showed the effect of adding different concentrations of glycerol on the extracellular GA yield during the 72h period of fusion enzyme expression induced by 1mM galactose at 24h of fermentation; b) showed the kinetic parameters under different glycerol conditions analyzed using the Michaelis-Menten model. Figure 18Comparison of RMSD between wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme; where a is the RMSD curve of wild-type dlGAS in a 100 ns molecular dynamics simulation, and b is the RMSD curve of M119N-GGGGS-ERG20 fusion enzyme in a 100 ns molecular dynamics simulation. Figure 19 RMSF analysis of protein residues for wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme, where a is the RMSF value of each residue of wild-type dlGAS in a 100 ns molecular dynamics simulation, and b is the RMSF value of each residue of M119N-GGGGS-ERG20 fusion enzyme in a 100 ns molecular dynamics simulation. Figure 20 It is a fusion enzyme protein of wild-type dlGAS and M119N-GGGGS-ERG20. R g The curve shows the change over time, where 'a' represents the wild-type dlGAS during molecular dynamics simulations. R g As time progresses, b represents the M119N-GGGGS-ERG20 fusion enzyme during molecular dynamics simulations. R g Changes over time; Figure 21 SASA analysis of the protein of wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme, where a is the change of hydrophobic surface area of ​​wild-type dlGAS with molecular dynamics simulation time, and b is the change of hydrophobic surface area of ​​M119N-GGGGS-ERG20 fusion enzyme with molecular dynamics simulation time. Figure 22 The number of hydrogen bonds between the wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme protein and the substrate changes over time. Here, a represents the change in the number of hydrogen bonds between wild-type dlGAS in molecular dynamics simulations, and b represents the change in the number of hydrogen bonds between M119N-GGGGS-ERG20 fusion enzyme in molecular dynamics simulations. Figure 23 Gibbs free energy geomorphic maps (FELs) of wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzymes were analyzed. In the FELs, a is the two-dimensional FEL of wild-type dlGAS, b is the two-dimensional FEL of M119N-GGGGS-ERG20 fusion enzyme, c is the three-dimensional FEL of wild-type dlGAS, and d is the three-dimensional FEL of M119N-GGGGS-ERG20 fusion enzyme. Detailed Implementation

[0026] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0027] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0028] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0029] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be obvious to those skilled in the art. This specification and embodiments are merely exemplary.

[0030] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0031] Unless otherwise specified, the term "parts" in this invention refers to parts by weight.

[0032] dlGAS The gene originates from the whorled anthracnose fungus (D.). loculata (NCBI gene ID: JAJKLC010000010, https: / / www.ncbi.nlm.nih.gov / nuccore / 2227269602), with a full length of 785,956 bp, encoding 510 amino acid residues, and a theoretical molecular weight of 59.1 kDa, its sequence is shown in SEQ ID NO.1.

[0033] SEQ ID NO.1: MDVTTLKLTLLSKATVPRWILLIIMILFAYAVVRCVYLLYFHPASKFPPGKLAAISNFYYAYSWLRGRYPWVVEQLAKEYGDIVRIAPNEVVFSTPQAAIDIYNPAVKRHELWVKTDLMDFGTGDLG FTWEIDPAKRREVAKKVLPAFSTKAIKAKEPTVQMYIDSFVEKMKELGSAPEGVSLNDWLNWLAMDMSADLAYSRKMHHMRDAKNTDFLETLLGTNFYGTVMQVSKRFPLLSPLSFLFAPPKIIRLVP KFFKMNSQEVQRRIDNRGKTTHPDFVDYMLPTNAPAPTNKKEKVHLEQVSMQLLIAGFDPIQITFFSSLFFLLKNQDAYKTLVEEIRNTFRNYTEITPDNLASLKYLNAVIHETLRTHVTAVTGLPRK SPGAMVAGTYVPKGVVCQTSNFNICRNPRYFYEPLNFHPERWLQSDHPKYDKRFSNDDLNCFYPFSIGPRACTGREIAWSQTRLFLAKVLWTFDLEMVRGHDKTFDKDFLIHVMWTRPEVRVRFLQR.

[0034] GenBank accession number: JAJKLC010000010.1, protein ID: KAI2782286.1 (Isotrichodermin C15 hydroxylase), UniRef ID: UPI0020C34236 (https: / / www.uniprot.org / uniparc / UPI0020C34236 / entry / EMBL_CON:KAI1648040).

[0035] dlGAS The amino acid sequence of the M119N mutant is shown in SEQ ID NO.2. The difference between it and SEQ ID NO.1 is that the 119th position is changed from M to N, and the rest are the same.

[0036] SEQ ID NO.2: MDVTTLKLTLLSKATVPRWILLIIMILFAYAVVRCVYLLYFHPASKFPGPKLAAISNFYYAYSWLRGRYPWVVEQLAKEYGDIVRIAPNEVVFSTPQAAIDIYNPAVKRHELWVKTDLNDFGTGDLGFTWEIDPAKRREVAKKVLPAFSTKAIKAKEPTVQMYIDSFVEKMKELGSAPEGVSLNDWLNWLAMDMSADLAYSRKMHHMRDAKNTDFLETLLGTNFYGTVMQVSKRFPLLSPLSFLFAPPKIIRLVPKFFKMNSQEVQRRIDNRGKTTHPDFVDYMLPTNAPAPTNKKEKVHLEQVSMQLLIAGFDPIQITFFSSLFFLLKNQDAYKTLVEEIRNTFRNYTEITPDNLASLKYLNAVIHETLRTHVTAVTGLPRKSPGAMVAGTYVPKGVVCQTSNFNICRNPRYFYEPLNFHPERWLQSDHPKYDKRFSNDDLNCFYPFSIGPRACTGREIAWSQTRLFLAKVLWTFDLEMVRGHDKTFDKDFLIHVMWTRPEVRVRFLQR.

[0037] The nucleotide sequence of the gene encoding the M119N mutant is shown in SEQ ID NO.3.

[0038] SEQ ID NO.3:

[0039] Example 1 The amino acid sequence of dlGAS was obtained from the NCBI database (GenBank: KAI2782306.1) and exported in FASTA format as the input sequence for subsequent modeling. Homology modeling was performed using the Swiss-Model online server. The BLAST and HHblits algorithms were used to automatically search for structural templates in the PDB and AlphaFold databases, and the optimal template was selected based on parameters such as sequence consistency, coverage, and GMQE. In this embodiment, the A0A1Y2WX98.1 model from the AlphaFold database was ultimately selected as the template to construct the initial three-dimensional model of dlGAS. The A0A1Y2WX98.1 model showed 89.22% sequence homology with dlGAS and exhibited high conservation in the catalytic core region.

[0040] The three-dimensional structure of the dlGAS protein obtained by the homology modeling Swiss-Model method is as follows: Figure 1 As shown, Figure 1 In the diagram, the red portion represents the α-helix, the cyan represents the β-sheet, the gray represents random coils, and the green area represents the linker loop structure. It is evident that this model uses the α-helix as the primary secondary structural unit, exhibiting typical folding characteristics of terpene synthases, while also containing a small amount of β-sheets and random coil structures. The overall structure is compact, with a stable spatial conformation. The active site is located in a hydrophobic cavity formed within the protein, which facilitates substrate binding and catalytic reactions, consistent with the structural characteristics of sesquiterpene synthases.

[0041] To evaluate the rationality and reliability of the constructed dlGAS protein model, the model was quality-checked using SAVE v6.0 SERVER. The dihedral distribution of each amino acid residue in the model was analyzed using Ramachandran plots to determine the proportions of rational, permissible, and aberrant conformations.

[0042] The results of the Ramachandran plot analysis are as follows: Figure 2 As shown in the figure, the dihedral angle of each amino acid residue corresponds to a specific (x, y) coordinate value. Influenced by the steric hindrance effect of neighboring residues, these coordinate values ​​exhibit a specific distribution pattern. The regions in the figure can be divided into three categories: red / dark regions represent high-density core allowed regions, yellow / light regions represent low-density edge allowed regions, and blank regions correspond to conformationally disallowed regions. The Swiss-Model predicts that residues falling into the most favorable region, residues in the additional allowed region, and residues in the broad allowed region account for 93.7%, 6%, and 0.3% of the total protein structure, respectively. The cumulative proportion of the confidence region and the maximum allowed region both exceed 90% of the total protein residues, indicating that the prediction model has reasonable stereochemical characteristics.

[0043] Example 2 The dlGAS three-dimensional structural model constructed in Example 1 and verified by Ramachandran diagrams was used as the receptor protein. Molecular docking of dlGAS with the substrate farnesyl pyrophosphate (FPP) was performed using AM-Dock software (based on the AutoDock Vina engine). Prior to docking, the receptor protein structure was preprocessed: water molecules and irrelevant cofactors were removed, polar hydrogen atoms were added, partial charges were calculated and atom types were assigned using the Gasteiger method, and the required PDBQT file for AutoDock was generated.

[0044] The three-dimensional structure of the substrate farnesyl pyrophosphate (FPP) was downloaded from the PubChem database (CID: 445713). After conformation optimization using Chem3D software, it was converted to PDBQT format using Open Babel, and the rotatable bonds in the ligand were automatically identified to ensure conformational freedom during docking.

[0045] Molecular docking was performed centered on the spatial location of key residues in the dlGAS active pocket, with a grid box size of 25 Å × 25 Å × 25 Å covering the entire substrate binding cavity. Docking parameters were set as follows: 50 docking runs (each independent), exhaustiveness of 16, and a maximum of 20 docking conformations output per binding site. Other parameters used the default AutoDock Vina settings.

[0046] After docking is completed, the binding energy ( The lowest standard is used to select the optimal binding conformation.

[0047] Calculate using the following formula: ; in, This represents the total free energy of the enzyme-substrate complex. and These are the free energies of the isolated enzyme and the substrate, respectively. The more negative the value, the tighter the binding between the ligand and the protein, and the more stable the complex.

[0048] Simultaneously calculate and estimate the suppression constant ( K i The formulas for calculating ligand efficiency (LE) are as follows: ; Where R is the gas constant (1.987 cal·mol⁻¹). -1 ·K -1T represents absolute temperature (298.15 K, 25 °C).

[0049] ; Where, N HA The number of heavy atoms in the ligand (i.e., the total number of atoms excluding hydrogen atoms); the unit of LE is usually kcal·mol. -1 Heavy atoms -1 .

[0050] Of the 50 independent docking results, Δ G bind The top ten FPP-dlGAS molecular docking results are shown in Table 1.

[0051] Table 1. Results of FPP-dlGAS molecular docking

[0052] The results showed that the Δ of the first ten binding sites G bind The energy ranged from -7.8 kcal / mol to -5.1 kcal / mol, with binding sites 4 and 3 showing values ​​of -7.8 kcal / mol and -7.7 kcal / mol, respectively, indicating that FPP binding was most robust at these two sites. Correspondingly... K i The values ​​of 1.92 μM and 2.27 μM further illustrate that these two binding conformations have a significant advantage in binding affinity. Meanwhile, the ligand efficiencies (LE) at binding sites 4 and 3 are -0.33 and -0.32, respectively, indicating that these conformations not only have low binding energies but also that each atom contributes a high percentage to the total binding energy.

[0053] Further, PyMOL was used to perform three-dimensional visualization of the six conformations with the lowest binding energies, such as... Figure 3 As shown in the figure, the substrate is displayed as a purple-red rod-like structure. From multiple spatial perspectives, the FPP remains stably located within the core region of the dlGAS active cavity, forming extensive van der Waals forces and hydrophobic interactions with key residues within the active pocket. The highly consistent localization of the FPP in the core binding region provides structural evidence for subsequent alanine scanning and virtual saturation mutagenesis.

[0054] Example 3 Using the Calculate Mutation Energy (Binding) module of Discovery Studio 4.5 software, an alanine (Ala) scan was performed on residues within a 5 Å radius centered on the FPP ligand. The target residues were replaced one by one with alanine, and the energy difference (ΔΔ) of the complex before and after the mutation was calculated. G bind ).

[0055] ; in, It is the free energy of wild-type protein binding to its substrate. It is the free energy of the mutant protein binding to the substrate.

[0056] When ΔΔ G bind When the value is less than -0.5 kcal / mol, the mutation is considered stable, indicating that replacing this residue with alanine enhances protein-ligand binding or overall folding stability; when ΔΔ G bind When the mutation is greater than 0.5 kcal / mol, it belongs to the destabilizing type, suggesting that this residue plays a core role in maintaining the function or structural integrity of the active pocket; when ΔΔ G bind A value between -0.5 and 0.5 kcal / mol is considered neutral, indicating that the mutation has a limited impact on protein-ligand binding or structural stability.

[0057] The results of the Ala scan of the first 16 key residues and their mutation energies and energy decomposition are shown in Table 2.

[0058] Table 2 Alanine Scan Results

[0059] Data show that the mutation energies of F267, F120, and M119 are -5.39, -4.61, and -2.90 kcal / mol, respectively, which are clearly stable mutations, suggesting that these residues play a central role in substrate localization and the stability of the active pocket structure through van der Waals forces and hydrophobic interactions. Residues such as C236, M264, S268, N271, and V349 also show negative mutation energies, indicating that they provide stabilizing contributions to substrate embedding and substrate channel formation.

[0060] Binding energy decomposition analysis revealed that van der Waals forces contribute the most to substrate binding, and hydrophobic interactions are a crucial factor in forming a tight core within the active pocket. While hydrogen bonds are limited in number, they provide directionality and specificity for the substrate, ensuring the correct localization of the FPP within the active pocket. N271, as a typical "dual-function residue," provides directional recognition through hydrogen bonds and enhances binding affinity through its hydrophobic side chain, serving as a vital bridge for optimizing the active pocket structure. The hydrogen bond distance of S268 is only 2.70 Å, with an angle close to 165°, falling into the category of strong hydrogen bonds, which is beneficial for stable substrate localization.

[0061] Some residues, such as N118, I117, and F235, are neutral, with mutation energies close to zero, indicating their limited impact on protein folding and substrate binding. These residues can serve as secondary targets for auxiliary modification without disrupting the core structure. Comprehensive analysis of mutation energies, van der Waals forces, electrostatic forces, and entropy terms can clarify the contribution strength of key sites in the active pocket, providing precise targets for subsequent virtual saturation mutagenesis and semi-rational design.

[0062] The three-dimensional complex structure of dlGAS and substrate FPP and the spatial distribution characteristics of the active pocket are as follows: Figure 4 As shown, the protein is predominantly α-helical, supplemented by a small amount of β-sheets and random coils, forming a typical folding pattern for terpene synthases. Red represents α-helices, cyan represents β-sheets, gray represents random coils, and green represents linker loop structures. The substrate is located in a hydrophobic catalytic cavity inside the protein, forming stable interactions with surrounding residues through hydrophobic interactions and van der Waals forces. The yellow area represents the spatial extent of the active pocket, and the dotted structures mark the pocket boundaries, indicating that dlGAS has good substrate embedding ability and catalytic potential.

[0063] Based on the alanine scan results, the Calculate MutationEnergy (Binding) module in Discovery Studio 4.5 was used to determine the key amino acids of each docking protein-small molecule complex and perform virtual saturation mutation analysis, mutating them to amino acids other than alanine. The results of the mutants were as shown in Table 3.

[0064] Table 3 Results of Virtual Saturation Mutation Analysis

[0065] As can be seen from Table 3, the ΔΔ of most mutated residues G bind Negative values ​​indicate that these substitutions enhance the protein's affinity for its substrate, suggesting they are potentially stable mutations. M119>G ΔΔ G bind The value was -10.6 kcal / mol, accompanied by a significant decrease in van der Waals energy (-21 kcal / mol), indicating that this mutation enhances the stability of the substrate in the active pocket by optimizing the hydrophobic core or reducing steric hindrance. Similarly, mutations such as F267>R / W / I also showed lower ΔΔC / mol values. G bindThis indicates that the hydrophobic and electrostatic environment of the active pocket can be optimized through these substitutions, potentially improving catalytic efficiency. On the other hand, some mutations can be positive or close to zero, indicating unfavorable or neutral effects. For example, V349>Q and M264>K mutations have limited impact on substrate binding affinity or may reduce binding capacity. These residues are generally not preferred targets for mutation to avoid disrupting the spatial conformation or stability of the active pocket.

[0066] Taking into account both mutation sensitivity and sequence conservation analysis, the M119 and F267 sites exhibit significant plasticity and functional sensitivity. Polar or small-volume residue substitutions in M119 (such as Gly, Tyr, Asp) can optimize the active site microenvironment, reduce steric hindrance, and enhance substrate binding; while substitutions in F267 (such as Arg, Trp, Ile, Gln) may affect substrate localization and transition state stability by adjusting hydrophobic interactions with the π-system.

[0067] Example 4 Clustal X2 software was used to perform multiple sequence alignment of dlGAS with GAS homologs from different species. During the alignment process, the software automatically constructed a sequence matrix based on the similarity between target sequences and the presence of insertions / deletions (indels), and output the alignment results and conservation information.

[0068] Sequence conservation maps were generated using the WebLogo tool, quantifying and visualizing the multiple sequence alignment data. The results of multiple sequence alignment and conservation analysis of dlGAS and homologous sesquiterpene synthases are shown below. Figure 5 As shown.

[0069] Figure 5 Part a presents the multiple sequence alignment results of the dlGAS protein and its homologous sesquiterpene synthases from different species. The aligned sequences include KAI2782306.1, KAK6948206.1, KAI8965853.1, and XP_033432897.1, all from different species but sharing the same core catalytic function. The results show that amino acid position 119 is highly conserved in most homologous enzymes, indicating its crucial role in the overall protein structure and function. As an important residue near the active pocket, position 119 not only participates in substrate localization and carbocation intermediate stability but may also affect the local stability and kinetics of protein folding. Therefore, selecting this site for mutation (e.g., M119N) is naturally feasible in semi-rational design: mutations can improve substrate binding or channelization efficiency without disrupting the overall enzyme folding and catalytic function, thereby optimizing enzyme performance.

[0070] Figure 5Figure b shows a sequence conservation map generated using the WebLogo tool, quantifying and visualizing multiple sequence alignment data. Residue position 119 has a relatively high height in the WebLogo map, but it is not absolutely conserved; it belongs to the so-called "semi-conserved site." This semi-conservative property means that this site has maintained the stability of its core function during evolution, while allowing a certain range of amino acid substitutions, thus providing modification potential in enzyme engineering design. Utilizing this characteristic, favorable mutations can be introduced while maintaining the protein's native folding and active pocket conformation, thereby optimizing substrate binding and catalytic efficiency.

[0071] Example 5 This study collected 100 known GAS amino acid sequences and performed phylogenetic analysis using MEGA X software. First, all sequences were imported into MEGA in Fasta format, and multiple sequence alignment (MSA) was performed using the Clustal X2 algorithm to obtain detailed information on conserved and variable regions, providing a foundation for subsequent functional and structural analysis. Based on the MSA results, a preliminary phylogenetic tree was constructed using the neighbor-joining (NJ) method, and amino acid substitution distances were calculated using the Poisson model. To assess branch reliability, the Bootstrap resampling method was used, repeated 1000 times to generate support percentages.

[0072] The constructed GAS phylogenetic tree is as follows Figure 6 As shown in the figure, dlGAS is clustered in the evolutionary branch of sesquiterpene synthases and is closely related to gemmaene synthases, whose functions are well-defined. This provides a reference for the prediction of key residues and the analysis of catalytic functions, and also reveals the pattern of functional differentiation and conservation of enzymes in the long-term evolution process.

[0073] Example 6 Construction and validation of the PESC-LEU expression vector: according to dlGAS Design of gene sequence-specific primers: PCR primers were designed and automatically calculated using SnapGene software. T m (°C) value, GC content, and predicted dimer formation probability. The designed amplified wild-type... dlGAS The PCR primers for 12 key mutants (A115Y, M119E, M119G, M119N, M119Q, M119R, G232N, C236V, F267C, F267T, F267N, and V349H) are shown in Table 4 below. All primers were synthesized by Beijing Qingke Biotechnology Co., Ltd., and their purity met the molecular biology grade standard (≥99%).

[0074] Table 4 Designed PCR primers

[0075] PCR amplification was performed using 2× PhantaMax Master Mix. The reaction mixture was prepared on ice, and each group was thawed, mixed thoroughly, and then used. The PCR reaction mixture is shown in Table 5 below.

[0076] Table 5 PCR Cloning dlGAS reaction system

[0077] PCR program: 95℃ pre-denaturation for 3 min; 95℃ denaturation for 15 sec, 60℃ annealing for 15 sec, 72℃ extension for 50 sec / kb, 25 cycles; 72℃ final extension for 5 min. After the PCR program, the PCR products were verified by agarose gel electrophoresis and purified and recovered. The results showed that each mutant and wild-type dlGAS showed a single band, the size of which was completely consistent with the expectation, indicating that the PCR reaction had good specificity and the amplified products were intact.

[0078] The PESC-LEU plasmid was subjected to double enzyme digestion, and the double digestion system is shown in Table 6. Funiculum inactivation was achieved by incubating with Funiculum BamHI and Funiculum SmaHI at 37°C for 15 min, followed by heating at 80°C for 20 min. The vector and insert fragment were mixed at a molar ratio of 1:2, and 5× lysis and extension buffer and Exnase were added. Recombination ligation was performed by incubation at 37°C for 30 min. The recombination reaction ligation system is shown in Table 7. The ligation product was transformed into *E. coli* DH5α competent cells and plated on LB agar plates containing ampicillin (100 μg / mL). The cells were incubated at 37°C for 12–16 h. Single colonies were picked for colony PCR verification. After confirming the correct band size by agarose gel electrophoresis, positive clones were sent for sequencing verification.

[0079] Table 6. PESC-LEU plasmid double enzyme digestion system

[0080] Table 7 Recombinant Reaction Linkage System

[0081] Include dlGAS A schematic diagram of the recombinant gene expression vector PESC-LEU-dlGAS is shown below. Figure 7 As shown in the figure, the key components of the vector are illustrated, including the LEU auxotrophic selection marker and the GAL1 / GAL10 promoter-driven... dlGAS Gene expression regions, as well as origin of replication and multiple cloning sites.

[0082] The constructed recombinant plasmid vector was introduced into GQ01 competent cells using a yeast transformation kit. Frozen GQ01 yeast cells were retrieved from glycerol cryopreservation at -80°C and streaked onto YPD agar plates, then incubated upside down at 30°C for 2-4 days. Single colonies were picked and inoculated into 10 mL of YPD liquid medium, and cultured overnight with shaking at 28-30°C and 250 rpm. The following day, the overnight culture was inoculated into 50 mL of fresh YPD liquid medium, with initial OD... 600 Maintain the pH at 0.2-0.3 to ensure the cells are in the logarithmic growth phase, and culture for 2-3 hours until OD500. 600 Reach a concentration of 0.4-0.6. Take 10 mL of culture medium, centrifuge at 3000 × g for 3 min at room temperature, remove the supernatant and collect the bacterial pellet. Add 7 mL of Buffer A to the pellet, resuspend thoroughly, and centrifuge again at 3000 × g for 3 min, removing the supernatant. Add 400 μL of Buffer B to resuspend, and incubate at room temperature for 10 min. After incubation, aliquot competent cells at 100 μL / tube into sterile 1.5 mL centrifuge tubes for later use.

[0083] Gently mix 0.1-1 μg of plasmid DNA with 10 μL of denatured vector DNA, add to 100 μL of competent cells, and gently tap to mix. Add 600 μL of Buffer C and incubate at 30°C for 30 min. After incubation, add 70 μL of LDMSO, gently invert to mix, and immediately heat shock at 42°C for 15 min, then quickly transfer to an ice bath for 3-5 min to cool. After heat shock, centrifuge the transformation product at room temperature at 19942 × g for approximately 10 s, remove the supernatant, and gently resuspend the cells in 100 μL of Buffer A. Spread the resuspended cell solution evenly on SD-LEU plates and incubate upside down at 30°C for 2-3 days until single colonies appear for subsequent colony PCR verification, sequencing, and plasmid extraction.

[0084] The results of agarose gel electrophoresis of plasmid DNA extracted and purified from recombinant Saccharomyces cerevisiae are as follows: Figure 8 As shown in the figure, lane 1 is the DNA marker, used to reference the size and migration position of the DNA fragment; lanes 2-4 correspond to the M119E, A115Y, and C236V mutant recombinant plasmids, respectively. Each lane shows a clear and bright single DNA band, approximately 1127 bp in size, consistent with the theoretically designed target gene fragment length, indicating that the dlGAS mutant fragment has been successfully cloned into the PESC-LEU vector, and the recombinant plasmid remains intact and stable.

[0085] The recombinant was verified by PCR and electrophoresis, showing that the size of the target gene fragment was consistent with the expectation. Sequencing results also confirmed the sequence accuracy, proving... dlGASThe mutants were successfully integrated into yeast expression vectors. The constructed vectors were uniformly named as follows: PESC-V349H, PESC-G232N, PESC-M119Q, PESC-F267C, PESC-F267N, PESC-M119N, PESC-M119R, PESC-M119E, PESC-A115Y, PESC-C236V, PESC-M119G, and PESC-F267T.

[0086] Example 7 To evaluate the expression efficiency and GA yield of exogenous dlGAS and its mutants in Saccharomyces cerevisiae: The recombinant yeast strain constructed in Example 6 was streaked on YPD plates and cultured upside down at 30°C for 2-3 days. A blank group (GQ01 strain not transformed into the host) and a GQ01 strain with an empty plasmid were set as controls.

[0087] A single colony was picked and inoculated into 3 mL of YPD liquid medium, and cultured with shaking at 30℃ and 220 rpm for 24 h to obtain a seed culture. The OD of the seed culture was measured using a spectrophotometer before inoculation. 600 To ensure the initial inoculum concentration is controlled within an appropriate range, the growth difference between different parallel groups should be minimized. Shake-flask fermentation was performed using 250mL bottles of 30mL YPD medium, with three replicates per recombinant yeast strain. The inoculum concentration was determined according to the OD value of the seed culture. 600 Adjust the initial value to 0.1 to ensure a consistent growth initiation state. Maintain the fermentation temperature at 30℃, the shaking rate at 220 rpm, and the shaking angle at approximately 25°-30° to increase the gas-liquid contact area, improve dissolved oxygen conditions, and maintain yeast activity in the logarithmic growth phase. The entire fermentation process lasted 96 hours. At 24 hours, add 2 mL of galactose solution (0.3 g / L) to the culture medium and immediately gently shake to mix, inducing the expression of the controllable promoter and activating the exogenous promoter. dlGAS Or mutant gene expression.

[0088] During fermentation, 1 mL of sample was taken every 12 hours for OD measurement. 600 After sampling, the sample was centrifuged at 16992×g for 3 min, and the supernatant was collected for GC-MS analysis.

[0089] β-elemene is a Cope rearrangement product of gemmaene A (GA). Since GA spontaneously undergoes a Cope rearrangement reaction to convert to β-elemene in a high-temperature column of GC-MS (approximately 250°C), β-elemene is used as an equivalent standard compound for quantitative analysis of GA.

[0090] Accurately weigh β-elemene standard and dissolve it in dodecane organic solution to prepare a series of gradient concentration solutions (5 mg / mL, 2.5 mg / mL, 1.25 mg / mL, 0.625 mg / mL, 0.3125 mg / mL, 0.15625 mg / mL) to cover the expected detection range. Subsequently, each concentration standard solution was injected into a GC-MS system using a DB-5 capillary column (30 m × 0.25 mm × 0.25 μm) with electron impact (EI) ionization. The column temperature program was as follows: initial 60 °C for 3 min, increased to 90 °C at 3 °C / min, then increased to 260 °C at 20 °C / min and held for 5 min to ensure complete elution of all components.

[0091] A standard curve was plotted by measuring the peak areas of the quantitative ions of β-elemene and using their ratios as the ordinate, with the concentration of the standard solution as the abscissa. The results showed that β-elemene exhibited good linearity in the concentration range of 0.15625-5 mg / mL, with the regression equation as follows: The Pearson correlation coefficient R = 0.998.

[0092] culture medium OD 600 Values ​​characterizing cell density; yeast growth curves for the blank group, empty vector group, and experimental group (transformed with M119N strain) are compared, for example... Figure 9 As shown, a represents the OD values ​​of the experimental group, the blank group, and the empty vector plasmid group. 600 The changes showed that each group exhibited a typical S-shaped growth curve throughout the culture period, with OD... 600 After reaching a stable plateau, the growth rate tends to saturate. b represents a linear fit during the logarithmic growth phase (0-24h), indicating that the introduction of exogenous genes or the presence of vectors have little impact on the overall growth rate and lag phase, with the maximum growth density OD... max The changes in growth constant k and the variation were not significant, providing a reliable growth basis for subsequent shake-flask fermentation product accumulation experiments. Throughout the entire culture period, the OD values ​​of the experimental group and the empty plasmid group were... 600 The values ​​were consistently slightly lower than those of the control group, but the difference was small. During the logarithmic growth phase (within 24 hours), the specific growth rate μ of the experimental group and the empty plasmid group did not show a significant decrease compared to the control group; the final OD of each group during the plateau phase was... 600 The values ​​are also quite close.

[0093] The above results indicate that the introduction of exogenous plasmids has a limited impact on host cell growth and does not significantly inhibit cell proliferation.

[0094] In GC-MS determination, the results of total ion chromatography (TIC) and mass spectrometry analysis are as follows: Figure 10As shown, a) is the total ion current chromatogram of the engineered yeast GQ01-PESC-WT expression system, and b) is the mass spectrum of the corresponding product, displaying the mass spectrometric fragment characteristics of GA in GC-MS analysis. The mass spectra were obtained by electron impact ionization (EI), with the x-axis representing the mass-to-charge ratio (m / z) and the y-axis representing ion abundance. The m / z values ​​corresponding to the peaks reflect the characteristic ion fragments generated by the breakage of GA molecules under high-energy electron bombardment. The results show that the peak at 19.6 min in the GQ01-PESC-WT expression product highly matches the retention time of the standard β-elemene (GA Cope rearrangement product). The mass spectrum shows that this peak has typical β-elemene fragment ion characteristics, and the main peak corresponds to the standard spectrum. Combined with the NISTII database search results, it can be confirmed that this peak is indeed β-elemene. The TIC and mass spectra of the expression products of each mutant also showed that the β-elemene peak was significantly enhanced in most of the highly active mutants, and the GA signal was most significant in some mutants such as M119N, indicating that the mutation promoted the enzyme's catalytic efficiency.

[0095] GC-MS analysis of GA production in wild-type dlGAS and its 12 mutant strains compared to... Figure 11 As shown, a) compares the extracellular GA production of wild-type dlGAS and 12 mutant strains after 72 h of shake-flask fermentation in engineered yeast GQ01, and b) is the heatmap corresponding to a). The results show significant differences in GA production among the mutant strains. The M119N mutant had the highest yield, reaching approximately 744 mg / L, an increase of about 1.89 times compared to the wild-type (approximately 393 mg / L), demonstrating a significant enhancement of enzyme catalytic efficiency by the mutation. p <0.0001). The mutants G232N and A115Y also significantly increased GA production ( p <0.01 or p <0.0001). However, the mutant M119E significantly reduced GA production ( p The result (<0.0001) suggests that this mutation has a negative impact on substrate binding or carbocation stability in the active pocket. Other mutants had no significant effect on GA yield.

[0096] Example 8 Molecular docking of wild-type dlGAS and M119N mutants with the substrate FPP was performed using the same method as in Example 2. Schematic diagrams of molecular docking between WT and M119N and the substrate FPP, and a schematic diagram of the non-covalent interaction between WT and M119N are shown below. Figure 12 As shown, although FPP can enter the catalytic core region in the wild-type active pocket, some areas suffer from steric hindrance and unfavorable collisions. Visualized schematic diagrams of the active pocket structures of wild-type dlGAS and the M119N mutant are shown below. Figure 13As shown, after the M119N mutation, the substrate forms a tight hydrophobic contact with residues such as A115, M119N, F120, T135, S268, V349, and Y356, with an interaction distance between 3.45 Å and 3.99 Å. Figure 13 Residues such as R227, D230, S268, and Q271 form multiple hydrogen bonds with the substrate. Among them, the hydrogen bond distance of S268 is 2.70 Å and the angle is close to 165°, which belongs to the category of strong hydrogen bonds. Q271 participates in substrate localization through a short-distance hydrogen bond of 2.41 Å, while enhancing binding affinity through a hydrophobic interaction of 3.99 Å. The results indicate that the M119N mutation optimizes the hydrophobic core and hydrogen bond network, enabling FPP to form a more stable binding conformation within the active pocket.

[0097] Example 9 Directed fusion of ERG20 with a highly active M119N mutant, including: Retrieve yeast from the NCBI database ERG20 The gene, designated J05091 (https: / / www.ncbi.nlm.nih.gov / nuccore / J05091), is 2139 bp in length, encodes 372 amino acid residues, and has a predicted molecular weight of approximately 40.5 kDa. Its amino acid sequence is shown in SEQ ID N.30.

[0098] SEQ ID N.30: MASEKEIRRERFLNVFPKLVEELNASLLAYGMPKEACDWYAHSLNYNTPGGKLNRGLSVVDTYAILSNKTVEQLGQEEYEKVAILGWCIELLQAYFLVADDMMDKSITRRGQPCWYKVPEVGEIAINDAFMLEAAIYKLLKSHFRNEKYYIDITELFHEVTFQTELGQLMDLITAP EDKVDLSKFSLKKHSFIVTFKTAYYSFYLPVALAMYVAGITDEKDLKQARDVLIPLGEYFQIQDDYLDCFGTPEQIGKIGTDIQDNKCSWVINKALELASAEQRKTLDENYGKKDSVAEAKCKKIFNDLKIEQLYHEYEESIAKDLKAKISQVDESRGFKADVLTAFLNKVYKRSK.

[0099] yeast ERG20The gene is identified as P08524 (ERG20_YEAST) in the Swiss-UniProt database and its corresponding structure prediction number in the AlphaFold database is AF-P08524-F1.

[0100] The M119N-GGGGS-ERG20 fusion gene was constructed using overlap extension PCR. Fusion primers were designed to introduce the coding sequence of the GGGGS linker peptide (SEQ ID N.31, GGTGGTGGTGGTTCT) into the 3′ end of the M119N gene fragment, and then ligated to the 5′ end of the ERG20 gene fragment.

[0101] In the first round of PCR, the M119N gene fragment (with the 3′ end of the linker peptide coding sequence) and the ERG20 gene fragment (with the 5′ end of the linker peptide coding sequence) were amplified separately. In the second round of PCR, the two fragments were mixed and used as a template, and the outer primers were added for fusion PCR amplification to obtain the full-length fusion gene M119N-GGGGS-ERG20.

[0102] Using the same method, fusion genes ERG20-GGGGS-M119N, M119N-GSG-ERG20, and ERG20-GSG-M119N were constructed to compare the effects of different linker peptides and fusion orientations on catalytic efficiency.

[0103] The recombinant expression vector and recombinant Saccharomyces cerevisiae engineered strain were constructed using the same methods as in Examples 7 and 8, and PCR and sequencing verifications confirmed that they were correct.

[0104] The GA yield of four fusion enzyme strains (ERG20-GSG-M119N, ERG20-GGGGS-M119N, M119N-GSG-ERG20, M119N-GGGGS-ERG20) and a control strain expressing M119N alone was determined using the method described in Example 8 to evaluate the effect of different fusion methods on GA yield. Results are as follows: Figure 14 As shown, a represents the GA yield of M119N and four fusion enzyme strains: ERG20-GGGGS-dlGAS, ERG20-GSG-dlGAS, M119N-GGGGS-ERG20, and M119N-GSG-ERG20; b is the corresponding heatmap.

[0105] The results showed that the M119N-GGGGS-ERG20 fusion enzyme, which was fused in the forward direction and used the GGGGS Linker, had the highest yield, approximately 953.4 mg / L, significantly higher than other fusion enzymes and the control group (**). p<0.01 (one-way ANOVA) indicates that the fusion direction and linker type have a key impact on substrate channelization efficiency and catalytic performance.

[0106] The heatmap further visualized the GA yield distribution among different fusion enzyme strains. The colors in the heatmap, from darkest to lightest, reflected the GA yield, with lighter colors indicating higher yields. The results showed that M119N-GGGGS-ERG20 (corresponding to the lightest colored area) had significantly higher yields than other fusion methods and the non-fusion enzyme, validating the results. Figure 14 The trend of quantitative determination of α-α was observed. Meanwhile, the forward GGGGS fusion enzyme showed a lighter overall color, indicating that substrate channelization and enzyme activity were superior to short-chain GSG or reverse fusion strategies. This visually demonstrates the significant impact of linker type and fusion direction on GA synthesis efficiency, providing a visual basis for subsequent enzyme engineering optimization.

[0107] Example 10 Optimization of GA synthesis conditions for the M119N-GGGGS-ERG20 fusion enzyme strain.

[0108] ① Optimal induction time of enzyme reaction: To evaluate the time dynamics of galactose induction on GA production, galactose at a final concentration of 10 mM was added to wild-type strain GQ01 (WT) and strain M119N-GGGGS-ERG20 for 24 h of shake-flask fermentation to induce expression. Simultaneously, 10% (v / v) dodecane was added as an organic capping layer for in-situ extraction and enrichment of GA produced during fermentation. The upper dodecane phase was collected at 24 h, 48 h, 72 h, 96 h, 120 h, 144 h, 168 h, and 192 h after fermentation, and the GA content was quantitatively analyzed using GC-MS.

[0109] The effect of fermentation time on GA synthesis by the M119N-GGGGS-ERG20 fusion enzyme strain is as follows: Figure 15 As shown, a) compares the GA yield of the M119N-GGGGS-ERG20 fusion enzyme strain at different fermentation times; b) shows the time-GA yield curve. The results indicate that GA yield increases with increasing induction time, reaching a peak of approximately 952 mg / L at 72 h, followed by a slight decrease. p <0.001; **** p <0.0001.

[0110] Figure 15 The linear fitting results for part b show that the final plateau value is 922.7 mg / L, and the reaction rate constant is 0.04044 h⁻¹. -1 The half-life was 17.14 h, indicating that the early GA synthesis rate was high and the product was generated rapidly.

[0111] ② Optimal galactose induction concentration for enzyme reaction: To investigate the effect of different galactose induction concentrations on GA yield, galactose was added to strains GQ01-WT and M119N-GGGGS-ERG20 in shake flasks at 24 h of fermentation to induce expression. Final concentrations of 0.01 mM, 0.25 mM, 1 mM, 2.5 mM, 5 mM, 10 mM, 25 mM, 50 mM, and 100 mM were added. Simultaneously, 10% (v / v) dodecane organic phase was added as an in-situ extractant to capture and enrich GA produced during fermentation. After 72 h of fermentation, the upper dodecane phase was collected, and the GA content was quantitatively analyzed using GC-MS.

[0112] The effects of different D-galactose concentrations on the GA production of the M119N-GGGGS-ERG20 fusion enzyme strain on, for example: Figure 16 As shown in the figure, a represents the effect of D-galactose concentration on GA production of the M119N-GGGGS-ERG20 fusion enzyme strain, and b is the corresponding heatmap. The results showed that low-concentration D-galactose induction effectively activated fusion enzyme expression and increased yield, reaching a peak at 1 mM. High-concentration D-galactose induction led to a gradual decrease in GA production due to increased metabolic load and carbon source inhibition. Statistical analysis showed significant differences among different D-galactose concentration groups (***). p <0.001, **** p <0.0001).

[0113] ③ Optimal glycerol concentration for enzyme reaction: To investigate the effect of different glycerol concentrations as auxiliary carbon sources on GA yield, galactose at a final concentration of 10 mM was added to shake-flask fermentation of strains GQ01-WT and M119N-GGGGS-ERG20 for 24 h to induce fermentation. Glycerol at volume fractions of 0%, 0.5%, 1%, 1.5%, 2%, and 3% was then added as an auxiliary carbon source. Simultaneously, 10% dodecane was added as an organic capping layer for in-situ extraction and enrichment of GA produced during fermentation. After 72 h of fermentation, the upper dodecane phase was collected, and the GA content was quantitatively analyzed using GC-MS.

[0114] The effects of glycerol concentration on the yield and reaction kinetics of the M119N-GGGGS-ERG20 fusion enzyme GA are as follows: Figure 17 As shown, a represents the effect of different concentrations of glycerol (0%, 0.5%, 1%, 1.5%, 2%, 2.5%) on extracellular GA production during the 72-hour period of fusion enzyme expression induced by 1 mM galactose at 24 h of fermentation. With increasing glycerol concentration, GA production showed an upward trend at low concentrations, reaching a peak of 1019 mg / L at 1.5% glycerol, slightly higher than the group without glycerol, but the difference was not statistically significant.p >0.05). When the glycerol concentration was further increased to 2-2.5%, the GA yield decreased slightly or remained basically stable, suggesting that glycerol has no significant promoting effect on GA synthesis.

[0115] b shows the kinetic parameters under different glycerol conditions analyzed using the Michaelis-Menten model. The fitting results indicate the maximum reaction rate. V max The concentration was 14.02 mg / L / h, indicating that the fusion enzyme had a high product production capacity under conditions of sufficient substrate supply; the Michaelis constant was also high. K m The yield was 0.008%, reflecting the high affinity of the fusion enzyme for glycerol or its metabolic substrates, indicating that efficient catalysis can still be achieved at low substrate concentrations. Analysis of the curves showed that the yield plateaued with increasing glycerol concentration, without significant inhibition or excess effects. This indicates that glycerol has no significant limiting effect on the kinetics of GA synthesis within the 0-2.5% concentration range. The fusion enzyme system can catalyze efficiently even at low substrate concentrations. V max and K m The ratio (catalytic efficiency) was superior, indicating that the combination of the M119N mutation and the GGGGS flexible linker optimized the spatial layout of the active pocket and improved substrate channelization efficiency. Furthermore, the fitting data were stable with small residuals, demonstrating good experimental reproducibility and indicating that the kinetic model is suitable for describing the reaction characteristics of this enzyme system.

[0116] Example 11 To elucidate the structural basis for the improved catalytic efficiency of fusion enzymes, molecular dynamics (MD) simulations were performed on wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzymes. Using the GROMACS 2025.3 software package, the system was placed in a periodic boundary cubic box, and solvation was performed using the AMBER14SB force field and the TIP3P water model, with a formal simulation lasting 100 ns.

[0117] Comparison of RMSD between wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme as follows: Figure 18 As shown, a) is the RMSD curve of wild-type dlGAS in a 100 ns molecular dynamics simulation, and b) is the RMSD curve of the M119N-GGGGS-ERG20 fusion enzyme in a 100 ns molecular dynamics simulation. The curve colors represent protein (black), substrate (red), and protein-substrate complex (green), respectively.

[0118] The results showed that WT dlGAS experienced significant fluctuations in the early stages of the simulation (0-40 ns), possibly reflecting frequent conformational transitions between different protein molecule configurations or large structural swings in its flexible regions. Around 40 ns, the RMSD of WT increased rapidly, indicating significant local structural adjustments in the system, such as local protein unfolding or slight ligand shifts from the binding site. In contrast, the M119N-GGGGS-ERG20 fusion enzyme showed a gradual increase in RMSD from 0-25 ns, with a maximum fluctuation of 0.2-0.4 nm. This fluctuation mainly reflected the transition of the protein from its initial crystal conformation to the equilibrium conformation under simulated conditions. As the simulation time increased to 25 ns, the fusion enzyme system reached equilibrium, with the RMSD stabilizing at approximately 0.38 nm, indicating that the overall conformation had entered a stable range.

[0119] Comprehensive analysis showed that the conformational fluctuation of the M119N-GGGGS-ERG20 fusion enzyme was lower than that of the wild type in the later stage, indicating higher overall stability. Combined with the results of in vitro catalytic experiments, the catalytic efficiency of the M119N-GGGGS-ERG20 fusion enzyme was significantly better than that of the wild type.

[0120] RMSF analysis results of protein residues of wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme are as follows: Figure 19 As shown in the figure, a represents the RMSF values ​​of each residue in the wild-type dlGAS during a 100 ns molecular dynamics simulation, and b represents the RMSF values ​​of each residue in the M119N-GGGGS-ERG20 fusion enzyme during a 100 ns molecular dynamics simulation. The results show that the M119N-GGGGS-ERG20 fusion enzyme exhibits relatively frequent fluctuations in the 0-80 site region. These fluctuating regions typically correspond to the enzyme's loop structure and unstructured regions, and may be closely related to substrate binding, catalytic reactions, and molecular function. Compared to the wild-type enzyme, this mutant shows slightly lower residue fluctuations near the key active site, indicating a more stable local active pocket structure, which is beneficial for precise substrate localization and improved catalytic efficiency. Simultaneously, the increased fluctuations in some regions may reflect the increased flexibility introduced by the mutation, allowing the protein to fine-tune around the active site, optimizing substrate binding and intermediate stability, thereby improving overall catalytic performance.

[0121] Wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme protein R g Curves of change over time as shown Figure 20 As shown, where a represents the wild-type dlGAS in molecular dynamics simulations. R g As time progresses, b represents the M119N-GGGGS-ERG20 fusion enzyme during molecular dynamics simulations.R g Changes over time.

[0122] The results showed that wild-type dlGAS exhibited [hyperactivity] in the early stages of simulation (0-10 ns). R g A slow rise may reflect unfolding or conformational adjustments in localized regions of the protein; subsequently, between 10-70 ns, R g A slight decrease indicates that the protein is gradually transitioning to a stable folded state. A slight fluctuation around 70 ns suggests transient unfolding or flexible conformational changes in certain regions of the protein, but overall... R g It remained at approximately 2.0 nm, indicating that its folded structure remained relatively stable. In contrast, the M119N-GGGGS-ERG20 fusion enzyme showed stability in the initial simulated phase (0-20 ns). R g The slow decrease reflects the gradual transition of the protein from its initial conformation to its folded state; it tends to stabilize after 20 ns. R g The wavelength remained between 2.15 and 2.25 nm, indicating that the overall structure of the fusion enzyme was relatively stable but slightly looser than that of the wild type.

[0123] Overall analysis shows that, R g The numerical changes reveal the folding and flexibility properties of proteins during dynamic simulations: lower R g Values ​​correspond to proteins that are more compact and have higher structural rigidity, while higher values... R g The value may indicate that the fusion enzyme has a slightly open spatial arrangement, which is beneficial for substrate entry into the active pocket and improves catalytic efficiency. Furthermore, initially... R g The decline and late-stage stabilization indicate that the mutation and fusion design enables the enzyme to maintain fold integrity in dynamic equilibrium while providing better substrate accessibility.

[0124] SASA analysis of the protein of wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme is as follows: Figure 21 As shown, a represents the change in hydrophobic surface area of ​​wild-type dlGAS over time during molecular dynamics simulations, and b represents the change in hydrophobic surface area of ​​the M119N-GGGGS-ERG20 fusion enzyme over time during molecular dynamics simulations. The results indicate that the hydrophobic surface area of ​​the fusion enzyme is relatively stable, and the hydrophobic core of the active pocket is well maintained, providing structural support for substrate localization and catalysis.

[0125] The number of hydrogen bonds between the wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme protein and substrate changes over time as follows: Figure 22 As shown, a represents the change in the number of hydrogen bonds between wild-type dlGAS and the target protein in molecular dynamics simulations, while b represents the change in the number of hydrogen bonds between the M119N-GGGGS-ERG20 fusion enzyme and the target protein in molecular dynamics simulations. The number of hydrogen bonds between the wild-type small molecule and the target protein ranges from 0 to 3, while the number of hydrogen bonds between the M119N-GGGGS-ERG20 fusion enzyme ranges from 0 to 4, which is more than that of the wild-type. Although there are some fluctuations in the number of hydrogen bonds in the system, they remain within a stable range overall, and there are no instances of long-term disappearance, indicating that the ligand has good binding stability in the binding pocket.

[0126] Analysis of the free energy geomorphic map (FEL) of wild-type dlGAS and M119N-GGGGS-ERG20 fusion enzyme Gibbs free energy is as follows: Figure 23 As shown, panel a represents the two-dimensional free energy (FEL) of wild-type dlGAS, panel b represents the two-dimensional FEL of the M119N-GGGGS-ERG20 fusion enzyme, panel c represents the three-dimensional FEL of wild-type dlGAS, and panel d represents the three-dimensional FEL of the M119N-GGGGS-ERG20 fusion enzyme. The results show that the system exhibits a distinct low-energy accumulation region, with the blue region representing the most stable conformational state, indicating that the complex tends to stabilize during the simulation and forms the dominant conformation. Panel b shows a more dispersed energy distribution than panel a, suggesting that the system has a certain conformational change capability. Panels c and d are the corresponding three-dimensional free energy landscapes, expressed in terms of RMSD and 2π / 4, ... R g The coordinates are for the reaction. The lowest energy point forms a clear energy valley, indicating that the protein-ligand complex can reach a thermodynamically stable state during the simulation. These results demonstrate that the system possesses good structural stability and conformational rationality.

[0127] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. An ERG20-dlGAS fusion enzyme, characterized in that: The fusion enzyme is a recombinant fusion protein formed by fusing Saccharomyces cerevisiae farnesyl pyrophosphate synthase ERG20 with the M119N mutant of Charcoalella vulgaris gemmaeene A synthase via a flexible linker peptide; the M119N mutant is formed by mutating methionine at position 119 to asparagine based on the amino acid sequence of wild-type dlGAS. The amino acid sequence of the wild-type dlGAS is shown in SEQ ID NO.1, the amino acid sequence of the M119N mutant is shown in SEQ ID NO.2, and the amino acid sequence of ERG20 is shown in SEQ ID NO.

30.

2. The ERG20-dlGAS fusion enzyme according to claim 1, characterized in that: The flexible linker peptide is GGGGS.

3. The ERG20-dlGAS fusion enzyme according to claim 1, characterized in that: The fusion enzyme is M119N-GGGGS-ERG20, with the M119N mutant located upstream and ERG20 located downstream. The M119N mutant and ERG20 are linked by the GGGGS linker peptide.

4. A nucleic acid molecule, characterized in that: The nucleic acid molecule encodes the ERG20-dlGAS fusion enzyme according to any one of claims 1-3.

5. A recombinant expression vector, characterized in that: The recombinant expression vector comprises the nucleic acid molecule of claim 4.

6. The recombinant expression vector according to claim 5, characterized in that: The recombinant expression vector is PESC-LEU.

7. A recombinant host cell, characterized in that: The recombinant host cell comprises the recombinant expression vector as described in claim 5 or 6.

8. The recombinant host cell according to claim 7, characterized in that: The recombinant host cell was Saccharomyces cerevisiae GQ01.

9. A method for constructing the ERG20-dlGAS fusion enzyme as described in any one of claims 1-3, characterized in that, Includes the following steps: S1. Obtain the gene fragment encoding ERG20 and the gene fragment encoding the dlGAS M119N mutant, respectively; S2. By overlapping extension PCR, the ERG20 gene fragment and the M119N gene fragment were fused using the coding sequence of a flexible linker peptide to obtain the fusion gene. S3. The fusion gene is cloned into the expression vector to obtain the recombinant expression vector; S4. Transform the recombinant expression vector into host cells and express the fusion enzyme.

10. The use of the ERG20-dlGAS fusion enzyme as described in any one of claims 1-3 in the preparation of gemmaene A.

Images