A method for rational modification of spermidine synthase (SpeE) in bacillus subtilis

Abstract

The present disclosure provides a rational modification method of spermidine synthase (SpeE) in Bacillus subtilis, which relates to the technical fields of enzyme engineering and computational structural biology. The modification method comprises the following steps: constructing a high-confidence three-dimensional model of Bacillus subtilis SpeE, determining the key action scope by double-substrate (S-adenosyl methionine and putrescine) docking, performing virtual saturation mutation on the residues in the region, screening the optimal mutant by combining function prediction, stability analysis, molecular dynamics simulation and binding free energy calculation, and explaining the mutation mechanism from the atomic level. The SpeE mutant (such as G85S, G85A) obtained by the modification method has significantly improved catalytic efficiency and substrate binding affinity, and when applied to a whole-cell catalytic system, it can greatly improve the yield of spermidine, and is suitable for the industrialized green biosynthesis of spermidine.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of enzyme engineering and computational structural biology, and in particular to a rational method for modifying spermidine synthase derived from Bacillus subtilis. Background Technology

[0002] Spermine, a natural polyamine, possesses various biological activities such as anti-aging, neuroprotection, and immune enhancement, and has broad application prospects in the fields of medicine, functional foods, and agriculture. Currently, the production methods of spermine suffer from problems such as complex chemical synthesis steps, low efficiency of natural extraction, and environmental pollution; therefore, microbial biosynthesis has become the preferred approach.

[0003] In the spermidine biosynthesis pathway, spermidine synthase (SpeE) is the key rate-limiting enzyme catalyzing the reaction of putrescine and S-adenosylmethionine to produce spermidine, and its catalytic efficiency directly determines the spermidine yield. Bacillus subtilis, as a commonly used safe strain in industrial fermentation, possesses advantages such as strong protein secretion capacity and good stress resistance. However, a complete high-resolution crystal structure of its derived SpeE enzyme has not yet been reported. Existing technologies mostly focus on modifying SpeE from other species such as Corynebacterium glutamicum and Bacillus amyloliquefaciens. Directly applying the mutation sites of these species may lead to the loss of SpeE enzyme activity in Bacillus subtilis.

[0004] Meanwhile, existing technologies for SpeE modification generally employ single-substrate docking analysis or random mutation strategies, lacking in-depth computational simulation and atomic-level theoretical guidance. The selection of mutation sites is often arbitrary, resulting in long development cycles, large workloads for blind testing, and an inability to accurately elucidate the structural biological mechanisms of mutation-enhanced effects. This limits the application of Bacillus subtilis in the industrial production of spermidine. Summary of the Invention

[0005] This invention provides a rational modification method for spermidine synthase (SpeE) in Bacillus subtilis. It can make up for the lack of structural information through homology modeling, accurately locate the key action domain through dual substrate docking, and obtain efficient mutants through multi-dimensional calculation, screening and verification, thereby overcoming the limitations of the prior art and achieving a significant increase in spermidine production.

[0006] On the one hand, the present invention provides a method for the rational modification of Bacillus subtilis SpeE, the method specifically including the following steps:

[0007] (1) Obtain the gene sequence and protein sequence of Bacillus subtilis spermine synthase (SpeE), and construct a high-confidence three-dimensional model of spermine synthase (SpeE) using homology modeling technology; (2) The two substrates synthesized from spermidine are simultaneously docked with the three-dimensional model described in step (1) to analyze the binding mode, interaction type and strength, small molecule torsion state and binding free energy of the substrate and enzyme, and to jointly determine the key action domain of the enzyme active site. The two substrates are S-adenosylmethionine and putrescine; (3) A virtual saturation mutation is performed on at least one residue in the key domain and the cooperating site at the edge of the active pocket, and the residues are replaced with the other 19 natural amino acids in sequence to generate a mutant library; the residue in the key domain includes glycine at position 85 and the cooperating site includes asparagine at position 120. (4) The mutant library is comprehensively screened in multiple dimensions through functional impact prediction, protein stability calculation, mutant three-dimensional structure modeling and residue interaction network analysis to obtain candidate mutants; (5) Perform molecular dynamics simulations and binding free energy calculations on the selected candidate mutants to verify the conformational stability of the mutants and their binding affinity with the two substrates, and determine the optimal mutant; (6) The catalytic performance and spermine production ability of the optimal mutant described in step (5) were verified by in vitro enzymatic property determination and whole-cell catalysis experiment.

[0008] Furthermore, the construction of the high-confidence three-dimensional model in step (1) requires the use of structural quality assessment tools to control the quality of the model. The quality control indicators include the global quality score of the model, the rationality of the amino acid residue conformation and the steric hindrance conflict, to ensure that the model meets the accuracy requirements of subsequent molecular docking.

[0009] Furthermore, the structural quality assessment tool includes at least one of QMEANDisCo, Ramachandran graph analysis tool, and MolProbity tool; the validation criterion for the global quality score is a GMQE value of the model ≥ 0.6.

[0010] Furthermore, the key active site of the enzyme active site described in step (2) is composed of amino acid residues at positions 83-90.

[0011] Further, the molecular docking operation in step (2) includes: performing docking adaptation pretreatment on the three-dimensional model and the two substrates, removing water molecules from the model and adding polar hydrogen atoms, setting a docking pocket to cover the region where amino acid residues at positions 83-90 are located, performing cluster analysis according to RMSD≤2 Å, and selecting the central conformation of each cluster as the optimal conformation.

[0012] Furthermore, the molecular docking software is any one of Autodock, Autodock Vina, Discovery Studio, or Glide.

[0013] Furthermore, the residues within the key domain described in step (3) include glycine at position 85.

[0014] Furthermore, the tools used for predicting the functional impact in step (4) include at least one of PolyPhen-2 and Provean; the tools used for calculating protein stability include at least one of PoPMuSiC, DUET and SAAFEC.

[0015] Furthermore, the duration of the molecular dynamics simulation in step (5) is not less than 100 ns, and the force field used in the simulation process includes the AMBER ff14SB force field, and the solvent model adopts the TIP3P water model.

[0016] Furthermore, the in vitro enzymatic properties measured in step (6) include specific enzyme activity, catalytic efficiency, and substrate affinity.

[0017] On the other hand, the present invention provides a Bacillus subtilis SpeE mutant, which is obtained by any of the above-described rational modification methods, and its mutation sites include a glycine mutation at position 85 of Bacillus subtilis SpeE, and may also include a mutation at position 120 of asparagine. Furthermore, the 85th glycine mutation includes a mutation of the 85th glycine to serine or alanine; Furthermore, the mutation at position 120 of the asparagine is a site-directed mutation at position 120 of the asparagine to serine.

[0018] In another aspect, the present invention also provides a method for the biosynthesis of spermidine, specifically: heterologously expressing the above-mentioned Bacillus subtilis SpeE mutant in a host strain, using S-adenosylmethionine and putrescine as substrates, and achieving the synthesis of spermidine through whole-cell catalysis.

[0019] Compared with the prior art, the present invention has the following beneficial effects: (1) Addressing the limitations of existing technologies that focus on enzyme modification of species such as Corynebacterium glutamicum and Blastochloris viridis, this invention focuses on SpeE from Bacillus subtilis, a commonly used industrial strain. By using homology modeling to compensate for the lack of high-resolution crystal structure, a dedicated and rational modification system is established, avoiding the problem of enzyme activity loss caused by applying cross-species mutation sites. The SpeE mutants obtained through this method show significantly improved performance. The specific enzyme activity of the G85S single-point mutant reaches 48.7 U / mg, which is 2.2 times that of the wild type (22.1 U / mg); the specific enzyme activity of the G85A mutant reaches 35.6 U / mg, which is 1.6 times that of the wild type; and the specific enzyme activity of the G85S / N120S combined mutant is further increased to 56.3 U / mg, providing a core tool enzyme for the application of Bacillus subtilis in the industrial production of spermidine.

[0020] (2) Existing technologies generally hold the inherent belief that "single substrate docking can meet the purpose of enzyme modification design," ignoring the natural characteristic of "synergistic binding and interaction of two substrates" in dual-substrate enzyme catalysis, resulting in high blindness in mutation site screening. This invention is based on the specificity of Bacillus subtilis SpeE—its active site conformation and substrate binding microenvironment are significantly different from those of other species of SpeE. Single substrate docking cannot restore the true binding mode of this enzyme with two substrates. Only through dual-substrate synergistic docking can the key action domain of its 83rd-90th amino acid residues be accurately located. Single substrate docking can identify up to 10 potential binding regions. After single-point mutation of these regions, the enzyme activity changes little or even reverses. At the same time, the synergistic site at the edge of the active pocket, asparagine, is screened through residue interaction network analysis. Combined with multi-dimensional calculation methods such as functional prediction, stability calculation, and molecular dynamics simulation, the optimal mutation site is screened at the atomic level, which greatly reduces the amount of invalid experimental verification, shortens the research and development cycle, and reduces manpower and material costs. When the G85S mutant was applied to a whole-cell catalytic system, spermidine production reached 15.8 g / L after 36 hours of fed-batch catalysis in a 5L fermenter. The production of the combined mutant G85S / N120S was further increased to 18.2 g / L, which was significantly higher than that of the wild-type enzyme (6.9 g / L).

[0021] (3) Compared to the limitations of existing technologies that only report yield or activity data, this invention elucidates the mechanism by which mutations affect enzyme-substrate binding modes and the conformational stability of the active pocket through residue interaction network analysis, conformational flexibility detection, and binding free energy decomposition. For example, the G85S mutation introduces a hydroxyl group to form a new hydrogen bond with the D83 residue, reducing the conformational flexibility of the key domain by 15% and lowering the binding free energy to -85.44 kJ / mol, which is significantly lower than the wild type (-60.30 kJ / mol). The N120S mutation introduces a hydroxyl group to fine-tune the electrostatic environment of the active pocket, forming a synergistic effect with G85S and further enhancing substrate binding stability. The clear causal logic enables the technical solution to form a transferable enzyme rational design principle, providing a methodological reference for the modification of similar dual-substrate enzymes.

[0022] (4) The SpeE mutant obtained by this invention can be heterologously expressed in host strains such as Escherichia coli, and is adapted to the whole-cell catalytic system of industrial fermentation. It is simple to operate and highly reproducible. At the same time, the core framework of the modification method can be extended to the modification of other enzymes of Bacillus subtilis or spermine synthases of different species by replacing parameters such as target enzyme structure and substrate molecules. It has a wide range of application scenarios and provides technical support for the systematic optimization of microbial cell factories. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0024] Figure 1 is a three-dimensional structural model of Bacillus subtilis SpeE in an embodiment of the present invention; Figure 2 This is a quality control result diagram of the three-dimensional model in the embodiment of the present invention (including QMEANDisCo score, Ramachandran plot and MolProbity analysis); Figure 3 This is a schematic diagram of the three-dimensional structure of S-adenosylmethionine in an embodiment of the present invention; Figure 4 This is a schematic diagram of the three-dimensional structure of putrescine in an embodiment of the present invention; Figure 5 The following is a diagram showing the results of the protein binding pocket scoring in an embodiment of the present invention, wherein Figure (A) shows the protein binding pocket score and Figure (B) shows the pocket structure parameters; Figure 6This is a diagram showing the docking results of SpeE-G85A with S-adenosylmethionine in an embodiment of the present invention. Figure 7 This is a diagram showing the docking results of SpeE-G85A with putrescine in an embodiment of the present invention; Figure 8 Figure (A) is a schematic diagram of the surface structure of Bacillus subtilis SpeE enzyme and the overall distribution of the active center region in the active center of Bacillus subtilis SpeE protein, and Figure (B) is a local magnified view of the region of residues 83-90 in the active center of Bacillus subtilis SpeE enzyme and a schematic diagram of the interaction of key residues. Figure 9 This is the final scoring diagram of the dual-substrate docking in an embodiment of the present invention; Figure 10 The graph shows the predicted stability changes after replacing the G85 position with 19 natural amino acids. Figure 11 This is a diagram showing the docking results of SpeE-G85S and S-adenosylmethionine in an embodiment of the present invention. Figure 12 This is a diagram showing the docking results of SpeE-G85S and putrescine in an embodiment of the present invention; Figure 13 This is a diagram showing the docking results of T158-S-adenosylmethionine in the comparative example of this invention; Figure 14 This is a diagram showing the docking results of Y62-corrosive amine in the comparative example of this invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1 1. 3D Model Construction and Quality Control: The gene sequence (as shown in SEQ ID NO:1) and protein sequence (as shown in SEQ ID NO:2) of Bacillus subtilis SpeE were obtained from the NCBI database; the SWISS-MODEL server was logged in, the SpeE protein sequence was input, and homology modeling was performed. After modeling, the closest protein template was selected as 1iy9.1.A to obtain the 3D structure data file of spermidine synthase (as shown in SEQ ID NO:2). Figure 1As shown in the figure, the global quality of the three-dimensional structural model of spermidine synthase was evaluated using the QMEANDisCo tool, and the GMQE value (0.69) and QMEANDisCo Global value (0.92) were recorded. The percentage of dominant regions (93.66%) and outliers (0.37%) were analyzed using the Ramachandran plot tool. The MolProbity Score (1.96) and Clash Score (1.78) were calculated using the MolProbity tool. All indicators met the requirements for subsequent calculation and analysis. The evaluation results are as follows. Figure 2 As shown.

[0027] 2. Substrate structure analysis and homologous protein function prediction: The three-dimensional structure of S-adenosylmethionine (CAS: 22365-13-5) was obtained from the PubChem database (e.g., Figure 3 The three-dimensional structures of putrescine (CAS: 110-60-1) and putrescine (e.g.) Figure 4 The functional groups (S-adenosylmethionine contains ribose and methionine groups, and putrescine is a diamine small molecule) and potential binding sites (polar groups such as amino and hydroxyl groups, which tend to form hydrogen bonds or electrostatic interactions with the polar residues of the enzyme) were analyzed. Using homologous sequence alignment tools, Bacillus subtilis SpeE was sequence aligned with spermidine synthase (template 1iy9.1.A) whose crystal structure has been resolved. It was found that the region of residues 83-90 is a functionally conserved region. This region directly participates in the binding of two substrates in the template enzyme (verified by the structure of the template enzyme-substrate complex) and contains multiple polar residues (such as D83). It has a high degree of matching with the polar functional groups of the two substrates, and this region is preliminarily identified as a potential active site.

[0028] 3. Screening core components combined with pocket: Full-protein pre-doping: The two substrates were pre-doped with the SpeE 3D model without any restrictions on the full protein (without setting the pocket range). After running the docking program, the top 5 binding pockets with the best scores were collected. Combined with pattern analysis: See the results of the pocket score analysis. Figure 5 Interaction analysis was performed on the high-scoring conformations (binding free energy ≤ -5.0 kcal / mol) of each pocket. It was found that the pocket containing residues 83-90 had the best binding mode with both substrates—forming ≥ 3 hydrogen bonds and having a binding free energy ≤ -6.0 kcal / mol. Furthermore, the functional groups of the substrates and the residues in this region showed the strongest polarity / charge complementarity. Other pockets either had high binding free energies or formed only 1-2 weak hydrogen bonds, lacking a stable binding microenvironment. Therefore, other regions were excluded, and the region containing residues 83-90 was identified as the core binding pocket.

[0029] 4. Dual-substrate molecular docking: The docking was performed using Autodock Vina 1.2.2 software. The docking pocket parameters were set as follows: the region of residues 83-90 was used as the center (coordinates were obtained by measuring with PyMOL 2.5 molecular visualization software (specific coordinates: x=65, y=86, z=71)), and the pocket size was set to x=20 Å, y=20 Å, z=20 Å to ensure complete coverage of the key action domain and the surrounding flexible region.

[0030] The docking programs for the two substrates and enzyme models were run separately, with 100 output conformations and a search intensity of 32. The top 100 optimal conformations for each substrate were collected.

[0031] 5. Analysis of docking results: Preliminary screening and cluster analysis of docking results: 100 conformations of each substrate were clustered according to "RMSD≤2 Å", and the central conformation of each cluster was selected for subsequent analysis to ensure the diversity of the binding modes were examined.

[0032] Quantitative and qualitative analysis of hydrogen bond interactions: Hydrogen bonds between the ligand and amino acid residues in the enzyme's active pocket were analyzed using molecular visualization software. The specific criteria were: the distance between the donor and acceptor atoms ≤ 3.5 Å, and the donor-hydrogen-acceptor angle ≥ 120°. The total number of hydrogen bonds, bond lengths, bond angles, and interacting residues were recorded to assess the rationality of the geometric configuration. Taking the docking of the G85A mutant with S-adenosylmethionine as an example (as shown in Figure 6), in this conformation, G85A forms two stable hydrogen bonds with the substrate (bond lengths of 2.9 Å and 3.1 Å, respectively), with bond angles within the reasonable range of 120° ± 10°, verifying the effectiveness of the hydrogen bond interaction between the mutant and the substrate.

[0033] Active center subregion segmentation and ligand localization analysis: The active pocket is divided into "S-adenosylmethionine binding groove" and "putrescine binding channel". The overall orientation and centroid position of the ligand molecule are analyzed, and the van der Waals contact area between the ligand and key residues in each subregion is calculated to determine whether the ligand is located at the expected catalytic reaction site. Figure 7 As shown, in the docking conformation of the G85A mutant with putrescine, the putrescine molecule is located exactly at the center of the 'putrescine binding channel', with a van der Waals contact area of ​​85 Ų with the residues in the channel, which meets the spatial positioning requirements of the catalytic reaction.

[0034] Small molecule conformation torsion state assessment: Extract the rotatable dihedral angle values ​​of the ligands in the docking conformation and compare them with the dihedral angles of the low-energy conformation in solution to assess whether the docking conformation is in a low-energy torsion state (energy difference ≤ 3 kcal / mol).

[0035] Comprehensive scoring and interaction energy analysis: The total binding free energy and the contribution of each component were calculated using the MM-GBSA method, with a focus on the electrostatic interaction energy and van der Waals interaction energy between the ligand and key residues in the active pocket, which were corroborated by the hydrogen bond analysis results.

[0036] Establish a comprehensive evaluation matrix: assign indicators such as docking score, number of hydrogen bonds, occupied sub-region, ligand torsional strain energy, and key residue interaction energy contribution to each important docking conformation, and comprehensively judge the most reasonable bioactive binding mode.

[0037] Key domain of action identified: Comprehensive analysis showed that both substrates formed stable binding with the region of residues 83-90. This region has ≥3 hydrogen bonds and a binding free energy ≤-6.0 kcal / mol, thus formally identifying the region of residues 83-90 as the key domain of the enzyme's active site (e.g., ...). Figure 8 As shown in Figure 9, this area is determined to be the optimal pocket based on the final score.

[0038] 6. Virtual saturation mutation and screening: (1) Select glycine at position 85 (G85) in the key domain and asparagine at position 120 (N120) at the edge of the active pocket, and replace them with 19 other natural amino acids in turn using a mutant modeling tool to generate a mutant library containing single-point mutants and combined mutants.

[0039] (2) Functional impact prediction: Using the PolyPhen-2 and Provean online tools, submit the amino acid sequence of the target protein and the G85A mutation to be analyzed, predict the possible biological impact of the point mutation on protein function, and screen for “benign” or “neutral” mutations.

[0040] (3) Stability Change Prediction: Stability prediction servers based on different algorithms, such as PoPMuSiC, DUET, MaestroWeb, and SAAFEC, were used. The protein's PDB ID and mutation information were input, and the predicted change in folding free energy (ΔG) was calculated. The stability change prediction results after replacing G85 with 19 natural amino acids are shown in the table below. Figure 10 Mutants with ΔΔG values ​​close to zero or slightly negative were screened: G85A, G85S, N120S, and the combined mutant G85S / N120S. The selection criteria for the folding free energy change value (ΔΔG) are as follows: Mutations with ΔΔG ≤ 0 kcal / mol (or ≤ 0 kJ / mol) are preferred (core screening targets). After mutation, the protein folding free energy is reduced, and the structure is more stable (the larger the negative ΔΔG value, the more significant the improvement in stability). Typical range: -2.0 ≤ ΔΔG ≤ 0 kcal / mol (or -8.4 ≤ ΔΔG ≤ 0 kJ / mol). This range ensures improved stability while avoiding conformational rigidity due to excessive stability (which affects the catalytic flexibility of the enzyme).

[0041] Mutations with 0 < ΔΔG ≤ 0.5 kcal / mol are acceptable (neutral mutations). These mutations have minimal impact on stability, and the protein conformation remains largely unchanged. They do not lead to loss of enzyme activity due to structural instability. If the mutation can significantly enhance substrate binding affinity (e.g., by creating new hydrogen bonds), even if ΔΔG is slightly positive (≤0.5 kcal / mol), it can still be considered a candidate (further molecular dynamics simulations are needed to verify conformational stability).

[0042] ΔΔG > 0.5 kcal / mol indicates the exclusion of mutations / harmful mutations. After mutation, the protein's folding free energy increases, its structural stability decreases, and it is prone to unfolding or conformational disorder.

[0043] If ΔΔG > 1.0 kcal / mol, it must be excluded directly.

[0044] (4) Residue interaction network analysis: Comparing the hydrogen bond and ionic bond interaction networks of the enzyme before and after mutation, it was confirmed that G85S can form a new hydrogen bond with D83, as shown in Figure 11. In the docking conformation of the G85S mutant with S-adenosylmethionine, the side chain hydroxyl group of Ser85 forms a stable hydrogen bond with the carbonyl oxygen atom of the D83 main chain with a bond length of 2.7 Å. At the same time, it forms an additional hydrogen bond with the ribosome of the substrate, which enhances the substrate binding stability. N120S can enhance the secondary interaction with the substrate without steric conflict.

[0045] (5) Comprehensive evaluation and priority ranking: Integrate the results of functional prediction, stability, structural change, flexibility change, and interaction network change, establish a multidimensional scoring table, rank the mutations according to the goal of "improving substrate affinity without compromising stability", and screen out the most promising mutations for subsequent validation.

[0046] 7. Simulation effect analysis of the optimal candidate mutation: Protein folding stability verification: Integrating the prediction results of multiple tools, the mean ΔΔG of the G85S mutation is -0.15±0.25 kcal / mol, the ΔΔG of the G85A mutation is close to zero, and the ΔΔG of the N120S mutation is -0.32 kcal / mol, all of which meet the stability requirements.

[0047] Local interaction network analysis of the active pocket: After the G85S mutation, the side chain hydroxyl group of Ser85 can form a new hydrogen bond with the carbonyl oxygen atom of the D83 main chain within a distance of 2.7 Å; the hydroxyl group introduced by the N120S mutation can form a secondary hydrogen bond with the amino group of the substrate putrescine, further stabilizing the substrate binding conformation.

[0048] Rigidity prediction of domain conformation: RMSF (root mean square fluctuation) analysis of molecular dynamics simulations showed that the atomic position fluctuation of residues 83-90 in the G85S mutant was reduced by an average of about 15% compared with the wild type, and the rigidity of this region was further enhanced in the G85S / N120S combined mutant.

[0049] Substrate binding affinity prediction: The binding free energy was calculated using the MM-GBSA method. The G85S mutant showed an increase of approximately -0.8 kcal / mol, as shown in Figure 12. In the docking conformation of the G85S mutant with putrescine, the substrate formed four hydrogen bonds with the active site residues. The decrease in binding free energy is directly related to the increase in the number of hydrogen bonds, confirming the enhanced binding affinity. The G85A mutant showed an increase of approximately -0.5 kcal / mol, and the G85S / N120S combined mutant showed an increase of approximately -1.5 kcal / mol, both demonstrating a significant enhancement in binding affinity.

[0050] 8. Molecular dynamics simulation: (1) Construction of simulation system: Using GROMACS 2019.6 software, based on the optimal docking conformation, a complex system of wild-type SpeE and candidate mutants with dual substrates was constructed. The AMBER ff14SB force field and TIP3P water model were adopted, and an ion neutralization system was added to make the ionic strength of the system 0.15 M.

[0051] (2) Energy minimization and equilibrium simulation: The steepest descent algorithm was used for energy minimization (maximum number of steps 50,000, energy convergence criterion 1,000 kJ / (mol)). nm), followed by 3ns NVT equilibration (temperature controlled at 300 K, using a V-rescale thermostat) and 3ns NPT equilibration (pressure controlled at 1 bar, using a Parrinello-Rahmanbarostat).

[0052] (3) Production simulation: Conduct 100 ns production simulation, save a frame trajectory every 10 ps, ​​and analyze the RMSD of main chain atoms and the RMSF of key regions (83-90 residues).

[0053] 9. Binding Free Energy Calculation: The gmx_MMPBSA program was used to calculate the binding free energy of the simulated trajectory, decomposing components such as van der Waals energy, electrostatic potential energy, and solvation energy. The difference in binding free energy between the wild type and the mutant was compared, and mutants with significantly reduced binding free energy were screened. Based on all the analysis results, G85S and G85A were determined to be the optimal single-point mutants, and G85S / N120S was determined to be the optimal combination mutant. Among them, the G85S mutant showed a 15% reduction in RMSF in the key region, and the average binding free energy with the two substrates was -85.44 kJ / mol, which was significantly better than that of the wild type (-60.30 kJ / mol).

[0054] Example 2 (1) Gene template preparation: Bacillus subtilis strain 168 (purchased from ATCC, catalog number 14580) was used to extract genomic DNA using a bacterial genomic DNA extraction kit (brand: TIANGEN, catalog number: DP302) according to the kit instructions. The DNA integrity was verified by 1% agarose gel electrophoresis (clear bands without degradation). The purity was detected by Nanodrop (A260 / A280=1.8-2.0). The DNA was stored at -20℃ for later use.

[0055] (2) Wild-type speE gene amplification: Specific primers were designed (upstream primer: 5'-CGGAATTCATGAGTGAACTTTGGTA-3', containing EcoRI restriction site; downstream primer: 5'-CCGCTCGAGTTATTTATCAGGTCAC-3', containing XhoRI restriction site), and the primers were synthesized by Sangon Biotech (Shanghai) Co., Ltd. PCR reaction system (50 μL): 25 μL 2×Taq PCR MasterMix, 2 μL each of upstream and downstream primers (10 μM), 1 μL genomic DNA template, and 20 μL sterile water. PCR reaction conditions: 95℃ pre-denaturation for 5 min; 95℃ denaturation for 30 s, 55℃ annealing for 30 s, 72℃ extension for 1 min, for a total of 35 cycles; final extension at 72℃ for 10 min. The amplification product was recovered by agarose gel electrophoresis (kit: Omega, catalog number: D2500-02) to obtain the wild-type speE gene fragment (approximately 800bp).

[0056] (3) Site-directed mutagenesis to construct the mutant gene: Construct the speE-G85S mutant gene: Design primers containing the mutation site (upstream: 5'-GGTGGTGACGGCGGCGTTATCAGAGAA-3', downstream: 5'-TTCTCTGATAACGCCGTCACCACC-3'), using the recovered wild-type speE gene as a template, and amplify using overlap extension PCR. First round PCR: Two groups of reactions were used to amplify the upstream and downstream fragments containing the mutation site, respectively. The reaction system was the same as before, and the annealing temperature was 57℃. Second round PCR: The products from the two groups of the first round were mixed as templates, and the complete mutant gene was amplified using full-length primers. The annealing temperature was 55℃.

[0057] Constructing the speE-G85A mutant gene: Designed mutation primers (upstream: 5'-GGTGGTGCTGGCGGCGTTATCAGAGAA-3', downstream: 5'-TTCTCTGATAACGCCGCCAGCACCACC-3'), and repeated the above overlapping extension PCR operation to obtain the speE-G85A mutant gene.

[0058] Mutant gene verification: Two mutant gene fragments were cloned into the pMD19-T vector (Takara, catalog number: D102A), transformed into E. coli DH5α competent cells, and positive clones were selected and sent to Sangon Biotech for sequencing to verify that the mutation sites were correct and that there were no other random mutations.

[0059] (4) Construction and transformation of recombinant vectors: Wild-type and sequence-verified mutant gene fragments, and pET-28a(+) vector (Novagen, catalog number: 69864-3) were digested with EcoRI and XhoI restriction endonucleases (NEB, catalog numbers: R0101S and R0146S) at 37℃ for 4 h, respectively. The digestion products were recovered by gel electrophoresis and ligated overnight at 16℃ using T4 DNA ligase (Takara, catalog number: D2011A). The ligation products were transformed into Escherichia coli BL21 (DE3) competent cells (Tiangen, catalog number: CB105), plated on LB solid medium containing 50 μg / mL kanamycin, and cultured at 37℃ for 12-16 h. Single colonies were picked for colony PCR verification, and positive clones were identified as recombinant engineered bacteria.

[0060] (5) Protein induction expression and purification: Seed culture: Select a single colony of the positive recombinant engineered bacteria, inoculate it into LB liquid medium containing 50 μg / mL kanamycin, and culture at 37℃ and 200 rpm for 12 h to obtain the seed culture.

[0061] Induction of expression: Seed culture was inoculated into 500 mL LB liquid medium (containing 50 μg / mL kanamycin) at a ratio of 1:100, and cultured at 37℃ and 200 rpm until OD600 = 0.6-0.8. IPTG was added to a final concentration of 0.5 mM, and expression was induced at 18℃ and 180 rpm for 16 h.

[0062] Protein purification: Induced bacterial cells were collected, resuspended in buffer A (50mM Tris-HCl pH 8.0, 300mM NaCl, 10mM imidazole), and sonicated (300W, 3s operation, 5s interval, total 30min). The supernatant was collected by centrifugation at 12000rpm for 20min at 4℃. The supernatant was passed through a Ni-NTA affinity chromatography column (GE Healthcare, catalog number: 17-5318-01). The column was first equilibrated with 5 column volumes of buffer A, then eluted with a gradient of buffer A containing 20mM, 50mM, and 250mM imidazole. The fraction corresponding to the 250mM imidazole elution peak was collected. Protein purity was verified by SDS-PAGE electrophoresis (single clear band, purity ≥90%). After purification, the protein was dialyzed overnight in buffer B (50mM Tris-HCl pH 8.0, 150mM NaCl) using a dialysis bag (molecular weight cutoff 10kDa) to remove imidazole. After aliquoting, the protein was stored at -80℃ for later use.

[0063] (6) In vitro enzyme activity assay: Standard reaction system (100 μL): 50 mM Tris-HCl pH 8.0, 2 mM DTT, 1 mM putrescine (Sigma, catalog number: P7505), 1 mM S-adenosylmethionine (Cayman, catalog number: 13519), 0.1 μM purified protein.

[0064] Reaction conditions: React in a 37℃ water bath for 30 min, add 100 μL of methanol to terminate the reaction, centrifuge at 12000 rpm for 10 min and collect the supernatant.

[0065] Detection method: HPLC (Agilent 1260) was used with a C18 column (4.6 mm × 250 mm, 5 μm), a mobile phase of methanol:water = 30:70 (v / v), a flow rate of 1.0 mL / min, a detection wavelength of 254 nm, and a column temperature of 30 °C. A standard curve was plotted using spermidine standard (Sigma, catalog number: S2626), and the amount of spermidine generated in the reaction system was calculated.

[0066] Results: The G85S mutant had a specific enzyme activity of 48.7 U / mg (defined as the amount of enzyme required to catalyze the production of 1 μmol spermidine per minute), which was 2.2 times that of the wild type (22.1 U / mg); the G85A mutant had a specific enzyme activity of 35.6 U / mg, which was 1.6 times that of the wild type. The experiment was repeated 3 times, and the RSD was ≤5%.

[0067] (7) Whole-cell catalysis verification: Seed culture: Select a single colony of recombinant engineered bacteria carrying the speE-G85S mutant gene, inoculate it into LB liquid medium containing 50 μg / mL kanamycin, and incubate at 37℃ and 200 rpm for 12 h.

[0068] Fermenter culture: The seed culture was inoculated into a 5L fermenter (brand: BIOTECH-5BG) at a 10% inoculation rate. The fermentation medium was LB medium (containing 50μg / mL kanamycin). The initial liquid volume was 3L. The fermentation was carried out at 37℃, 200rpm and 1vvm aeration until OD600=10.0 to obtain whole-cell catalyst.

[0069] Catalytic reaction: Add putrescine and S-adenosylmethionine to the fermenter to a final concentration of 30 mM each, adjust the pH to 7.0, maintain the temperature at 30℃, the stirring speed at 200 rpm, and the aeration rate at 0.8 vvm, and carry out batch feeding catalysis. Add substrate (15 mM each) every 6 hours to avoid substrate depletion.

[0070] Product detection: After 36 hours of reaction, 1 mL of sample was taken, centrifuged at 8000 rpm for 10 min, and the supernatant was collected. The spermidine yield was detected by HPLC method in step (6). The result was 15.8 g / L, which was significantly higher than the wild-type engineered bacteria's 6.9 g / L.

[0071] Example 3 Construction and Validation of Combinatorial Mutants (1) Construction of combined mutant gene: Using the speE-G85S mutant gene recombinant vector verified in Example 2 as a template, primers specific to the N120S mutation site were designed (upstream: 5'-GAGAACATGCTTTTTTTGGACGG-3', downstream: 5'-CCGTCCAAAAAAAGCATGTTCTC-3'). The double-point mutant speE-G85S / N120S was constructed using a site-directed mutagenesis kit (Agilent, catalog number: 200518). The operation was performed according to the kit instructions: the primers, template vector, and enzyme mixture were placed in a 50 μL system for PCR amplification (95℃ for 30s, 55℃ for 1min, 68℃ for 10min, for a total of 18 cycles). The amplification product was digested with DpnI enzyme (NEB, catalog number: R0176S) at 37℃ for 1h to remove the original template vector.

[0072] (2) Recombinant vector verification and transformation: The digested PCR product was transformed into Escherichia coli DH5α competent cells, and positive clones were selected for sequencing verification to confirm that the G85S and N120S double mutation sites were correct and there were no other mutations. The recombinant vector with correct sequencing was transformed into Escherichia coli BL21 (DE3) competent cells according to the method in step (4) of Example 2 to obtain the combined mutant recombinant engineered bacteria.

[0073] (3) Protein induction and purification: Seed culture, low temperature induction and Ni-NTA affinity chromatography purification were carried out according to the method of step (5) in Example 2. The protein purity was verified by SDS-PAGE electrophoresis to be ≥90%. After dialysis, the protein was stored at -80℃ for later use.

[0074] (4) In vitro enzyme activity assay: The enzyme activity was assayed according to the standard reaction system, reaction conditions and HPLC detection method of step (6) in Example 2. The results showed that the in vitro specific enzyme activity of the speE-G85S / N120S combined mutant reached 56.3 U / mg, which was 19.7% higher than that of the G85S single-point mutant, demonstrating a synergistic effect. The experiment was repeated 3 times, and the RSD was ≤5%.

[0075] (5) Whole-cell catalysis verification: Seed culture, fermentation tank expansion culture and batch feeding catalysis were carried out according to the method of step (7) in Example 2. After 40 hours of reaction, samples were taken for testing. The spermidine yield reached 18.2 g / L, which is 15.2% higher than the 15.8 g / L of the G85S single-point mutant, further verifying the synergistic effect mechanism of the double mutation site.

[0076] Comparative Example 1 The procedure was the same as in Example 1, except that only one of the main substrates, S-adenosylmethionine, was coupled to the active pocket of Bacillus subtilis SpeE for analysis.

[0077] Cluster analysis of docking results shows (e.g.) Figure 13 The adenosine and sulfur atom of S-adenosylmethionine are stably bound in a conserved hydrophobic pocket, while its propylamine side chain extends into a region composed of residues such as threonine at position 158 (T158). Interaction analysis revealed a potential hydrogen bond interaction (approximately 3.2 Å) between the hydroxyl group of T158 and an oxygen atom on the S-adenosylmethionine molecule. Based on this single-substrate binding model, T158 was arbitrarily identified as a key "hotspot residue" influencing the binding of S-adenosylmethionine.

[0078] To verify this prediction, T158 was mutated to alanine (T158A) to eliminate the hydroxyl group and investigate its effect on binding. Following the same procedure as in Example 2, the recombinant plasmid pET28a-SpeE(T158A) was constructed and expressed and purified in *E. coli* BL21(DE3) to obtain the mutant protein.

[0079] In vitro enzyme activity assays showed that the specific activity of the T158A mutant was only 8.5 U / mg, a significant decrease of approximately 61% compared to the wild-type enzyme (22.1 U / mg). This result indicates that the mutation at the T158 site, selected based on the S-adenosylmethionine single-substrate docking model, severely impaired enzyme function. Retrospective analysis revealed that under true dual-substrate catalysis, T158 may not directly form a strong hydrogen bond with S-adenosylmethionine, but rather plays a more complex role: its side-chain hydroxyl groups may indirectly stabilize the substrate through a water molecule network, or more critically, it plays a vital spatial or polar interaction with neighboring residues (such as regions 83-90) in the dual-substrate binding-induced conformational closure of the active pocket. The single-substrate docking model completely failed to capture this cooperative conformational change, leading to a misjudgment of T158 function and resulting in harmful modification decisions.

[0080] Comparative Example 2 The procedure is the same as in Example 1, except that only the putrescine molecules are docked to the SpeE active pocket. The docking results show stable performance (e.g., Figure 14 The two ammonium cations of putrescine form strong electrostatic interactions with conservative acidic residue clusters (such as D83 and E89), while its alkane chain is embedded in a hydrophobic channel. In this model, the aromatic ring side chain of tyrosine at position 62 (Y62) forms a significant π-alkyl hydrophobic contact and van der Waals interaction with the end of the putrescine alkane chain, and is identified as a key residue for stabilizing putrescine binding.

[0081] To "optimize" this hydrophobic interaction, Y62 was mutated to phenylalanine (Y62F) to retain the hydrophobic properties of the aromatic ring while removing the phenolic hydroxyl group, thus providing a simpler hydrophobic interface. Following the method in Example 2, plasmid pET28a-SpeE(Y62F) was constructed and the protein was expressed and purified.

[0082] Enzyme activity assays showed that the specific activity of the Y62F mutant was 28.5 U / mg, an increase of approximately 29% compared to the wild type (22.1 U / mg). While this increase exists, it is far less significant than that of the G85S mutant (approximately 120% increase) obtained through dual-substrate analysis in this invention. This indicates that mutations guided solely by the putrescine binding model can only fine-tune the binding environment of a single substrate, representing a localized optimization. The Y62 site may primarily affect the putrescine entry channel or its initial localization within the pocket, but it has no decisive impact on the precise placement of subsequent S-adenosylmethionine or the formation of the transition state between them.

Claims

1. A method for the rational modification of spermidine synthase (SpeE) in Bacillus subtilis, characterized in that, Includes the following steps: S1: Obtain the gene and protein sequences of Bacillus subtilis spermine synthase (SpeE) and construct a high-confidence three-dimensional model of spermine synthase (SpeE) using homology modeling technology. S2: Simultaneously perform molecular docking of the two substrates for synthesizing spermidine with the three-dimensional model described in S1, analyze the binding mode, interaction type and strength, small molecule torsion state and binding free energy of the substrate and enzyme, and collaboratively determine the key action domain of the enzyme active site. The two substrates are S-adenosylmethionine and putrescine; S3: Perform a virtual saturation mutation on at least one residue within the key functional domain, and replace it with one of the other 19 natural amino acids in sequence to generate a mutant library; S4: Through functional impact prediction, protein stability calculation, mutant three-dimensional structure modeling, and residue interaction network analysis, the mutant library is comprehensively screened from multiple dimensions to obtain candidate mutants; S5: Perform molecular dynamics simulations and binding free energy calculations on the screened candidate mutants to verify the conformational stability and binding affinity with the two substrates, and determine the optimal mutant; S6: The catalytic performance and spermidine production ability of the optimal mutant described in S5 were verified by in vitro enzymatic property determination and whole-cell catalysis experiments.

2. The rational transformation method according to claim 1, characterized in that, The construction of the high-confidence 3D model described in S1 requires the use of structural quality assessment tools to validate the model. Validation indicators include the model's global quality score, the conformational rationality of amino acid residues, and steric hindrance conflicts, ensuring that the model meets the accuracy requirements for subsequent molecular docking. The structural quality assessment tools include at least one of QMEANDisCo, Ramachandran analysis tools, and MolProbity tools. The validation criterion for the global quality score is a GMQE value ≥ 0.6 for the model.

3. The rational transformation method according to claim 1, characterized in that, The key active site of the enzyme described in S2 is composed of amino acid residues at positions 83-90. The molecular docking operation includes: converting the three-dimensional model and the two substrates into PDBQT format, removing water molecules from the model and adding polar hydrogen atoms, setting a docking pocket to cover the region where amino acid residues at positions 83-90 are located, performing cluster analysis according to RMSD≤2 Å, and selecting the central conformation of each cluster as the optimal conformation.

4. The rational transformation method according to claim 3, characterized in that, The molecular docking software is any one of Autodock, Autodock Vina, Discovery Studio, or Glide.

5. The rational transformation method according to claim 3, characterized in that, The residues within the key domain described in S2 include glycine at position 85.

6. The rational transformation method according to claim 1, characterized in that, The tools used for predicting the functional impact described in S4 include at least one of PolyPhen-2 and Provean; the tools used for calculating protein stability include at least one of PoPMuSiC, DUET, and SAAFEC. The molecular dynamics simulation described in S5 shall last for no less than 100 ns, and the force field used in the simulation process shall include the AMBERff14SB force field, and the solvent model shall use the TIP3P water model. The in vitro enzymatic properties measured in S6 include specific enzyme activity, catalytic efficiency, and substrate affinity.

7. A Bacillus subtilis spermidine synthase (SpeE) mutant obtained by the method according to any one of claims 1-6, characterized in that, The mutant has mutation sites including the glycine mutation at position 85 of Bacillus subtilis SpeE, and may also include the asparagine mutation at position 120.

8. The Bacillus subtilis spermidine synthase (SpeE) mutant according to claim 7, characterized in that, The mutation at position 85 refers to the mutation of glycine at position 85 into serine or alanine.

9. The Bacillus subtilis spermidine synthase (SpeE) mutant according to claim 7, characterized in that, The mutation at position 120 of asparagine refers to a site-directed mutation of asparagine at position 120 into serine.

10. A method for the biosynthesis of spermidine, characterized in that, The Bacillus subtilis spermidine synthase (SpeE) mutant according to any one of claims 7-9 is heterologously expressed in a host strain, and spermidine is synthesized through whole-cell catalysis using S-adenosylmethionine and putrescine as substrates.

Images