| Adipic acid is one of the most valuable medium-chain dicarboxylic acids in the chemical industry and it is mainly used in the synthesis of nylon 6,6 or polyurethane materials.Adipic acid production from glucose by microorganisms possesses the great significance in the current context of global pollution reduction and carbon reduction.At present,the work of microbial synthesis of adipic acid focused on strain modification and fermentation conditions optimization,while the understanding was limited in the modification and catalytic mechanism of key enzymes in the pathway.Our laboratory has previously developed an efficient synthesis pathway of adipic acid(the reverse adipic acid degradation pathway).Among them,the key rate-limiting enzymesβ-ketothiolase Tfu_0875 and acyl-Co A dehydrogenase Tfu_1647 catalyzed Claisen condensation and double bond reduction,respectively.Therefore,it is helpful to further enhance the synthesis of adipic acid by analyzing their catalytic mechanisms and rationally modifying them and improving their catalytic performance.In this study,the expression conditions,crystallization conditions and crystal structure of Tfu_0875 and Tfu_1647 were optimized,and their catalytic mechanisms were clarified.By using machine learning to rationally modify the substrate tunnel of Tfu_0875,its enzyme activity and substrate specificity were significantly improved.In addition,the principle of local cationic domain modulated rule of theβ-ketothiolase substrate tunnel was proposed.Based on the structural characteristics of Tfu_1647,the substrate specificity was improved by semi-rational design of the substrate tunnel,and the mutant Tfu_1647T370Q with increased substrate specificity was obtained.Finally,the yield of adipic acid was increased by fermentation using the optimal mutants.The main research contents are listed bellow:(1)Analysis of the stereostructure and catalytic mechanism of Tfu_0875 in the adipic acid synthesis pathway.In view of the unclear catalytic mechanism of Tfu_0875,it was first heterologously expressed in E.coli,and the protein solution that met the crystallization requirements was obtained by a two-step purification method.The crystal of Tfu_0875 was obtained by gas phase diffusion method and analyzed by molecular replacement.The three-dimensional structure of Tfu_0875 was obtained and stored in the protein database(PDB ID:7VTR).The crystal structure of Tfu_0875 was an asymmetric tetramer formed by four identical monomers,and the catalytic active center of each monomer was a catalytic triad composed of Cys90-His348-Cys378.The substrate must pass through the substrate tunnel to approach the active center for catalyzing.According to the structural characteristics of the substrate,the substrate tunnel of Tfu_0875 can be divided into three regions:cationic domain,adenine binding pocket and pantetheine loop.(2)Machine learning-assisted rational design of the substrate tunnel of Tfu_0875 to improve enzyme activity.The reaction was carried out in the substrate tunnel of Tfu_0875.In order to explore the effect of substrate tunnel on enzyme activity of Tfu_0875,alanine scanning of 5(?)amino acids around the substrate tunnel was carried out,and 26 potential key amino acid sites were identified.DLKcat was used to predict the dominant mutants for potential key amino acid sites.Combined with enzyme activity detection,the best mutant Tfu_0875L163H was obtained,which increased the enzyme activity by 313%compared with Tfu_0875.Molecular dynamics simulation was used to compare the changes of the substrate tunnel,and it was found that the mutation of L163 to H163 could significantly improve the enzyme activity of Tfu_0875by affecting the volume of the cationic domain and the stability of the reaction intermediate or affecting the anchoring of the substrate.The local cation domain modification strategy(LCDMR)for the substrate tunnel of Tfu_0875 was further established,which could improve the activity of otherβ-ketothiolases.(3)Machine learning-assisted rational design of Tfu_0875 to improve substrate specificity.In order to improve the substrate specificity of Tfu_0875 catalyzing succinyl-Co A,alanine scanning was performed on the amino acids in 5(?)range of substrate tunnel around succinyl-Co A,and 16 potential key amino acid sites were identified.Then,DLKcat was used to screen the virtual effective mutants,and the optimal single mutant Tfu_0875N249W was obtained,which was144%higher than that of Tfu_0875.The effective single mutants were further modified using the greedy accumulated iterative site-specific mutagenesis strategy(GRAISM)with Tfu_0875N249Was a template.Subsequently,the optimal mutant Tfu_0875N249W/L163H/E217L was obtained and its enzyme activity was 221%higher than that of Tfu_0875.Moreover,the affinity for the substrate succinyl-Co A was increased by 3.55-fold.MD simulation was used to analyze the structure changes of the mutants,and it was found that the shape changes of the substrate binding pocket and the enhancement of the interaction between enzyme and substrate,which could help to improve the substrate specificity of succinyl-Co A.(4)Stereostructure analysis and substrate-specific modification of Tfu_1647.In view of the problem of Tfu_1647 lacking the three-dimensional structure,Tfu_1647 was heterologously expressed in E.coli.After purification,a protein solution that met the crystallization requirements was obtained,and then the crystal of Tfu_1647 was obtained by gas phase diffusion method.The three-dimensional structure of Tfu_1647 was obtained by molecular replacement and stored in the protein database(PDB ID:7W0J).The crystal structure analysis showed that Tfu_1647 was a tetramer,whose monomer was composed ofα-helix,β-sheet andα-helix,and its catalytic activity center was Glu368.Meanwhile,the cofactor FAD contained in the monomer was non-covalently embedded between theα-helix and theβ-sheet.The enzyme properties of Tfu_1647 were determined,it was indicated that Tfu_1647 with low substrate affinity for C6 substrate.In order to improve the specificity of Tfu_1647 to C6 substrate,the key sites were determined by alanine scanning of the amino acids within 5(?)range of the substrate tunnel of Tfu_1647,and then the key sites were semi-rationally mutated.The catalytic ability to C6 substrate was improved by changing the volume of substrate tunnel,and the enzyme activity of the optimal mutant Tfu_1647T370Q was increased 239%than Tfu_1647.(5)Reconstructing the synthetic pathway of adipic acid to produce adipic acid by fermentation.In order to explore the effect of the Tfu_1647 mutant after substrate-specific modification on adipic acid synthesis,E.coli K12 MG1655Δ8(Δato B,Δsuc D,Δpfl B,Δarc A,Δadh E,Δldh A,Δpox B,Δpta)was used as the chassis cell,and the mutant Tfu_1647T370Q replaced Tfu_1647 in the reverse adipic acid degradation pathway.The yield of adipic acid in shake flask was increased by 1.35-fold.Subsequently,in order to explore the effect of Tfu_0875 mutant after substrate-specific modification on adipic acid synthesis,Tfu_0875 in the reverse adipic acid degradation pathway was replaced by Tfu_0875N249W,Tfu_0875N249W/L163H and Tfu_0875N249W/L163H/E217L,respectively.The shake flask production of adipic acid in the chassis cell E.coli K12 MG1655Δ8 increased from 174.7 mg·L-1 to 188.0 mg·L-1,209.1 mg·L-1 and235.0 mg·L-1,respectively,which were 1.08-fold,1.20-fold and 1.35-fold of the control strain.Therefore,the synthesis of adipic acid could be promoted by modifying the substrate tunnel of Tfu_0875 and Tfu_1647 to improve its substrate specificity and enzyme activity. |