Improving The Thermostability Of Aspergillus Terreus Transaminase By Coevolutionary Network

Transaminases have become the most prominent biocatalysts for the generation of optically pure chiral amines through asymmetric synthesis and kinetic resolution because of their high stereoselectivity and environmentally mild reaction conditions.With the continuous expansion of the chiral drug market,transaminases have great potential for the bio-industry applications.However,most of the transaminases have poor thermostability,which severely limits their commercialization.Improving their thermostability is not only a challenging task for transaminase protein engineering,but also a critical problem to be addressed in industrial application of transaminases.In this study,the R-ω-transaminase derived from Aspergillus terreus was used as model.A coevolution network ofω-transaminase was generated by analyzing their homologous protein sequences using MISTIC server.The mutual information between residues was calculated to predict functionally important residues in the coevolutionary network.A subnetwork consisting of eight strongly interaction residues was filtered by setting the thresholds for the four metric parameters.Firstly,alanine scanning was performed on those predicted eight residues to construct mutant libraries.The specific activity of the mutant enzymes was determined,and the mutant L118A showed2.57-fold enhancement of specific activity compared to wild type.By thermostability analysis,the T_m,T₅₀¹⁰ and t_1/2 at 40°C values of the mutant L118A showed a 2.1°C,3.7°C and 13.7 min increase relative to the wild-type enzyme,respectively.Saturation mutagenesis library was constructed at residue position L118.Nearly300 colonies were randomly selected.The residual activity of enzymes which treated at50°C for 10 min was used to identify positive mutants.Subsequently,sequencing displayed that the most thermostable single mutant was L118T.By thermostability analysis,the T_m,T₅₀¹⁰ and t_1/2 at 40°C values of the mutant L118T showed a 5.3°C,5.0°C and 19.2 min increase relative to the wild-type enzyme,respectively.Furthermore,the most stable mutant L118T displayed the highest catalytic efficiency of the substrate（R）-α-methylbenzylamine and pyruvate.More importantly,UPLC-MS analysis proved that the substitution of Leu118 with Thr had no effect on the chiral selectivity of AT-ATA.Molecular dynamics simulation revealed that the overall average RMSD value of the L118T structure was lower than that of wild-type protein structure during 100 ns simulation at 313 K.The region of residues Val170-Gly190 in mutant L118T structure is less flexible than in the wild-type structure according to RMSF,indicating that more stable conformation was achieved for these residues of the mutant L118T.To further improve enzyme thermostability,the stabilized single mutant L118T was used as a template for iterative saturation mutagenesis combined with other three related residues（F115,N181 and W184）.After screening in the same heat treatment conditions,the best double variant was identified as L118T-F115L.The t_1/2 at 40°C of L118T-F115L was 65.9 min,which was approximately 9.55-fold higher than the value of the wild-type enzyme.The T₅₀¹⁰ values of L118T-F115L increased to 47.3°C,which is approximately 8.8°C increase relative to wild type.The T_m of L118T-F115L was7.7°C higher than that of the wild type.Molecular dynamics simulation analysis indicated that the double mutation L118T-F115L has lower RMSD than wild type,and the mutation makes its structural rigidity enhanced.An irrational design method,analyzing the mutual information between residues in the co-evolutionary network to infer functionally important residues,was used to guide the molecular modification ofω-transaminase and significantly improved their thermostability.This study not only provides excellent biocatalyst for the asymmetric synthesis of chiral amines,but also guidance for further improving the thermostability and catalytic efficiency of other enzymes.