Study The Thermal Stability Of Lipase Based On Complex Network Theory And Molecular Dynamics Simulation

The Lipase has vital applications in industry.However,due to the poor thermal stability of wild type lipase,it makes its application subject to many restrictions.This makes it necessary to study the factors that affect the thermal stability of lipases.In this study,long time molecular dynamics simulations were carried out to study salt bridges,Hydrogen Bond(HB)dynamics and HB networks of WTL(WTL,PDB ID:1I6W,T50=56℃)and its two mutants(2D9,PDB ID:3D2B,T50=67.4℃,6B,PDB ID:3QMM,T50=78.2℃).To fully analysis the difference between the WTL and mutants,full-atom network of amino acids were constructed.We computed the network parameters,and used the time series and BP neural network algorithm to predict the trend of amino acid network topology at different temperatures.The main contents of this paper are as follows:1.Molecular dynamics simulations were performed to simulate the protein kinetic profiles of wild-type lipase and mutants at 300 K,325K,350 K,375K,400 K at 300 ns.The stability of the hydrogen bond of the wild type and mutants were analyzed at different simulated temperatures.The stability of the hydrogen bond of mutants formed by N166 Y are found to has significantly difference from that of WTL.In addition,the hydrogen bonds formed by four mutations,A15 S,A20E,G111 D,M137P(only found in the more thermostable mutant 6B)are more stable than those of WTL and mutant 2D9.In addition,we analyzed the stability differences of hydrogen bonds at non-mutation sites.The results indicate that these differences affect the secondary structure stability of proteins,which is a direct factor for their differences in thermal stability.2.The persistence values of all hydrogen bonds in WTL,2D9,and 6B(the persistence values of hydrogen bonds must be greater than 0.3)were calculated.These hydrogen bonds form six hydrogen-bonding networks,and we focus on three of them,because some hydrogen bonds persistence values of them are significantly higher than that of the WTL as the temperature increasing.With the increase of temperature,the first network shows stronger ability to maintain the mutant secondary structure β3,β5,loop and 310 helix,and the second hydrogen bonding network makes the loop,αC,β3 and β5 of mutants more stable,the third hydrogen bonding network enhances the interaction of loop,αB and αC.3.The persistence values of salt-bridges at each temperature of wild-type lipase and mutants were calculated.The stability of salt bridges and salt bridge networks were analyzed to study the relationship between salt bridge and lipase thermal stability.The stability of the salt bridges Lys35-Asp34,Asp118-Lys70,Glu171-Arg147 and Asn181-Lys122 at mutants are found to be quite different from those of wild type lipase.In addition,due tothe mutations,four new salt bridge networks are formed in the mutants.The first salt bridges network(only presents in the mutant 6B)enhances the internal interaction of the helix αA,and makes the αA,loop and 310-helix connect to each other.The second salt bridges network reduces the flexibility of helix αB of the mutant 2D9 and 6B.The third salt bridges network strengthens the relationship between the loop region,αE and αD of the mutant 6B.The fourth salt bridge network enhances the rigidity of the loop and the amino acid sequence terminal of the mutant 2D9.4.Based on the three-dimensional information of WTL and mutants,the network was constructed and the parameters of complex network were calculated,such as average degree,characteristic path length and so on.The values of characteristic path length of the mutants are generally lower than that of WTL,while other parameters are slightly higher than those of WTL.5.Amino acid network parameters of WTL and mutants were predicted using time series algorithm and neural networks.The exponential smoothing and ARMA models were used respectively for time series algorithm and BP neural network was used for neural network.Three methods were used to predict the topological parameters of the wild-type and the mutant at each temperature,showing high accuracy.It indicates that the time series algorithm and BP neural network can be used to predict the lipase network parameters,and then analyze the thermal stability of lipase.It is found that the stability of the hydrogen bonds in the lipase are directly or indirectly changed by the mutations,and influences the stability of the secondary structure of the lipase.In addition,the dynamic changes of the topological properties of the amino acid network also show the insensitivity of lipase mutants to temperature,which is consistent with the better thermal stability of the mutants.Finally,the time series algorithm and the neural network are used to predict the topological properties of the network at different temperatures and obtain high accuracy,which furtherly means the relationship between network topology and lipase thermostability.