Improvement Of The Protein Thermal Stability By The Computer Aided Molecular Design

As the natural biological catalysts, enzymes usually have the high catalytic efficiency and substrate specificity. They have great potential and good prospects in industrial production. However, many enzymes have poor thermal stability, which greatly restricts the application of the enzymes in industrial production. Therefore, the studies on the mechanism of protein thermal stability and the rational design methods for improving the protein thermal stability have become the hotspots in the field of computational biology and protein engineering. It can not only expand the application scope of enzymes, but also help us to understand the relationship between protein structure and its function.In this study, we designed two strategies at the micro and macro level to screen the critical amino acids that could affect the protein thermal stability. These strategies have been applied to a methyl parathion hydrolase (MPH-Ochr) isolated from Ochrobactrum sp. M231, which aimed to test the effects of these two strategies and improve the thermal stability of the enzyme.The first strategy is to find the unstable regions with high conformational fluctuations out of the protein active center by the molecular dynamics simulations, identify the flexible amino acids (e.g. glycine), mutate the flexible residues (e.g. glycine) to the rigid ones (e.g. proline), and evaluate the effect of the mutation on the unstable region with high conformational fluctuations. MPH-Ochr was used to test the effect of the strategy. At first, an unstable region (residues 186–193) out of the protein active center was detected by the molecular dynamics simulations and two glycines (G194 and G198) were identified at the downstream of the region. Then three mutants (G194P, G194P/G198P and G198P) were constructed by the homology modeling. Finally, the influence of the mutants on the conformationally unstable region was tested by the method of molecular dynamics simulations. Only the mutant G194P was proved to improve the stability of the unstable region with high conformational fluctuations. Furthermore, the thermostability and kinetic behavior of the wild-type and three mutant enzymes were measured. The experimental results were consistent with the computational predictions. The T_m and T₅₀of G194P were higher 3.3°C and 4.6°C than that of the wild type enzyme respectively. The thermodynamic data revealed that the enthalpy change of G194P was similar to that of the wild type enzyme, but the entropy change of G194P was lower than that of the wild type enzyme. It indicated that the conformation of the mutant became stable, when it was dealt with high temperature. The experimental results revealed that this strategy was an effective method to improve the thermal stability of proteins.The second strategy is to select the residues related to the protein thermal stability based on the unfolding free energy difference (ΔΔG) between the mutant and wild type enzyme. Software, Prethermut, was developed to predict theΔΔG based on the protein structure featrues. The software had two distinct characteristics. Firstly, it was able to predict the effect of the single or multiple point mutations on the protein thermal stability. And secondly, a protein standard feature was introduced into the feature construction. The mutant structure feature was compared to the protein standard feature instead of the wild type structure feature. The S-dataset was used to evaluate the performance of Prethermut and compare it with other popular software. As a result, the accuracy of the classification and regression predictions was superior to that of existing commonly used software on the test dataset (S-dataset).The strategy was also been applied to MPH-Ochr in order to improve the thermal stability of the enzyme and test the effectiveness of the strategy. Seven sites in MPH-Ochr that might affect the thermal stability but not influence of the protein function were screened by the evolutionary entropy (E), the classification and regression results predicted by Prethermut. The seven site saturation mutagenesis libraries were constructed. The positive clones were screened from the libraries, and the thermal stabilities of the mutant enzymes were evaluated. As a result, there were six libraries that found the clones exhibiting better thermal stability than that of the wild type enzyme. It indicated that the strategy was an effective method to search the sites related to the protein thermal stability.A combination of mutation strategy was proposed based on the positive mutants selected from the six site saturation mutagenesis libraries. At first, the six sites were sorted according to the thermal stability of mutants. Then, the double, three and four point mutants were selected from the libraries that were constructed using the best single, double and three point mutant as the template at the second, third and fourth priority sites respectively. Finally, a mutant with four point mutations (274-183-197-192) was selected, which exhibited the best thermal stability in all of the mutants. The T_m and T₅₀ of the mutant 274-183-197-192 (S274Q/T183E/K197L/S192M) were higher than 11.7°C and 10.2°C of the wild type enzyme respectively. The results indicated that the protein thermal stability could be greatly enhanced by the rational combination of single point mutations.