Design, Selection And Application Of Specific Molecular Interactions With Proteins

The specific interaction between the protein and its substrate is one of the material basements of protein functions. The design, improvement and application of the specific recognition are well concerned in protein engineering and synthetic biology. Here, we tried to theoretically and experimentally develop the more efficiency strategies in the design, improvement and application of the specific interaction between the protein and the substrate.In the strategy to design the specific recognition pairs. we focus on the theoretical method of protein interface design to the ligand, which is to design a protein interface that can specifically bind the target ligand. It is an important issue in protein functional design. The existed score function to screen the candidates is to evaluate the binding affinity between the protein and the target ligand. However, this score function can not confirm that the ligand is the only target with the design result can interact with. This issue can be addressed by negative design to some extent, but how to choose the competing ligands, which is necessary to negative design, becomes a new issue we should addressed. We introduced local interaction optimization in protein functional design, which is to optimize every local interaction between the protein and the ligand. If the aim can be reached, the results will show good performance in binding affinity, as the whole interaction is the sum of every local ones. Further more, the results will also show good performance in binding specificity, as the perturbation on the ligand will disturb the interaction network and decrease the binding affinity between the protein and the ligand. From this point, we developed a dual-phase design strategy based on genetic algorithm to design the protein-ligand interface with binding affinity and specificity. The strategy is trained and examined by the complex structures from PDBB1ND2005 and PDBBIND2007. X-score is employed to evaluate the binding affinities and the binding specificities of the results, comparing with the native structures and the results obtained with single-phase strategy, which is employed the binding affinity with the protein and the ligand as the score function. The results show that the binding affinity of the design results with dual-phase strategy, single-phase strategy and the native structure are at the same level. However, for specificity, the design results with dual-phase strategy show better performance than the results with single-phase strategy and the native structures. Comparing with the native structures, the binding specificity of the results by dual-phase design strategy improves 1-2 magnitudes in average, and the maximum improvement is 6 magnitudes. In this way, a new protein-ligand interface design strategy for binding specificity is developed.In the strategy to improve the binding affinity of the specific recognition pairs, we focus on the binding affinity between the antibody and the antigen, and tried to find a new strategy to improve the efficiency of antibody maturation based on new technologies and computational tools. Antibody maturation is a key step for an antibody from discovered to clinical application. The existed strategy is to construct mutated library based on the naked antibody, and screen the mutation library with high-throughput screen protocol, finally get the mutants with improved binding affinities with se(?)-quantitative and quantitative method. The screen results highly depend on the efficiency and the quality of the mutant library. However, it is hard to construct the required library with traditional oligo synthesize and also hard to measure the quality of the library. We develop a new strategy to construct the mutant library based on high-throughput oligo synthesize and high-throughput sequencing. First, with the bioinformatics tools, the mutation range of each position has been defined based on the CDR regions of the naked antibody and the sequences in the same cluster with each CDR. Then the mutant library is obtained by high-throughput oligo synthesize by introducing ambiguity codes. And the next step is sequencing the library with high-throughput sequencing and the synthesize strategy is optimized according to the sequencing results, synthesize again. Finally, the high'quality mutant library is obtained. With A21 as the template, we construct position-saturation perturbation library with same copies of each mutant. At the same time, we also tried to get the improvement of binding affinity of the mutants based on the high-throughput sequencing results of the mutant libraries before and after the screening. The verified experiments are improving.In the strategy to apply the specific recognition pairs, we focus on the specific interaction between the protein and the DNA, and tried to find the general strategy for developing more protein-DNA recognition pairs and new modes to realize more complex dynamic behaviors based on these pairs. The key positions for protein and DNA recognition are found by analyzing the 3-dimensional structure of the Lacl and its binding sites. The saturation mutant library is constructed theoretically and experimentally. The new pairs are found out based on the screen. And then, by theoretical model, we defined the range of the parameters for "NAND" and "NOR" gate based on protein-DNA interactions. These restrictions are successfully realized by experiments and the dynamic behaviors of "NAND" and "NOR" gates are successfully seen in vivo. So based on these results. a new strategy to lind out more elements for synthetic biology are developed.