Construction Of Plant Cytochrome P450 Enzyme Database

The cytochrome P450 genes,as the "universal biocatalyst" in the biosynthesis of plant natural products,have been widely concerned in synthetic biology.The P450 gene family is one of the largest gene families in living life,and play important roles in the plant developments and secondary metabolism.So far,there are only about 200 plant P450 genes with function resolved and annotated.Therefore,digging out the correct P450 gene from several candidate genes is the bottleneck of the synthetic pathway analysis and heterologous synthesis of important plant natural products.The gene function is usually determined by the tertiary structure of the expressed protein.However,as a kind of membrane protease,the P450 enzymes have disadvantages like low expression efficiency and difficulty in crystallization which limit the advancement of P450 structural analysis.At present,only 6 plant P450 enzymes have been crystallized in PDB database,the limited number of P450 structures severely limits our understanding of plant P450 structures and functions,making it impossible to mine out proper P450 enzymes for plant natural product synthesis.Therefore,the development of an efficient plant P450 structure prediction method and a sequence and structure database dedicated to plant cytochrome P450 enzymes have become the key to the development of synthetic biology-related to plant natural products.First,this study developed a structure prediction method specific to plant P450 enzymes.In this study,we constructed accurate conformations of heme-surrounding residues in homologous models by collating and analyzing the structural information of known P450s.A molecular dynamics simulation was performed to optimize the whole structure to guarantee the robustness.The cysteine-ligated heme group and the initial protein structure constitute a complete structural model of the plant P450 enzyme.Finally,we constructed a plant P450 enzyme structure database by organizing the structural models of the function resolved P450 genes.Secondly,this study developed a method exclusively for plant P450 enzyme function prediction.The RosettaLigand was used in this specific P450s system with the consideration of the CpdI and the required ligands.To analyze the relationship between plant P450 enzyme structure and function,we performed cross-docking to 10 substrates and 173 plant P450 enzymes based on the predicted functions and structures of plant P450 enzymes.In the docking experiments,a score ranking system was obtained through the analysis of the cross-docking results.The ligand complex structures and scores are arranged to construct a database of plant P450 complex with ligands.Finally,this study also built a web database dedicated to plant P450 enzymes.This website integrated the sequences,structures,ligands and annotations for plant P450 enzymes.The users can browse and download related information from this webserver,also can perform structure prediction and ligand docking experiments by uploading the required data.The prediction results of structure or function will be sent to the user by email.In conclusion,in this study,the construction of a database website can share the sequence,structure and docking information of plant P450 enzymes,and the designed structure and function prediction methods can promote the mining of plant P450 enzymes and the analysis of plant natural product biosynthetic pathways.