The prediction of protein-protein interaction of A. thaliana and X. campestris pv. campestris based on protein domain and interolog approaches

摘要

Plant pathogenic bacteria cause many serious disease symptoms by injecting a variety of effector proteins to reprogram the host defense mechanism. Protein-protein interaction is an essential process playing a crucial role in host-pathogen interactions and pathogenicity. However few have been known for how pathogen bacteria interact with their hosts. In this study, the interactions of A. thaliana proteins and Xanthomonas campestris pv.campestris (Xcc) pathogen bacteria proteins are identified by two different approaches: the domain-based approach which infers interspecies protein-protein interactions by known domain-domain interactions recorded by various databases; and the interolog approach that identifies protein-protein interactions based on homologous pairs of protein interactions across different organisms. The results from these two methods are integrated and the dense protein interaction regions are specified by clique percolation analysis. In particular, the plant resistance genes (PRG) information and the bacterial effector proteins are studied to provide new insights into the molecular mechanism of plant immunity systems against bacteria. From our findings, we demonstrate that a pathogen employs five strategies to reprogram the host defense mechanism. First, ome Xcc proteins tend to interact with A. thaliana’s hub proteins or the PRGs. Second, Xcc proteins tend to interact with many A. thaliana proteins indicating that a pathogen mutates its genes to infect the host. Third, some Xcc proteins target a group of A. thaliana proteins that are involved in responding to cadmium ions, a significant plant biological process against pathogen. Fourth, many Xcc proteins target a few A. thaliana proteins which are involved in the plant-pathogen interaction pathways. Finally, the pathogen may make use of a type III effector protein to reprogram the host protein-protein interactions. Host-pathogen interactions remain a largely unexplored area in computational biology. The present work may provide some key information useful for revealing the biological mechanism of a plant’s immune system against bacteria. A web-based interface has been established (http://ppi.bioinfo.asia.edu.tw/AtXccPPI) where investigators can pose queries.