同源蛋白質交互作用與複合體剖析蛋白質交互作用體行為

Abstract

透過蛋白質交互作用網路（protein interaction network）可以對於複雜的生物系統有更進一步的了解，例如探討不同生化途經之間的協同作用、蛋白質上特定殘基對功能的影響。因此，大量的交互作用資料庫（如：IntAct、DIP和BioGRID等）被建立來探討蛋白質交互作用網路。然而，這些資料庫中的蛋白質交互作用資料往往集中在少數的物種，而且也缺乏對於交互作用介面機制的解釋。 針對此議題，我們提出了蛋白質交互作用家族的觀念（包含蛋白質－蛋白質交互作用家族(protein-protein interaction family)和蛋白質複合體家族(protein complex family)）以協助多個物種中建立具有結構解析的蛋白質交互作用網路，並探討單一物種的蛋白質交互作用網路行為。蛋白質交互作用家族為一群有擁有保留性交互作用區塊、結合環境和相似生物途徑的蛋白質交互作用所組成。而透過“3D-domain interolog mapping＂與一個新的能量函式，我們將透過已知結構的模板（template），來探索多個物種間所有的同源蛋白質相互作用。此外，我們也在多個物種中找尋同源蛋白複合體，並描述了結合模型（例如在交互作用介面上的氫鍵和保留性氨基酸配對）、功能模塊、保留性相互作用區塊（interacting domain）和Gene Ontology。 透過由“3D-domain interolog mapping＂與蛋白質複合體家族，我們在人類、老鼠與斑馬魚中建立了具有結構解析的蛋白質交互作用網路。在每一個網路中，這些具有結構解析的蛋白質交互作用與Gene Ontology相似度有很高的一致性。此外，這些網路也都具有之前對於生化網路研究中所指出的拓譜特性(scale-free network)。而透過蛋白質交互作用家族我們也可指出網路中在跨物種間具有高度保留的蛋白質與交互作用，而這些蛋白質往往是生存必須基因(essential gene)或是跟疾病相關。更進一步的，這些跟疾病相關的基因突變往往位於蛋白質交互作用介面，並擔任重要的交互作用（例如氫鍵）。此外，對於單一蛋白質交互作用網路，我們提出了一個新的概念“MS-matrix＂來描述網路上重要的蛋白質以及模組化特性。基於上述這些研究，我們認為透過交互作用家族所構建的結構解析交互作用網路對於了解生化途徑的機制是很有幫助的。

Protein-protein interaction (PPI) networks provide key insights into complex biological systems, from how different processes communicate to the function of individual residues on a single protein. Therefore, several large network databases (e.g. IntAct, DIP, and BioGRID) record hundreds of thousands of physical and genetic interactions from a wide variety of organisms have been purposed. However, these PPI databases are dominated by few species and usually could not provide the binding mechanisms. Therefore, constructing the structure resolved PPI networks across multiple organisms should provide a great value for investigating the behavior of PPI network. To address the issues, we proposed the concepts of protein interaction family (i.e. protein-protein interaction family and protein complex family) to construct a structure resolved PPI networks and study the behaviors of a specific PPI network. The protein interaction family is a group of protein interactions (PPI or protein complex) which share the consensus interacting domain, binding environment, and have similar biological processes. According to the concept "3D-domain interolog mapping" with a scoring system, we are able to explore all homologous protein-protein interaction pairs (protein-protein interaction family) between two homolog families, derived from a known 3D-structure dimmer (template), across multiple species. Then, we also identify the homologous protein complexes with the binding models (e.g. hydrogen bonds and conserved amino acids in the interfaces), functional modules, and the conserved interacting domains and Gene Ontology annotations in multiple organisms. Based on the PPIs derived from "3D-domain interolog mapping" and "protein complex family", we are able to construct structure resolved PPI networks in multiple organisms (e.g. Homo sapiens, Mus musculus, and Danio rerio). In each network, the PPIs with residue-based binding models have a highly agreement in Gene Ontology similarities. Furthermore, the architecture (i.e. scale-free network properties) of these networks is consistent with some cellular networks of previous studies. In addition, the consensus proteins and PPIs derived form on our method are highly related to the essential genes and disease related proteins recorded in OMIM. We also indicate that the disease related mutations are more enrichment on the interacting residues, especially on the hydrogen bond residues. In addition, for a given PPI network, we also provided a new characterization (named MS-matrix) to describe the modularity and relative importance of proteins. We believe that structure resolved PPI networks derived from the PPI family would provide the insight for understanding the mechanism of biological processes within a given PPI network.