Topological Analysis And Experimental Study Of Protein-protein Interaction Network Associated With Esophageal Squamous Cell Carcinomas

Objective: Esophageal squamous cell carcinomas as other cancers is a complex disease. In recent years, complex system theory is employed to biomedical area, especially to explore the mechanism such as cancer. Protein-protein interaction network is a important biological molecule network. The studies in model organism show that hub proteins in PPI network play special biological role. The cancer proteins tend to be hub node in human disease network. This study aim to: 1) construction and analysis the PPI network related to ESCC; 2) investigation the expression pattern of hub genes in ESCC tumor tissue and paired normal tissue; 3) study the effect on expression of hub gene in ECA109 cell dealing with 5-aza-DC, TSA and 5-aza-DC combined with TSA, study the effect on expression of hub gene in ECA109 after over expression Lrig1. Methods: 1) Firstly, the genes or proteins related to ESCC are extracted from Pubmed database and Chinese biomedical database and they are divided into two groups according to their source. The genes from Chinese biomedical database are restricted to the Kazak people in Xinjiang area. Secondly, the networks at five grades are constructed according to two fundamental topological parameters degree and BC value by complex network tool called Pajek and Cytoscape. Lastly, the relationship between topological and biomedical parameters is compared. 2) The total RNA is extracted from tumor tissue and paired normal tissue of 48 Han people and 40 Kazak people with ESCC. Then, the expression of hub genes at mRAN level is detected by semi quantitative RT-PCR, the differential expression rate and difference of expression quantity in tumor tissue and paired normal tissue are compared. Then the relationship between positive expression rate of tumor tissue and clinical factor is analyzed. 3) culture ECA109 cell line treat with 5-aza-DC and TSA, Over expression Lrig1 in ECA109 cell line then expression of hub genes aredetected. Results: 1) there is difference in number of nodes in the largest connected sub-graph(214) and number of sub-graph(83) between core network CNCA and random network. 2) There are 27 pathways in KEGG database with statistical relevant to proteins in extended network ECNCA and the pathways called paths in cancer has the lowest p-value. Analysis of five grades backbone network show that the nodes with higher BC value have more chance to be the members of the pathways called paths in cancer and more chance to be genes associated with ESCC. The overlap rate of networks derived from two groups initial genes are higher than it of two group initial genes. 2) The positive expression rate of the most genes in Han and Kazak people is higher than 50%. There are 3 genens including AR, EGFR and SPP1 with statistical difference in positive expression rate between tumor and paired normal tissue of Han people with ESCC and 2 genes including AR and EGFR within Kazak people. There are 11 genes with up-regulation in tumor tissue including EP300, VIM, KDM5 B, MDM2, SFN, AR, CUL1, EGFR, PTEN, SHC and SPP1, 2 genes with down-regualtion including TP53 and Sumo2 within Han people. There are 7 genes with up-regulation in tumor tissue including EGFR, NEDD8, VCP, UBD, AR, MMP2 and UBE2 I, 2 genes with down-regualtion including BCL2 and CUL3 Sumo2 within Kazak people. After Wilcoxon rank sum tes, the 5 genes including AR, EGFR, MMP2, UBD and VIM have statistical difference at protein level. 3) There are 5 genes with down-regultion including AR, ARRB2, CAND1, MAPK14 and MYC, 3 genes with up-regulation including UBD, UBE2 Iand PTEN in 5-aza-dc group. There are 6 genes with down-regultion including ARRB2, CAND1, HSP90AA1, MAPK14, UBD and MYC, 2 genes with up-regulation including AR and UBE2 I in TSA group. There 4 genes with interactive effect including AR, MMP2, HSP90AA1, MGMT and SUMO2 in 5-aza-d C combined with TSA group. There are 5 genes with down-regultion including AR, MMP2, HSP90AA1, MGMT and SUMO2 Lrig1 over-expression group. Conclusion: 1) the result from complex network analysis suggests that the hub genes are related to tumorgenesis and progress of ESCC. 2) The most of hub gene express in tumor and normal tissue, which suggests that these genes have important function. Thus, the aberrant expression of them will occur in stages of tumorgenesis and progress of ESCC. In this study, the several genes with aberrant expression distribute in many pathways associated with cancer. 3) After culturing with 5-aza-DC, the low expression of genes AR, ARRB2, CAND1, MAPK14 and MYC in ECA109 cell line maybe relate to the transcription activity restoring of some tumor suppressor genes and high expression of genes UBD, UBE2 I and PTEN maybe relate to the change of methylation status of theirpromoter. With the over expression of Lrig1 in CA109 cell line, the possible mechanism of low expression of genes AR, MMP2, HSP90AA1, MGMT and SUMO2 is Lrig1 suppress the EGFR pathway.