1- Department of Animal Science, Faculty of Animal Science and Fisheries, Sari Agricultural Sciences and Natural Resources University
Abstract: (168 Views)
Background: In recent years, the influenza virus has caused severe economic losses in the poultry industry, particularly in meat production, and remains one of the most important pathogens in the poultry industry worldwide despite vaccination programs. Avian influenza, especially H5N1, has become a major global health problem due to its potential as a zoonotic disease and its devastating impact on poultry populations. Identifying the molecular mechanisms of response to infections is crucial for the control, treatment, and prevention of epidemic diseases. A central goal of biological research is the coherent identification of all molecules in a living cell and their interactions. Co-expression of genes provides valuable information for understanding living systems because co-expressed genes often act in the same biological pathways or are linked through protein-protein interactions (PPI). The need for effective therapeutic strategies prompted us to investigate protein-protein interaction networks using systems biology approaches as a suitable method to discover candidate proteins and key biological pathways related to this disease. Since centrality indices measure the influence of nodes in an interaction network, we used betweenness, closeness, and degree centrality indices to analyze protein networks. This study aims to identify important biomarkers for the diagnosis, treatment, or control of the losses caused by avian influenza using biological network and signaling pathway analysis methods.
Methods: In the present study, the COXPRESdb gene co-expression database was searched using the keyword “response to the virus.” As a result, 21 genes were identified as involved in the biological processes of virus response in chickens. Of these 21 genes, seven were presented in two separate gene co-expression networks. Two gene co-expression networks were developed, and a total of 148 genes were extracted from the COXPRESdb database in 12 distinct gene clusters with the goal of identifying other gene co-expression components. To annotate the studied network, gene expression data from the GSE53932 series were extracted from the GEO database of NCBI. After statistical analysis, gene expression analysis was performed using GEO2R software to identify differentially expressed genes (P<0.05 and -2
Results: From the 21 virus-responsive genes in chickens, seven genes were presented in two separate co-expression networks. These two networks included three genes (TTR, ALB, and RBP4A) and four genes (SAMHD1, MX1, IRF7, and MYD88). In addition to the genes annotated during the virus response in chickens, a gene list of 148 possible virus-responsive genes with a co-expression score deviation (Z>3) was identified, as the co-expression exceeded the normal distribution of random co-expression. Expression data were extracted from the COXPRESdb gene co-expression database. After statistical analysis of the expression data, the initial interaction network for genes with significant differential expression was created with 61 nodes (proteins) and 306 edges using Cytoscape software. Two PPI networks were connected by three proteins: PLAC8L1, LBFABP, and IFI6. SERPINA10 and AHSG proteins, with node scores of 22, and PLG, with a node score of 21, were identified as hub proteins with the highest interaction rates in the entire network of differentially expressed proteins. Using the MCODE plugin, two high-density clusters were identified. These high-density regions may contain proteins that function as complexes within the cell. AMBP and DDX60 were identified as seed proteins in clusters 1 and 2, respectively. After adding expression data to network nodes in Cytoscape software, clustering and calculations with the jActiveModules plugin identified five active expression subnetworks. Two significant gene clusters (P<0.05) were identified using the CytoCluster 2.1.0 plugin (ClusterOne algorithm, P<0.05). The first significant cluster, with 25 nodes, includes proteins involved in the phenylalanine, tyrosine, and tryptophan biosynthesis pathway. The second significant cluster, with 20 nodes, includes proteins related to the influenza A signaling pathway, herpes simplex virus infection, and the RIG-I receptor signaling pathway, which are involved in the immune response to the virus. Centrality indices were calculated for active expression subnetworks obtained using the jActiveModules algorithm after adding expression data and for clusters obtained with the MCODE algorithm before adding expression data. Finally, the IFIT5, IFIH1, and RSAD2 proteins in the clusters obtained by the MCODE algorithm (highest cluster score) and the IFIH1 protein in the active expression subnetwork obtained by jActiveModules had the highest betweenness, closeness, and degree values. In KEGG gene enrichment analysis, these three proteins, along with STAT1, EIF2AK2, TRIM25, and PLG, were significantly (P<0.05) associated with the influenza A signaling pathway, herpes simplex virus 1 infection, and other related pathways.
Conclusion: This study identified the proteins IFIH1, IFIT5, and RSAD2 as having the highest centrality index values for active expression clusters and subnetworks, and STAT1, PLG, EIF2AK2, DHX58, and TRIM25 as key biomarkers for influenza A response. These proteins, along with the associated signaling pathways, including the RIG-I receptor signaling pathway, offer potential targets for the diagnosis, treatment, and control of avian influenza in poultry.
Methods: In the present study, the COXPRESdb gene co-expression database was searched using the keyword “response to the virus.” As a result, 21 genes were identified as involved in the biological processes of virus response in chickens. Of these 21 genes, seven were presented in two separate gene co-expression networks. Two gene co-expression networks were developed, and a total of 148 genes were extracted from the COXPRESdb database in 12 distinct gene clusters with the goal of identifying other gene co-expression components. To annotate the studied network, gene expression data from the GSE53932 series were extracted from the GEO database of NCBI. After statistical analysis, gene expression analysis was performed using GEO2R software to identify differentially expressed genes (P<0.05 and -2
Results: From the 21 virus-responsive genes in chickens, seven genes were presented in two separate co-expression networks. These two networks included three genes (TTR, ALB, and RBP4A) and four genes (SAMHD1, MX1, IRF7, and MYD88). In addition to the genes annotated during the virus response in chickens, a gene list of 148 possible virus-responsive genes with a co-expression score deviation (Z>3) was identified, as the co-expression exceeded the normal distribution of random co-expression. Expression data were extracted from the COXPRESdb gene co-expression database. After statistical analysis of the expression data, the initial interaction network for genes with significant differential expression was created with 61 nodes (proteins) and 306 edges using Cytoscape software. Two PPI networks were connected by three proteins: PLAC8L1, LBFABP, and IFI6. SERPINA10 and AHSG proteins, with node scores of 22, and PLG, with a node score of 21, were identified as hub proteins with the highest interaction rates in the entire network of differentially expressed proteins. Using the MCODE plugin, two high-density clusters were identified. These high-density regions may contain proteins that function as complexes within the cell. AMBP and DDX60 were identified as seed proteins in clusters 1 and 2, respectively. After adding expression data to network nodes in Cytoscape software, clustering and calculations with the jActiveModules plugin identified five active expression subnetworks. Two significant gene clusters (P<0.05) were identified using the CytoCluster 2.1.0 plugin (ClusterOne algorithm, P<0.05). The first significant cluster, with 25 nodes, includes proteins involved in the phenylalanine, tyrosine, and tryptophan biosynthesis pathway. The second significant cluster, with 20 nodes, includes proteins related to the influenza A signaling pathway, herpes simplex virus infection, and the RIG-I receptor signaling pathway, which are involved in the immune response to the virus. Centrality indices were calculated for active expression subnetworks obtained using the jActiveModules algorithm after adding expression data and for clusters obtained with the MCODE algorithm before adding expression data. Finally, the IFIT5, IFIH1, and RSAD2 proteins in the clusters obtained by the MCODE algorithm (highest cluster score) and the IFIH1 protein in the active expression subnetwork obtained by jActiveModules had the highest betweenness, closeness, and degree values. In KEGG gene enrichment analysis, these three proteins, along with STAT1, EIF2AK2, TRIM25, and PLG, were significantly (P<0.05) associated with the influenza A signaling pathway, herpes simplex virus 1 infection, and other related pathways.
Conclusion: This study identified the proteins IFIH1, IFIT5, and RSAD2 as having the highest centrality index values for active expression clusters and subnetworks, and STAT1, PLG, EIF2AK2, DHX58, and TRIM25 as key biomarkers for influenza A response. These proteins, along with the associated signaling pathways, including the RIG-I receptor signaling pathway, offer potential targets for the diagnosis, treatment, and control of avian influenza in poultry.
Keywords: centrality index, gene co-expression, influenza, KEGG, protein-protein interaction network
Type of Study: Research |
Subject:
ژنتیک و اصلاح نژاد دام
Received: 2024/09/22 | Accepted: 2024/11/23
Received: 2024/09/22 | Accepted: 2024/11/23
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |