Cross-species gene expression analysis reveals gene modules implicated in human osteosarcoma

Abstract

Background: Osteosarcoma (OS) is one of the malignant bone tumors occurring in both human and canine, and in both of them, it is characterized by a high rate of metastasis and poor prognosis. Cross-species analysis reveals previously neglected molecular or signaling pathways involved in the progression of diseases, and dogs are genetically comparable to humans and live in similar environments. Therefore, the aim of this study was to find out OS hub genes through a cross-species analysis. Materials and Methods: All the human and canine OS gene expression data obtained by the Affymetrix platform were collected. After quality assessment and normalization, co-expression network was performed using weighted gene co-expression network analysis (WGCNA). Species-specific modules and consensus modules were identified. Protein-protein interaction (PPI) networks analysis was performed based on consensus gene modules. Then, consensus modules were functionally annotated and correlated with clinical traits. Hub nodes were identified by a subnetwork analysis of PPI network and WGCNA module membership. Modules of interest and hub nodes were validated in an external data set. Results: Three modules for the human network, seven modules for the canine network, and four consensus modules were identified. The consensus module 3 (C3) showed a significant correlation with the metastatic status in the training data set and a significant correlation with metastasis-free survival in the external data set. Cluster of differentiation 86 (CD86) was identified as the hub gene of C3, showing a significant correlation with metastasis-free survival. Conclusion: Genes in C3 play an important role in OS metastasis, whereas CD86 might be a potential molecular biomarker for OS metastasis.