Outcome-Dependent Global Similarity Analysis of Imbalanced Core Signaling Pathways in Ischemic Mouse Hippocampus

Abstract

Analysis of the diverse interactions of multiple signaling pathways is an emerging challenge in the era of networking pharmacology. To reveal imbalanced signaling pathways and pharmacological mechanisms involved in ischemic process, we designed systemic experiments from top-down to bottom-up for investigating the variations of multiple pathways in mouse hippocampal cells. A total of 711 focal cerebral ischemia-reperfused animals (504 mice and 207 rats), induced by occlusion of the middle cerebral artery, were obtained to conduct 4 experiments. The mice were used to analyze the pharmacological effects of four single compounds, baicalin (BA), jasminoidin (JA), ursodeoxycholic acid (UA) and concha margaritifera (CM) and two combination therapies (BA+JA, and JA+UA). Moreover, the mouse models were also used for microarray and western blotting test. The rat models were used for infarction volume test, magnetic resonance imaging (MRI) test and neurological score analysis to validate the pharmacological effects in another species. The results of western blotting confirmed that the expression of the key proteins involved in the ischemiaactivated Wnt and nuclear factor κB (NF-κB) pathway was markedly altered. In addition, based on the screened gene expression profiles of ischemia hippocampus, a variety of altered genes contributed to the 9 stroke-related pathways based on literature review [Wnt, extracellular signal-regulated kinase (ERK), janus kinase (JAK), mitogen-activated protein kinase (MAPK), gonadotropin-releasing hormone (GnRH), calcium/calmodulin-dependent protein kinase (CaMK), vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), and platelet-derived growth factor (PDGF)] in different groups. Thus, we believed that the 9 signaling pathways were significantly imbalanced in different groups. However, analysis of overlapping genes was insufficient to reveal the expression profiles of imbalanced pathways between or within various conditions treated with different compounds or compound mixtures. Therefore, global similarity index (GSI) is introduced to quantify the genotypic outcomes of gene expression profiles. Independent experiments in mice on the effects of infarction volume, neurologic deficit score and the results of MRI in rats showed that GSI was suitable for the spectral measurement of imbalance in those 9 biochemical pathways with a predictive accuracy of 81.0% as assessed by leave-one-out cross-validation.