A Multidimensional Comparison Of Pharmacological Mechanisms Of Different Compound Treatments On Cerebral Ischemia Models

BackgroundCerebral ischemia is a process related to a series of biochemical and molecular mechanisms involving changes of multiple genes, pathways and protein targets. It has been demonstrated that a number of genes, proteins and signaling pathways such as MAPK, PI3K/Akt, NF-κB, JAK-STAT and Wnt/beta-catenin have an impact on cerebral ischemia-reperfusion injury. Network pharmacology suggests that, targeting whole networks is maybe the key behavior of pharmacological intervention. Deletion of individual nodes has little effect on disease networks, modulating multiple proteins may be required to perturb robust phenotypes. Many drugs hit multiple targets, which exist within a complex network. The effects of the drug, both therapeutic actions and adverse events, are a result of perturbation of the complex network. However, most previous studies were limited to a single gene or a single pathway. Therefore, it is extremely important and necessary to identify functional modules in disease-and drug-associated networks accurately. Otherwise, module identification is also an essential step in the framework of Modular Pharmacology (MP) proposed recently. Currently, researchers have proposed a lot of methods or algorithms for module identification, however, there are still many problems for application, which due to lack of systematic and reasonable classification criteria. Therefore, we collected a large number of literatures, and preliminary established reasonable classification criteria to classify existing methods. Notably, we do not know the number of functional modules contained in a given network in advance, therefore, when the decomposition of the same network using different methods, it often leads to different results. Then, how can we determine the results of module identification? Previous studies have shown that BA, CA and JA were effective in reducing the ischemic infarct volume compared with the vehicle group (P<0.05). Our study intends to further reveal the pharmacological mechanisms of different compound treatments on cerebral ischemia.ObjectiveThe aim of our study was to reveal the overlapping and diverse pharmacological protective mechanisms of different compound treatments on cerebral ischemia at the signaling pathway, functional module and network level.MethodsFirstly, in order to find signaling pathways and networks from gene expression profiles of hippocampus of ischemic mice treated with baicalin (BA), cholic acid(CA) and jasminoidin (JA), a microarray comprising16,463genes, FDA Arraytrack software and Ingenuity Pathway Analysis, were employed. Secondly, drug-target networks were constructed by integrating gene expression data and protein interaction data, and then functional modules were identified using Affinity propagation (AP), Markov Cluster algorithm (MCL) and Molecular Complex Detection (MCODE), respectively. Then, we could determine the relative optimal module identification result based on the minimum entropy criterion. GO functional enrichment analysis was performed with DAVID6.7software. Finally, both module (Mrml-Gukl-Hrsp12) and modules (Fos-Cebpg-Atf2) could be validated using RT-PCR and Western blot, respectively.Results1A total of5,8,11,9IPA networks were found in vehicle (vs sham), BA, CA and JA (vs vehicle), respectively (P<0.05).2A total of78,70,77,74IPA network functions were found in vehicle (vs sham), BA, CA and JA (vs vehicle), respectively (P<0.05). The total of60overlapping functions could be approximately divided into diseases and disorders, molecular and cellular functions, physiological system development and function as categories with ratio of1:1:1.3A total of6,7,40,16canonical pathways were found in vehicle ((vs sham), BA, CA and JA (vs vehicle), respectively (P<0.05).4and7overlapping pathways were shared between BA and CA, CA and JA, respectively. BA, CA and JA affected some growth factor signaling pathways, including EGF signaling, PDGF signaling, IGF-1signaling. BA, CA and JA all acted on Ca2+-dependent signaling cascades in diverse links. BA intervened in arachidonic acid metabolism. JA might decrease oxidative damage via NRF2-mediated oxidative stress response. CA mainly affected eicosanoid signaling, cyclin-dependent kinase5(CDK5) signaling, nuclear factor kappa B (NF-κB) activation, IL-2Signaling as well as TNF-a receptor signaling.4Topological attributes of global networks among vehicle, BA, CA and JA groups were similar to each other, except that there was small difference in network size.5After the optimization of minimum entropy,50,49,41, and42modules (nodes≥3) were identified from related target networks by MCODE analysis.6GO functional enrichment analysis revealed218significantly enriched biological processes.185,153,147,192GO biological processes were enriched in vehicle, BA, CA and JA groups, respectively. BA, CA and JA were enriched for neurotransmitter-related functions. BA was enriched for L-ascorbic acid transport, neurogenesis, etc. CA was enriched for cation transport.7There were no significant linear correlations between the number of GO biological processes and the number of nodes or the number of edges in modules. There was significant linear correlation between the degree of each node and the number of GO biological processes in the same module, and the correlation coefficient value was 0.18. There were different degrees of overlap between modules among groups8The results of RT-PCR showed that, compared with the sham group, the mRNA level of Mrml was significantly down-regulated in vehicle group (P<0.01). Compared with the vehicle group, Mrml mRNA was significantly up-regulated by BA, while down-regulated by CA (P<0.01). The mRNA levels of Gukl and Hrsp12were not significantly different between groups (P>0.05). The results of Western blot showed that, compared with the sham group, the protein expression levels of Atf2, Fos, Cebpg were significantly up-regulated in vehicle group (P<0.01). Compared with the vehicle group, both Fos and Cebpg protein expression levels were significantly down-regulated by JA, BA and CA (P<0.05). There was no significant difference on Atf2protein expression among four groups (P>0.05).Conclusion1There were both overlapping and diverse pharmacological mechanisms of different compounds in the intervention of cerebral ischemia-reperfusion injury. BA, CA and JA exerted therapeutic effects in cerebral ischemia through multiple targets, multiple pathways, and multiple functions.2Analysis of signaling pathways, functional modules and networks may be complementary paradigms for revealing potential pharmacological mechanisms. At signaling pathway level, BA, CA and JA all acted on Ca2+-dependent signaling cascades in diverse links. At functional module level, BA was involved in antioxidation and neurogenesis, CA focused on regulating cation transport. BA, CA and JA affected neurotransmitter-related functions.3The method of modular analysis is suitable for multi-targeted drugs treatment on complex diseases, and provides a new method for pharmacological study in future.