Drug Repurposing Using Gene Co-Expression and Module Preservation Analysis in Acute Respiratory Distress Syndrome (ARDS), Systemic Inflammatory Response Syndrome (SIRS), Sepsis, and COVID-19

Simple Summary The COVID-19 pandemic has caused a global standstill. The advent of vaccines has significantly hindered viral transmission, paving the way back to normalcy. However, post-COVID-19 treatment is still of utmost concern as people continue to suffer from cytokine storms or excessive inflammation, leading to death. In this paper, we utilize computational biology or bioinformatics techniques to identify possible drugs that may be repurposed to treat COVID-19 symptoms. We do so by identifying the genetic similarities and differences in three symptoms strongly associated with COVID-19: acute respiratory distress syndrome (ARDS), systemic inflammatory response syndrome (SIRS), and sepsis. As these diseases are related by the fact that they affect the immune system, similarities in their mechanisms may be exploited to find potential therapeutics. Using such an approach, we identify seven possible repurposable drugs that have been previously shown to treat immune-system related ailments. Our findings may potentially aid in driving drug design and discovery towards more effective post-COVID-19 therapeutics, and show how computational techniques can hasten the otherwise lengthy process of bringing drugs into clinical use. SARS-CoV-2 infections are highly correlated with the overexpression of pro-inflammatory cytokines in what is known as a cytokine storm, leading to high fatality rates. Such infections are accompanied by SIRS, ARDS, and sepsis, suggesting a potential link between the three phenotypes. Currently, little is known about the transcriptional similarity between these conditions. Herein, weighted gene co-expression network analysis (WGCNA) clustering was applied to RNA-seq datasets (GSE147902, GSE66890, GSE74224, GSE177477) to identify modules of highly co-expressed and correlated genes, cross referenced with dataset GSE160163, across the samples. To assess the transcriptome similarities between the conditions, module preservation analysis was performed and functional enrichment was analyzed in DAVID webserver. The hub genes of significantly preserved modules were identified, classified into upregulated or downregulated, and used to screen candidate drugs using Connectivity Map (CMap) to identify repurposed drugs. Results show that several immune pathways (chemokine signaling, NOD-like signaling, and Th1 and Th2 cell differentiation) are conserved across the four diseases. Hub genes screened using intramodular connectivity show significant relevance with the pathogenesis of cytokine storms. Transcriptomic-driven drug repurposing identified seven candidate drugs (SB-202190, eicosatetraenoic-acid, loratadine, TPCA-1, pinocembrin, mepacrine, and CAY-10470) that targeted several immune-related processes. These identified drugs warrant further study into their efficacy for treating cytokine storms, and in vitro and in vivo experiments are recommended to confirm the findings of this study.

Automated summary

This paper utilizes computational biology techniques to identify possible repurposed drugs for post-COVID-19 treatment. By analyzing the genetic similarities and differences in symptoms associated with COVID-19, the study identifies seven candidate drugs that have been shown to treat immune-system related ailments. The findings have the potential to drive drug design and discovery for more effective post-COVID-19 therapeutics, and demonstrate the importance of computational techniques in speeding up the drug development process.