Reversal of Infected Host Gene Expression Identifies Repurposed Drug Candidates for COVID-19

Xing, Jing; Shankar, Rama; Drelich, Aleksandra; Paithankar, Shreya; Chekalin, Eugene; Dexheimer, Thomas; Rajasekaran, Surender; Tseng, Chien-Te Kent; Chen, Bin

Repurposing existing drugs is a timely option to cope with COVID-19. We predicted therapeutic candidates that could reverse the gene expression of coronavirus-infected host cells. Thirteen expression signatures computed from various experimental conditions and preclinical models could be reversed by those compounds known to be effective against SARS- or MERS-CoV, as well as the drug candidates recently shown to be effective against SARS-CoV-2. We selected ten novel candidates to further evaluate their in vitro efficacy against SARS-CoV-2 infection. Four compounds bortezomib, dactolisib, alvocidib and methotrexate inhibited the formation of virus infection-induced cytopathic effect in Vero E6 cells at 1 μM, yet such a concentration seems toxic to the cells as well. While the evaluation in other permissive cells and the prediction of toxicity are needed to optimize and minimize their antiviral activity and cytotoxicity, respectively, this computational approach has the potential to rapidly and rationally identify drug candidates against COVID-19.