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

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

We manually inspected the top candidates and selected ten representative candidates according to their reversal scores ( Figure 2D ), clinical applicability, MoA, administration routes and safety (Table S3 ). Finally, we evaluated their cytotoxicity and ability of the prevention of cytopathic effect (CPE) in the Vero E6 cell line ( Table 1) . Instead of measuring EC50, we used more stringent criteria, which required a 4-day complete prevention of CPE observed under the microcopy. Six of the proposed drugs showed less than 10 µM minimum concentration that could completely prevent CPE. However, five of them presented unfavorable cytotoxicity at their effective concentration. In the following confirmatory experiment, we examined the efficacy and toxicity for the four most effective drugs (CPE preventing concentration <= 1 µM), bortezomib, methotrexate, nvp-bez235 and alvocidib as well as chloroquine, a drug being used as the first-line treatment for SARS-CoV-2. None of the predicted drugs could completely prevent CPE, though chloroquine could at the dose of 15 µM. Figure 1E ), the signature from the comparison between 24 h and 0h could not capture the biology of viral infection, thus failed to recover known drugs. Those findings also reiterate the challenge of choosing appropriate experimental models and the right time points for drug treatment in COVID-19 25 .