Drug-induced phospholipidosis is not correlated with the inhibition of SARS-CoV-2 - inhibition of SARS-CoV-2 is cell line-specific

Recently, Tummino et al. reported that 34 compounds, including Chloroquine and Fluoxetine, inhibit SARS-CoV-2 replication by inducing phospholipidosis, although Chloroquine failed to suppress viral replication in Calu-3 cells and patients. In contrast, Fluoxetine represses viral replication in human precision-cut lung slices (PCLS) and Calu-3 cells. Thus, it is unlikely that these compounds have similar mechanisms of action. Here, we analysed a subset of these compounds in the viral replication and phospholipidosis assays using the Calu-3 cells and PCLS as the patient-near system. Trimipramine and Chloroquine induced phospholipidosis but failed to inhibit SARS-CoV-2 replication in Calu-3 cells, which contradicts the reported findings and the proposed mechanism. Fluoxetine, only slightly induced phospholipidosis in Calu-3 cells but reduced viral replication by 2.7 orders of magnitude. Tilorone suppressed viral replication by 1.9 orders of magnitude in Calu-3 cells without causing phospholipidosis. Thus, induction of phospholipidosis is not correlated with the inhibition of SARS-CoV-2, and the compounds act via other mechanisms. However, we show that compounds, such as Amiodarone, Tamoxifen and Tilorone, with antiviral activity on Calu-3 cells, also inhibited viral replication in human PCLS. Our results indicate that antiviral assays against SARS-CoV-2 are cell-line specific. Data from Vero E6 can lead to non-transferable results, underlining the importance of an appropriate cell system for analysing antiviral compounds against SARS-CoV-2. We observed a correlation between the active compounds in Calu-3 cells and PCLS.