Understanding the cell-dependent activation of ProTides and its implications in optimizing the nucleoside antiviral prodrug design for treating pulmonary infections

Abstract

<b>Abstract ID 16698</b> <b>Poster Board 185</b> Nucleoside (Nuc) analogs are an essential class of antivirals. Nuc analogs are often structurally modified into a phosphoramidate prodrug form (ProTide) to boost their intracellular activation efficiency: generating more active metabolite triphosphate Nuc (TP-Nuc). To date, three ProTides have been approved by the FDA: tenofovir alafenamide (TAF), sofosbuvir (SBV), and remdesivir (RDV). After entering cells, TAF, SBV, and RDV are hydrolyzed by the intracellular esterases carboxylesterase 1 (CES1) and cathepsin A (CatA) to release their ester moieties. The abundance of CES1 and CatA in the cells determines the amount of TP-Nuc generated. In the human lung, CatA is the major ProTide activating enzyme. We hypothesize that a higher susceptivity to CatA could improve the activation of a ProTide in the lung. In this study, we evaluated the activation profiles of TAF, SBV, RDV in various cell lines on a molar equivalent dose basis (except for RDV in Huh-7 cells due to cytotoxicity). The cell lines tested were those commonly used in SARS-CoV-2 studies, including the human lung cell lines A549-ACE2-TMPRss2, Calu-3, and normal bronchial epithelial cells BEAS-2B; the colon cell line Caco-2; the liver cell lines Huh-7 and HepG2; and the African green monkey kidney epithelial cells Vero E6. After 6, 12, and 24 hours of drug incubation, the intracellular concentrations of the ProTides and their active metabolites were determined by LC-MS/MS. The expression profiles of activating enzymes (e.g., CES1 and CatA) and transporters (e.g., OATP1B1 and P-gp) were retrieved from available datasets. The contribution of CatA and CES1 to the activation of TAF, SBV, and RDV in BEAS-2B cells was further evaluated using CES1 and CatA inhibitors. The results showed that the three ProTides were activated in a cell-dependent manner. Overall, the ProTides achieved higher levels of TP-Nuc in the liver cell lines (i.e., Huh-7 and HepG2) than the other cell lines tested, which was associated with the higher expressions of OATP1B1 and CES1 in Huh-7 and HepG2 cells. In contrast, the three ProTides generated significantly less TP-Nuc in Vero E6 cells compared to the other cell lines, which was related to the abundant P-gp expression in Vero E6 cells. Among the human lung cell lines, the ProTide activation was comparable between the A549-ACE2-TMPRss2 and BEAS-2B cells and higher than the Calu-3 cells. Similar to the type II pneumocytes, the major SARS-CoV-2 infection target, the human lung cell lines exhibit a very low abundance of CES1 but a considerable expression of CatA. Moreover, our CES1 and CatA inhibition study confirmed the predominant role of CatA in activating ProTides in BEAS-2B cells. Interestingly, in all the human lung cell lines, TAF generated 2-3- and 6-10- folds higher amounts of TP-Nuc than RDV and SBV, respectively. This observation is highly correlated with the susceptivity of the three ProTides to CatA. Moreover, the TAF prodrug had a higher accumulation in the lung cell lines than RDV and SBV, indicating that cell permeability is another important factor for intracellular prodrug activation. In conclusion, TAF, SBV, and RDV bare considerable cell-dependent activation properties, which are associated with the expression patterns of activating enzymes and transporters across the different cell lines. Higher susceptivity to CatA and greater cell membrane permeability will enhance the activation of a ProTide in the human lung.