The potential of remdesivir to affect function, metabolism and proliferation of cardiac and kidney cells in vitro.

Remdesivir is a prodrug of a nucleoside analog and the first antiviral therapeutic approved for coronavirus disease. Recent cardiac safety concerns and reports on remdesivir-related acute kidney injury call for a better characterization of remdesivir toxicity and understanding of the underlying mechanisms. Here, we performed an in vitro toxicity assessment of remdesivir around clinically relevant concentrations (Cmax 9 µM) using H9c2 rat cardiomyoblasts, neonatal mouse cardiomyocytes (NMCM), rat NRK-52E and human RPTEC/TERT1 cells as cell models for the assessment of cardiotoxicity or nephrotoxicity, respectively. Due to the known potential of nucleoside analogs for the induction of mitochondrial toxicity, we assessed mitochondrial function in response to remdesivir treatment, early proteomic changes in NMCM and RPTEC/TERT1 cells and the contractile function of NMCM. Short-term treatments (24 h) of H9c2 and NRK-52E cells with remdesivir adversely affected cell viability by inhibition of proliferation as determined by significantly decreased 3H-thymidine uptake. Mitochondrial toxicity of remdesivir (1.6-3.1 µM) in cardiac cells was evident by a significant decrease in oxygen consumption, a collapse of mitochondrial membrane potential and an increase in lactate secretion after a 24-48-h treatment. This was supported by early proteomic changes of respiratory chain proteins and intermediate filaments that are typically involved in mitochondrial reorganization. Functionally, an impedance-based analysis showed that remdesivir (6.25 µM) affected the beat rate and contractility of NMCM. In conclusion, we identified adverse effects of remdesivir in cardiac and kidney cells at clinically relevant concentrations, suggesting a careful evaluation of therapeutic use in patients at risk for cardiovascular or kidney disease.

Mono-ADP-ribosylation sites of human CD73 inhibit its adenosine-generating enzymatic activity.

CD73-derived adenosine plays a major role in damage-induced tissue responses by inhibiting inflammation. Damage-associated stimuli, such as hypoxia and mechanical stress, induce the cellular release of ATP and NAD+ and upregulate the expression of the nucleotide-degrading purinergic ectoenzyme cascade, including adenosine-generating CD73. Extracellular NAD+ also serves as substrate for mono-ADP-ribosylation of cell surface proteins, which in human cells is mediated by ecto-ADP-ribosyltransferase 1 (ARTC1). Here we explored, whether human CD73 enzymatic activity is regulated by mono-ADP-ribosylation, using recombinant human CD73 in the presence of ARTC1 with etheno-labelled NAD+ as substrate. Multi-colour immunoblotting with an anti-etheno-adenosine antibody showed ARTC1-mediated transfer of ADP-ribose together with the etheno label to CD73. HPLC analysis of the enzymatic activity of in vitro-ribosylated CD73 revealed strong inhibition of adenosine generation in comparison to non-ribosylated CD73. Mass spectrometry of in vitro-ribosylated CD73 identified six ribosylation sites. 3D model analysis indicated that three of them (R328, R354, R545) can interfere with CD73 enzymatic activity. Our study identifies human CD73 as target for ARTC1-mediated mono-ADP-ribosylation, which can profoundly modulate its adenosine-generating activity. Thus, in settings with enhanced release of NAD+ as substrate for ARTC1, assessment of CD73 protein expression in human tissues may not be predictive of adenosine formation resulting in anti-inflammatory activity.