Abstract
Over the past several decades, multiple coronaviruses (CoVs) have emerged as highly infectious, lethal viruses in humans, most notably in the pandemic outbreak of COVID-19, the disease caused by SARS-CoV-2. To date, there are no known therapeutic or preventative agents to target CoVs. Though age and comorbidities severely increase case fatality rates, the host factors that influence resistance or susceptibility to infection with highly pathogenic human CoVs are unknown. Innate immune responses to CoVs are initiated by recognition of double-stranded (ds) RNA and induction of interferon, which turns on a gene expression program that inhibits viral replication. SARS-CoV-2 conserves an ADP-ribosylhydrolase domain previously shown to counteract innate immunity to both mouse hepatitis virus (MHV), a model CoV, and SARS-CoV. Here we show that SARS-CoV-2 infection of cell lines, infected ferrets, and a deceased patient’s lung consistently and strikingly dysregulates the nicotinamide adenine dinucleotide (NAD+) gene set with respect to NAD+ synthesis and utilization. SARS-CoV-2 induces a set of poly(ADP-ribose) polymerase (PARP) family members; these PARPs include enzymes required for the innate immune response to MHV. Further, we show that MHV infection induces an attack on host cell nicotinamide adenine dinucleotide (NAD+) and nicotinamide adenine dinucleotide phosphate (NADP+). The data indicate that overexpression of a virally induced PARP, PARP10, is sufficient to depress host cell NAD metabolism and that NAD+ boosting strategies differ in their efficacy to restore PARP10 function. Gene expression and pharmacological data suggest that boosting NAD+ through the nicotinamide and nicotinamide riboside kinase pathways may restore antiviral PARP functions to support innate immunity to SARS-CoV-2, whereas PARP1,2 inhibition may be less likely to restore antiviral PARP functions.
Competing Interest Statement
Charles Brenner is chief scientific adviser of ChromaDex and owns shares of ChromaDex stock. Other others declare no competing interests.
Footnotes
we corrected an error in figure 2 that was caused by our plotting the wrong time point of MHV with respect to the 12 hour control sample. Figure 2 and the supplementary information files have been corrected.
