Spike protein of SARS-CoV-2 kills cells by targeting fumarase
Diabetics tend to have more severe COVID-19 disease and COVID-19 infection increases the risk of non-diabetic patients developing new-onset hyperglycaemia and diabetes. SARS-CoV-2 clearly reprogrammes glucose metabolism in host cells– increasing aerobic glycolysis but inhibiting oxidative phosphorylation– possibly to favour viral replication through the Warburg effect. However, it is uncertain how the virus disrupts mitochondrial respiration, how this increases pathogenicity in diabetics, and why infection predisposes to diabetes. Saccharomyces cerevisiae yeast and human cells have functionally similar glucose metabolic enzymes. Moreover, the SARS-CoV-2 viral surface protein, known as Spike, is cytotoxic to both yeast and human cells. Hence, genetically-modified yeast expressing Spike were used to model human cell responses to metabolic and cytotoxic effects of Spike. The first project finding was that Spike interacts with the Krebs cycle enzyme fumarase. Spike also inhibited the enzymatic activity of fumarase, consistent with the project hypothesis that Spike targets fumarase to disrupt oxidative metabolism in host cells. Fumarase-deficient yeast had poorer survival than wild-type yeast in the presence of either S1 or S2 domains of Spike. Fumarase inhibition is likely to be linked to cell death because Krebs cycle inhibition limits ATP generation through oxidative phosphorylation– this increases mitochondrial reactive oxygen species and oxidative stress, causing cell death. Such oxidative stress is also a central mechanism for pathology in diabetes. Targeting and inhibition of fumarase by Spike is a novel finding that suggests a possible mechanism for clinical observations of higher COVID-associated morbidity in diabetics. Mitochondrial oxidative stress and dysfunctional sugar metabolism, associated with Spike-mediated inhibition of aerobic respiration, can explain the prevalence of post-COVID diabetes. Spike-interacting domains of fumarase were identified through alanine-scanning mutagenesis of fumarase, and testing mutants for loss of interactions with Spike. A catalytically active fumarase mutant strain which interacted poorly with Spike was more resistant to both enzymatic inhibition and cytotoxicity caused by Spike, than strains expressing wild-type fumarase. These data support a novel therapeutic strategy of designing molecules to block Spike-fumarase interactions, aided by knowledge of the interacting domains identified. Such therapy could potentially reduce Spike-induced cell death and mitigate ‘long COVID’ symptoms and the diabetes risk.