In addition to the cellular necrosis COVID-19 causes, releasing mitochondrial DNA into the bloodstream, it also has the ability to mutate mitochondrial DNA.

This is done via two separate mechanisms. One mechanism of mitochondrial disruption employed by SARS-CoV-2 involves ferritin as evidenced by the high levels of ferritin in those with severe outcomes. A normally functioning mitochondrion uses this iron to make heme, create iron-sulfur clusters, and store as mitochondrial ferritin, but an overload of iron can lead to oxidative stress and impair mitochondrial function by reducing oxygen consumption by the mitochondria.

The other mechanism is that the virus directly localizes to the mitochondria. A study found 5′ and 3′ untranslated regions on SARS-CoV-2 unique for mitochondrial localization, although further work needs to be done on this finding. When comparing SARS-CoV-1 and SARS-CoV-2, both are found to contain open reading frame ORF-9b, ORF-7a, and ORF-8b, which localize to the mitochondria, in the case of SARS-CoV-1, to alter MAVS function and mitochondrial function.

If mitochondrial DNA maintenance is impaired, it activates Astrocytes, leading to spongiotic encephalopathy. This may be the cause of the spongiosis we are observing in COVID autopsies, and not prions.

Mice with the inability to maintain mitochondrial DNA had wide-spread chronic microglial and astrocyte activation accompanied by prominent myelin disorganization. Eventually, the astrocytes cause the brain to be “eaten” creating vacuoles (holes).

This may be the mechanism by which SARS-CoV-2 causes early-onset, severe neurological disease.