Mitochondria and cytokine storm in COVID-19
As mitochondria are shown to be one of the critical components in eliciting the innate immune response, specifically in response to viruses, the SARS-CoV-2 mediated immune response could involve mitochondria. In support of this hypothesis, a master regulator analysis of combined SARS-CoV-2 specific interactome with MERS and SARS- CoV transcriptome using the human lung RNA-sequence data set validates the interaction with ACE-2 and TMPRSS, in addition to certain mitochondrial proteins (Guzzi, Mercatelli, Ceraolo & Giorgi, 2020). These mitochondrial proteins are MCL-1, a regulator of apoptosis, as well as the Complex I subunit NDUFA10 network, which is downregulated (Guzzi, Mercatelli, Ceraolo & Giorgi, 2020). However, the proposed mechanism is that the virus strategizes to down-regulate mitochondrial function and uses host cells for replication. This is not executed by mitochondrial-dependent elimination, controlling host cell metabolism and apoptosis. At normal physiological conditions, the MAV protein present on the mitochondrion’s outer membrane interacts with mitofusin-2 (Mfn2), which is essential for mitochondrial fusion. During viral entry, mitochondrial associated membranes at the ER tether mitochondria using Mfn2 and RIG-1. Using other protein recruits, the virus binds to MAVS, which then induces phosphorylation and nuclear translocation of IRF3, resulting in the production of cytokines and interferon I/III through activation of NFKba and IRFs (3/7), respectively (Lin, Heylbroeck, Pitha & Hiscott, 1998). As in vitro data showed that MAVS is required for the induction of IFN production by activating NFKb and IRF3, recent evidence also shows that mice lacking MAVS failed to induce IFN production in response to viral infection. This suggests the involvement of mitochondria in the viral-mediated immune response (Sun et al., 2006) (Figure 3) .
A clue from a previous study carried out in peripheral blood mononuclear cells (PBMCs) infected with SARS-CoV (Li et al., 2003; Li et al., 2004) showed that upregulation of genes encoding mtDNA, as well as genes involved in oxidative stress, heat shock, and transcription, coincides with cytokine elevation, compared to PBMCs from control samples (Shao et al., 2006). The mtDNA genes included 16S rRNA (a ribosomal subunit), NADH dehydrogenase subunit 1 (ND1), and cytochrome c oxidase subunit I (COX1), whereas the genes involved in oxidative stress are peroxiredoxin 1 (PRDX1) and ferritin heavy polypeptide 1 (FTH1), heat shock response are DnaJ (Hsp40), homolog subfamily B member 1 (DNAJB1), and cytokine IL-1B as well (Shao et al., 2006). These SARS-CoV-infected patients are also shown to have a significantly increased number of mitochondria in their PBMCs compared to PBMCs from control subjects. This may be one of the reasons why more mtDNA gene expression is observed. Interestingly, the electron microscopic structure of PBMCs from SARS-CoV infected patients showed increased lysosome-like granules, which was not seen in the control. This could possibly be the activation of mitochondria in PBMCs, causing an increased immune response and cytokine storm. We hypothesize that this could be more cell-specific as it depends on whether the cell is more immunogenic or not. Likewise, a recent study has shown a similar observation of SARS-CoV-2 infection in PBMCs and lymphoma along with the cytokine storm, but the involvement of mitochondria, which is likely, was not investigated in this study (Liu et al., 2020). This study further showed that cytokines IL-6, IL-10, IL-2 and IFN-γ are significantly increased in severe COVID-19 cases than mild cases, displaying that these levels are associated with the disease severity. Therefore, a possible cytokine storm induced by mitochondria in COVID-19 could be heterogeneous among the infected population. It could also be influenced by preexisting metabolic dysfunction, which further determines the intensity of the cytokine storm and whether it eliminates the virus or causes multi-organ failure.