SARS-CoV-2 cell entry leads to IFN-I response, antagonized

Upon entry of a virus into the host cell (KE1738), the virus is unpackaged from the structural nucleocapsid (N), envelope (E), and membrane (M) proteins. The viral RNA is detected by Pattern Recognition Receptor (PRR) proteins including RIG-I and MDA5 but the M proteins can interact with these PRRs directly, and block this initial host reaction (Fu et al., 2021). The viral genomic RNA can then be translated directly at the host ribosomes. The viral proteins are processed through cleavage by viral protease enzymes. This releases a repertoire of non-structural proteins (NSPs) and accessory open reading frame (ORF) proteins that has evolved, for example in the SARS-CoV-2 virus, to bind and block the proteins in the interferon I (IFN-I) antiviral cascade (KE1901). The normal function of the host’s IFN-I response to other viruses is the expression of IFN-I which in turn stimulates the expression of many interferon-stimulated gene (ISG) proteins with antiviral functions. The SARS-CoV-2 antagonism of the IFN-I pathway delays or curtails the expression of IFN-I and ISG proteins.

The COVID-19 pandemic began in early 2020, after the cause of a respiratory illness first seen in Wuhan, China in December 2019 was identified as a severe acute respiratory syndrome-related coronavirus and the viral genome was sequenced in January 2020 (Zhu et al., February 2020; Gorbalenya et al., March 2020). Early literature characterizing the transcripts and proteins from SARS-CoV-2 started to become available soon after (e.g., Dongwan Kim et al., April 2020). Of two other papers published in April 2020, one indicated the angiotensin converting enzyme 2 (ACE2) cell entry receptor and the TMPRSS2 enzyme priming mechanism for the virus to enter the host cell (Hoffmann et al.), and another provided a map of protein-protein interactions of the host with the translated viral genome proteins after entry (Gordon et al.). These sources established a course that led to literature indicating the specific effects of protein-protein interactions between viral and host proteins. Other sources made comparisons with data from previous studies of the SARS and MERS coronaviruses (Channappanavar et al., 2019). Many of the host proteins engaged by the viral non-structural proteins (NSPs) and accessory open reading frame (ORF) proteins were part of the IFN-I antiviral response pathway (Xia et al., 2020; Xia and Shi, 2020). Therefore, IFN-I response antagonism was considered a key event and literature search strings targeted the specific NSPs and ORFs with IFN-I pathway proteins (e.g., SARS-CoV-2 NSP3 + interferon regulatory factor 3 [IRF3]). Literature searches utilized the large established digital libraries including NCBI, Google and Google Scholar. Database resources for 3D molecular structures and viral-host protein binding information include Aquaria-COVID Structural Models of COVID-19 Proteins at https://aquaria.ws/covid#Map, the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (PDB) at https://www.rcsb.org/ and Alphafold Protein Structure Database at https://alphafold.ebi.ac.uk/. General IFN and innate immune antiviral response literature was reviewed, and primary literature cited in review articles was also investigated.

Pollutant exposures. Most studies have been conducted with the endpoints to determine whether prior or concurrent exposure to chemical or air particulate pollutants exacerbates COVID-19 symptoms resulting in more severe disease or higher mortality rates. This would point to effects downstream of viral replication usually relating to antibody suppression, inflammation and organ/tissue damage. Fewer studies can be found that study pollutant effects on susceptibility to infection, which are relevant to this KER, specifically cell entry or interferon response antagonism.

Marques et al., (2022) reviews associations between COVID-19 and outdoor air pollutants including PM_2.5, PM₁₀, O₃, NO₂, SO₂ and CO, reporting that environmental air pollution increases both disease incidence and severity. Physiological mechanism is not investigated for most studies. One relevant study estimated significant odds ratios for increased risk of severe COVID-19 and gene transcriptional analysis showing downregulation of genes associated with the IFN-I pathway in patients with high short-term NO₂ exposure (Feng et al., 2023).

Per- and polyfluoroalkyl substances (PFAS) are a large group of contaminants of current concern, due to their potential for toxicity, ubiquitous presence in the environment and consumer products, as well as their resistance to degradation. Although most community exposure to PFAS is through diet and drinking water, airborne and dermal exposures may also occur, especially in the workplace (CDC/NIOSH 2022). Statistical links between high measured serum or urine concentrations of specific PFAS compounds or mixtures and higher rates of COVID-19-positive cases have been found. One study in Sweden calculated a sex- and age-Standardized Incidence Ratio (SIR) for the town of Ronneby that had highly PFAS-contaminated drinking water compared to a demographically matched town with background PFAS levels (Nielsen et al. 2021). Serum PFAS concentrations were previously measured in 2014-15 for 3507 participants (Xu et al. 2021), after the Ronneby drinking water contamination issue was identified in 2013. Ronneby residents had higher infection risk, with a SIR of 1.19 [95% CI: 1.12-1.27]. Ji et al. (2021) measured urine and serum in a smaller study in China with 160 subjects. They reported statistically significant odds ratios for infection of 1.94 [95% CI: 1.39–2.96] for perfluorooctane sulfonate (PFOS), 2.73 [1.71–4.55] for perfluorooctanoic acid (PFOA), and 2.82 [1.97–3.51] for Σ (12) PFASs, after controlling for age, sex, body mass index (BMI), comorbidities, and urine albumin-to-creatinine ratio (UACR). These odds of infection were clearly higher even though the PFAS-exposed subjects in China had serum concentrations lower than in the Ronneby study participants. Additionally, the risk of infection was similar for residents in a significantly more contaminated section of Ronneby compared with a less contaminated section, so there was no dose-response relationship (Nielsen et al. 2021). However, these associations warrant more study to determine causality. Ji et al. (2021) also found elevated PFAS to be associated with altered mitochondrial metabolism. A potential consideration is that inhibition of mitochondrial oxidative phosphorylation impairs MAVS-mediated induction of IFNs, indicating the coordination between antiviral response and mitochondrial metabolism (Yoshizumi et al., 2017). Another study proposes modulation of ACE2 and TMPRSS2 expression in the lungs of PFAS-exposed mice may play a role in PFAS-associated immune suppression (Yang et al. 2022). Houck et al., (2022) report testing 147 PFAS substances in screening platforms including the BioMAP® Diversity PLUS panel, which is used to model complex tissue adverse effects of pharmaceuticals and environmental chemicals. Toxicity Signatures within the BioMAP profile indicated the Skin Rash (MEK-Associated) Signature for PFOA, with IFNα/β as one of the target mechanisms. While not specific to COVID-19, one study found that exposure to aryl hydrocarbons and dioxins may block IFN production (Franchini and Lawrence, 2018).

The current quantitative understanding of this relationship is described below.

Sex and age applicability

It has been shown that in human populations males are more likely to suffer severe infections and deaths due to COVID-19 than females. However, in the viral entry and infection phase, one study found that women of working age had higher infection rates than men, but the suggested cause was higher contact rates among women (Doerre and Doblhammer, 2022). Contact rate increase is an important transmission factor but would not constitute a gender-based biological difference in viral entry or IFN-I pathway antagonism. A biological basis for females having higher levels of Type I IFN has been proposed concerning Toll-like receptor (TLR) 7. TLR7 is expressed in plasmacytoid dendritic cells (pDCs), an immune cell type that on infection with SARS-CoV-2 migrates from peripheral blood into the respiratory tract epithelium. TLR7 stimulates higher IFN-I production in pDCs in women than in men (Van der Sluis et al. 2022). It is proposed that this is due to the TLR7 gene being on the X chromosome, and that X inactivation in males is incomplete regarding the TLR7 gene, creating a double gene-dose effect in females (Spiering and de Vries, 2021). In a mouse SARS-CoV model, XY males had more adverse outcomes than XX females and XXY males (Gadi et al. 2020). Additionally, loss-of-function TLR7 mutations have been identified that are associated with increased COVID-19 severity (Szeto et al. 2021). However, these results focus on disease outcome as the endpoint, where factors beyond the initial antiviral response could be involved. Also note that the nasal and upper respiratory tract (URT) epithelial cells express ACE2 receptors for SARS-CoV-2 entry while the pDCs do not, relying on viral endocytosis (Van der Sluis et al. 2022). There is not a clear picture in the literature of the timing of pDC arrival in the epithelium after exposure, and the role of TLR7 in sex differences is currently hypothetical (Spiering and de Vries, 2021).

In a large study modelling URT viral load dynamics drawn from measurements in 605 human subjects, variations over 5 orders of magnitude in URT viral load from the time of symptom onset was not explained by age, sex, or severity of illness. Additionally, these variables did not explain modelling results concerning control of viral load by immune responses in the early (innate) or late (adaptive) phases (Challenger et al. 2022). Other sources also support that rate of infection and measured viral load does not differ by gender (e.g., Arnold et al. 2022; Qi et al. 2021; Cheemarla et al. 2021). Therefore, evidence exists that the components of cell entry and the early antiviral response are not influenced by gender specific differences such as sex hormone levels or sex chromosomes to the extent of affecting viral load. Elderly people are more susceptible to severe disease than children and young adults, but Challenger et al. (2022) found no evidence indicating that life stage increases infectability or early viral load generation. However, Sharif-Askari et al. (2022) reported that children had higher expression of IFN-I and associated ISGs than adults.

Taxonomic applicability

Generally, most mammals are likely susceptible to the SARS-CoV-2 virus based on reports of naturally and experimentally infected animals (See AO 1939). No infections have been reported in other classes of vertebrates. Other than bioinformatic studies on the ACE2 sequence across vertebrates however, there have been few studies on the mechanisms of susceptibility to infection of non-human hosts. Three studies were found on protein targets in the IFN-I innate immune response pathway that included other vertebrates. Rui et al. (2021) showed that SARS-CoV-2 3CLpro and ORF3a inhibit vertebrate (human, mouse, and chicken) STING ability to induce IFNβ promoter activity in a dose-dependent manner in HEK293T cells transfected with IFNβ-luciferase reporter plasmid vectors, together with tagged STING and cGAS vectors and increasing amounts of the SARS-CoV-2 3CLpro or ORF3a expression vectors. This study shows that the vulnerability of the host IFN-I pathway protein components to inhibition by SARS-CoV-2 protein stressors is not limited to humans, however Rui et al. (2021) did not determine the specific amino acids involved in the STING-ORF3a or STING-3CLpro interactions. Mostaquil et al. (2020) studied the cleavage site of IRF3 by PLpro (SARS-CoV-2 NSP3) and compared sequences across mammals. They determined that the IRF3 cleavage site in mammalian species in the taxonomic orders of primates, carnivora, artiodactyla, chiroptera (bats) and a few other mammals was conserved and would generally be susceptible to cleavage, and therefore IFN-I antagonism, but rodentia IRF3 would likely not be susceptible. Hameedi et al. (2022) compared molecular dynamic simulations of 3CLpro cleavage of NEMO in humans and mice showing a decrease in the average number of contacts between mNEMO and 3CLpro compared to hNEMO. Also, hNEMO may be more strongly bound to the catalytic site, and the mNEMO/3CLpro interaction appears more prone to destabilization (Hameedi et al., 2022).