Authors: Jacob Class (1Department of Microbiology and Immunology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA), Lacy M. Simons (2Department of Medicine, Division of Infectious Diseases, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.; 3Center for Pathogen Genomics and Microbial Evolution, Havey Institute for Global Health, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Ramon Lorenzo-Redondo (2Department of Medicine, Division of Infectious Diseases, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.; 3Center for Pathogen Genomics and Microbial Evolution, Havey Institute for Global Health, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Jazmin Galván Achi (1Department of Microbiology and Immunology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA), Laura Cooper (1Department of Microbiology and Immunology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA), Tanushree Dangi (4Department of Microbiology-Immunology, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Pablo Penaloza-MacMaster (4Department of Microbiology-Immunology, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Egon A. Ozer (2Department of Medicine, Division of Infectious Diseases, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.; 3Center for Pathogen Genomics and Microbial Evolution, Havey Institute for Global Health, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Sarah E. Lutz (5Department of Anatomy and Cell Biology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA), Lijun Rong (1Department of Microbiology and Immunology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA), Judd F. Hultquist (2Department of Medicine, Division of Infectious Diseases, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.; 3Center for Pathogen Genomics and Microbial Evolution, Havey Institute for Global Health, Feinberg School of Medicine, Northwestern University, Chicago, IL 60611, USA.), Justin M. Richner (1Department of Microbiology and Immunology, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA)
Categories: Article, SARS-CoV-2, Viral Evolution, COVID-19, NeuroCOVID, Furin Cleavage Site (FCS), Central Nervous System (CNS)
Source: Nature microbiology
Authors: Jacob Class, Lacy M. Simons, Ramon Lorenzo-Redondo, Jazmin Galván Achi, Laura Cooper, Tanushree Dangi, Pablo Penaloza-MacMaster, Egon A. Ozer, Sarah E. Lutz, Lijun Rong, Judd F. Hultquist, Justin M. Richner
Severe COVID-19 and post-acute sequelae of SARS-CoV-2 infection are associated with neurological complications that may be linked to direct infection of the central nervous system (CNS), but the selective pressures ruling neuroinvasion are poorly defined. Here, we assessed SARS-CoV-2 evolution in the lung versus CNS of infected mice. Higher levels of viral divergence were observed in the CNS than the lung after intranasal challenge with a high frequency of mutations in the Spike furin cleavage site (FCS). Deletion of the FCS significantly attenuated virulence after intranasal challenge, with lower viral titers and decreased morbidity compared to the wild-type virus. Intracranial inoculation of the FCS-deleted virus, however, was sufficient to restore virulence. After intracranial inoculation, both viruses established infection in the lung, but dissemination from the CNS to the lung required the intact FCS. Cumulatively, these data suggest a critical role for the FCS in determining SARS-CoV-2 tropism and compartmentalization.
SARS-CoV-2 is an enveloped, positive-sense, single-stranded RNA virus and the causative agent of COVID-19. The virus has caused more than 7 million deaths. Eleven different WHO approved vaccines have been developed to date, significantly reducing global hospitalizations and deaths^1^. The respiratory system is the primary location of SARS-CoV-2 infection where the virus replicates in lung epithelial cells^2^. SARS-CoV-2 can also disseminate to distal tissues including the heart, gastrointestinal tract, and the central nervous system (CNS)^3,4^. COVID-19 is associated with a number of extrapulmonary pathologies including neurological post-acute sequelae, acute kidney injury, gastrointestinal distress, myocarditis, thromboembolism, and acro-cutaneous lesions^5^. The viral and host factors which contribute to these extrapulmonary pathologies are not well understood.
SARS-CoV-2 entry into a host cell is mediated by the viral spike glycoprotein (S). S is produced as a full-length immature form before it undergoes proteolytic cleavage and maturation. S is cleaved into the S1/S2 subunits by furin proteases at the furin cleavage site (FCS; 681-PRRAR-685)^6^. The RBD within S1 binds to ACE2 on the surface of susceptible host cells and undergoes a conformational shift. In the canonical entry pathway, this shift exposes the S2/S2-prime cleavage site, which is cleaved by a surface bound protease, TMPRSS2. S2/S2-prime cleavage exposes the fusion peptide (FP) domain of S2, which results in detachment of the S1 domain^7^. The exposed FP inserts itself into the plasma membrane of the cell, before a conformational change of S2 initiates membrane fusion and release of the genome into the cytoplasm^8^. Alternatively, SARS-CoV-2 can enter the cell through the endosomal entry pathway following ACE2 receptor binding. Within the maturing endosome, the cysteine proteases cathepsin B/L cleave the S protein at the S2/S2-prime site allowing for S1 release and membrane fusion^9,10^.
The continued evolution of SARS-CoV-2 has resulted in the emergence of new viral variants with enhanced fitness^11–16^. Variants associated with an increase in transmissibility, disease severity, or immune evasion are considered ‘variants of concern’ (VOCs)^17^, and most of the mutations responsible for these fitness benefits lie within the S protein. Mutations that alter FCS cleavage efficiency and stability of the S1/S2 interaction are commonly found in VOCs. The first mutation found to enhance SARS-CoV-2 transmissibility was S:D614G, which enhanced S1/S2 stability following furin cleavage and promoted sampling of the open confirmation, which exposes the RBD for ACE2 binding^18,19^. Subsequent mutations in the Alpha (S:P681H) and Delta (S:P681R) variants within the FCS increased furin cleavage and promoted viral entry at the cell surface^20^. Proteolytic cleavage of the FCS removes the covalent bond between the S1 and S2 domains and can occur throughout various steps of the viral lifecycle^21^. It has been demonstrated that cleavage of the FCS results in altered cellular tropism, a high TMPRSS2 processing rate, and increased viral transmission^22^. Consistent with this, deletion of the FCS in SARS-CoV-2 results in attenuated disease compared to ancestral virus in mouse and hamster models^23^. However, these studies were largely limited to tracking viral load in the lung; the dynamics of ΔFCS viral variants and how these mutations may alter viral tropism and pathogenesis remains unknown.
Here, we asked if preexisting immunity shapes viral evolution and how SARS-CoV-2 evolves within different host tissues. Utilizing two different mouse models, we discovered increased viral divergence within the CNS regardless of vaccination status, suggesting that immune privileged sites such as the CNS may serve as incubators for SARS-CoV-2 diversification. We also found that the FCS is positively selected for in the respiratory tract and negatively selected for in the CNS, suggesting a critical role for this site in determining SARS-CoV-2 tropism and compartmentalization.
To better understand the determinants driving viral evolution, we performed viral whole genome sequencing of SARS-CoV-2 in the lungs and brains of mice which had received different vaccine formulations. We hypothesized that vaccine-induced immunity against S and/or nucleocapsid (N) would suppress viral diversity independent of compartment due to a suppression of viral replication. K18-hACE2 mice were intramuscularly vaccinated with Ad5-vector vaccines that encoded either the SARS-CoV-2 S open reading frame (Ad5-S) or the N open reading frame (Ad5-N) from the SARS-CoV-2 isolate USA-WA1/2020, similar to our previous studies^24–28^. Mice were vaccinated with either 1×10^9^ PFU of Ad5-S, Ad5-N, both Ad5-S and Ad5-N each, or with PBS as a control (n = 5 mice per condition). The vaccines elicited robust humoral and cellular immune responses as measured by anti-S and anti-N IgG and CD8+ T cell responses, respectively (Supplemental Fig 1). After three weeks, mice were challenged intranasally with 5×10^4^ (PFU) of USA-WA1/2020 and ultimately euthanized at 5- dpi (Fig. 1a). Total RNA from lung or brain homogenate was analyzed for the presence of viral RNA by qRT-PCR and used for viral whole genome sequencing using an amplicon based approach as previously described^29–31^.
Whole genome sequencing and phylogenetic analysis of the viral isolates from each compartment and condition revealed that vaccination status and formulation had no effect on viral divergence in the lung. However, while the lung isolates were highly similar across all vaccination groups, the brain isolates were much more divergent from the original input virus (Fig. 1b). To better understand the distribution of these changes, Shannon entropy was calculated at every position along the genome (Fig. 1c). Cumulatively, diversity was higher in the lung than the brain of PBS and Ad5-N vaccinated mice while there was no difference in overall diversity by compartment in the Ad5-S and Ad5-N + Ad5-S mice. Notably, this was not attributable to changes in the lung as there was no statistically difference in diversity in the lung across different vaccination groups. In the brain however, diversity in the Ad5-S and Ad5-N + Ad5-S mice was higher than in the PBS controls. Taken together, this suggests that Ad5-S lowers viral diversity in the lung, but that higher diversity is maintained in the brain. This is reflected in increased viral divergence in the brain regardless of vaccination.
To better understand these results, we examined diversity across the genome. While several positions varied within both compartments, there was a notable enrichment in diversity in S with the most diversity occurring in and around the FCS (Fig. 1d). Examining the consensus sequences of each isolate in this region, we found all 20 lung isolates had a majority consensus sequence that matched the reference with an intact FCS (Fig. 1e) regardless of vaccination. In stark contrast to this, 15 of the 20 brain isolates had a majority consensus sequence with a substitution mutation, frameshift, or deletion in or near the FCS. The 5 brains isolated that did not have a change in the consensus sequence at this site were distributed among all four vaccination groups, again suggesting that this selection is not dependent on prior vaccination.
Notably, the parental isolate USA-WA1/2020 used in the above experiment was found to be polymorphic at position 23606 in the FCS. Sequencing of the USA-WA1/2020 inoculate yielded 47.9% reference sequence at that position while 52.1% of reads contained a C23606T (R682W) mutation that renders the FCS nonfunctional. Furthermore, SARS-CoV-2 is known to accumulate to very high titers in the tissues of K18-hACE2 mice^32,33^, which may impact the selective pressures underlying compartmentalization. To test our findings in an alternate model, we intranasally challenged immunocompetent BALB/c mice with a mouse adapted strain of SARS-CoV-2 (MA10)^34^ and euthanized at 5 dpi (Fig. 2a). Total RNA was extracted from lung or brain homogenate to determine viral load and perform viral whole genome sequencing. Viral RNA was readily detected in both homogenates of the MA10 infected BALB/c mice (Fig. 2b). Whole genome sequencing and phylogenetic analysis of the viral isolates from each compartment revealed that the lung isolates again more closely matched the initial inoculate compared to the brain isolates (Fig. 2c). Two of the five brain isolates showed substantial divergence from the input inoculate with one isolate’s majority consensus sequence harboring an R682G mutation in the FCS (Fig. 2d). Furthermore, Shannon entropy per position over the genome was significantly increased in the brain compared to the lung isolates (Fig. 2e). This diversity was again concentrated in the S open reading frame, with the highest peaks in entropy at sites in and around the FCS (Fig. 2f). Taken together, these data - in two independent mouse models with varying levels of CNS pathology and viral replication - suggest that neuroinvasion elicits a selective pressure for deletion or mutation of the FCS independent of prior immunity.
A previous study reported that deletion of the FCS in SARS-CoV-2 attenuates viral pathogenesis after a respiratory inoculation and reduces viral titers in the lung^35^, but it is not clear if this occurs in other organs of the body. The CNS has lower expression levels of ACE2 and TMPRSS2 compared to lung epithelial cells^36–38^. The lack of TMPRSS2 in the CNS suggests that the virus might use an alternative entry pathway to target cells in the brain. ACE2 can be found on epithelium, neurons, and vascular cells within the brain, which also has high levels of cathepsin B and L expression^39^. We thus hypothesized that deletion of the FCS is due to tissue specific selective pressure in the CNS to utilize the endosomal entry pathway.
To evaluate the entry pathway of a virus lacking the FCS, we generated luciferase expressing pseudoviruses with an HIV backbone and SARS-CoV-2 S glycoprotein on the surface. Pseudoviruses were generated with the S amino acid sequence from the parental SARS-CoV-2 isolate USA-WA1/2020 (WT) with or without deletion of amino acids 681-PRRA-684 (ΔFCS). Loss of the FCS prevents furin-mediated processing of the S protein, resulting in higher levels of full-length S protein compared to the cleaved S1 and S2 domains as demonstrated by Western blot on the purified WT and ΔFCS pseudoviruses (Fig. 3a). To interrogate whether the ΔFCS pseudovirus exploits an alternative entry pathway, we challenged VeroE6 cells overexpressing hACE2 and TMPRSS2 (VAT cells) with each pseudovirus in the presence of increasing concentrations of either a TMPRSS2 inhibitor (camostat mesylate) or a cathepsin B/L protease inhibitor (aloxistatin/E64d) using luciferase activity as a readout. Camostat mesylate inhibited the WT pseudovirus approximately 3-fold better than the ΔFCS pseudovirus (Fig 3b). Aloxistatin inhibited the ΔFCS pseudovirus approximately 100-fold better than the WT pseudovirus. This data indicates that the ΔFCS mutant is dependent on endosomal cathepsin B/L proteases for efficient viral entry, whereas the WT virus can enter more efficiently use TMPRSS2 on the cell surface. We also compared the entry efficiency of the WT and ΔFCS pseudoviruses on Calu3 and VAT cells. Both WT and ΔFCS pseudoviruses infected VAT cells more efficiently than Calu3 cells. WT pseudovirus infection of Calu3 cells, however, was >10-fold more efficient than the ΔFCS virus (Fig 3c). Together, these data demonstrate that mutation of the FCS inhibits TMPRSS2-mediated entry in favor of the endosomal entry pathway.
Due to the increased prevalence of ΔFCS mutations in the CNS, we hypothesized that these mutations may pose a selective advantage in this compartment. To address this, we obtained an infectious clone of wild-type USA-WA1/2020 virus and with the ΔFCS mutation (deletion of 681-PRRA-684)^23^. We confirmed the expected furin processing patterns by western blot (Fig. 3d). Sequencing determined the WT viral stock had 91.5% WT FCS sequence while 8.5% of reads contained the aforementioned R682W mutation in the FCS. The ΔFCS stock had 93.6% of reads with the predicted ΔPRRA deletion and 6.4% of reads with wild-type FCS. We inoculated K18-hACE2 mice intranasally (Fig. 4a). As seen previously^23^, the ΔFCS mutant caused significantly less weight loss in intranasally inoculated mice (Fig. 4b). Intranasally inoculated mice were euthanized at 2- and 5-dpi to provide a time course of viral dissemination during early infection. The ΔFCS mutant had an impaired growth rate compared to the WT virus with a 25-fold reduction in genomes/lung at 2 dpi (Fig. 4c) and a 2.5-fold reduction in infectious viral particles (FFU)/lung (Fig. 4d). At 2-dpi, viral nucleocapsid was widely dispersed throughout the lung of WT infected mice, whereas nucleocapsid staining was more punctate in the ΔFCS inoculated lungs (Fig. 4e), By 5-dpi, viral RNA titers remained elevated in the WT inoculated lungs (6-fold increase WT vs ΔFCS), whereas WT infectious virus titers were decreased in the lungs, likely due to clearance of infectious virus. Nucleocapsid staining remained widespread in the lungs of WT infected mice at 5-dpi. ΔFCS virus had spread more throughout the lungs, although nucleocapsid staining was not distributed widely throughout the lung as observed in the WT infected mice (Fig. 4e). We then asked if the deletion of the FCS provided improved entry into the CNS. However, we found that at 2dpi the ΔFCS had 58-fold lower genome/organ in the CNS compared to the WT virus (Fig 4f), although we could not detect infectious virus in the brains of either WT or ΔFCS inoculated mice (Fig 4g). By 5-dpi viral RNA and infectious virus titers were highly variable in the brain. The ΔFCS inoculated mice had slightly higher viral RNA and infectious virus titers, although these differences were not significant. We observed more wide-spread nucleocapsid staining in sites throughout the brain in ΔFCS inoculated mice (hippocampus and premotor cortex in Fig 4h).
To assess CNS specific replication dynamics, we inoculated K18-hACE2 mice directly into the brain (Fig. 5a). Following inoculation with either strain, mice reached humane endpoint criteria by 3-dpi with no clear difference in clinical symptoms or weight loss between the groups (Fig. 5b). At 1-dpi, the ΔFCS mutant virus had increased viral RNA loads in brain compared to the WT, indicating that ΔFCS virus has a slight early growth advantage in the CNS (Fig. 5c). Infectious ΔFCS titers were higher than WT titers, however these differences did not reach significance (Fig 5d). We observed foci of infected cells in the corpus callosum by 1-dpi (Fig. 5e). Intriguingly these clusters of infected cells were only observed in the ΔFCS inoculated mice and not the WT mice, indicating more efficient infection of the ΔFCS virus at this site. We did not observe comparable infectious foci at other sites in the brain at 1-dpi. By 3-dpi, there was no significant difference in the viral RNA or infectious virus loads in the brains of ΔFCS and WT infected mice. We also quantified virus in the lungs of mice infected via the intracranial route. Low levels of viral genome copies in the lung were detected and these values increased from day 1 to day 3, suggesting that viral dissemination can occur not only from the respiratory tract to the brain, but also from the brain to the lung (Fig. 5f). We could not detect infectious virus in the lung likely due to the limit of detection of our assay (Fig. 5g). However, in vitro infection of VAT cells with lung homogenate from intracranially infected mice resulted in cytopathic effect (Fig. 5h), indicating the presence of infectious virus. Collectively, these data demonstrate that, in contrast to the findings in the lung, the ΔFCS virus is not attenuated in the brain, and that virus originating from the brain can traffic back out to the respiratory tract.
To better understand the compartmental viral population dynamics following intracranial or intranasal challenge, each isolate was again subject to viral whole genome sequencing and phylogenetic analysis. Mice infected intranasally with WT virus (left panels) displayed the previously observed pattern with the lung isolates (green squares) grouping closely with the input inoculate (blue circle) and with the brain isolates showing more divergence (green triangles) (Fig. 6a). Following intranasal inoculation, the consensus sequence of the FCS in each lung isolate matched the reference whereas 3 of the 5 brain isolates had substitution mutations or deletions near or in the FCS (Fig. 6b). On the other hand, mice infected intracranially with WT virus had some lung isolates (red squares) cluster more closely with the brain isolates (red triangles) with more divergence from the input inoculate (Fig. 6a). Looking at the consensus sequences, 4 of the 5 brain and lung isolates maintained wild-type sequence across the FCS after intracranial inoculation, with one isolate in each compartment gaining a substitution or deletion mutation in that region (Fig. 6b). Quasispecies analysis confirmed that divergence of viral subpopulation after intracranial inoculation was maintained in the lung after trafficking (Fig. 6c).
Mice infected with the ΔFCS virus (right panels) showed a distinct pattern of viral evolution. Upon intranasal inoculation, some divergence is observed relative to the input inoculate (Fig. 6a), but all brain and lung isolates maintain the FCS deletion (Fig. 6b). Upon intracranial inoculation, the FCS deletion is similarly conserved in all brain isolates. However, the lung isolates of intracranially ΔFCS inoculated mice were highly divergent from the inoculating virus (red squares, Fig. 6a). Furthermore, each lung isolate was found to have partially (n = 2) or completely (n = 3) re-acquired the FCS sequence (Fig. 6b), which is similarly reflected in the quasispecies analysis (Fig. 6c). Taken together, these data suggest that trafficking of SARS-CoV-2 between the lung and CNS elicits a differential selective pressure for or against an intact FCS in Spike, respectively.
We report that viruses lacking the FCS are attenuated in the lung and that ΔFCS pseudovirus have decreased entry efficiency in Calu3 compared to VAT cells, consistent with prior reports^23^. We present evidence that this is due to an increased reliance of ΔFCS viruses on the TMPRSS2-independent endosomal entry pathway. We postulate that attenuated growth of ΔFCS viruses after intranasal inoculation in vivo is triggered by decreased viral entry in respiratory cells (represented by the Calu3 cells in vitro), which leads to lower viral titers in the lungs and reduced pathology. However, we also observe selective loss of the FCS in the brain, where we postulate that low TMPRSS2 levels favor the endocytic pathway for entry. Previous studies have demonstrated that furin-cleavage decreases overall stability of the spike protein, but increases binding affinity towards the ACE2 receptor^40^. We hypothesize that the more stable ΔFCS virus is better able to survive the endocytic entry pathway in target cells of the CNS.
Similar to our findings, the absence of the FCS in other coronaviruses has been associated with increased CNS tropism. The human alphacoronavirus OC-43 naturally lacks an FCS, and when a neuroinvasive strain of OC-43 gained an FCS, dissemination into the CNS was decreased^41^. We would hypothesize that deletion of the FCS is selected for because the target cell type required for neuroinvasion has low TMPRSS2 expression, putting pressure on the virus to favor endosomal-mediated entry. However, it remains unclear if this selective pressure is driven by tropism of a specific trafficked cell type or by population bottlenecking that amplifies selective pressure within a compartment. On one hand, inter-compartment trafficking may be mediated by a specific immune cell type that elicits a tropism-specific selective pressure, but the selection for an FCS when trafficking to the lung and for loss of an FCS when trafficking to the brain would require the involvement of different cellular intermediates dependent on directionality. On the other hand, population bottlenecking at the time of seeding may better amplify selective pressures within compartments that might not be observed upon direct inoculation due to sufficient initial challenge to overcome these barriers. Immunohistochemistry in the CNS of the intranasally inoculated mice showed that WT and FCS mutant viruses could both infect neuronal cells, however the FCS mutant spread more rapidly, suggesting that the selective pressure occurs within the target cells of the CNS itself.
Several viruses are known to acquire compartment-specific adaptations that improve fitness in a given anatomical site, including Human Immunodeficiency Virus (HIV)^42^ and other coronaviruses. Murine hepatitis virus predominantly replicates within the liver, but mutations within the S protein can alter cellular tropism and cause neurovirulence. Similar to our data, replication in the CNS is associated with the acquisition of variants with mutations and deletions within the S protein that alter S dynamics and affinity to the host-cell entry receptor^43,44^. While our studies were limited to the brain, we hypothesize that tissue-specific patterns of viral diversity may develop across other anatomical sites. Indeed, sequencing of SARS-CoV-2 from heart tissue found an overrepresentation of the Spike Q675H mutation, which has been suggested to enhance furin cleavage^45,46^. On the other hand, the immune privileged status of the CNS may facilitate longer infection time courses in those tissues, which would enable a higher degree of genetic diversification specifically in that compartment. The role of compartmentalization on emergence of novel variants warrants further investigation.
Taken together, these data define a selective pressure acting on SARS-CoV-2 Spike during neuroinvasion and compartmental trafficking. Whether direct infection of the CNS is responsible for the neurological complications observed during acute COVID-19 and long-term PASC and how the properties of the virus influence CNS pathology remains unclear^47^. These data may be important for understanding the mechanisms governing intra-host coronavirus evolution, shedding light into the factors that influence the emergence of novel variants and the development of neurological pathologies.
All animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Illinois College of Medicine (Protocol # 21–084). K18-hACE2 transgenic mice (B6.Cg-Tg(K18-ACE2)2Prlmn/J) male mice (Jax Strain # 034860) aged 4–8 weeks, were purchased from Jackson Laboratories and were maintained as hemizygotes through breeding at UIC. K18-hACE2 expression was validated through genotyping as described by Jackson Laboratories. Mice were randomly separated into experimental groups, then given ad libitum access to food and water and kept on a 12h light/dark cycle in microisolator cages (Allentown – BCU2) equipped with HEPA filters. Ambient temperature of 68–76°F maintained in the animal facility. BALB/c mice were obtained from Jackson Laboratories (Jax Strain # 00651). For intranasal infection mice were anesthetized with isoflurane and then intranasally inoculated with virus in a 50μL droplet placed on the left and right nostrils of the mouse. Inhalation of the droplet was confirmed for each mouse. For intracranial infection, anesthetized 5-week-old mice were inoculated via direct injection of 1×10^2 PFU (plaque forming units) of ΔFCS or WA-1 stock in 10μL of PBS through the top of the skull at a depth of 1–2mm with a 29G needle. Mice were euthanized and organs were collected and homogenized in 1mL PBS. Homogenization was achieved with 1mm silicon beads at 5m/s for 60 sec on brain tissue and 3mm silicon beads for 120sec on lung tissue. For immunohistochemistry, mice were anesthetized with aerosolized isoflurane and perfused transcardially with 4% PFA. Tissues were fixed in 4% PFA for 24h, then submerged in 70% ethanol. Tissues were embedded in paraffin and 0.5uM sections were fixed onto slides and stained 2500 with SARS Nucleocapsid protein (Novus Biologicals NB56576). Images were taken on a Keyence BZ-X imager. Images were randomized prior to analysis. No statistical methods were used to pre-determine sample sizes but our sample sizes are similar to those reported in previous publications^23,32,33^
6–8-week-old K18-hACE2 mice were purchased from Jackson laboratories (Stock No: 034860). Mice were immunized intramuscularly (50μL per each quadricep) with an adenovirus type 5 (Ad5) vector expressing SARS-CoV-2 spike protein (Ad5-S), or nucleocapsid protein (Ad5-N), or both vectors combined; diluted in sterile PBS, at 10ˆ9 PFU per mouse. Ad5-N was a kind gift of the Masopust/Vezys laboratory^48^.
SARS-CoV-2, Isolate USA-WA1/2020, NR-52281 was deposited by the Centers for Disease Control and Prevention and obtained through BEI Resources, NIAID, NIH. Virus was propagated on Vero-E6 cells (ATCC, Catalog no. CRL 1586). SARS-CoV-2, Mouse-Adapted MA10 Variant infectious clone, NR-55329 was deposited by RS Baric and obtained through BEI Resources, NIAID, NIH. SARS-CoV-2 ΔFCS infectious clone and its parent USA-WA1/2020 clone (WA-1) were obtained from the Vineet Menachery Lab (University of Texas Medical Branch). Infectious clone derived viruses were propagated on Vero-E6 cells expressing ACE2 and TMPRSS2 (VAT cells. In brief, Vero and VAT cells were passaged in DMEM with 10% Fetal bovine serum (FBS) and Glutamax. Cells less than 20 passages were used for all studies. Viral stocks were used after a single expansion (passage = 1) to prevent genetic drift.
Focus forming assay (FFA) were performed as described previously^49^. Briefly, serial dilutions of tissue homogenate from inoculated mice were added to a monolayer of Vero-E6 cells in a 96-well plate. 1h after infection, cells were overlaid with 1% (wt/vol) methylcellulose in 2% fetal bovine serum (FBS), 1× minimal essential medium (MEM). 24 h after infection, plates were fixed for 15 min with 4% paraformaldehyde (PFA) followed by 1h with 10% neutral buffer formalin (NBF). Staining involved primary antibody polyclonal anti-SARS-CoV-2 guinea pig (BEI Resources – NR10361 15000) and secondary antibody goat anti-guinea pig–HRP (Thermo Cat# A16104 5000) in PermWash buffer (0.1% saponin, 0.1% BSA, in PBS). Treatment with TrueBlue peroxidase substrate (SeraCare – 5510–0030) produced focus-forming units that were quantified on an ImmunoSpot ELISpot plate scanner (Cellular Technology Limited).
RNA was isolated using RNeasy Mini Kit (Qiagen Cat# 74104) protocol and eluted in a volume of 40μL of RNase free water. Real-time quantitative reverse transcription PCR was performed using TaqMan 1-step RNA to Ct (Thermo Cat# 4392938) with CDC primer/probe kit (IDT - 10006713) against the N1 gene. Samples were analyzed using Viia7 (ThermoFisher) along with Quantstudio and DA2 analysis software (ThermoFisher). Genomes/mL were interpolated using Ct values and genomic standard (BEI - NR-52358) run in triplicate.
cDNA synthesis was performed with SuperScript IV (Thermo 18090200) using random hexamers according to manufacturer’s specifications. Direct amplification of viral genome cDNA was performed as previously described using the Artic Network version 4 primers. Sequencing library preparation of amplicon pools was performed using the SeqWell plexWell 384 kit (Seqwell PW384A) per manufacturer’s instructions. Pooled libraries were sequenced on the Illumina MiSeq using the V2 500 cycle kit. To generate consensus sequences, reads were trimmed to remove adapters and low-quality sequences using Trimmomatic v0.36. Trimmed reads were aligned to the reference genome sequence of SARS-CoV-2 (accession MN908947.3) using bwa v0.7.15. Pileups were generated from the alignment using samtools v1.9 and consensus sequence determined using iVar v1.2.2 with a minimum depth of 10, a minimum base quality score of 20, and a consensus frequency threshold of 0 (i.e., majority base as the consensus).
Consensus sequences assembled for each sample were aligned using MAFFT v7.453 software. A Maximum Likelihood (ML) phylogeny with all consensus sequences were inferred with IQ-Tree v2.0.5 using its ModelFinder function before each analysis to estimate the nucleotide substitution model best-fitted for each dataset by means of Bayesian information criterion (BIC). We assessed the tree topology for each phylogeny both with the Shimodaira–Hasegawa approximate likelihood-ratio test (SH-aLRT) and with ultrafast bootstrap (UFboot) with 1000 replicates each. Additionally, with the assembled reads from each sample we performed probabilistic inference of intra-host viral quasispecies of the Spike gene for each sample using QuasiRecomb. The sequences of the inferred viral haplotypes from each quasispecies were also aligned and ML phylogenies inferred using the same approach as the consensus analysis. All final tree representation was performed with the R package ggtree v3.2.1.
To study and compare intra-host diversification in different animals and tissues, Shannon Entropy was calculated using the nucleotide frequencies obtained from iVar applying the Sh = SUM[-(pi)·log2(pi)]; where Sh is Shannon Entropy calculated for each position and pi is the frequency of each nucleotide in each position. To ensure a robust estimation of diversity, Shannon Entropy calculations were limited to positions with a minimum read depth of 100 reads to ensure robustness of the measurement. To test for significant differences in overall genetic entropy between compartments and, in the case of the vaccine studies between groups and the group and tissue interaction, we used the genetic entropy values for each nucleotide position in every animal and tissue and fitted a log-transformed linear mixed effects model, using animal and nucleotide position as random effects. When multiple comparisons were tested, p-values were False Discovery Rate (FDR) adjusted using the Benjamini-Hochberg procedure. For these analyses we used lme4 version 1.1–34 in R version 4.0.3.
Pseudoviruses were created using plasmids for SARS-CoV-2 WA1/2020 spike, SARS-CoV-2 ΔFCS and HIV-1 proviral vector pNL4–3.Luc.R-E-(from the NIH AIDS Research and Reference Reagent Program) containing a luciferase reporter gene. Pseudovirions were created following a polyethylenimine (PEI)-based transient co-transfection on 293T cells (ATCC CRL-3216). After 5h, cells were washed with PBS and the medium was replaced with phenol red-free DMEM. 16 h post-transfection, supernatants were collected and filtered through 0.45μm pore size filters. Vero E6-TMPRSS2-T2A-ACE2 cells (NR-54970, BEI resources) or Calu3 (ATCC HTB-55) were seeded (1 × 10^4^ cells/well) in 96-well white-bottom plates and incubated at 37 °C and 5% CO2 for 24h. Compounds Aloxistatin (MedChem Express – HY-100229) and Camostat mesylate (MedChem Express – HY-13512), used for pseudovirion inhibition, were dissolved in DMSO. Cells were transduced with the pseudovirions with or without compounds while maintaining a 1% DMSO concentration. Compounds were serially diluted to obtain and calculate the IC50 and CC50 values. The treated cells were incubated for 48h. The degree of viral entry was determined by measuring the luciferase activity using the Neolite Reporter Gene Assay System (PerkinElmer). Pseudovirion inhibition and compound cytotoxicity were normalized using the 1% DMSO-only control. IC50 and CC50 values were obtained by analyzing the dose–response data, using the four-parameter logistic regression analysis in GraphPad Prism (version 10.1.1).
To isolate protein, viral stock or pseudovirus were prepared using 4x Laemmli sample buffer (Thermo – NP0007) and reducing agent (Thermo – NP0004). Samples were inactivated and denatured by 70°C heat for 10 min and loaded onto a 4–12% Bi-Tris gel (Thermo – NP0321BOX) submerged in 1x MEM SDS Running buffer (Thermo – NP0002). Protein was transferred to polyvinylidene difluoride (PVDF) membrane and stained with 4000 dilution of polyclonal anti-SARS-CoV-2 rabbit antibody (Novus Biologicals - 56578) followed by probing with 10000 HRP-conjugated anti-rabbit antibody (Invitrogen – 65–6120). Signal was developed by treating membranes with Clarity Western ECL Substrate (Bio-Rad – 170–5060) and imaged on a ChemiDoc MP Imaging System (Bio-Rad).
Lung Homogenate was added to a monolayer of VeroE6-ACE2-TMPRSS2 cells in 12 well plate. Cells were infected for 1h at 37°C, washed with PBS, then given fresh DMEM with 10% FBS and GlutaMax. For UV inactivation, 500μL of homogenate was placed in 10mm dish and exposed to direct UV light for 1h at room temperature.
Statistical analyses were performed using Graphpad Prism (v10.2.3). Statistical tests are defined within each figure legend. No statistical method was used to predetermine sample size. No data were excluded from the analyses. All in vitro studies were performed in at least three independent experiments and individual replicates verified results. All in vivo data points are shown in graphs. Images of viral nucleocapsid staining in Figures 4 and 5 are representative of 3–5 mice per group. For RNA and viral titer analysis, all samples were assigned an identification code and analysis performed on coded samples. Samples were decoded after data analysis. Researchers were blinded to sample name of IHC data upon initial analysis.