Pandemic Host Response and Disease

Why does SARS-CoV-2 produce the pattern of disease it does, and what part of that is the host's own response?

In one paragraph

Eighteen publications between 2020 and 2023 pursue one question, how much of COVID-19 is the virus and how much is the response to it. The founding observation, Blanco-Melo 2020, positioned SARS-CoV-2 against five other respiratory viruses in matched systems and found low type I and type III interferon induction alongside vigorous chemokine and IL-6 output, with the inflammatory arm independent of interferon signalling. Nilsson-Payant 2021 then showed that the inflammatory arm is not an escape from viral control but a requirement of the virus, since NF-kappa B-driven transcription is needed for replication. Work in the golden hamster extended the description into whole animals and across time, and Carrau 2023 showed that interferon generated by airway replication circulates and keeps the infection respiratory, so that blunting or bypassing the airway response permits viremia and productive infection of distal organs. In parallel, collaborations led by other groups mapped host requirements and druggable signalling, mapped which human cell types are permissive, and built models. Three studies then found a persistent inflammatory program in tissues where no virus remains detectable, and one of them, Zazhytska 2022, traced loss of smell to nuclear reorganization in neurons the virus never enters. Much of this work was done at speed and a substantial part of it rests on entry reporters, correlations without intervention, and animal models whose relationship to human disease is assumed rather than shown.

The defining scientific question

The question the area asks is not what SARS-CoV-2 encodes but what the host does when it arrives, and which part of the resulting pathology belongs to which. Stated as the corpus asks it, why does this virus produce the pattern of disease it does, and what part of that pattern is the host's own response. Two subsidiary questions follow. The first is whether the inflammatory character of the disease reflects a failure of viral antagonism or a viral dependency, which determines whether an intervention should restore the missing arm or suppress the intact one. The second is whether the effects seen away from the airway require the virus to be there, which determines whether extrapulmonary disease is a tropism problem or a signalling problem.

Origins

This area begins abruptly. There is no earlier coronavirus programme in the corpus leading into it, and the framing of the central question in Blanco-Melo 2020 is explicitly that a matched comparison across systems was missing for a pathogen that had emerged weeks earlier. What made a fast answer possible was that the necessary capabilities existed for other reasons.

The records support naming several. The comparative transcriptional profiling approach, in which viral and host reads are sequenced from the same libraries so that viral load and host response are read together, had been developed by the laboratory for influenza A virus and other RNA viruses, and Blanco-Melo 2020 applies it unchanged. The reference standard for an unantagonised response was an influenza A virus lacking its NS1 protein, used in that paper as a positive control for what a full interferon response in primary bronchial cells looks like. The influenza strains themselves are cited to Langlois 2013, work originally undertaken to mitigate the risk of gain-of-function influenza studies. The ferret system used for the animal arm comes from the laboratory's influenza transmission bottleneck work, cited to Varble 2014. The conceptual framing of the cellular response is taken from the laboratory's own review, tenOever 2016, and the assay for detection, activation of TBK1, traces to Sharma 2003 from the senior author's doctoral training period. Reading these as a continuous research arc would be synthesis across papers rather than something any of them claims.

One collaborative capability also predates the pandemic explicitly. Si 2021 reports that the airway chip programme was funded by DARPA and the National Institutes of Health two years earlier in anticipation of biothreat challenges, and that the group's COVID-19 work began on 13 January 2020, one day after the genome sequence was released.

Major findings

The response phenotype. Blanco-Melo 2020 established low interferon induction with partial interferon-stimulated gene induction alongside strong chemokine and interleukin expression, reproduced in cell lines, primary bronchial epithelium, ferrets, post-mortem lung and patient serum, and showed with ruxolitinib that the chemokine arm does not require interferon signalling. The paper identifies no viral antagonist and does not adjudicate between an antagonist being overcome at high multiplicity of infection and high multiplicity generating patterns that may not form physiologically.

The inflammatory arm as a viral requirement. Nilsson-Payant 2021 found NF-kappa B engaged at chromatin, transcript and protein level with the interferon factors untouched, and showed that loss of p65 or p50 blocks viral protein accumulation while a chimeric RelA activator restores it, placing the requirement at the level of NF-kappa B-driven transcription.

Interferon as a local intervention and as a systemic signal. Hoagland 2021 showed intranasal type I interferon, or a double-stranded RNA mimetic, lowering viral load and disease burden and preventing transmission in three of five exposed animals, and documented antiviral transcription in olfactory bulb, brain and small intestine at viral loads far below the airway. Carrau 2023 then showed by three independent manipulations that airway replication is the source of circulating interferon that primes every organ, and that dexamethasone, intravenous delivery and deliberate priming move distal infection in the predicted directions.

Host requirements and drug candidates. Daniloski 2021 in Cell ranked every protein-coding gene by the effect of its loss and converged on endosomal machinery, with cholesterol biosynthesis as a shared consequence and RAB7A controlling surface ACE2. Bouhaddou 2020, led by the Krogan laboratory, showed the host response over 24 hours enacted through phosphorylation rather than abundance and converted inferred kinase activity into tested inhibitors. Yaron 2022, led from Duke and Weill Cornell, resolved nucleocapsid phosphorylation into an ordered kinase cascade reachable with an approved drug. Si 2021, led at the Wyss Institute, showed that dosing at clinically achievable concentrations under flow eliminated hydroxychloroquine and chloroquine while retaining amodiaquine, which then worked in hamsters.

Permissiveness and entry. Yang 2020, led by the Chen, Evans and Schwartz groups, found that ACE2 protein does not predict permissiveness across human lineages. Eriksen 2021 placed the ocular surface among directly infectable human tissues and localized the vulnerability to the limbus. Daniloski 2021 in eLife isolated Spike D614G from its linked ORF1b variant and showed it raises entry efficiency without altering S1 affinity for ACE2.

Persistence. Frere 2022 benchmarked against pandemic influenza and found sustained olfactory interferon signalling and myeloid activation at 31 days with no detectable virus, together with altered behaviour and correlated signatures in recovered human tissue. Zazhytska 2022 showed that infection of sustentacular cells dissipates the interchromosomal compartments holding olfactory receptor genes in uninfected neurons, an effect reproduced by virus-free serum. Serafini 2023 found viral RNA without infectious virus in sensory ganglia followed by a late neuropathic transcriptome and returning hypersensitivity.

Host variation. Horiuchi 2021 showed memory clearing a beta variant its own serum neutralized poorly, while protected animals still infected every cohoused partner. Oishi 2022 in Cell Reports showed age diminishing the duration and quality of the response without raising viral load. Oishi 2022 in the Journal of Virology showed influenza suppressing SARS-CoV-2 for up to two weeks after its own clearance.

How the work evolved

Three movements are visible. The first is from description to dependency, running from Blanco-Melo 2020 through Nilsson-Payant 2021, which inverts the reading of the same phenotype. The second is from the site of infection to the site of response, running from Hoagland 2021, where distal inflammation was attributed speculatively to disseminated viral material, to Carrau 2023, which tested that proposal, found for circulating interferon instead, and showed the response to be protective rather than merely present. The third is from acute to persistent, where Frere 2022, Zazhytska 2022 and Serafini 2023 find an inflammatory program decoupled from detectable virus in three different tissues.

Some threads do not continue. No viral antagonist responsible for the low interferon phenotype is identified anywhere in this corpus. The haematopoietic progenitor signature noted in ferret trachea in Blanco-Melo 2020 is flagged as requiring further work and is not returned to. The NF-kappa B dependency established in culture by Nilsson-Payant 2021 could not be reproduced in the laboratory's hamster model, which the authors state. The drug candidates arising from Bouhaddou 2020, Daniloski 2021 and Yaron 2022 are not followed up within this corpus. ILF3, nominated in Serafini 2023, was validated in mouse pain models and never tested in infected animals.

Principal publications

• Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19. Lab-led. The founding observation for the area.
• The NF-κB Transcriptional Footprint Is Essential for SARS-CoV-2 Replication. Lab-led. Reframes the inflammatory arm as a viral requirement.
• Leveraging the antiviral type I interferon system as a first line of defense against SARS-CoV-2 pathogenicity. Lab-led. Builds the hamster platform and tests interferon as a local intervention.
• Delayed engagement of host defenses enables SARS-CoV-2 viremia and productive infection of distal organs in the hamster model of COVID-19. Lab-led. Separates the site of replication from the site of response by intervention.
• SARS-CoV-2 infection in hamsters and humans results in lasting and unique systemic perturbations after recovery. Co-led with the Zachariou laboratory. Establishes persistence with a benchmark virus.
• Non-cell-autonomous disruption of nuclear architecture as a potential cause of COVID-19-induced anosmia. Co-led with the Lomvardas and Overdevest groups. Nuclear architecture as the route to anosmia.
• Immune memory from SARS-CoV-2 infection in hamsters provides variant-independent protection but still allows virus transmission. Lab-led. Dissociates host protection from interruption of spread.
• Identification of Required Host Factors for SARS-CoV-2 Infection in Human Cells. Co-led with the Sanjana laboratory. Genome-scale host dependency map.
• The Global Phosphorylation Landscape of SARS-CoV-2 Infection. Led by the Krogan laboratory. Phosphoproteomic landscape and inhibitor candidates.
• Host protein kinases required for SARS-CoV-2 nucleocapsid phosphorylation and viral replication. Led from Duke and Weill Cornell. Ordered nucleocapsid phosphorylation cascade.
• A human-airway-on-a-chip for the rapid identification of candidate antiviral therapeutics and prophylactics. Led at the Wyss Institute. Clinically relevant dosing as a candidate filter.
• A Human Pluripotent Stem Cell-based Platform to Study SARS-CoV-2 Tropism and Model Virus Infection in Human Cells and Organoids. Led by the Chen, Evans and Schwartz groups. Permissiveness map across human lineages.
• Mouse genome rewriting and tailoring of three important disease loci. Led by Zhang and Boeke. Genomically humanized ACE2 and TMPRSS2 mice.

Connections to other areas

Four of these studies also sit in innate immune signalling, where the interferon system is the object rather than the variable, and Blanco-Melo 2020 reaches back into that area by citing Sharma 2003 for TBK1. The comparative use of influenza A virus as a calibrated reference runs through Blanco-Melo 2020, Hoagland 2021, Horiuchi 2021, Frere 2022, Serafini 2023 and both Oishi 2022 papers, and connects the area to influenza genome regulation, where Nilsson-Payant 2021 on nucleoprotein availability belongs. Daniloski 2021 in Cell is also assigned to programmable virology under screening through fitness, and Zhang 2023 to synthetic virology as method, since both treat a genome as something to be rewritten in order to pose a question.

Current implications

Three conclusions from this work are robust enough to act on and should be stated with their evidentiary character attached. Blocking inflammation rather than supplying interferon systemically follows from the imbalanced response, but it is an interpretation of correlated observations across systems, and the therapeutic suggestions in Blanco-Melo 2020 are flagged by its authors as untested. Local airway delivery of interferon or of a pattern recognition receptor agonist is supported by intervention in an animal, prophylactically and one day after challenge, and says nothing about dosing or efficacy in humans. The finding that speed of early airway engagement determines where the virus can go suggests that individuals whose interferon response is blunted by age, immunosuppression, inborn error or autoantibody are the ones in whom transient viremia seeds distal organs, which the authors of Carrau 2023 present as speculation. For drug development, the clearest practical lesson is Si 2021, that a candidate nominated in a static cell line assay at convenient concentrations is not yet a candidate.

Open questions

Which viral product limits interferon induction is unanswered across the corpus. Which NF-kappa B target gene the virus requires is unidentified. What drives persistent olfactory inflammation once virus is undetectable is unresolved, with residual defective genomes and barrier breach admitting commensal organisms offered as untested alternatives. The identity of the circulating factor that disrupts nuclear architecture in olfactory neurons is unknown, as is the neuronal pathway that receives it, and whether the compartment loss is reversible is untested. Whether persistent inflammation causes the behavioural changes measured in hamsters is not demonstrated. Whether the cell types found permissive in stem cell derived systems are meaningful sites of infection in patients requires primary patient material. Whether the ocular surface is a transmission route or a reservoir is raised and not answered. What nucleocapsid phosphorylation does for the virus remains undefined. Whether the coinfection interference reflects residual interferon-stimulated gene expression or an increased resident immune population was not separated, and whether it affects severity is unknown.

Publications referenced

• Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19
• The Global Phosphorylation Landscape of SARS-CoV-2 Infection
• A Human Pluripotent Stem Cell-based Platform to Study SARS-CoV-2 Tropism and Model Virus Infection in Human Cells and Organoids
• Identification of Required Host Factors for SARS-CoV-2 Infection in Human Cells
• The Spike D614G mutation increases SARS-CoV-2 infection of multiple human cell types
• SARS-CoV-2 infects human adult donor eyes and hESC-derived ocular epithelium
• Leveraging the antiviral type I interferon system as a first line of defense against SARS-CoV-2 pathogenicity
• Immune memory from SARS-CoV-2 infection in hamsters provides variant-independent protection but still allows virus transmission
• The NF-κB Transcriptional Footprint Is Essential for SARS-CoV-2 Replication
• A human-airway-on-a-chip for the rapid identification of candidate antiviral therapeutics and prophylactics
• SARS-CoV-2 infection in hamsters and humans results in lasting and unique systemic perturbations after recovery
• A diminished immune response underlies age-related SARS-CoV-2 pathologies
• The Host Response to Influenza A Virus Interferes with SARS-CoV-2 Replication during Coinfection
• Host protein kinases required for SARS-CoV-2 nucleocapsid phosphorylation and viral replication
• Non-cell-autonomous disruption of nuclear architecture as a potential cause of COVID-19-induced anosmia
• Delayed engagement of host defenses enables SARS-CoV-2 viremia and productive infection of distal organs in the hamster model of COVID-19
• SARS-CoV-2 airway infection results in the development of somatosensory abnormalities in a hamster model
• Mouse genome rewriting and tailoring of three important disease loci
• Triggering the Interferon Antiviral Response Through an IKK-Related Pathway
• MicroRNA-based strategy to mitigate the risk of gain-of-function influenza studies
• Influenza A Virus Transmission Bottlenecks Are Defined by Infection Route and Recipient Host
• The Evolution of Antiviral Defense Systems
• Reduced Nucleoprotein Availability Impairs Negative-Sense RNA Virus Replication and Promotes Host Recognition

Explore the themes

Connected discoveries

Publications in this area