What is distinctive about the transcriptional response to SARS-CoV-2?
In the first weeks of 2020 there was no matched description of what SARS-CoV-2 does to a host cell. The general framework was settled, in which cytoplasmic sensors detect aberrant viral RNA, activate interferon regulatory factors together with nuclear factor kappa B, and launch two programs, one that restricts the virus through type I and type III interferon and the interferon-stimulated genes, and one that recruits leukocytes through chemokines. It was also understood that respiratory viruses encode antagonists of this system. What nobody could say was where SARS-CoV-2 sat relative to other respiratory viruses in the same cells, and whether any feature of that position bore on the inflammatory character of the disease.
A second problem followed once the first was answered. If a virus is evidently capable of dampening host transcriptional responses, why does the cytokine arm remain so active. Reading that activity as a failure of viral control assumes the virus would suppress it if it could.
Both publications assigned to this theme carry the contribution character lab-led, with tenOever as senior and corresponding author.
Blanco-Melo 2020 is the founding observation. Rather than characterising SARS-CoV-2 alone, it placed the virus alongside SARS-CoV-1, MERS-CoV, influenza A virus, human parainfluenza virus 3 and respiratory syncytial virus, using the same cell systems and the same differential expression framework, and then repeated the question at four levels of biological system. The recurring pattern was low type I and type III interferon with only a subset of interferon-stimulated genes induced, alongside pronounced chemokine and interleukin expression. In primary bronchial epithelial cells SARS-CoV-2 and wild-type influenza A virus produced responses similar in magnitude but shared only eight significantly induced genes. Ruxolitinib abolished interferon-stimulated gene induction while leaving chemokine induction largely intact, which is the direct evidence that the inflammatory arm does not require interferon signalling. Ferrets showed a cytokine response that outlasted viral clearance, patient lung showed chemokines without detectable interferon, and patient serum showed raised IL-6, IL1RA and multiple chemokines.
Nilsson-Payant 2021 supplied a transcription factor level account of the same pattern and then inverted its interpretation. In clonal ACE2-expressing A549 cells the dominant signature from nine hours onward was tumour necrosis factor alpha signalling through NF-kappa B, with no interferon signature and, at the protein level, no phosphorylation of STAT1 or IRF3. Single-cell sequencing placed that signature predominantly in infected rather than bystander cells. Chromatin accessibility profiling showed infection opening regions enriched for REL, RELA and NFKB1 motifs and not for IRF3 or IRF7 motifs, with the largest changes at distal regulatory elements rather than promoters. Silencing RelA reduced viral nucleocapsid protein and silencing NF-kappa B1 abolished it, and infection of RELA knockout cells was rescued by a chimeric RelA DNA-binding domain fused to a VPR activator, which places the requirement at the level of NF-kappa B-driven transcription rather than the protein itself. Four chemically distinct inhibitors reduced infection.
The movement is from a described phenotype to a claimed dependency. Blanco-Melo 2020 defines a reproducible response and explicitly declines to assign a mechanism, offering instead two unadjudicated readings of why interferon induction is low, that an antagonist is overcome at high multiplicity of infection or that high multiplicity generates pathogen patterns which may not form physiologically. No viral gene product is named. Nilsson-Payant 2021 does not close that question either, and its trigger for IKK activation is a proposal rather than a demonstration, but it reframes the inflammatory arm as something the virus requires rather than something it fails to suppress.
Both studies carry boundaries worth keeping in view. The human tissue analysis in Blanco-Melo 2020 rests on two post-mortem COVID-19 lungs and two healthy lungs, all from males over 60, the serum study on 24 cases and 24 controls which the authors describe as not necessarily representative, and comparison viruses were run at differing multiplicities and durations, so magnitude comparisons between viruses carry that caveat. Interferon induction proved strongly dependent on multiplicity, and the authors themselves note that high multiplicity may not reflect physiological infection, which limits how firmly the low interferon phenotype can be stated. The claim that the observed pattern drives COVID-19 pathology is an interpretation of correlated observations across systems and not a demonstrated causal chain. Nilsson-Payant 2021 is bounded to transformed lines overexpressing ACE2, the required NF-kappa B target gene is not identified, and the authors state that they could not reproduce the in vitro NF-kappa B inhibitor effect in their hamster model, so the dependency is shown in culture only.
This theme supplies the reference phenotype that the rest of the area is measured against. Interferon as intervention takes the low interferon observation as a premise and asks whether supplying interferon early helps, which Hoagland 2021 and Carrau 2023 pursue in the hamster. Tropism and permissive tissues reuse the same signature as a calibration standard, with Yang 2020 comparing infected pancreatic and hepatic models against the Blanco-Melo 2020 lung autopsy data and Eriksen 2021 comparing the ocular surface against it. Immunity, age and reinfection carries the framing into hosts that differ, where Oishi 2022 in Cell Reports finds the interferon-stimulated gene response falling away early in older animals and Horiuchi 2021 finds the NF-kappa B response delayed relative to influenza. The NF-kappa B dependency also connects outward to influenza work in the corpus, where a similar requirement has been reported with the mechanism unresolved.
SARS-CoV-2 infection of human lung epithelial cells engages NF-κB at chromatin, transcriptional, protein and post-translational levels without engaging the type I interferon transcription factors, and loss of p65 or p50 abolishes viral replication in a manner rescued by reconstituting RelA transcriptional activity.
Across cell lines, primary bronchial epithelium, ferrets and patient material, SARS-CoV-2 infection produces a transcriptional response distinguishable from that to other respiratory viruses, combining low type I and type III interferon induction with a moderate interferon-stimulated gene response and strong chemokine and IL-6 expression.