Respiratory infection leaves a persistent inflammatory program in tissues where no virus remains, and olfactory loss is produced in cells the virus never enters

Status co-led. Frere 2022 and Serafini 2023 are co-corresponding with the Zachariou laboratory, Zazhytska 2022 was led with the Lomvardas and Overdevest groups, and Heaton 2014 is co-led with Peter Palese and tenOever as joint senior and corresponding authors Areas Pandemic Host Response and Disease, Innate Immune Signaling and the Interferon Response, Programmable Virology

Persistent symptoms after COVID-19 were well documented clinically with no biological account and no small animal system. Two things were needed. A comparator, because an inflammatory change a month after a severe respiratory infection means nothing until it is known whether another respiratory virus leaves the same mark, and a way to ask how a tissue the virus barely enters changes what it does. The problem had a precedent in influenza. Every label for an infected cell reports on a viral product and therefore decays, so whether any cell survives a productive infection and what it then does could not be asked until Heaton 2014 put Cre recombinase on the PB2 segment and marked infected cells permanently in reporter mice. Marked cells persisted at 10 and 21 days after infectious virus was undetectable, were confined to the epithelium of larger airways, retained Cc10 as their only lineage marker, carried an amplified interferon-stimulated gene signature with elevated Cxcl10, Ccl20 and Ccl5, and their ablation reduced bronchiolar epithelial necrosis.

Frere 2022 made the comparator a requirement, infecting hamsters with SARS-CoV-2 or 2009 pandemic H1N1 influenza at doses matched for peak titre and profiling lung, heart, kidney and six nervous system regions at peak, one week after clearance and at 31 days. The discipline of that design is that several findings which would have read as specific to SARS-CoV-2 turn out to be shared with influenza, including the acute interferon response across tissues, peribronchiolar metaplasia, renal tubular atrophy and loss of olfactory receptor transcripts. What survived was narrower. At 31 days the olfactory bulb of influenza animals had returned to baseline while SARS-CoV-2 animals retained interferon signatures, elevated CXCL10 and CCL5 and microglial and myeloid activation, with no viral RNA detectable and no difference in apoptotic nuclei, alongside altered behaviour at 26 days and correlated inflammatory programs in tissue from recovered human donors. Serafini 2023 found the same shape in the peripheral sensory nervous system, with viral nucleocapsid transcripts and Isg15 rising in dorsal root ganglia and spinal cord within a day while no infectious virus was recoverable from any neural tissue, followed at 31 days by 1065 differentially expressed genes of neuropathic character and returning mechanical hypersensitivity in both sexes.

Zazhytska 2022 supplied a route by which an infection reaches cells it does not enter. The virus infected sustentacular cells, which were transiently depleted and restored, while neuronal representation stayed constant. Uninfected olfactory sensory neurons nonetheless mounted an antiviral response and then lost expression of olfactory receptor genes and of the signal transduction machinery for odour detection, still absent ten days after clearance. In situ Hi-C showed that the long-range contacts among receptor gene clusters, which normally converge into specialised interchromosomal compartments, were reduced from one day and remained reduced at ten days, and ultraviolet-inactivated serum from infected animals given intranasally to naive animals for 12.5 hours reproduced the loss of contacts with no viral genome transferred. Human autopsy tissue showed the same receptor downregulation and, in sorted neuronal nuclei, the same loss of contacts.

Taken together the corpus supports a general statement that much of the damage attributed to this virus is done by the host response rather than by infection of the damaged cells, and that statement is synthesis across these papers rather than a claim any one makes. What none of them establishes is causality. Frere 2022 does not link persistent brain inflammation to the behavioural change, offers residual defective genomes and barrier breach as untested alternatives, and notes that absence of virus at 31 days is a limit of its assays. Zazhytska 2022 did not identify the circulating molecule or the receiving neuronal pathway, did not test smell in either species, and states its nuclear memory proposal as a hypothesis, with human data resting on two control and four infected autopsies. Serafini 2023 describes the association between viral material and persistent hypersensitivity as correlative, and its nominated target ILF3 was validated in mouse pain models and never tested in infected animals. Heaton 2014 states that the link between the interferon-stimulated gene signature of surviving club cells and their survival is correlative, and its ablation removes all marked survivors rather than club cells specifically.

Substantiated by - Long-term survival of influenza virus infected club cells drives immunopathology, shows that some airway cells survive productive influenza infection and remain inflammatory after clearance, and that removing them reduces damage - SARS-CoV-2 infection in hamsters and humans results in lasting and unique systemic perturbations after recovery, benchmarks against pandemic influenza and isolates what persists in the olfactory bulb at 31 days with no detectable virus - Non-cell-autonomous disruption of nuclear architecture as a potential cause of COVID-19-induced anosmia, shows nuclear architecture holding olfactory receptor genes dissipating in neurons the virus does not enter, reproducible with virus-free serum - SARS-CoV-2 airway infection results in the development of somatosensory abnormalities in a hamster model, extends the pattern to sensory ganglia and couples a late neuropathic transcriptome to returning hypersensitivity

Research areas

Supporting publications