A Cre recombinase-expressing influenza A virus combined with Cre-responsive reporter and ablation mouse strains showed that a subpopulation of directly infected lung cells, predominantly club cells, survives productive infection, sustains elevated interferon-stimulated gene and chemokine expression, and contributes to bronchiolar epithelial damage after virus is cleared.
Heaton NS, Langlois RA, Sachs D, Lim JK, Palese P, tenOever BR. Long-term survival of influenza virus infected club cells drives immunopathology. Journal of Experimental Medicine. 2014. Volume 211, issue 9, pages 1707-1714.
DOI 10.1084/jem.20140488. PMID 25135297. PMCID PMC4144728.
A Cre recombinase-expressing influenza A virus combined with Cre-responsive reporter and ablation mouse strains showed that a subpopulation of directly infected lung cells, predominantly club cells, survives productive infection, sustains elevated interferon-stimulated gene and chemokine expression, and contributes to bronchiolar epithelial damage after virus is cleared.
Influenza A virus infection of the respiratory tract produces tissue damage and proinflammatory signalling that can outlast detectable virus. Conventional tracking of infected cells relies on viral RNA or viral protein, both short-lived, so the fate of an infected cell after virus clearance has been hard to establish. Heaton and colleagues built an H1N1 virus carrying Cre recombinase fused to PB2 through a PTV-1 2A site and infected mice carrying Cre-activated reporter cassettes, which permanently marks any cell in which viral replication and protein synthesis occurred. Reporter-positive cells were present not only during active replication but also at 10 and 21 days after infection, past the point at which infectious virus could be recovered from lung. Histology placed the surviving marked cells in the epithelium of larger airways and never in alveoli. Transcriptional profiling of sorted reporter-positive and reporter-negative cells identified Cc10 as the only cell type marker retained in survivors, implicating club cells, and showed that survivors carried a higher magnitude interferon-stimulated gene signature and elevated Cxcl10, Ccl20 and Ccl5. Ablating the surviving cells with a Cre-inducible diphtheria toxin receptor reduced bronchiolar epithelial necrosis. The work reframes a subset of infected airway cells as durable local sources of inflammatory signal rather than as cells that are uniformly destroyed.
Influenza A virus infects several cell types in the respiratory tract, including ciliated epithelial cells, type I and type II alveolar cells and immune cells. Infected cells were understood to be eliminated either by replication-driven apoptosis and necrosis or by innate and adaptive immune clearance, and acute infection in humans resolves within about two weeks. The paper states the methodological gap directly. Infected cells are normally identified through virus-derived products or reporters whose half-lives are short, so those approaches cannot define infected cell types at times well after replication has ceased. Separately, severe influenza outcomes have been associated with sustained proinflammatory responses, which raised the unresolved question of what cellular source maintains that signalling once virus has been cleared.
Do any lung cells survive a productive influenza A virus infection, and if so, what lineage are they, what transcriptional state do they hold after virus clearance, and do they contribute to the immunopathology that persists beyond the acute phase.
The design rests on converting a transient viral event into a permanent host-genome mark. Cre recombinase was inserted downstream of a PTV-1 2A site at the 3 prime end of the PB2 segment of A/Puerto Rico/8/1934, so that Cre is produced only where the viral polymerase segment is expressed. Mice carrying a lox-stop-tdTomato cassette then label any cell in which that expression occurred, and the label persists whether or not the cell still contains virus. Specificity controls established that reporter activation requires active replication rather than uptake of infected cell debris, and virulence comparison with the parental strain established that the recombinant virus still produces influenza-like disease. Once the tool was validated, three readouts were layered onto it. Flow cytometry over a time course established whether marked cells persist past clearance, measured against plaque assay for infectious virus. RNA sequencing of sorted reporter-positive and reporter-negative cells from the same lungs supplied both lineage assignment through cell type marker transcripts and functional state through interferon-stimulated gene and chemokine levels, with the paired sorting controlling for shared tissue environment. Finally, substituting a Cre-inducible diphtheria toxin receptor strain for the reporter strain converted the labelling system into an ablation system, allowing the surviving population to be removed after replication had largely ended and the consequence for lung pathology to be scored by an independent pathologist.
IAV-Cre activates the host reporter only under productive infection. Mock-infected and wild-type-infected reporter fibroblasts showed no tdTomato, IAV-Cre infection produced red fluorescence, pretreatment with type I interferon before IAV-Cre abolished the signal, and lysed debris from IAV-Cre infected cells applied under neutralizing antibody produced no signal (Figure 1 C). The authors read this as evidence that viral RNA replication and protein expression are required for labelling.
The recombinant virus retains influenza pathogenicity. Intranasal IAV-Cre produced morbidity comparable to the parental strain, with the median lethal dose shifting from 50 to 240 PFU (Figure 1 D). The paper notes this experiment was performed once.
Reporter-positive cells persist beyond virus clearance. Marked cells were detected at 5, 10 and 21 days after infection (Figure 2 A), while infectious virus recoverable from lung homogenate was high at days 2 and 5 and undetectable by day 10 (Figure 2 C). No marked cells appeared in uninfected mice or in reporter mice given wild-type virus (Figure 2, A and B).
Surviving marked cells localise to the epithelium of larger airways. At day 10, tdTomato-positive cells were found in bronchial epithelium and never in alveoli (Figure 2 D).
Surviving cells carry a transcriptional profile distinct from their reporter-negative neighbours in the same lung (Figure 3 A), and replicate sorting and sequencing of day 10 samples were reproducible (Figure 3 B).
Viral mRNA is lost from marked cells by day 10 (Figure 3 C), and day 5 viral reads mapped across all eight segments (Figure 3 D). The authors interpret the eight-segment coverage as arguing against reporter activation by defective interfering particle entry. Sorted day 5 marked cells injected into embryonated eggs yielded virus in 11 of 12 eggs, while day 10 marked cells did not (Figure 3 E), which is read as showing the day 10 population has genuinely cleared a productive infection.
Club cells are the predominant surviving lineage. Early after infection marked cells expressed both Sftpc and Cc10, but Sftpc was almost entirely lost by day 5 while Cc10 was maintained (Figure 3 F), and no other cell type specific marker was detected at appreciable levels in the survivor population (Table S1). This is a lineage inference from marker transcripts rather than from a lineage-restricted genetic label, and the authors support it with the concordance between club cell anatomical distribution and the observed localisation of marked cells.
Survivors hold an amplified interferon-stimulated gene response. Interferon-stimulated genes were higher in magnitude in tdTomato-positive than in tdTomato-negative cells from the same lungs at day 5 (Figure 3 G) and the signature persisted across the time course (Table S1). In cell lines, a murine club cell line showed significantly higher Irf7 and Isg15 induction than a lung epithelial line after interferon treatment or infection (Figure 3, H and I). The authors state explicitly that the relationship between interferon-stimulated gene upregulation and cellular survival is only correlative, and describe the survival explanation as a favoured hypothesis.
Depleting surviving cells reduces airway pathology. Diphtheria toxin given at day 5 to Cre-inducible diphtheria toxin receptor mice infected with IAV-Cre produced a significant reduction in bronchiolar epithelial pathology at day 10, with non-depleted airways showing frequent segmental necrosis and depleted airways showing only minor lesions and often unaffected terminal bronchioles (Figure 4, A and B). Overall cell infiltration scores did not differ significantly between groups. The authors flag that interpretation is complicated by possible contributions from surviving non-club cells.
Survivors are a selective source of proinflammatory chemokines. Cxcl10, Ccl20 and Ccl5 were highly induced in marked cells relative to uninfected controls while Ccl2 was not appreciably induced (Figure 4 C). Infection of the murine club cell line raised secreted CXCL10, CCL20 and CCL5 protein without a significant change in CCL2 (Figure 4 D), and the human club cell line H441 behaved similarly (Figure 4 E).
The study does not establish a definitive mechanism for either survival or pathology. What the data constrain is the sequence of states. Cells that go on to survive were productively infected, as shown by the eight-segment viral read coverage and the recovery of infectious virus from day 5 sorted cells. Those cells clear virus, as shown by loss of viral mRNA and by failure of day 10 sorted cells to amplify in eggs. They are predominantly club cells by marker transcript, and they retain an elevated interferon-stimulated gene and chemokine programme after clearance. Removing them reduces bronchiolar epithelial necrosis.
What the data do not constrain is causation in either direction. The authors state that the link between interferon-stimulated gene levels and survival is correlative, and the club cell line comparison establishes a higher intrinsic interferon responsiveness in that lineage without showing that this responsiveness is what permitted survival in vivo. Likewise the depletion experiment shows that removing surviving cells reduces damage, but it does not identify which secreted factor or which recruited cell population mediates that damage, and the authors note that non-club survivors may contribute. The model the authors propose is that club cells directly infected by influenza A virus but surviving that infection establish a local proinflammatory environment in the bronchi, useful for initial antiviral control but detrimental to bronchial remodelling once the pathogen is gone. That is presented as a proposal consistent with the data rather than as a demonstrated causal chain.
The reporter virus converts a transient infection event into a permanent heritable mark on the host cell, which makes the post-clearance fate of infected cells an accessible experimental question rather than an inference. Pairing that label with a Cre-inducible diphtheria toxin receptor turns the same genetic logic into a functional test, so the population defined by having been infected can be removed and its contribution to disease measured. Conceptually, the finding that an acute lytic respiratory infection leaves behind a surviving, transcriptionally altered epithelial population makes it testable whether post-viral inflammatory states in the airway are maintained by cells that were themselves infected, rather than solely by immune infiltrate or by residual antigen.
The Cre-expressing influenza virus extends a line of engineered influenza tools from the same groups that use viral genome modification to interrogate host biology, including the hematopoietic-specific targeting virus of Langlois and colleagues in 2012 and the microRNA-based biocontainment strategy of Langlois and colleagues in 2013, both cited here. The coupling of a recombinant virus to host transcriptional profiling by sequencing, with sorted infected and uninfected populations compared within the same tissue, is a recurring design in the corpus.
Category 3 synthesis. Recurring questions about how a localised infected cell population shapes tissue-level inflammatory outcome, and about post-acute sequelae of respiratory virus infection, appear elsewhere in this corpus, including in later hamster and human airway work on SARS-CoV-2. Establishing that connection properly requires those records side by side and is not something this paper supports on its own.
The infection model is a single laboratory-adapted H1N1 strain, A/Puerto Rico/8/1934, delivered intranasally to inbred C57BL/6 mice, and the conclusions are bounded to that combination. The recombinant virus is mildly attenuated relative to the parental strain, with the median lethal dose shifting from 50 to 240 PFU, and the virulence comparison was performed once. Lineage assignment to club cells rests on cell type marker transcripts in sorted populations and on anatomical concordance rather than on an independent club cell specific genetic label, and the authors describe it as a hypothesis supported by those observations. The depletion experiment removes all Cre-marked survivors rather than club cells specifically, and the authors state that contributions from surviving non-club cells complicate interpretation. The link between the elevated interferon-stimulated gene signature and cellular survival is explicitly described by the authors as correlative. Time points extend to 21 days after infection, so nothing is established about longer-term persistence or resolution of the survivor population. Chemokine secretion in the human system was measured only in the H441 cell line, not in primary human airway tissue, and no human in vivo data are presented. Several key experiments are stated to be representative of two independent experiments, and pathology was scored on two lung sections 100 micrometres apart per animal.
Some lung cells survive influenza infection rather than dying. These surviving airway cells keep signalling inflammation after virus is gone, and removing them reduces airway damage in mice.
Influenza is usually thought to kill the cells it infects. Using a virus that permanently tags any cell it replicates in, this work showed that a small population of mouse airway cells, mostly club cells, clears the virus and lives on. Those survivors keep antiviral and inflammatory genes switched on after virus is undetectable. Genetically removing them lessened damage to the airway lining, indicating that surviving infected cells contribute to post-infection lung injury.
An influenza A virus carrying Cre recombinase on the PB2 segment was used with Cre-responsive mouse strains to permanently mark cells that supported viral replication, independent of whether virus remained. Marked cells persisted at 10 and 21 days after infection, past the point when infectious virus was recoverable, and were confined to the epithelium of larger airways. RNA sequencing of sorted marked and unmarked cells from the same lungs identified Cc10 as the only retained lineage marker, implicating club cells, and showed that survivors carry an amplified interferon-stimulated gene signature and elevated Cxcl10, Ccl20 and Ccl5 without induction of Ccl2. A murine club cell line was intrinsically more interferon responsive than a comparison lung epithelial line, and murine and human club cell lines secreted the same selective chemokine set upon infection. Ablating marked survivors with a Cre-inducible diphtheria toxin receptor significantly reduced bronchiolar epithelial necrosis without altering overall infiltration scores.
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