Kids’ noses resist COVID-19

In a paper published in this issue (Watkins et al., 2024), Watkins and colleagues report that children bear high loads of viruses and/or bacteria in the nose, causing potentiated innate immune responses that are distinct and depend on the dynamic nature of the microorganisms present in their upper airways.

Children have been notoriously resistant to the development of severe coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory coronavirus 2 (SARS-CoV-2). Multiple groups demonstrated that resistance to severe COVID-19 in children was associated with a heightened immune response in the upper respiratory tract. In particular, children were characterized by an elevated antiviral state associated to a potentiated interferon response (reviewed in Svensson Akusjarvi and Zanoni, 2024). It has been speculated that age-specific immunological factors and/or environmental exposure to respiratory viruses explain the presence of this augmented interferon response.

The team led by Ellen Foxman sheds new light on how the presence in the nose of viruses or bacterial pathobionts determines unique innate immune programs that may drive resistance to severe COVID-19.

The authors initially analyzed nasopharyngeal swabs of children taken during a period with low COVID-19 incidence or during the Omicron surge, which reported the highest pediatric infection rate in the U.S. The authors analyzed a total of 16 respiratory viruses (SARS-CoV-2, rhinovirus [RV], respiratory syncytial virus [RSV], etc.). They also tested the presence of the three most common bacterial pathobionts of the upper respiratory tract: Moraxella catarrhalis, Streptococcus pneumoniae, and Hemophilus influenzae. Across the periods analyzed, almost one-third of the children resulted positive for a virus, and between one-fourth and one-third for a pathobiont. Children <5 years old reached the highest level of infection with either viruses, pathobionts, or both. Of note, asymptomatic children also reported very high infection rates, with individuals <5 years old being the most asymptomatic but presenting the highest infection rates.

To understand how the presence of distinct microorganisms affected the nasal interferon response, the authors initially assessed the protein levels of CXCL10, which in a previous publication were shown to directly correlate with the transcriptional response driven by interferons (Cheemarla et al., 2021). The nasal levels of CXCL10 significantly correlated with the viral load, but not with the levels of pathobionts, and both CXCL10 and viral levels were generally higher in symptomatic, compared to asymptomatic, children. Several COVID-19–related studies found that the SARS-CoV-2 load does not always correlate to the interferon response, especially in children (Zanoni, 2021; Loske et al., 2022; Yoshida et al., 2022; Wimmers et al., 2023). To test whether co-infection with other viruses, rather the SARS-CoV-2 alone, could explain the interferon response in the upper airways of children, the samples described above were integrated with additional pediatric samples collected during an RSV resurgence. A total of 634 samples were analyzed, of which 42 presented multiple viral co-infections, demonstrating that the CXCL10 levels correlated with the overall viral load, either due to a single virus or to multiple co-infecting viruses.

The authors next focused on pathobionts. Re-analyses of publicly available RNA sequencing datasets of children infected with RV revealed that pathobionts (found in 50% of the samples) led to the upregulation of transcriptional pathways associated to neutrophils/granulocytes migration and activation, immune response to fungi, pyroptosis, and pro-inflammatory cytokine induction. Multiplex protein measurement of cytokines showed that interleukin-1 (IL-1) and TNF were among the most enriched cytokines. Watkins and colleagues then correlated levels of viruses and/or photobionts with CXCL10, TNF, and IL-1β in their cohort of <5-year-old children, which have the highest presence of photobionts. While CXCL10 correlated with the presence of viruses, independently of pathobionts, IL-1β was strongly or weakly upregulated by pathobionts or viruses, respectively. TNF was induced by either class of microorganisms but reached maximal levels during virus/pathobiont co-infections.

Finaly, paired nasopharyngeal swabs were longitudinally collected from 1-year-old infants that were undergoing routine healthy child visits. Although the children were healthy, viruses and/or pathobionts were detected in more than half of the samples. Levels of CXCL10 increased upon acquisition of a viral infection, or decreased when the kids cleared the infection, and always correlated to the viral load. TNF and IL-1β only increased when pathobionts were present in children that acquired a viral infection. Overall, these data demonstrated that the nasal innate immune response is dynamic and changes with acquisition or clearance of viruses and that the nasal immune status is determined by virus, pathobiont, or co-infections.

The findings reported in this paper are important for several reasons. First, they demonstrate that the nasal innate immunity of children dynamically changes based on the acquisition or clearance of different microorganisms, and also that viruses and pathobionts drive distinct immune programs, independently of the presence of overt infection symptoms. The presence of a respiratory virus is the only predictor of the induction of CXCL10, while both pathobionts and viruses can drive TNF and IL-1, with pathobionts being stronger inducers. Finally, the authors demonstrate that there is a very strict correlation between the interferon response and the total nasal viral load, either driven by a single virus or by multiple viruses.

Some key questions remain, though, open. The nature of the interferons that drive the response in the nasal mucosa was not assessed in this investigation, and CXCL10 was used as the proxy for a response that is very complex and can be driven by type I, II, or III interferons. Of note, 5–12% of severe COVID-19 cases have been associated with the presence of anti–type I interferon autoantibodies (Bastard et al., 2020), suggesting that type I interferons play a major role in protecting against SARS-CoV-2. Nevertheless, the picture is complicated by several factors. First, the induction of interferons is dynamically regulated upon SARS-CoV-2 encounter (Lucas et al., 2020). Second, the landscape of interferons across the respiratory tract varies based on COVID-19 severity (Sposito et al., 2021). The high expression of distinct interferons in the nose and lungs of patients with severe COVID-19 (Sposito et al., 2021), together with mouse studies demonstrating that interferons decrease the barrier function of the lungs (Major et al., 2020; Broggi et al., 2020a), suggest that the nature, rather than presence/absence, of interferons is fundamental. In general, type I and III interferons drive very similar transcriptional programs, and the necessity to maintain over evolution both families remains an unsolved question. Part of the answer to this question may reside in the strict compartmentalization of the responses driven by these two interferon families. In their previous paper, Foxman and colleagues demonstrated that nasal epithelial cells produce almost exclusively type III interferons (Cheemarla et al., 2021). Epithelial cells are also the major responders to type III interferons, that with few relevant exceptions, do not signal in immune cells (Broggi et al., 2020b). In contrast, type I interferons signal in almost all immune and non-immune cells. Beyond anti-microbial roles, interferons can also have strong inflammatory roles, especially when acting on immune cells. A possible explanation for the maintenance of multiple interferon families is that respiratory viruses are initially sensed by upper airways epithelial cells that produce, and respond to, type III interferons in the tentative to contain and neutralize the virus. If/when this first level of defense is not efficacious, type I interferons are produced to engage a broader group of (immune) cells to avoid the spread of the virus, even at the cost of excessive inflammation or damage. The longitudinal samples utilized by Watkins et al. (2024) for their study can become the ideal tool to give an answer to this long-standing question.

Another open question is whether, beside the factors described in their paper by Foxman and colleagues, there are also age-related intrinsic immune features that explain the resistance of children to SARS-CoV-2. Two recent papers demonstrated that pediatric nasal epithelial cells grown at the air–liquid interphase in in vitro cultures are intrinsically more resistant to SARS-CoV-2 and are characterized by heightened interferon responses (Woodall et al., 2024; Zhu et al., 2022). These findings are not in contrast to the ones discussed above and instead point to either long-lasting (possibly epigenetic) changes induced by frequent infections in the nasal mucosa and/or to a non-mutually exclusive capacity of children’s nasal epithelial cells to produce/respond to interferons. Zhu et al. (2022) also identified interesting differences in the capacity of nasal epithelial cells to respond to distinct variants of concern of SARS-CoV-2, another aspect that, due to the nature of their samples, Watkins and colleagues could not explore. Finally, Woodall et al. (2024) report very distinct programs activated in different cell types among nasal epithelial cell cultures of children and elderly. This is another point of great interest, especially in the context of virus and bacteria co-infections. While in adults and/or elderly virus and bacteria co-infections, known as superinfections, are often detrimental and life threatening, data reported by the group of Ellen Foxman show that co-infected children are often asymptomatic. Whether the neutrophilic inflammation driven by pathobionts positively or negatively affects the anti-viral responses in children is another aspect that longitudinal sampling will help to understand in the future.

Overall, this work poses the base to better understand how frequent respiratory infections in children impact their immune response to pathogens or to mucosal vaccines.