Diverging patterns in innate immunity against respiratory viruses during a lifetime: lessons from the young and the old

Abstract

Respiratory viral infections frequently lead to severe respiratory disease, particularly in vulnerable populations such as young children, individuals with chronic lung conditions and older adults, resulting in hospitalisation and, in some cases, fatalities. The innate immune system plays a crucial role in monitoring for, and initiating responses to, viruses, maintaining a state of preparedness through the constant expression of antimicrobial defence molecules. Throughout the course of infection, innate immunity remains actively involved, contributing to viral clearance and damage control, with pivotal contributions from airway epithelial cells and resident and newly recruited immune cells. In instances where viral infections persist or are not effectively eliminated, innate immune components prominently contribute to the resulting pathophysiological consequences. Even though both young children and older adults are susceptible to severe respiratory disease caused by various respiratory viruses, the underlying mechanisms may differ significantly. Children face the challenge of developing and maturing their immunity, while older adults contend with issues such as immune senescence and inflammaging. This review aims to compare the innate immune responses in respiratory viral infections across both age groups, identifying common central hubs that could serve as promising targets for innovative therapeutic and preventive strategies, despite the apparent differences in underlying mechanisms.

Shareable abstract

Comparing innate immune responses to respiratory viral infections in different age groups identifies common central hubs that serve as targets for innovative therapeutic and preventive strategies, despite the apparent differences in underlying mechanisms. https://bit.ly/3xRmRrn

Introduction

Respiratory tract infections (RTIs), particularly in infants and older adults, pose a substantial risk for severe disease, but the underlying mechanisms can vary significantly. In the case of young children, the maturation of innate immunity is a critical factor, whereas older adults experience a decline in immune function along with other physiological impairments, such as reduced mucociliary clearance [1, 2], diminished elastic recoil of the lung, decreased respiratory muscle capacity and heightened aspiration [3].

Microbial exposure drives immune development in newborns

Early life is a period of dynamic change for the immune and respiratory systems, as both are undergoing a period of maturation. Alveolarisation of the lung continues after birth for the first 2 years of life [4]. For the immune system, this involves the development of immunological memory against pathogens, which may take several years [5]. Due to the post-natal development of innate antimicrobial defence pathways, which initially is heavily dependent on the activity of neutrophils, monocytes, dendritic cells (DCs) and epithelial barrier cells, the very young are more susceptible to respiratory pathogens [6], posing an increased risk of hospitalisation due to lower RTIs. Interestingly, abundant microbial exposure in early life seems to reduce this risk. Studies from areas rich in microbial exposure, e.g. living on a farm, suggest that this profoundly affects innate immune responses and poses a protective effect [7].

From birth onwards, the lungs and gut are seeded with microbes. Results from a longitudinal study of 100 children during the first 3 months of life [8], employing mass cytometry and proteomics on peripheral blood, suggested stereotypical immune maturation driven by the development of the gut microbiome. The relevance of early-life microbial exposure is further underlined in children born via caesarean section [9, 10] or those who received antibiotics in early life [11, 12], which were associated with an increased susceptibility to respiratory viruses and a heightened risk of developing childhood asthma. To protect the vulnerable mucosal surfaces in vital organs such as lungs and gut, maintaining immune balance and a state of tonic noninflammatory immune readiness is essential. Host–microbe interactions play a crucial role in establishing this balance [13].

Inflammaging and immunosenescence

As individuals age, the functionality of the immune system declines, a process known as immunosenescence [14–16]. Consequently, older people are at an increased risk for cancer and infectious diseases, alongside a suboptimal development of vaccine responses [17–19]. The molecular mechanisms that drive immunosenescence have been defined well in mice, but their validation in humans is less extensive [20, 21]. Immunosenescence is multi-faceted, affecting the function and phenotype of almost all immune populations. The primary lymphoid organs produce fewer (naïve) immune cells [22]. In addition, proliferating memory cells lead to strong clonal expansions and a reduced pool of cells remaining capable of responding to novel antigens [23, 24]. The accumulation of damage, metabolic changes and chronic exposure to viruses and bacteria translocating from the gut further contributes to immunosenescence. Many of these mechanisms converge through the process of inflammaging, which is the elevated presence of inflammatory cytokines in the body [25]. Many of the inflammatory cytokines found during inflammaging are produced by innate immune populations, such as monocytes, DCs and innate lymphocytes; however, adipocytes can also contribute [26]. There are profound differences in the composition, activation state and functional profile of the innate immune system with advanced age, many of which are expected to impact the response to RTIs [26–31].

Both infants and older adults exhibit a heightened susceptibility to RTIs and experience distinctive variations in innate infection control. This review examines the existing evidence concerning these immune deviations and their significance in viral control and the progression of disease. Ultimately, we explore potential strategies that target innate immune components to enhance the management of RTIs effectively.

Search strategy

This manuscript is structured as a narrative review. The published data in the area is nonhomogeneous and not conducive to meta-analysis. In addition, various high-quality reviews have been published on either respiratory viral infection, on immune development in neonates or immune senescence in elderly. Here, we focus on overlapping aspects in innate immunity against respiratory viral infection between neonates and elderly. To restrict our findings to the state of the art, the search period covers 2000–2023. To find relevant English-language publications in peer-reviewed journals, the following databases were interrogated: PubMed, Google, Scopus and Medline (Ovid). Search terms were variably deployed, depending on the area of focus in the current review. The search algorithm was based on combinations of the following medical subject headings (MeSH): (“Respiratory virus” OR SARS-CoV-2 OR COVID-19 OR “Respiratory Syncytial Virus” OR Rhinovirus OR Influenza) AND (“Innate Immunity’ OR “Trained Immunity” OR “Innate immune cells” OR Neutrophils OR Macrophages OR “Dendritic cells” OR “MAIT cells”) AND (lung OR nose OR Nasal OR “respiratory tract”) AND ((“older adults” OR elderly OR aged) OR (infants OR neonates OR pediatric OR Young)) AND ((resolution OR resolving OR clearance) OR (“lung microbiome”) OR (“Bronchial epithelial cells” OR “nasal epithelial cells”) OR (BCG OR “bacterial lysates” OR “microbial metabolites”) OR (Interferons OR “IFN-stimulated genes”)). This resulted in 864 papers, including 108 reviews, 33 clinical studies and various opinion pieces, of which 207 have been referenced in this review.

Immune surveillance and initial responses to respiratory viruses

Neutrophils in the respiratory tract

It is challenging to gain an insight into early immune events during the initiation of natural infections in humans, as this should be examined prior to or at least at the first signs of infection. Controlled human infection models are helpful in this regard as the infection initiation is known and planned. Several studies have investigated older adults in these models. For example, controlled infection of older adults with respiratory syncytial virus (RSV) led to a twice higher attack rate compared to young adults, with significantly higher viral titres [32]. The number of baseline neutrophils was related to infection susceptibility and was associated with inhibited tumour necrosis factor-α (TNF-α) and interleukin (IL)-17 responses early post-exposure [33]. Consequently, local inflammatory responses following exposure can be impaired by a pre-existing inflammaging state. In a mouse model, RSV infection in older mice also led to increased viral loads at day 5 [34]. Neutrophils were also found to be less abundant in the nasal mucosa of children compared to adults [35]. However, at an older age, the neutrophil to lymphocyte ratio increases both in blood and the nasal mucosa [35, 36], even in the absence of disease [37]. This might relate to an increased susceptibility to severe viral infections in older individuals. In contrast to older mice, neonatal mice showed a relatively low neutrophilic infiltration in the lungs early after RSV infection, while this was clearly more enhanced in neonatal mice with influenza infection [38, 39]. However, neonatal neutrophils showed poor chemotaxis, endothelial adhesion, phagocytic function and neutrophil extracellular trap (NET) formation in vitro, suggesting that they may not act as effectively as neutrophils from adults [40–42]. However, a recent study [43] showed a greater migration of cord blood neutrophils across RSV-infected epithelial cells compared to adult neutrophils in an in vitro culture model. At baseline, cord blood neutrophils had a higher expression of CD11b, CD64, neutrophil elastase and myeloperoxidase, suggesting that infant neutrophils are not fully impaired in their function, but may respond different to virus-infected epithelium. Some studies have suggested an enhanced viability of migrated neutrophils from infants (e.g. in bronchoalveolar lavage (BAL) [44]), while others report enhanced apoptosis [45, 46]. As these processes will affect protective mechanisms against the virus by clearing obsolete cells and curbing ongoing inflammation, changes in or failure of these mechanisms may affect disease severity and resolution. Therefore, while in neonates the behaviour of neutrophils in the lung may be different from those in adults, in older adults their frequency is much higher, which is strongly associated with an unfavourable outcome (figure 1).

FIGURE 1.

Innate responses during viral infections in young and older individuals. Impairment or deviations in innate immune cell numbers and/or functionalities during different stages of viral infection: 1) during immune surveillance and initiation of infection, 2) viral clearance and damage control, and 3) disease effects with uncontrolled and persistent viral infections in both very young children and older adults. Ab: antibody; IFN: interferon; MAIT: mucosal-associated invariant T; MF: macrophage; NK: natural killer; pDC: plasmacytoid dendritic cell.

Interferons (IFNs) during early infection

An important component of innate immunity in the respiratory tract is the IFN response. This consists of type I, II and III IFNs (IFN-I, IFN-II and IFN-III). IFN-I (IFN-α/IFN-β) is an antiviral cytokine that binds to ubiquitously expressed INF-α/β receptors and induces the expression of hundreds of IFN-stimulated genes (ISGs) that impair viral replication in infected cells and protect the lung from further spreading of viruses [47]. However, IFN-I is also central to inflammatory responses, by inducing recruitment and activation of immune cells to the lungs. This helps to control virus burden but can also cause detrimental immunopathology and contribute to disease severity. IFN-III [48, 49] or IFN-λ acts on similar genes to IFN-I, but it primarily targets lung epithelial cells and protects them against respiratory viruses. Indeed, IFNs could inhibit severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza infection and replication in primary human airway epithelial cells when given before infection [50, 51]. Type II IFN refers to IFN-γ, which is primarily produced by immune cells, such as T- and natural killer (NK) cells. It has antiviral activity and promotes the activation of macrophages and other immune cells [52].

At birth, IFN responses are low and still need to develop. In a human cohort followed prospectively, IFN-α production by DCs increased to adult levels by 1 year of age [53]. In addition, in response to RSV infection, IFN-α production in plasmacytoid DCs (pDCs) from cord blood or peripheral blood mononuclear cells (PBMCs) from children (<1 year) were lower compared to adults [54]. Similar findings were described in RSV-infected neonatal mice, where recruitment of pDCs and production of IFN-I was strongly reduced [55]. Interestingly, deletion of insulin-responsive aminopeptidase (IRAP), a protein involved in regulation of IFN-I production, substantially enhanced IFN-I in neonatal alveolar macrophages, but not in adult ones. Neonatal IRAP-deficient mice infected with RSV cleared the virus more efficiently compared to wild-type mice [56]. This suggests that IRAP is an age-dependent regulator of IFN-I expression and influences protection and immunity to respiratory viruses.

During older age, monocytes/macrophages and conventional DCs (cDCs) showed reduced IFN-I production [57]. In particular, pDCs decline, produce fewer IFN-I and IFN-III responses to influenza virus, and are present at lower frequencies at older age [28, 58]. In contrast to these human data, viral load and IFN-I responses were not different between young and aged mice in the early stages following influenza virus infection [59]. However, this likely depends on the viral agent. A mouse-adapted SARS-CoV-2 strain showed that increased early viral load in aged mice led to death and was linked to impaired IFN responses [60]. Indeed, daily sampling of SARS-CoV-2 infection followed by mathematical modelling indicated that, with increasing age, innate immune responses are less effective in rendering target cells refractory to infection [61]. There was no association between age and viral clearance or growth rate. Early IFN responses in the lung seem key to control viral infection and replication.

Respiratory microbiome

Various bacterial communities are associated with an increased risk of severe RTIs in young children [62, 63]. Nasopharyngeal microbiota in healthy infants showed the biggest dynamics in gene expression profiles during the first days of life, which included Toll-like receptor (TLR) and inflammasome signalling, representing the first colonisation of airway microbiota [64]. This was accompanied by a gradual increase in regulatory responses, such as IL-10. Asymptomatic viral infections were associated with increased IFN-I and IFN-γ expression and preceded the development of risky respiratory microbiota (e.g. Moraxella and Haemophilus species) and increased susceptibility to severe RTI [64]. Indeed, a causal link between respiratory viral infections, IFN-I and changes in microbiota was demonstrated in influenza-infected mice showing that nasal microbiota was modified in an IFN-dependent manner, leading to staphylococcal persistence [65]. In contrast, asymptomatic viral infections were associated with increased IFN signalling in upper respiratory tract and protected against live attenuated influenza vaccine replication in a cohort of children from the Gambia, indicating that asymptomatic viral infections can also protect against secondary incoming viruses [66]. Influenza-like illness was also found to change the nasal microbiota of frail older adults in long-term care homes [67]. However, although gut microbiota was associated with immunosenescence in older adults and the respiratory tract microbiota was shown to change with advancing age, the effect of respiratory tract microbiota on viral infections remains poorly defined in this group [68, 69]. Interestingly, commensal bacteria can also stimulate IFN-I/III expression in respiratory epithelial cells, leading to enhanced antiviral resistance for influenza-A virus [70], as shown by antibiotic treatment and faecal transplantation in mice [71]. Moreover, Streptococcus pneumoniae colonisation was able to dampen nasal inflammatory responses to live attenuated influenza virus in a controlled human co-infection model [72].

These studies suggest that bidirectional interactions between colonised bacteria, asymptomatic viral infections and epithelium influence early IFN-I levels and affect the course of viral RTIs and the composition of microbiota. The timing of the first IFN wave and coincident induction of regulatory responses may be crucial for the clinical outcomes of viral RTIs.

Lessons from the coronavirus disease 2019 (COVID-19) pandemic

Further insights into age-related differences in innate antiviral immunity have emerged from the COVID-19 pandemic. Age is the strongest risk outcome for severe SARS-CoV-2 infections; children have a substantially lower risk of developing severe COVID-19. Of note, approx. 15% of the severe and critically ill COVID-19 patients can be explained by either inborn errors of type 1 IFNs or pre-existing auto-antibodies neutralising IFNs [73–75]. Large-scale epidemiological studies have yielded contradictory results on the effect of old age on SARS-CoV-2 loads [76, 77]. SARS-CoV infection in aged African green monkeys was associated with increased viral load at day 1 post-infection, but not at later time-points [78]. Viral replication of SARS-CoV was similar in aged and young macaques [79].

To understand why SARS-CoV-2 viral load is better controlled in children [77], several single-cell transcriptomics studies were performed comparing airway mucosa and peripheral blood from children and adults. Airway epithelium of children had a higher steady-state expression of IFN-response genes, which may restrict viral replication in children [80]. Indeed, IFN-λ was also upregulated in nasal swabs from infected children, while other IFNs and antimicrobial genes were unaffected, contrasting with findings in infected adults [81]. Children also showed higher basal expression of viral pattern recognition receptors (PRRs) (i.e. melanoma differentiation-associated protein 5 (MDA5)/retinoic acid-inducible gene I (RIG-I)) in epithelial and immune cells, resulting in stronger innate antiviral responses compared adults [82]. Whether this is a consequence of more recent infections in children (and/or changes in microbiota composition which may induce IFNs) remains to be established. Furthermore, nasal populations such as killer cell lectin like receptor C1 (NKG2A)⁺ cytotoxic T-cells and CD8⁺ T-cells with a memory phenotype were found in children, suggesting that children mounted stronger early antiviral responses to SARS-CoV-2 infection compared to adults [82]. In contrast, adults displayed a higher cytotoxic immune compartment in peripheral blood, probably because of viral spreading and containment difficulties. This also was accompanied with ISGs in multiple immune cells, adding an inflammatory state to the already cytotoxic immune response and amplifying the pathological effects of systemic immune responses [80]. These studies indicate that a higher level of infection readiness locally in (nasal) mucosal tissues, but without a systemic inflammation-inducing component, is beneficial to curb viral infections early and effectively prevents spreading to lower airways. Notably, higher IFN-III expression in patients with COVID-19 correlated with lower viral load in bronchial aspirates and faster viral clearance. In addition, a higher IFN-III/IFN-I ratio correlated with an improved outcome for critically ill patients [83].

In summary, depending on the virus and host bacterial species, baseline differences in neutrophilic inflammatory status and IFN responses are related to an increased susceptibility to infection (figure 1).

Viral clearance and damage control

Respiratory epithelium: an important gatekeeper

Pathogen clearance and infection resolution in the lung is a delicate and multifaceted process, orchestrated by innate immune cells and respiratory tract cells. Even though an effective immune response is needed to eliminate the pathogen, exaggerated and/or prolonged immune responses can damage the lung tissue and should be avoided. The respiratory epithelium is an important player in the defence against viral infections [84]. Epithelial cells express PRRs and cytokine receptors, produce antimicrobial proteins and mucins, as well as produce IFN responses to viruses [84]. Furthermore, interactions between resident airway macrophages and epithelial cells drive local homeostasis and immune tolerance to harmless stimuli, support antimicrobial responses to dangerous invaders and drive tissue repair following pathogen clearance [85].

IFN-III is considered an important player in immune homeostasis at mucosal sites, where it stimulates pathogen clearance while curbing inflammation to maintain barrier integrity [86]. The activity of IFN-III is restricted by the selective expression of IFN lambda receptor-1 (IFNLR1) on respiratory epithelial cells [87] and certain immune cells [88]. For example, IFN-λ signalling in DCs was crucial for optimal antiviral CD8 T-cell responses as well as an enhanced IL-10 regulatory network in murine influenza virus infections [89]. Importantly, IFN-III was able to restrict the virus within the upper respiratory tract and prevent it from spreading to the lungs [90].

In response to rhinovirus (RV)-C, a subset of ciliated epithelial cells was identified with minimal IFN production despite high levels of virus replication, while other subsets of ciliated cells produced high IFN levels with moderate viral replication. This suggests that ciliated airway epithelial cell composition and IFN responses of infected and uninfected cells influence the risk of more severe viral RTIs, possibly explaining differences found in young versus older individuals [91]. Indeed, bronchial epithelial cultures from paediatric rhesus monkeys showed more viral replication and less IFN-I production in response to influenza (H1N1) virus compared to adult ones. Infected infant rhesus monkeys showed high viral titres on day 1 with prolonged viral presence, as well as prolonged bronchiolitis and alveolitis despite viral clearance [92]. Interestingly, tissues samples collected from children with nonsymptomatic SARS-CoV-2 infections produced IFN-λ faster in vitro than the ones from symptomatic infected children [45], suggesting that an instant readiness of IFN-III may help to protect against viral infections such as SARS-CoV-2. These contrasting data suggest that differential epithelial responses to viruses may explain different disease outcomes in children. In older adults, however, primary nasal epithelial cells cultures infected with influenza virus showed a stronger inflammatory response, but reduced antigen presentation at similar viral loads compared to those from young adults [93]. Single-cell RNA sequencing from older adults infected with SARS-CoV-2 also suggested that interactions between epithelial cells and immune responses were altered, with increased epithelial-derived transforming growth factor-β (TGF-β)/macrophage TGF-β receptor crosstalk and reduced IFN-γ interaction [94]. Epithelium-derived factors, including hyaluron, also drive murine alveolar macrophage dysfunction with age, reducing the ability to respond to granulocyte−macrophage colony-stimulating factor, as shown by adoptive cell transfer and parabiont experiments [95]. In addition, for RSV infection, differences were shown in viral kinetics with a slower viral clearance and enhanced inflammatory responses in epithelial cell cultures from older individuals [96]. Interestingly, similar findings were noted in nasal epithelial cells infected with RV in young children, showing higher viral titres and lower host immune responses [97]. However, no differences were reported for SARS-CoV2 viral titres or immune factors when comparing paediatric and adult bronchial epithelial cells [98]. However, more detailed Single-cell RNA sequencing analysis of the effects of SARS-CoV-2 infection on the differentiation of human nasal epithelial cells in children (<10 year), adults and elderly people (>70 year) revealed age-specific and epithelial cell-specific responses. Notably, there was a strong IFN response in infected paediatric goblet inflammatory cells, reducing viral spread and the appearance of elderly basaloid-like cells (ITGB6^hi progenitors) that sustain viral replication and are associated with fibrotic signalling pathways [99]. The studies above in young and older subjects show that epithelial perturbations in protective antiviral inflammatory responses affect their capacity to quickly clear viruses, as well as modulate immune responses affecting clearance and resolution of virus, and pose a threat to the barrier integrity of the lung.

Delayed inflammatory response

Aging is linked to a decreased control and clearance of influenza infections. For example, aged mice showed reduced influenza titres in their lungs compared to younger mice, although the aged mice had prolonged shedding, indicating reduced clearance capacity [100]. Delayed influenza virus clearance was also observed in aged compared to younger macaques in both nasal swabs and BAL fluid [101]. Influenza clearance was also reduced in individuals >65 years in a hospital setting [102]. There are multiple mechanistic studies underway that aim to understand local responses underlying this phenomenon. Older mice showed reduced inflammasome activation in DCs and overcoming this by using nigericin treatment or performing adoptive transfer of DCs from young to older mice improved survival and increased immune infiltration post-infection [103]. Another study found a delayed infiltration of immune cells (cDC2, NK and granulocytes) in the lungs of aged mice following influenza infection, together with delayed cytokine responses [100]. Finally, in aged macaques, higher viral loads in the lungs correlated with increased IFN responses and innate populations [101]. Thus, a weaker initial control could lead to higher and prolonged viral load and then to even higher innate responses.

In a cohort of naturally influenza-infected individuals, similar viral loads in adults and children were observed; however, children showed higher innate responses in nasal fluid. When adjusting for age and viral load, increased nasal monocyte chemoattractant protein-3 (MCP-3), IFN-α and plasma IL-10 at enrolment predicted progression towards severe disease, while increased plasma IL-10, MCP-3 and IL-6 predicted hospitalisation; this correlated with monocyte recruitment to the site of infection [104]. Thus, local enhanced innate responses, due to uncontrolled viral infection and high-level viral replication, contribute to disease progression irrespective of age.

NK cell functionality

Another key mechanism of early viral control is NK-cell recruitment to the respiratory tract [105]. In severely ill RSV-infected children, a highly active NK-cell subset (CD94^-) was detected in peripheral blood, while pro-inflammatory plasma cytokines were lower than in healthy children [106]. Neonatal NK cells show diminished cytotoxicity (i.e. reduced perforin secretion), compared to NK cells from adults in co-culture with RSV-infected cells, despite similar upregulation of expression of CD107 [107]. When explanted lungs were infected with influenza virus, the lung-resident CD56^brightCD49a⁺ NK-cell population showed high CD107 upregulation, indicating responsivity. The NK cells provided significant antiviral and cytotoxic activity following contact with influenza-infected cells (i.e. IFN-γ and granzyme-B release) resulting in cell death [108]. Intranasal delivery of IFN-γ to RSV-infected neonatal mice improved RSV clearance and lung inflammation and was dependent on activation of alveolar macrophages [109]. Following influenza vaccination of older adults, the ability of antibodies to induce NK-cell activation was correlated with protection from later infection. Senescent peripheral blood NK cells, as measured by CD57 expression, were less able to be activated by influenza vaccine-induced antibodies [110]. NK-cell numbers in the lungs of aged mice at steady state and their function (degranulation and IFN-γ production) following influenza infection were decreased [111]. In contrast, lung NK cells were not decreased in a human cohort study; however, relatively few people >80 years were analysed [112]. Most lung NK cells were CD57⁺, which was correlated with age [112]. Thus, alterations in NK-cell numbers, maturity and/or functionality in very young and in older individuals might contribute to reduced clearance of viral pathogens and resolution of inflammation (figure 1).

Disease effects of respiratory viral infections

Secondary bacterial infections

Uncontrolled viral infections can progress towards pneumonia, with or without secondary bacterial infections. These secondary bacterial infections are often the cause of death; a clear example is provided by the “Spanish flu” [113]. Pneumonia is usually related to increased neutrophils in the lungs, whereby neutrophils are required for initial control of bacterial and fungal pathogens, but sustained recruitment can lead to tissue damage [114]. Following severe COVID-19, granulocytes were also found in the nasal mucosa and were negatively associated with hypoxia. A subset of granulocytes was elevated for at least 6 weeks after hospital discharge [115]. Respiratory viral infections also impair neutrophilic function. For example, even though more neutrophils were recruited to the lungs of older adults during influenza infection, they showed reduced phagocytic capacity [114]. Neutrophilic dysfunction enhanced the susceptibility to secondary infection with S. pneumoniae [116]. The reduced innate control of this bacterium was associated with virus-induced TNF-α, which inhibits neutrophilic killing [117]. Aged alveolar epithelial cells in mice secrete more neutrophil-attracting chemokines during influenza infection and this increased neutrophilic recruitment underlies their susceptibility to influenza mortality [118]. This is likely related to outcome following infection as older adults that had a high level of inflammaging had an increased risk of developing community-acquired pneumonia [119]. In addition, in severely ill RSV-infected children, neutrophils are the predominant cell type in the lung lavage [120]. When neutrophils were co-cultured with aspirate fluid from children with viral/bacterial co-infections, a decreased respiratory burst and killing activity was found against Haemophilus influenzae and Staphylococcus aureus [121]. Furthermore, in the blood and lung lavage of RSV-infected children with a bacterial co-infection, increased suppressive neutrophils (CD16^hiCD62L^lo) were found [122]. These data suggest that neutrophils are similarly increased in severely ill young and older individuals and linked to mortality, while their functionally seems reduced.

Disease resolution

The removal of apoptotic neutrophils from the lungs is mediated by macrophages through efferocytosis, which was impaired in older adults due to reduced Tim-4 expression and could be reversed by p38 mitogen-activated protein kinase (MAPK) inhibition [123]. Adoptive transfer of alveolar macrophages from young into old mice reduced lung damage following influenza infection [124]. Another mediator of resolution of inflammation are specialised pro-resolving lipid mediators (SPMs), which are produced by macrophages, and which reduced inflammation and damage during viral and bacterial lung infections [125]. These SPMs were reduced in homeostasis and following zymosan challenge in aged mice, which led to increased neutrophil influx and delayed resolution of inflammation [126]. The resolution of inflammation measured in circulation during pneumonia was also reduced in older adults [127].

An impaired IFN response with aging or in the very young might also be related to increased inflammation and damage, irrespective of its effects on viral control. IFN-I can reduce the excessive recruitment of neutrophils to the lung, potentially via modifying monocyte-derived cytokines [57]. IFN-α production by interstitial macrophages was able to suppress in situ inflammatory monocyte proliferation during influenza virus infection and confer disease tolerance [128]. IFN-I also induced neutrophil apoptosis, efferocytosis and induction of anti-inflammatory cell types, such as “M2 macrophages” and myeloid-derived suppressor cells in later stages of infection [129–132]. Indeed, individuals with severe COVID-19 showed reduced IFN-α and ISG expression in blood and reduced ISG responses in nasal mucosa [133, 134]. In contrast, children showed faster resolving responses, possibly due to robust and early IFN responses to SARS-CoV-2 [135]. Individuals with severe influenza infection showed a similar reduced IFN-I response accompanied with a progression to neutrophilic-associated patterns [136]. In neonatal IFNLR1-deficient mice, infection with pneumonia virus of mice (rodent equivalent of RSV), induced early and pronounced neutrophilia, with enhanced reactive oxygen species production, NETosis and mucus in the airways [137]. This indicates that enhanced IFN-III signalling in neutrophils will diminish their functionality, dampening lung inflammation and damage.

Mucosal associated invariant T (MAIT) cells are an abundant and evolutionarily conserved class of innate-like lymphocytes [138], which comprise up to 10% of human peripheral blood and respiratory mucosal T-cells [139]. Circulating MAIT cells are low in newborns and increase with age into adulthood, but then decrease again in older adults [140]. MAIT cells recognise microbial-derived metabolites derived from the riboflavin biosynthesis pathway, presented by nonclassical MHC-related protein 1 (MR1). Pulmonary MAIT cells have shown some protection from lethal influenza infection by preventing excessive immune infiltration, while not affecting viral load [141]. Of note, this was independent of MR1 sensing and dependent on IFN-γ production, as IFN-γ-deficient MAIT cells showed no protective effect in adoptive transfer. Monocytic IL-18 production following influenza-infected A549 coculture triggered MAIT cell activation [142]. Peripheral MAIT cell numbers were found to be lower in people who succumbed to H7N9 influenza infection and severe COVID-19 [142, 143]. These cells likely homed to the airways as highly activated MAIT cells and invariant NK T-cells were found in the lungs of individuals with COVID-19 where they were associated with an improved clinical outcome [143]. Peripheral MAIT cells are reduced in older adults; however, mucosal numbers still need to be assessed in older adults [144]. MAIT cells are responsive to signals from local microbiota in early life and may have longer lasting consequences for tissue repair and antiviral immunity [145]. Although MAIT cells are abundantly present in the lungs of adults, their presence and role in viral infections in the paediatric lung remains an open question [139, 146].

Lung function decline and chronic inflammatory lung diseases

Viral RTIs can have prolonged effects, persisting long after clearance and resolution of infection. A pooled analysis of five birth cohorts across Europe showed that children with early-life lower RTIs had significantly lower lung function and a higher risk of developing asthma [147]. Furthermore, reduced lung function gain during childhood into adulthood also increased the risk of developing COPD later in life [148]. Two of the most common viral infections in the first year of life include RSV and RV, both associated with the development of recurrent wheeze and childhood asthma [149, 150]. RSV prophylaxis in early life did reduce the risk of developing asthma; however, this was not significant in a meta-analysis of eight combined studies [149]. Possibly a more prolonged delay or avoidance of early life infection is necessary to reach protection against asthma secondary to RSV infection [151]. Prophylaxis for RV is challenging, since >170 different species exist, of which multiple circulate simultaneously [152]. Interestingly, even asymptomatic RV infections in the first year were associated with increased susceptibility of recurrent RTIs [153]. Although viruses, including RSV and RVs, typically induce type 1 responses, various mouse studies have shown that type 2 (inducing) cytokines and allergic sensitisation occur following infection with these viruses in neonatal models. For example, RSV induced TSLP production in epithelial cells and type 2 innate lymphoid cells (ILC2) accumulation in the lung [154]. Recurrent infections in neonatal mice boosted epithelial damage, further potentiating the development of local type 2 immunity [155]. In children hospitalised with RSV, increased levels of thymic stromal lymphopoietin (TSLP), IL-33 and IL-13 were found in nasal aspirates [156]. RV infection with neonatal mice showed increased type 2 inflammatory cytokines such as IL-25, derived from airway epithelial cells, and IL-13, produced by ILC2, leading to mucus hypersecretion and airway hyperresponsiveness [157]. These effects could be blocked by a neutralising antibody against IL-25 [158].

Thus, impaired resolution of inflammation and efferocytosis by alveolar macrophages, altered trafficking of neutrophils and reduced IFNs contribute to prolonged inflammatory signals and pathology in the lung of older adults following infection. In young children, type 2 inflammatory signals and cells are also boosted, increasing the risk of developing chronic inflammatory diseases such as asthma (figure 1).

Innovative solutions to boost effective innate responses

With the clear involvement of innate responses in susceptibility to infections, several innovative solutions have emerged to boost innate immunity in the setting of RTIs, either as a vaccine or immunotherapy (figure 2).

FIGURE 2.

Innovative therapeutic approaches targeting innate immunity during respiratory tract infections (RTIs). Several strategies have evolved targeting reduced interferon (IFN) production, innate mucosal immune populations, trained innate immunity and excessive inflammation during RTIs. BCG: bacille Calmette–Guérin; MF: macrophage; NK: natural killer.

Immunological agents

While reduced efferocytosis is associated with pneumonia, one study found that patients who were taking statins, a commonly prescribed medication for atherosclerosis and coronary heart diseases, showed increased efferocytosis following pneumonia [159]. This suggests that alveolar macrophage-mediated clearance of apoptotic neutrophils can be improved with a relatively simple solution.

IFN-III drives strong antiviral activity against various respiratory viruses in mice but also helps to curb unwanted and prolonged local inflammation [160]. Pegylated-IFN-λ has been given therapeutically to mice both before and after influenza infection. It was able to prevent infection and reduce inflammation [88], i.e. by enhancing disease resolution [161] through alveolar macrophages [162], and by enhancing humoral and adaptive immune responses when given as adjuvant during vaccination [163]. Similar observations were made in mice with SARS-CoV-2, when IFN-λ was given prophylactically or therapeutically via the nose [164]. During the COVID-19 pandemic, clinical trials with pegylated-IFN-λ showed a more rapid decline of SARS-CoV-2 transcripts over time [165], as well as a significantly reduced risk of hospitalisation [166]. Analysis of immune cells expressing IFNLR1 showed enhanced ISGs, while the development of specific antibodies and virus-specific T-cells was not hampered. The authors observed a delayed T-cell immunity development in older adults compared to young adults, while the decline of viral transcripts was equal, suggesting that pegylated-IFN-λ can overcome delays in adaptive immunity to accelerate viral clearance in older adults [167].

Vaccines and microbial products

Intranasal and, especially, live-attenuated vaccines might also induce innate immunity. For example, live-attenuated influenza vaccination of young adults induced not only influenza-specific CD4 and CD8 T-cells in the lung, but also increased IFN-γ-producing T-cell receptor-γδ T-cells upon in vitro restimulation with influenza vaccine [72].

One of the best-known examples of a vaccine that can induce innate responses is the bacille Calmette–Guérin (BCG) vaccine. Understanding its effects on innate immunity led to a new field, called trained innate immunity [168]. In mouse models, BCG vaccination led to ISG expression in lung myeloid and epithelial cells, secondary to CD4 T-cell-derived IFN-γ. This conferred protection against SARS-CoV-2 infection [169]. A double-blind randomised clinical trial vaccinating older adults with BCG or placebo upon hospital discharge following COVID-19 infection found a protective effect of BCG vaccine against re-infection [170]. However, a placebo-controlled clinical trial that tested BCG administration in the absence of infection did not find any protective effect on the cumulative incidence of RTIs during a year follow-up. Although it improved cytokine responses and antibody induction following infection, this suggests that timing of administration might be key in conferring protection [171]. BCG was also tested against RSV infection in neonatal calves and in mice. Administration of recombinant BCG-expressing RSV nucleoprotein reduced clinical disease upon RSV challenge, although viral burden was unaffected in calves but reduced in mice. The vaccination strategy induced IgA and virus-neutralisation activity as well as T-cell immunity in both animals [172, 173].

Orally administered bacterial lysates of mixed respiratory pathogenic bacteria have been used for decades to prevent RTIs [174] in children [175, 176] and in vulnerable adults [177–179]. Infection incidence, duration and severity were studied in randomised clinical trials. Wheezing episodes and/or asthma exacerbations in children were reduced [180]. In adults with chronic bronchitis and COPD, bacterial lysates reduced exacerbations triggered by RTIs [178, 179]. The underlying mechanism is only partly understood. Initially it was sought to act as an oral “vaccine”, mounting humoral and adaptive immunity against specific respiratory pathogens. However, recently, it was suggested that reduced frequency and duration in RTIs in children [181] was linked to innate immune training. Lipopolysaccharide (LPS) stimulation of PBMCs showed reduced production of pro-inflammatory cytokines (TNF-α, IL-6 and CCL3) following treatment. Pathway analysis and differentially gene correlation analysis on RNA-transcriptomics showed 1) rewiring of immune responses following LPS stimulation resulting in increased coordination of TLR4 with IFN signalling pathways and 2) segregation of TNF-α and IFN-γ in separate expression modules in the treatment group – two cytokines that were collectively shown to exaggerate inflammatory sequelae in the respiratory tract [182, 183]. Uncoupling these processes would facilitate swift innate responses to pathogens, but without the risk of uncontrolled inflammatory pathways. A similar innate immune training was suggested in studies with mice [184, 185].

In one murine study, the MAIT ligand 5-OP-RU was administered together with intraperitoneal influenza, which led to MAIT cell proliferation and accumulation in multiple tissues, including the lungs of both young and old mice [186]. This was associated with heterosubtypic protection from subsequent influenza A virus challenge and similar MAIT cell induction was seen when 5-OP-RU was administered together with an intranasal influenza vaccine or an intramuscular recombinant vesicular stomatitis virus-based SARS-CoV-2 vaccine.

Gut microbial metabolites

The gut microbiota seem crucial for defence against respiratory viruses. Antibiotic-treated mice displayed impaired defence against influenza infection [187], which was dependent on microbiota-driven tonic IFN signals in lung epithelium [188]. Microbiota can act at distal sites, since metabolites, such as short-chain fatty acids (SCFAs) produced in the gut, can enter the circulation and influence processes the lung; this is called the gut–lung axis [189]. Indeed, SCFAs can affect viral infections in the airways, e.g. a fibre-rich diet increased specific monocyte precursors, leading to more anti-inflammatory and tissue-protective macrophages in the lungs through propionate [190], as well as enhanced, lung cytotoxic T-cell metabolism, instrumental against experimental influenza infections [191]. Both oral and nasal administration of acetate reduced viral load and pulmonary inflammation in RSV-infected mice and induced antiviral responses in cells from children with RSV bronchiolitis [192, 193]. In addition, the flavonoid-derived metabolite desaminotyrosine reduced lung pathology in antibiotic-treated mice prior to influenza infection [194]. Several studies point towards roles for IFNs, intracellular PRRs (e.g. RIG-I) and reduced host defence causing pathology and tissue damage [191, 193]. SCFAs have been tested in patients, for example to improve metabolic (obesity and diabetes) or colonic health. However, systemic immune effects showed mixed results. A lower inflammatory status of circulating monocytes following oral ethyl-butyrate supplementation in obese individuals was noted [195], while in patients with long-standing type 1 diabetes, innate peripheral blood immune responses were unchanged [196]. Several studies used high-fibre diets in healthy subjects, older people or pre-diabetics to increase butyrate production. Decreased circulating inflammatory cytokines (IL-1β, IL-6, TNFα), as well as increased IL-10 (in healthy subjects) were found [197–199]. However, there were also studies in which fibre-rich diets did not affect circulating inflammatory markers [200]. Here, elevated gut SCFA levels did not always translate into elevated systemic SCFA levels, in particularly in patients. This may be related to their inflammatory status, as inflammatory cytokines, such as TNF-α, reduce butyrate uptake from the gut [201].

Thus, several immune or microbial agents can be recognised to act on various aspects of innate immunity, i.e. boosting IFN responses and/or innate mucosal immune populations, training innate immunity and/or dampening excessive inflammation, and harbour potential as host-directed therapies.

Summarising discussion

In this review, our focus is on innate responses throughout various stages of viral infections. We acknowledge that these responses are complemented by adaptive cellular and humoral immune components, each with specific shortcomings in young infants and older adults, as well as by adaptive immune cell subsets that employ innate functions, such as neonatal regulatory B-cells [202]. However, detailed discussions on these aspects are available elsewhere [203, 204]. Impaired innate responses are closely associated with the loss of viral control, significantly contributing to exaggerated and harmful immune responses during severe RTIs and may offer interesting possibilities for new treatment options. Additionally, optimising innate defences at mucosal surfaces as a host-directed prevention strategy holds great promise. A comparison between the young and the old reveals several central innate hubs.

During immune surveillance at mucosal surfaces and at the initiation of viral responses, neutrophils take centre stage. In the young, neutrophils in the lung exhibit a different behaviour to viruses, while in older adults, baseline levels are elevated. Although neutrophils are not the primary defenders against viral infections, their presence and functionality are an important factor in early infections and linked to severe disease. Tonic IFN production serves as a guarding system at mucosal surfaces. In both young children and older adults, IFN-I (predominantly pDC-derived) is reduced, whereas IFN-III is more and mostly reduced during aging. Importantly, IFN-III is instrumental in fighting the virus, but without exaggerated and damaging inflammatory responses. This divergence, with children expressing more IFN-III, could partially explain why they experience less severe SARS-CoV-2 infections than adults. In children, the lung microbiota may contribute to better or less mucosal immune fitness, while in older adults their role in worse viral infections is less clear.

The persistence of viral presence at mucosal surfaces affects both children and older adults. The epithelium, either or not infected with the virus, may play an important role in clearing the infection and curbing the downstream inflammation. In children, continued viral presence may be associated with reduced IFN-I production, while in older adults, the picture is more complex. Next to reduced IFN-I, factors include diminished antigen presentation capacity by epithelial cells, distorted interaction with macrophages, reduced immune cell recruitment, reduced inflammasome activity and reduced IFN-γ levels. An additional contributing factor is attributed to NK cells. In young children, NK cells exhibit lower cytotoxicity, whereas in older adults, their antibody-dependent functionality is diminished. Overall, a picture is emerging of local enhanced innate responses, due to uncontrolled viral infection and viral replication, in which disease progression occurs irrespective of age.

Uncontrolled infections progress to severe disease with potential long-term consequences. While the risk of secondary bacterial infections is not age-dependent, it is linked to severe and uncontrolled viral infections. In children, prolonged viral presence is associated with type 2 alarmin production, inducing ILC2 and other type 2 immune cells, leading to an increased risk of wheeze, childhood asthma and decreased lung function. In older adults, impaired immune components important for resolving inflammation, such as IFNs, macrophages, neutrophil clearance and pro-resolving lipids, may contribute to virus-induced pneumonia and prolonged disease.

Various strategies have been proposed to address these impairments, which were tested in preclinical models and in humans. These approaches include pegylated-IFN-λ, bacterial lysates, BCG vaccination, intranasal vaccination, MR-1 ligands and microbial metabolites. These aim to enhance IFN responses, induce innate trained immunity, boost mucosal immune populations and reduce excessive inflammation. Additionally, biologicals such as anti-IL-6 or anti-TNF may help mitigate cytokine storm development [205, 206]. The effectiveness and necessity of these approaches in different age groups, particularly in young children versus older adults, requires further investigation. Nonetheless, these strategies offer unique opportunities to bolster pandemic preparedness and protect vulnerable populations.