3.4 Interleukin-25 (IL-25)
IL-25 is a member of the interleukin-17 cytokine family that recognizes a receptor composed of IL-17RB and IL-17RA subunits [58]. IL-17RB receptor is mainly expressed on ILC2, and its activation is involved in type 2 effector response [58-61]. In C57BL/6J mice, exposure to RSV induced the expression of IL-25 and IL-17RB lung transcripts and of other potentially pathogenic cytokines, including IL-13 [61]. Treatment with an anti-IL-25 antibody significantly reduced Th2 cytokine production, mucus-associated gob5  gene expression and airway hyperresponsiveness. Moreover, IL-17RB−/− mice showed increased clearance of the virus and diminished pathology [61]. In a different set of experiments, IL-17RB−/− mice were sensitized to allergen, infected with RSV during the active allergic responses, and then challenged with allergen [62]. As compared to wild-type mice, decreased inflammatory response, cytokine production and IL-13 and gob5 gene expression was detected in IL-17RB−/− mice, possibly reflecting enhanced clearance of RSV, leading to decreased immune activation [61]. Immune and morphological responses to RSV infection were also evaluated in wild-type and NK cell-depleted BALB/c mice [62]. NK cells play a critical role in the development of an effector immune response in RSV infection [63]. Depletion of NK cells led to increased IL-25 expression in the respiratory epithelium, higher IL-4, IL-13 and eotaxin lung mRNA levels and higher serum IgE, tissue eosinophil numbers and mucus-secreting cells. This increased Th2 pathology was reflected in a delayed viral clearance in the later stages of infection [61]. RSV-induced Th2 responses were IL-25 dependent. Treatment of NK-depleted mice with anti-IL-25 antibodies led to attenuation of the Th2 cell responses, suppression of inflammation and histopathological changes in the lungs [62].
4. GUT MICROBIOTA, RSV INFECTION, ALARMINS AND ILC2
Gut microbiota composition might affect the severity of respiratory virus infections, but the interaction can be bidirectional, since infections can induce gut dysbiosis [64,65]. The high concentration of microorganism- and pathogen-associated molecular patterns physiologically present in the gut and the huge amounts of cytokines and chemokines produced during dysbiosis can activate local ILC that can then migrate to other sites of the body [19-21]. In an integrated microbiota dysbiosis mice model, it was demonstrated that gut microbiota can modulate ILC2 directional migration to the lung via regulation of select cytokines [66]. In these mice, Proteobacteriaabundance was associated with increased IL-33 production which promoted ILC2 migration and accumulation in the lung. Blocking the IL-33 receptor with anti-ST2 antibodies, abolished the observed increased percentages of lung ILC2 in this animal model [65]. Moreover, tissue and circulating ILC2 can recognize microbial ligands through their TLR, and directly produce a variety of cytokines, chemokines which can fight or promote infections [19-21, 67-70]. In stool samples collected from infants hospitalized during a bronchiolitis season, four microbiota profiles were detected:Escherichia -dominant, Bifidobacterium -dominant,Enterobacter /Veillonella -dominant, andBacteroides -dominant [71]. The proportion of infants with bronchiolitis (related to RSV infection in 65% of them) was lowest in the Enterobacter /Veillonella -dominant profile and highest in the Bacteroides -dominant profile [71]. To determine whether a specific gut microbial profile could be associated with RSV severity, stool samples were collected in 95 infants during an RSV season: 37 were healthy babies and 58 were hospitalized with RSV bronchiolitis [18]. Out of the RSV positive infants, 53 remained in the pediatric ward and 5 later moved to the pediatric intensive care unit. There was a significant enrichment inBacteroides , and a decrease in Firmicutes in RSV infants vs. healthy controls. In addition, infants with severe RSV disease had slightly lower alpha diversity (richness and evenness of the bacterial community) of the gut microbiota compared to infants with moderate RSV and controls. Beta diversity (overall microbial composition) was significantly different between all RSV patients compared to controls [18]. These results were confirmed in BALB/mice in which, after RSV infection, showed a significant increase in the relative abundance of Bacteroidetes  and a corresponding decrease in Firmicutes was detected (figure 3A) [72]. Interestingly, many members of the Bacteroidetes  phylum use mucus as an energy source [73] and Muc5ac mucin levels were significantly increased in the airways and colon of RSV-infected mice, but not in control mice. Changes in gut microbiota composition following RSV infection may also indirectly regulate ILC2 function through the production of alarmins. 6-to-8-week-old BALB/c male mice were randomly divided into a control (CON) group, an ovalbumin (OVA) sensitized group, and an OVA + RSV group [73]. Compared with the CON group, the OVA group had lower abundance of bothBacteroidetes and Firmicutes (figure 3B), whereas higher abundance of these phyla was detected in the OVA + RSV group, compared with the OVA group (figure 3C). RSV-infected asthmatic mice had increased expression of IL-25 and IL-33 and of the Th2 cytokines IL-4, IL-5, and IL-13 [74]. Prevotellaceae_NK4A136 _group, which belongs to the Bacteroides species, was significantly associated with IgE, IL-33, IL-25, IL-5, and IL-13 levels, whereasLachnospiraceae_NK4A136_ group which belongs to theFirmicutes species, was significantly associated with IgE and IL-33 levels [74]. Aggravation of bronchial hyperresponsiveness to methacholine and airway inflammation was observed in asthmatic mice following RSV infection-induced alteration of gut microbiota. Interestingly, in a study comparing children with recurrent respiratory tract infections (RRTI) and a healthy control group, distinct gut bacterial community structures between the groups were observed with an enrichment in Bacteroidetes in the RRTI group [75]. Whether probiotics and/or bacterial derived products, potentially involved in immune training, could affect gut dysbiosis, prevent GIT epithelial dysfunction and the related negative influence on the immune system is an interesting hypothesis that needs to be adequately addressed and demonstrated [76,77].