Background Human respiratory syncytial virus (RSV) is the most common cause of bronchiolitis and acute lower respiratory tract infections (LRTI) among children worldwide. RSV is the leading cause of childhood hospitalisation and a financial burden on national health systems. No vaccine that protects against RSV infection is available. Children with bronchiolitis are at high risk of developing wheeze and asthma later in childhood. Recent data suggest that innate immunity plays a vital role in childhood, especially when maternal antibodies' levels are low and adaptive immune responses are immature. Innate lymphoid cells (ILCs) are among the most recently discovered innate immune cell types that could play a pivotal role in innate responses to RSV bronchiolitis. ILCs have been classified into three groups: ILC1, ILC2 and ILC3 depending on their function, transcription factor expression and phenotype. Although ILC2 numbers have been found to be elevated in bronchiolitis, the phenotypic and functional differences between ILCs from healthy infants and those with bronchiolitis have not been explored.
Hypothesis I propose that ILC frequencies and function are associated with disease severity and aetiology in children with acute lower respiratory tract infection.
Sub-hypothesis 1 ILCs from neonates are functionally different from healthy adult subjects.
Sub-hypothesis 2 Alterations in ILC frequencies and numbers in blood and airway samples are associated with disease severity, pathogen type and load.
Sub-hypothesis 3 Alterations in ILC function are associated with disease severity, pathogen type and load.
Methods To determine differences in ILCs in early life, lymphocyte populations, including ILC subset frequencies and numbers in blood samples collected from adults and neonates were investigated by flow cytometry. To study ILCs in bronchiolitis, blood and upper and lower airway samples were collected from infants with and without a respiratory viral infection. An optimised multicolour immunophenotyping panel (OMIP), intracellular staining panels (ICS) and other systems of analytical approach such as UMAP (Uniform Manifold Approximation and Projection), heatmaps, correlograms and principal component analysis (PCA) were used. ILCs were defined using the following gating strategy: live, single, CD45+, lineage negative (CD3, CD14, CD16, CD56, CD1a, CD123 FcεR1α)- and CD127+ cells. ILC1 cells were defined as CD117−CRTH2−, ILC2 cells were defined as CRTH2+ CD117int and ILC3s were defined as CD117+CRTH2−. At the same time, other lymphocyte and granulocyte populations were enumerated. Alongside cellular work, using the Meso Scale Discovery (MSD) platform, 29 immune mediators and cytokines were measured in airway samples.
Results I successfully optimised flow cytometry panels to determine frequencies and function of ILCs in small volumes of difficult to obtain clinical samples. Results suggest that ILC numbers were higher in whole blood obtained from neonates compared to healthy adults (P=0.0032), however, they produced significantly less IFN-γ (P=0.0015). Furthermore, there were significantly lower numbers of NKT (P=0.0001) and MAIT cells (P=0.0031) in neonates compared to healthy adults. In bronchiolitis, ILC numbers were significantly decreased in peripheral blood (P=0.005) and airways samples (P=0.025) obtained from bronchiolitic compared to control subjects; however, their capacity for IFN-γ production was similar between those groups. Interestingly, there was a significantly decreased production of IFN-γ in CD4+ (P=0.042) and CD8+ T cells (P=0.006) from bronchiolitic compared to control patients. Finally, levels of almost all mediators were higher in bronchiolitic babies than controls, including both type-1, type-2 and type-17 associated cytokines. Only IL-4 production in lower and upper airways was increased in rhinovirus- (RV) but not RSV-associated bronchiolitis (P=0.006).
Conclusions Small volumes of whole blood and paediatric respiratory samples can be used for enumerating ILCs (NK cells, ILC1, ILC2, ILC3), granulocytes (eosinophils and neutrophils), T – cells (CD4+ and CD8+), MAIT cells and NKT-like cells. The comparison of neonate and adult functional responses suggests that innate lymphoid cells are mature in early life and might play a pivotal role in the neonatal type-2 bias via the production of IL-13. Bronchiolitis was characterised not only by CD3+ T cell but also by CD3- lymphopenia. ILCs numbers were much lower in infants with bronchiolitis compared to healthy controls; however, ILC had the ability to produce IFN-γ. My data suggest that RSV and RV can lead to different phenotypes of bronchiolitis. Finally, I demonstrated the importance of studying the lower and upper airways simultaneously, discovering that inflammatory signatures were different between these two compartments. For the first time all members of the ILC family were studied in viral bronchiolitis. Determining why some infants develop bronchiolitis and the pathogenesis of lower airway disease in infants could lead to a better understanding of early life immunity and the development of new biomarkers and clinical prophylactics.