AUTHOR=Li Bobby W. S. , Stadhouders Ralph , de Bruijn Marjolein J. W. , Lukkes Melanie , Beerens Dior M. J. M. , Brem Maarten D. , KleinJan Alex , Bergen Ingrid , Vroman Heleen , Kool Mirjam , van IJcken Wilfred F. J. , Rao Tata Nageswara , Fehling Hans Jörg , Hendriks Rudi W. TITLE=Group 2 Innate Lymphoid Cells Exhibit a Dynamic Phenotype in Allergic Airway Inflammation JOURNAL=Frontiers in Immunology VOLUME=8 YEAR=2017 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2017.01684 DOI=10.3389/fimmu.2017.01684 ISSN=1664-3224 ABSTRACT=

Group 2 innate lymphoid cells (ILC2) are implicated in allergic asthma as an early innate source of the type 2 cytokines IL-5 and IL-13. However, their induction in house dust mite (HDM)-mediated airway inflammation additionally requires T cell activation. It is currently unknown whether phenotypic differences exist between ILC2s that are activated in a T cell-dependent or T cell-independent fashion. Here, we compared ILC2s in IL-33- and HDM-driven airway inflammation. Using flow cytometry, we found that surface expression levels of various markers frequently used to identify ILC2s were dependent on their mode of activation, highly variable over time, and differed between tissue compartments, including bronchoalveolar lavage (BAL) fluid, lung, draining lymph nodes, and spleen. Whereas in vivo IL-33-activated BAL fluid ILC2s exhibited an almost uniform CD25+CD127+T1/ST2+ICOS+KLRG1+ phenotype, at a comparable time point after HDM exposure BAL fluid ILC2s had a very heterogeneous surface marker phenotype. A major fraction of HDM-activated ILC2s were CD25lowCD127+T1/ST2low ICOSlowKLRG1low, but nevertheless had the capacity to produce large amounts of type 2 cytokines. HDM-activated CD25low ILC2s in BAL fluid and lung rapidly reverted to CD25high ILC2s upon in vivo stimulation with IL-33. Genome-wide transcriptional profiling of BAL ILC2s revealed ~1,600 differentially expressed genes: HDM-stimulated ILC2s specifically expressed genes involved in the regulation of adaptive immunity through B and T cell interactions, whereas IL-33-stimulated ILC2s expressed high levels of proliferation-related and cytokine genes. In both airway inflammation models ILC2s were present in the lung submucosa close to epithelial cells, as identified by confocal microscopy. In chronic HDM-driven airway inflammation ILC2s were also found inside organized cellular infiltrates near T cells. Collectively, our findings show that ILC2s are phenotypically more heterogeneous than previously thought, whereby their surface marker and gene expression profile are highly dynamic.