The main aims of the proposal were to perform a metabolic analysis of NK cells and ILC2s isolated from different tissues under homeostatic or stress conditions; and to study ILCs functional “plasticity”, that is, how ILC subsets adapt to the microenvironment and acquire new functional capacities in the context of disease.
Part 1. We started by establishing the techniques and characterizing the metabolic profiles of NK cells (more abundant compared to ILC2). Human NK cells can be divided into two subpopulations based on differential expression of CD56 (also known as neural cell adhesion molecule or NCAM) and CD16 (or FcγRIIIA, the low-affinity receptor for the Fc portion of immunoglobulin G). In the blood of healthy donors, these subpopulations are identified as CD56DimCD16+ NK cells (NKDim), that are highly cytotoxic and CD56Bright CD16- NK cells (NKBr) which are less cytotoxic but more efficient in producing cytokines, such as IFNγ and TNFα, upon activation. We analyzed the epigenome and the transcriptome and we found that NKDim cells at steady state are metabolically active compared the NKBr. The analysis of the metabolites by mass spectrometry revealed an upregulation of pathways involved in the energetic metabolism in NK cells. We then set up an in vitro system of cytokine stimulation (IL-15-12-18) to monitor the mechanism underlying NKBr and NKDim activation. Using the Seahorse™ bioanalyzer, that provides a real-time measurement of the bioenergetics metabolism, we could observe that upon activation NKDim cells rely mainly on glycolysis, while NKBr cells increase oxidative phosphorylation. Inhibition of the oxidative phosphorylation complexes, or modification in the mitochondrial networks led to a reduction in cytokine production upon activation by both subsets, highlighting the fundamental role of these pathways. Moreover, phenotypic analysis of the two subsets by flow cytometry revealed suggested that mitochondrial remodeling associated with enhanced oxidative phosphorylation represents a potential signature for functionally mature NK cells.
Part 2. ILCs, like T cells, have been described to be “plastic” cells. The ability to adapt phenotype and functions to changes in the environment is crucial during the different phases of the immune response in tissues. It has been reported that human and mouse ILC3s and ILC2s can differentiate into ILC1 by down-regulating respectively the transcription factor RORγt and GATA-3, upregulating T-bet while enhancing the production of IFNγ. Upon chronic inflammation, IFNγ-producing ILC2s and ILC3s accumulate, causing the exacerbation of the pathology. For example, ILC2s are known to regulate adipose tissue homeostasis by controlling eosinophils and alternatively activated macrophages. In the context of obesity, ILC2s modify their function and were found to decrease in abundance, while NK cells become activated and produce IFNγ participating to tissue inflammation and insulin resistance. T cells plasticity has been partially elucidated, however the exact mechanism by which ILCs became plastic and are controlled when inflammation resolves is yet to be clarified. We applied the techniques described in Part 1 to ILC2s and monitored their responses at steady state, upon cytokine activation (IL-1b) or inducing plasticity to type 1 immunity (IL-1b + IL-12). We found that ILC2 rely on different metabolic pathway depending on the given environment and stimuli. Urea cycle and glutamine metabolism resulted to be important pathways for the production of IFNγ. Clarifying the steps to ILC2 activation and plasticity might be highly relevant especially in the context of complex diseases and metabolic disorders.
Data from the two parts have been used to write two manuscripts that are in the course of submission and revision processes. The data have also been presented in national and international meetings. Some of the data regarding the NK cells metabolism have been used in a presentation to general public, in the context of students’ formation.
