UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Elucidating the mechanism of type 2 innate lymphoid cell-mediated tumour elimination de Lucía Finkel, Pablo 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2020_november_de_lucia_finkel_pablo.pdf [ 4.03MB ]
JSON: 24-1.0392343.json
JSON-LD: 24-1.0392343-ld.json
RDF/XML (Pretty): 24-1.0392343-rdf.xml
RDF/JSON: 24-1.0392343-rdf.json
Turtle: 24-1.0392343-turtle.txt
N-Triples: 24-1.0392343-rdf-ntriples.txt
Original Record: 24-1.0392343-source.json
Full Text

Full Text

    ELUCIDATING THE MECHANISM OF TYPE 2 INNATE LYMPHOID CELL-MEDIATED TUMOUR ELIMINATION   by PABLO DE LUCÍA FINKEL B.Sc., Hon., University of Glasgow, 2017  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Microbiology and Immunology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2020 © Pablo de Lucía Finkel, 2020ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the thesis entitled:  Elucidating the Mechanisms of Type 2 Innate Lymphoid Cell-Mediated Tumour Elimination  submitted by Pablo de Lucía Finkel  in partial fulfillment of the requirements for the degree of Master of Science in Microbiology and Immunology  Examining Committee: Dr. Wilfred Jefferies, Professor, Department of Microbiology and Immunology, UBC. Supervisor  Dr. Kenneth Harder, Associate Professor, Department of Microbiology and Immunology, UBC. Supervisory Committee Member  Dr. Laura Sly, Associate Professor, Department of Pediatrics, Faculty of Medicine, UBC. Additional Examiner     Additional Supervisory Committee Members: Dr. Michael Cox, Associate Professor, Department of Urologic Sciences, UBC. Supervisory Committee Member     iii  Abstract Besides amplifying type 2 immune responses, the role of type 2 innate lymphoid cells (ILC2s) in cancer remains ambiguous. While some studies suggest that ILC2s exert immunosuppressive functions, others present evidence they enhance anti-cancer immunity. However, their successful isolation from mammal tissues and in vitro expansion continues to pose challenges. This is partly due to their scarcity compared to other leukocyte populations, but also because our current knowledge on ILC2 biology is incomplete. In this study, we focused on ST2+ IL-25Rlo lung resident ILC2s and demonstrate for the first time how mouse type 2 innate lymphoid cells can be cultured, grown and expanded in serum-free media containing interleukin-33 (IL-33), TSLP, IL-2 and IL-7. Moreover, we report that ILC2s from the lungs of mice with primary tumours are able to cross-present the ovalbumin peptide OVA257-264 and effectively cross-prime a population of OT-I cells. Our data also demonstrates that ILC2s from naïve mice are less efficient at antigen cross-presentation, suggesting a plastic cellular identity in the presence or absence of a tumour. To further our understanding of ILC2s harvested from mice with primary tumours, our single-cell RNA sequencing analyses reveal a highly heterogeneous population with several subsets. One of them downregulates markers associated to type 2 identity such as GATA-3, IL-5 or IL-13, but also begins to express pro-inflammatory markers like granzyme B, IFNγ or interferon-induced membrane proteins. We hypothesize that systemic inflammation or a tumour environment can trigger differentiation of a portion of the lung resident ILC2 population into a pro-inflammatory subset. The existence of these cells challenges current paradigms of ILC2 biology and implies the participation of a distinct agent in the process of cancer immune surveillance.       iv  Lay Summary Type 2 innate lymphoid cells (ILC2) are an unconventional population of lymphocytes that amplify immune responses directed against helminthic parasites but are also capable of driving allergen sensitization and thus, asthma or allergic lung inflammation. These is known as type 2 immunity. However, the role that they play in cancer remains controversial. Given their scarcity in mammalian tissues, we first optimize protocols of ILC2 isolation and culture. Furthermore, we report the ability of ILC2s to perform antigen cross-presentation, an elegant immunological mechanism designed to activate CD8+ T-cells, which are professional tumour cell killers. Finally, we demonstrate that purified, lung ILC2s are a plastic cell population, meaning that in response to certain environmental cues, they can shift their type 2 identity into a pro-inflammatory type 1 response, which is directed to antiviral functions and anti-tumour immunity.             v  Preface Dr. Wilfred Jefferies was responsible for the idea of ILC2 cross-presentation and CD8+ T-cell cross-priming, as well as the idea of performing single-cell RNA sequencing on ILC2s.  This thesis is based on the work of Dr. Iryna Saranchova, who has demonstrated a role for ILC2s in cancer, and hence provided a starting point for this thesis.  Cell sorting was performed by Andy Johnson and Justin Wong, from The University of British Columbia (UBC) FLOW Cytometry facility at the Life Science Centre. Bulk RNA sequencing was performed by Tara Stach and Ryan Vander Werff at the Biochemical Research Centre Sequencing Core at UBC. Single cell RNA sequencing and data processing was done by Anne Haegert and Yen-Yi Lin respectively at the Laboratory for Advanced Genomics at the Vancouver Prostate Centre.  Part of the bioinformatics analyses for the single cell RNA sequencing data was done by Clara Xia from the Jefferies group at the Michael Smith Laboratory, UBC.  All animal breeding (A18-0298/A14-0267) and experimental (A15-0284/A20-0066) protocols were approved by the UBC Animal Care Committee, which is overseen by the Canadian Council on Animal Care (CCAC). Experiments were performed under the UBC Biosafety protocol B11-0066 and B17-0002. Pablo L. Finkel performed all other work completed in this thesis.    vi  Table of Contents  Abstract .................................................................................................................................................. iii Lay Summary .......................................................................................................................................... iv Preface ...................................................................................................................................................... v Table of Contents .................................................................................................................................... vi List of Tables ........................................................................................................................................ viii List of Figures ......................................................................................................................................... ix List of Abbreviations ................................................................................................................................ x Acknowledgments ................................................................................................................................. xii Dedication ............................................................................................................................................ xiii Literature Review. .................................................................................................................................... 1 Innate immunity and innate lymphoid cells ......................................................................................... 2 Type 2 innate lymphoid cell development ........................................................................................... 3 Stimulation and functions of conventional ILC2s ................................................................................ 5 ILC2 plasticity and subsets .................................................................................................................. 8 Memory ILC2s .................................................................................................................................. 9 Regulatory ILC2s ........................................................................................................................... 10 Inflammatory ILC2s ....................................................................................................................... 12 Type 1 ILC2s .................................................................................................................................. 14 The role of ILC2s in cancer................................................................................................................ 17 Pro-tumorigenic evidence .............................................................................................................. 17 Anti-tumorigenic evidence .............................................................................................................. 19 Cross-presenting ILC21 .................................................................................................................. 21 Concluding remarks ........................................................................................................................... 22 Hypothesis and Objectives ..................................................................................................................... 24 Materials and Methods ........................................................................................................................... 25 Tissue processing and ILC2 isolation ................................................................................................ 25 Mouse ILC2 in vitro culture and analysis .......................................................................................... 26 Calculation of ILC2 growth rate and doubling time .......................................................................... 27 Tumour cell culture and establishment .............................................................................................. 27 ILC2 in vitro culture and cross-priming assay ................................................................................... 28 Dendritic cell harvest and differentiation ........................................................................................... 29 CD8+ T-cell isolation and culture ...................................................................................................... 29 vii  ILC2 and CD8+ lymphocyte co-culture and in vitro cross-priming assay ......................................... 30 Intracellular staining for flow cytometry ........................................................................................... 30 Cytokine production assays ................................................................................................................ 31 RNA isolation and sequencing ........................................................................................................... 31 Single cell RNA sequencing sample preparation ............................................................................... 31 The scRNA-seq and data analysis ...................................................................................................... 32 Chapter 1. Defining Parameters in Serum-Free Medium for the Expansion and Study of Murine     Lung Type 2 Innate Lymphoid Cells .................................................................................................. 33 Introduction ........................................................................................................................................ 33 Results ................................................................................................................................................ 35 Mouse ILC2s can be optimally expanded and sub-cultured in serum-free media ......................... 35 Discussion .......................................................................................................................................... 39 Conclusion.......................................................................................................................................... 44 Chapter 2. Type 2 Innate Lymphoid Cells Cross-Prime CD8+ T-Lymphocytes Like Dendritic Cells. . 45 Introduction ........................................................................................................................................ 45 Results ................................................................................................................................................ 48 Type 2 innate lymphoid cells are of lymphoid origin and functional ............................................. 48 pILC2s from mice with IL-33+ primary tumours successfully process exogenous, whole OVA protein and cross-present the peptide SIINFEKL on their MHC-I ................................................ 50 ILC2s derived from tumour – bearing mice cross-prime population of OT-I cells by presenting SIINFEKL peptide from exogenous OVA ....................................................................................... 53 Discussion .......................................................................................................................................... 56 Conclusion.......................................................................................................................................... 61 Chapter 3. Single-Cell Analysis Reveals a Heterogeneous Population of Type 2 Innate Lymphoid  Cells from Tumour-Bearing Mice that Lost Type 2 Identity in Favour of Inflammatory Markers .... 63 Introduction ........................................................................................................................................ 63 Results ................................................................................................................................................ 66 ILC2s derived from mice with primary tumours (pILC2s) lose type 2 identity at the mRNA but   not the protein level ........................................................................................................................ 66 A subset of pILC2s that lose type 2 identity possess a pro-inflammatory phenotype ..................... 69 ILC2s derived from primary tumour-bearing mice are heterogeneous ......................................... 73 Discussion .......................................................................................................................................... 77 Conclusion.......................................................................................................................................... 85 Final Conclusion .................................................................................................................................... 87 Bibliography ........................................................................................................................................... 88 Appendix .............................................................................................................................................. 102 viii  List of Tables Table 1. Upregulated genes in cluster 6 differentially expressed from the other clusters in a              non-activated population of pILC2s harvested 2 weeks post tumour establishment.. ........................... 72     ix  List of Figures Figure 1.  Serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 promotes female mouse ILC2 growth and expansion better than RPMI 1640 with 10% fetal bovine serum..37 Figure 2.  Subcultures of primary ILC2s from female mice using serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 ......................................................................................... 38 Figure 3.  Sorted ILC2s from female C57Bl/6J mice are a pure and functional population of   lymphoid origin. ................................................................................................................... 49 Figure 4.  Non IL-33 pre-treated pILC2s from female mice succesfully process OVA protein and cross-present the peptide SIINFEKL. .................................................................................. 52 Figure 5.  OT-I proliferation in co-culture with different types of ILC2s. ........................................... 55 Figure 6.  pILC2s from female mice with IL-33+ primary tumours treated with OVA cross-prime   OT-I cells on day 3. .............................................................................................................. 55 Figure 7.  pILC2s from female mice lose type 2 identity at the RNA but not the protein level on      day 3 of culture. .................................................................................................................... 68 Figure 8.  A subpopulation of pILC2s lose type 2 identity, express type 1-associated genes and        has potential to have acquired APC-like properties. ............................................................ 71 Figure 9.  The scRNA-seq of the week 2, non-activated pILC2 population reveals a cluster with potential APC-like properties and transitioning to a pro-inflammatory phenotype. ............ 75 Figure 10.  Week 3, non-activated pILC2s.  ........................................................................................... 73 Figure 11 Week 3, activated pILC2s .................................................................................................... 76 Figure S1. Heat map comparing the most differentially expressed genes for every cluster in each sample of pILC2s ............................................................................................................... 102 Figure S2. Heat map comparing most differentially expressed genes by pathway in pILC2s. ............ 103   x  List of Abbreviations APC – Antigen Presenting Cell // Allophycocyanin  AT – Adipose Tissue BV – Brilliant Violet CD – Cluster of Differentiation CFSE – Carboxyfluorescein Succinimidyl Ester CHILP – Common Helper Innate Lymphoid Progenitor CILP – Common Innate Lymphoid Progenitor CLP – Common Lymphoid Progenitor CTL – Cytotoxic T-Cell DC – Dendritic Cell DMEM – Dulbeccos’s Modified Eagle Medium  DPBS – Dulbecco's Phosphate-Buffered Saline ELISA – Enzyme Linked Immunosorbent Assay FACS – Fluorescence-Activated Cell Sorting FBS – Fetal Bovine Serum  FITC – Fluorescein Isothiocyanate FPKM – Fragments per Kilobase Million GATA-3 – [A/T] GATA [A/G]-Binding Protein 3 GM-CSF – Granulocyte Macrophage-Colony Stimulating Factor HBSS – Hank’s Balanced Salt Solution IFITM – Interferon-Induced Transmembrane Protein IFN - Interferon IL – Interleukin  xi  ILC – Innate Lymphoid Cell ILC2 – Type 2 Innate Lymphoid Cell LN – Lymph Node MDSC – Myeloid-Derived Suppressor Cell MHC – Major Histocompatibility Complex  mILC2 – Metastatic Type 2 Innate Lymphoid Cell nILC2 – Naïve Type 2 Innate Lymphoid Cell NK – Natural Killer OT-I – C56Bl/6-Tg(TcraTcrb)1100Mjb/J Mice OVA – Ovalbumin  P/S – Penicillin/Streptomycin PE – Phycoerythrin PerCP – Peridinin-Chlorophyll-Protein pILC2 – Primary Type 2 Innate Lymphoid Cell RNA – Ribonucleic Acid RPMI – Roswell Park Memorial Institute TCR – T-Cell Receptor TSLP – Thymic Stromal Lymphopoietin  tSNE – t-Distributed Stochastic Neighbour Embedding  UBC – The University of British Columbia UMI – Unique Molecular Identifier WT – Wild Type    xii  Acknowledgments I would like to thank my committee members Dr. Michael Cox and Dr. Ken Harder for their wisdom and their guidance. I would also like to thank Dr. Iryna Saranchova for teaching me the most important wet laboratory techniques, and to KB, Cheryl Pfeifer and Lonna Munro for easing every experiment for me. Every other member from the Jefferies Laboratory has made my experience memorable. Special thanks to Andy Johnson and Justin Wong for their experience and patience when sorting ILC2s, to the RNA sequencing staff Tara Stach and Ryan Vander Werff, and of course to Anne Haegert, Yen-Yi Lin, Robert Bell and Clara Xia for their support in data processing. I would like to give a special mention to the mice sacrificed for this work. Without them, research and progress would be an impossibility. My deepest thanks to Dr. Wilfred Jefferies for providing me the opportunity of living in Vancouver, for his wisdom, his humour, and most importantly, for teaching me the value of ambition and hard work.   xiii  Dedication This work is dedicated to my mother Lucila, my father César, my grandmothers Susy and Carmen, my grandfather Carlos (Pexete), who is no longer with us, and to Mario. Thank you for bringing me back home with every phone call and thank you for your infinite trust in every new challenge I face.  1  Literature Review Type 2 innate lymphoid cells are a group of effectors that mediate type 2 immunity by driving the expulsion of helminthic parasites but also sensitizing the organism to allergens and driving allergic, lung inflammation. As innate cells, they do not recognize protein peptide, instead, they are sensitive to the alarmin response, especially IL-33, IL-25 and TSLP. Upon alarmin engagement they produce type 2 cytokines that amplify adaptive TH2 immunity and regulate tissue homeostasis. Their innate but also lymphoid identity appoints them as an intriguing group of unconventional cells; however, an increasing amount of evidence is unmasking them as more than type 2, tissue-specific effectors by unravelling a series of unprecedented functions that less than five years ago were unthinkable for ILC2s. Their highly plastic transcriptional programs have allowed researchers to classify and characterize ILC2s in more detail than ever, however, the novel and exciting nature of ILC2 biology requires constant updates and recapitulations. This review attempts to summarize where the field is in terms of ILC2 function, but most importantly it provides an overview of the current subsets that research has discovered. Eventually, the role of ILC2s in cancer is re-evaluated by taking into account their highly plastic nature to try to understand what the current knowledge on type 2 innate lymphoid cells is.        2  Innate immunity and innate lymphoid cells The agents that comprise innate immune responses have limited specificity compared to the adaptive response, but they still distinguish and sense foreign structures through evolutionarily conserved pathogen-recognition receptors (Akira & Takeda, 2004). Although phagocytes, polymorphonuclear leukocytes, the complement system or the release of a wide variety of defensins tend to act immediately upon danger encounter (Medzhitov & Janeway, 2000), their role in supporting and linking adaptive immunity is often overlooked. So much, that it took almost 50 years since the discovery of the lymphocyte (Cooper & Miller, 2019) to unmask a whole family of unconventional innate effectors that orchestrate immunity, inflammation and homeostasis across multiple tissues (Artis & Spits, 2015). These are the innate lymphoid cells (ILCs), the immune’s system ultimate blend between the innate and the adaptive branches.  What makes every ILC innate is a lack of recombination activating gene-dependent rearranged antigen receptors (Spits et al., 2013). These group of cells have no antigen specificity of any kind, instead, they respond to a wide variety of cell-derived factors or alarmins such hormones and cytokines, that are secreted both in homeostasis and inflammatory conditions (Artis & Spits, 2015). Upon alarmin engagement with specific membrane receptors, ILCs usually respond by secreting cytokines that not only will help unleash the full potential of the immune response but will also orient it towards its specific needs (Eberl et al., 2015). Curiously, that same cytokine response is what discriminates the three main types of ILCs. Based on differential expression of cell surface markers and distinct secretion profiles, ILCs can be type 1, 2 or 3. On the other hand, what makes these cells lymphoid is their developmental origin, as they are all derived from a common lymphoid progenitor (CLP) committed to multiple cell lineages, including B- and T-lymphocytes (Artis & Spits, 2015). One of the possible differentiations of the CLP is the ILC/NK precursor, or α-lymphoid precursor, and it can be distinguished from the CLP itself thanks to the expression of the integrin α4β7 and the chemokine receptor CXCR6 (Klose et al., 2014; 3  Cherrier et al., 2012; Possot et al., 2011). Eventually, ILCs reach their final effector stages guided by transcriptional regulators that are retained even when terminally differentiated, and their expression can also be used to distinguish between types 1, 2 and 3.  Type 1 ILCs play important roles in inflammation, and after extensive studies this subset has been split into two groups, one comprising a highly cytotoxic innate cell, the natural killer (NK), and another one that has been established as the ILC1, which promotes inflammation but has less cytotoxic potential (Jiao et al., 2016). NK cells can be considered the innate counterpart of the CD8+ cytotoxic T-cell, whereas the T-bet+ ILC1 immune and transcriptomic profile resembles that of a T-bet+ CD4+ TH1 cell. Type 3 ILCs are yet another example of ILC heterogeneity within a subset. They are defined by the expression of the transcription factor RORγt, like their TH17 counterparts, and can also be split into two different cell types, the lymphoid tissue inducer (LTi) and the ILC3 (Ardain et al., 2019). In the mammalian fetus, the former helps develop the lymph nodes and Peyer’s Patches, whereas the latter participates in maintaining intestinal homeostasis and defends against extracellular pathogens (Spits et al., 2013). The focus of this review will be the remaining group – type 2 innate lymphoid cells (ILC2s) – which are perhaps the least understood ILC subset given the increasing evidence that recently situates them as a highly heterogeneous population that play more roles in immunity than previously thought. Type 2 innate lymphoid cell development Like every other ILC, type 2 innate lymphoid cells (ILC2) are also derived from a CLP in the mammalian bone marrow and fetal liver, which in turn can differentiate into the previously mentioned α-lymphoid precursor (αLP), which has potential to give rise to any imaginable ILC, but not T- or B-lymphocytes (Artis & Spits, 2015). This separation from adaptive lineages that occurs so early in development is mediated by expressing the transcriptional repressor Id2, which inhibits transcription of E proteins that are essential for downstream differentiation into B- and T-cells (Berrett et al., 2019). 4  The existence of the αLP has been confirmed because knockout mice for Id2 lack every ILC lineage (Satoh-Takayama et al., 2010). In addition, Notch signalling is also fundamental to produce αLP cells and to add another level of blockade towards B-cell development (Radtke et al., 1999). Then, there is a major lineage divergence in ILC development. If the αLP maintains Id2 expression but also upregulates IL-7Rα and GATA-3, the cell has transitioned into a common helper innate lymphoid progenitor, or CHILP, that can give rise to every ILC type except NK cells. If the αLP cell loses Id2 expression and remains IL-7Rα-, then it has differentiated into a NK cell precursor instead of a CHILP (Wang et al., 2020). The expression of GATA-3 at the CHILP stage, however, is debatable. Other groups have reported that this transcription factor is expressed one step further in development, at the common innate lymphoid progenitor stage (CILP), and immediately after the CHILP (Yagi et al., 2014). Nonetheless, in both cases GATA-3 is being expressed before reaching terminal ILC2 differentiation, which is why it is a good intranuclear identity marker for cell characterization. What designates the transition from the CHILP to a CILP cell is the expression of the transcription factor PLZF and consistent TGF-β signalling (Wang et al., 2020). At this point, the PLZF+ CILP loses potential to give rise to LTi cells. Instead, it only gives rise to ILC1s, ILC2s and ILC3s. This is why the CILP is the ILC precursor that is most skewed towards ILC development, in other words, the progenitor that produces the highest frequencies of ILCs (Cortez et al., 2015). LTi cells, however, differentiate from PLZF- CHILP cells (Constantinides et al., 2014). Interestingly, mice deficient in zbtb16 (the gene for PLZF) showed mild reduction but not an absence of ILC1s, ILC2s and CCR6- ILC3s (or non LTi ILC3s). This demonstrated that PLZF is not strictly required for ILC development (Klose et al., 2014). On the other hand, zbtb16-/- mice showed normal development of NK cells.  The CILP is the closest and therefore last progenitor cell before reaching terminal ILC2 differentiation. To reach the final effector stage, the expression of GATA-3, RORα and TCF-1 is essential. The first one, GATA-3, has been retained since it was first upregulated at some point 5  between the CHILP and the CILP, and it is also the most important one. Hoyler and colleagues demonstrated in 2012 that a gata3 deletion using Id2CreERt2/+ mice strongly impaired ILC2 differentiation. In addition, Klein Wolterink et al. (2013) showed that overexpressing GATA-3 in ILC2s led to an increase in the expression of the alarmin receptor IL-33R (ST2), which translates into enhanced type 2 cytokine secretion. This suggests that GATA-3 is not only essential for ILC2 generation but also important to maintain effector functions. GATA-3 is in turn dependant on Gfi, a transcription factor that regulates its expression and suppresses the IL-17A gene, and thus, ILC3 identity (Spooner et al., 2013). The second transcription factor required is RORα. Mice lacking the RORα gene have very few ILC2s in the bone marrow, lung, intestines and other tissues (Halim et al., 2012), and although they are not completely absent, they are still insensitive to triggering alarmins (Wong et al., 2012). RORγt+ ILC3s were still present and functional in the absence of RORα. The third transcription factor that CILP cells need to give rise to ILC2s is TCF-1 (encoded by tcf7). Mice lacking the gene have ST2lo ILC2s that are functionally impaired but are still able to differentiate from the bone marrow (Mielke et al., 2013).  Stimulation and functions of conventional ILC2s The emerging plasticity in what was used to be considered the most homogeneous group of ILCs forces us to address non-plastic ILC2s as conventional versus the unconventional subsets that we present later in this review. What remains undoubted is that ILC2s are experts at mediating the expulsion of helminthic parasites and are implicated in the development of asthma and allergic inflammation. These are known as type 2 immune responses (Walker & McKenzie, 2013). ILC2s are stimulated by alarmins, which are chemotactic or activating proteins that are released by tissue-resident or lining cells in response to anoikis, cell injury, death or degranulation (Yang et al., 2017). In particular, the alarmins that stimulate ILC2s are IL-33, IL-25 and thymic stromal lymphopoietin (TSLP), leading to ILC2 activation and a subsequent type 2 cytokine response (Konya & Mjösberg, 6  2016). Barlow et al. demonstrated in 2013 that IL-33 is far more potent than IL-25 and TSLP when it comes to lung ILC2 activation due to its rapid expression and release compared to other alarmins. Despite this, the basis of ILC2 heterogeneity relies in the multiple tissues they occupy in mammals. Gut or skin ILC2s, for example, express low levels of IL-33R (ST2), but are IL-25Rhi, so far more responsive to IL-25 stimulation (Ricardo González et al., 2018). In fact, ILC2s were first identified as c-kit+ FcεR1- cells in IL-25-/- mice that were lacking a proper type 2 response against the protease allergen papain (Fallon et al., 2006), hence the importance of IL-25. In addition, IL-2 has proven to be essential for ILC2 expansion (Roediger et al., 2015) and IL-7 for survival in the lungs (Halim et al., 2012), but both cytokines are ineffective at ILC2 stimulation in the absence of IL-33 or IL-25 (Martínez-González et al., 2015). Upon alarmin engagement, ILC2s across most tissues secrete signature type 2 cytokines such as IL-5, IL-13, amphiregulin, moderate levels of IL-9 and low levels of IL-4 (Martínez-González et al., 2015). Interestingly, several studies have identified novel ILC2 activators such as the neuropeptide NMU, TGF-β1 or several lipid mediators that trigger type 2 cytokine release in levels that are equivalent to IL-33 stimulation (Wallrapp et al., 2017; Wang et al., 2020; Denney et al., 2015; Konya & Mjösberg, 2016). IL-5 promotes eosinophil differentiation from the bone marrow and activates them to release granule contents specific to eosinophils that are toxic to helminths and protozoa, but also to host cells in severe cases of eosinophilia (Abbas et al., 2018). IL-13 is directed to a wider range of functions. It can promote smooth muscle contractibility and hyperactivate goblet cells to produce high amounts of protective mucus in the gut or in the airway. This has been consistently linked to severe asthma (Kuperman et al., 2002). These effects of IL-5 and IL-13 are observable in RAG mice, demonstrating a T cell-independent role for ILC2s in health and disease (Halim et al., 2014). However, via IL-13, ILC2s also link innate and adaptive immune responses by stimulating the migration of IL-13R+ allergen-activated dendritic cells (DCs) that have been “licensed” to migrate from the lung to the mediastinal lymph nodes, where they will elicit CD4+ TH2 cell proliferation 7  (Halim et al., 2014). ILC2s are already a proven source of IL-4, which is an important driver of TH2 differentiation (Pelly et al., 2016) and immunoglobulin class switch to IgE, which is a mediator of mast cell degranulation and allergy (Abbas et al., 2018). Finally, IL-9 supports both innate and adaptive responses by promoting the growth of smooth muscle cells, epithelial cells, mast cells, B-cells and CD4+ T-cells (Goswami & Kaplan, 2011). Despite being seen as type 2 innate effectors in disease, conventional ILC2s play distinct roles in homeostasis across tissues. IL-25Rhi ST2lo gut ILC2s, for example, are highly responsive to tuft cell-derived IL-25, leading to IL-13 production. In 2016, von Moltke and colleagues demonstrated a feed forward circuit where ILC2-derived IL-13 promotes further differentiation of tuft and goblet cells from epithelial crypt progenitors, highlighting the role of gut ILC2s in tissue homeostasis. Furthermore, production of IL-13 but especially IL-4 is essential for macrophage polarization into the M2 phenotype, which is a crucial agent in the process of wound healing by promoting epithelial cell turnover, encapsulation and barrier formation, increased luminal fluids, angiogenesis and myofibroblast activation (Gause et al., 2013). ILC2s can also participate in wound healing in more direct ways. In response to influenza infection in the lungs, ILC2s present genetic profiles enriched for extracellular matrix genes involved in lung tissue remodelling (Monticelli et al., 2011). Upon viral infection, lung ILC2s also produce amphiregulin, which has been reported to be implicated in lung tissue repair (Enomoto et al., 2009; Fukumoto et al., 2010). In the skin, elevated numbers of ILC2s upon cutaneous inflammation contribute to pathology (Kim et al., 2013; Roediger et al., 2013; Salimi et al., 2013). Yet, an absence of IL-33 in mice and humans presents poor levels of type 2 cytokines, severely impaired re-epithelialization and deficient wound closure, demonstrating a novel role for the IL-33/ILC2 axis in restoration of the cutaneous barrier (Rak et al., 2015).  If these non-canonical roles of ILC2s were not remarkable enough, they also seem to play an important part in adipose tissue (AT) homeostasis. In mammals, white AT stores excess energy and contributes to fat accumulation, whereas brown AT is involved in dissipating energy through the 8  production of heat, and therefore linked to lower body weight (Cautivo & Molofsky, 2016). By expressing a certain proprotein convertase and an endopeptidase, IL-33 – dependant white AT-resident ILC2s can induce the upregulation of the uncoupling protein 1 (UCP1) in white AT (Brestoff et al., 2015). UCP1 is involved in dissipation of energy through the production of heat, therefore converts white AT into beige and eventually brown AT. Therefore, the “browning” of AT by ILC2s can prevent obesity (Cautivo & Molofsky, 2016). Also, by reducing white AT where multiple immune cells accumulate, ILC2s prevent chronic low grade AT inflammation and systemic insulin resistance that usually evolves into type 2 diabetes mellitus (Cildir et al., 2013). As with lung tissue, alternatively activating macrophages promotes AT remodelling and prevents systemic inflammation. These macrophages are further supported by ILC2-derived IL-4 competent eosinophils that accumulate in response to ILC2-derived IL-5 (Wu et al., 2011). ILC2 plasticity and subsets So far, we have outlined what we consider the conventional roles of a member of an already unconventional family of innate lymphocytes. In other words, the roles that ILC2s exert based on their “standard” transcriptomic program, but like every other cell, ILC2s respond to environmental stimuli. As a consequence, they re-wire their default programming and adapt to changing circumstances in an unexpectedly flexible manner. This is known as cellular plasticity. Given the resemblance between the ILC and the T-cell transcriptome, ILC plasticity was, to an extent, expected. The ability of CD4+ T-cell subsets to interchange identities with each other in response to external cues was already an established phenomenon (Geginat et al., 2014; Caza & Landas, 2015; Zhou et al., 2009) and was already found to be implicated in cancer and autoimmune disease, but also a new advantage that the immunology community could harness as a therapy (Huber et al., 2017; DuPage & Bluestone, 2016). Why would their innate counterparts be any less plastic? So far, studies on ILC plasticity are becoming a common trend in ILC biology. Not only we consider inter-subset plasticity, 9  where cells abandon their default programming towards a different ILC identity, but also intra-subset reprogramming that generates more subpopulations within a given ILC subset. In this section, we explore beyond the unconventional. We focus on the different ILC2 subsets that have been identified so far. Some of them are clear examples of inter-subset plasticity and others reflect the deeper understanding that we have gained over recent years on type 2 innate lymphoid cells.  Memory ILC2s Innate effectors of this kind do not need to become an entirely different cell to challenge the current dogma of ILC biology. Martínez-González and her group have demonstrated in 2016 the existence of a pool of memory ILC2s (ILC2m) that responds more vigorously to secondary stimulation after several months, defying the notion of immunological memory being defined by antigen specificity. Upon intranasal injection of IL-33 or papain in mice, there is a rapid expansion in lung ILC2 numbers followed by a contraction phase. However, Martínez-González and colleagues demonstrated that a fraction of the ILC2s can persist in the lungs for 2 weeks and draining mediastinal lymph nodes (mLN) for up to 160 days after primary challenge. In addition, these ILC2s were IL-5+ IL-13+, but the cytokines were not being secreted, instead, they were stored within the cell (Martínez-González et al., 2016). This reduced pool that survived the contraction phase was also responding more vigorously to secondary papain challenge but also other allergens that they have not encountered before. ILC2m secreted more cytokines than other ILC2s that were being stimulated for the first time. Except recognizing an antigen directly, ILC2m are displaying the same characteristics as a memory T-cell response: quick expansion, a contraction phase and a small, long-lived pool with rapid and enhanced effector functions, plus they expressed a similar gene profile to that of memory T-cells (Martínez-González et al., 2017).  ILC2m residing in mLNs are highly responsive to IL-25 instead of IL-33, as opposed to unstimulated lung ILC2s. Upon primary IL-33 engagement, lung ILC2s upregulate IL-25R. The cells that survive 10  the contraction phase will remain in the lung and mLN as IL-25Rhi, like gut ILC2s. If mLN ILC2m can persist up to five months, it is possible that they can migrate back to the lung upon secondary challenge to display an enhanced cytokine release (Martínez-González et al., 2018). This forces us to reconsider ILC2s as a purely tissue resident cell. Some researchers address this peculiar allergen experience as “trained immunity”. Macrophages, for example, can be trained to be long-lived and better prepared to subsequent infections, however, it remains unknown if they can survive more than 45 days (Martínez-González et al., 2017). Also, as opposed to macrophages and T-cells, ILC2s do not need to undergo differentiation to reach the memory stage, its functions as ILC2m remain very similar to those of unstimulated ILC2s, meaning that they stay committed as type 2 cytokine secretors (Martínez-González et al., 2017). The existence of this population of ILC2m might sensitize individuals to certain allergens and facilitate the development of T-cell independent asthma, or non-allergic asthma. However, the increased production of IL-13 by ILC2m could license DCs with increased efficiency, expand a larger population of TH2 cells and lead to an adaptive response in asthma, or allergic asthma. Perhaps, ILC2m could be a target for future asthma therapies (Martínez-González et al., 2017). Regulatory ILC2s ILC2s serve critical roles in the development of allergic inflammatory reactions like asthma, atopic dermatitis or allergic lung inflammation. To prevent these reactions from “spinning” out of control, the immune system employs mechanisms of strict ILC2 regulation (Duerr et al., 2015). The global theory of TH1 and TH2 responses counter-regulating each other to grant balance and successful homeostasis arose in 1986 and provides the first level of control of ILC2s (Kidd, 2003), which are, in principle, clear representatives of type 2 immunity (of note, their type 2 identity, like every other defining signature of ILC2s, will be challenged later in this review). Type 1 immunity will therefore inhibit the progression of type 2 responses. Recently, two different groups have demonstrated that ILC2 proliferation and function can be effectively suppressed by the pro-inflammatory cytokines 11  IFNα/β, IFNγ and IL-27 (Duerr et al., 2015; Moro et al., 2015). The most potent suppressor of ILC2 function was IFNγ, and this was demonstrated using ifng-/- mice that displayed strong type 2 immunopathology (Duerr et al., 2015). In fact, single nucleotide polymorphisms in the ifng gene or the chains encoding its receptor have been associated to asthma in humans (Kumar & Ghosh, 2008; Pinto et al., 2007; Nakao et al., 2001). An immunosuppressive role for IFNγ in ILC2s implies that they express the IFNγ receptor (IFNγR1/2). The IFNγ receptor IFNγR1 is detectable in low levels in ILC2s (Stehle et al., 2015), but its upregulation mediated by IL-1β could result to an IFNγR1hi ILC2 subset prone to suppression in a pro-inflammatory environment (Shirey et al., 2006). Our laboratory has detected increased expression of this receptor in ILC2s harvested from the lungs of mice with primary tumours grown from a murine, lung epithelial tumour cell line (unpublished data). These results suggest that a systemic pro-inflammatory environment could attenuate ILC2 type 2 functions. Even more intriguing is the capacity of self-regulation that ILC2s have. In 2016, Kim et al. revealed a non T- non B-lymphocyte whose surface markers and tissue residency resembled an innate lymphoid cell. This ILC, however, was a new effector that produced notable amounts of the immunosuppressing cytokine IL-10. It was named ILC10. A year later, this ILC10 was unmasked and revealed an alternatively activated ILC2 subset that produced large amounts of IL-10 (Seehus et al., 2017). This new subset, the ILC210, was also activated by IL-33, but it persisted after a contraction phase when the stimulus was removed, and then was easily expanded upon secondary challenge (Seehus et al., 2017). Interestingly, the ILC210 displays features similar to that of the previously described memory ILC2. Whether they are the same cell or not remains to be elucidated, however, the fact that ILC2m respond with enhanced capabilities suggests a need for tighter regulation, and hence IL-10 production. The frequency of ILC210 correlated to stronger ILC2 activation and elevated eosinophilia, suggesting an immunosuppressive role that can act on itself or other immune cells (Seehus et al., 2017). 12  In 2019, Morita and colleagues reported striking yet compelling evidence of a population of ILC2s that not only produced high amounts of IL-10, but also upregulated the regulatory T-cell (TREG) markers CTLA-4 and CD25 in response to retinoic acid (RA). Although they retained IL-13 and GATA-3 expression, their genetic profile was so deviated from conventional ILC2s that they removed the “2” and termed it ILCREG (Morita et al., 2019). Also, they were demonstrated to genetically differ from the previously mentioned ILC210 subset. RA is known to drive TREG differentiation from naïve T-cells (Sun et al., 2007; Mucida et al., 2009), but it seems to also have a role involving ILCs, where ILC2s transition from a type 2 identity towards a regulatory phenotype. Here, we introduce the first case of full ILC2 inter-subset plasticity. ILCREGs were capable of fully suppressing CD4+ T-cell differentiation (Morita et al., 2019), and thus, they present themselves as a novel ILC2-derived innate regulatory subset. Inflammatory ILC2s An undeniably effective way to define ILC2 plasticity is by bringing together a comprehensive map of ILC2 subsets clustered by tissue-residency, or their differences between tissues (Ricardo-González et al., 2018). In this review we opt for differencing ILC2 subsets based on the acquisition of novel, non-type 2-associated functions and highly unconventional roles. Both approaches however benefit from new emerging technologies like mass cytometry or single-cell RNA sequencing. ILC1 to ILC3 plasticity has been studied in detail to a point where researchers have enough criteria to identify a complex spectrum of identities that end in either fully differentiated ILC1s or ILC3s, and where the direction of such differentiation depends on multiple environmental cues from tissues (Cella et al., 2019; Vonarbourg et al., 2010; Bernink et al., 2013, 2015). Despite this, plasticity between ILC2 and ILC3 subsets remains a large mystery, and so far, the only reported phenotypical shifts are confined to a limited number of interchangeable traits. Notch signalling has proven to be essential both for ILC2 and ILC3 development, however it seems that the strength of the signalling determines the digression into one lineage or the other (Cherrier et al., 2012; Lee et al., 2012). Nevertheless, both 13  effectors strictly depend on TCF-1 to develop and exert their roles (Mielke et al., 2013). Given these developmental similarities, both ILC2s and ILC3s are expected to show some degree of flexibility under the right conditions. As previously mentioned, lung ILC2s are characterized as Lin- IL-25Rlo ST2+ cells that will expand after IL-33 engagement. However, Huang and colleagues identified a scarce and distinct lung resident subset that expanded upon IL-25 stimulation instead, and that expressed high levels of the killer cell lectin receptor KLRG1 (Huang et al., 2015), described as Lin- ST2- KLRG1+ ILC2s. Since this cells were only encountered after IL-25 administration or infection, they were called inflammatory ILC2s (iILC2). This new subset expressed more GATA-3 than conventional ILC2s and secreted similar amounts of IL-13 and IL-5, but more IL-4 (Huang & Paul, 2016). Although the exact progenitor for iILC2s has not yet been identified, what is interesting about these cells is that if cultured under TH2 conditions or during parasite infection in vivo, they evolve into ST2+ conventional ILC2s that downregulate IL-25R, suggesting that iILC2s could be transient progenitors (Huang et al., 2015). Intriguingly, iILC2s also express intermediate amounts of the master regulator RORγt, and begun to resemble an initial ILC3-like identity by producing some IL-17 (but also IL-13), which was drastically increased when cultured under TH17 conditions (Koyasu, 2015). These iILC2 were also highly effective at fighting Candida albicans infection in mice compared to conventional ILC2s, and in vivo they shut down IL-13 production and upturn IL-17 levels, proving an increased induction of plasticity upon fungal infection (Huang et al., 2015). Overall, iILC2s are an excellent example of a highly inducible and plastic subset of ILC2s that develop into an ILC2- or ILC3-like phenotype depending on its environment. The studies by Huang et al. on these novel, inflammatory ILC2s were furthered by three independent laboratories in 2019 that confirmed induction of ILC2 plasticity towards an ILC3 phenotype mediated by IL-1β, IL-23 and TGF-β (Golebski et al., 2019; Ting Cai et al., 2019; Bernink et al., 2019). The presence of RORγt, the downregulation of GATA-3 the co-expression of IL-13 and IL-17A and the 14  reduction in IL-5 in these cells coincide with the iILC2-derived ILC3-like cells described by Huang and colleagues (Golebski et al., 2019). Bernink et al. expanded the knowledge on iILC2-to-ILC3 induction of plasticity by determining the importance of c-Kit expression in ILC2s. Apparently, c-Kit+ ILC2s already express RORγt at baseline, and based on their lower IL-13 production after activation it is safe to assume that they are more prone to transit into an ILC3-like phenotype. On the other hand, c-Kit- ILC2s retain a stronger type 2 identity and require larger amounts of IL-1β, IL-23 and TGF-β to acquire ILC3 properties (Bernink et al., 2019), whereas c-Kit+ ILC2s only need IL-1β and IL-23. These studies add another level of complexity to inflammatory ILC2s.  The implications of these levels of plasticity in asthma were clear since 2010, when Wang et al. discovered an IL-17A+ TH2 effector that contributed to disease severity by promoting TH2 and TH17 responses simultaneously. The innate counterpart of this potent effector was unmasked by Ting Cai and colleagues also in 2019, and it was so powerful at promoting lung inflammation via eosinophil and neutrophil recruitment that it was addressed as the ILC217, a new effector capable of IL-5, IL-13 and IL-17 production in vast quantities (Ting Cai et al., 2019). ILC217s lack RORγt and therefore cannot be identified as iILC2s. Instead, their acquisition of ILC3 identity (retaining type 2 identity) depends on the expression of the aryl hydrocarbon receptor (Ahr) along with sustained IL-33 signalling, as opposed to a single stimulation (Ting Cai et al., 2019). It is fascinating that within the realm of ILC2-ILC3 plasticity research has identified at least two different subsets of unconventional ILC2s: the iILC2 and the ILC217.  Type 1 ILC2s The abandonment of the classical type 2 identity by ILC2s towards a more pro-inflammatory, TH1-like phenotype has also been reported by several studies (Ohne et al., 2016; Silver et al., 2016; Lim et al., 2016; Bal et al., 2016). The phenomenal abilities of ILC2s to acquire ILC3 properties also apply to ILC1 cells, and again, a carefully measured series of cues in the form of cytokines can make this 15  shift happen. In an attempt to prove this possible, Ohne and colleagues stimulated human ILC2 cultures with IL-1α and IL-1β, central mediators of the inflammatory response (López-Castejón & Brough, 2011). Unexpectedly, and in the presence of IL-2, IL-1β via IL-1R1 induced significant expression of IL-5 and IL-13 just like IL-33 did, as well as notable levels of the immune suppressor IL-10 and GM-CSF (Ohne et al., 2016). This situates IL-1β as a perfectly valid ILC2 activator besides IL-33, IL-25 and TSLP. Interestingly, our group has also detected important levels of IL-10 and GM-CSF in culture supernatant cytokine profiling analyses. Besides classical ILC2 activation, important genes involved in adaptive immunity like MHC-II, CD80 and CD40L were upregulated. In the presence of IL-1β with IL-12, however, GATA-3 expression is reduced in favour of an increment in T-bet and abundant IFNγ, confirming attainment of ILC1 identity by ILC2s (Ohne et al., 2016). Soon after, it was clarified that such induction of plasticity is directly mediated by IL-12, but it is IL-1β the responsible for IL-12Rβ2 upregulation in conventional ILC2s. Furthermore, inclusion of IL-4 in culture reversed these plastic properties by promoting STAT6, which in turn induces GATA-3 (Bal et al., 2016, 2020; Lim et al., 2016).  An innate effector that is heavily involved in mediating immunity against parasites with capacity to acquire ILC1 pro-inflammatory properties has tremendous repercussions in disease, but also in the global ILC paradigm. In lung disorders such as tobacco smoke-mediated chronic obstructive pulmonary disease (COPD) or severe influenza infection, a systemic inflammatory environment rich in IL-1β and IL-12 is necessary to mount an adaptive TH1 response. As previously mentioned, the action of these cytokines on conventional lung ILC2s will result in GATA-3 downregulation, a halt in IL-5 and IL-13 production, T-bet expression and IFNγ secretion. In other words, the backbone of TH1 immunity. Silver and colleagues have studied this kind of plasticity on the context of disease, and their results demonstrate that there is a clear derivation of fully transformed ILC1 cells from the local ST2+ ILC2 pool, and that the newly generated ILC1 population dramatically amplifies anti-viral immunity in the lungs of mice. Furthermore, these ex-ILC2s gathered around the influenza-infected 16  foci in the lungs to combat viral expansion (Silver et al., 2016). If lung ILC2s have indeed a significant role in antiviral immunity, it goes without saying that characterizing their response in the context of the novel 2019 coronavirus disease (COVID-19) is a matter of urgent interest and a unique opportunity to pioneer an absolutely unexplored field of ILC2 biology.  Despite the fascinating evidence that unveils a new pro-inflammatory subset of ILC2s, there is still not a unified nomenclature for it. Perhaps because the plasticity towards the type 1 phenotype is so polarized that it can be directly addressed as an ILC1. Nevertheless, as inflammatory ILC2 is already taken, this review proposes the term Type 1 ILC2 or ILC21 for those conventional ILC2s with potential to acquire ILC1 properties, in an intermediate state of plasticity, or as a substitute for ILC1 ex-ILC2s. All the previous work on ILC21s gives rise to a new subset that fiercely challenges the current dogma that situates ILC2s as a homogeneous group of type 2 innate effectors. These cells have potential to acquire an IFNγ+ pro-inflammatory phenotype that could be harnessed to mitigate strong allergic inflammations, atopic disorders or asthma. In addition, they are proven effective agents in the pulmonary antiviral response, but do they participate at all in cancer?  Our group reveals that type 1 ILC2s comprise a subpopulation of ILC2s that may be responsible for tumour elimination in mice by directly activating CD8+ T-cells (Saranchova et al., 2016, 2018) through peptide antigen cross-presentation, the mechanism that DCs employ in order to mount adaptive anti-tumour and antiviral responses (unpublished data). Using single-cell RNA sequencing analysis, we have characterized this population and identified previously mentioned ILC21 markers such as MHC-II and IFNγ, but also granzyme B (unpublished data), which has only been reported once to be expressed intracellularly by ILC2s (Camelo et al., 2017).  The addition of granzyme B to this pro-inflammatory profile pushes the boundaries of ILC2 plasticity towards the NK cell realm. We re-evaluate how ILC2s participate in cancer in the next and final section.  17  The role of ILC2s in cancer Over the last few years, most of the scientific work allocates ILC2s a marked pro-tumorigenic character. While ILC1s and ILC3s keep presenting a controversial role in cancer (Gao et al., 2017; Vacca et al., 2018; Mattner & Wirtz, 2017), there is consensus on the idea that ILC2s are foes in the anti-tumour response, where their type 2 functions derive in detrimental responses for immunity and tumour containment. ILC2s have continuously surprised the scientific community with novel roles that seem beyond unconventional, and cancer is no exception, however, the evidence that situates them in the tolerogenic side of anti-tumour immunity is abundant. Pro-tumorigenic evidence The IL-33/ST2 axis seems to promote tumour progression in several types of cancer. In human and mouse colorectal cancer, for example, IL-33 released from the tumour epithelium activate subepithelial myofibroblasts and mast cells, which in turn enhance metalloproteinase and growth factor production, minimized apoptosis and promote angiogenesis (Maywald et al., 2015). This is also the case for breast cancer, where the involvement of IL-33 in a newly discovered signaling pathway promoted epithelial cell transformation and breast cancer progression (Kim et al., 2015). Studies with the breast cancer model T41 revealed that IL-33 expression in tumour cell increased as it progressed to metastasis, and that resulted in reduced tumoricidal infiltration of NK cells and elevated numbers of myeloid-derived suppressor cells (MDSCs) and M2 polarized macrophages. Higher concentrations of circulating IL-13 and IL-10 also contributed to metastasis in the liver and lungs (Jovanovic et al., 2014). The fact that ST2 deletion by microRNAs leads to gastric cancer metastasis or the observation of increased IL-33 in the plasma of early lung cancer patients suggests a role for IL-33 in tumour progression, but does not explain the mechanism behind it (Kim et al., 2015; Yu et al., 2015;).  18  The evidence presented for the IL-33/ST2 axis in tumour progression led the research community directly to ILC2s. Coming back to gastric cancer, a group of researchers identified that patients experienced a powerful induction of the TH2 phenotype, including elevated amounts of ILC2s and type 2-associated cytokines, which were blamed for a hypothesized suppression of the TH1 response (Bie et al., 2014). The presence of activated conventional ILC2s in the peripheral blood of acute promyelocytic leukaemia patients was associated to increased IL-13 release upon engagement between the NKp30 receptor expressed on ILC2s and its ligand B7H6, along with tumour-derived prostaglandin-2 binding the CRTH2 receptor on ILC2s. Elevated concentrations of IL-13 led to recruitment of IL-13Rα1+ MDSCs known to have a potent immunosuppressive potential (Trabanelli et al., 2017). This IL-13 – mediated ST2/MDSC axis was reinforced by a study that involved 131 patients that showed a strong correlation between IL-13 and MDSC numbers in gastric, pancreatic and esophageal cancer (Gabitass et al., 2011). Even though correlation does not necessarily entail causation, one must take into account that the immunosuppressive capacity of MDSCs rockets in the presence of IL-13 in vitro (Highfill et al., 2010). More recently, the ILC2/MDSC axis was extended to lung and bladder cancer and was associated to its recurrence in recovered patients (Wu et al., 2017; Chevalier et al., 2017).  Another detrimental role of IL-13 is its facility to promote and maintain tumour-associated macrophage function, which are highly related to M2 macrophages when it comes to tumour promotion (Munn & Bronte, 2016). In addition, in its homeostatic efforts to remodel and maintain healthy lung tissue during infection or in the presence of a tumour, the ability of IL-13 to drive the development of fibrosis in lung or liver tissue could culminate in cancer development (McHedlidze et al., 2013; Mattner & Wirtz, 2017). Increased pulmonary fibrosis in turn expresses larger amounts of IL-25, which will further activate ILC2s (Hams et al., 2014). Another potential pro-tumorigenic role of ILC2s involves the production of amphiregulin, a ligand for the epidermal growth factor with 19  demonstrated capabilities to endorse lung tumour cell growth and survival, but also an enhancer of TREG function (Busser et al., 2011; Zaiss et al., 2013).  Anti-tumorigenic evidence As expected from a cell whose heterogeneity has been progressively unmasked, and like the other ILC family members, ILC2s and the cytokines related to their activation also seem to play very interesting roles in anti-cancer immunity. IL-5 is a type 2 cytokine secreted by ILC2s that plays a major role in eosinophil recruitment and activation (Abbas et al., 2018), and is responsible for the first evidence demonstrating that ILC2s can participate in the process of immune surveillance (Trabanelli et al., 2019). To trace IL-5 producers in the context of malignant melanoma, Ikutani and colleagues (2012) generated IL-5/Venus knock-in mice. Most of the Venus+ cells were located in the lungs and produced high amounts of IL-5 and IL-13, resembling an ILC2 phenotype. These mice had a reduced number of metastatic nodules compared to the mice that lacked IL-5 production by Venus+ cells (Ikutani et al., 2012). This, along with the study by Simson et al. in 2009 proposing a role for eosinophils in tumour surveillance, suggests that the eosinophilia mediated by IL-5 can play a role against malignant melanoma.  Slowly hinting a role for ILC2s against tumours, three other studies have found compelling evidence for IL-33 aiding in the mounting of an adaptive response mediated mostly by CD8+ cytotoxic lymphocytes (CTL). First, Gao and colleagues declared that IL-33 could be a double-edged sword when it comes to tumour immunology (Gao et al., 2013). By over-expressing IL-33 in B16 melanoma and LLC lung carcinoma, metastasis was inhibited. This was possible due to a marked infiltration of CD8+ T-cells and NK cells in the tumour. They attributed this TH1-like response to IL-33 only by activating the NF-κB pathway in CD8+ T-cells (Gao et al., 2013). However, this alone is insufficient to mount a CTL proliferative response. A year later, a different group obtained a very similar set of results, where overexpression of IL-33 in B16 melanoma cells not only led to a specific CTL response, 20  but also incremented the expression of MHC-II in melanoma cells (Gao et al., 2014). This intriguing, contradictory role of IL-33 in tumour immunity was finally elucidated by our group in 2016, where Saranchova and colleagues discovered a new mechanism of metastatic immune escape related to the loss of IL-33 expression in tumour cells (Saranchova et al., 2016). While other groups were overexpressing IL-33 to understand its effects in different cancers, our laboratory realized that when primary lung and prostate murine cell lines (TC1 and PA respectively) transit into their metastatic derivatives (A9 and LMD respectively), they downregulate the expression of IL-33. Interestingly, we also discovered that MHC-I expression is tied to that of IL-33, meaning MHC-I levels are rescued upon IL-33 overexpression in metastatic cell lines leading to a reverse of metastasis in vivo (Saranchova et al., 2016). A common mechanism of cancer immune evasion is by downregulating MHC-I in tumour cells and hence rendering the tumour invisible to the immune system (Hanahan & Weinberg, 2011). While the mechanisms of this transcriptional link remained unknown, this mechanism of immune escape towards metastasis explains the CTL responses induced by IL-33 overexpression. In accordance to previous work, we have also detected tumour infiltration of eosinophils and CD8+ and CD4+ T-cells in IL-33+ tumour cells.  This discovery involving IL-33 took us to the next step: determining whether ILC2s are actually a component of IL-33 – mediated anti-tumour immunity. In 2018, Saranchova and colleagues established, for the first time, a direct role for ILC2s in CTL-mediated immune surveillance (Saranchova et al., 2018). As expected, IL-33+ primary tumour cells and metastatic cells overexpressing IL-33 displayed elevated numbers of intratumoral ILC2s. In a mouse model that lacks ILC2 cells (RORα-/- mice), the volume of the IL-33+ A9 metastatic tumour was significantly higher than in WT mice, demonstrating a role for ILC2s attacking the tumour in direct or indirect ways in vivo. The percentage of circulating tumour cells, an indicative of metastasis, was also higher in RORα-/- mice. Furthermore, ILC2 numbers in the lungs of mice with metastatic tumours were very low, suggesting a migratory pattern of recruited ILC2s towards the IL-33 – expressing tumour (Saranchova 21  et al., 2018). Again, challenging their condition of tissue-resident effectors. Furthermore, we also proved a direct link between ILC2s and CD8+ T-cells by observing elevated levels of granzyme B and increased cellular killing (seen by microscopy) for CD8+ T-cells in the presence of ILC2s but not in their absence (Saranchova et al., 2018). Although another study by Kim et al. in 2016 has also demonstrated the halt of tumour growth by intratumoral ILC2s, they attribute it to IL-33 – mediated release of CXCR2 chemokine ligands from ILC2s, which initiate a pro-apoptotic pathway in CXCR2+ EL4 lymphoma cells, but not a CTL response.  Cross-presenting ILC21 ILC2s are type 2 mediators of allergic lung inflammation and helminth elimination, but their extraordinary plastic properties have yielded worldwide evidence that situates them as a phenomenally heterogeneous population with multiple subsets. This, of course, is also reflected in cancer, which is certainly one of the most physically and mentally devastating conditions a patient can endure. The IL-33/ILC2 axis is a double-edged sword. It is implicated in promoting a powerful immunosuppressive environment, and thus, pro-tumorigenic, but also a new and effective agent in the anti-tumour response. Following up on Saranchova’s work, our laboratory has been working on the elucidation of the ILC2-mediated mechanism of tumour clearance. If ILC2s can initialize a CTL response, how do they do it? Our work has identified a population of ILC2s that are derived from primary and metastatic tumour-bearing mice capable of antigen cross-presentation to CD8+ T-cells. Unlike those conventional ILC2s derived from wild type animals, those harvested from the lungs of tumour-bearing mice have proven effective at OVA-derived SIINFEKL peptide (OVA257-264) cross-presentation and subsequent cross-priming of CD8+ OT-I cells, which displayed a proliferative response (unpublished data). Furthermore, we performed single cell RNA sequencing analysis on those ILC2s derived from tumour-bearing mice, specifically from the lungs of mice with IL-33+ primary tumours, and 22  discovered a highly heterogeneous population identified by our t-SNE plots. One of the clusters downregulated GATA-3, IL-5 and IL-13, but showed higher expression of IFNγ and granzyme B, suggesting either a pro-inflammatory role, cytotoxicity, or both. Moreover, that same subpopulation upregulated MHC-II related genes and the invariant chain CD74 (unpublished data), both genes essential to establish dialogue with T-cells and to cross-present (Oliphant et al., 2014; Basha et al., 2012).  Based on previous work, we suspect that when ILC2s are in the presence of a tumour that, as a by-stander effect, drives systemic inflammation, or as Saranchova and colleagues proved, recruits ILC2s, receive cues from the microenvironment that induce the initiation of a plastic transcriptional program. In our laboratory, the ILC2s that we have observed, and their properties qualify them as mediating agents of the TH1 immune-surveillance response, far from their conventional type 2 identity, but closer to the ILC1 phenotype, and thus the type 1 ILC2 or ILC21 subset that we previously proposed. However, our cells express granzyme, which may grant cytotoxicity, but most importantly cross-present. Our ILC2s from tumour-bearing mice are therefore a potential new subset within the ILC21s, or perhaps an entirely novel ILC type, that has the capacity to activate a CTL response directly that targets tumour with antigen specificity. Concluding remarks Once again, the field of Immunology proves us that every new discovery entails uncountable levels of complexity and more questions than answers, but also beauty and an infinity of opportunities to learn. Here, we have defined an unconventional group of effectors: the innate lymphoid cells, with special focus on the remarkable type 2 innate lymphoid cell or ILC2. Their developmental origin qualifies them as lymphoid, and most of their functions allocates them in the innate branch of immunity. For now, not much more can be assured about ILC2s. Recently classified as type 2 innate effectors, conventional ILC2s are proven fundamental in other non-canonical roles such as tissue 23  remodelling and wound healing across numerous tissues. In theory, ILC2s are a fixed, homogeneous population, but in practice they are flexible. Their transcriptional program is highly responsive to environmental cues, triggering the induction of intra- and inter-subset plasticity. Research has already identified memory ILC2s, regulatory ILC2s, inflammatory ILC2s with ILC3-like properties, type 1 ILC2s involved in antiviral and anti-tumour responses and cross-presenting ILC2s. Even without acquisition of plasticity, conventional ILC2s are still immensely heterogeneous across tissues. Such a flexible identity is reflected in the way they are involved in cancer: pro-tumorigenic cells, but also anti-tumorigenic. This review has dived beyond the unconventional, and leaves the initial question as open as ever. What do we really know about ILC2s?   24  Hypothesis and Objectives  Type 2 innate lymphoid are a heterogeneous population of cells that responds to changes in the surrounding environment, such as the presence of a tumour in the organism. This will shift the type 2 identity of a subset of the lung ILC2 population towards a pro-inflammatory phenotype that is also capable of cross-presenting antigen to CD8+ T-cells, and hence contributing in the process of immune surveillance as a type 1 innate effector.  Objective 1: To optimize ILC2 isolation and in vitro culture to find a way to work with manageable numbers. Objective 2: To assess the cross-presenting and cross-priming potential of ILC2s derived from wild type and tumour-bearing mice. Objective 3: To analyze a population of ILC2s from the lungs of mice with primary tumours via single cell RNA sequencing and demonstrate high cell heterogeneity.         25  Materials and Methods Tissue processing and ILC2 isolation ILC2s were isolated from the lungs of C57Bl/6 wild type mice treated with intraperitoneal recombinant mouse IL-33 injections to maximize ILC2 differentiation and harvest. Mice were treated on days 0, 2 and 4 with 2ng/µL of rmIL-33 (R&D Systems #3626-ML-010) prior to isolation on day 6. Lungs were then harvested and cut into small pieces using a razor blade and digested for 30 minutes at 37°C in a shaker platform (200rpm) in 10mL of digestion media per 5 pairs of lungs. The digestion medium employed contained RPMI 1640 (Gibco #11875-093) with penicillin-streptomycin (P/S - Gibco #15140-122) and 10% fetal bovine serum (FBS - Gibco #A3160501), as well as collagenase/hyaluronidase and DNase I (StemCell #07912 and #07900 respectively). The pieces of lung tissue were placed onto a 70µm cell strainer. Using the plunger end of a 3mL syringe, the tissue was mashed through the strainer and rinsed with 5mL RPMI 1640 to a total of 15mL. Cells were centrifuged for 6 minutes at 1637rpm. The supernatant was carefully removed, and the pellet was re-suspended in 20mL ammonium chloride solution (StemCell #07800) and incubated at room temperature for 5 minutes to lyse the erythrocytes. After neutralization with 30mL of FACS buffer cells were counted and washed (6 minutes, 1637rpm) in a total volume of 50mL (full 50mL Falcon Tube). FACS buffer was made of DPBS (Gibco #14190-136) with 2% FBS.  After removing the supernatant and re-suspending in the appropriate volume of FACS buffer (108 cells/mL), cells were enriched for ILC2s using an EasySep™ Mouse ILC2 Enrichment Kit (StemCell #19842), which reduces sorting time and increases ILC2 recovery both from naïve and IL-33 – treated lungs (Martínez-González et al., no date).  Prior to sorting, the cells were stained with FITC-conjugated lineage marker mouse antibodies purchased from Invitrogen (CD3ε/γ, CD4, CD8α, CD19, TCRβ, NK1.1, TER119, CD11c, CD11b and Ly-6G/C) and the ILC2 markers purchased from BioLegend: PE-conjugated CD127 (IL-7 26  receptor), PerCP-Cy5.5-conjugated ST2 (IL-33 receptor), BV605-conjugated Thy1.2 and BV421-conjugated CD45. ILC2s were then phenotypically assessed and sorted using a BD FACS Aria II machine to obtain a pure population. Mouse ILC2 in vitro culture and analysis For optimal growth ILC2s were cultured in serum-free StemSpan™ SFEM II (StemCell #09655) supplemented with 100ng/mL of recombinant mouse IL-33 and TSLP for ILC2 activation (Invitrogen #14-8332-80 and #14-8498-90 respectively), 25ng/mL of rmIL-7 to promote survival (R&D Systems #407-ML-005) and 1000 Units/mL of recombinant human IL-2 to enhance proliferation (StemCell #78036.3). The media also contained P/S.  For experimental purposes, ILC2s were also cultured in RPMI 1640 with 10% FBS supplemented with the same cytokines and antibiotics. In both scenarios cells were kept at 37°C during experiments. Cells were cultured in U-bottom, non-treated 96- or flat bottom 24-well plates. Fed-batch wells had half of their volume of media changed every other day and 50% of the wells were triturated with a pipette to avoid cell clumps in all experiments.  ILC2s in culture were counted every day using a Cedex HiRes® automated cell counter. Prior to loading a 300µL sample into the Cedex Sample Cups, cultures were pipetted to disaggregate cell clumps. Samples were usually diluted with fresh media to 1:2 ratios to facilitate reads. The precision level was set to superior, which allows for approximately 8 images for analysis per sample and the cell type was set as lymphocyte, which is the minimum size and appropriate for analysis of cells with a diameter of about 6µm, such as ILC2s.  Glucose and lactate levels were measured every day by analyzing 90µL of ILC2 culture supernatant using a Stat Profile® pHOx® Ultra Bloodgas Analyzer from Nova Biomedical. Photos of the ILC2 culture in StemSpan™ were taken using a Canon EOS DSLR camera attached to a Motic confocal 27  microscope. Images were processed using the software Camera Control Pro from Nikon and ImageJ v1.8.0. and figures produced in GraphPad Prism v8.0.0. Calculation of ILC2 growth rate and doubling time The average cell growth rate was calculated based on the exponential phase of three different ILC2 cultures in 1mL of media in a 24-well plate. The equation employed is the following: 𝐶𝑓 = 𝐶𝑖 × 𝑒𝑘(𝑡𝑓−𝑡𝑖) Where 𝑘 represents the cellular mean growth constant or growth rate, 𝐶𝑖 the initial cell number and 𝐶𝑓 the final cell number at the end of the exponential phase. Initial time (h) is depicted by 𝑡𝑖 and represents the beginning of the culture, whereas 𝑡𝑓 is the time that has passed at the end of the exponential phase. Doubling time (𝑡21) was calculated with the following equation: 𝑡21=  𝑙𝑛2𝑘 Tumour cell culture and establishment The murine, lung primary tumour cell line (TC1) and its metastatic derivative (A9) were derived from lung epithelial cells and were grown at 37°C in Dulbecco’s modified Eagle medium (DMEM - Gibco #11995-065) supplemented with 10% FBS and P/S. 24h prior to tumour establishment, media was changed to DMEM with 10% FBS with no antibiotic supplementation. The day of tumour establishment, adherent cells were washed with DPBS, lifted using TrypLE™ Express Enzyme 1X (Gibco #12605-010), spun at 1500rpm for 3 minutes and resuspended in Hank’s Balanced Salt solution 1X (HBSS - Gibco #14170-112) for injection.  TC1 and A9 cells were injected via subcutaneous injection in the right flank of WT C57Bl/6 mice. All mice were female, between 4-8 weeks of age, kept in pathogen-free facilities and their use was approved by the Animal Care Committee at UBC. For all experiments, 50µL of 6 × 105 cells were 28  used per mouse. Tumour growth was monitored by measuring tumour volumes with calipers three times a week using the following formula: 𝑇𝑢𝑚𝑜𝑢𝑟 𝑣𝑜𝑙𝑢𝑚𝑒 =  𝑙𝑒𝑛𝑔𝑡ℎ × 𝑤𝑖𝑑𝑡ℎ2 With the length (mm) being the longer axis of the tumour. The mouse weight was also recorded at the time of the measurement. Humane endpoint followed by euthanasia was based on a minimum of 20% reduction in body weight, a tumour volume equal or higher than 1cm3 or ulceration of the tumour. Tumour monitoring up to humane endpoint lasted typically 21 days for A9 cells and up to 38 days for TC1 cells. ILC2 in vitro culture and cross-priming assay For optimal growth ILC2s were cultured in serum-free StemSpan™ SFEM II (StemCell #09655) supplemented with 100ng/mL of recombinant mouse IL-33 and TSLP for ILC2 activation (Invitrogen #14-8332-80 and #14-8498-90 respectively), 25ng/mL of rmIL-7 to promote survival (R&D Systems #407-ML-005) and 1000 Units/mL of recombinant human IL-2 to enhance proliferation (StemCell #78036.3). The media also contained P/S. For experimental purposes, ILC2s were also cultured in RPMI 1640 with 10% FBS supplemented with the same cytokines and antibiotics. In both scenarios cells were kept at 37°C for 5-7 days prior to use. After 5-7 days in culture (depending on their growth) and 24h before flow cytometric analysis, ILC2s were pumped with 0.5mg/mL of full ovalbumin protein (OVA – Worthington #P5B7908) or 10µg/mL of the OVA peptide (257-264) known as SIINFEKL (GenScript #RP10610-1). The next day, ILC2s were stained with an antibody that detects SIINFEKL in the context of MHC-I only (BioLegend #141605) and analyzed by flow cytometry to detect any processed exogenous whole OVA presented 29  on MHC-I as SIINFEKL. The program FlowJo v.8.10 was always used for flow cytometry data analysis. Dendritic cell harvest and differentiation Bone marrow (BM) precursors were harvested from C57Bl/6 WT femur BM on day 0 and cultured at 37°C in RPMI 1640 supplemented with 10% FBS, P/S, L-Glutamine (Gibco #25030-081) and 2-Mercaptoethanol (Gibco #21985-023). BM precursors were treated with 6ng of mouse GM-CSF (StemCell #78017.1) on days 1, 3 and 6 to promote differentiation into dendritic cells (DCs), which were then collected on day 10 to obtain a population at an intermediate stage of maturation. Bone marrow DCs were adherent in culture and it required TrypLE™ Express Enzyme 1X to lift them for downstream applications. BM-DCs were analyzed by flow cytometry for CD11c and MHC-II surface markers (antibodies: Invitrogen #11-0116-41 and BioLegend #107613 respectively). CD8+ T-cell isolation and culture CD8+ OT-I lymphocytes were harvested from the spleens of female OT-I mice with a C57Bl/6 background. OT-I cells have a transgenic T-cell receptor that is restricted to the processed OVA peptide (257-264) SIINFEKL (Clarke et al., 2000).  Spleens were placed on a 70µm cell strainer inside a 100 x 20mm Petri dish. Using the plunger end of a 3mL syringe, the tissue was mashed through the strainer and rinsed with 5mL of DPBS. The cell suspension was then transferred to a 15mL Falcon tube and the OT-I cells were isolated using an EasySep™ Mouse CD8+ T-Cell Isolation Kit (StemCell #19853A). Cells were then cultured in Advanced RPMI 1640 (Gibco #12633-012) supplemented with 10% FBS, P/S, L-Glutamine, 100 Units/mL of rhIL-2 and 25ng/mL of IL-7 at 37°C for 1-5 days depending on the assay. Cells were phenotypically assessed by flow cytometry with antibodies against surface markers CD8α and TCRβ (Invitrogen #MCD0817 and #11-5961-82 respectively). 30  ILC2 and CD8+ lymphocyte co-culture and in vitro cross-priming assay Freshly isolated CD8+ OT-I cells were stained with proliferation marker CellTrace™ CFSE (Invitrogen #C34570) prior to co-culture at 37°C for 3 days in Advanced RPMI 1640 with ILC2s that have been pumped with exogenous OVA and SIINFEKL 24h before. ILC2s were harvested as previously described from the lungs of C57Bl/6 mice with primary tumours (TC1 cell line), metastatic tumours (A9 cell line) or no tumours. Co-cultures took place at a 1:4 ILC2 to OT-I ratio. If co-cultures started at day 0, OT-I cell proliferation was assessed on days 2 and 3 by flow cytometry to detect the rounds of proliferation flagged by CFSE in the OT-I T-cells. To further ensure that the proliferation displayed does not include any ILC2, co-cultures were also stained with anti-CD8 antibodies. Intracellular staining for flow cytometry ILC2s derived from TC1 tumour-bearing mice (pILC2s) were fixed and permeated using an intracellular fixation and permeabilization buffer set (eBioscience #88-8824-00) and then stained in PBS (2% FBS) with the following mouse antibodies: PE-conjugated granzyme B antibody (BioLegend #372207) and APC-conjugated granzyme C (BioLegend #150803). The IFITM1 mouse monoclonal antibody was purchased from Proteintech (#60074-1-Ig) and conjugated with a FITC antibody labeling kit (Mix-n-Stain #92295). The CD3γ mouse monoclonal antibody was purchased from Bioss Antibodies (#bsm-54300R) and conjugated with an APC-CF 750T antibody labeling kit (Mix-n-Stain #92311A). Both antibodies were conjugated at The University of British Columbia Antibody Laboratory.    31  Cytokine production assays  The secretion of IL-5 and IL-13 was assessed by enzyme-linked immunosorbent assays (ELISAs) according to the manufacturer’s protocol. Recombinant Murine IL-5 (PeproTech #215-15) and Recombinant Murine IL-13 (PeproTech #210-13). RNA Isolation and sequencing Total cellular RNA was extracted from A9, TC1 cells and ILC2s using a RNeasy™ Mini Kit (Qiagen #74104). Samples were sent to the Biomedical Research Centre Sequencing Core at UBC. Single cell RNA sequencing sample preparation For this experiment we only employed TC1 cells. After tumour establishment, tumours were grown in groups of 5 WT C57Bl/6 mice (30 mice in total) for two, three and four weeks. After two weeks, ILC2s from a group of five mice were pooled together and cultured in StemSpan™ SFEM II (StemCell #09655) with no supplemental cytokines except for 1000 Units/mL of recombinant human IL-2 (R&D Systems #407-ML-005) and P/S. This would be the non-activated population and was sent immediately for analysis. Simultaneously, ILC2s from another group of five mice were isolated and cultured in StemSpan™ SFEM II supplemented with rhIL-2, 100ng/mL of recombinant mouse IL-33 and TSLP for ILC2 activation (Invitrogen #14-8332-80 and #14-8498-90 respectively) and 25ng/mL of recombinant mouse IL-7 to promote survival (R&D Systems #407-ML-005). This was the activated population and was cultured for 3 days at 37°C prior to scRNA-seq. The same procedure was repeated at weeks 3 and 4.    32  The scRNA-seq and data analysis ILC2s from mice with TC1 primary tumours were sorted based on ST2, CD127, CD45 and Thy1.2 expression. Sorted cells were sequenced using the 10x Genomics platform and the Cell Ranger pipeline was used to align reads and generate matrices. Our sample quality checks, and further downstream analyses were ran using the R software package Seurat. We followed the recommended default setting of the standard workflow (Butler et al., 2018) for each set of sequencing data, except for our week 2 non-activated sample. Due to the number of cells sequenced, we relaxed the filtering parameter to maintain a robust number of cells for downstream analysis: we filtered cells with unique feature counts over 7000 and less than 50 (instead of over 2500 and less than 200 for default). The resolution for the granularity of the clustering is set to 0.5 – 0.6 so the number of clusters for each dataset is between 5-7. Data visualization is performed with Seurat embedded functions and EnhancedVolcano package. For the integrated cohort analysis, we followed the standard integration workflow (Butler et al., 2018) to combine all datasets into the cohort data for downstream analysis.   33  Chapter 1 Defining Parameters in Serum-Free Medium for the Expansion and Study of Murine Lung Type 2 Innate Lymphoid Cells Introduction Innate lymphocytes (ILCs) are the innate equivalents of T lymphocytes. Based on their function and protein expression, they can be grouped into three different categories of ILCs: types 1, 2 and 3 (Vivier et al., 2016). Type-2 innate lymphoid cell (ILC2) are rare cells that are known as type 2 cytokine secretors in response to helminthic infections, but also major drivers of allergen sensitization and allergic, lung inflammation (Bernink et al., 2014). ILC2 engagement is initiated at mucosal sites by epithelial cell – derived alarmins: thymic stromal lymphopoietin (TSLP), IL-25 and IL-33. (Carriere et al., 2007; Martínez-González et al., 2015). Lacking adaptive antigen receptors, ILC2s sense the microenvironment via cytokine receptors, and in response regulate immunity by secretion of a large number of cytokines (Mirchandani et al., 2014; Gasteiger & Rudensky, 2014; Halim et al., 2014; Barlow et al., 2011; Roediger et al., 2013; Moro et al., 2010; Ikutani et al., 2012). ILC2s require IL-33 in the microenvironment for their development and Activation (Moro et al., 2010; Neill et al., 2010; Eberl et al., 2015). IL-33 not only activates ILC2s, but also promotes eggress of newly generated ILC2s from the bone marrow to the lungs (Stier et al., 2018) and leads to systemic ST2+ (IL-33R) ILC2 expansion (Riedel et al., 2017). Also, ILC2s modulate adaptive immune responses by cell-to-cell contact through the expression of MHC-II molecules (Mirchandani et al., 2014; Oliphant et al., 2014).  ILC2s originate from a Lin- CD127+ Flt3+ common lymphoid progenitors from the bone marrow that can be isolated and induced to differentiate into ILC2s via IL-33, IL-7 and Notch signaling (Licona-Limón et al., 2013; Wong et al., 2012). ILC2s are generally tissue resident cells that can be found in the lungs, adipose tissues, mesenteric lymph nodes, intestine, liver, spleen, blood and meninges 34  (Walker et al., 2013; Krabbendam et al., 2018; Zhang et al., 2017; Silver et al., 2016). The most common source for “conventional” ILC2s is the lungs, where most studies harvest from (Halim et al., 2012, 2014; Kabata et al., 2013; Motomura et al., 2014; Halim & Takei, 2014). ILC2s express transcription factor GATA-3 and in its absence the development and function of ILC2s is impaired (Hoyler et al., 2012; Vivier et al., 2018; Furusawa et al., 2013; Mjösberg et al., 2012). In a departure from the existing dogma, the function of ILC2s in TH1 responses against tumours was recently discovered (Saranchova et al, 2018). The data supports a new paradigm for cancer immunology where the IL-33/ILC2 axis acts as a TH1 effector pathway that assists and mediates anti-cancer CTL immune responses and reduces tumour metastasis. ILC2s thus may have important therapeutic utility amongst the multitude of emerging cancer immunotherapies and this motivated our studies on how to culture and expand ILC2s. ILC2s are a challenging cell population to explore experimentally due to their scarcity (Duerr & Fritz, 2017). In mice, ILC2s represent 0.25-1% of the total leukocyte population in the lung (Drake & Kita, 2014), which appears to be a depot for these cells as the numbers in the lung decrease after peripheral activation. Female mice that have been reported to harbour greater ILC2 numbers than male potentially due to a lack of the maturity marker KLRG1, which inhibits ILC2 function upon binding to E-cadherin (Kadel et al., 2018). Intraperitoneal or intranasal IL-33 pre-treatment has proven to increase ILC2 frequency in the lungs that increases the  yield during isolations (Halim & Takei, 2014). In addition to the low number of ILC2s that current isolation protocols yield, it has proven difficult to culture and expand primary ILC2s in vitro, where sub-optimal media conditions or inadequate cytokine combinations can easily plummet cell viability and provoke a decreased longevity in culture  (Moro et al., 2015). Here we isolated  ST2+ IL-25Rlo lung resident ILC2s and expanded them in serum-free media containing IL-33, TSLP, IL-2 and IL-7. This culturing methodology will enable future research and applications for ILC2s in adoptive transfer approaches and other settings. 35  Results Mouse ILC2s can be optimally expanded and sub-cultured in serum-free media  The need to continuously work with type 2 ILC2s and the considerable difficulties that their isolation and culture require lead us to explore alternative protocols to yield higher numbers of usable ILC2s from mouse lungs. We also examined the optimal cell concentrations we could seed using the best possible culture media with the right supplementation.  In mice, ILC2s are harvested from the lungs, but the yield is usually very low. To solve this issue, many research groups pre-treat the mice prior to isolation with intra-peritoneal injections of alarmins such as TSLP, IL-25 and IL-33, which significantly increases the yield per mouse (Bruce et al., 2017; Camelo et al., 2017). In our experience, non-pre-treated isolations yield approximately 1,000 cells per mouse, whereas from alarmin pre-treated mice it is possible to obtain 150,000 – 1,000,000 cells per animal (dose dependent). For these experiments, we have decided to pre-treat mice with 600ng of murine IL-33 one week prior to isolation injecting every other day for a total of three days (one 200ng injection each day).  We batch-cultured ILC2s in 24- and round bottomed 96-well plates both in RPMI supplemented FBS with IL-33, TSLP, IL-2 and IL-7 and serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 and as shown in Figure 1A, the cell concentration only increased where serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 was employed, reaching up to 106 viable cells/mL in 5 days from an initial explant of 105 cells/mL. To further support this finding, we measured the glucose consumption and subsequent lactate production in both cultures (Figure 1B), and only the wells containing serum-free media displayed a shift in the concentration of both molecules. Glucose and lactate levels remained static in the ILC2 cultures with RPMI supplemented FBS with IL-33, TSLP, IL-2 and IL-7. The cell concentration densities in both media is illustrated in Figures 1C & 1D. In addition, cell morphology in serum-free medium supplemented with IL-33, TSLP, IL-2 and 36  IL-7 loses the classic lymphocytic round shape to acquire a water drop-like shape that we attribute to either mitosis or cell mobility (Figure 1E). We do not observe this change in morphology in ILC2s cultured in RPMI supplemented with FBS, IL-33, TSLP, IL-2 and IL-7 (not shown).  We then seeded different ILC2 concentrations in a 24-well plate (105 and 1.94×105 cells/mL) to assess potential disparities in cell growth, however both curves illustrated a similar growth pattern (Figure 2A). Moreover, we evaluated the viability of sub-populations derived only from the 105 cells/mL explant (P0) by sub-culturing two different cell concentrations (P1) on days 6 and 7 after the exponential phase. ILC2 sub-populations were viable and expanding slowly but they started to die five days later, proposing that both sub-cultures resembled a late stationary phase of cellular growth (Figure 2A). We confirmed ILC2 functionality by individually measuring cell IL-5 and IL-13 production (Figure 2B) based on Figure 2A and an ELISA assay (not included). In a different experiment we sub-cultured at the early exponential phase of the original explant (days 1.5 and 3), but this time assessing the same cell concentration per sub-culture (82,000 cells/mL). Figure 2C illustrates how ILC2s grow appreciably better in the earliest sub-culture (day 1.5) than in the later ones.  Finally, in a 24-well plate, we examined a lower initial cell concentration (66,000 cells/mL) to see if it has any detrimental effect on growth. At the same time, we compared two conditions: batch culture versus a fed-batch approach, were half of the conditioned media (0.5mL) was removed and replaced by fresh media every other day. Growth curves for both conditions were very similar and also reached a concentration of 106 cells/mL in five days, indicating that feeding and the initial concentration of the inoculum does not have a significant impact on expansion. This is illustrated in Figure 2D. Overall, our results imply that serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 is a better culture media than RPMI supplemented with FBS, IL-33, TSLP, IL-2 and IL-7 for mouse ILC2s, as cells clearly expand for almost a week and at the same time allowing for viable sub-cultures from the original, primary explant. 37                  Figure 1. Serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 promotes female mouse ILC2 growth and expansion better than RPMI 1640 with 10% fetal bovine serum. A) Starting at the same cell concentration, Serum-free media promotes culture expansion significantly better than RPMI in a round bottomed 96- and 24-well plate. This difference in growth is also reflected in B) in terms of glucose consumption and lactate production. Cell density is also clearly different between both media in the C) round bottomed 96- and D) 24-well plates. E) Representative image of healthy ILC2 morphology in a 24-well plate (10x magnification, serum-free media). Immediately after harvest, cells are small and round. This morphology is maintained in RPMI 1640 whereas in serum-free media they become irregular and acquire a water drop-like shape.  38                      Figure 2. Subcultures of primary ILC2s from female mice using serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7. A) The starting cell concentration of the original explant (P0) does not have a significant effect on expansion. The three time points highlighted in red represent the exponential phase of cellular growth in the linear and log version of the same growth curve. ILC2 sub-cultures (P1) at different concentrations seeded on days 6 and 7 after the exponential phase are viable, grow slowly and cells start dying after five days. B) Amount of IL-5 (left) and IL-13 (right) secreted per ILC2 individual cell based on Figure 2A. We have included two different initial cell concentrations that have been diluted differently in an ELISA assay (not included) C) Two ILC2 sub-cultures (P1) derived from the original explant (P0) seeded at the same cell concentration on different days. Although growth is represented linearly and measured in cell numbers, expansion is clearly improved when sub-culturing from the early exponential phase (day 1.5) rather than at the end (day 3) or after (days 6 or 7, Figure 2A). As stated before, cell viability drops after the 4th or 5th day of culture. D) ILC2 original explants seeded at lower concentrations experience a significant fold expansion in less than 7 days. In addition, changing culture media every other day (FED-Batch) does not have a noticeable impact in expansion compared to batch cultures, where media was never changed. 39  Discussion In this report we identify conditions to grow and expand Type 2 innate lymphoid cells in culture. This has significant impact for studying ILC2s and for facilitating further downstream applications of ILC2s. We found an improved methodology to maximize yields to gather lung ILC2s from female mice. Kadel and colleagues identified a plausible explanation for this gender bias in murine ILC2 numbers. Due to sex hormone-related factors, female mice appear to accumulate a KLRG1- ILC2 population in the lungs. A lack of this receptor renders ILC2s insensitive to a halt in cell function and therefore more prone to activation and division. Interestingly, in the presence of androgens, these KLRG1- ILC2s are reduced in males. Also, due to differential lung environments, female ILC2s are also more responsive to IL-33 and thus are better type 2 cytokine producers (Mathä et al., 2019). A second way to improve yields is to alarmin pre-treat mice before ILC2 isolation. Although the use of IL-7 and TSLP have been reported (Ricardo-González et al., 2018), IL-33 pre-treatment is a more obvious choice for isolating ILC2s, but there is no consensus on what the recommended dose is. In the simplest terms the assumption is that the higher the administered dose, the higher the yields, however. Although IL-33 activates ILC2s, enhancing effector functions and proliferation in the lungs, it also mobilizes gut resident IL-25R+ ST2- ILC2s cells (Martínez-González et al., 2017, 2018) and egresses differentiated ILC2s from the bone marrow to the lung (Stier et al., 2018). Despite the efficacy of this pre-treatment, caution is recommended however, as it may lead to the accumulation of functionally heterogeneous ILC2s that can interfere with certain studies.   Although isolation protocols for ILC2s are already well established, we recommend a series of optimizations that can accelerate the process and improve cell viability. During dissection, several protocols recommend a lung perfusion step in ice-cold PBS to clear as many erythrocytes as possible (Saranchova et al., 2018; Duerr & Fritz, 2017). Treating the lung homogenate (chopped lungs) with red blood cell lysis buffer should suffice however, therefore we recommend skipping lung perfusion for harvesting ILC2s from large numbers of mice for saving time in processing. A digestion step is 40  crucial to fully liberate the ILC2s from the lung tissue, and thus, the labware and selection of enzymes is essential to preserve ILC2 viability and expression of surface markers (Moro et al., 2015). We employ a mixture of DNase, collagenase IV and hyaluronidase while shaking the lung homogenate at 200rpm at 37°C. Interestingly, Duerr & Fritz noticed that using different enzymes like liberase, dispase II or a lung disassociation kit may yield more cells and preserve surface markers.  Whether researchers use large groups of mice to pool ILC2s or smaller groups with IL-33 treated lungs, the final homogenate will probably take several hours of fluorescent activated cell sorting that can affect recovery and viability (Fitzgerald, 2001). To save time we chose a mouse ILC2 Enrichment Kit, which treats the lung homogenate with a lineage cocktail of ten antibodies that will bind to magnetic streptavidin beads in a 5mL polystyrene tube. Once the tube is inserted in a magnet, magnetic separation of other leukocytes will greatly purify the ILC2 population prior to staining and sorting, and therefore accelerating the procedure (Martínez-González et al., no date). For a list of the flow cytometry antibodies see methods section.  Following our protocol and after treating mice with IL-33, we were able to harvest between 100K and 200K ILC2s from groups of five C57Bl/6 female mice per experiment. In Figures 1A & 1B we demonstrate that serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 supports ILC2 growth better than RPMI 1640 containing 10% FBS supplemented with IL-33, TSLP, IL-2 and IL-7. Camelo and colleagues have detected very low concentrations of IL-1RA, LIF, HGF, NGF, VEGF-A and VEGF-D in serum-free medium (StemSpan™) in an attempt to find IL-1β or IL-12 in their ILC2s, which expressed intracellular granzyme B. The low concentration of IL-1RA, LIF, HGF, NGF, VEGF-A and VEGF-D does not explain the large difference between media in growth. Interestingly, our group also reports granzyme B+ ILC2s in serum-free medium (StemSpan™), and since such media does not contain the pro-inflammatory cytokines responsible for previously reported induction of plasticity towards an NK phenotype (Lim et al.; Bal et al.; Ohne et al., 2016), we attribute this identity shift to other biological factors.  41  The cell densities displayed in Figures 1C & 1D were immediately achieved as soon as we implemented serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 into our protocols. Lymphocytes are non-adherent cells and can be generally grown in flat or round bottomed wells, however with ILC2s, as the yields are generally low, these tend to be cultured in U-bottom 96-well plates to increase cell proximity (Saranchova et al., 2018; Halim & Takei, 2014; Camelo et al., 2017). There is no published evidence regarding ILC2 culture mechanics and the advantages of keeping cells in close contact, but our group has explored the effects of clump pipette trituration (and the absence of it) in culture expansion. We have not observed any significant differences nor trends in population dynamics when we consistently apply trituration to keep ILC2s apart (data not included). Although we have not achieved high culture confluency with RPMI containing 10% FBS supplemented with IL-33, TSLP, IL-2 and IL-7, other studies demonstrate with that by supplementing with IL-2 and different cytokine combinations and an automated tissue dissociation kit higher densities may be achievable (Moro, no date). We have also calculated the average ILC2 growth rate from three different cultures in serum-free media in a 24-well plate. We understand the growth constant k as the number of cell generations per unit time, and in mouse ILC2s it turns out to be 0.025 generations per hour, or 0.025h-1. This value is of importance to calculate the doubling time for ILC2s, which take exactly 27.72 hours to double the size of their population, a little bit more than a day. In RPMI, however, ILC2s have a doubling time of almost 70 hours. For reference, laboratory grown Escherichia coli have a doubling time of 19.5 minutes (ranges between 15 and 20 minutes), for which they need a significantly higher growth constant of 2.13h-1 (Marr, 1991). Interestingly, the murine tumour cell line B16 divide at a rate of 24.9h per cell division, slightly faster than ILC2s, but IL-2 – sensitive and co-stimulated CD8+ T-cells have an impressive doubling time of 5.3h (Hwang et al., 2006), essential to mount effective cellular responses against tumours or pathogens. As type 2 cytokine secretors and recently discovered 42  CD8+ T-cell activators (Saranchova et al., 2018), like dendritic cells, ILC2s do not have the urgency to replicate at such speeds.  On an individual level, ILC2 cell morphology post-sorting is small and circular, which is also the case for recently purified T-lymphocytes. After 2-3 days in serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 but not RPMI supplemented with FBS and IL-33, TSLP, IL-2 and IL-7, ILC2s acquire an irregular shape (Figure 1E) that we have also observed for CD8+ T-lymphocytes that have interacted with antigen-bearing dendritic cells. We hypothesize that acquisition of irregularity in ILC2s is a sign of growth, differentiation and activation and this change in shape may result from a rapid increase in sensitivity to Ca2+ signaling upon alarmin engagement. Although calcium channel dynamics and signaling are still yet to be explored in ILC2s, it is known that intracellular Ca2+ concentrations dramatically change in T-lymphocytes upon antigen presentation, leading to differential morphologies and changes in the direction of crawling (Negulescu et al., 1996). Given the similarities observed in ILC2s, acquisition of irregularity may be related to an increased cell motility instead of processes associated with cell division. This may have interesting implications for ILC2s as migratory cells as opposed to the current notions that understand them as purely tissue-resident cells (Walker et al., 2013).  We demonstrate that it is possible to obtain viable ILC2 sub-cultures by re-seeding on days 6 and 7, where the cell concentration increases steadily (see log curve) but quickly deteriorates at the same time as the primary explant (Figure 2A). Cell quality is reflected in Figure 2B, where we can observe that the amounts of IL-5 and IL-13 secreted per cell increments with time, as the cell concentration in culture increases. Furthermore, the IL-13 secreted per ILC2 cell belonging to an initial concentration of 194K cells/mL in a higher dilution (x1120) reflect the initial contraction phase in culture, probably due to the transfer from an in vivo system to an in vitro one. ILC2 concentration increases more significantly if sub-culturing is carried out early in the exponential phase, on day 1.5, as opposed to days 3, 6 or 7 (Figure 2C). Yet, growth in Figure 2C is represented by a linear y-axis 43  and only reflects the early stages of an exponential phase of cellular growth. Our data, however, is still aligned with other non-adherent cell culture protocols that recommend sub-culturing during the exponential phase to maximize growth and to prevent genomic instability that can affect the quality of experiments (Oh et al., 2012; Sigma-Aldrich, 2020). We also observe that starting with a high cell concentration (194K cells/mL) and a lower one (100K cells/mL) yields a similar culture growth in serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 (Figure 2A), meaning that the initial concentration of the inoculum does not have a significant impact on expansion. If the initial cell concentration is lower, such as 66K cells/mL (Figure 2D), we also observe an exponential phase of growth (Figure 2A). Thus, we find that culturing with low initial ILC2 concentrations is attainable in serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7, however the lower limit for a viable expansion remains unknown.  Finally, we compared the effect of two different culture strategies on ILC2s for the course of a week: batch vs fed batch. The former involves a lack of media changes, meaning that the cells will eventually meet a limited supply of nutrients that will cause a decline in viability and cell numbers. The latter is commonly used in large scale bioreactors and involves a continuous supply of one or more nutrients. Technically, our approach is not purely fed-batch because we changed the media every other day instead of continuously, but ILC2s still have constant access to fresh nutrients. As illustrated in Figure 2D, in the span of seven days there is no visible difference in culture expansion when comparing these techniques. Our group has used this approach to batch-culture ILC2s for the experiments and have obtained consistent results. This reduces the cost of media and therefore contributed to economize experiments with ILC2s.   44  Conclusion In this study we present an alternative serum-free media to RPMI 1640 to culture murine type 2 innate lymphoid cells. Although serum-free medium was initially recommended for human hematopoietic cells, it has only been tested in one study using human ILC2s (Camelo et al., 2017), and never with murine ILC2s. Our results demonstrate that culturing mouse, lung ILC2s in serum-free medium supplemented with IL-33, TSLP, IL-2 and IL-7 yields highly viable cells but also a significantly larger expansion than RPMI supplemented with FBS and IL-33, TSLP, IL-2 and IL-7 with the added benefit that there is no need for media changes over the course of a week. The observed increase in cell concentration was enough to culture these ILC2 in wells of larger formats than the 96-well plate, and to sub-culture viable cells from original explants. To fully optimize culture, a cytokine study will be necessary to identify the best combination for ILC2 growth. In addition, we recommend improving freeze-thaw protocols to minimize the number of ILC2 isolations from mice. Given the low numbers of ILC2s yielded by existing methods our studies can improve ILC2 culture and further downstream applications for ILC2s.         45  Chapter 2 Type 2 Innate Lymphoid Cells Cross-Prime CD8+ T-Lymphocytes Like Dendritic Cells. Introduction Adaptive immune responses heavily rely on the interaction between highly specific, genetically rearranged antigen receptors with major histocompatibility complex (MHC) molecules that have a peptide antigen bound to them by necessity (Janeway et al., 2001). MHC molecules can be type I and found in all nucleated cells. In general, only peptides derived from endogenous proteins are presented in the MHC-I and it exclusively interacts with the T-cell receptor (TCR) of a CD8+ cytotoxic lymphocyte (CTL). This is useful to detect and eliminate modified or infected cells (Joffre, 2012). On the other hand, just a limited number of cell types constitutively express MHC-II molecules on their surfaces. Hitherto the most import of these cell types that act as gate-keepers in initiating primary immune response are dendritic cells. These professional antigen-presenting cells (APC) such as dendritic cells process and present peptides derived from proteins outside the cell in a process termed exogenous antigen processing or cross-presentation of antigens in the context of MHC-I or MHC-II to prime naïve CD8+ or CD4+ T-lymphocytes (Lizée et al., 2003; Basha et al., 2008; Reinicke et al., 2009; Basha et al., 2012).  First, naïve CD8+ T-cells mostly recirculate the secondary lymphoid compartment. This maximizes their presence in the blood, spleen and lymph nodes; meaning that other tissues that have encountered a pathogen might have minimal interaction with them (Heath & Carbone, 2001). This limitation is solved by cross-presentation, an elegant process where an APC internalizes a peptide, processes and presents it in the MHC-I (despite being exogenous) to the CD8+ T-cell (Embgenbroich & Burgdorf, 2018). Professional APCs such as dendritic cells (DC) will acquire the antigen at the site of infection and migrate to the draining lymph nodes, easing antigen accessibility to CD8+ T-cells. Cross-46  presentation is also the solution for the second scenario, where an MHC-II – restricted response is directed at exogenous antigens by presenting them to CD4+ T-cells. APCs can capture pathogen products and initialize a T-helper response, or can simultaneously cross-present said products in the context of MHC-I to CD8+ T-cells (Lizée et al., 2003; Basha et al., 2008; Reinicke et al., 2009; Basha et al., 2012).  Historically, cross-presentation has normally been associated to class I – restricted CTL priming given it was the focus of the first experiments (Bevan, 1976; Yewdell, 1999). The molecular pathways involved in this process are in constant evolution, but two well established routes are known: the cytosolic and the vacuolar. The former involves the access of internalized proteins to the cytosol, where they then follow the classical class I pathway for peptide loading (Kovacsoviks-Bankowski & Rock, 1995; Ackerman et al., 2003; Guermonprez et al., 2003). The vacuolar pathway is sensitive to lysosomal proteolysis and comprises loading of peptide fragments onto MHC-I molecules already residing in the endocytic loading compartment first identified by Lizée and colleagues in 2003. MHC-I molecules can access this peptide loading compartment by recycling from the cell surface (Lizée et al., 2003; Basha et al., 2008; Reinicke et al., 2009) or directly interacting with the invariant chain (CD74) that associate with MHC-I molecules and aid in their trafficking towards the vacuolar-endocytic compartment (Basha et al., 2012). CTL cross-priming is an essential step in the process of immunosurveillance via neoantigen cross-presentation (Hiraoka et al., 2006). However, although CTL responses and competent tumour endogenous antigen processing are both required for the defense against tumours (Alimonti et al., 2000), weak and inefficient responses are not rare, rendering CTL cross-priming by DCs insufficient (Stumbles et al., 2004). This could be due to the lack of inflammation and/or danger signals in the tumour tissue (Lyman et al., 2004) combined with the mechanisms of immune evasion that tumours employ (Alimonti et al., 2000). In addition, there is a dependency on high neoantigenic loads to reach more naïve CD8+ T-cells through APCs, and by the time they do, it could be late, clinically speaking 47  (Ochsenbein, 2005). These issues highlight the importance of early, innate responses and complementary ways to activate CD8+ T-cells against tumours. We previously demonstrated that an unconventional cell from the innate branch of the immune system was not only participating in the anti-tumour response, but also enhancing CTL – mediated killing of murine, tumour cell lines (Saranchova et al., 2018). Type 2 innate lymphoid cells (ILC2) are well known as type 2 cytokine secretors in response to helminthic infections, but also major drivers of allergen sensitization and allergic, lung inflammation (Bernink et al., 2014). Any functions they exert to mediate adaptive immunity in the context of cancer had never been described before. The frequency of ILC2s appears elevated in tumours expressing interleukin 33 (IL-33), a powerful ILC2 activator (Martínez-González et al., 2015). Interestingly, ILC2 knockout mice (RORα-/- mice) display higher tumour volumes, more circulating tumour cells (indicator of metastasis) and less granzyme B production by CD8+ T-cells (Saranchova et al., 2018). The process of CTL activation is energetically and immunologically taxing and must involve cross-presentation of antigens by professional APCs. Provided the importance of this process, in this study we explore whether ILC2s have cross-presenting capabilities and if so, if they have any potential to cross-prime CD8+ T-cells. We use ILC2s from three different sources: naïve ILC2s from wild type mice (nILC2), ILC2s from mice with IL-33+ primary tumours (pILC2) using a murine IL-33+ lung tumour epithelial cell line and ILC2s from mice inoculated with its metastatic IL-33- derivative (mILC2). Despite the unorthodox nature of this idea, these data provide an elegant explanation to the observations that ILC2s participate in anti-tumour TH1 responses.    48  Results Type 2 innate lymphoid cells are of lymphoid origin and functional  Prior to sorting, ILC2s were stained for four surface markers that they are known to express (Saranchova et al., 2018): the IL-33 receptor (ST2), the leukocyte tyrosine phosphatase CD45, the pan-thymocyte marker Thy1.2 and the IL-7 receptor (CD127). Cells were also stained for 10 non-ILC2 lineage markers in the same colour (FITC). The gating strategy is shown in Figure 3A. Once sorted, we tested the RNA expression levels of the transcription factor GATA-3, essential for ILC2 development. GATA-3 is expressed early in the ILC2 lineage, specifically at the early innate lymphoid progenitor stage, which is immediately preceded by the common lymphoid progenitor (Ebihara & Taniuchi, 2019). This makes GATA-3 a good identity marker. In addition, and as a negative control, we also measured the RNA levels of the spi1 gene (Figure 1B, left), which encodes the master regulator PU.1, essential for myeloid fate and B-cell development (Kueh et al., 2013) and absent in our sorted sample.  To assess ILC2 functionality we then analyzed the RNA expression levels of IL-5, IL-13 and IFNγ. When activated by alarmins, ILC2s produce type 2 cytokines, especially interleukins 5 and 13. Interferon γ (IFNγ), on the other hand, is a pro-inflammatory cytokine produced in type 1 immune responses and not secreted by ILC2s (Artis & Spits, 2015). Our results in Figure 3B (right) show that the il5 and il13 genes are expressed by the sorted sample, whereas the ifng gene that encodes IFNγ is absent, indicating that the sample functionality matches that of conventional ILC2s. We then performed an ELISA assay to test for the protein levels of IL-5 and IL-13 in ILC2 conditioned media at different time points during culture (Figure 3C). Both cytokines were detected and as expected, their concentration increased with time. This was also reflected at an individual level, where cytokine secretion per cell increased with time (not included). Note that low levels of both cytokines were detected in a control sample of fresh RPMI 1640 with serum.  49  So far, our results demonstrate that our sorted sample retains lymphoid identity and express type 2 cytokines when seeded in media supplemented with the alarmin IL-33, which were then detected at a protein level, implying that the sample is likely a purified population of ILC2s.                Figure 3. Sorted ILC2s from female C57Bl/6J mice are a pure and functional population of lymphoid origin. A) Representative gating strategy for Lin- CD45+ ST2+ Thy1.2+ CD127+ cells. B) At an RNA level, purified ILC2s express GATA-3 (left), which is upregulated early in the lineage and retained until terminal differentiation. The myeloid factor PU.1 is not expressed. ILC2s also express the genes for their signature type 2 cytokines IL-5 and IL-13 (right), but not the type 1 cytokine IFNγ, as expected. FPKM = fragments per kilobase million. Error bars represent standard error of the mean and n=3 for both experiments (unpaired t-test). C) ELISA assay (left) illustrating the concentration of the cytokines IL-5 and IL-13 at the protein level. Interleukin production increases with time in culture. Culture media SFEM II and RPMI 1640 (10% FBS) were also tested and RPMI has basal level of both cytokines (n=1).   50  pILC2s from mice with IL-33+ primary tumours successfully process exogenous, whole OVA protein and cross-present the peptide SIINFEKL on their MHC-I During the process of immune editing or at the core of an anti-cancer response, the cells in charge of tumour cell elimination are mostly CD8+ T-cells and NK cells (Mittal et al., 2014). As mentioned before, the former is succesfully activated after DCs cross-present neoantigen to the TCR/CD8 complex along with CD4+ T-cell help (Basha et al., 2012). Our hypothesis that ILC2s could also act as APCs and cross-prime antigen was reinforced by a study carried out by Oliphant and colleagues in 2014 where, for the first time, MHC-II+ ILC2s were shown to establish a dialogue with antigen-specific T-cells, promoting a TH2 response. For these reasons, we tested whether ILC2s can cross-present the OVA peptide (257-264) SIINFEKL. In this experiment we employed nILC2s, pILC2s and mILC2s. Each culture was treated with whole, exogenous OVA protein and 24h later stained with an antibody that detects SIINFEKL only in the context of MHC-I. The positive control consisted of pumping purified SIINFEKL into the culture supernatant. The negative control involved staining with the antibody but without any OVA/SIINFEKL treatment. The results were visualized using flow cytometry.  We first analyzed nILC2s both from IL-33 pre-treated and non-pre-treated C57Bl/6 mice. Figure 4C shows the differences in yield between pre-treatment and no condition in mice. The mean fluorescence of the antibody detected in Figure 4A illustrates how those nILC2s that were derived from non-pre-treated mice were more effective at cross-priming OVA peptide compared to those nILC2s derived from pre-treated animals (2B). The “shoulder” in the ILC2 + OVA sample (4A) depicts a population that does not cross-present versus one that does. From here on, we only isolated ILC2s from animals not pre-treated with IL-33 as we expected newly generated ILC2s not to interact with the tumour microenvironment as much, and therefore possibly display a different phenotype.   51  Interestingly, the ILC2 population that displays the most noticeable cross-presentation capabilities are pILC2s, in other words, when they derive from the lungs of animals with primary tumours that express IL-33 and express MHC-I. Figure 4D shows a clear population of pILC2s treated with OVA. The shift in brightness that we detected in the ILC2 + OVA sample was very noticeable when compared to our positive control (ILC2 + SIINFEKL). We account these successful results to the use of serum-free media for ILC2 culture for significant expansion and to the lack of IL-33 pre-treatment in tumour-bearing mice.  Taken together, these results suggest two important points. First, IL-33 pre-treated mice yield ILC2s that are incapable of antigen cross-presentation. Second, those pILC2s that are derived from mice with primary tumours expressing IL-33 and MHC-I successfully cross-present SIINFEKL from exogenous OVA compared to nILC2s. In the following experiments, pILC2s from non-pre-treated mice were utilized as they are the most efficient subset, under the conditions we characterised for cross-presentation assays. The findings are summarized in Figure 4E.      52                 Figure 4. Non-IL-33 pre-treated pILC2s from female mice successfully process OVA protein and cross-present the peptide SIINFEKL. A) ILC2s from WT mice (n=10) that have not been pre-treated with mouse IL-33 can cross-present OVA to an extent. The orange curve shows a “shoulder” that represents the cross-presenting population. The blue curve is a positive control where ILC2s directly bind exogenous SIINFEKL onto their MHC-I. B) ILC2s from WT mice (n=4) that have been pre-treated with IL-33 to obtain higher yields. The OVA-treated sample (orange) does not cross-present as evidently as the non-pre-treated population. C) Injecting mice with murine IL-33 three times a week every other day (600ng total) leads to higher yields from the lungs. Given the scarcity of these cells, it is a commonly employed technique in ILC2 research (n=4 for mice and results analyzed using an unpaired t-test). D) ILC2s from mice (n=20) with primary IL-33, MHC-I expressing tumours (TC1 cell line) contain a cross-presenting population when incubated with OVA. There is a clear shift in the OVA-treated population of pILC2s (orange) where cross-presentation is evident and similar to the positive control (blue). E) Summarize of the findings of pILC2 cross-presentation illustrating average mean fluorescence intensity (MFI) where experiment n=3 (unpaired t-tests).   SIINFEKL bound to MHC-I 53  ILC2s derived from tumour – bearing mice cross-prime population of OT-I cells by presenting SIINFEKL peptide from exogenous OVA After studying cross-presentation in ILC2s, the next step was to evaluate whether they were able to cross-prime a population of CD8+ OT-I lymphocytes that have a transgenic TCR restricted to the OVA peptide (257-264) SIINFEKL (Clarke et al., 2000). To maintain the dynamics of previous experiments (Figure 2) and 24h prior to co-culture with OT-Is, we kept pumping OVA and purified SIINFEKL to ILC2 culture media as a positive control. As soon as the OT-I cells were isolated from spleen they were stained with CFSE to trace their proliferation rounds. Co-cultures were kept for a maximum of three days to prevent ILC2 culture deterioration and OT-I proliferation was assessed using flow cytometry on days 2 and 3. For our first experiment, we separately co-cultured OT-Is with ILC2s from mice with primary and metastatic tumours (pILC2s and mILC2s). As illustrated in Figure 5, these were either treated with SIINFEKL (middle column), OVA (right column) or untreated (left column). In every case, cell division was only observed mostly on day 2 but also on day 3. In regard to untreated ILC2s, despite detecting clear proliferation rounds, OT-I numbers do not increase for both types of ILCs. SIINFEKL – treated ILC2s on the other hand provoked a marked T-cell expansion on both days as expected. The most notable aspect of OT-I divisions in the presence of OVA – treated ILC2s is that proliferation rounds do not show cell numbers plummeting as clearly as with untreated ILC2s, although these do not increase as visibly as with SIINFEKL. So far, pILC2s and mILC2s have not displayed notable differences in cross-priming, however, in the OVA – treated sample and against our expectations, OT-I CD8+ T-lymphocytes co-cultured with mILC2s or pILC2s (day 2) are approximately same at triggering OT-I CD8+ T-lymphocytes cell division and therefore Cross-priming.  To have a better frame of reference for CTL division, we repeated the cross-priming assay using OVA – primed DCs co-cultured with OT-Is as a positive control instead of pumping purified SIINFEKL into the media. The first assay is shown in Figure 6A where we observe effective OT-I CD8+ T-54  lymphocytes cross-priming by DCs, where the CFSE dilution in the OT-I CD8+ T-lymphocytes population markedly increases with every new round of proliferation as opposed to the negative control, which was a well with OT-I CD8+ T-lymphocytes only (red histogram) where cell numbers plummeted after each division. In accordance with our previous experiment (Figure 5), we observed that in the presence of pILC2s there is a clear OT-I CD8+ T-lymphocytes expansion (orange histogram). We observe the similar proliferation of OT-I CD8+ T-lymphocytes and mILC2 co-culture (black histogram) and also in agreement with our previous results. Finally, figures 6B and 6C illustrate the purified OT-I cells and bone marrow-derived DCs respectively that we purified prior to our cross-priming assays with ILC2s. Overall, these results show that ILC2s from the lungs of tumour-bearing mice have significant cross-priming capabilities, however the differences between primary and metastatic ILC2s in this aspect are yet to be determined, since they cross-prime a population of CD8+ cells with the same effectiveness. From here we conclude that although cross-presentation is necessary for cross-priming, what happens at one level does not have to necessarily reflect at the other one given the many more factors involved in antigen processing or CTL activation.       55                 Figure 5. OT-I proliferation in co-culture with different types of ILC2s. Top charts display OT-I cell proliferation in co-culture with mILC2s and bottom charts co-cultures of OT-I cells with pILC2s. The left column shows proliferation of OT-Is that have been co-cultured with untreated ILC2s (no OVA/SIINFEKL) as a negative control. Proliferation rounds were detected on days 2 and 3, but the cell number did not increase. Central column focuses on OT-Is co-cultured with SIINFEKL-treated ILC2s as a positive control. Proliferation was also detected from day 2 and OT-I numbers increased, especially in the presence of pILC2s. The right column represents the co-culture with OVA-treated ILC2s to test for cross-presentation. Proliferation rounds detected on day 2 were the most prominent. OT-I cell numbers did not decrease as markedly as with untreated ILC2s. For each histogram, n=20 for mice.           Figure 6. pILC2s from female mice with IL-33+ primary tumours treated with OVA cross-prime OT-I cells on day 3. A) CD8+ cells in monoculture (red) proliferated but numbers did not increase. When co-cultured with OVA-treated dendritic cells (DC), OT-I cells were activated and displayed notable proliferation rounds (blue). OT-I cells displayed a marked proliferation in the presence of pILC2s (orange). Interestingly, mILC2s also initiated an equivalent proliferative response (black). B) Isolated CD8+ T-cells and C) differentiated bone marrow – derived DCs. Mouse n=50 for ILC2 isolation (25 for TC1 mice + 25 for A9 mice).  56  Discussion This line of study began when we demonstrated that MHC-Ihi primary tumours (TC1 cells) were also IL-33+, and that MHC-Ilo metastatic tumours (A9 cells) were IL-33- (Saranchova et al., 2016). We subsequently demonstratred that IL-33 expression in patient kidney and prostate cancers is correlated to patient outcomes, where the loss of IL-33 expression leads to more rapid death (Saranchova et al., 2016). The idea that these genes were transcriptionally linked was demonstrated when A9 cells were induced to express IL-33, resulting in a rescue of MHC-I levels (Saranchova et al., 2018). This novel mechanism of metastatic immune escape by downregulating IL-33 led us to examine the role of ILC2s in cancer. We then demonstrated that ILC2s are directly involved in tumour immunosurveillance and elimination in mice (Saranchova et al., 2018). These data revised the dogma of ILC2s are only innate mediators of type 2 immunity and established their role in TH1 CTL responses and cancer immunosurveillance. Furthermore, ILC2s appear to have an unexplained role in dramatically reducing the metastasis of tumour by reducing the number of circulating tumour cells (Saranchova et al., 2018).  These studies are the first to propose a mechanism of ILC2-mediated tumour elimination and also the first to explore cross-priming of CD8+ T-cells by ILC2s. Within the intricacies of the tumour microenvironment it is difficult to spot a pathway were effector ILC2s initiate an anti-tumour response, but it was relatively easy to predict the outcome: CTL activation. For these reasons, the goal of this study became the identification of an ILC2-mediated mechanism of CTL activation.  Initially, the most immediate interaction between ILC2s and T-cells appears to be indirect. ILC2s are avid IL-13 producers, and although TH2 lymphocytes do not express IL-13R, DCs do. The interaction with IL-13 licenses DCs to migrate to the lymph nodes where they can prime and differentiate CD4+ T-cells into a TH2 population (Martínez-González et al., 2015). Perhaps an unidentified mechanism of TH1 priming by ILC2s could lead to subsequent CTL expansion, however there is no current evidence of such interaction. In fact, ILC2 responses will disbalance the TH1/ TH2 ratio in favour of 57  severe allergic asthma or lung inflammation (Lu et al., 2018). Remarkably, ILC2s can express class II MHC molecules that allow for direct presentation of exogenous antigens to a TCR. This leads to IL-2 production by the T-cell and further ILC2 expansion (Oliphant et al., 2014). We hypothesized that if ILC2s can process and present exogenous antigen for exogenous MHC-I antigens this would be a direct mechanism underlying ILC2-driven tumour elimination. The whole ILC compartment comprises less than 1% of lung or mucosal tissue, which guarantees low yields. In addition, ILC2s do not posess unique surface markers, which is why a combination of them is necessary (Halim & Takei, 2014). We stain lung homogenates for ST2 (IL-33R), CD45, Thy-1.2 and IL-7R, as well as a combination of lineage markers to isolate ILC2s from the homogenate (Figure 3A). Other groups have tested for different combinations with other positive markers such as ICOS, c-Kit or CD25. This can depend on the tissue researchers isolate from (Moro et al., 2015). To further cell characterization, assessing the expression of the transcription factor GATA-3 is key. This can be done via flow cytometry or RNA sequencing. As a control, we assessed the presence of the gene for the master regulator PU.1 to discard any myeloid cell interference in downstream cross-presentation experiments (Figure 3B, left). Since ILC2s also depend on GATA-3 functionality, testing for their signature cytokines IL-5 and IL-13 was our next step (de Grove et al., 2016). The genes for both cytokines were present in our sample, but not the gene for IFNγ (Figure 1B, right), which is largely produced by ILC1s (Bernink et al., 2017). We also performed an ELISA assay to ensure that IL-5 and IL-13 were also present at the protein level, since we did not stain for intranuclear GATA-3 (Figure 3C). During our experiments we transitioned from RPMI 1640 (10% FBS) to serum-free media for ILC2 culture. We do not recommend the use of serum as it has some basal levels of IL-5 and -13 in our analysis, rendering it inappropriate for these kind of assays.  In our first cross-presentation assays we examined nILC2s from the lungs of WT mice (C57Bl/6 background). Given the low number of cells we obtain per mouse, we decided to treat mice with IL-58  33 a week prior to isolation. This technique enhances in vivo pulmonar ILC2 responses in mice and entail a rapid cell expansion (Barlow et al., 2013) allowing resarchers to work with manageable yields (Figure 4C). The control in these experiments is a negative control where ILC2s remain untreated in culture but are still stained with the antibody that detects the MHC-I – SIINFEKL complex (red histograms) and a positive control where we pumped purified SIINFEKL into ILC2 cultures. The strong shifts we observe in Figures 4A & 4B (blue histograms) is due to the high specificity of the peptide for MHC-I molecules, which will ensure the direct binding with no need for previous internalization or processing, even with potential to displace other already bound peptides (Hearn et al., 2009). The weak shift in the ILC2 + OVA sample (orange histogram) in Figure 4B shows a population of nILC2s that has not successfully cross-presented OVA. We changed the experiment parameters and instead, we started isolating nILC2s from mice that had not received an IL-33 pre-treatment. Although this yielded low numbers of nILC2s post-sorting, we found that a portion of these nILC2s were able to cross-present SIINFEKL from OVA (Figure 2A). We believe that there could be three non-mutually exclusive explanations for this difference between ILC2s from pre-treated and untreated mice.  Firstly, Martínez-González et al. (2016) identified a population of ILC2s that expanded upon allergen encounter, but that also went through a contraction phase where some cells persisted for a long time. These “allergen-experienced” ILC2s responded more vigorously to subsequent challenges and were nicknamed memory ILC2s. Pre-treating mice with IL-33 will prime lung resident ILC2s to expand and give birth to new generations of “inexperienced” ILC2s that will eclipse the initial memory population in size, and perhaps also hide any potential phenotypic traits related to cross-presentation and antigen processing. This might explain the “shoulder” in Figure 2A, where two populations were clearly detected, and mice were not pre-treated. Second, memory ILC2s are IL-25Rhi, which is also a defining marker for gut ILC2s (Artis & Spits, 2015; Fang & Zhu, 2017; Huang et al., 2018) and 59  memory ILC2s identified in mesenteric and mediastinal lymph nodes (LN). These would be in closer contact with CD8+ T-cells amongst other cells and are also hypothesized to migrate to the lungs upon alarmin release (Martínez-González et al., 2017, 2018). Thirdly, Huang and colleagues (2015) characterized a population of ILC2s that mostly responded to IL-25 termed inflammatory ILC2s. These appeared to behave as transient progenitors giving rise to “normal”, IL-33 – responding ILC2s but also displayed plasticity towards the ILC3 phenotype. These cells might also be the previously described migratory, IL-25Rhi memory phenotype that resides in LNs that we miss when we pre-treat with IL-33. Following cross-presentation assays with nILC2s we focused on examining ILC2s from the lungs of primary (pILC2) tumour-bearing mice. Figure 4D illustrates how pILC2s displayed striking capabilities at cross-presentation, even elevated compared to nILC2s (Figures 4D & 4E). These results were reflected in Figure 5, where OT-I cells were successfully and equally primed by OVA-treated pILC2s. We also introduced ILC2s from the lungs of metastatic tumour-bearing mice (mILC2s), which were also able to cross-prime upon OVA-treatment. As expected, OT-I numbers experienced a larger expansion in our positive control, where SIINFEKL-treated pILC2s and mILC2s presented unprocessed, purified OVA257-264 peptide to the highly specific OT-I TCR. The negative control, which consisted of a co-culture with untreated pILC2s and mILC2s, reflected poor but existing OT-I proliferation.  We observed a similar result in another negative control from Figure 6A where OT-I numbers plummeted in (red histogram), however in this case, the control lacked ILC2s and was just an OT-I monoculture. We attribute the proliferation in our negative controls to the presence of IL-2 in the media, especially when working with CD8+ T-cells that produce less autocrine IL-2 than CD4+ T-cells, which is why they need to be exquisitely sensitive to the cytokine to proliferate (Au-Yeung et al., 2019), even in the absence of co-stimulation or antigen presentation. When mouse OT-I cells were in co-culture with OVA-treated pILC2s and mILC2s, OT-Is proliferated in both cases (Figure 6A), 60  showing a striking increase in cell numbers that competes with the DC positive control, but with less defined peaks or new generations. These data demonstrate for the first time that ILC2s are capable of MHC-I cross-priming. This became even clearer when we compared ILC2s with DCs as control instead of a peptide control, like in Figure 5, suggesting ILC2s have similar capabilities to DCs in cross-presentation and cross-priming. One of the main limitations to these studies are the poor ILC2 yields we obtain from tumour-bearing mice compared to wild types. We have shown that in the presence of a tumour, ILC2 migratory patterns are altered (Saranchova et al., 2018). The current understanding is that pulmonary ILC2s are tissue-resident cells (Mindt et al., 2018) but depend on IL-33 to egress from the bone marrow to the lungs (Stier et al., 2018), bringing newly generated cells that overlap with the population of interest, hence the halt to mouse IL-33 pre-conditioning in our studies. This low amount of lung ILC2s in IL-33 untreated, tumour-bearing animals might be explained by two mechanisms: as previously mentioned, gut and LN ILC2s are IL-25Rhi and migrate to the lungs upon alarmin responses, but they are also ST2- (Huang et al., 2015; Martínez-González et al., 2017, 2018). If the systemic response against a tumour and subsequent circulation of alarmins can attract this population to lung tissue it will mask our targets, as we use ST2 but not IL-25R as a positive isolation marker. Another reason could be the egress of ILC2s from the lungs as shown by Saranchova et al. (2018). ILC migration between distinct peripheral tissues in the context of cancer is a largely unexplored topic, however in homeostatic conditions gut ILC2s are known to travel to and from mesenteric LNs (Huang et al., 2015). More profound ILC2 tracking is key to determine whether ILC2s exit the lung to access other LNs or tumour sites. We suspect that mILC2s but especially pILC2s undergo differential transcriptomic pathways compared to conventional ILC2s that may enhance their cross-presenting and cross-priming potential. We can assume that a late metastatic environment resembles homeostatic, non-inflammatory conditions, since the original tumour has evaded immunity and that there is no longer a sustained and 61  systemic immune response (Hanahan & Weinberg, 2011). If there are any appreciable shifts in the dynamics of ILC2-driven anti-tumour responses it may be at the initial tumour stage or early metastatic stages, which explains the cross-priming capabilities observed in figures 5 and 6A. In our primary tumour model, for example, the lung epithelial cell line TC1 is IL-33+. On the other hand, the metastatic derivate A9 line is IL-33-. This explains a role for ILC2s at the primary stage (Saranchova et al., 2016), but not during metastasis. We suspect that A9 cells may recover some levels of IL-33 expression upon inoculation in mice, but not in vitro. Aligned with our hypotheses, other studies have demonstrated that tumours expressing IL-33 but not their stroma promote type 1 immunity mediated in principle by NK cells and CD8+ T-cells (Lu et al., 2016). Others propose novel roles for IL-33 where it synergizes with the TCR and IL-12 to mount potent CTL anti-tumour responses (Yang et al., 2011). That, along with the previously established downregulation of MHC-I linked to IL-33 (Saranchova et al., 2016), would provide a complementary explanation as of why IL-33- tumours become insensitive to TH1 responses and are allowed to metastasize due to a lack of anti-tumour immunity.   Conclusion Currently, most ILC2 studies that departed from the existing dogma on the role of ILC2s in type 2 responses points to ILC2s as immunosuppressive and pro-tumorigenic (Bie et al., 2014; Salimi et al., 2018; Trabanelli et al., 2017; Chevalier et al., 2017; Jovanovic et al., 2014). These groups used naïve ILC2s from WT mice that were in homeostatic conditions prior to harvest and are not equivalent to our studies. We, on the other hand, demonstrate that in the context of cancer, ILC2s have potential to be immunoenhancing if they are derived from IL-33+ tumour-bearing animals. We propose a mechanism where ILC2s from these animals acquire exogenous tumour antigen and cross-present it to naïve CD8+ lymphocytes, activating a tumour-specific CTL response and supporting DC cross-62  priming. Further research will be necessary to elucidate whether these ILC2s egress the lung and access the LNs or tumour tissue and if they are able to cross-present other antigens besides OVA257-264. We hope that this establishes a new dichotomy in the way we understand ILC2 biology in cancer.     63  Chapter 3 Single-Cell Analysis Reveals a Heterogeneous Population of Type 2 Innate Lymphoid Cells from Tumour-Bearing Mice that Lost Type 2 Identity in Favour of Inflammatory markers Introduction Innate lymphoid cells (ILCs) integrate innate and adaptive immune responses and control numerous physiological processes (Artis & Spits, 2015) by producing cytokines in response to danger signals. However, they do not possess a T-cell receptor that allows for an antigen-specific response (Walker et al., 2013). Generally, ILCs can be split into three different groups depending on their function: ILC1s are pro-inflammatory cells and can be further categorized by the presence or absence of cytotoxic granules. ILC2s secrete TH2-type cytokines and amplify type 2 responses and ILC3s have been associated with lymphoid tissue induction and release of interleukin 17 (IL-17) and IL-22 (Simoni et al., 2017).  Cytotoxic type 1 ILCs, also known as natural killer (NK) cells, are key agents in the anti-tumour response. So much, that they were actually named on the basis of their capacity to kill tumour cells in vitro (Trinchieri, 1989). NKs kill via perforin/granzyme-dependent cell death or by apoptosis, which can be a granule-independent killing mostly mediated by surface TNF-ligand family members such as FasL or TRAIL (Zamai et al., 2007). The receptors for these ligands are found on the surface of the target cell. NK cells have inhibitory receptors like KIR or Ly49 that form complexes with MHC-I molecules leading to subsequent NK inactivation, therefore a loss of MHC-I in the target can render it sensitive to NK killing. Another mechanism of NK priming is through activating receptors like NCR or NKG2D that bind stress-inducible ligands expressed by tumour cells (Waldhauer & Steinle, 2008). There are also less cytotoxic ILC1s that specialize in the production of mediators of inflammation like TNFα and IFNγ that tilt immunity into a type 1-like environment. This would be favourable for tumour elimination (Dredge et al., 2002).  64  As with ILC3s, less understood is the role that ILC2s play in cancer immunosurveillance. There are considerable amounts of evidence that situate ILC2s in the tolerogenic side of immunity. Large amounts of secreted IL-13, IL-5 and IL-4 are related to poor prognosis (Di Stefano et al., 2010; Zhou et al., 2013). Furthermore, IL-13 production has been recently linked to the activation of myeloid-derived suppressor cells (MDSCs) hence generating an ILC2-MDSC immunosuppressive axis that facilitates tumour immune escape (Trabanelli et al., 2017). This skewing towards type 2 immunity and active MDSC recruitment has also led to recurrence in bladder cancer in humans (Chevalier et al., 2017). In the contrary, studies from Saranchova and colleagues in 2016 demonstrated that IL-33, the ILC2 activator by excellence, is downregulated in certain tumour cell lines when they progress to metastatic stages, reducing the expression of MHC-I, which is linked to the levels of IL-33. The loss of IL-33 enhancing tumour immune evasion and recent evidence demonstrates that ILC2s cooperate with dendritic cells (DCs) to mount TH1 tumour-specific cytolytic T-lymphocyte responses.  ILC1 and NK cell development is regulated, in part, by the same transcription factors. It is the balance between two of them, EOMES and T-bet, what could determine whether the cell becomes a natural killer or an ILC1 (Zook & Kee, 2016), so we can assume that they are closely related and that such plasticity does not involve radical transcriptomic changes, but ILC plasticity goes beyond. Bernink et al. reported that retinoic acid production in CD103+ DCs drives IL-23 and IL-1β secretion by reducing T-bet expression in favour of RORγt, which in turn differentiates ILC1s into IL-23 – producing ILC3s in the intestinal Lamina Propria. In the context of cancer, IL-23 – producing pulmonary squamous cell carcinomas also convert ILC1s into ILC3s (Koh et al., 2019). These shifts in identity are reversible, however both phenotypes are just the ends of a complex spectrum filled with ILC1- and ILC3-related subsets (Cella et al., 2019). In a tumour environment rich in IL-15, ILC1 granzyme B expression is highly upregulated (Dadi et al., 2016), shifting the identity of these innate effectors towards NK-like phenotypes, which is not their “normal” state. A TGF-β – rich tumour on the other hand forces changes in NK cell surface receptors that resemble an ILC1 identity (Gao et al., 2017), 65  triggers the upregulation of several inhibitory receptors (CTLA4, KLRG1, LAG3) and downregulates the expression of CCL5 and IFNγ (Chiossone et al., 2018). It is possible that other ILC subsets possess similar plasticity. ILC2s were previously seen as the most homogenous and stable group within the ILCs, however recent studies are uncovering the existence of additional ILC2 subpopulations (Krabbendam et al., 2018). The flagship of ILC2 heterogeneity is a study carried out by Huang et al. where they demonstrate that there is a second subset of ILC2s that respond primarily to IL-25 instead of IL-33. These inflammatory ILC2s seem undetectable in steady state but can be elicited upon alarmin release and can differentiate into natural IL-33 – responding ILC2s or can upregulate RORγt and acquire ILC3 functions. This is accompanied by an increase in IL-17 secretion and what appears to be central in ILC2 plasticity: loss of GATA-3 and a halt to the production of IL-5 and IL-13 (Krabbendam et al., 2018). Inflammatory ILC2s are closely related to ILC3 phenotypes but not to ILC1s. In this context, Ohne et al. discovered that IL-1β is a potent ILC2 activator by causing upregulation of IL-5, IL-13, ST2, IL-25R but also T-bet and IL-12Rβ2, which in response to IL-12 and an accompanying inflammatory environment can convert ILC2s into IFNγ-producing ILC1s. In addition, MHC-II molecules and CD80 also increased in response to IL-1β, suggesting an increase in ILC-T cell interactions. This and other studies focused on ILC2-to-ILC1 plasticity (Zhang et al., 2017; Silver et al., 2016) are all carried out in homeostatic conditions. ILC2 identity shifts in the context of cancer remain largely undescribed beyond our initial report (Saranchova et al., 2018). Here, we examined ILC2s that were derived from murine lungs including ILC2s from naïve mice (nILC2), those derived from mice with primary tumours termed primary ILC2s (pILC2) and ILC2s isolated from mice with antecedent metastatic tumours termed metastatic ILC2s (mILC2). We find that ILC2s harvested from the lungs of mice with primary tumours lose type 2 identity and transit into a pro-inflammatory phenotype similar to that of ILC1s or NK cells that may account for the 66  observations in previous studies that connect ILC2s with tumour elimination (Saranchova et al., 2018). These data provide the initial view of ILC2 plasticity in the context of a cancer.  Results ILC2s derived from mice with primary tumours (pILC2s) lose type 2 identity at the mRNA but not the protein level To assess ILC2 viability and functionality we extracted and sequenced RNA from nILC2s, pILC2s (TC1-derived) and mILC2s (A9-derived). We only measured on day 3 of culture due to low cell numbers. We measured the levels of the gata3 gene to confirm the correct lineage identity (Ebihara & Taniuchi, 2019) versus spi1, which encodes the master regulator PU.1 that determines myeloid development (Kueh et al., 2013) and that was absent from our sample. Interestingly, pILC2s expressed lower amounts of the GATA-3 gene compared to the two other populations (Figure 7A), however this was not assessed at the protein level. Driven by these intriguing results, we also assessed the expression of the IL-5 and IL-13 genes in our three ILC2 populations and it was interesting to find that in accordance to the downregulation of gata3, pILC2s had very low expression of il5 and il13 (Figure 7B). This suggests that the common type 2-associated functionality of pILC2s might be altered compared to nILC2s and mILC2s and perhaps caused by a slight drop in GATA-3 expression. In addition, we included the gene for an IFNγ receptor that is highly expressed on macrophages, associated with inflammatory responses and not expected to be found in the surface of ILC2s. However, pILC2s surprised us again by expressing significant amount of the ifngr1 gene compared to the other two ILC2 types (Figure 1B). To assess if this was reflected at the protein level, we examined protein arrays of culture supernatants. Figure 7C displays the log10 concentration of 32 cytokines of interest potentially secreted by ILC2s in their third day of culture. These include IL-5 and IL-13, which did not reflect the RNA expression 67  levels. However, even though ILC2s from tumour-bearing mice appear to produce less IL-5 than those from WTs, the difference is not statistically significant. As for IL-13, naïve ILC2s produce the least amount of this cytokine.  The difference in production of IL-6, a pro-inflammatory cytokine, was however unexpected. mILC2s from metastatic mice produced more IL-6 than WTs (nILC2s). In the same context of inflammation, mILC2s from tumour-bearing mice notably produced more MIP-2, also known as chemokine CXCL2, a powerful polymorphonuclear cell chemoattractant that acts in the early phases of the innate response (Al-Alwan et al., 2013). On the contrary, nILC2s from the lungs of WT mice secrete higher amounts of the immunoinhibitory cytokine IL-10. This knowledge may have important implications for the inflammatory model of cancer etiology. Although statistically not significant, we observed that pILC2s and mILC2s produced high amounts of IL-3 compared to nILC2s, whereas the three populations produced GM-CSF, which mediates proliferation of cells from the myeloid lineage (like IL-3). Secretion of GM-CSF is in accordance with previous cytokine detection assays we did (not included). Finally, we were surprised to observe high amounts of IL-7 produced by the three populations of ILC2s, as it is a cytokine that is not usually secreted by lymphocytes (Tang et al., 1997). Overall, these results demonstrate that at a transcriptional level pILC2s are fundamentally different to mILC2s and nILC2s by downregulating the expression of identity markers such as gata3, il5 and il13 and suggests that nILC2s may reflect the immune-evasion phenotype of metastatic tumours. However, the mRNA expression levels of cytokines were not reflected at the protein level, where we unexpectedly detected secretion of mediators involved in myeloid proliferation, lymphocyte development, immune regulation and inflammation.     68     Figure 7. pILC2s from female mice lose type 2 identity at the RNA but not the protein level on day 3 of culture. A) At the RNA level, purified ILC2s from WT and tumour-bearing mice express the gene for GATA-3 but not PU.1, which is the master regulator for myeloid fate. The GATA-3 gene is also significantly downregulated in pILC2s compared to the other two. FPKM = fragments per kilobase million. B) pILC2s are also the only of the three ILC2 populations that expresses very low amounts of il5 and il13, furthering the loss of type 2 identity. In addition, they express higher amounts of the gene for the IFNγ receptor, suggesting that they are somehow more involved in type 1 immunity. Error bars represent standard error of the mean and n=3 for both experiments (unpaired t-tests). C) Results from a 32-Plex mouse cytokine and chemokine array from Eve Technologies for day 3 culture supernatants from three types of ILC2s. Naïve ILC2s secrete less of the pro-inflammatory cytokines IL-6 and MIP-2 compared to those ILC2s derived from tumour-bearing mice. On the other hand, nILC2s produce more immunoinhibitory IL-10. The three populations produce large amounts of IL-5 and IL-13, as opposed to what was expected from the RNA data, as well as IL-7, which is necessary for ILC survival but not known to be secreted by lymphocytes. Finally, the three populations secreted GM-CSF and only the ILC2s from tumour-bearing mice produced high amounts of IL-3. Both cytokines are involved in myeloid cell recruitment. Error bars represent standard error of the mean and experiment n=3. Statistical tests per cytokine are unpaired t-tests.    69  A subset of pILC2s that lose type 2 identity possess a pro-inflammatory phenotype To deepen the transcriptomic analysis, we performed single cell RNA sequencing (scRNA-seq) on pILC2s at three different time points: primary tumour-bearing mice were terminated and ILC2s isolated at 2, 3 and 4 weeks post tumour establishment. ILC2s were then immediately sent to scRNA-seq (non-activated sample) or cultured for 2-3 days in the presence of IL-33 and TSLP (activated sample). The experimental procedure is summarized in Figure 8A. Here we will focus on describing the non-activated pILC2s for week 2, which are representative for a loss in type 2 identity but display the most obvious skewing towards an ILC type 1 phenotype. This appears to be an activated population that losses gata3, il5 and il13 expression and gains IL-6 expression. Figure 8B illustrates the t-SNE plot from the non-activated population in week 2. The analysis was performed in Cell Ranger and visualized in Loupe Cell Browser and shows six different clusters based on differential gene expression. Cluster 6, which is more distant from the rest in terms of differential gene expression, is also where the expression of gata3, il5 and il13 is the lowest, hinting a lack of type 2 identity (8C, left). In addition, it also contains the most unique molecular identifiers (UMI) expressing ifng (8C, right). Furthermore, we assessed the scRNA-seq data for antigen processing and cross-presentation machinery and found that the UMIs in cluster 6 are not only CD74+ (Figure 8D, left), which plays a role in cross-presentation of MHC-I and MHC-II (Basha et al., 2012), but also MHC-I+ and MHC-II+ (8D, right). This reinforces our suspicion that pILC2s have antigen-presenting cell (APC)-like properties and are able of cross-presentation (see accompanying manuscript). Interestingly, the globally distinguishing features for cluster 6 with lowest P value (Table 1) are the genes for the granzymes B & C, the CD3γ subunit and IFITM1, an IFNα/γ-induced transmembrane protein involved in the transduction of antiproliferative signals (Yang et al., 2006). Given the large pro-inflammatory interest of these genes and the intriguing presence of a CD3 subunit in an innate lymphoid cell, we tried to identify this gzmb+ gzmc+ ifitm1+ cd3g+ pILC2 population by staining with available fluorescently-tagged antibodies for three markers gzmb+ gzmc+ ifitm1+ (antibodies for CD3γ 70  were not available) and then visualized the samples via flow cytometry. Figure 8E (left) depicts pILC2s positive for intracellular granzyme B and C. The high expression of granzyme B was surprising, as well as the two populations that produce more or less of this granule (8E, right). We also detected a small number of intracellular IFITM1+ cells (Figure 8F). We directly assessed previous RNA data and found that pILC2s express more of the CD3γ subunit than nILC2s and mILC2s (Figure 8G), but the they also express the genes of the CD3δ/ε subunits.  In conclusion, these set of results demonstrate that a subset of pILC2s not only lose expression of genes tightly linked to type 2 identity, but also express markers of inflammation and cytotoxicity. In addition, it seems that these pro-inflammatory ILC2s also express genes demonstrated to be involved in cross-presentation and ILC-T cell dialogue, which explain data in the accompanying experiments where pILC2s are successful at cross-presenting and cross-priming CD8+ T-cells.71                    Figure 8. A subpopulation of pILC2s lose type 2 identity, express type 1-associated genes and has potential to have acquired APC-like properties. A) Experimental procedure prior to scRNA-seq analysis using Cell Ranger and Loupe Cell Browser for visualization in B) t-SNE plot for a non-activated population of pILC2s isolated 2 weeks after TC1 tumour establishment in mice. The cluster of interest, cluster 6, is highlighted inside a red circle. C) UMIs expressing gata3, il5 and il13 except in cluster 6, which in turn expresses the IFNγ gene. D) Several UMIs in cluster 6 also express the genes for CD74 and four murine MHC-II genes, suggesting they could be involved in dialogues with other T cells and cross-presentation. Statistics calculated through a negative binomial exact test for lower UMI counts and an asymptotic beta test for higher counts. E) Intracellular stain for granzymes B & C where a portion of pILC2s are positive for both, but mostly for granzyme B, where two populations can be observed based on the amount of granzyme B they store (histogram: red curve is a blank control). F) A low percentage of pILC2s were also positive for intracellular IFITM1. G) pILC2s express significantly higher amounts of CD3 genes than nILC2 and mILC2 cells. For currently unknown reasons, cd3g is the gene that they express the most. FPKM = fragments per kilobase million. Error bars represent standard error of the mean and experiment n=3. Statistical analysis: unpaired t-tests.72   Table 1. Upregulated genes in cluster 6 differentially expressed from the other clusters in a non-activated population of pILC2s harvested 2 weeks post tumour establishment. In a list comprising the 50 most upregulated genes in cluster 6 compared to the rest, only the genes encoding granzymes B & C, IFITM1 and the CD3γ subunit are upregulated with statistical significance. These four globally distinguishing features visualized in Loupe Cell Browser drove us to find this possibly pro-inflammatory subset of pILC2s via flow cytometry, as previously depicted in Figures 2E, F. We could not detect any fluorescent signal for CD3γ (not included) so we checked the RNA data as shown in Figure 2G, where it appears evident that as expected, ILC2s do not express CD3 molecules unless they are harvested from mice with primary tumours. The explanation to these intriguing results remains unclear. Statistical analysis in Cell Ranger was done through a negative binomial exact test for lower UMI counts and an asymptotic beta test for higher counts.           73  ILC2s derived from primary tumour-bearing mice are heterogeneous  To further resolve the heterogeneity of pILC2s we performed another sc-RNAseq downstream analysis using the R software package Seurat on the same samples as in Figure 8. As stated in Figure 8A, we sorted lung pILC2s and prepared activated and non-activated samples from weeks 2, 3 and 4 post tumour inoculation. From week 4 we only analyzed a non-activated population. We pooled lung pILC2s from five mice per sample and between 200 and 2,500 cells for each sample were characterized. Unsupervised clustering by t-SNE in the non-activated, week 2 pILC2 population identified eight discrete clusters (numbered 0-7) based on the ten most differentially expressed genes, as shown in Figure 9A. The corresponding heat map is shown in Figure 9B. Among these, the clusters with most cells were 0 and 1. Cells in cluster 5 and to some extent in cluster 1 had the highest expression of il5 and il13, suggesting a more conventional ILC2 phenotype. On the other hand, cells in cluster 3 show a marked downregulation in these two genes, again demonstrating the heterogeneity of these cells. The dynamics in the expression of these two cytokines are summarized in Figure 3C. Of interest is cluster 6 (Figure 9B), which not only shows a lack of il5 and il13, but also shows high expression of the invariant chain cd74. Also, although MHC-II genes were not in the top five differentially expressed genes (there were less cells for cluster 6), they were still significantly upregulated, as displayed in the volcano plot in Figure 9D. Furthermore, ifitm3 was also highly expressed in cluster 6. The gene for CD3γ was upregulated in cells from cluster 5 instead, but also present in clusters 0-3. Cluster 7 was differentiated for the multiple mitochondrial genes that were upregulated. These findings are summarized in Figure 3E showing t-SNE plots with several genes of interest and in Figure 9D. In summary, non-activated samples from week 2 show similar gene expression patterns despite the different software employed. This adds reliability to our studies. We have also included the t-SNE clustering for the non-activated (Figure 10) and activated samples (Figure 11) from week 3. Different UMI distributions already suggests differential gene expression in an activated or quiescent state. Clusters 0 and 3 in Figure 10 highly express the genes for the three 74  CD3 subunits, whereas il5 expression was detectable in clusters 1, 4 and to some extent in 2. The t-SNE for Figure 10 suggests that these clusters are heterogeneous relative to each other, and that clusters 0 and 3 approach the phenotype we detected in Cell Ranger with a lack of type 2 cytokines and the puzzling upregulation of CD3 genes. In accordance to our previous analysis, the activated population in Figure 11 only displays one cluster with high type 2 cytokine expression (cluster 5), but also an IL-1R gene and ccr10, which are both related to inflammation. Cluster 7 was enriched with CD3 genes and TCR constant region-related genes, which is an intriguing gene expression profile for a member of the ILC family. These last results show that ILC-type 2 identity is altered when the pILC2 population is activated (Figure 9C), but the potentially pro-inflammatory il5lo il13lo pILC2 subset can be found in every sample. These results demonstrate that whether pILC2s are activated or not, they are a heterogeneous population, which supports the idea of having a subset that is more skewed towards an ILC type 1 identity and overall reinforces previous studies on ILC2 plasticity. A summary heat map of gene expression in every sample over time and a summary heat map based on immunological pathways is presented in the appendix (Figures. S1 and S2). 75       Figure 9. The scRNA-seq of the week 2, non-activated pILC2 population reveals a cluster with potential APC-like properties and transitioning to a pro-inflammatory phenotype. A) Unsupervised t-SNE analysis of the sample to define clusters. B) Heat map showing the top ten differentially expressed genes per cluster except for cluster 6, which only shows five genes due to lower UMI count. C) Violin plots comparing IL-5 and IL-13 gene expression between activated and non-activated populations through time. D) Volcano plot displaying upregulation and downregulation of the most significantly expressed genes in the whole sample. E) t-SNE of representative expressed genes involved in antigen-presenting machinery, T-cell dialogue, cross-presentation (top two rows) and inflammation (bottom row). These are mostly expressed in cluster 6, which also downregulates il5 and il13.  76            Figure 10. Week 3, non-activated pILC2s. The t-SNE plot identifies 8 different clusters whose ten most differentially expressed genes are shown in the heat map. Figure 11 Week 3, activated pILC2s shows the same analysis for a week 3, activated population of pILC2s, with 7 distinguished clusters. The highly different t-SNE distributions demonstrates a difference between the quiescent and the activated state, mediated by alarmins such as IL-33 and TSLP in ILC2 culture media. 77  Discussion These studies in ILC2 heterogeneity have been motivated by the discovery of the importance of IL-33 in cancer immunosurveillance and immune-evasion (Saranchova et al., 2016; Alimonti et al., 2000) later termed immune-editing, and by the demonstration of the important role that ILC2s play in anti-tumour immunity (Saranchova et al., 2018). The immediate goal was to uncover a mechanism of action. In the accompanying paper, we show that pILC2s that are derived from mice enduring a primary tumour are capable of MHC-I cross-presentation to CD8+ T-cells, a function conventionally reserved for dendritic cells. pILC2s are better at cross-presentation than naïve ILC2s from WT mice and mILC2s derived from metastatic mice. Interestingly, mILC2s are as effective as pILC2s at CD8+ cross-priming. This led us to address if molecular differences exist between nILC2s, pILC2s and mILC2s that enable ILC2 immunosurveillance against tumours. We, therefore, examined if pILC2s skew from the conventional definition of ILC2, a type 2 innate mediator, towards a more inflammatory phenotype capable of initializing dialogue with TH1 lymphocytes. The first indicators of plasticity that we observed were at the RNA level, where pILC2s expressed less GATA-3 gene than their metastatic and naïve counterparts (Figure 7A). Given the importance of GATA-3 in ILC2 development and function (Lim et al., 2017), it is now well established that downregulation of this transcription factor is a common trend in ILC2 identity shifts, wherever that plasticity is heading to (Krabbendam et al., 2018; Lim et al., 2016; Ohne et al., 2016; Silver et al., 2016; Zhang et al., 2017). Next, and also at the RNA level, we detected astonishingly low expression of il5 and il13 in pILC2s, compared to mILC2s and nILC2s (Figure 7B). Both cytokines are a signature of conventional ILC2 functionality (Zhu, 2015) and under inflammatory conditions (specifically in the presence of IL-1β or IL-12) both are downregulated (Ohne et al., 2016). We suspect that a primary tumour, expressing IL-33 that has not yet evaded immunity through reduction of MHC-I expression, triggers an immune response that entails systemic circulation of pro-inflammatory mediators, affecting the transcriptional program of lung resident ILC2s as stated. 78  Furthermore, IL-13 appears to enhance immunosuppressive functions in tumours by recruiting MDSCs associated to poor prognosis, as well as by polarizing macrophages into the M2 phenotype (Trabanelli et al., 2017; Chevalier et al., 2017; Dhakal et al., 2014). This aligns with the idea of an anti-tumour response mediated by pILC2s. Also interesting was the expression of the IFNγ receptor gene ifngr1 in pILC2s (Figure 7B). Enhancing the upregulation of T-bet paralleled to a loss in GATA-3 expression forces ILC2s to be IFNγ+ IL-12Rβ1+ IL-18R+ cells (Belz, 2016), therefore sensitive to inflammation induction but not to IFNγ. In the context of inflammation and plasticity there is not much evidence for IFNGR+ ILC2s, however ILC2 identity shifts do not always involve the transition into another ILC type. Like in any other area of the immune system, type 2 responses must be regulated, and thus there is a mechanism for ILC2 regulation. As type 1 and type 2 immunity counter-regulate each other, it has been reported the presence of IFNGR+ ILC2s that downregulate cytokine production upon IFNγ binding (Stehle et al., 2015). The pro-inflammatory environment in the presence of a primary tumour may enhance ifngr1 transcription in pILC2s to render them sensitive to IFNγ and in turn suppress potential immunoinhibitory ILC2 responses. This hypothesis and the presence of a pro-inflammatory subset of pILC2s are not mutually exclusive. We analyze a set of 32 cytokines and chemokines in culture supernatant from pILC2s but also naïve and metastatic ILC2s (Figure 7C). We continued the analysis at the protein level with an exploratory focus, hence the many cytokines included. The most striking and unexpected result was the high concentrations of IL-5 and IL-13 for the three populations, including the pILC2s, where we detected a low number of transcripts for these cytokines in Figures 7A & 7B. The cells in the media we analyzed were used for downstream applications other than RNA sequencing, meaning they were not from the same batch of mice. Coming from a distinct group of mice bearing primary tumours could explain this mRNA/protein contradiction in our results, where less ILC1-like IL-5- IL-13- cells differentiated, leading to the IL-5+ IL-13+ ILC2 majority in Figure 7C.  79  Other contributing factors could be low protein turnover and therefore high stability. IL-13 mRNA, for example, is highly stabilized by a 3’ untranslated region-binding protein called HuR that promotes stability and translation (Casolaro et al., 2008). Even though the FPKM count is low, one transcript could translate into multiple proteins. In addition, IL-5 mRNA has been shown to have an absolute dependency on high protein synthesis for further il5 transcription (Umland et al., 1997). This suggests that it could take minimum 3 days of culture (Figure 7C) to achieve high enough IL-5 concentrations to promote mRNA transcription and increase the low fragment counts we observed.  Other cytokines of interest are the pro-inflammatory chemokine MIP-2 that mobilizes polymorphonuclear leukocytes and hematopoietic stem cells and the acute phase reactant IL-6. Both of them secreted at higher concentrations from tumour-derived ILC2s than nILC2s. Interestingly, the latter produces significantly higher concentrations of the immunomodulatory cytokine IL-10. Reports of IL-10+ ILC2s are not new. In 2017, Seehus et al. identified a subpopulation of ILC2s that produce IL-10 in response to IL-33 stimulation that retained type 2 cytokine production such as IL-13. These were named ILC210, and are lung resident cells capable of participating in anti-inflammatory processes. The higher expression of IL-10 in nILC2s aligns with previous models of ILC2s exerting pro-tumorigenic roles in cancer (Di Stefano et al., 2010; Zhou et al., 2013). Also, although the three ILC2 populations we identify produced GM-CSF, which we detected in previous cytokine assays (not included), only pILC2s and mILC2s produced IL-3. It seems that when derived from mice with tumours, ILC2s shift their cytokine secretome.  The obvious differences between pILC2s and the nILC2s and mILC2s at the RNA level were explored in more detail by performing scRNA-seq analysis using the 10X Genomics platform, clustering through Cell Ranger and visualizing with Cell Loupe Browser. Most intriguing was the sample from a non-activated population of pILC2s from week 2, where the t-SNE identified six discrete clusters (Figure 8B). Cluster 6 lost expression of the genes for GATA-3, IL-5 and IL-13 (Figure 8C) and upregulated IFNγ (not significant), which confirms what we have observed in previous RNAseq 80  experiments and suggests a shift to an inflammatory role. Consistent with our data in the accompanying paper demonstrating pILC2s functioning in cross-presentation, we were interested to see that cells in cluster 6 were CD74+ and MHC-II+ (Figure 8D). Oliphant and colleagues already confirmed that a subset of ILC2s express MHC-II and initiate a cross-talk with TH2 cells that promotes mutual growth and enables foreign antigen presentation, like professional APCs. Basha et al., on the other hand, demonstrated that the invariant chain CD74 not only prevents premature binding of peptides in the MHC-II, but it also associates with the MHC-I and shuttles a fraction of these molecules to the endocytic compartment, where it can incorporate exogenous peptides and cross-present. With this data we already suspect that the cross-presenting and hence cross-priming subset of pILC2s is the same one that downregulates type 2 markers and upregulates pro-inflammatory genes. Table 1 summarizes the four globally distinguishing genes that separate cluster 6 from the rest: gzmb, gzmc, ifitm1 and cd3g. Loss of type 2 identity and upregulation of granzyme RNA is pointing towards NK identity. There could be two explanations for this observation. First, transcriptional commonalities in ILC2 and NK developmental programmes are responsible for early and low expression of granzyme B and NKG2 in differentiating ILC2s (Zamai et al., 2015) and second, ILC2 plasticity towards NK- or ILC1-like phenotypes. Cluster 6 could therefore be a subset of ILC2 progenitors in earlier phases of development or a more plastic population. We discard the possibility of having fully differentiated NK cells as we use a lineage antibody against NK1.1 (KLRB1), which is present in every NK cell from C57Bl/6 mice (Walzer et al., 2007). Nevertheless, Camelo and colleagues have already reported an increase in cell granularity due to augmented levels of intracellular granzyme B in ILC2s cultured in the same serum-free media we employed, supplemented with IL-33. Instead, in the presence of IL-1β, their ILC2s expressed NK markers such as NKp30 or NKG2D. Their data, along with our results, suggest an induction of plasticity mediated by IL-33 and perhaps in combination with IL-2. 81  Granzyme-expressing ILC2s could exert very important roles in cancer and viral infections. The obvious thought is that tumour or infected cells can be targeted by a classic perforin-dependant granzyme killing pathway, but there is no evidence for perforin-expressing ILC2s. Granzymes, however, have immunomodulatory functions that go beyond cytotoxicity, which stresses the statement that not all granzyme-expressing cells are cytotoxic. A report by Hoves et al. in 2012 postulates that granzymes can also regulate cross-presentation by DCs by acting as “eat me” signals on dying tumour cells, easing antigen uptake, CTL priming, and subsequent anti-tumour responses. As we demonstrate in our accompanying paper, ILC2s can cross-present antigen and cross-prime CD8+ T-cells, so perhaps the presence of granzyme B in ILC2s is not only a consequence of a change in the transcriptional program due to plasticity induction, but also a potential mechanism to facilitate themselves the acquisition of antigen from dying tumour cells and cross-present it; as well as a potential novel pathway to support DC cross-presentation and overall, TH1 responses. The fact that multiple non-cytotoxic cells besides ILC2s (basophils, mast cells, platelets…) have been found to express granzyme B implies that this protease might have other extracellular roles (Afonina et al., 2010). Given the large variety of extracellular matrix components, cell surface markers or serum proteins that can act as granzyme B substrates (Buzza et al., 2005), their proteolysis by granzyme B could lead to cell detachment, increased leukocyte extravasation, alarmin release or even macrophage activation (Kuhns et al., 2007; Okamura et al., 2001). In other words, granzymes can also be mediators of inflammation, which supports the idea of a newly identified pro-inflammatory ILC2 subset. In an attempt to find the gzmb+ gzmc+ ifitm1+ cd3g+ subset of pILC2s that we previously described, we intracellularly stained the cells for these markers. Figure 8E shows two populations of pILC2s that are granzyme B+ (histogram), but also granzyme C+. There is little to no evidence about IFITM1 expression in type 2 ILCs, but its transcript has been often associated to the NK cell response (Bezman et al., 2012) and a small percentage of our ILC2s expressed it (Figure 8F). What is truly unexpected is the expression of CD3 molecules in pILC2s. CD3δ mRNA expression in ILC2s has been detected 82  in the past (Robinette et al., 2015; Wallrapp et al., 2017). However, our RNAseq and scRNA-seq from two different experiments coincide to find CD3γhi CD3δ+ CD3ε+ pILC2s (Figure 8G). During the staining phase, our lineage markers ensured that these ILC2s were TCRβ- CD3ε/γ- CD4- CD8α- at sorting, suggesting that T-cell identity is highly unlikely and that they could represent an entirely novel population. In 2007, De Smedt and her group demonstrated that Notch signalling induced CD3ε expression in differentiating NK cells. Notch signalling is essential to differentiate common lymphoid progenitors into the first established ILC progenitor, the α-lymphoid precursor, that in turn gives rise to the first committed NK precursor. We therefore hypothesize that CD3 expression may be acquired and retained in early phases of ILC development, when developmental programs are still not too divergent from each other. Future studies will be essential to elucidate the biological implications of this novel population of CD3+ ILC2s. To deepen our understanding, we employed the R software package Seurat to re-do the analysis from the same samples analyzed in Figure 2 with the same week 2 non-activated population. The t-SNE identified eight discrete clusters summarized in a heat map (Figures 9A & 9B). Coincidentally, cluster 6 was our cluster of interest again, being low in il5 and il13 but high in ifitm3 and cd74. Although it is not IFITM1, which is the one we identified using Cell Ranger, the genes for ifitm1, 2, and 3 are regulated by the same IFN-induced enhancer (Li et al., 2017). Interestingly, cluster 3 was one of the few where il5 and il13 were clearly downregulated and at the same time expressing cd3g. From a broader perspective and according to Figure 9C, the genes for type 2 cytokines are lost over time in non-activated populations, but rescued in activated populations overall, even though other samples also show high pILC2 heterogeneity, especially regarding il5 and il13 expression. Although class II MHC genes are not in the top 10 differentially expressed genes (therefore not in the heat map), they are still significantly upregulated in the whole sample and not distinguishing between clusters, as shown in the volcano plot (Figure 9D). On the other hand, type 2 cytokines are downregulated. However, the t-SNEs in Figure 9E informs us that MHC-II and cross-presentation genes, as well as 83  those related to pro-inflammatory processes, localize to cluster 6, which is our subset of interest in this sample.  In samples from week 3, Clusters 0 and 3 upregulate the genes for all of the main CD3 subunits, whereas clusters 1, 2 and 4 downregulate them (Figure 10) and provide an overview of the heterogeneity within a purified population of ILC2s at this time point and demonstrate the dynamic changes that ILC2s are capable of. In Figure 11, where pILC2s were activated, clusters 0, 1 and 3 express type 2 cytokines. Undeniably, the expression of these genes varies across samples and therefore through time, but what remains clear is the high heterogeneity and also the presence of a small inflammation-related, non-type 2 cluster.  A fascinating yet perplexing finding is the presence of the pre-TCRα surrogate chain gene (ptcra) along with the three CD3 components cd3g, cd3e and cd3d in clusters 0, 3 and 5 from Figure 10. Especially given the fact that they are among the ten most upregulated genes for those clusters. Furthermore, in Figure 5, cluster 6 highly expresses other T-cell markers such as tcf7 (encodes TCF-1), cd3g, the gene for the TCRδ constant region (trdc) and the gene for the TCRβ constant domain 1 (trbc1).  We are the third group to identify TCR gene signatures in innate lymphoid cells (Björklund et al., 2016; Li et al., 2019). To discard contamination, Björklund and colleagues found that their TCR and CD3-related genes detected in the three types of ILCs were scattered throughout their sample and not confined to one cluster. Detecting these signatures for a third time suggests that they are common among ILCs. Coming from the same progenitor that generates T-cells might also explain the presence of TCR-related genes in ILC2s. Highly similar transcriptional profiles between ILCs and T-cells could imply a strong conservation of the TCR recombination machinery by ILCs (Retière et al., 1999), and would explain why they mimic the cytokine transcriptional profile of their adaptive counterparts (Li et al., 2019). We presume that in order to ensure successful terminal differentiation of non B- non T- innate lymphocytes and to avoid predisposition to certain recombination events there must be some 84  sort of developmental blockade of TCR rearrangements in developing ILCs. Still, this could explain the presence of the TCRβ/δ constant region genes that us and other groups have detected.  Notwithstanding, we are the first group to report significant expression of the pre-TCRα chain gene (ptcra) in an innate lymphoid cell. Ptcra gene transcription depends on E-box binding proteins like E47 strongly synergizing with Pip-1 (Petersson et al., 2002). Interestingly, E47 is a splice variant of the transcription factor E2A, which synergizes with another E protein, HEB, to suppress Id2 and thus, deny an opportunity for ILC digression in favour of T-cell development (or B-cell development if it interacts with Pip-1) in early stages of differentiation (Nagulapalli & Atchison, 1998; Miyazaki et al., 2017). Also, in T-cells, the signalling of the pre-TCRα chain inhibits further TCR germline rearrangements (Abbas et al., 2018). We suspect that suppressing the ILC lineage via synergistic transcriptional complexes with E proteins at early common precursor stages, might result, as a by-stander effect, in the upregulation of a pre-TCRα chain in ILC progenitors. Such progenitors might have upregulated Id2 (an E-protein repressor that promotes ILC lineage) shortly after E47 expression. This would allow a brief period of existence for E47 to promote expression of a pre-TCRα chain that might in turn block further TCR rearrangements in ILCs. Back to our single cell study, we confirm that ILC2s are a heterogeneous cell type that shift their expression profile depending on the stage of the tumour in animals. Our focus was on pILC2s given the cross-presenting capabilities they displayed for MHC-I cross-presentation. Studies on cancer immunosurveillance by Saranchova et al. demonstrate that ILC2s play a novel, fundamental role in cancer. We suspect that this role is to provide support to other APCs on neoantigen cross-presentation to activate a population of CD8+ lymphocytes to mount an effective and specific anti-tumour response. Our data suggests that only a fraction of the lung resident ILC2s will be able to do so. Here we demonstrate that the subset that upregulates genes associated to T-cell dialogue and cross-presentation is the same pILC2 population that downregulates type 2-associated genes and upregulate pro-inflammatory genes. 85  We know that this pro-inflammatory subset of ILC2s arises when they are harvested from mice with primary tumours, meaning that a pro-inflammatory, systemic environment could be responsible for the priming of this subset. There is already evidence of such phenomenon. Silver and colleagues have reported that in influenza-infected mouse lungs, the tissue-resident ILC2s that gathered around infected foci displayed GATA-3 downregulation and expressed IFNγ in response to IL-12, which is a remarkable example of a pro-inflammatory environment directly tailoring ILC2 identity into a more useful phenotype to fight the threat. In our case, we suspect that an expanded population of these type 1 ILC2s has been a key agent in the anti-tumour response that Saranchova and colleagues observed. This pro-inflammatory subset may contribute to the inflammatory model of cancer etiology. Further studies will be necessary to identify what are the exact factors that give rise to these unconventional ILC2 phenotypes. Normally, ILC2 development and function is studied under allergic or helminthic conditions. In future ILC2 studies, it will be necessary to trigger inflammatory responses in mice and characterize lung resident ILC2s to observe in more detail what is the effect of systemic and/or chronic inflammation in these type 2 innate mediators. In addition, future studies focused on the TCR germline in innate lymphoid cells could put to the test the “I” in ILCs.  Conclusion In recent years, immunologists around the world have been characterizing ILC1 and ILC3 populations in detail, unmasking unprecedented levels of plasticity between these cells (Bernink et al., 2015; Koh et al., 2019; Viant et al., 2016; Verrier et al., 2016). Immediately after came ILC2s, where plasticity towards ILC3 and ILC1 phenotypes was also uncovered (Krabbendam et al., 2018; Zhang et al., 2017; Silver et al., 2016). Here, we report conventional ILC2s that abandon the well-established type 2 phenotype towards a pro-inflammatory identity that appears to mediate CD8+ T-cell cross-priming that supports effective anti-tumour responses (Saranchova et al., 2018). The subset we identified was detected in ILC2s harvested from the lungs of mice with a primary MHC-I +, Il-33+ tumour at 2, 3 86  or 4 weeks, suggesting that these pro-inflammatory pILC2s – or type 1 ILC2s – arise in the context of cancer or inflammation. Whether these cells can be exploited as a therapy remains unclear, but we have now revealed the presence of an new Immune cell type that functions in immune surveillance.   87  Final Conclusion This thesis has updated the current knowledge on type 2 innate lymphoid cells, but as a nascent field that it is, there is a lot of work ahead. Our group, along with the Piret Laboratory, has demonstrated that serum-free media is a better choice than RPMI 1640 to culture lung ILC2s from mice, reaching up to 10-fold expansion from original explants of diverse cell concentrations. Although our cytokine combination is efficient, the field still lacks a thorough study on ILC2 cytokine complementation cultured in serum-free media specifically. Furthermore, there are no published ILC2 freeze/thaw protocols, and thus, it should be a major milestone. More importantly, we have reported cross-presentation and cross-priming activity from lung ILC2s, however, we are the first group to work with ILC2s derived from tumour-bearing mice, and perhaps this is why we have observed such an unconventional function in a cell that until recently was considered well characterized. Since we have tested OVA257-264 as an antigen for the highly specific OT-I TCR, a strong corroboration of ILC2 cross-presentation would be testing a variety of different antigens, including potential neoantigens, in a population of CD8+ T-cells. We attribute such cross-priming capabilities not to the whole ILC2 pool, but to a pro-inflammatory subset that we believe expands in vivo after two weeks in animals with tumours, where systemic inflammation or active immune surveillance may affect the lung microenvironment and eventually the resident ILC2 transcriptome. Our single cell RNA sequence analysis has revealed high heterogeneity across time within these tumour-bearing mice – derived populations of ILC2s, where a portion of the population loses type 2 identity markers in favour of pro-inflammatory type 1 markers and genes related to the antigen presentation machinery. In the future, it is of utmost importance to study lung ILC2s under different inflammatory triggers, systemic or local, such as distal or lung tumours, viral infection or induced systemic inflammation. In summary, conditions opposite to classic type 2 immunity such as allergen or helminthic stimulation. We hope that the future work in the field identifies the exact triggers that induce ILC2 plasticity in vivo.    88  Bibliography 1. Abbas, A.K. et al (2018). Cellular and Molecular Immunology. 9th Ed. Elsevier. Philadelphia, PA. 2. Ackerman, A.L. et al (2003). Early phagosomes in dendritic cells form a cellular compartment sufficient for cross-presentation of exogenous antigens. Proceedings of the National Academy of Sciences. 100. p. 12889-12894. 3. Afonina, I.S. et al (2010). Cytotoxic and non-cytotoxic roles of the CTL⁄NK protease granzyme B. Immunological Reviews. 235. p. 105-116. 4. Akira, S. & Takeda, K. (2004). Toll-like receptor signalling. Nature Immunology. 4. p. 499-511. 5. Al-Alwan, L.A. et al (2013). Differential Roles of CXCL2 and CXCL3 and Their Receptors in Regulating Normal and Asthmatic Airway Smooth Muscle Cell Migration. The Journal of Immunology. 191 (5). p. 2731-2741.  6. Alimonti, J. et al (2000). TAP expression provides a general method for improving the recognition of malignant cells in vivo. Nature Biotechnology. 18. p. 515-520.  7. Ardain, A. et al (2019). Type 3 ILCs in Lung Disease. Frontiers in Immunology. 10 (92). p. 1-8. 8. Artis, D. & Spits, H. (2015). The biology of innate lymphoid cells. Nature. 517 (7534). p. 293-301. 9. Au-Yeung, B. et al (2019). IL-2 modulates the TCR signaling threshold for CD8 but not CD4 T cell proliferation on a single cell level. The Journal of Immunology. 198 (6). p. 2445-2456. 10. Bal, S.M. et al (2016). IL-1b, IL-4 and IL-12 control the fate of group 2 innate lymphoid cells in human airway inflammation in the lungs. Nature Immunology. 17 (6). p. 636-648. 11. Bal, S.M. et al (2020). Plasticity of innate lymphoid cell subsets. Nature Reviews Immunology.  12. Barlow, J. et al (2012). Innate IL-13-producing nuocytes arise during allergic lung inflammation and contribute to airways hyperreactivity. Journal of Allergy and Clinical Immunology. 129 (191–198). e191–194. 13. Barlow, J.L. et al. (2013). IL-33 is more potent than IL-25 in provoking IL-13–producing nuocytes (type 2 innate lymphoid cells) and airway contraction. Journal of Allergy and Clinical Immunology. 132 (4). p. 933-941. 14. Basha, G. et al (2008). MHC Class I Endosomal and Lysosomal Trafficking Coincides with Exogenous Antigen Loading in Dendritic Cells. Public Library of Science ONE. 3(9). e3247. 15. Basha, G. et al (2012). A CD74 - dependent MHC class I endolysosomal cross-presentation pathway. Nature Immunology. 13 (3). p. 237-246. 16. Belz, G.T. (2016). ILC2s masquerade as ILC1s to drive chronic disease. Nature Immunology. 17. p. 611-612. 89  17. Bernink, J.H. et al (2013). Human type 1 innate lymphoid cells accumulate in inflamed mucosal tissues. Nature Immunology. 14. p. 221-229. 18. Bernink, J.H. et al (2014). The role of ILC2 in pathology of type 2 inflammatory diseases. Current Opinion in Immunology. 31. p. 115-120. 19. Bernink, J.H. et al (2015). Interleukin-12 and -23 Control Plasticity of CD127+ Group 1 and Group 3 Innate Lymphoid Cells in the Intestinal Lamina Propria. Immunity. 43 (1). p. 146-160. 20. Bernink, J.H. et al (2017). Human ILC1: To Be or Not to Be. Immunity. 46. p. 756-757. 21. Bernink, J.H. et al (2019). c-Kit-positive ILC2s exhibit an ILC3-like signature that may contribute to IL-17-mediated pathologies. Nature Immunology. 20. p. 992-1003. 22. Berrett, H. et al (2019). Development of Type 2 Innate Lymphoid Cells Is Selectively Inhibited by Sustained E Protein Activity. ImmunoHorizons. 3 (12). p. 593-605.  23. Bevan, M.J. (1976). Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. The Journal of Experimental Medicine. 143 (5). p. 1283-1288. 24. Bezman, N.A. et al (2012). Molecular definition of the identity and activation of natural killer cells. Nature Immunology. 13. p. 1000-1009. 25. Bie, Q. et al (2014). Polarization of ILC2s in peripheral blood might contribute to immunosuppressive microenvironment in patients with gastric cancer. Journal of Immunology Research. 2014: 923135. 26. Björklund, A.K. et al (2016). The heterogeneity of human CD127+ innate lymphoid cells revealed by single-cell RNA sequencing. Nature Immunology. 17. p. 451-460. 27. Brestoff, J.R. et al (2015). Group 2 innate lymphoid cells promote beiging of white adipose tissue and limit obesity. Nature. 519. p. 242-246. 28. Bruce, D.W. et al (2017). Type 2 innate lymphoid cells treat and prevent acute gastrointestinal graft-versus-host disease. The Journal of Clinical Investigation. 127 (5). p. 1813-1825. 29. Busser, B. et al (2011). The multiple roles of amphiregulin in human cancer. Biochimica et Biophysica Acta – Reviews on Cancer. 1816 (2). p. 119-131.  30. Butler, A. et al (2018). Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nature Biotechnology. 36. p. 411-420.  31. Buzza, M.S. et al (2005). Extracellular matrix remodeling by human granzyme B via cleavage of vitronectin, fibronectin, and laminin. Journal of Biological Chemistry. 280. p. 23549-23558. 32. Camelo, A. et al (2017). IL-33, IL-25, and TSLP induce a distinct phenotypic and activation profile in human type 2 innate lymphoid cells. Blood Advances. 1 (10). p. 577-589. 33. Carriere, V. et al (2007). IL-33, the IL-1 cytokine ligand for ST2 receptor, is a chromatin-associated nuclear factor in vivo. Proceedings of the National Academy of Science. 104 (1). p. 282-287. 90  34. Casolaro, V. et al (2008). Posttranscriptional regulation of IL-13 in T cells: Role of the RNA-binding protein HuR. Journal of Allergy and Clinical Immunology. 121 (4). p. 853-859. 35. Cautivo, K.M. & Molofsky, A.B. (2016). Regulation of metabolic health and adipose tissue function by group 2 innate lymphoid cells. European Journal of Immunology. 46. p. 1315-1325. 36. Caza, T. & Landas, S. (2015). Functional and Phenotypic Plasticity of CD4+ T Cell Subsets. BioMed Research International. 521957. 37. Cella, M. et al (2019). Subsets of ILC3−ILC1-like cells generate a diversity spectrum of innate lymphoid cells in human mucosal tissues. Nature Immunology. 20. p. 980-991.  38. Cherrier, M. et al (2012). Notch, Id2, and RORγt sequentially orchestrate the fetal development of lymphoid tissue inducer cells. Journal of Experimental Medicine. 209. p. 729-740. 39. Chevalier, M.F. et al (2017). ILC2-modulated T cell-to-MDSC balance is associated with bladder cancer recurrence. The Journal of Clinical Investigation. 127 (8). p. 2916-2929. 40. Chiossone, L. et al (2018). Natural killer cells and other innate lymphoid cells in cancer. Nature Reviews Immunology. 18. p. 671-688. 41. Cildir, G. et al (2013). Chronic adipose tissue inflammation: all immune cells on the stage. Trends in Molecular Medicine. 19. p. 487-500. 42. Clarke, S.R. et al (2000). Characterization of the ovalbumin-specific TCR transgenic line OT-I: MHC elements for positive and negative selection. Immunology and Cell Biology. 78 (2). p. 110-117. 43. Constantinides, M.G. et al (2014). A committed precursor to innate lymphoid cells. Nature. 508. p. 397-401. 44. Cooper, M.D. & Miller, J.F.A.P. (2019). Discovery of 2 Distinctive Lineages of Lymphocytes, T Cells and B Cells, as the Basis of the Adaptive Immune System and Immunologic Function. Journal of the American Medical Association. 322 (13). p. 1247-1248. 45. Cortez, V.S. et al (2015). Innate lymphoid cells: new insights into function and development. Current Opinion in Immunology. 32. p. 71-77. 46. Dadi, S. et al (2016). Cancer immunosurveillance by tissue-resident innate lymphoid cells and innate-like T cells. Cell. 164. p. 365–377. 47. De Grove, K. C. et al (2016). Characterization and Quantification of Innate Lymphoid Cell Subsets in Human Lung. Proceeding of the National Academy of Sciences one. 11 (1), e0145961. 48. De Smedt, M. et al (2007). Notch signaling induces cytoplasmic CD3ϵ expression in human differentiating NK cells. Blood. 110 (7). p. 2696-2703. 49. Denney, L. et al (2015). Pulmonary Epithelial Cell-Derived Cytokine TGF-β1 Is a Critical Cofactor for Enhanced Innate Lymphoid Cell Function. Immunity. 43 (5). p. 945-958. 50. Dhakal, M. et al (2014). IL-13Ralpha1 is a surface marker for M2 macrophages influencing their differentiation and function. European Journal of Immunology. 44 (3). p. 842-855. 91  51. Di Stefano, A.B. et al (2010). Survivin is regulated by interleukin-4 in colon cancer stem cells. Journal of Cell Physiology. 255. p. 555-561. 52. Drake, L.Y. & Kita, H. (2014). Group 2 Innate Lymphoid Cells in the Lung. Advances in Immunology. 124. p. 1-16. 53. Dredge, K. et al (2002). Adjuvants and the promotion of Th1-type cytokines in tumour immunotherapy. Cancer Immunology, Immunotherapy. 51. p. 521-531. 54. Duerr, C.U. & Fritz, J.H. (2017). Isolation of Group 2 Innate Lymphoid Cells from Mouse Lungs. Innate Antiviral Immunity. Methods in Molecular Biology. vol. 1656. 55. Duerr, C.U. et al (2015). Type I interferon restricts type 2 immunopathology through the regulation of group 2 innate lymphoid cells. Nature Immunology. 17. p. 67-75. 56. DuPage, M. & Bluestone, J.A. (2016). Harnessing the plasticity of CD4+ T cells to treat immune-mediated disease. Nature Reviews Immunology. 16. p. 149-163. 57. Eberl, G. et al (2015). Innate lymphoid cells. A new paradigm in immunology. Science. 348 (6237). p. 879-887. 58. Ebihara, T. & Taniuchi, I. (2019). Transcription Factors in the Development and Function of Group 2 Innate Lymphoid Cells. International Journal of Molecular Sciences. 20 (6). 1377. 59. Embgenbroich, M. & Burgdorf, S. (2018). Current concepts of antigen cross-presentation. Frontiers in Immunology. 9. 1643.  60. Enomoto, Y. et al (2009). Tissue remodeling induced by hypersecreted epidermal growth factor and amphiregulin in the airway after an acute asthma attack. Journal of Allergy and Clinical Immunology. 124. p. 913-920. 61. Fang, D. & Zhu, J. (2017). Dynamic balance between master transcription factors determines the fates and functions of CD4 T cell and innate lymphoid cell subsets. Journal of Experimental Medicine. 214 (7). p. 1861-1876. 62. Fitzgerald, D.A. (2001). Cell Sorting: An Enriching Experience. The Scientist. 15 (15). p. 28. 63. Fukumoto, J. (2010). Amphiregulin attenuates bleomycin-induced pneumopathy in mice. American Journal of Physiology - Lung Cellular and Molecular Physiology. 298. L131-L138. 64. Furusawa, J. et al (2013). Critical role of p38 and GATA3 in natural helper cell function. The Journal of Immunology. 191 (4). p. 1818-1826. 65. Gabitass, R.F. et al (2011). Elevated myeloid-derived suppressor cells in pancreatic, esophageal and gastric cancer are an independent prognostic factor and are associated with significant elevation of the Th2 cytokine interleukin-13. Cancer Immunology, Immunotherapy. 60. p. 1419-1430. 66. Gao, K. et al (2013). Transgenic expression of IL-33 activates CD8+ T cells and NK cells and inhibits tumor growth and metastasis in mice. Cancer Letters. 335. p. 463-471. 67. Gao, X. et al (2014). Tumoral Expression of IL-33 Inhibits Tumor Growth and Modifies the Tumor Microenvironment through CD8+ T and NK Cells. The Journal of Immunology. 194. p. 438-445. 92  68. Gao, Y. et al (2017). Tumor immunoevasion by the conversion of effector NK cells into type 1 innate lymphoid cells. Nature Immunology. 18. p. 1004-1015. 69. Gasteiger, G. & Rudensky, A.Y. (2014). Interactions between innate and adaptive lymphocytes. Nature Reviews Immunology. 14. p. 631-639. 70. Gause, W.C. et al (2013). Type 2 immunity and wound healing: evolutionary refinement of adaptive immunity by helminths. Nature Reviews Immunology. 13. p. 607-614. 71. Geginat, J. et al (2014). Plasticity of human CD4 T cell subsets. Frontiers in Immunology. 5 (630). p. 1-10. 72. Golebski, K. et al (2019). IL-1β, IL-23, and TGF-β drive plasticity of human ILC2s towards IL-17-producing ILCs in nasal inflammation. Nature Communications. 10 (2162). p. 1-15. 73. Goswami, R. & Kaplan, M.H. (2011). A Brief History of IL-9. The Journal of Immunology. 186 (6). p. 3283-3288. 74. Guermonprez, P. et al (2003). ER-phagosome fusion defines an MHC class I cross-presentation compartment in dendritic cells. Nature. 425. p. 397-402. 75. Halim, T. & Takei, F. (2014). Isolation and characterization of mouse innate lymphoid cells. Current Protocols in Immunology. 106. 3.25.1.  76. Halim, T.Y. et al (2012). Lung Natural Helper Cells Are a Critical Source of Th2 Cell-Type Cytokines in Protease Allergen-Induced Airway Inflammation. Immunity. 36 (3). p. 451-463.  77. Halim, T.Y. et al (2012). Retinoic-Acid-Receptor-Related Orphan Nuclear Receptor Alpha Is Required for Natural Helper Cell Development and Allergic Inflammation. Immunity. 37. p. 463-474. 78. Halim, T.Y. et al (2014). Group 2 Innate Lymphoid Cells Are Critical for the Initiation of Adaptive T Helper 2Cell-Mediated Allergic Lung Inflammation. Immunity. 40. p. 425-435. 79. Hams, E. et al (2014). IL-25 and type 2 innate lymphoid cells induce pulmonary fibrosis. Proceedings of the National Academy of Sciences. 111 (1). p. 367-372. 80. Hanahan, D. & Weinberg, R.A. (2011). Hallmarks of Cancer: The Next Generation. Cell. 144 (5). p. 646-674. 81. Hearn, A. et al (2009). The Specificity of Trimming of MHC Class I-Presented Peptides in the Endoplasmic Reticulum. The Journal of Immunology. 183 (9). p. 5526-5536. 82. Highfill, S.L. et al (2010). Bone marrow myeloid-derived suppressor cells (MDSCs) inhibit graft-versus-host disease (GVHD) via an arginase-1-dependent mechanism that is up-regulated by interleukin-13. Blood. 116. p. 5738-5747. 83. Hiraoka, K. et al (2006). Concurrent infiltration of CD8+ T cells and CD4+ T cells is a favourable prognostic factor in non-small-cell lung carcinoma. British Journal of Cancer. 94. p. 275-280. 84. Hoves, S. et al (2012). A novel role for granzymes in anti-tumor immunity. OncoImmunology. 1 (2). p. 219-221. 93  85. Hoyler, T. et al (2012). The transcription factor GATA-3 controls cell fate and maintenance of type 2 innate lymphoid cells. Immunity. 37. p. 634-648. 86. Huang, Y. & Paul, W.E. (2016). Inflammatory group 2 innate lymphoid cells. International Immunology. 28 (1). p. 23-28. 87. Huang, Y. et al (2015). IL-25-responsive, lineage-negative KLRG1hi cells are multipotential ‘inflammatory’ type 2 innate lymphoid cells. Nature Immunology. 16. p. 161-169. 88. Huang, Y. et al (2018). S1P-dependent interorgan trafficking of group 2 innate lymphoid cells supports host defense. Science. 359 (6371). p. 114-119. 89. Huber, S. et al (2017). CD4+ T Helper Cell Plasticity in Infection, Inflammation, and Autoimmunity. Mediators of Inflammation. 7083153.  90. Hwang, L.N. et al (2006). The In vivo Expansion Rate of Properly Stimulated Transferred CD8+ T Cells Exceeds That of an Aggressively Growing Mouse Tumor. Cancer Research. 66 (2). p. 1132-1138. 91. Ikutani, M. et al (2012). Identification of Innate IL-5–Producing Cells and Their Role in Lung Eosinophil Regulation and Antitumor Immunity. The Journal of Immunology. 188. p. 703-713. 92. Janeway, C.A. et al (2001). Immunobiology: the immune system in health and disease. 5th Ed. Garland Science. New York, USA. 93. Jiao, Y. et al (2016). Type 1 Innate Lymphoid Cell Biology: Lessons Learnt from Natural Killer Cells. Frontiers in Immunology. 7 (426). p. 1-8. 94. Joffre, O.P et al (2012). Cross-presentation by dendritic cells. Nature Reviews Immunology. 12. p. 557-569.  95. Jovanovic, I.P. et al (2014). Interleukin-33/ST2 axis promotes breast cancer growth and metastases by facilitating intratumoral accumulation of immunosuppressive and innate lymphoid cells. International Journal of Cancer. 134 (7). p. 1669-1682. 96. Kabata, H. et al (2013). Thymic stromal lymphopoietin induces corticosteroid resistance in natural helper cells during airway inflammation. Nature Communications. 4. 2675.  97. Kadel, S. et al (2018). A Major Population of Functional KLRG1– ILC2s in Female Lungs Contributes to a Sex Bias in ILC2 Numbers. ImmunoHorizons. 2 (2). p. 74-86.  98. Kidd, P. (2003). Th1/Th2 Balance: The Hypothesis, its Limitations, and Implications for Health and Disease. Alternative Medicine Review. 8 (3). p. 223-246. 99. Kim, B.S. et al (2013). TSLP elicits IL-33-independent innate lymphoid cell responses to promote skin inflammation. Science Translational Medicine. 5. p. 170ra16.  100. Kim, H.S. et al (2016). A novel IL-10-producing innate lymphoid cells (ILC10) in a contact hypersensitivity mouse model. BMB Reports. 49 (5). p. 293-296.  101. Kim, J. et al (2016). Intratumorally Establishing Type 2 Innate Lymphoid Cells Blocks Tumor Growth. The Journal of Immunology. 94  102. Kim, J.Y. et al (2015). Interleukin-33/ST2 axis promotes epithelial cell transformation and breast tumorigenesis via upregulation of COT activity. Oncogene. 34. p. 4928-4938. 103. Kim, M.S. et al (2015). Circulating IL-33 level is associated with the progression of lung cancer. Lung Cancer. 90 (2). p. 346-351.  104. Klein-Wolterink, R.G. et al (2013). Essential, dose-dependent role for the transcription factor Gata3 in the development of IL-5+ and IL-13+ type 2 innate lymphoid cells. Proceedings of the National Academy of Sciences. 110. p. 10240-10245.  105. Klose, C.S. et al (2014). Differentiation of type 1 ILCs from a common progenitor to all helper-like innate lymphoid cell lineages. Cell. 157. p. 340-346.  106. Koh, J. et al (2019). IL23-Producing Human Lung Cancer Cells Promote Tumor Growth via Conversion of Innate Lymphoid Cell 1 (ILC1) into ILC3. Clinical Cancer Research. 25 (13). p. 4026-4035. 107. Konya, V. & Mjösberg, J. (2016). Lipid mediators as regulators of human ILC2 function in allergic diseases. Immunology Letters. 179. p. 36-42.  108. Kovacsoviks-Bankowski, M. & Rock, K.L. (1995). A phagosome-to-cytosol pathway for exogenous antigens presented on MHC class I molecules. Science. 267. p. 243-246. 109. Koyasu, S. et al (2015). Inflammatory ILC2 cells: disguising themselves as progenitors? Nature Immunology. 16. p. 133-134. 110. Krabbendam, L. et al (2018). New insights into the function, development, and plasticity of type 2 innate lymphoid cells. Immunological Reviews. 286 (1). p. 74-85.  111. Kueh, H.Y. et al (2013). Positive Feedback Between PU.1 and the Cell Cycle Controls Myeloid Differentiation. Science. 341 (6146). p. 670-673. 112. Kuhns, D.B. et al (2007). Induction of human monocyte interleukin (IL)-8 by fibrinogen through the toll-like receptor pathway. Inflammation. 30. p. 178-188. 113. Kumar, A. & Ghosh, B. (2008). A single nucleotide polymorphism (A → G) in intron 3 of IFNγ gene is associated with asthma. Genes & Immunity. 9. p. 294-301. 114. Kuperman, D.A. et al (2002). Direct effects of interleukin-13 on epithelial cells cause airway hyperreactivity and mucus overproduction in asthma. Nature Medicine. 8. p. 885-889. 115. Lee, J.S. et al (2012). AHR drives the development of gut ILC22 cells and postnatal lymphoid tissues via pathways dependent on and independent of Notch. Nature Immunology. 13. p. 144-151.  116. Li, P. et al (2017). Coordinated regulation of IFITM1, 2 and 3 genes by an IFN-responsive enhancer through long-range chromatin interactions. Biochimica et Biophysica Acta. Gene Regulatory Mechanisms. 1860 (8). p. 885-893. 117. Li, S. et al (2019). Gene expression signatures of circulating human type 1, 2, and 3 innate lymphoid cells. The Journal of Allergy and Clinical Immunology. 143 (6). p. 2321-2325. 118. Licona-Limón, P. et al (2013). TH2, allergy and group 2 innate lymphoid cells. Nature Immunology. 14. p. 536-542. 95  119. Lim, A. I. et al (2016). IL-12 drives functional plasticity of human group 2 innate lymphoid cells. Journal of Experimental Medicine. 213 (4). p. 569-583. 120. Lim, A. I. et al (2017). Developmental options and functional plasticity of innate lymphoid cells. Current Opinion in Immunology. 44. p. 61-68. 121. Lizée, G. et al (2003). Control of dendritic cell cross-presentation by the major histocompatibility complex class I cytoplasmic domain. Nature Immunology. 4 (11). p. 1065-1073. 122. López-Castejón, G. & Brough, D. (2011). Understanding the mechanism of IL-1β secretion. Cytokine & Growth Factor Reviews. 22 (4). p. 189-195. 123. Lu, B. et al (2016). Interleukin-33 in tumorigenesis, tumor immune evasion, and cancer immunotherapy. Journal of Molecular Medicine. 94. p.535-543. 124. Lu, X. et al (2018). Lipoxin A4 regulates PM2.5-induced severe allergic asthma in mice via the Th1/Th2 balance of group 2 innate lymphoid cells. Journal of thoracic disease, 10 (3). p. 1449–1459.  125. Lyman, M. A. et al (2004). Spontaneously arising pancreatic tumor does not promote the differentiation of naive CD8+ T lymphocytes into effector CTL. The Journal of Immunology. 172. p. 6558–6567. 126. Marr, A.G. (1991). Growth rate of Escherichia coli. Microbiology Reviews. 55 (2). p. 316-333. 127. Martínez-González, I. et al (2015). Lung ILC2s link innate and adaptive responses in allergic inflammation. Trends in Immunology. 36 (3). p. 189-195.  128. Martínez-González, I. et al (2016). Allergen-Experienced Group 2 Innate Lymphoid Cells Acquire Memory-like Properties and Enhance Allergic Lung Inflammation. Immunity. 45 (1). p. 198-208. 129. Martínez-González, I. et al (2017). Immunological Memory of Group 2 Innate Lymphoid Cells. Trends in Immunology. 38 (6). p. 423-431. 130. Martínez-González, I. et al (2018). ILC2 memory: Recollection of previous activation. Immunological Reviews. 283. p. 41-53. 131. Martínez-González, I. et al (no date). Magnetic Enrichment of Mouse ILC2s from the Lung [Poster]. 132. Mathä, L. et al (2019). Female and male mouse lung group 2 innate lymphoid cells differ in gene expression profiles and cytokine production. Public Library of Science ONE. 14(3). e0214286. 133. Mattner, J. & Wirtz, S. (2017). Friend or Foe? The Ambiguous Role of Innate Lymphoid Cells in Cancer Development. Trends in Immunology. 38 (1). p. 29-38. 134. Maywald, R.L. et al (2015). IL-33 activates tumor stroma to promote intestinal polyposis. Proceedings of the National Academy of Sciences. 112 (19). E2487-E2496. 135. McHedlidze, T. et al (2013). Interleukin-33-Dependent Innate Lymphoid Cells Mediate Hepatic Fibrosis. Immunity. 39 (2). p. 357-371.  96  136. Medzhitov, R. & Janeway, C. (2000). Innate Immunity. The New England Journal of Medicine. 343. p. 338-344.  137. Mielke, L.A. et al (2013). TCF-1 Controls ILC2 and NKp46+ RORgt + Innate Lymphocyte Differentiation and Protection in Intestinal Inflammation. The Journal of Immunology. 191. p. 4383-4391. 138. Millipore Sigma-Aldrich (2020). Cell Culture Protocol 5: Subculture of Suspension Cell Lines [Online]. Viewed April 20th 2020. https://www.sigmaaldrich.com/technical-documents/protocols/biology/subculture-of-suspension.html. 139. Mindt, B.C. et al (2018). Group 2 Innate Lymphoid Cells in Pulmonary Immunity and Tissue Homeostasis. Frontiers in Immunology. 9 (840). p. 1-17. 140. Mirchandani, A.S. et al (2014). Type 2 innate lymphoid cells drive CD4+Th2 cell responses. The Journal of Immunology. 192. p. 2442-2448. 141. Mittal, D. et al (2014). New insights into cancer Immunoediting and its three component phases – elimination, equilibrium and escape. Current Opinion in Immunology. 27. p. 16-25. 142. Miyazaki, M. et al (2017). The E-Id Protein Axis Specifies Innate and Adaptive Lymphoid Cell Fate.  Immunity. 46 (5). p. 818-834. 143. Mjösberg, J. et al (2012). The transcription factor GATA3 is essential for the function of human type 2 innate lymphoid cells. Immunity. 37 (4). p. 649-659. 144. Monticelli, L.A. et al (2011). Innate lymphoid cells promote lung tissue homeostasis following acute influenza virus infection. Nature Immunology. 12 (11). p. 1045-1054. 145. Morita, H. et al (2019). Induction of human regulatory innate lymphoid cells from group 2 innate lymphoid cells by retinoic acid. Journal of Allergy and Clinical Immunology.  146. Moro, K. (no date). ILC2 isolation from different mouse tissues. MACS Miltenyi Biotec. 147. Moro, K. et al (2010). Innate production of T(H)2 cytokines by adipose tissue-associated c-Kit(+) Sca-1(+) lymphoid cells. Nature. 463. p. 540-544. 148. Moro, K. et al (2015). Interferon and IL-27 antagonize the function of group 2 innate lymphoid cells and type 2 innate immune responses. Nature Immunology. 17. p. 76-86. 149. Moro, K. et al (2015). Isolation and analysis of group 2 innate lymphoid cells in mice. Nature Protocols. 10 (5). p. 792-806. 150. Motomura, Y. et al (2014). Basophil-derived interleukin-4 controls the function of natural helper cells, a member of ILC2s, in lung inflammation. Immunity. 40. p. 758-771. 151. Mucida, D. et al (2009). Retinoic Acid Can Directly Promote TGF-β-Mediated Foxp3+ Treg Cell Conversion of Naive T Cells. Immunity. 30 (4). p. 471-472. 152. Munn, D.H. & Bronte, V. (2016). Immune suppressive mechanisms in the tumor microenvironment. Current Opinion in Immunology. 39. p. 1-6. 97  153. Nagulapalli, S. & Atchison, M.L. (1998). Transcription Factor Pip Can Enhance DNA Binding by E47, leading to Transcriptional Synergy Involving Multiple Protein Domains. Molecular and Cellular Biology. 18 (8). p. 4639-4650. 154. Nakao, F. et al (2001). Association of IFN-γ and IFN regulatory factor 1 polymorphisms with childhood atopic asthma. Journal of Allergy and Clinical Immunology. 107. p. 499-504. 155. Negulescu, P.A. et al (1996). Polarity of T Cell Shape, Motility, and Sensitivity to Antigen. Immunity. 4 (5). p. 421-430. 156. Neill, D.R. et al (2010). Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature. 464. p. 1367-1370. 157. Ochsenbein, A.F. (2005). Immunological ignorance of solid tumors. Springer Seminars in Immunopathology. 27. p. 19-35. 158. Oh, J.H. et al (2012). Genotype instability during long-term subculture of lymphoblastoid cell lines. Journal of Human Genetics. 58. p. 16-20. 159. Ohne, Y. et al (2016). IL-1 is a critical regulator of group 2 innate lymphoid cell function and plasticity. Nature Immunology. 17. p. 646-655. 160. Okamura, Y. et al (2001). The extra domain A of fibronectin activates Toll-like receptor 4. Journal of Biological Chemistry. 276. p. 10229-10233. 161. Oliphant, C.J. et al (2014). MHCII-Mediated Dialog between Group 2 Innate Lymphoid Cells and CD4+ T Cells Potentiates Type 2 Immunity and Promotes Parasitic Helminth Expulsion. Immunity. 41. p. 283-295. 162. Pelly, V.S. et al (2016). IL-4-producing ILC2s are required for the differentiation of TH2 cells following Heligmosomoides polygyrus infection. Mucosal Immunology. 9. p. 1407-1417. 163. Petersson, K. et al (2002). The pTα promoter and enhancer are direct targets for transactivation by E box‐binding proteins. European Journal of Immunology. 32 (3). p. 911-920. 164. Pinto, L.A. et al (2007). STAT1 gene variations, IgE regulation and atopy. Allergy. 62. p. 1456-1461.  165. Possot, C. et al (2011). Notch signaling is necessary for adult, but not fetal, development of RORγt+ innate lymphoid cells. Nature Immunology. 12. p. 949-958. 166. Radtke, F. et al (1999). Deficient T cell fate specification in mice with an induced inactivation of Notch1. Immunity. 10. p. 547-558. 167. Rak, G.D. et al (2015). IL-33-Dependent Group 2 Innate Lymphoid Cells Promote Cutaneous Wound Healing. Journal of Investigative Dermatology. 136. p. 487-496. 168. Reinicke, A.T. et al (2009). Dendritic Cell Cross-Priming Is Essential for Immune Responses to Listeria monocytogenes. Public Library of Science ONE. 4(10). e7210. 169. Retière, C. et al (1999). The Mechanism of Chromosome 7 Inversion in Human Lymphocytes Expressing Chimeric γβ TCR. The Journal of Immunology. 162. p. 903-910. 98  170. Ricardo-González, R.R. et al (2018). Tissue signals imprint ILC2 identity with anticipatory function. Nature Immunology. 19. p. 1093-1099. 171. Riedel, J. et al (2017). IL-33–Mediated Expansion of Type 2 Innate Lymphoid Cells Protects from Progressive Glomerulosclerosis. Journal of the American Society of Nephrology. 28 (7). p. 2068-2080. 172. Robinette, M.L. et al (2015).  Transcriptional programs define molecular characteristics of innate lymphoid cell classes and subsets. Nature Immunology. 16. p. 306-317. 173. Roediger, B. et al (2013). Cutaneous immunosurveillance and regulation of inflammation by group 2 innate lymphoid cells. Nature Immunology. 14. p. 564-573. 174. Roediger, B. et al (2015). IL-2 is a critical regulator of group 2 innate lymphoid cell function during pulmonary inflammation. Journal of Allergy and Clinical Immunology.  175. Salimi, M. et al (2013).  A role for IL-25 and IL-33-driven type-2 innate lymphoid cells in atopic dermatitis. Journal of Experimental Medicine. 210. p. 2939-2950. 176. Salimi, M. et al (2018). Activated innate lymphoid cell populations accumulate in human tumour tissues. BMC Cancer. 18 (1). p. 341. 177. Saranchova, I. et al (2016). Discovery of a Metastatic Immune Escape Mechanism Initiated by the Loss of Expression of the Tumour Biomarker Interleukin-33. Scientific Reports. 6 (30555). 178. Saranchova, I. et al (2018). Type 2 Innate Lymphocytes Actuate Immunity Against Tumours and Limit Cancer Metastasis. Scientific Reports. 8 (2924). 179. Sato-Takayama, N. et al (2010). IL-7 and IL-15 independently program the differentiation of intestinal CD3-NKp46+ cell subsets from Id2-dependent precursors. Journal of Experimental Medicine. 207. p. 273-280.  180. Seehus, C.R. et al (2017). Alternative activation generates IL-10 producing type 2 innate lymphoid cells. Nature Communications. 8 (1900). p. 1-13. 181. Shirey, K.A. et al (2006). Upregulation of IFN-γ Receptor Expression by Proinflammatory Cytokines Influences IDO Activation in Epithelial Cells. Journal of Interferon and Cytokine Research. 26 (1).  182. Silver, J.S. et al (2016). Inflammatory triggers associated with exacerbations of COPD orchestrate plasticity of group 2 innate lymphoid cells in the lungs. Nature Immunology. 17. p. 626-635. 183. Silver, J.S. et al (2016). Inflammatory triggers associated with exacerbations of COPD orchestrate plasticity of group 2 innate lymphoid cells in the lungs. Nature Immunology. 17. p. 626-635. 184. Simoni, Y. et al (2017). Human Innate Lymphoid Cell Subsets Possess Tissue-Type Based Heterogeneity in Phenotype and Frequency. Immunity. 46 (1). p. 148-161. 185. Spits, H. et al (2013). Innate lymphoid cells – a proposal for uniform nomenclature. Nature Reviews Immunology. 13. p. 145-149. 99  186. Spooner, C.J. et al (2013). Specification of type 2 innate lymphocytes by the transcriptional determinant Gfi1. Nature Immunology. 14. p. 1229-1236. 187. Stehle, C. et al (2015). Putting the brakes on ILC2 cells. Nature Immunology. 17. p. 43-44. 188. Stier, M.T. et al (2018). IL-33 promotes the egress of group 2 innate lymphoid cells from the bone marrow. Journal of Experimental Medicine. 215 (1). p. 263-281. 189. Stumbles, P. A. et al (2004). Cutting edge: tumor-specific CTL are constitutively cross-armed in draining lymph nodes and transiently disseminate to mediate tumor regression following systemic CD40 activation. The Journal of Immunology. 173. p. 5923–5928. 190. Sun, C.M. et al (2007). Small intestine lamina propria dendritic cells promote de novo generation of Foxp3 T reg cells via retinoic acid. Journal of Experimental Medicine. 204 (8). p. 1775-1785. 191. Tang, J. et al (1997). TGF-beta down-regulates stromal IL-7 secretion and inhibits proliferation of human B cell precursors. The Journal of Immunology. 159 (1). p. 117-125. 192. Ting Cai, B.S. et al (2019). IL-17–producing ST21group 2 innate lymphoid cells play a pathogenic role in lung inflammation. Journal of Allergy and Clinical Immunology. 143 (1). p. 229-244. 193. Trabanelli, S. et al (2017). Tumour-derived PGD2 and NKp30-B7H6 engagement drives an immunosuppressive ILC2-MDSC axis. Nature Communications. 8 (1). 593. 194. Trabanelli, S. et al (2019). The pro- and anti-tumor role of ILC2s. Seminars in Immunology. 41. 101276. 195. Trinchieri, G. (1989). Biology of natural killer cells. Advances in Immunology. 47. p. 187-376. 196. Umland, S.P. et al (1997). Interleukin-5 mRNA Stability in Human T Cells Is Regulated Differently than Interleukin-2, Interleukin-3, Interleukin-4, Granulocyte/Macrophage Colony-stimulating Factor, and Interferon- γ. American Journal of Respiratory Cell and Molecular Biology. 18 (5). p. 631-642. 197. Vacca, P. et al (2018). Human natural killer cells and other innate lymphoid cells in cancer: Friends or foes? Immunology Letters. 201. p. 14-19.  198. Verrier, T. et al (2016). Phenotypic and Functional Plasticity of Murine Intestinal NKp46+ Group 3 Innate Lymphoid Cells. The Journal of Immunology. 196 (11). p. 4731-4738. 199. Viant, C. et al (2016). Transforming growth factor–β and Notch ligands act as opposing environmental cues in regulating the plasticity of type 3 innate lymphoid cells. Science Signaling. 9 (426). p. ra46. 200. Vivier, E. et al (2016). The evolution of innate lymphoid cells. Nature Immunology. 17. p. 790-794.  201. Vivier, E. et al (2018). Innate Lymphoid Cells: 10 Years On. Cell. 174 (5). p. 1054-1066. 202. Von Moltke, J. et al (2016). Tuft-cell-derived IL-25 regulates an intestinal ILC2–epithelial response circuit. Nature. 529. p. 221-225. 100  203. Vonarbourg, C. et al (2010). Regulated Expression of Nuclear Receptor RORγt Confers Distinct Functional Fates to NK Cell Receptor-Expressing RORγt+ Innate Lymphocytes. Immunity. 33 (5). p. 736-751. 204. Waldhauer, I. & Steinle, A. (2008). NK Cells and Cancer Immunosurveillance. Oncogene. 27. p. 5932–5943. 205. Walker, J.A. & McKenzie, N.J. (2013). Development and function of group 2 innate lymphoid cells. Current Opinion in Immunology. 25. p. 148-155.  206. Walker, J.A. et al (2013). Innate lymphoid cells – how did we miss them? Nature Reviews Immunology. 13. p. 75-87. 207. Wallrapp, A. et al (2017). The neuropeptide NMU amplifies ILC2-driven allergic lung inflammation. Nature Immunology. 549. p. 351-365. 208. Walzer, T. et al (2007). Identification, activation, and selective in vivo ablation of mouse NK cells via NKp46. Proceedings of the National Academy of Sciences. 104 (9). p. 3384-3389. 209. Wang, L. et al (2020). TGF-β induces ST2 and programs ILC2 development. Nature Communications. 11 (35). p. 1-15. 210. Wang, Y.H. et al (2010). A novel subset of CD4(+) T(H)2 memory/effector cells that produce inflammatory IL-17 cytokine and promote the exacerbation of chronic allergic asthma. Journal of Experimental Medicine. 207. p. 2479-2491. 211. Wong, S.H. et al (2012). Transcription factor RORα is critical for nuocyte development. Nature Immunology. 13. p. 229-236. 212. Wu, D. et al (2011). Eosinophils sustain adipose alter-natively activated macrophages associated with glucose homeostasis. Science. 332. p. 243-247. 213. Wu, Y. et al (2017). Enhanced circulating ILC2s and MDSCs may contribute to ensure maintenance of Th2 predominant in patients with lung cancer. Molecular Medicine Reports. 15 (6). p. 4374-4381. 214. Yagi, R. et al (2014). The Transcription Factor GATA3 Is Critical for the Development of All IL-7Ra-ExpressingInnate Lymphoid Cells. Immunity. 40. p. 378-388.  215. Yang, D. et al (2017). Alarmins and Immunity. Immunological Reviews. 280 (1). p. 41-56. 216. Yang, G. et al. (2006). IFITM1 plays an essential role in the antiproliferative action of interferon-γ. Oncogene. 26. p. 594-603. 217. Yang, Q. et al (2011). IL‐33 synergizes with TCR and IL‐12 signaling to promote the effector function of CD8+ T cells. European Journal of Immunology. 41. p. 3351-3360. 218. Yewdell, J.W. (1999). Mechanisms of exogenous antigen presentation by MHC class I molecules in vitro and in vivo: implications for generating CD8+ T-cell responses to infectious agents, tumours, transplants and vaccines. Advances in Immunology. 73. p. 1-77. 219. Zaiss, D.M.W. et al (2013). Amphiregulin Enhances Regulatory T Cell-Suppressive Function via the Epidermal Growth Factor Receptor. Immunity. 38 (2). p. 275-284. 101  220. Zamai, L. et al (2007). NK Cells and Cancer. The Journal of Immunology. 178 (7) p. 4011-4016. 221. Zhang, K. et al (2017). Cutting Edge: Notch Signaling Promotes the Plasticity of Group-2 Innate Lymphoid Cells. The Journal of Immunology. 198. p. 1798-1803. 222. Zhou, L. et al (2009). Plasticity of CD4+ T Cell Lineage Differentiation. Immunity. 30 (5). p. 646-655. 223. Zhou, R. et al (2013). Interleukin-13 and its receptors in colorectal cancer. Biomedical Reports. 1. p. 687-690. 224. Zhu, J. (2015). T helper 2 (Th2) cell differentiation, type 2 innate lymphoid cell (ILC2) development and regulation of interleukin-4 (IL-4) and IL-13 production. Cytokine. 75 (1). p. 14-24. 225. Zook, E.C. & Kee, B.L. (2016). Development of innate lymphoid cells. Nature Immunology. 17. p. 775-782.   102  Appendix   Figure S1. Heat map comparing the most differentially expressed genes for every cluster in each sample of pILC2s. This heat map allows for a broad overview and a comparison between activated and non-activated samples and through time. Genes of interest are, for example, IL-9, which seems to be only upregulated in a week 2 non-activated sample of pILC2s. A week later, non-activated pILC2s express high amounts of CD3 and TCR-related genes. Class II MHC genes (bottom) might be more present in activated samples.           103      Figure S2. Heat map comparing most differentially expressed genes by pathway in pILC2s. Activated populations seem to express the most notable differences with their non-activated counterparts when considering genes associated to Th1 and Th2 responses. This indicates the loss of a quiescent state. Although some genes are expressed in every sample, differences are best observed on a gene-by-gene basis. Interesting examples are IL-10, TNF, Bcl3, Il1rl1, IL-4 or IL-6 or the IL-18R.            


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items