Innate lymphoid cells (ILCs) are among the most recently identified populations of cells that contribute to immune responses. ILCs lack the antigen-specific receptors that characterize adaptive T and B lymphocytes but are exquisitely attuned to their local tissue environment and can respond rapidly to tissue disturbances and pathogenic threats. Chronic activation of ILCs is linked to various immunopathologies, which has prompted the need for fuller understanding of these cells. In this issue of Nature Immunology, we present a collection of commissioned articles that review the current understanding of the development and evolution of ILCs and their roles in mediating immune responses and contributions to pathogenesis and present frank discussions of many unanswered questions that preclude the manipulation of ILCs for therapeutic intervention (http://www.nature.com/ni/focus/ILC).

Several subgroups of ILCs have been identified that are best characterized by their effector functions, which include cytokine secretion or cytotoxic killer potential; this has prompted the description of ILCs as the innate cell equivalents of helper and cytotoxic T lymphocytes. These subgroups include the interferon-γ (IFN-γ)-expressing group 1 ILCs, which include natural killer (NK) cells; the interleukin 5 (IL-5)- and IL-13-expressing ILC2s; and the heterogeneous group 3 ILCs, which include lymphoid-tissue-inducer (LTi) cells and IL-17- and IL-22-expressing cells. With the exception of circulating NK cells, most ILCs are sessile tissue-resident cells that seem to replicate in situ as conditions warrant.

Zook and Kee describe the developmental paths from progenitor cells to functionally mature cell subsets. ILCs descend from hematopoietic common lymphoid progenitors in the bone marrow; however, full ILC maturation might occur as the immature ILCs take up residence in peripheral tissues. A hierarchy of transcription factors including NFIL3, ETS1 and Id2 regulates ILC development as cells progressively lose multi-lineage potential. The mature subsets are distinguished by the signature transcription factors T-bet, for ILC1s, GATA-3, for ILC2 cells, and RORγt, for ILC3 cells; these act to maintain cell identity and function.

Vivier et al. use comparative transcriptomics, leveraging gene-expression profiles across many vertebrate species from jawless lamprey to mammals, to trace the evolution of NK cells and ILCs. Surprisingly, adaptive lymphocytes, NK cells and ILC2 cells seem to be more 'ancient', arising some 500 million years ago, than the ILC1 and ILC3 subsets. LTi cells also are a relatively late arrival, emerging approximately 100 million years ago, coincident with the appearance of lymphotoxin, whose signaling is essential for the LTi-cell-mediated formation of peripheral lymph nodes. In contrast, development of the spleen and other primordial lymphoid tissues is independent of lymphotoxin, which might explain the earlier emergence of adaptive lymphocytes.

Bando and Colonna review the various mouse genetic models that have given insight to how ILCs contribute to host immunity. Many of these studies, however, used RAG-deficient mice to avoid interpretative complications of functional redundancy. Conditional models are needed for probing ILC function in lymphocyte-replete hosts. Nevertheless, ILCs have non-redundant tissue-resident roles, including the elaboration of IFN-γ by uterine ILC1 cells for the vascular remodeling that accompanies placentation to support pregnancies, and the production of IL-5 and IL-13 by ILC2 cells for the regulation of adipocyte metabolism and intestinal tuft cells, respectively.

Although ILCs reside in most tissues, they serve prominent roles in barrier tissues such as the skin, intestine and lungs. ILCs respond to tissue damage mediated by microbial infection much faster than T lymphocytes do. As reviewed by Klose and Artis, ILCs are able to rapidly secrete cytokines after being activated by soluble mediators released by stromal cells. In particular, ILC2 and ILC3 cells express receptors for various inflammatory mediators and cellular alarmins. These cells integrate multiple simultaneous signals that arise within the tissue, and by their elaboration of mediators, they can act as immunological 'first responders' to repel parasites and contain bacterial and fungal infection. Their actions seem to condition and regulate the ensuing adaptive immune response. ILC2 and ILC3 cells also promote the tissue-repair process for the return to tissue homeostasis. However, ILC2s are also linked to various disease states, such as asthma and tissue fibrosis, and ILC3s are associated with inflammatory bowel disease.

Spits, Bernink and Lanier review the development and function of group 1 ILCs. This group of IFN-γ-producing ILCs mediates protection against intracellular pathogens and tumors. Whereas NK cells exhibit potent cytotoxicity against infected cells or transformed cells and were discovered much earlier than ILC1 cells, phenotypically distinguishing these two cell types in non–fate-reporter mice is difficult. This conundrum prompts re-examination of much of the earlier literature on the development and function of NK cells. Further complicating the analysis of ILC1s is the fact that this population shows considerable heterogeneity and plasticity.After exposure to IL-12, IL-18 or IL-1β, ILC2 and ILC3 cells can transdifferentiate to become phenotypically ILC1s. These ILC1 populations have been associated with inflammatory autoimmune diseases such as chronic obstructive pulmonary disease.

The Comment by Eberl and colleagues shows that manipulating ILCs for therapeutic use would be clinically advantageous, given the manifold roles of ILCs. How to specifically target ILCs remains a challenge, as does attempting to induce beneficial activation of the immune system without eliciting immunopathology. Clearly, much more insight into the activation and regulation of ILCs is needed.

The Focus also includes an online animation on ILC biology in the gut mucosa (http://www.nature.com/ni/multimedia/ilc/index.html). The Focus is provided free to interested readers via sponsorship from MedImmune, for which we are grateful.