The distinct features of ILCs and T cells allow for complementarity and redundancy between these innate and adaptive immune systems. Whereas T cells are activated through MHC-peptide-TCR interactions and co-stimulatory signals, ILCs characteristically lack expression of rearranged antigen receptors. Instead, these cells are primed by surrounding cytokines, hormones, and lipid mediators and may additionally be susceptible to environmental stimuli [
30,
45,
107,
108]. In temporal space, the differing modes of activation allow ILCs to act as first responders, with T cells picking up the pace following activation by antigen-presenting cells and clonal expansion. The ability of ILCs to act as first responders is further compounded by the observation that many tissues, at least in the mouse, harbor resident ILC populations [
13], preceding the necessity for these cells to migrate prior to eliciting an immune response. In contrast, the majority of T cells must first acquire expression of homing receptors and migrate from secondary lymphoid organs to the effector site. On a situational level, the complementary localization of these two immune subsets also translates to the capacity for ILCs to respond rapidly and robustly at local sites, whereas T cells may respond both locally and systemically. Lastly, whereas T cells have evolved to respond towards attacks on the immune system, recent publications highlight the capacity for ILCs to additionally respond to more subtle alterations in immune homeostasis. For example, constitutive IL-5 secretion by ILC2 was demonstrated to be influenced by circadian rhythms and food intake [
109], whereas IL-22 production by ILC3 is regulated by AHR, a transcription factor responsive to both xenobiotics and organic compounds [
45,
110]. Furthermore, lL-13 production by ILC2 was shown to promote tuft cell development [
111], and IL-22 expression by ILC3 is known to induce proliferation and survival of epithelial cells [
112]. Therefore, it seems reasonable to argue that ILCs and T cells are orchestrated to act in harmony with one other, complementing one another in spatial, temporal, and functional aspects [
113]. This has recently been the focus for intensive research, primarily in mice, but also in humans, which we will summarize below.
Lessons from immune-compromised and immune-competent mice and humans
In 2015, a study by Song et al. addressed the contribution of ILC and T cells to anti-CD40-mediated colitis, making effective use of engineered mice harboring severely reduced numbers of NKp46
+ ILC3, all ILC3, T cells or both ILC3 and T cells. Interestingly, anti-CD40 colitic mice lacking both T cells and ILC3 displayed milder histopathology than mice lacking T cells only, where ILC3 production of GM-CSF was shown to be required for recruitment of pro-inflammatory monocytes to the site of inflammation [
114]. More recently, Brasseit et al. have expanded on these findings by demonstrating that mice which lack ILC3 display milder colitis. Therefore, ILC appear to contribute uniquely to the innate immune response in the anti-CD40-mediated colitis model [
115].
In turn, an in-depth study of ILC and T cells in the CD4 T cell transfer colitis model, in which the adaptive immune compartment contributes significantly, has shed a more nuanced light on the contribution of innate and adaptive immune cells to colitis development. Transfer of colitogenic CD4 T cells to
Rag1−/− mice depleted of ILC highlighted that CD4
+ T cells, but not ILCs, are critical for induction of colitis [
115]. Nonetheless, absence of ILC3 exacerbated histopathological signs of colitis, arguing for a non-redundant and time-specific function of ILC3 in the CD4 T cell transfer colitis model [
115].
A notable limitation of these and other studies is the examination of ILC function in the context of immune-compromised mice. However, a number of recent publications have successfully dissected the individual contributions of ILC and T cells in specific immune settings. In a model of
C. rodentium infection developed to examine the contribution of ILC to bacterial infection, ILC were shown to exacerbate pathogenesis of
C. rodentium-mediated infection in mice lacking T cells. Specifically, lack of non-NKp46
+ ILC3, but not NKp46
+ ILC3, accelerated body weight loss and mortality of mice as compared with mice lacking T cells only [
114]. These findings were subsequently corroborated by a second study wherein deletion of key ILC3 genes confirmed IL-22 production by NKp46
+ ILC3 to be redundant for the control of
C. rodentium infection in the presence of T cells [
116]. Importantly, however, NKp46
+ ILC3 were shown to be essential for cecal homeostasis, where
C. rodentium-infected mice lacking NKp46
+ ILC3 presented with a decreased cecum size and histopathological signs of hyperplasia and inflammation.
Another intriguing recent publication has additionally shed new light on the complementarity and redundancy of ILCs and T cells in maintaining the delicate balance between bacterial control and gut homeostasis. Studying intestinal epithelial cells (IEC) and ILC3 activation through pSTAT3 phosphorylation, Mao et al. observed microbiota-dependent pSTAT3 signaling in ILC3 and IEC in mice lacking T cells but not WT mice [
101]. Subsequently, a more detailed analysis showed that neonatal mice had neither pSTAT3
+ ILC3 nor IEC, and appearance of pSTAT3
+ cells was linked to weaning of mice. Furthermore, as the adaptive immune system evolved, ILC3 in WT mice lost their activated state, whereas activation of ILC3 in adult mice lacking T cells persisted. In this, Tregs were shown to prevent ILC3 activation through suppression of IL-23 production from CCR2
+ myeloid cells, whereas Th
17 cells decreased microbial burden and as such indirectly inhibited ILC3 activation. In addition, activated ILC3 and T cells differentially regulated segmented filamentous bacteria (SFB) in the small intestine, where ILC3 prohibited the development of SFB into long filamentous forms and T cells prevented attachment of SFB to IEC. Thus, providing evidence that ILCs carry out complementary and non-redundant functions in the intestine of young and adult mice [
101].
In humans, little research has been conducted examining ILC redundancy. However, one recent study by Vely et al. showed that severe combined immunodeficiency (SCID) patients with mutations in the genes
IL2RG and
JAK3 are deficient in circulating helper ILCs and NK cells [
113]. This can be explained by the requirement for IL-7 and IL-15 signaling in the survival of these cells, where IL-7 and IL-15 signals are integrated by the common γc cytokine receptor and the downstream JAK3 tyrosine kinase. Interestingly, ILCs and NK cells are not properly reconstituted after bone marrow transplantation, although their T and B cell pools are replenished [
117]. Alongside, the authors could detect little or no tissue-resident CD3
− NKp46
+ or CD3
− CD11b
− ICOS
+ ILC as examined through staining of skin and gut paraffin-embedded and frozen tissue sections. A subsequent study of the long-term medical history of these
IL2RG- and
JAK3-deficient SCID patients demonstrated no significant increased risk for a number of major medical afflictions, such as HPV, respiratory infections, or disease as compared with control patients. Taken together, this data argues for some level of ILC redundancy in humans. Nonetheless, as staining of skin and gut tissue sections could not exclude the presence of all tissue-resident ILCs, and both patient cohort size and the number of medical conditions examined were limited, caution should be taken when drawing conclusions. This is particularly true when considering the large number of studies that have shown that ILCs are major sources of effector cytokines in human disease [
9].
Overall, it seems reasonable to conclude that most ILC populations carry out unique functions, whereby the impact of ILC function should be viewed in the context of the stage of immune development, immune competency, and localization of the immune response. The existing literature also demonstrates the need for continued studies of human subjects, to ultimately determine the complementarity and redundancy of human ILCs.