Richard M. Locksley | Sandler Distinguished Professor of Medicine, UCSF and HHMI, San Francisco, USA

Immune cells infiltrate all organs of the body. Traditionally identified by roles in host defense, tissue immune cells are now known to contribute to homeostasis of stem cell niches in organs throughout life. Positioned during embryogenesis, innate immune cells derived from yolk sac and fetal liver are gradually replaced by cells derived from adult bone marrow, creating cellular layers through ontogeny that interact to respond to local perturbations ranging from circadian to metabolic to pathologic events. Helper lymphocytes elaborate cytokines using canonical modules that orchestrate interactions among lymphoid, myeloid and non-hematopoietic tissue cells. Although lacking antigen receptors, innate helper lymphocytes, or ILCs, are abundant in small intestine, where allergy-associated ILC2s contribute to small intestine physiology by altering epithelial cell fate after feeding using a pathway hijacked by intestinal helminths. Uncovering immune pathways involved in organ homeostasis may lead to understanding diseases of increasing prevalence, such as food allergy, and might be exploited to enhance healthspan.

The foundations of immunology were advanced by recognition of the fatal infectious complications accompanying genetic loss of rearranging T and B cell antigen receptors that pioneered bone marrow, hematopoietic stem cell, and gene replacement therapies that are covered by others in this Symposium. Successes drove the focus of immunology towards understanding the processes of adaptive immunity that generate the rearranged receptors necessary to protect from environmental pathogens, and that remain the goal of successful vaccines. A prescient contribution by Charlie Janeway from a 1989 Cold Spring Harbor Symposium addressed immunologists’ ‘dirty little secret’ that successful immunization required administration of the antigen with an ‘adjuvant’, usually containing microbial products from mycobacteria or pertussis with alum (1). Janeway pointed out that randomly generated receptors could not alone discriminate ‘non-self’ from ‘self’ antigens and reasoned that evolutionarily encoded receptors that recognized conserved molecules from infectious organisms likely occur at the time of antigen receptor engagement, thus guiding the T cell towards meaningful commitment and avoidance of autoimmunity. Janeway’s insights accounted for the adjuvant requirement and led to discovery of what was later designated ‘signal 2’ required for successful T cell signaling. Later studies from his lab and others led to the discovery of Toll-like receptors, a family of leucine-rich repeat proteins that decorate macrophages, dendritic cells and B cells, which constitute the major cells that present peptide antigens to helper T cells. Toll-like receptors (TLRs), deeply rooted in evolution and named for Drosophila Toll involved in dorsal-ventral patterning and (shown later) host defense, sparked a frenzy of research, resulting in not only Nobel Prizes but also discovery of families of Pattern Recognition Receptors (PRRs) arrayed across the surface, cytosol and endosomal vacuoles of human cells, including 10 TLRs, ~15 C-type lectin, ~15 nucleotide oligomerization domain (NOD)-like and ~15 RIG-I-like and AIM2-like nucleic acid receptors, that engage the spectrum of constituents from bacteria, fungi and viruses, and lead to elaboration of signal 2 required for optimal T cell activation. Modifying mRNA to bypass PRR engagement was a critical engineering feat in developing RNA vaccines against SARS-CoV-2 as recognized by the 2021 Lasker Award to Kariko and Weissman (2).

Seminal studies by Mossman and Coffman in the late 1980s from DNAX Institute at Schering-Plough identified subsets of T helper cells, designated type 1 and type 2, as distinguished by the groups of cytokines they produced (3). T helper 1 (Th1) cells secreted interferon-gamma (IFNg), involved in classical activation of myeloid cells for host defense, whereas T helper 2 (Th2) cells secreted IL-4, IL-5 and IL-13 involved in allergic immunity, including the alternative activation of macrophages, recruitment of eosinophils, and B cell switching to IgE antibodies that arm mast cells and basophils. Two decades later, Cua and Kastelein from DNAX pioneered studies leading to discovery of the third T cell subset and further work quickly elucidated the factors important for establishing and maintaining Th17 cells in mice and humans (4). Th17 cells produce IL-17 cytokines, which mediate neutrophil accumulation associated with clearance of cell debris and microbes during acute injury, and IL-22, which induces epithelial cell proliferation and secretion of antimicrobial peptides.

The discovery of subsets of T helper cells pushed identification of the transcriptional programs that establish the three effector cytokine modules, eventually resulting in the epigenetic model of T helper cell differentiation by which naïve CD4 helper T cells are instructed by dendritic cells and PRR-elicited signals to mature into effector T cells that migrate to tissues to elaborate their respective cytokines. Further study revealed that not only adaptive CD4 T cells selected by peptides embedded in classical MHC molecules, but also unconventional CD4 T cells, including NKT cells, MAIT cells and gd T cells expressing receptors selected on noncanonical MHC molecules, could be grouped according to the three helper cell subsets (5). Finally, a group of innate lymphoid cells, later designated ILCs (6), which lack expression of antigen receptors altogether, was recognized over years of study, culminating with the discovery of ‘allergic-like’ ILC2s in 2010, to complete the three canonical groups based on core transcription factor dependence and cytokine outputs that mirrored those of adaptive and unconventional T cell subsets. Thus, three stereotyped outputs characterize effector mechanisms by which helper lymphocytes confront perturbations in body tissues (here, we won’t consider regulatory and follicular T cells, which mainly communicate with other lymphocytes). The lack of antigen receptors on ILCs creates the opportunity to uncover tissue signals that activate these pathways and align with the demands of basal homeostasis (7).

Outputs from tissue resident helper lymphocytes organize acute response to injury (type 3 associated with IL-17/IL-22-associated immunity) and recall (type 1 associated with IFNg and TNFa), and can be conceptualized respectively as acute neutrophilic responses, whether sterile or infectious, and memory, by which host tissues respond more quickly to a second challenge as designated by ‘trained’ immunity (8). Type 3 and type 1 responses are initiated by PRRs arrayed on multiple cell types, including at epithelial barriers, which induce inflammatory cytokines to activate these distinct immune modules in innate and adaptive immune cells. The third output, associated with allergy, is less obviously associated with host health, and is proposed to organize aversive responses (e.g.; itch, cough, vomiting, diarrhea, etc.) associated with noxious environmental insults but also epithelial adaptations (e.g.; increased mucus) (9). Type 2 responses are initiated by a group of cytokines designated ‘alarmins’, that are expressed from distinct types of sentinel cells in response to perturbations of homeostasis. Receptors for alarmin cytokines are expressed constitutively on tissue resident ILC2s, positioning these cells as integrated sensors of tissue disruption. Multiple functional and GWAS studies have implicated alarmin cytokines like IL-33, TSLP, IL-25 and IL-18 or their receptors in allergic pathology, and therapeutics targeting alarmins are active against allergic diseases.

Innate lymphoid cells emerge from fetal liver around E14 in the mouse and at comparable periods in human fetal development around the time of villus initiation in the small intestine. Like macrophages, ILC2s expand in overlapping waves during fetal, postnatal and adult life, enter tissues, proliferate and activate tissue-specific transcriptomes before adopting a predominantly tissue-resident state (10). Largely maintained by locally deposited and self-renewing precursors, turnover from adult-derived ILC2 precursors occurs with variable kinetics that generally reflect epithelial turnover in the tissue of residence, with more rapid turnover in intestine and skin and slower turnover in adipose and lung. Upon perturbations that stimulate continued activation, ILC2s proliferate and enter the blood, resulting in circulating ILC2s and cytokines that mediate effects at distal tissues (11). The potent cytokine potential of ILC2s has implicated these cells in pathologic allergic states while increasing studies have begun to implicate these cells in tissue homeostasis (12). Numbers of tissue ILC2s are relatively unaffected in germfree mice and fetal human small intestine contains IL-13+ ILC2s, suggesting that tissue residency and cytokine profiles are independent of the microbiota.

Our interest in the role of ILC2s in small intestine physiology began with investigations of eosinophil circadian variation, which was driven by metabolic rather than circadian cues (13). Activated ILC2s produce the eosinophil survival factor, IL-5, and IL-13, which induces release of eosinophilic chemotactic factors by stromal cells, suggesting that activation of intestinal ILC2s with feeding underpinned the oscillations by which blood eosinophils diminish with feeding and increase with fasting. Indeed, fasting was associated with diminished IL-5 production by lamina propria ILC2s, which increased and was accompanied by IL-13 production with feeding, driven in part by induction of the neuropeptide VIP. Thus, feeding induced small intestinal ILC2 activation, production of IL-5 and IL-13, and entry of blood eosinophils into gut tissue, consistent with the decrease in circulating eosinophils in response to nutrient intake. Intestinal eosinophils may be important in tissue remodeling in response to luminal perturbations through release of proteases and growth factors, and further research is needed to understand the basic biology of these cells and their involvement in normal gut physiology.

Although ILC2s integrate multiple synergistic signals from tissues to activate and produce cytokines, the alarmin cytokines IL-33, IL-25 and TSLP constitute key contributors; in the absence of all 3 signaling pathways, activation of both innate ILC2s and adaptive Th2 cells is greatly attenuated in most tissues (14). Single-cell sequencing studies revealed high expression of the IL-25 receptor on small intestine lamina propria ILC2s (15), and epithelial tuft cells were subsequently identified as the unexpected source of IL-25 (16,17). Tuft cells and goblet cells expand remarkably after challenge with parasitic helminths, which elicits proliferation and activation of small intestinal ILC2s (Fig. 1). Recognized over 60 years ago by their blunt, long apical microvilli extending into the hollow lumen, tuft cells are rare chemosensory cells in most mucosal epithelia of vertebrate organs, including upper and lower respiratory tract, gastrointestinal and parts of the genitourinary tract, and within endodermal-derived medullary epithelial cells of the thymus. Intestinal tuft cells are post-mitotic cells derived from columnar crypt stem cells and typically turnover during the normal 3-5 days of villus transit from crypt to apex. Tuft cells in other tissues are long-lived, reflecting slower epithelial turnover in organs like lung or gallbladder. Tuft cells use taste-associated signaling from upstream GPCRs via the Ca⁺⁺-activated cation channel TRPM5 to depolarize and activate canonical effector outputs, including IL-25, eicosanoids including leukotrienes and prostaglandin D2, ATP and acetylcholine, reflecting unique outputs among epithelial cells (18).

Although historically given distinct names in different tissues, such as tuft cells in gut, brush cells in trachea, and microvillus cells in the oronasal epithelia, all tuft cells depend on expression of the transcription factor Pou2F3, suggesting that epithelial cells that depend on Pou2F3 and express typical taste cell signaling pathways and these canonical outputs can all be considered members of the tuft cell family, a designation that includes type 2 taste cells which sense umami, sweet and bitter (19). Expression of various taste and vomeronasal receptors, as well as multiple other GPCRs, suggests that tuft cells are chemosensory cells arrayed to detect luminal signals and transfer information to host cells, which in small intestine serves to change the epithelial boundary by altering cell fate among transit amplifying cells.

Constitutive expression of IL-25 in tuft cells suggests that ILC2 activation initiates a feed-forward circuit to increase tuft and goblet cell numbers until the luminal signal becomes attenuated. ILC2 activation is driven both by IL-25 from the increased numbers of tuft cells but also by release of cysteinyl leukotrienes from tuft cells that synergize to drive proliferation and cytokine production. The resultant epithelial alterations account for the increased goblet cell and mucus response that accompany intestinal helminth infection. These effects were genetically traced to IL-13 generated by ILC2s downstream of IL-25 in tuft cells and upstream of direct effects of IL-13 on epithelia, as shown the ability of exogenous IL-13 to increase tuft cell numbers in organoids that was dependent on epithelial expression of IL-4Ra, the signaling component of the IL-13 receptor. Deleting the ubiquitin-modifying enzyme A20 (TNFAIP3), a negative regulator of IL-25 signal transduction, in ILC2s caused spontaneous IL-25-mediated small intestinal adaptation in mice characterized by lengthening and increased muscle mass with elevated numbers of ILC2s and tuft cells, thus attesting to the constitutive activity of this pathway in situ (20). These small intestinal adaptations were stable over time and associated with persistent alterations in tuft cell and ILC2 numbers. The affected mice were resistant to subsequent helminth infection explaining an observation termed ‘concomitant’ immunity by which intestinal helminth infection establishes a resistant state that impedes maturation of new eggs or larvae from the same or even different helminths. Intriguingly, resistance to injury at distal mucosal sites, such as lung and conjunctivae, also occurs, reflecting the circulation in blood of ILC2s and their cytokines that accompany proliferation and egress from perturbed tissues (11,21).

The appearance of mice with increased numbers of tuft cells in several academic research facilities led to identification of infection by unsuspected cecal Tritrichomonas muris, a parabasalid protist widespread in feral animals, that, like helminths, induces the small intestinal tuft cell-ILC2 circuit (22). T. muris are obligate anaerobes with an array of enzymes that degrade complex plant polysaccharides; related Tritrichomonads constitute the cellulose-degrading organisms found in various species of termites. Rather than mitochondria, protists use hydrogenosomes, which lack the electron transport chain, to generate ATP via decarboxylation and oxidation of pyruvate resulting in production of the metabolites acetate and succinate. Small intestine tuft cells prominently express GPR91, the succinate receptor, and succinate was sufficient to induce tuft cell IL-25, ILC2 activation and initiation of circuit amplification that was lost after deletion of tuft cells, TRPM5 or IL-25, establishing a role for small intestine tuft cells in luminal succinate sensing (20). Additional GPCRs are being discovered that define the chemosensory spectra of not only intestinal tuft cells, but also tuft cells in other organs that express diverse GPCRs that establish the repertoire by which these epithelial sensors deconvolute their microenvironment.

Helminth parasites and protists, widespread in the animal kingdom, likely evolved to induce the adapted intestinal state to facilitate generation of eggs and larvae while avoiding immune attack and generating a host niche resistant to further colonization. While intestinal parasitism has decreased, the rising prevalence of diseases like food allergy in Westernized cultures is ascribed to the ‘hygiene hypothesis’, which proposes aberrant immune deviation driven by exposure to microbiota and nutrients that deviate from developmental programs initiated during embryogenesis and early perinatal life as established earlier in evolutionary history (23). Food allergy, an immune attack on ingested nutrients resulting in symptoms ranging from bloating and hives to life-threatening anaphylaxis, impacts almost 10% of children in the United States at great economic cost. Animal models and human studies implicate type 2 immunity in food allergy, often driven by IgE-mediated recognition of harmless ingredients shared by common food groups. The rising prevalence of food allergy raises the possibility that this pathologic response reflects dysregulation of a physiologic role for type 2 immunity in monitoring food quality, perhaps driven by metabolites or unnatural chemical constituents in modern processed diets (24). In this model, positive food qualities are sensed by enterocytes and enteroendocrine cells whereas negative food qualities are sensed by chemosensory cells like tuft cells and serotonin- and histamine-secreting enterochromaffin cells; both become ‘coded’ and reinforced by neural circuitry. Whereas some ingestants can be directly sensed by genetically encoded receptors like GPR91 for succinate or bitter receptors on type 2 taste cells, akin to PRRs in type 1 and type 3 immunity, others become ‘associated’ with negative outcomes during ingestion and ‘tagged’ through induction of adaptive type 2 immunity, leading to IgE and the arming of mucosal mast cells that degranulate and mediate aversive responses upon subsequent exposure. Thresholds for detection of high-quality nutrients or noxious ingestants likely differ, with higher sensitivity for the latter necessary to protect the host from harmful entities hidden in food. As such, the ILC2-tuft cell feed-forward circuit would represent a food-triggered amplification system enacted to increase detection of potentially toxic constituents once the threshold for ingestion has occurred. Intestinal pathobionts have hi-jacked the process to drive intestinal adaptation and enhance host resilience in support of their reproductive niche. Indeed, intestinal parasitic infection promoted colonization resistance from bacterial pathogens and attenuated inflammatory bowel disease pathology (25,26).

An additional component of the small intestinal feeding response consists of anticipatory propagation of information via hormone and neural circuitry to the distal small intestine to coordinate the response to food intake. A second ILC2 population in the small intestine resides in the muscularis mucosa embedded within the dense network of nerves and macrophages constituting the enteric nervous system. Expression of the IL-13 receptor on populations of enteric neurons (27) and macrophages, which communicate to regulate intestinal physiology, raises the possibility that the second ILC2 population plays a role in propagation of information to the enteric nervous system. In contrast to lamina propria ILC2s, which express the IL-25 receptor, muscularis ILC2s express the IL-33 receptor and produce IL-5 and IL-13 in response to this alarmin cytokine. Intestinal IL-33 is expressed in multiple cell types near the crypt base, including fibroblastic reticular cells and lymphatic endothelial cells in proximity to the enteric plexus. While we continue to refine this area of research, our working hypothesis suggests a three-part process by which luminal information regarding food quality is transferred to ILC2s and crypt transit-amplifying cells to ensure optimal detection and handling of both positive- and negative-quality ingestants by the small intestine, revealing an intrinsic system commandeered by luminal helminths to enable parasitism (Fig. 2).

In summary, an unusual subset of innate lymphocytes that produce cytokines typically associated with allergic diseases populates the small intestine in the lamina propria and muscularis during development and becomes activated to secrete cytokines after ingestion of food. Through a pathway that involves epithelial nutrient detection and relay through lamina propria support cells, intestinal ILC2s are activated to mediate a secretory cell bias among transit amplifying cells, thus increasing goblet and tuft cell production by a forward-amplifying circuit. Increased mucus and enhanced surface area and peristalsis impact gut physiology to optimize nutrient extraction while also increasing the detection capacity for ingested toxins and irritants. Further understanding the links between innate immune cells, stem cell outputs and small intestinal physiology promises to reveal insights into intestinal health and disease. Innate tissue-resident immune cells, including macrophages and ILCs, interact with stem cell niches and affect physiology in many organs (28), and further research defining these pathways are likely to provide substantial opportunities for improving human health.

Acknowledgements

The author thanks lab members and colleagues for discussion, Chang Liao (and BioRender) and Satoshi Koga for help with Fig. 2, and funding from HHMI, the National Institutes of Health and the Sandler Asthma Basic Research Center at UCSF.

References

1. Janeway, Jr., CA. 1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harbor Symp Quant Biol 54:1-13.
2. Kariko K, Buckstein M, Ni H, Weissman D. 2005. Suppression of RNA recognition by Toll-like receptors: the impact of nucleotide modification and the evolutionary origin of RNA. Immunity 23:165-75.
3. Mosmann TR, Coffman RL. 1989. Th1 and Th2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 7:145-73.
4. Weaver CT, Hatton RD, Mangan PR, Harrington LE. 2007. IL-17 family cytokines and the expanding diversity of effector T cell lineages. Annu Rev Immunol 25:821-52.
5. Mayassi T, Barreiro LB, Rossjohn J, Jabri B. 2021. A multilayered immune system through the lens of unconventional T cells. Nature 595:501-10.
6. Vivier E, Artis D, Colonna M, Diefenbach A, Di Santo JP, Eberl G, Koyasu S, Locksley RM, McKenzie ANJ, Mebius RE, Powrie F, Spits H. 2018. Innate lymphoid cells: 10 years on. Cell 174:1054-66.
7. Kotas ME, Locksley RM. 2018. Why innate lymphoid cells? Immunity 48:1081-90.
8. Netea MG, Joosten LAB, Latz E, Mills KH, Natoli G, Stunnenberg HG, O’Neill LA, Xavier RJ. 2016. Trained immunity: a program of innate immune memory in health and disease, Science 352:aaf1098.
9. Palm NW, Rosenstein RK, Medzhitov R. 2012. Allergic host defenses. Nature 484:465-472.
10. Schneider C*, Lee* J, Koga S, Ricardo-Gonzalez R, Nussbaum JC, Smith LK, Villeda SA, Liang H-E, Locksley RM. 2019. Tissue-resident group 2 innate lymphoid cells differentiate by layered ontogeny and in situ perinatal priming. Immunity 50:1425-1438.
11. Huang Y, Mao K, Chen X, Sun M-A, Kawabe T, Li W, Usher N, Zhu J, Urban, Jr, JF, Paul WE, Germain RN. 2018. S1P-dependent interorgan trafficking of group 2 innate lymphoid cells supports host defense. Science 359:114-119.
12. Rankin LC, Artis D. 2018. Beyond host defense: emerging functions of the immune system in regulating complex tissue physiology. Cell 173:554-67.
13. Nussbaum JC, Van Dyken SJ, von Moltke J, Cheng LE, Mohapatra A, Molofsky AB, Thornton EE, Krummel MF, Chawla A, Liang H-E, Locksley RM. 2013. Type 2 innate lymphoid cells control eosinophil homeostasis. Nature 502:245-248.
14. Van Dyken SJ, Nussbaum JC, Lee J, Molofsky AB, Liang H-E, Pollack JL, Gate RE, Haliburton GE, Ye CJ, Marson A, Erle DJ, Locksley RM. 2016. A tissue checkpoint regulates type 2 immunity. Nat Immunol 17:1381-1387.
15. Ricardo-Gonzalez RR, Van Dyken SJ, Schneider C, Lee J, Nussbaum JC, Liang H-E, Vaka D, Eckalbar WL, Molofsky AB, Erle DJ, Locksley RM. 2018. Tissue signals imprint ILC2 identity with anticipatory function. Nature Immunol 19:1093-1099.
16. von Moltke J, Ji M, Liang H-E, Locksley RM. 2016. Tuft cell-derived IL-25 regulates an intestinal ILC2-epithelial response circuit. Nature 529:221-225.
17. Gerbe F, Sidot E, Smyth DJ, Ohmoto M, Matsumoto I, Dardalhon V, Cesses P, Garnier L, Pouzolles M, Brulin B, Bruschi M, Harcus Y, Zimmermann VS, Taylor N, Maizels, RM, Jay P. 2016. Intestinal epithelial tuft cells initiate type 2 mucosal immunity to helminth parasites. Nature 529:226-230.
18. Schneider C, O’Leary CE, Locksley RM. 2019. Regulation of immune responses by tuft cells. Nature Reviews Immunol 19:584-593.
19. O’Leary CE, Schneider C, Locksley RM. 2019. Tuft cells – systemically dispersed sensory epithelia integrating immune and neural circuitry. Annu Rev Immunol 37:47-72.
20. Schneider C, O’Leary CE, von Moltke J, Liang H-E, Yan Ang Q, Turnbaugh PJ, Radhakrishnan S, Pellizzon M, Ma A, Locksley RM. 2018. A metabolite-triggered tuft cell-ILC2 circuit drives small intestinal remodeling. Cell 174:271-284.
21. Campbell L, Hepworth MR, Whitingham-Dowd J, Thompson S, Bancroft AJ, Hayes KS, Shaw TN, Dickey BF, Flamar A-L, Artis D, Schwartz DA, Evans CM, Roberts IS, Thornton DJ, Grencis RK. 2019. ILC2s mediate systemic innate protection by priming mucus production at distal mucosal sites. J Exp Med 216:2714-2723.
22. Howitt MR, Lavoie S, Michaud M, Blum AM, Tran SV, Weinstock JV, Gallini CA, Redding K, Margolskee RF, Osborne LC, Artis D, Garrett WS. 2016. Tuft cells, taste-chemosensory cells, orchestrate type 2 immunity in the gut. Science 351:1329-1333.
23. Lambrecht BN, Hammad H. 2017. The immunology of the allergy epidemic and the hygiene hypothesis. Nat Immunol 18:1076-1083.
24. Florsheim EB, Sullivan ZA, Khoury-Hanold W, Medzhitoz R. 2020. Food allergy as a biological food quality control system. Cell 184:1440-54.
25. Ramanan D, Bowcutt R, Lee SC, Tang MS, Kurtz ZD, Ding YI, Honda K, Gause WC, Blaser MJ, Bonneau RA, Lim YAL, Loke P, Cadwell C. 2016. Helminth infection promotes colonization resistance via type 2 immunity. Science 353:608-612.
26. Broadhurst MJ, Leung JM, Kashyap LV, McCune JM, Loke P. 2010. IL-22 CD4 T cells are associated with therapeutic Trichuris trichiura infection in an ulcerative colitis patient. Sci Transl Med 2:60ra88.
27. Drokhlyansky E, Smillie CS, Van Wittenberghe N, Ericsson M, Griffin GK, Eraslan G, Dionne D, Cuoco MS, Goder-Reiser MN, Sharova T, Kuksenko O, Aguirre AJ, Boland GM, Graham D, Rozenblatt-Rosen O, Xavier RJ, Regev A. 2020. The human and mouse enteric nervous system at single-cell resolution. Cell 182:1606-1622.
28. Naik S, Larsen SB, Cowley CJ, Fuchs E. 2018. Two to tango: Dialog between immunity and stem cells in health and disease. Cell 175:908-920.