Type 2 innate immune responses and the natural helper cell


  • Senior author: Shigeo Koyasu

S. Koyasu, Department of Microbiology and Immunology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. Email: koyasu@sc.itc.keio.ac.jp


The T helper type 2 (Th2) immune response, characterized by the production of interleukin-4 (IL-4), IL-5 and IL-13, is a critical immune response against helminths invading cutaneous or mucosal sites. It also plays a critical role in the pathophysiology of allergic diseases such as asthma and allergic diarrhoea. The Th2 cytokines are induced soon after helminth infection, even before a pathogen-specific adaptive immune response is established. Recent studies have shed light on such innate Th2 cytokine production by formerly uncharacterized innate immune cells such as natural helper cells capable of producing Th2 cytokines in response to IL-25 and IL-33 independently of adaptive immune responses. These cells produce large amounts of Th2 cytokines, most notably IL-5 and IL-13, leading to eosinophilia and goblet cell hyperplasia. We discuss here the mechanisms of innate production of Th2 cytokines in host immune responses against helminth infection as well as allergic immune responses and the similarities and differences between recently identified Th2-cytokine producing cells.


The T helper type 2 (Th2) inflammatory responses are characterized by the recruitment and activation of mast cells, basophils and eosinophils, and goblet cell hyperplasia in airway and intestinal epithelia.1,2 These immune responses are induced against helminths invading cutaneous or mucosal sites and function as protective immunity against those pathogens. Interestingly, Th2 cytokines are induced soon after helminth infection and before pathogen-specific Th2 cells are established. Although the final expulsion of helminths usually requires Th2-mediated immunity, an early Th2-type innate immune response is important in the restriction of helminth invasion before the adaptive immune response initiates. Similar Th2 responses play important roles in the pathophysiology of allergic diseases such as asthma and allergic diarrhoea.

Studies on the relatively new cytokines, thymic stromal lymphopoietin (TSLP), interleukin-25 (IL-25) and IL-33, revealed that epithelial cells play important roles in inducing Th2 immune responses against allergens and helminths.3–5 Various types of epithelial cells produce TSLP, IL-25 and IL-33 in response to allergens and helminths.6,7 For example, house dust mites stimulate lung epithelial cells through Toll-like receptor 4.8

Interleukin-4 is a key cytokine that triggers antigen-specific Th2 responses. Although the initial source of IL-4 is still unknown, CD1d-restricted natural killer T (NKT) cells,9,10 mast cells11 and basophils12,13 are able to produce IL-4 upon stimulation. Basophils produce IL-4 and accumulate in the liver and lung after Nippostrongylusbrasiliensis infection.14 Many allergens are associated with cysteine protease activities. Incubation of basophils with one of cysteine proteases, papain, leads to the production of IL-4.15 Intriguingly, basophils are transiently recruited to the draining lymph node where Th2 cells are predominantly induced after Trichurismuris and Shistosoma mansoni infection16–18 but it is controversial whether basophils present antigens to T cells.19–21

Although the antigen-specific Th2 response plays a central role in protective immunity against helminths and antigen-specific allergic responses, accumulating evidence indicates the involvement of innate immune cells in the onset of Th2 responses. Administration of IL-25 and IL-33 induces rapid production of Th2 cytokines such as IL-5 and IL-13 in mice independently of T or B cells.22–24 Recent studies identified a previously unrecognized cell population(s) capable of producing large amounts of IL-5 and IL-13 in response to IL-25 and IL-33. We discuss here the characteristics and functions of such Th2 cytokine-producing innate cells. In particular, we focus on natural helper (NH) cells found in fat-associated lymphoid clusters (FALCs) present in visceral adipose tissue.25

Epithelial cell-derived Th2-inducing cytokines

Accumulating evidence has shown the importance of epithelial cells in the production of cytokines such as TSLP, IL-25 and IL-33 in response to allergens and helminths.3–7

Thymic stromal lymphopoietin (TSLP)

Thymic stromal lymphopoietin was originally identified from thymic stromal cells and thought to support growth and differentiation of T and B cells but is now considered to be a Th2-inducing cytokine.6 It acts to induce dendritic cells (DCs) capable of differentiating naive CD4+ T cells to Th2 cells.26,27 The TSLP-activated DCs produce CXCL8 and CCL24 attracting neutrophils and eosinophils, respectively. Th2-attracting chemokines, CCL17 and CCL22, are also produced by DCs activated by TSLP. Interestingly, however, these DCs do not produce tumour necrosis factor, IL-1β, IL-6, IL-10 or IL-12. Induction of Th2 cells by TSLP-stimulated DCs depends on OX40 ligand specifically induced by TSLP.28


Interleukin-25 is a member of the IL-17 family, which is also called IL-17E.22 The IL-25 receptor (IL-25R) is composed of IL-17RA and IL-17RB whereas the receptor for other IL-17 family members is a heterodimer of IL-17RA and IL-17RC.29 Interleukin-25 is produced by Th2 cells and epithelial cells.22,23,30,31 For example, lung epithelial cells infected with Aspergillus fumigatus23 or sensitized with ovalbumin30 as well as gut epithelial cells infected with N. brasiliensis31 produce IL-25. Mast cells also produce IL-25 upon cross-linking of FcεRI.32

Unlike other family members that induce neutrophil-mediated inflammatory responses, administration of IL-25 induced eosinophilia and IgE production in mice, a typical Th2 response.22 Repeated nasal administration of IL-25 resulted in the expression of IL-5 and IL-13 in the lung and induced airway hypersensitivity.23 Similarly, transgenic mice expressing IL-25 exhibited increased levels of serum IL-5 and IL-13 and eosinophilia.33,34In vitro, IL-25 induces the production of IL-5 and IL-13 from a CD4 CD8 MHC-IIhigh CD11cdull F4/80low non-T and non-B (NTNB) population.22 Both IL-5 and IL-13 are also produced in Rag2−/− mice upon IL-25 administration, confirming the production of these cytokines by NTNB cells.23 Intriguingly, IL-25 induces the expression of IL-5 and IL-13 but not IL-4. Mice lacking the cytokine receptor common γ chain (γc) are incapable of producing IL-5 and IL-13 in response to IL-25.23

Interleukin-25-deficient mice were unable to elicit a Th2 response upon T. muris infection and are therefore unable to control the infection.35,36 Similarly, the expulsion of N. brasiliensis was significantly delayed in IL-25-deficient mice but administration of recombinant IL-25 leads to the induction of IL-5 and IL-13 from NTNB cells and to the expulsion of N. brasiliensis.23


Interleukin-33 is a member of the IL-1 family37,38 expressed in a variety of cells including fibroblasts, epithelial cells, endothelial cells and adipocytes.39–41 Surprisingly, IL-33 is localized in the nucleus and functions as a heterochromatin-associated transcriptional repressor like IL-1α.42,43 Caspases 3 and 7 cleave IL-33 at the IL-1-like domain and inactivate IL-33 during apoptosis.44,45 In contrast, full-length IL-33 is released from cells upon necrotic cell death and functions as an alarmin to stimulate various types of cells.44 Similar alarmins that are also present in the nucleus and are released upon cellular damage include high mobility group box-1,46 IL-1α47 and spliceosome associated protein 130.48

Interleukin-33R is composed of ST2 and IL-RAcP39 and is expressed on various types of cells including Th2 cells,49 mast cells50 and basophils.51–53 Interleukin-33R is also expressed on CD34+ haematopoietic progenitor cells and IL-33 induces the differentiation of mast cells from CD34+ cells.54 Interestingly, a truncated form of ST2 named soluble ST2 (sST2) is secreted and acts as a negative regulator of IL-33.55

Interleukin-33 induces the production of IL-5 and IL-13 from mast cells56,57 and CD34+ cells54in vitro. Basophils also respond to IL-33 to produce IL-4, IL-5, IL-6 and IL-13.51 Administration of IL-33 induces Th2 cytokine production and associated physiological changes in mice including IgE production, eosinophilia and goblet cell hyperplasia.24 Interleukin-33 is also involved in Th2-type inflammatory responses induced by Pseudomonas aeruginosa and respiratory syncytial virus.58,59 Mice deficient for IL-33 are resistant to allergen-induced airway hypersensitivity.60 Administration of sST2 blocks the allergic response in the allergic airway inflammation model.55,61 Many studies have shown that IL-33 is involved in allergic diseases in humans. Polymorphisms of IL-33 and ST2 are associated with asthma in humans and patients with asthma have a lower level of sST2 than healthy controls.62 The expression levels of IL-33 in epithelial cells are higher in patients with asthma than in healthy individuals63 and even higher in treatment-resistant patients.64 The levels of IL-33 were also higher in patients with allergic conjunctivitis,65 rhinitis66 and atopic dermatitis67 than in healthy individuals.

Innate cells responsible for the production of Th2 cytokines

Although administration of TSLP, IL-25 and IL-33 induces Th2 cytokine production and associated physiological changes in mice such as IgE production, eosinophilia and goblet cell hyperplasia, the identity of the cell(s) responsible for the production of Th2 cytokines has been obscure. As mentioned, mast cells, basophils and CD34+ haematopoietic progenitor cells produce Th2 cytokines. A fraction of NKT cells express IL-25R and respond to IL-25 to produce IL-13.68 However, the amounts of IL-5 and IL-13 produced by these cells were lower than those produced in vivo, suggesting the presence of yet to be discovered cell type(s). We have recently identified a previously unrecognized lymphocyte population in the visceral adipose tissue of the peritoneal cavity that we named natural helper cells.25 Cells sharing similar characteristics with NH cells have also been reported by others.69–71

Natural helper cells present in FALC

We noted a previously unrecognized γc-dependent lymphoid tissue located along the blood vessels in the mouse and human mesentery, an adipose tissue in the peritoneal cavity.25 We named these lymphoid clusters ‘fat-associated lymphoid clusters’ or FALCs (Fig. 1a). Cells in FALC are in direct contact with ambient adipocytes and no fibrous capsule is present around FALC. The FALC is also present in visceral adipose tissues around the kidney and genitalia but very few are found in the subcutaneous fat tissue or the omentum. They are structurally similar to the ‘milky spot’ in the omentum, which is considered a gateway of cells between the circulation and the peritoneal cavity.72 However, unlike the milky spot,73 T-cell and B-cell zones or germinal centre structures are not observed in FALC. Conversely, NH cells are not present in milky spots. The FALC is present in retinoic acid-related orphan receptor γ (RORγ) -deficient and aly/aly mice, suggesting that the differentiation pathway of FALC is distinct from those of lymph nodes or Peyer’s patches involving lymphoid tissue inducer cells.74

Figure 1.

 Fat-associated lymphoid clusters (FALC) and natural helper (NH) cells. (a) Haematoxylin & eosin staining of FALC. Note that cells in FALC directly attach to surrounding adipocytes. Bar indicates 200 μm. (b) Giemsa staining of isolated NH cells. The NH cells are small in size and round in shape with a dark nucleus and scanty cytoplasm. Bar indicates 20 μm. These pictures are reproduced by courtesy of Nature (ref. 22).

The FLAC contains T cells, B cells, macrophages and DCs. In addition, NH cells expressing c-Kit, Sca-1, IL-2R, IL-7R and IL-33R but no conventional lineage (Lin) markers make up 20–40% of cells in FALC.25 A cell population similar to the NH cells expressing c-Kit and IL-7Rα is also present in human FALC. Giemsa staining and electron microscopy showed that NH cells have the characteristics of lymphoid cells (Fig. 1b). The NH cells were present in Rag2−/− and nu/nu mice but absent from γc−/− and IL-7−/− mice, indicating that NHs cell are of lymphoid lineage with differentiation dependent on IL-7. Stem cell factor and IL-7 supported the survival of NH cells and IL-2 induced the proliferation of NH cells.

Natural helper cells constitutively produce IL-5 and IL-6, which are involved in antibody production.75,76 Interleukin-5 is a critical growth factor for B1 cells present in the peritoneal cavity and plays an important role in innate-type immune responses by producing natural antibodies.77,78 Indeed, NH cells support the production of IgA from B cells and self-renewal of B1 cells, indicating that NH cells have helper functions under steady-state conditions (Fig. 2a).

Figure 2.

 Functions of natural helper (NH) cells. (a) Interleukin-5 (IL-5) and IL-6 constitutively produced by NH cells support IgA production by B cells. Interleukin-5 supports the survival of B1 cells in the peritoneal cavity. (b) Upon helminth infection, IL-33 produced by epithelial cells, endothelial cells and adipocytes induce high level of IL-5 and IL-13 by NH cells, inducing eosinophilia and goblet cell hyperplasia, which play important roles in an early phase of anti-helminth immune responses.

The NH cells produce even larger amounts of IL-5, IL-6 and IL-13 in response to IL-33 or a combination of IL-2 and IL-25. Of note, NH cells do not respond to IL-25 without IL-2. Five thousand NH cells are able to produce microgram amounts of IL-5 and IL-13 in response to IL-33 during a 5-day culture period, which is much higher than the amounts produced by mast cells,50 basophils51–53 or polarized Th2 cells.49 Unlike basophils, NH cells do not produce IL-4 in response to IL-33 or a combination of IL-2 and IL-25. Notably, however, NH cells are capable of producing IL-4 by a combination of 12-O-tetradecanoylphorbol-13-acetate and ionomycin. So far, this is the only stimulation leading to the production of IL-4 by NH cells.

Administration of IL-33 or Nbrasiliensis infection induced the production of IL-5 and IL-13 as well as goblet cell hyperplasia in the intestine of Rag2−/− but not γc−/− Rag2−/− mice devoid of NH cells. Adoptive transfer of NH cells into γc−/− Rag2−/− mice restored the production of IL-5 and IL-13 and goblet cell hyperplasia, demonstrating that NH cells indeed play an important role in the production of IL-5 and IL-13 in vivo.25 However, expulsion of helminths still requires T cells as Rag2−/− mice are not able to expel Nbrasiliensis (ref. 79 and our unpublished observations) and adoptive transfer of IL-4−/− IL-13−/− CD4+ T cells resulted in the clearance of Nbrasiliensis in Rag2−/− mice.79 The NH cells probably provide IL-5 and IL-13 to limit the helminth invasion before helminth-specific T cells are induced (Fig. 2b).

Multipotent progenitor type 2 cells

Using knock-in mice in which green fluorescent protein (GFP) was inserted into the IL-4 allele, termed ‘4get’ mice,80 Saenz et al. identified IL-4-producing NTNB cells (GFP+ cells) in gut-associated lymphoid tissues upon IL-25 administration.69 Adoptive transfer of GFP+ cells into IL-25-deficient mice resulted in Th2 differentiation in mesenteric lymph nodes, mucin secretion and protective immunity against Tmuris. Cells induced by IL-25 administration shared cell surface markers with haematopoietic stem cells or multipotent progenitor cells and were named MPPtype2. This population seems to be heterogeneous and c-Kit+ GFP+ cells differentiate into mast cells whereas c-Kit+ GFP cells have the potential to differentiate into basophils and macrophages.


Using another knock-in mouse in which GFP was inserted into the IL-13 allele, Neill et al. reported that NTNB cells producing IL-5 and IL-13 (GFP+ cells) were induced in the mesenteric lymph nodes, spleen and small intestine after administration of IL-25 or IL-33.70 They named these GFP+ cells ‘nuocytes’, nu being the thirteenth letter of the Greek alphabet although these cells also produce IL-5. As observed for NTNB cells in an earlier report,22 nuocytes express MHC class II and respond to IL-25 alone, which are characteristics distinct from those of NH cells (Table 1). Neill et al. established mice lacking IL-17RB that could not respond to IL-25 or control Nbrasiliensis infection. Adoptive transfer of wild-type nuocytes into IL-17RB-deficient mice restored their ability to expel Nbrasiliensis. According to their report, 2 million/ml nuocytes produced microgram amounts of IL-5, IL-6 and IL-13 upon 1-week culture with a combination of IL-7 and IL-33.

Table 1.   Characteristics of innate T helper type 2 cytokine-producing cells as examined by surface phenotypes
NH cellNuocyteMPPtype2Ih2 cell
  1. 1Based on microarray analysis.

  2. 2GFP cells expressed MHC class II after cultivation with a combination of interleukin-3 and stem cell factor.

  3. NH, natural helper; MPP, multipotent progenitor; Ih2, innate helper type 2; IL-7R, interleukin-7 receptor.

MHC class II+?2?

Innate helper type 2

Price et al. recognized a Lin c-Kit+ GFP+ population in naive 4get mice.71 It was known that eosinophils, basophils, mast cells, NKT cells and Th2 cells are GFP+ in naive 4get mice,81–83 but the Lin c-Kit+ GFP+ population they noticed was distinct from those reported previously. Price et al. also used IL-13 reporter mice and observed that such c-Kit+ NTNB cells produce IL-5 and IL-13 in response to IL-25 or IL-33 administration and N. brasiliensis infection. Intriguingly, IL-4-GFP+ cells do not produce IL-4 at the protein level. The authors named these cells innate helper type 2 (Ih2) cells. These Ih2 cells are absent in γc−/− Rag2−/− mice incapable of controlling Nbrasiliensis infection but adoptive transfer of Ih2 cells followed by administration of IL-25 produced eosinophilia and anti-helminth activity. Whether Ih2 cells are able to produce cytokines in response to IL-25 alone and the quantity of cytokines produced by Ih2 cells is currently unknown.

At the moment, the relationship between NH cells, nuocytes, MPPtype2 and Ih2 cells is unclear. There are similarities and differences, as shown in Table 1. One important difference between NH cells and the other cells is that NH cells do not respond to IL-25 alone.25,69 It is intriguing that there is no evidence of the production of IL-4 as a protein in these cells even though the il-4 gene is active as determined by microarray analyses and knock-in approaches.


Protective immunity against helminths

Invasion of helminths results in tissue destruction, leading to the secretion of alarmins from necrotic cells including IL-25 and IL-33. These cytokines act on innate immune cells such as NH cells to induce IL-5 and IL-13, resulting in eosinophilia and goblet cell hyperplasia, respectively (Fig. 2b). Eosinophilia is important in controlling helminths in the lung stage of Nbrasiliensis and Strongyloidesvenezuelensis and goblet cell hyperplasia in the intestine is involved in the blocking of attachment of helminth during the early phase of the intestinal phase. Both IL-25 and IL-33 induce IL-4 from basophils, mast cells and possibly MPPtype2 cells that support the induction of Th2 differentiation. The Th2-mediated adaptive immune responses will eventually result in the expulsion and clearance of helminths.

These pictures are in a way similar to the Th1-type innate immune response before the differentiation of pathogen-specific Th1 cells. Natural killer cells support the phagocytes by providing interferon-γ to limit the growth of intracellular pathogens including protozoan parasites and viruses.84 The IL-22-expressing NKp46+ NK cells deal with extracellular mucosal pathogens in the intestine as well as Mycobacteriumtuberculosis in the lung until Th1 cells become ready to work.85,86 From this viewpoint, it is of interest that IL-33 induces Th1 cytokines from NK cells and NKT cells,51,87 implying that IL-33 induces a Th1-type innate immune response under certain circumstances.

Contribution to allergic responses

Given that IL-5 and IL-13 are critical factors for the induction of eosinophilia and goblet cell hyperplasia, respectively, and given that these two processes are involved in the pathogenesis of allergic diseases such as asthma and allergic diarrhoea, it stands to reason that the innate Th2 cytokine producers described here play important roles in these allergic diseases.1 Of note, FALC are present in humans and it is likely that NH cells play important roles in human allergic diseases as well.

The NH cells do not respond to Toll-like receptor ligands such as lipopolysaccharide or lectins such as concanavalin A (ref. 25 and our unpublished observations). It is unknown at the moment whether the innate Th2 cytokine producers discussed here directly respond to allergens or microbial/parasitic antigens. It remains to be shown how these Th2 producers are triggered in vivo.

Inflammatory reactions in the adipose tissues

In addition to their roles in allergic diseases and protection against helminth infection, NH cells may have additional roles in adipose tissues. Accumulating evidence demonstrates the importance of interaction between adipocytes and immune cells in the regulation of homeostasis in adipose tissues. Impaired regulation under pathogenic conditions such as obesity results in inflammatory responses in adipose tissues leading to insulin resistance.88 One subset of macrophages called M1 macrophages, is involved in the inflammation in adipose tissues.88,89 Interleukin-33 induces the alternative M2 subset of macrophages but high levels of Th2 cytokines convert them to M1 macrophages.53,90,91 Because such phase transition in macrophages plays a critical role in the course of inflammation in adipose tissues, it will be of interest and of importance to elucidate the functions of NH cells in adipose tissues.

Finally, it will be important to elucidate the developmental and functional relationship between NH cells, nuocytes and Ih2 cells in future studies.


We thank our collaborators and members of our laboratory. This work was supported by a Grant-in Aid for Young Scientist (B) (20790378 to K.M.), a Grant-in Aid for Young Scientist (A) (22689013 to K.M.), Grants-in-Aid for Scientific Research (B) (14370116, 16390146, 18390155 to S.K.), a Grant-in-Aid for Scientific Research (A) (22249013 to S.K.) and a Grant-in-Aid for Scientific Research (S) (22229004 to S.K.) from the Japan Society for the Promotion of Science. K.M. is a postdoctoral fellow of the Global COE programme supported by the Ministry of Education, Culture, Sports, Science and Technology, Japan.


S.K. is a consultant for Medical and Biological Laboratories, Co. Ltd. The authors otherwise have no financial conflicts of interest.