Striking Similarity: GATA-3 Regulates ILC2 and Th2 Cells

Abstract

Innate lymphoid cells (ILCs) and T helper (Th) cells share similar effector functions. In this issue of Immunity, Hoyler et al. (2012) and Mjosberg et al. (2012) demonstrate that GATA-3, a Th2 regulator, controls ILC2 differentiation, maintenance, and function.

Innate lymphoid cells (ILCs) represent an emerging family of cell types, including natural killer (NK) cells, type 2 ILCs (ILC2s), and RORγt⁺ ILCs that play key roles in tissue remodeling and innate immunity, particularly at barrier surfaces. These cells are characterized by a lymphoid morphology and an absence of T cell receptors but can secrete a variety of effector cytokines that are also produced by T helper (Th) cell subsets. The transcription factor Id2 is required for all innate lymphoid cell development. However, the development of ILCs is independent of the genes that encode the Rag protein and thus ILCs are abundantly present in Rag-deficient mice (Spits and Cupedo, 2012). Unlike RORγt⁺ ILCs, ILC2 cells do not express RORγt but express type 2 helper T (Th2) cell-related cytokines (e.g., interleukin-13 [IL-13], IL-5, and IL-4). ILC2s are involved in protective immunity against helminth infection (e.g., N. brasiliensis) and in tissue repair after influenza virus infection, whereas dysregulation of ILC2s may lead to inflammation and autoimmunity (e.g., allergic airway inflammation) (Chang et al., 2011; Monticelli et al., 2011; Moro et al., 2010; Neill et al., 2010; Price et al., 2010; Saenz et al., 2010).

Previous data suggest that ILCs and Th cells share a number of common features (e.g., transcription factor requirement, cytokine profile, and anatomic location). As an example, both RORγt⁺ ILCs and Th17 cells require nuclear receptor RORγt for their differentiation and/or development and function. Accordingly, RORγt-deficient mice lack RORγt⁺ ILCs and have marked reduction of Th17 cells. Both RORγt⁺ ILCs and Th17 cells produce IL-17 and/or IL-22, are abundantly present in mucosal tissues, and are involved in similar immune functions (e.g., clearance of Citrobacter rodentium) (Spits and Cupedo, 2012). It is logical to hypothesize that other ILC populations (e.g., ILC2) may also share similar transcriptional programs with their corresponding Th cells (e.g., Th2 cells). In this issue of Immunity, Hoyler et al. (2012) and Mjosberg et al. (2012) show that, indeed, similar to RORγt for RORγt⁺ ILC lineage specification, GATA-3 is a bona fide regulator of ILC2 development and function.

Using elegant genetic and cellular approaches, Hoyler et al. (2012) demonstrated that GATA-3, an important Th2 cell regulator, is a critical transcription factor for ILC2 lineage specification and maintenance (Figure 1). High GATA-3 expression was detected in bone marrow (BM) progenitor cells that are Lin⁻Sca1^hiId2^hi (LSIG). Gene profiling analysis indicated a close relationship between LSIG and ILC2. LSIG can give rise to mature ILC2 that express high amounts of the lectin-like receptor KLRG1 in vitro and in vivo. Using an OP9 cell system, Hoyler et al. (2012) further showed that even under conditions that favor the development of other ILC lineages (e.g., NK or RORγt⁺ ILCs), LSIG seemed to be “prewired” for ILC2 programming, suggesting their limited plasticity at least in vitro. LSIG after transferring into alymphoid Rag2^−/−Il2rg^−/− mice proliferated and developed into mature ILC2s, which were functional in gut immunity against helminth infection (e.g., N. brasiliensis). Thus, LSIG populations in the bone marrow represent lineage-specified precursors to ILC2 (i.e., ILC2P).

Figure 1. GATA-3 Is Essential for ILC2 Development and Function in Mice.

GATA-3 determines the lineage specification of ILC2. GATA-3 is not only important for the maturation of ILC2P but also essential for the maintenance of ILC2P and mature ILC2. Bone marrow resident progenitors contain a population of cells that are Lin⁻Sca1^hiId2^hi. These cells have high expression of GATA-3 and Id2 and show similar gene profile to mature ILC2, thus representing ILC2-specific progenitors (ILC2P). After adoptive transfer, ILC2P cells migrate to the gut preferentially and differentiate into functional ILC2 with production of effector cytokines (IL-13, IL-5, and IL-4), consistent with the expression of gut homing receptor CCR9. Mature ILC2 cells in the gut do not express α4β7 but upregulate KLRG1. ILC2 cytokines stimulate gut epithelial cells to produce protective factors (e.g., angiogenin and mucin) to control helminth infection.

Using an inducible Cre-Lox system (Gata3^f/f Id2CreER^t2) to temporally ablate GATA-3 in Id2⁺ cells including all ILCs and T cells by administration of tamoxifen, the authors further showed that continuous expression of GATA-3 is essential for the maintenance of mature ILC2 and their effector functions. Temporal deletion of GATA-3 also affected the ILC2 progenitor pool in the bone marrow, suggesting that GATA-3 is required for the maintenance of ILC2P. Culturing GATA-3-deficient Lin⁻CD25⁺Sca1^high progenitors failed to generate mature ILC2 in vitro, indicating that GATA-3 is essential for differentiation of ILC2P to ILC2. These data collectively demonstrate that GATA-3 expression is crucial for the development and maintenance of ILC2 lineage.

Intriguingly, adoptive transfer experiments suggest that ILC2Ps migrate to the intestine preferentially before maturation to ILC2s, consistent with the expression of gut homing receptor CCR9 on ILC2P. Given that substantial numbers of intestinal ILC2 were found in CCR9-deficient animals, these data argue that additional mechanisms besides CCR9 (e.g., the presence of integrin α4β7 required for lymphocyte gut homing and/or retention) may account for the apparent gut homing property of ILC2Ps. Unexpectedly, the expression of integrin α4β7 actually declined in mature ILC2s when mice aged, suggesting that other adhesion molecules than α4β7 may be required for the retention of mature ILC2s in the intestine of adult mice. Consistent with this notion, KLRG1, uniformly expressed by intestinal ILC2s in adult mice at high levels, binds to classical cadherins, including E-cadherin that is highly expressed by intestinal epithelial cells. The role of KLRG1 in recruitment, retention, and function of ILC2 remains to be determined in future studies. Of note, the number of ILC2 was prominent in the lamina propria of intestine under the steady state. However, during inflammation or infection (e.g., IL-25, IL-33 injection, or N. brasiliensis infection), ILC2 numbers significantly increased in other organs (e.g., the lung, liver, and spleen). It remains to be determined whether this increase of ILC2 in extraintestinal sites represents an enhanced recruitment (e.g., active diversion of ILC2 from the gut to other organs), increased proliferation and/or survival of other tissue resident ILC2, and/or facilitated de novo differentiation of ILC2P in situ.

Using gain-of-function approach (overexpression by retroviral transduction) and loss-of-function approach (gene silencing by short-hairpin RNA), Mjosberg et al. delineated the signaling pathways that regulate ILC2 function and convincingly showed that GATA-3 is essential for the production of type 2 cytokines (e.g., IL-13, IL-5, and IL-4) by human ILC2s.

ILC2, defined as Lin⁻CD127⁺CRTH2⁺ in humans, are highly enriched in inflamed nasal polyps of patients with chronic rhinosinusitis (CRS). Interestingly, although the TSLP-TSLPR axis is dispensable for intestinal ILC2 development and/or maintenance in mice, it is important for the function of human ILC2 (Hoyler et al., 2012). Consistent with the potential role of microbes in CRS pathogenesis, various TLR ligands activated nasal polyp epithelial cells to increase Tslp and Il33 gene expression. TSLP and IL-33 produced by epithelial cells in turn act on ILC2 by binding to TSLP receptor (TSLPR) and IL-33 receptor subunit (ST2), respectively. The authors further showed that TSLP activated STAT5 but not STAT3 to markedly enhance GATA-3 expression. Although IL-33 acts synergistically with TSLP to induce ILC2 effector cytokines (e.g., IL-13, IL-5, and IL-4), it seemed to have less of an effect on GATA-3 induction. It has been shown that IL-33 enhances ILC2 proliferation and function, but the signaling pathways underlying these processes remain to be determined (Neill et al., 2010; Price et al., 2010). It is of special interest to determine how the synergy between TSLP (e.g., via JAK-STAT pathway) and IL-33 (e.g., via NF-κB and AP-1 pathways) is achieved in the induction of ILC2 effector cytokines.

It is interesting to note that ectopic expression of GATA-3 in Lin⁻CD127⁺CD117⁺NKp44⁻CRTH2⁻ cells by retroviral transduction resulted in the induction of surface marker of human ILC2 and Th2 cells, CRTH2. However, overexpression of GATA-3 was not able to directly induce ILC2 effector cytokines, which required additional stimulation of GATA-3-expressing cells with IL-33, TSLP, and IL-2. Consistently, expression of TSLPR and ST2 was significantly enhanced by GATA-3 overexpression. These data suggest a positive amplifying mechanism underlying effector function of ILC2. Induction of GATA-3 by TSLP upregulates TSLPR and ST2, resulting in the ability of ILC2 to respond to TSLP and IL-33 for further enhancement of GATA-3. The additional requirement of TSLP and IL-33 for induction of ILC2 effector cytokines suggests that factor(s) induced by TSLP and/or IL-33 may function cooperatively with GATA-3 to induce effector cytokine gene transcription. It is tempting to speculate that a permissive chromatin structure set up by TSLP and/or IL-33 at the IL13, IL5, or IL4 loci in ILC2 may be required for GATA-3 binding and transactivation.

ILC2 were recently shown to depend on the transcription factor RORα (Halim et al., 2012; Wong et al., 2012). Both Hoyler et al. (2012) and Mjosberg et al. (2012) showed that the expression of RORα is not limited to ILC2, thus raising an interesting question about the cell-intrinsic role of RORα in ILC2. The development and characterization of RORα mice with ILC-specific deletion will help elucidate the role of RORα in the future. Although it was shown that RORA transcription is not regulated by GATA-3 (Mjosberg et al., 2012), it remains to be determined whether RORα works in parallel to interplay with GATA-3 (e.g., by protein-protein interaction), thereby promoting ILC2 development and function.

ILCs share a striking similarity to Th cells in effector cytokine production. Coevolution of two systems may be a fail-safe mechanism to implement redundancy in host immunity against certain infections especially at the mucosal surfaces. Active cross-regulation between individual ILCs and Th cells may also be present in the host. Recent data suggest that Th cells can display significant plasticity in certain conditions (e.g., during infection and inflammation). It remains to be determined whether this plasticity is also associated with ILCs. Collectively, the two papers published in the issue shed novel light on the ILC biology and demonstrate that ILC2 and Th2 cells share the same master transcriptional regulator GATA-3.