Transcription Factors Directing Th2 Differentiation: Gata-3 Plays a Dominant Role

From the 1970s to the early 1980s, there was increasing evidence for functional heterogeneity among CD4⁺ T cells from the work of numerous researchers. Mosmann and Coffman (1), working then at the DNAX Research Institute for Immunology in California, advanced and clarified the field of CD4⁺ T cell heterogeneity by their seminal discoveries that distinct CD4⁺ Th cell subsets, at the clonal level, show clear differences in the production of key hallmark cytokines. Importantly, their work linked the distinct cytokine profiles to the differences in function of the Th cell clones. In particular, they showed how all of the major features of allergic and anti-helminth responses (known to include IgE, mast cell, and eosinophil responses, induced by then undefined mechanisms) were produced by clones of a specific subset: the Th2 subset. In contrast, factors (mainly IFN-γ) mediating delayed-type hypersensitivity responses and macrophage activation were made by an alternative subset: the Th1 subset. The fact that these distinct functions could be mediated by a single T cell subset, along with the attribution of which cytokines were responsible for the activities, represented key discoveries. They and many other investigators went on to show that the ability of Th1 and Th2 cells to produce distinct sets of cytokines was of fundamental importance for protection against different pathogens or, conversely, to mediate immune pathologies, such as colitis or allergy, in both mouse and man (reviewed in Refs. 2–5). Since then, other subsets of CD4⁺ T cells have been described, including the Th17 subset, which is important for control of extracellular bacterial and fungal infections but can also mediate immune pathologies (reviewed in Ref. 6), and regulatory Foxp3⁺ T cells, which control immune responses and, thus, inhibit immune pathologies (reviewed in Ref. 7).

The initial description of Th1 cells producing IFN-γ, important for the activation of macrophages and cell-mediated immunity (8), and of Th2 cells promoting the activation of eosinophils, mast cells, and IgE-producing B cells (1), was followed by a flood of publications and suggested that naive CD4⁺ T cells could develop into multiple distinct types of cytokine-producing T cells. It became clear then that deciphering the molecular basis of the development of Th1 and Th2 cells from a naive CD4⁺ T cell precursor was the next step. The development of in vitro systems was instrumental in identifying the factors underlying the differentiation of Th1 and Th2 subsets from a naive CD4⁺ T cell and for enhancing subsequent understanding of the molecular basis of the differentiation of Th1 and Th2 cells. Using T cells stimulated with polyclonal activators or T cells expressing transgenic TCRs of known specificities cultured with their cognate Ag and APCs, it was shown that Th1 and Th2 subsets develop from the same T cell precursor, which is a mature, naive CD4⁺ T lymphocyte producing mainly IL-2 first upon Ag-specific stimulation. The groups of Swain (9) and Paul (10) demonstrated that naive CD4⁺ T cells stimulated with polyclonal TCR activators in the absence of APCs, but in the presence of IL-4, differentiated into Th2 cells, themselves producing IL-4 within a few days of culture in vitro. This was extended to cultures of TCR-transgenic CD4⁺T cells stimulated in the presence of their cognate Ag and APCs, to which the addition of IL-4 was also shown to direct the development of Th2 cells (11, 12). The groups of Murphy and O’Garra (13) collaborated further to show that TCR-transgenic CD4⁺ T cells stimulated with their cognate Ag presented by APCs could differentiate into Th1 cells, provided that macrophages within the APC population were stimulated simultaneously with pathogen-derived products (heat-killed Listeria monocytogenes or LPS) to produce IL-12. They then showed that IL-12 was the major driver of Th1 cell differentiation and that the addition of IL-12 to APC-independent cultures of naive CD4⁺ T cells stimulated with polyclonal activators of the TCR also led to the development of Th1 cells (13).

These fundamental findings of what factors directed CD4⁺ T cell differentiation (14) laid the stepping-stones for elucidating the underlying molecular mechanisms of Th cell specification. In the featured Pillars article by Zheng and Flavell (15), published in Cell in 1997, they used the in vitro systems described above to show that Gata3 was selectively highly expressed at the mRNA level in naive CD4⁺ T cells and in differentiated Th2 cells, but not in Th1 cells. Zheng and Flavell first identified this selective expression of Gata3 by examining differences in mRNA expression between Th2 and Th1 cells by representational difference analysis of cDNA. They went on to confirm that Gata3 is preferentially expressed by Th2, and not Th1, cells, with probes made from Gata3 cDNA and hybridized to the original Th1 and Th2 representations and to the Th2 representational difference analysis difference products. These data showed that strong signals were detected in the original Th2 representations, with only a very weak band detected in Th1 representations. This was further validated by RT-PCR, with which strong expression of Gata3 was observed in Th2 clone D10, with minimal expression in Th1 clone AE7. In keeping with a previous report that Gata3 was expressed in thymocytes (16), they observed expression of Gata3 in naive CD4⁺ T cells, which was maintained at a high level during the differentiation of Th2 cells but dramatically decreased after 2 d of Th1 differentiation.

Zheng and Flavell went on to test whether GATA-3 was actually required for Th2 cell differentiation, a possibility supported by the existence of potential GATA-3 binding sites in the promoters of all Th2 cytokine genes. Because GATA-3 deficiency causes embryonic lethality (17), and Gata3^−/− embryonic stem cells could not reconstitute the T cell compartment in Rag2-deficient mice (18), an antisense Gata3 construct was transfected into the Th2 clone D10 to knock down GATA-3 expression. The expression of Il4, Il6, and Il13 mRNA in the antisense Gata3 D10 Th2 cell clones was abolished or significantly reduced, and the expression of Il10 mRNA was also greatly inhibited. Il5 gene expression appeared to be inhibited as well, albeit to a lesser extent, as measured at the mRNA level and at the protein level. The investigators also ruled out the possibility that the inhibition of cytokine transcription in the antisense Gata3 transfectants resulted from inadequate stimulation of the cells caused by low TCR expression, a possibility made plausible because GATA-3 binding sites were found in the TCR δ and α gene enhancers, and GATA-3 could transactivate these enhancers (19, 20). Zheng and Flavell showed that TCR expression levels, measured by flow cytometry, were not altered in the antisense Gata3 transfectants compared with control cells, although their Ag-specific proliferation was reduced by the loss of IL-4 autocrine-induced proliferation. Thus, they concluded that the reduction in cytokine production in D10 cells by antisense Gata3 was specific rather than a secondary result of the reduced capability to respond to Ag stimulation. They went on to show that ectopic expression of GATA-3 activated the Il4 promoter in the B cell lymphoma M12; however, this was only observed when the cells were treated with PMA plus ionomycin in the presence of GATA-3, suggesting that GATA-3 is a potent transactivator of the Il4 promoter. Finally, they demonstrated that constitutive expression of GATA-3 caused Th2 cytokine gene expression in Th1 precursors of CD4–Gata3-transgenic mice.

Concurrent and subsequent studies using alternate technologies to those of the Pillars of Immunology article by Zheng and Flavell (15) clearly support the premise of this paper, confirming that GATA-3 is the key transcription factor regulating the differentiation and maintenance of Th2 cytokine production. A separate study by Ray et al. was also published in 1997, and showed that the transcription factor GATA-3 was upregulated in murine Th2 cells (while repressed in Th1 cells) and controlled Th2-specific expression of the Il5 gene, independently and concurrently demonstrating the importance of this transcription factor in Th2 differentiation (21). Later, two groups led by Murphy (22) and O’Garra (23) independently showed that retroviral transduction of Gata3 into naive CD4⁺ T cells during their differentiation into Th1 cells in the presence of IL-12 not only led to induction of Th2 cytokines but that this was preceded by the inhibition of Th1 production of IFN-γ. Similarly, mice transgenic for a dominant-negative form of GATA-3 showed modest diminution in IL-4 production in bronchoalveolar lavage fluid after aerosol challenge of sensitized mice with OVA (24). However, IFN-γ production was also inhibited when such transgenic cells were primed in Th1 conditions (24), suggesting that levels of GATA-3 and the context of its expression determine the degree to which it induces Th2 cytokines and inhibits IFN-γ production. Because deletion of Gata3 leads to embryonic lethality (17), and complementation analysis of Rag2-deficient embryos showed that GATA-3–deficient cells fail to contribute to the T cell lineage (18) and that GATA-3 is, in fact, essential at multiple stages of CD4⁺ T cell development (25), direct genetic evidence to support the idea of specific involvement of GATA-3 in Th2 responses was needed. To address the function of GATA-3 in Th2 responses, particularly in STAT5 and Nippostrongylus brasiliensis models, in which IL-4 is not required and GATA-3 is expressed at low abundance, Paul and colleagues (26) generated conditional GATA-3–deficient cells. They showed, using conditional deletion of Gata3 in vitro or in vivo in CD4⁺ T cells, that GATA-3 is very important for initiating and maintaining Th2 responses, even when it is underexpressed. They also found GATA-3 to be important for Th2, but not Th1, cell population expansion and that deletion of Gata3 was sufficient to initiate Th1 responses in the absence of both IL-12 and IFN-γ (26), in keeping with the previously reported negative effect of GATA-3 on Th1 IFN-γ production (23), suggesting that GATA-3 serves as a principal switch for Th1-Th2 responses.

Undisputedly, this pioneering Pillars article by Zheng and Flavell, which showed that GATA-3 appears to be a key regulator of the collective activity of the transcription factors related to Th2 cytokine gene expression, was momentous for subsequent studies identifying the Th1 (27) and Th17 (28) lineage hallmark transcription factors that then instructed the concept of Th cell lineage specification. A recent paradigm of how Th cell identity is established involves a structured, modular network of transcription factors. Pioneer transcription factors induced downstream of the TCR are able to bind closed chromatin and, together with specific STATs, are initially responsible for the establishment of permissive epigenetic patterns across a spectrum of Th cell target genes (29). Specificity of the Th responses arises from the hallmark transcription factors whose binding sites are thought to be more confined to lineage-determining genes, including those limiting alternative fates (29). In developing thymocytes, in which GATA-3 was found to be an essential and specific regulator (18), GATA-3 binding can precede enhancer activation (30), suggesting that it may function as a pioneer-like transcription factor. Indeed, GATA-3 was shown to exhibit pioneering functions by mediating the remodeling of the Il4 gene locus (31, 32) in addition to transactivating the Il4 promoter (15). GATA-3 also binds and facilitates remodeling of the Il10 gene locus, but it does not appear to transactivate it (33, 34). Furthermore, GATA-3 and STAT6 were shown to directly remodel the Th2 locus control region (35, 36) and, together with additional transcription factors, are essential for its establishment, maintenance, and three-dimensional organization (reviewed in Ref. 37), collectively promoting the expression of Th2-associated cytokines and ultimately contributing to specification of Th2 cells.

GATA-3 biology has clearly demonstrated that, within the hematopoietic system, GATA-3 has multiple and diverse roles (37). This is highlighted by further evidence for the participation of GATA-3 in the development of the recently appreciated innate lymphoid cells (ILCs), including ILC1, ILC2 (thus further contributing to Th2 immunity), and ILC3 (reviewed in Ref. 37), as well as thymic NK cells, including their maturation, homing to the liver, and, paradoxically, IFN-γ production (37, 38). Therefore, the 1997 Cell paper by Zheng and Flavell was pivotal for our understanding of the complexity of GATA-3’s effects that are far reaching and likely mediated in a multifaceted, dose-dependent, developmental stage–specific and cell lineage–specific fashion (30). Crucially, the featured Pillars article by Zheng and Flavell (15) formed the foundation for molecular characterization of Th cell specification.