Cytokines of the γ_c family control CD4⁺ T cell differentiation and function

Abstract

Naive CD4⁺ T cells undergo massive proliferation and differentiation into at least four distinct helper T cell subsets after recognition of foreign antigen–derived peptides presented by dendritic cells. Each helper T cell subset expresses a distinct set of genes that encode unique transcription factor(s), as well as hallmark cytokines. The cytokine environment created by activated CD4⁺ T cells, dendritic cells and/or other cell types during the course of differentiation is a major determinant for the helper T cell fate. This Review focuses on the role of cytokines of the common γ-chain (γ_c) family in the determination of the effector helper T cell phenotype that naive CD4⁺ T cells adopt after being activated and in the function of these helper T cells.

Cytokines regulate a variety of cellular responses, including proliferation, differentiation and survival. Among the several different classes of cytokines, type I cytokines have a particularly important role. Type I cytokines have a four α-helix bundle structure and bind transmembrane proteins whose extracellular regions contain a hematopoietin receptor domain. This evolutionarily conserved, 200–amino acid region derived from a tandem of two ancestral fibronectin-like domains has four conserved cysteine residues in the amino-terminal segment and a tryptophan-serine doublet near the carboxy-terminal end¹. Although they were first defined as cytokines, the type I cytokines also include hematopoietic factors and endocrine hormones. Their receptors belong to the type I cytokine receptor superfamily.

An important subfamily of the type I cytokines are those that use the common γ-chain (γ_c) to generate signaling receptor complexes. These include interleukin 2 (IL-2), IL-4, IL-7, IL-9, IL-15 and IL-21 (ref. 2). In addition, the cytokines IL-13 and TSLP are closely related to IL-4 and IL-7, respectively, and, although they do not use γ_c, they use an alternative chain (IL-13Rα1 for IL-13, and TSLPR for TSLP) that may have comparable features. In general, the receptor chain that binds the cytokine with high affinity is designated the α-chain (for example, IL-4Rα); the IL-2 receptor (IL-2R) and IL-15R complexes are exceptions to this rule. IL-2R consists of three subunits: the IL-2 receptor β-chain (IL-2Rβ (CD122)), which is the analog of the α-chains in other receptor complexes; the common γ-chain (γ_c (CD132)); and a third chain (IL-2Rα (CD25)) that is not a member of the structural family of type I cytokine receptors. IL-2Rα alone binds IL-2, although it does so with low affinity and no signaling ability. An intermediate-affinity IL-2R complex is composed of IL-2Rβ and γ_c, and all three subunits form the high-affinity IL-2R complex, which has both a rapid on rate and a slow off rate³. The IL-15R complex is similar to the IL-2R complex, with IL-2Rβ and γ_c but with the IL-15Rαhomolog of IL-2Rα. The IL-15R system, although structurally homologous to the IL-2R system, is unique in that not all the receptor subunits are necessarily expressed on the same cells. IL-15 may be captured by cells that express an IL-15Rα chain and then may be presented in trans (that is, as an IL-15-IL-15Rα complex) to neighboring cells that express IL-2Rβ and γ_c chains, which leads to IL-15-mediated immune responses⁴.

Each of the receptors for the γ_c family of cytokines transduces signals through the kinases Jak1 and Jak3, but different members of the family activate different transcription factors of the STAT family². IL-2, IL-7, IL-9 and IL-15 activate mainly STAT5 (both STAT5A and STAT5B); IL-4 activates STAT6 and, to a lesser extent, STAT5; and IL-21 activates mainly STAT3. In this Review, we will discuss the role of cytokines of the γ_c family in the fate of peripheral CD4⁺ T cells during their differentiation into effector helper T cells after exposure to their cognate antigens and in their function. We emphasize results from mouse systems, recognizing that although the principles of helper T cell differentiation in mice and humans are similar, there are some differences in detail⁵.

T cells originate in the thymus and undergo a process of selection in which cells able to bind complexes of self peptide and major histocompatibility complex (MHC) with some threshold affinity are rescued from apoptosis (positive selection), whereas cells with high affinity for self peptide–MHC complexes are eliminated (negative selection). Cytokines of the γ_c family have a crucial role in this selection process; excellent reviews of this topic have been published elsewhere^6–8.

When naive CD4⁺ T cells recognize foreign antigen–derived peptides presented in the context of MHC class II on dendritic cells (DCs) in the periphery, these cells undergo a process that includes massive proliferation and differentiation into distinct helper T cell subsets. There is still considerable uncertainty about the number of these subsets, the precursor-product relationships among them and their ability to convert from one to another. At least four different subsets of helper T cells (T_H1, T_H2, T_H17 and regulatory T cells (T_reg cells)) have been studied in great detail. Each expresses a distinct set of regulatory transcription factors, including what is sometimes designated a ‘master regulator’, as well as hallmark cytokines. Indeed, the cytokine environment created by activated CD4⁺ T cells themselves, by ‘partner’ DCs and/or by other cell types during the course of differentiation is one of the key determinants for the differentiation into distinct helper T cell subsets⁹.

T_H1 differentiation: the role of IL-2

The master regulatory factor T-bet has a central role in T_H1 differentiation. In developing T_H1 cells, IL-2-driven activation of STAT5 controls the binding of T-bet to the conserved noncoding sequence CNS-1 in the promoter of the gene (Ifng) encoding interferon-γ (IFN-γ) and thereby regulates IFN-γ production and the positive feedback process through which T_H1 differentiation proceeds¹⁰. In addition, IL-2 regulates induction of the IL-12Rβ2 subunit of the IL-12 receptor, an event critical to the completion of T_H1 differentiation because of the central role of IL-12 in both the induction of T-bet and the production of IFN-γ¹¹. This IL-2 effect also depends on STAT5 activation¹². There is a still unresolved controversy about whether the induction of T-bet expression requires the IL-2–STAT5 pathway^10,12. The key cytokines involved in T_H1 differentiation are IFN-γ and IL-12, with the latter being a type I cytokine that does not use γ_c.

T_H2 differentiation: the role of IL-2 and IL-4

In vitro T_H2 differentiation requires stimulation via the T cell antigen receptor (TCR) and IL-4-mediated activation of STAT6, which jointly induce the expression of GATA-3, a T_H2 master regulatory transcription factor¹³. However, the very first report on in vitro T_H2 differentiation pointed out the requirement for both IL-4 and IL-2 for optimal T_H2 development¹⁴. Although the role of IL-2 in T_H2 differentiation was unclear at that time, subsequent studies have indicated a central role for IL-2 in T_H2 differentiation. Neutralization of endogenous IL-2 results in the failure of naive CD4⁺ T cells to undergo T_H2 development without affecting their proliferation. In such experiments, the ‘differentiation’ and ‘proliferation and survival’ functions of IL-2 can be distinguished because IL-4, a key component of in vitro T_H2 differentiation, can provide the second two functions in place of IL-2 but cannot provide its differentiation function¹⁵. Indeed, ectopic expression of constitutively active STAT5A in activated CD4⁺ T cells leads to robust T_H2 differentiation even when IL-4 activity is blocked and exogenous IL-12 is provided¹⁶. STAT5 binds to DNase I–hypersensitivity site II in the second intron of Il4 (Fig. 1), although the importance of this second intron site relative to that of other potential STAT5-binding sites in the larger Il4 genetic region has not been established. Thus, the IL-2–STAT5 pathway may function to maintain the Il4 locus in an open configuration so that transcription factors required for T_H2 differentiation gain access to their binding sites.

Figure 1.

A model for early T_H2 differentiation. The recognition of an antigen–MHC class II complex (Ag-MHCII) by the TCR on a naive CD4⁺ T cell leads to the upregulation of GATA-3 expression independently of IL-4 during the early stages of T_H2 differentiation. The NF-ºB1-Bcl-3 complex, the Notch-CSL pathway and the Wnt–β-catenin–TCF-1 pathway have been each proposed to have a critical role in TCR-driven GATA-3 expression. GATA-3 binds to several sites on loci encoding T_H2 cytokines, ncluding DNase I–hypersensitivity sites II and III (HS II-III) and site V (HS V_A) in the Il4 locus. Stimulation via the TCR also induces the production of IL-2 and expression of the IL-2R complex, which results in the activation of STAT5. Activated STAT5 binds to DNase I–hypersensitivity sites II and III, which act together with GATA-3 to induce small amounts of early IL-4 production, and also binds to the IFN-γ-activated GAS3 motif in the Il4ra locus to upregulate IL-4Rα expression.

IL-2 also upregulates IL-4RαC expression on activated CD4⁺ T cells in a STAT5-dependent manner, which enhances the capacity of IL-4 to induce signals in developing T_H2 cells. The activation of STAT5 by IL-2 results in the binding of STAT5 to the IFN-γ-activated motif GAS3 in the Il4ra locus during the early stage of T_H2 differentiation¹⁷ (Fig. 1). It should be noted that IL-4 itself also upregulates IL-4Rα expression, so the relative importance of IL-2 and IL-4 in enhancing the sensitivity of the developing CD4⁺ T cells to IL-4 almost certainly depends on the timing of the availability of these cytokines during the T_H2 differentiation process.

In vivo, differentiation of CD4⁺ T cells down the T_H2 pathway may proceed through either IL-4-dependent routes or IL-4-independent routes. Identification of the sources of IL-4 that initiate IL-4-dependent T_H2 differentiation in vivo has been of great interest for many years; among the cell types that have been considered are memory T_H2 cells, natural killer T cells, mast cells, basophils, eosinophils and the naive CD4⁺ T cells themselves¹⁸. It has been reported that basophils serve as T_H2-inducing professional antigen-presenting cells in vivo, given their ability to express functional MHC class II molecules and to produce a large amount of IL-4 (refs. 19–21). However, the methodology used to establish such a function for basophils involves depletion of these cells through the use of a monoclonal antibody to the FcεRI receptor for immunoglobulin E. It has been shown that this antibody results in the depletion of not only basophils but also a population of ‘inflammatory’ DCs expressing Fc∈RI that induce mainly T_H2 differentiation²². Furthermore, in another mouse model of depletion of basophils selectively in vivo in the offspring of the crossing of mice that express Cre recombinase in basophils (Basoph8 mice) with mice expressing diphtheria toxin α-chain from the ubiquitous Rosa26 locus (Rosa-DTα mice), the absence of basophils does not affect the ability of CD4⁺ T cells to produce IL-4 after infection with Schistosoma mansoni²³. Whether basophils are dispensable for all aspects of T_H2 differentiation remains to be established.

A second main determinant of T_H2 differentiation is the strength of signals generated by engagement of the TCR. In vivo immunization with a low concentration of antigen favors antibody production over delayed-type hypersensitivity²⁴. T_H1 differentiation is favored when agonist peptides are used for immunization, whereas altered peptide ligands, which interact with the TCR with lower affinity, favor T_H2 differentiation²⁵. Indeed, in a wide variety of mutant mice in which TCR signals are partially impaired, T_H2 differentiation is favored^26–28. In vitro differentiation of cells stimulated with low and high concentrations of peptide has shown that naive CD4⁺ T cells that receive weak TCR signals increase their expression of GATA-3 in an IL-4-independent manner at ∼14–24 hours after activation, whereas cells that receive strong TCR-mediated signals do not²⁹ (Fig. 2). The NF-κB1–Bcl-3 complex, the Notch-CSL pathway and the Wnt–β-catenin–TCF-1 pathway have each been proposed to serve a critical role in TCR-driven GATA-3 expression^30–34. However, gain or loss of function of these pathways, which are the modalities used to study the role of these molecules in this process, may lead to either abnormal thymic development or poor T cell proliferation and survival, even if T cell development seems phenotypically normal^35–37. Therefore, it is conceivable that the impairment of both GATA-3 expression and subsequent T_H2 differentiation may not be due to the direct regulation of GATA-3 expression by these pathways but may instead be secondary to either abnormal development or insufficient activation of CD4⁺ T cells derived from donors in which the genes encoding these molecules have been manipulated.

Figure 2.

A model for the regulation of early T_H2 differentiation by TCR signal strength. When naive CD4⁺ T cells receive strong TCR signals (right), prolonged and intense activation of the Erk pathway results in not only the failure to upregulate TCR-driven early expression of GATA-3 but also transient inhibition of IL-2-mediated activation of STAT5, at least during the first 24 hours after stimulation, despite the abundant IL-2 production and IL-2R expression. This defect in both GATA-3 upregulation and STAT5 activation leads to a lack of both early production of IL-4 and subsequent T_H2 differentiation. In contrast, when naive CD4⁺ T cells receive weak TCR signals (left), the degree of activation of the Erk pathway is not strong enough to suppress TCR-driven early expression of GATA-3 or to block STAT5 activation in response to small amounts of IL-2, which allows T cells to generate early production of IL-4 and to undergo subsequent T_H2 differentiation.

Naive CD4⁺ T cells that have received strong TCR signals not only fail to upregulate TCR-driven expression of GATA-3 but also do not activate STAT5 in response to IL-2 endogenously produced by activated CD4⁺ T cells themselves, at least during the first 24 hours after stimulation. This defect in both GATA-3 induction and STAT5 activation through the action of the kinase Erk pathway results in a failure to produce IL-4 or to undergo subsequent T_H2 differentiation²⁹ (Fig. 2). The physiological relevance of the TCR signal strength–mediated regulation of in vivo T_H1 differentiation versus T_H2 differentiation has been an open question. However, two studies have shown that omega-1, a T2 RNase derived from S. mansoni egg antigen, acts on DCs to suppress IL-12 production and to diminish the intensity of TCR-mediated signals that naive CD4⁺ T cells receive, which indicates that manipulation of DC function may result in weak TCR signals even if the antigen amount is not limiting and thus favor in vivo T_H2 differentiation^38,39.

Indeed, although IL-4 is essential for the differentiation of naive CD4⁺ T cells into T_H2 cells in vitro^14,40, its requirement for in vivo T_H2 differentiation has been a matter of intense debate, given the fact that mice deficient in either IL-4 or STAT6 show normal T_H2 differentiation in vivo in response to infection with Nippostrongylus brasilienis or Trichuris muris^41,42. In contrast, the deletion of Gata3 specifically in activated T cells results in the failure to undergo T_H2 differentiation and instead results in the appearance of IFN-γ-producing CD4⁺ T cells after infection with N. brasilienis⁴³, which indicates an indispensable role for GATA-3 in T_H2 development in vivo. Future investigations should focus on clarifying the mechanisms by which helminth infection regulates the induction of GATA-3 and suppression of transcription factors for differentiation into other helper T cell subsets during the early stages of the activation of naive CD4⁺ T cells. It will also be of particular interest to identify helminth-derived substances similar to S. mansoni egg antigen–derived omega-1 that may be responsible for lowering TCR-mediated signals.

T_reg cell development and function: role of IL-2, IL-7 and IL-15

Most thymocytes that express TCRs that recognize self peptide in the context of MHC molecules with very low or high affinity are eliminated by failure of positive selection or by negative selection in the thymus, respectively. However, some cells with high-affinity TCRs escape the negative selection process and thus are potentially able to mediate destructive autoimmunity. To prevent such autoreactive T cells from being activated in the periphery, the immune system has evolved several means, one of which involves T_reg cells, a specialized subset of CD4⁺ T cells with high expression of CD25 and the master regulatory transcription factor Foxp3 (ref. 44).

Similar to autoreactive T cells, thymus-derived T_reg cells, often called ‘natural T_reg cells’ (nT_reg cells), express a TCR with higher affinity for self peptide–MHC than that of TCRs on the bulk of conventional CD4⁺ T cells, which indicates that nT_reg cells escape clonal deletion by negative selection during thymic development⁴⁵. It has been proposed that there are two steps in the development of nT_reg cells, as follows: a fraction of immature CD4⁺ single-positive thymocytes with self-reactive TCRs of sufficiently high affinity gives rise to Foxp3⁻CD25⁺ T_reg precursor cells that then give rise to cells that express Foxp3 in response to cytokines of the γ_c family without further activation through the TCR⁴⁶. Indeed, mice deficient in IL-2, CD25 or IL-2Rβ have 50% fewer Foxp3⁺ thymocytes as wild-type mice have^47–49. Although the loss of IL-7 or IL-15 alone does not perturb the generation of thymic Foxp3⁺ cells, the combined elimination of IL-2, IL-7 and IL-15 leads to complete abrogation of Foxp3⁺ thymocytes, which indicates a substantial role for IL-2 and compensatory roles for IL-7 and IL-15 in nT_reg cell development^50,51.

Whether transforming growth factor (TGF-β) is required for the thymic development of nT_reg cells remains controversial^52,53. However, TGF-β is essential for the differentiation of peripheral naive CD4⁺ T cells into Foxp3⁺ cells with regulatory ability, called ‘induced T_reg cells’ (iT_reg cells)⁵⁴. IL-2 is required for TGF-β-mediated iT_reg cell differentiation in vitro⁵⁵. Interestingly, the vitamin A metabolite retinoic acid promotes TGF-β-mediated induction of Foxp3 expression in CD4⁺ T cells in vitro independently of the IL-2–STAT5 pathway⁵⁶. Moreover, retinoic acid derived from CD103⁺ DCs in the mesenteric lymph nodes and the lamina propria of the small intestine has an essential role in the conversion of naive CD4⁺ T cells into Foxp3⁺ T cells with suppressive activity^57–59. However, the proposal of a role for retinoic acid in generating iT_reg cells in the gut has been challenged by the observation that retinoic acid does not act as a cofactor for TGF-β-mediated induction of Foxp3 but is required for optimal activation of the TCR-proximal signaling cascades in CD4⁺ T cells⁶⁰.

How nT_reg cells suppress the immune responses of conventional T cells has been of great interest. It has been shown that nT_reg cells abrogate the induction of Il2 mRNA expression in CD4⁺CD25⁻ responder cells in vitro without affecting the initial activation of responder T cells, and that nT_reg cells require IL-2 for their ability to block Il2 transcription in the target T cells⁶¹. Given that nT_reg cells do not produce IL-2 in response to stimulation via the TCR, the source of IL-2 required for nT_reg cells to demonstrate suppressor activity may well be the conventional CD4⁺ T cells that will be the eventual targets of the activated T_reg cells. Interestingly, detailed kinetic analysis by IL-2-capture assay has shown that nT_reg cells do not begin to suppress IL-2 production by CD4⁺CD25⁻ responder cells until 6 hours of coculture, so nT_reg cells can receive IL-2 signals until then to activate their suppressor function⁶². However, that model has been challenged by the observation that under certain circumstances, nT_reg cells do not abolish IL-2 production by CD4⁺CD25⁻ responder cells but instead compete for IL-2 and for other cytokines that are essential for T cell survival in vivo, which leads to apoptosis of the responders owing to cytokine deprivation⁶³. Overexpression of the antiapoptotic protein Bcl-2 or loss of the proapoptotic protein Bim in CD4⁺CD25⁻ responder cells renders these cells resistant to nT_reg cell–mediated suppression⁶¹. Further investigation of these contradictory findings in terms of ‘abrogated IL-2 production’ versus ‘cytokine deprivation’ will provide better understanding of the mechanism through which nT_reg cells suppress the immunological responses of CD4⁺CD25⁻ T cells.

Generation of T_H17 and follicular helper T cells: role of IL-21

T_H17 cells are important in protection from bacterial and fungal infection and in the development of autoimmunity. It has been proposed that the following three steps control the differentiation of naive CD4⁺ T cells into T_H17 cells: differentiation induced by TGF-β and IL-6; IL-21-driven amplification; and IL-23-mediated stabilization^64,65. During the differentiation step, TGF-β and IL-6 in combination with stimulation via the TCR cause naive CD4⁺ T cells to express IL-23R and to induce the T_H 17 master regulatory transcription factor RORγt, and to produce IL-17A, IL-17F and IL-21 (refs. 66–69). Induction of IL-21 production depends on IL-6-driven activation of STAT3 and stimulation via the inducible costimulator ICOS^70,71. During the amplification step, IL-21 acts together with TGF-β to further upregulate IL-17 production and IL-23R expression in a STAT3-dependent manner. Genetic loss of either IL-21 or its receptor IL-21R results in less T_H17 differentiation both in vitro and in vivo^70,72,73. If mice are depleted of T_reg cells by treatment with antibody to CD25 before immunization, IL-21 is able to induce T_H17 responses in vivo even in the absence of IL-6, although to a lesser degree than if IL-6 is available. The cellular source of IL-21 in this setting remains unclear⁷². Thus, IL-21 seems to have an important role in the positive feedback regulation of T_H17 differentiation through its activation of STAT3. However, that conclusion has been challenged by a study reporting that mice deficient in either IL-21 or IL-21R are still able to mount T_H17 responses in vivo, which indicates a dispensable role for IL-21 in T_H17 differentiation when proinflammatory cytokines such as IL-6, IL-1 and TNF are abundantly available⁷⁴.

Follicular helper T cells (T_FH cells) promote T cell–dependent humoral immune responses by providing T cell help to B cells and thereby promote the formation of germinal centers, affinity maturation of antibody-secreting B cells and long-lived antibody responses⁷⁵. T_FH cells express the transcriptional repressor Bcl-6 as their master regulator^76–78 and the CXC chemokine receptor CXCR5 (refs. 79,80) and have high expression of the costimulatory molecules ICOS^79,80, PD-1 (ref. 81) and BTLA⁸², as well as the signaling adaptor molecule SAP⁸³, but downregulate their expression of the transcription factor Blimp-1 (ref. 77). It is still a point of considerable controversy whether T_FH cells originate directly from naive CD4⁺ T cells as a distinct subset, similar to T_H1, T_H2, T_H17 and iT_reg cells, or whether T_FH cells emerge from CD4⁺ T cells that have adopted a T_H1, T_H2 or T_H17 cell fate. One key issue about the acquisition of T_FH identity is clarification of the timing and mechanism by which responding CD4⁺ T cells acquire expression of Bcl-6 and CXCR5. CXCR5 expression is essential for the migration of activated CD4⁺ T cells to the follicles, where they interact with B cells that express the same cognate peptide presented by DCs to undergo maturation into functional T_FH cells.

Mice with T cell-specific deletion of Stat3 have a much lower frequency of CD4⁺CXCR5⁺ cells that arise in vivo in response to immunization with keyhole limpet hemocyanin in complete Freund's adjuvant, a phenotype that resembles that seen in deficiency in IL-6 or IL-21 (ref. 84). In contrast, mice deficient in IL-6 or IL-21 generate T_FH cells normally after infection with lymphocytic choriomeningitis virus (LCMV), which indicates a redundant role for these cytokines in the development of T_FH cells⁸⁵. Although the combined absence of IL-6 and IL-21 results in a failure to secrete antigen-specific immunoglobulin G after infection with LCMV, such mice have only slightly less generation of T_FH cells than that of their wild-type counterparts⁸⁶. As the dependence of LCMV-generated T_FH cells on STAT3 has not been determined, it is uncertain whether IL-6 and IL-21 may be replaced by another activator of STAT3 in this case or whether a STAT3-independent pathway exists for the generation of T_FH cells.

T_H17 and T_FH differentiation: role of IL-2

The IL-2–STAT5 pathway has been demonstrated to block T_H17 differentiation. When IL-2 is exogenously provided to T_H17-polarizing cultures, the generation of IL-17-producing cells is impaired, whereas there is an increase in the frequency of Foxp3⁺ cells⁸⁷. Naive CD4⁺ T cells from mice of the OT-II strain (with transgenic expression of an MHC class II–restricted, ovalbumin-specific TCR) that are deficient in recombination-activating gene 1 and are on the scurfy background, which lack functional Foxp3, still fail to undergo T_H17 differentiation under T_H17-polarizing conditions in the presence of exogenous IL-2 (ref. 88). STAT5 competes with STAT3 for binding to multiple sites in the Il17a-Il17f locus; binding of STAT5 to these sites is associated with repressive epigenetic marks across the Il17a promoter region and enhancer elements⁸⁸, which suggests a mechanism through which IL-2 directly represses T_H17 differentiation (Fig. 3).

Figure 3.

A model for early determination of iT_reg fate versus T_H17 fate controlled by TCR signal strength. During early T_H17 differentiation phase, naive CD4⁺ T cells require a combination of the cytokines IL-6 and TGF-β and costimulation (including ICOS), as well as strong TCR signals, to induce the expression of RORγt and T_H17 cytokines (right). When naive CD4⁺ T cells receive weak TCR signals under T_H17-polarizing conditions, the differentiation of Foxp3-expressing iT_reg cells is favored (left). Although weak TCR signals induce only small amounts of IL-2 production and IL-2R expression, IL-2-mediated STAT5 activation blocks IL-17A production and induces Foxp3 expression, which suppresses the induction of RORγt and thereby favors iT_reg differentiation. In contrast, despite abundant IL-2 production and IL-2R expression, transient inhibition of STAT5 activation by strong TCR signals leads to the failure to induce Foxp3 but allows expression of genes encoding RORγt and T_H17 cytokines and thereby favors T_H17 differentiation. Smad, signal-transducer protein(s) downstream of the receptor for TGF-β (TGF-βR).

Two days after infection with LCMV, T cells from SMARTA mice (which have transgenic expression of a TCR specific for LCMV glycoprotein) CD4⁺ develop into the following two subpopulations: CD25^hi cells, which have high expression of Blimp-1 and undergo a program of differentiation into effector T cells (T_eff cells); and CD25^int cells, which express Bcl-6 and CXCR5 and undergo differentiation into T_FH cells⁸⁹. The failure of CD25^hi cells to become T_FH cells suggests an inhibitory role for in vivo IL-2-generated signals in the development of T_FH cells. In addition, the IL-2–STAT5 pathway blocks the generation of T_FH cells by inducing Blimp-1, which results in the suppression of Bcl-6 expression^90–93 (Fig. 4). Interestingly, the crucial role of the IL-2–STAT5 pathway in the fate ‘decisions’ to develop into either T_eff cells or T_FH cells through regulation of the expression of the two mutually exclusive transcriptional regulators Blimp-1 and Bcl-6 resembles that observed for the differentiation of iT_reg cells and T_H17 cells, in which Foxp3 and RORγt reciprocally regulate each other (Figs. 3 and 4).

Figure 4.

A model for early determination of T_eff cell fate versus T_FH cell fate controlled by TCR signal strength. During the early phase of differentiation into the T_FH cell subset, naive CD4⁺ T cells require the cytokine IL-6 and costimulation (including ICOS), as well as strong TCR signals, to induce the expression of Bcl-6 (right). Although strong TCR signals induce abundant production of IL-2 and expression of the IL-2R complex, transient inhibition of STAT5 activation, presumably through the action of the Erk pathway, leads to the failure to express Blimp-1 and allows Bcl-6 expression induced by IL-6 and stimulation via ICOS and thereby results in differentiation into the T_FH cell subset. In contrast, T cells that have received weak TCR signals under T_FH cell–polarizing conditions can activate STAT5 in response to small amounts of IL-2, which induces Blimp-1 expression and suppresses Bcl-6 expression and thereby results in ‘preferential’ differentiation into the T_eff cell subset (left).

In addition to the mutual regulation of helper T cell differentiation as a result of the induction of distinctive transcription factors, TCR signal strength is also a key element in determining fate ‘decisions’ to develop into iT_reg cells or T_H17 cells, as well as T_eff cells or T_FH cells. If naive CD4⁺ T cells receive weak TCR signals, they do not differentiate into T_H17 cells and express Foxp3 instead, even though they are exposed to T_H17-inducing cytokines^94–96. Furthermore, the generation and function of T_FH cells depends on the strength of the binding of the TCR to a foreign peptide–MHC class II complex⁹⁷. Notably, strong TCR signals transiently inhibit IL-2-driven activation of STAT5, despite abundant production of IL-2 by activated CD4⁺ T cells, which thereby blocks T_H2 development^29,98. Thus, it is conceivable that the generation of T_H17 cells and T_FH cells may require strong TCR signals to block the IL-2–STAT5 pathway so that the inhibitory effects of IL-2 can be abrogated (Figs. 3 and 4). Clarifying the molecular basis that underlies TCR signal strength–mediated control of the IL-2–STAT5 pathway may provide better understanding of the delicate balance between helper T cell subset fates that are reciprocally regulated.

Concluding remarks

Cytokines of the γ_c family have crucial roles in the fate ‘decisions’ of naive CD4⁺ T cells to differentiate into distinct helper T cell subsets and in the function of these effector helper T cells. We have emphasized here the roles of cytokines of the γ_c family in this process through the activation of STAT proteins and the genes targeted by activated STAT proteins, particularly the so-called master regulatory transcription factors. Cytokines of the γ_c family signal through other signaling pathways as well, including activation of the metabolic checkpoint kinase mTOR. The two mTOR pathways, mTORC1 and mTORC2, have a unique function in influencing the fate of cells developing into distinct helper T cell subsets^99,100. Whether these differences can be accounted for by the action of the cytokines that determine the ‘choice’ of helper T cell phenotype is not yet clear but is an important area for continued study.

Comprehensive analysis of the cytokine dependence of in vivo helper T cell differentiation as well as the regulation of the change in phenotype of differentiated cells (‘plasticity’) is of great potential importance because this may provide opportunities for the development of drugs that can ensure that appropriate responses are mounted to given challenges and, possibly, to alter ‘inappropriate’ responses.