Navigating double negatives: new pathways for regulating T_FH differentiation

The transcription factors Foxp1 and Foxo1 inhibit the differentiation of follicular helper T cells. On the other hand, the E3 ligase Itch targets Foxo1 for degradation to promote such differentiation.

Help provided by T cells is necessary for efficient B cell responses. Investigation of helper T cell differentiation has so far focused largely on identifying positive regulators of each subset and how they antagonize each other to stabilize lineage commitment. However, studies by Wang et al.¹ and Xiao et al.² in this issue of Nature Immunology demonstrate that negative regulators also have an important role in the differentiation of helper T cells.

Cognate help from T cells is required for B cells to undergo affinity maturation and generate B cell memory, which happens mainly in a specialized lymphoid structure: the germinal center. Follicular helper T cells (T_FH cells) are a subset of helper T cells specialized to support the proliferation and selection of B cells and their differentiation in germinal centers^3,4. Sufficient T_FH cell function is required for robust antibody responses to control infections, whereas excessive T_FH cell activity promotes the production of autoantibodies and can lead to autoimmune diseases^5,6,7.

The differentiation of T_FH cells is induced by antigen stimulation in synergy with costimulation from CD28, CD40L, the inducible costimulator ICOS, receptors of the SLAM ('signaling lymphocyte activation molecule') family and a cytokine milieu that includes interleukin 21 (IL-21), IL-6 and IL-12. This differentiation program is directed by the key transcription factor Bcl-6 and is facilitated by several other transcription factors, including STAT3, c-Maf and Ascl2 (refs. 4,7).

In contrast to the considerable body of knowledge about the positive regulators of T_FH cell differentiation, understanding of the negative regulation of this process is limited. IL-2–STAT5 signaling has been reported to suppress T_FH cell differentiation, probably through a mechanism that induces expression of the transcriptional repressor Blimp-1, which subsequently antagonizes Bcl-6 function⁷. Now Wang et al. and Xiao et al. identify previously unknown pathways that suppress T_FH cell differentiation^1,2. These new reports substantially extend the knowledge of the sophisticated molecular program that regulates T_FH cell differentiation. Indeed, while Wang et al. study the involvement of the transcription factor Foxp1 in negatively regulating T_FH cell differentiation¹, Xiao et al. investigate the mechanism by which a second negative regulator, the transcription factor Foxo1, is itself inhibited by via its degradation by the E3 ubiquitin ligase Itch² (Fig. 1a).

Figure 1: Dynamic regulation of T_FH cell differentiation by negative regulators.

Debbie Maizels/Nature Publishing Group

(a) Foxp1 suppresses T_FH cell differentiation by directly inhibiting IL-21 expression and indirectly limiting ICOS expression, presumably through the MEK-Erk pathway. Foxp1A is constitutively expressed, while Foxp1D expression is transiently induced by signaling via the TCR and costimulation. Foxo1 also suppresses T_FH cell differentiation by a mechanism that remains to be elucidated. ICOS signaling activates the phosphatidylinositol-3-OH kinase (PI(3)K)-Akt pathway, which induces phosphorylation (red 'P') of Foxo1 and its export from the nucleus. Itch in the cytosol then targets Foxo1 for ubiquitination and degradation and promotes T_FH cell differentiation. (b) Bcl-6 is a positive regulator that directs T_FH cell differentiation. Bcl-6 expression is upregulated during the priming of T cells by dendritic cells (DCs). Cognate interaction with B cells induces a second wave of Bcl-6 upregulation for the terminal differentiation of T_FH cells. Foxp1A and Foxp1D are negative regulators of T_FH cell differentiation. Foxp1A is constitutively expressed during T_FH cell differentiation, while Foxp1D expression is transiently induced during T cell priming. Foxo1 is a negative regulator of T_FH cell differentiation and its expression decreases during this differentiation. Itch targets Foxo1 for degradation after Foxo1 is exported to the cytosol and is therefore a positive regulator of T_FH cell differentiation. The expression of Itch persists during T_FH cell differentiation. GC, germinal center.

Foxp1, which belongs to the forkhead box ('Fox') family of transcription factors, maintains the quiescence of naive T cells⁸. Foxp1-deficient CD4⁺ or CD8⁺ T cells in the periphery spontaneously acquire an activated phenotype, increase their proliferation and rapidly produce cytokines upon stimulation via the T cell antigen receptor (TCR). However, as this prevents investigation of the role of Foxp1 in peripheral CD4⁺ T cell responses, Wang et al. use a strategy in which deletion of Foxp1 is induced by tamoxifen¹. After mice are immunized with protein antigens, naive CD4⁺ T cells in which Foxp1 has been deleted 'preferentially' differentiate into T_FH cells. The increased T_FH differentiation of Foxp1-deficient cells supports considerably enhanced formation of germinal centers and leads to more production of antigen-specific isotype-switched antibodies.

Notably, among the four alternative splice isoforms of Foxp1 (Foxp1A–Foxp1D), Foxp1A is constitutively expressed in CD4⁺ T cells and Foxp1D is the only isoform induced upon stimulation with antibody to the invariant signaling protein CD3 (anti-CD3) and anti-CD28. In mice with transgenic expression specifically of Foxp1A or Foxp1D, overexpression of either isoform suppresses T_FH cell differentiation. Furthermore, deletion of one copy of Foxp1 (which results in reduced expression of both Foxp1A and Foxp1D) or selective deletion of Foxp1D enhances T_FH cell differentiation, although it is less pronounced than that achieved by complete deletion of Foxp1. These observations allow Wang et al. to propose that Foxp1 serves as a rate-limiting 'double-check' negative regulator for T_FH cell differentiation: sustained Foxp1A is a homeostatic limiting factor for T_FH cell differentiation, and induced Foxp1D is a 'gatekeeper' for such differentiation¹.

To delineate the mechanism of Foxp1-mediated negative regulation of T_FH cells, Wang et al. compare the gene-expression profiles of wild-type and Foxp1-deficient CD4⁺ T cells and find that IL-21 expression is increased in Foxp1-deficient CD4⁺ T cells¹. Foxp1-deficient CD4⁺ T cells produce more IL-21, and chromatin immunoprecipitation reveals that Foxp1 directly binds to the Il21 promoter region, which suggests that Foxp1 directly targets Il21 to inhibit its expression. In addition, Foxp1-deficient CD4⁺ T cells have higher expression of ICOS than do wild-type cells upon activation, although the expression of Icos mRNA is unchanged. Activation of the kinases MEK and Erk has been shown to amplify ICOS expression⁴, and MEK-Erk signaling is enhanced in Foxp1-deficient T cells⁸. Therefore, Foxp1 acts to limit the expression of IL-21 (directly) and ICOS (indirectly) to suppress T_FH cell differentiation.

Foxp1 is not the only member of the Fox family of transcription factors that has been shown to suppress T_FH cell differentiation. Conditional deletion of Foxo1 in CD4⁺ T cells leads to enhanced T_FH cell differentiation and spontaneous germinal center formation in mice^9,10. Although how Foxo1 regulates T_FH cell differentiation remains elusive, Xiao et al. demonstrate that Itch targets Foxo1 for ubiquitination and degradation². By downregulating this negative regulator, Itch promotes T_FH cell differentiation.

After infection with vaccinia virus, Itch-deficient mice have fewer T_FH cells, germinal centers and plasma cells and lower titers of virus-specific antibodies than do wild-type mice. The phenotype is entirely recapitulated in mice with conditional deletion of Itch in T cells, which suggests that Itch regulates the antibody response in a T cell–intrinsic manner. Itch is required in CD4⁺ T cells for T_FH cell differentiation from its early stage. In further studies, Xiao et al. infect mice with lymphocytic choriomeningitis virus (LCMV) and track CD4⁺ T cells expressing a TCR specific for LCMV in the infected mice; they find that T_FH cell differentiation is severely impaired from day 3 after infection². The ICOS–Bcl-6 axis 'instructs' early T_FH differentiation^3,4,7. ICOS upregulates Bcl-6 expression to promote T_FH differentiation through the phosphatidylinositol-3-OH kinase (PI(3)K)–kinase Akt pathway⁴. How activation of PI(3)K-Akt induces Bcl-6 remains unknown. Interestingly, Itch deficiency in CD4⁺ T cells substantially reduces Bcl-6 expression without impairing ICOS expression, which suggests that Itch acts downstream of ICOS. Identifying the function of Itch might help to reveal the regulatory circuit of ICOS–PI(3)K–Akt–Bcl-6.

Itch has been shown to target the transcription factors JunB and c-Jun for degradation to inhibit differentiation into the T helper type 2 subset of helper T cells¹¹. Xiao et al. establish that Itch targets Foxo1 for ubiquitination and degradation². Coimmunoprecipitation analyses show that Itch directly interacts with Foxo1. In addition, overexpression of Itch enhances the ubiquitination of Foxo1. Notably, stimulation with anti-CD3 and anti-ICOS enhances the Foxo1-Itch interaction and subsequently induces degradation of Foxo1. The induced degradation of Foxo1 is forfeited in Itch-deficient cells. This elegant study reveals how ICOS regulates Foxo1 expression. Although Itch is constitutively expressed in the cytosol of CD4⁺ T cells, it mediates the degradation of Foxo1 only upon signaling via the TCR and ICOS, presumably through a mechanism of nuclear-to-cytosolic translocation of Foxo1, which then 'opens Foxo1 up' to Itch-mediated degradation. Signaling via the TCR and ICOS activates Akt, which has been reported to phosphorylate Foxo1 to trigger its export from the nucleus¹⁰. Finally, a functional study proves that genetic ablation of Foxo1 restores the ability of Itch-deficient CD4⁺ T cells to differentiate into T_FH cells and to support the formation of germinal centers.

Signaling via the TCR and ICOS is the main driving force for T_FH cell differentiation, but the mechanism of this action remains a puzzle⁴. With the key pieces identified by these two studies, the whole picture should be revealed soon. Coincidentally, both studies demonstrate that signaling via the TCR and costimulation induces T_FH cell differentiation by removing negative regulation. Naive CD4⁺ T cells have high expression of the negative regulators Foxo1 and Foxp1A. Foxp1A is a homeostatic controller that prevents T cell activation and remains at steady levels during T_FH cell differentiation¹. In contrast, Foxo1 expression is reduced soon after activation, and this is required for the first wave of T_FH cell differentiation². Foxo1 decreases gradually throughout T_FH cell differentiation, and mature T_FH cells have a low abundance of Foxo1, an effect that now seems to occur mainly via Itch-mediated degradation. Although there is high expression of Itch even in naive T cells, Itch mediates Foxo1 degradation only upon stimulation via the TCR and ICOS. The expression of Foxp1D is low in naive T cells and is transiently induced by priming¹. Foxp1D might work in synergy with Foxp1A to limit cell activation and ICOS expression. Intriguingly, after reaching its peak at the priming stage, Foxp1D expression decreases in a manner similar to that of Foxo1 expression. Such a reduction in Foxp1D is considered to be induced by signaling via the TCR and costimulation through cognate interactions with B cells. Removal of the negative regulator Foxp1D probably promotes the production of IL-21 and terminal differentiation of T_FH cells. Together these new reports represent a substantial advance in the understanding of the mechanism that regulates two waves of T_FH cell differentiation (Fig. 1b).

Several interesting questions are raised by these studies. Are these regulatory mechanisms unique to T_FH cells and not present in other helper T cell subsets? How does Foxo1 suppress T_FH cell differentiation? Why does signaling via the TCR and costimulation regulate Foxp1D expression but not Foxp1A expression? Discoveries of previously unknown regulatory circuits have highlighted negative regulation as an important checkpoint in T_FH cell differentiation. Targeting pathways of negative regulation, such as CTLA-4 or PD-1, has achieved great success in cancer immunotherapy. Researchers can also now start to consider how to apply the new knowledge acquired from these papers to potentially modulate T_FH cell function to treat autoimmune disease, enhance vaccination and control chronic infection⁷.