TCF-1: A maverick in T cell development and function

Abstract

The T cell-specific DNA binding protein TCF-1 is a central regulator of T cell development and function along multiple stages and lineages. Because it interacts with β-catenin, TCF-1 has been classically viewed as a downstream effector of the canonical Wnt signaling, although there is strong evidence for β-catenin-independent TCF-1 functions. TCF-1 co-binds accessible regulatory regions containing or lacking its conserved motif and cooperates with other nuclear factors to establish, context-dependent epigenetic and transcription programs, that are essential for T cell development and for regulating immune responses to infection, autoimmunity, and cancer. While it has mostly been associated with positive regulation of chromatin accessibility and gene expression, TCF-1 has the potential to reduce chromatin accessibility and thereby suppress gene expression. In addition the binding of TCF-1 bends the DNA and affects the chromatin conformation genome-wide. This review discusses the current understanding of the multiple roles of TCF-1 in T cell development and function and their mechanistic underpinnings.

Introduction

T lymphocyte development is a highly ordered stepwise process that depends on the cooperative and highly orchestrated action of multiple transcription and epigenetic regulators¹. These regulators organize cooperating complexes to establish the specific chromatin landscape and transcription profile required in each T cell lineage and developmental stage. The DNA binding protein TCF-1, encoded by the Tcf7 gene, has emerged as a central player in these processes^2–4. A recent review by Xue and colleagues elegantly discusses findings on TCF-1 across the fields of T cell immunity reflecting on the potential application of this knowledge to therapeutic intervention in viral infections and antitumor immunity⁵. The present review focuses on our current understanding of the molecular mechanisms through which TCF-1 leverages its diverse functions to shape T cell immunity.

TCF-1 is a member of the TCF/LEF family of high-mobility group (HMG) domain-containing proteins that have been conventionally viewed as effectors of the canonical Wnt signaling pathway (for reviews see^6–8). The Wnt cascade is activated in response to signals induced by the binding of extracellular Wnt ligands to Frizzled and LRP5/6 receptors on the cell surface. This results in the stabilization of β-catenin, which is then transported to the nucleus where it binds to TCF/LEF factors and promotes chromatin accessibility and gene expression. In the absence of Wnt signals, DNA-bound TCF/LEF factors interact with repressors of the Grg/TLE family that reduce chromatin accessibility and suppress transcription. This simplified view does not consider that in addition to the full-length TCF-1 protein, which can interact with β-catenin, TCF-1 also expresses isoforms that lack the β-catenin-interacting domain⁹. These short isoforms originally presumed to have dominant-negative regulatory functions¹⁰, are sufficient to support thymocyte maturation^11,12 and the generation of memory CD8⁺ T cells in response to acute infection¹³. The long isoform, on the other hand, supports thymocyte survival and is needed for the optimal maturation of central memory CD8⁺ T cells¹³. However, it is unclear whether the specific functions of the long isoform involve the canonical Wnt signaling, since ablating β-catenin, γ-catenin, or manipulating the expression and function of Wnt pathway components does not impair normal T cell development and function^14–18. Altogether these findings suggest that the functions of TCF-1 in T cell development are largely independent of the classical Wnt signaling.

By contrast, in leukemia, autoimmunity, and cancer, uncontrolled pathological activation of Wnt signaling in T cells, engages TCF-1 to promote aberrant developmental progression and transformation of thymocytes as well as immune imbalance^19–25. Stabilizing mutations of β-catenin and activation of the Wnt signaling pathway have been reported in human T cell malignancies, including precursor (T-ALL), peripheral (PTCL), cutaneous (CTCL), and adult T cell leukemia (ATL)^26–30. Similarly, conditional stabilization of β-catenin in mouse double-positive (DP) thymocytes induced leukemias with recurrent chromosomal translocations like the ones seen in human T-ALL, providing mechanistic validation of the findings in human leukemias^20,25,31. Importantly, we found that TCF-1 is responsible for the transformation of DP thymocytes with stabilized β-catenin since conditional ablation of TCF-1 in these cells abrogates leukemogenesis (submitted for publication). Furthermore, in chronic inflammatory conditions including Inflammatory Bowel Disease (IBD) and Multiple Sclerosis (MS) as well as in Colon Cancer, the regulatory T (T_REG) cells were found to express high levels of β-catenin and to acquire proinflammatory properties^22–24,32. Conditional stabilization of β-catenin in mouse T_REG cells induced an IPEX-like syndrome and experimental autoimmune encephalomyelitis (EAE) which models MS. Mechanistically, the pathologies were linked to β-catenin/TCF-1-mediated changes in chromatin accessibility and gene expression^23,24.

Altogether, these findings urge the need to better understand how TCF-1 leverages its Wnt-dependent versus Wnt-independent functions in T cells under physiological conditions and in the context of autoimmunity and cancer.

TCF-1 in T cell lineage specification and thymocyte differentiation

Notch activation enforces T cell specification on early thymic progenitors (ETPs) upon their entry into the thymus^33,34 (Fig 1). Tcf7 is among the first genes upregulated in ETPs in direct response to Notch signals^35–38. We and others have established that the specification of ETPs to the T cell lineage requires TCF-1 and its loss results in an early block of T cell development^35,36. TCF-1 positively regulates T lineage genes including Gata3, Il2ra, and Bcl11b leading to T cell commitment. In an elegant recent study, Rothenberg and colleagues validated these original findings by single-cell CRISPR perturbation of the genes encoding early acting transcription factors combined with single-cell RNA-seq³⁹. These analyses established that TCF-1 was profoundly needed for progression through the earliest phase of the ETP stage, while GATA-3 becomes especially important after the initiating role of TCF-1 in T cell specification. TCF-1 was further found to play a critical role in the γδ versus αβ T cell lineage separation. While the gradual upregulation of TCF-1 facilitates the αβ T cell fate after assembly of the pre-T cell antigen receptor (pre-TCR)^40,41, reduced TCF-1 expression favors the γδ T lineage fate^42,43. In the γδ T cell lineage TCF-1 also controls the γδ T cell effector fate, and essentially all TCF-1-deficient γδ cells convert into interleukin 17 (IL-17)-producing γδ effectors⁴³. More recent investigations using conditional CD4-Cre mediated ablation of Tcf7, which targets CD4⁺CD8⁺ DP thymocytes, revealed functions of TCF-1 in these later stages of thymic development. Our own studies established that TCF-1 co-binds DNA regulatory sites together with other nuclear factors including Ikaros, Runx1, HEB, and in particular it cooperates with HEB to establish the molecular profile of DP thymocytes⁴⁴. Following positive selection, the DP thymocytes face a lineage choice decision to the CD4⁺ or the CD8⁺ T cell lineage. This binary decision is guided by the transcription factors T_H-POK and RUNX3, respectively^45–47. Loss of TCF-1 impairs the CD4 lineage choice through insufficient T_H-POK induction and elevated RUNX3 expression⁴⁸. Thus, TCF-1 is a T cell specification and fate-determining factor at multiple stages of thymocyte development.

Figure 1: T cell development and the stages of TCF-1 implication.

Notch signaling upregulates Tcf7 expression in early ETPs ^35–39. TCF-1 induces the expression of genes encoding transcription factors critical for T cell specification, including Gata3 and Bcl11b. The levels of TCF-1 increases progressively up to the CD4⁺CD8⁺ (DP) stage. In CD4⁺CD8⁺ (DP) thymocytes TCF-1 cooperates with HEB to define their epigenetic landscape and trasncription profile ⁴⁴. Following thymic selection the DP cells become either CD4⁺ or CD8⁺ SP cells. TCF-1 fosters the CD4⁺ T cell fate by promoting Zbtb7b (T_H-POK) expression, and although TCF-1 is not required for commitment to the CD8⁺ T cell lineage it ensures CD8⁺ T cell stability by cooperating with RUNX3 to suppress Cd4 gene expression ⁴⁸. TCF-1 also suppresses Rorc (RORγT) and Il17 expression in DP thymocytes preventing their conversion to CD4⁺ T_H17 and CD8⁺ T_C17 cells ^50,113. After thymic egress CD4⁺ T cells differentiate into T_H subsets. TCF-1 promotes T_FH differentiation by inducing Bcl6 and supressing Prdm1 (BLIMP-1) expression and limits T_H1 differentiation by suppressing Tbx21 (T-bet) expression ^63–65. TCF-1 promotes T_H2 differentiation by directly upregulating Gata3 ⁵⁹. TCF-1 is expressed by stem cell-like CD27⁺ T_H17 cells, which persist in models of MS ⁵⁴. In T_REG cell subsets, TCF-1 cooperates with Foxp3 to limit expression of T cell activation and T_H17 differentiation genes. It also enhances Bcl6 expression to promote T_FR differentiation. CD8⁺ T cells are critical for defense against acute and chronic viral infections and cancers. In acute viral infection TCF-1 is important in driving the development of T_MP and T_MEM cells. Both in acute and chronic LCMV infection downregulation of TCF-1 is needed for the differentiation of T_EFF and T_EX cells. In chronic viral infections and cancer, TCF-1 and BCL-6 support the differentiation and maintenance of T_EX-_STEM cells that express PD-1 and can respond to anti-PD-1 therapy.

TCF-1 in differentiation of peripheral CD4⁺ T helper cell lineages

After thymic egress, the circulating naïve CD4⁺ T cells have the potential to differentiate into several T helper (T_H) cell lineages in response to antigen stimulation. These include the T_H1, T_H2, T_H17, and T follicular helper (T_FH) cell lineages that act to control the source of the foreign antigen⁴⁹. TCF-1 orchestrates the development, equilibrium, and function of all these CD4⁺ T cell lineages (for reviews see^5,50,51). T_H17 cells have critical roles in host defense against bacteria and specific pathogens and also drive pathogenic inflammation in autoimmunity and cancer⁵². Early studies suggested that germline deletion of Tcf7 resulted in increased IL-17 gene expression both in thymus and peripheral T cells, which led to enhanced Th17 differentiation ⁵³. However, these mice also have abnormall T-cell development due to their consitutive lack of TCF-1. Moreover, in the context of experimental autoimmune encephalomyelitis (EAE), TCF-1 expression was shown to mark a CD27⁺ T_H17 progenitor-like cell subset, which upon activation can give rise to a disease-promoting TCF-1^lo CD27⁻ T_H17 cell subset⁵⁴. T_H2 cells are critical for the immune response against extracellular parasites, and are activated by allergens and toxins⁵⁵. They can induce allergy and cooperate with T_H17 cells to mount tumor-promoting inflammation involving IL-33 and IL-10^56–58. TCF-1 promotes the T_H2 cell polarization by transactivating GATA-3, the signature regulator of the T_H2 cell lineage⁵⁹. T_H1 cells, differentiate in response to infection by intracellular pathogens, and have a shared transitional stage during their early differentiation steps with T_FH cells, which help B cells produce antibodies^60–62. TCF-1 controls the bifurcation between the T_H1 and T_FH cell lineages in favor of T_FH, by upregulating the BCL-6, which drives T_FH cell differentiation and downregulating BLIMP-1, which normally suppresses BCL-6 (Fig 1). BCL-6 limits T_H1 cell differentiation by directly suppressing the expression of critical T_H1 cell differentiation genes, including Tbx21 (encoding T-bet), which is the signature T_H1 lineage transcription factor^63–65. Thus, starting from a common naïve CD4⁺ T cell TCF-1 guides the maturation and the equilibrium of T helper cell lineages by selectively modulating the expression of key regulatory genes.

The specific roles of TCF-1 in T regulatory cells

Among CD4⁺ T cells, T_REG cells, have indispensable roles in resolving tissue inflammation and preserving tolerance to self-antigens^66,67. The majority of T_REG cells are generated in the thymus while a smaller group is converted in the periphery from naïve CD4⁺ T cells. The function of TCF-1 in T_REG cells is complex. Earlier studies using a germline Tcf7 knockout model showed that heterozygous deletion of Tcf7 increases the number of thymically generated T_REG cells⁶⁸, suggesting that TCF-1 negatively regulates T_REG cell development and T cell receptor affinity. Consistent with this finding, the T_REG cell lineage determining factor FOXP3 was reported to suppress Tcf7 gene expression⁵⁵. More recent studies demonstrated that TCF-1 interacts with Foxp3 and chronic activation of Wnt signaling impairs Foxp3 functions through TCF-1⁶⁹. Subsequently, several studies verified that in T_REG cells, TCF-1 shares a large number of DNA binding sites with Foxp3 and regulates the expression of FOXP3 target genes to fine tune T_REG properties^23,70,71. In line with these findings, we found that the targeted ablation of TCF-1 in T_REG cells enhanced their suppression of CD8⁺ T cell cytotoxicity but compromised their anti-inflammatory functions⁷². Interestingly, scRNA-Seq analysis identified numerous T_REG cell subsets and established that loss of TCF-1 has greater impact in the peripherally induced as compared to the thymic T_REG cells⁷². T_REG cell-specific ablation of both TCF-1 and LEF1 induced autoimmune pathologies⁷⁰, whereas loss of TCF-1 alone did not by itself render the mice unhealthy. However, in mice predisposed to polyposis the T_REG cell-specific loss of TCF-1 exacerbated pathogenic inflammation driving aggressive tumor growth⁷². Consistently, T_REG cells infiltrating human colorectal cancer tumors expressed lower levels of TCF-1 compared to those in the healthy margins or peripheral blood⁷². These findings suggest that TCF-1 alters T_REG cell functions in response to environmental cues to produce appropriate or, in the case of autoimmunity and cancer, inappropriate and pathogenic immune responses⁷². A distinct sub-category of T_REG cells are the T follicular regulatory (T_FR) cells that act in the germinal centers (GCs) to maintain immune homeostasis by suppressing excessive T_FH and B cell responses⁷³. T_REG cell-specific loss of TCF-1 diminished the number of T_FR cells and impaired immune regulation within GCs⁷⁰. This could contribute to the enhanced inflammation observed in mice with TCF-1-deficient T_REG cells^70,72. Thus, TCF-1 expression in T_REG and T_FH cells is essential to selectively regulate T_REG functions that control inflammation.

TCF-1 in CD8⁺ T cell survival and progenitor states

Expression of TCF-1 in CD8⁺ T cells distinguishes antigen-stimulated cells that have progenitor potential from their terminally differentiated couterparts. This distinction applies to T cells that are subjected to acute or chronic TCR stimulation.

TCF-1 is downregulated in mouse CD8⁺ T cells cells that respond to acute lymphocytic choriomeningitis virus (LCMV) infection, as they progress from naïve to terminally differentiated effector cells (T_EFF). However, expression of TCF-1 is maintained in a subset of T cell precursors with stem cell-like features⁷⁴, that differentiate to long-lived central memory cells (T_CM)^75–77. Several subsets of antigen experienced CD8⁺ T cells are defined based on their longevity, progenitor properties, and effector functions as well as expression of nuclear factors and cell surface markers (for review see⁷⁸). Hybrid phenotypes have also been described based on the expression of markers for migration and tissue residence (for review see⁷⁹). TCF-1 deficiency alone diminishes but does not ablate CD8⁺ T_CM cells, while loss of both TCF-1 and LEF1 almost completely eliminates the CD8 T_CM precursors (T_MP), suggesting that TCF1 together with LEF1 contribute to the T_MP cell phenotype⁸⁰.

In mice with chronic LCMV infection, persistent TCR signaling drives the downstream activation of transcription factors that promote the expression of inhibitory receptors, such as PD-1, LAG-3, and TIGIT, and reduce expression of KLRG1⁸¹, resulting in exhausted CD8⁺ T cells (T_EX). These transcription factors include the calcineurin-dependent nuclear factor of activated T cells (NFAT)⁵, interferon regulatory factor-4 (IRF4), basic leucine zipper transcription factor, ATF-like (BATF), nuclear receptor subfamily 4 group A (NR4A), and thymocyte selection-associated HMG BOX (TOX). TOX is the master regulator of the exhausted epigenetic state. T_EX cells were initially thought to be dysfunctional, however the loss of TOX also limits the percistence of functionally active antigen-specific CD8⁺ T cells, indicating that a subset of T_EX cells have progenitor properties^82–88. This subset has elevated expression of TCF-1 and controls viral spread in chronically infected mice. These observations are consistent with TCF-1 modulating the epigenetic outcomes of TOX activity.

The tumor-induced T cell dysfunction occurs in early stages of carcinogenesis and is driven by the presence and persistence of tumor antigen⁸⁹. Tumor infiltrating CD8⁺ T cells become increasingly dysfunctional with time, expressing high levels of genes associated with reduced immune function such as those encoding the transcriptional repressors Egr1, Batf, Blimp-1, and the inhibitory receptors PD1, LAG3, 2B4, and TIM3. Dysfunctional CD8⁺ T cells fail to produce effector cytokines in response to cognate antigen. By contrast, non-tumor-specific T cells that infiltrate the same tumor do not upregulate inhibitory receptors, and produce interferon-γ (IFN-γ) and tumor necrosis factor (TNF)⁸⁹. This finding suggests that chronic TCR stimulation by tumor antigens drives the dysfunctional CD8⁺ T cell phenotype. There is growing consensus that elevated expression of TCF-1 identifies the tumor antigen-specific CD8⁺ T cell subset that maintains long-term functional responses (T_EX-STEM) from the terminally differentiated T_EX subset^90–92 (for review see⁵). This concept may be oversimplified as discussed below.

TOX and TCF-1 are both members of the family of transcription factors that contain the conserved high mobility group box (HMG-box) region. TCF-1 partners with TOX to maintain an epigenetic state in CD8⁺ T_EX-STEM cells that supports long term survival and progenitor potential but not necessarily their functionality^83,85,93,94. Expression of TCF-1 is under the control of TOX. In the absence of TOX chronically stimulated CD8⁺ T cells maintain elevated levels of TCF-1 and fail to upregulate exhaustion markers. However, loss of TOX is not enough for expression of IFN-γ and TNF by tumor-infiltrating CD8⁺ T cells. This finding has led to the suggestion that the expression of exhaustion markers is uncoupled from the loss of effector functions, but rather serves to prevent the overstimulation of T cells and activation-induced cell death⁸⁵. This finding, brings into question the immediate benefits of targeting TOX or TCF-1 for immunotherapy of cancer⁹⁵. Further research is needed to establish the mechanisms responsible for functionall inactivity of tumor-infiltrating cells, as well as the similarities and differences between dysfunctional CD8⁺ T cells in cancer and exhausted T cells in chronic viral infection⁹⁶.

Functions of TCF-1 in the Wnt cascade and the option of Wnt independence

It is intriguing that TCF-1 is critically required in all stages of T cell development and maturation including various context-dependent T cell functions. The developmental stage specific and T cell lineage-specific transcriptional profiles and biological outcomes attributed to the action of TCF-1 raise intriguing questions about its’ Wnt dependent versus independent functions, and the roles of molecular partners of TCF-1 that help mediate diverse outcomes.

In the logic of the WNT cascade, DNA-bound TCF-1 enables directly interacting transcription and epigenetic factors to access specific genomic regions in order to regulate the chromatin landscape and transcription of the associated genes. In this scenario β-catenin interacts with the full-length TCF-1 protein in complex with epigenetic and transcription regulators that enhance chromatin accessibility and gene transcription (Fig 2a). In the absence of β-catenin, repressors of the Grg/TLE family, that bind TCF-1 at a region proximal to the HMG DNA binding domain^6,97, reduce chromatin accessibility, and suppress gene transcription (Fig 2a)⁹⁸. The extent to which β-catenin or repressors of the Grg/TLE family are involved in T cell development and function is uncertain and needs further study (for review see⁵). It is possible that other regulators, which can bind TCF-1 outside the context of Wnt signaling, also follow this scenario to modulate the chromatin landscape and transcription profile of the cells, but the identities of such regulators remain to be elucidated.

Figure 2. Understanding the fundamental molecular functions of TCF-1.

TCF-1 has multiple molecular functions that are superimposed on its’ ability to bend the DNA upon binding and regulate the 3D chromatin conformation.

(a) In the absence of β-catenin, TCF-1 can bind with the repressors of the Grg/TLE family at a region proximal to the HMG DNA binding domain ^6,97 to reduce chromatin accessibility, and suppress gene transcription ⁹⁸. Extracellular Wnts stabilize β-catenin by disrupting its’ degradatioin complex, allowing its’ nuclear translocation and interaction with the full-length TCF-1 protein. This results in the recruitment of epigenetic and transcription regulators that enhance chromatin accessibility and gene transcription. (b) TCF-1 cooperates with other transcription factors to establish context and developmental stage-specific epigenetic and transcription profiles through co-binding to common DNA sites. (c) TCF-1 could function as a pioneer-like factor and initiates changes in the chromatin landscape that establish the T cell lineage identity¹⁰¹. (d,e) TCF-1 may function as a “Place Holder”, to promote/maintain T cell-specific chromatin accessibility, independently of Wnt signals. (f) TCF-1 has intrinsic histone deacetylase activity (HDAC) and can directly reduce chromatin accessibility.

Cooperation of TCF-1 with other transcription and epigenetic regulators

In addition to its direct interaction with regulators, TCF-1 also cooperates with other transcription factors to establish context and developmental stage-specific epigenetic and transcription profiles through co-binding to common DNA sites. Regulation through co-binding is supported by findings that accessible chromatin sites along thymocyte development are not only enriched for the TCF/LEF motif, but also motifs for ETS, RUNX and E2A families of transcription factors ^44,99. Comparing the genome-wide binding of TCF-1 to that of IKAROS, RUNX1 and HEB in DP thymocytes demonstrated an extensive overlap between TCF-1-occupied sites and sites bound by these transcription factors. Detailed studies focusing on the co-binding of TCF-1 and HEB, showed that TCF-1 cooperates with HEB to establish and maintain the epigenetic and transcription profile of DP thymocytes (Fig 2b)⁴⁴. Furthermore, in T_REG cells TCF-1 binding significantly overlaps with the binding of Foxp3^23,71. We found that this overlapping binding limited the expression of genes involved in T_H17 inflammation, transforming growth factor-β (TGF-β) signaling, and T cell activation⁷². The genetic ablation of Tcf7 led to the upregulation of these genes without compromising the core T_REG gene expression signature, and even upregulated Foxp3 expression ⁷². The interaction of TCF-1 with Foxp3 was further shown to involve Foxp3-mediated downregulation of Tcf7⁷¹. Thus, these new findings establish that TCF-1 engages transcription factors from several different families as its cooperating partners to shape the molecular profiles of T cells during development and function.

TCF-1 preserves chromatin accessibility

It has been suggested that in T cell development TCF-1 promotes or preserves chromatin accessibility^44,71,99,100. Vahedi and colleagues found that the upregulation of TCF-1 in early thymocytes undergoing T cell specification, coincides with increased accessibility of chromatin sites that are enriched in the conserved TCF-1 DNA binding motif⁹⁹. Also during hematopoietic development, from hematopoietic stem cells to CD4⁺ and CD8⁺ single-positive (SP) thymocytes, the TCF-1 binding motif becomes progressively more enriched at accessible chromatin sites. The increase in chromatin accessibility parallels the progressive upregulation of Tcf7 gene expression in developing thymocytes⁹⁹. These findings led to suggestions that TCF-1 may act as a pioneer-like factor¹⁰¹ that initiates the T cell-specific chromatin landscape and establishes the T cell lineage identity (Fig 2c). A detailed molecular analysis of TCF-1 sufficient and deficient lymphoid precursors cultured for a short time in T cell differentiation conditions (as described in³⁹) could help clarify the precise role of TCF-1 in T cell specification and may provide direct support for the role of TCF-1 as a pioneer factor. Other studies showed that loss of TCF-1 in DP thymocytes⁴⁴, or in CD8⁺ T cells¹⁰⁰, resulted in overall reduced chromatin accessibility at sites previously bound by TCF-1. At least in DP thymocytes, these accessibility changes were more evident in TCF-1 bound enhancer sites that contained its conserved motif, and loss of TCF-1 correlated with an overall reduction of target gene expression ⁴⁴. Along this line, Schietinger and colleagues showed that the transition of CD8⁺ T_EX-STEM tumor-infiltrating lymphocytes (TILs), which express TCF-1, into terminal CD8⁺ T_EX TILs, which do not, is associated with extensive changes in chromatin accessibility. Sites that progressively lose accessibility during this process are highly enriched for the TCF/LEF binding motifs ⁹³, suggesting that TCF-1 sustains the open chromatin state in T_EX-STEM TILs (Fig 2d), and can serve as a “place holder” for other transcription factors (Fig 2e). Similarly, following CD4⁺ and CD8⁺ T cell activation, TCF-1 is downregulated and its canonical binding sites lose accessibility. Exceptions are TCF-1 sites that are co-bound by Ets1, or by activation induced transcription factors, indicating that TCF-1 may act as a place-holder in these sites¹⁰².

Collectively, these findings suggest that TCF-1 predominantly maintains and potentially also promotes chromatin accessibility in T cell development and support the notion that this function underlies its ability to maintain some differentiation potential or “stemness” in T cells. These functions are likely independent of Wnt signals since in contrast to the critical need for TCF-1, β-catenin has minimal impact on normal T cell development.

TCF-1 directly modifies the chromatin through its intrinsic HDAC activity

Deciphering how TCF-1 functions in T lymphocytes has become more challenging by the finding of Xue and colleagues that TCF-1, in addition to shaping the chromatin landscape through its interacting partners it also has an intrinsic histone deacetylase activity (HDAC). This activity, which was mapped to a region between the N terminal β-catenin binding domain and the central HMG DNA-binding domain directly upstream of the Grg/TLE binding domain, can directly reduce chromatin accessibility (Fig 2f). It has been suggested that the TCF-1 HDAC activity is essential for establishing the CD8⁺ T cell identity⁹⁸ and for sustaining the ability of T_FH cells to provide B cell help¹⁰³. Therefore, in order to determine the context-dependent functions of TCF-1 on chromatin accessibility, it is important to understand how its intrinsic HDAC activity orchestrates with the activities of regulators that directly interact with TCF-1 particularly the ones that promote chromatin accessibility.

TCF-1 shapes the 3D chromatin conformation

The HMG domain family of proteins, including TCF-1, bind to the minor groove of the DNA helix and have the capacity to bend the DNA (reviewed in¹⁰⁴). Early studies by Grosschedl and colleagues established that LEF-1, the relative of TCF-1 that is also expressed in T cells induces a significant DNA bend to its binding site^105,106. In particular, LEF-1, likely through its ability to bend the DNA upon binding, facilitated interactions between proteins bound at nonadjacent sites and coordinated the assembly of a complex at the enhancer of the TCRα gene. Based on the similarity of their HMG domain, it is expected that TCF-1 also bends the DNA upon binding, and this could have significant implications on the ability of TCF-1 to regulate the 3D chromatin conformation (Fig 2). Newly developed technologies have made it possible to assess chromatin conformation changes genome-wide and associate them with lineage commitment stages. In this context, a comprehensive study combining high throughput chromatin conformation capture (HiC), with chromatin accessibility, and gene expression analyses, established that T cell commitment at the DN2 to DN3 thymocyte stage is associated with major chromatin conformation changes ¹⁰⁷. Although this study did not implicate TCF-1, more recently Xue and colleagues integrated Hi-C with epigenetic and transcription data, to show that TCF-1 and LEF-1 promote the formation of extensively interconnected hubs by enforcing chromatin interaction and accessibility in CD8⁺ T cells ¹⁰⁰. These findings open the way for further studies to independently assess the role of TCF-1 versus LEF-1 on chromatin conformation and to compare their differential impact in the various T cell lineages.

TCF-1 leverages its functions through abundance

The expression levels of TCF-1 in developing T cells are highly regulated. TCF-1 starts expressing at low levels in ETP thymocytes, and its progressively upregulated up to the DP stage where it reaches an expression level that is unusually high for a transcription factor. Past the DP stage TCF-1 expression levels are reduced but remain relatively high in all peripheral T cell lineages with the exception of the T_REG cells, which express low levels of TCF-1. Polarization of CD4⁺ T cells to the T_H lineages is invariably marked by downregulation of TCF-1. Similarly, progression of naïve CD8⁺ cells to the terminal effector or terminal exhaustion stages is associated with downregulation of TCF-1, while the progenitor exhausted CD8+ T-cells that are long lived and have progenitor propertors express high levels of TCF-1.

Recent studies have highlighted the molecular impact of TCF-1 downregulation. In particular, the reduced expression of TCF-1 after CD4 and CD8 T cell activation, was suggested to underlie the observed accessibility loss of sites uniquely bound by TCF-1¹⁰². The increased generation of thymic T_REG cells after heterozygote TCF-1 deletion⁶⁸, and the downregulation of TCF-1 by FOXP3, which reduces chromatin accessibility and associated gene expression⁷¹ in T_REG cells, further support the suggestion that the levels of TCF-1 determine its functions. TCF-1 is also downregulated in colon tumor infiltrating T_REG cells, indicating that T_REG cells adapt to their environment by regulating the levels of TCF-1, which in this case contributed to their enhanced tumor promoting properties ⁷². Therefore, variations in TCF-1 levels can significantly impact the interplay between TCF-1 and other molecular partners affecting key biological processes including differentiation, cell fate decision, and function in health and disease.

Future perspectives

Deciphering how TCF-1 orchestrates its diverse functions during T cell development represents a challenging puzzle. This challenge is because TCF-1 has essential roles in multiple T cell lineages and developmental stages and its activity is context dependent. Therefore, any effort to dissect the molecular functions of TCF-1, must take into consideration in each T cell lineage and developmental stage: 1) the epigenetic and transcription profile of the cells, 2) the cooperating and interacting partners, and 3) the physiological levels of TCF-1. An additional parameter to consider in each context is the redundant functions of the transcription factor LEF1, the WNT responsive close relative of TCF-1 that is expressed in all T cells and shares significant structural homology with TCF-1. TCF-1 and LEF1 share the ability to bind β-catenin and Grg/TLE factors, and they both express shorter isoforms that do not bind β-catenin. The genome-wide DNA binding of LEF1 overlaps with that of TCF-1 in cells where it has been assessed, including DP thymocytes⁴⁴ and T_REG cells⁷⁰. Simultaneous loss of both TCF-1 and LEF1 in many T cell lineages, and stages mainly strengthens the phenotypes observed by simply deleting TCF-1^48,63,70. It is even puzzling why the loss of TCF-1 has so much stronger impact on T cell development than the loss of LEF1 and it is unclear whether this is due to the lower levels of LEF1 expression, like in DP thymocytes, or to yet unknown functional differences. Although studies up to now have mainly identified functional similarities between TCF-1 and LEF-1 in T cell development it would be interesting to uncover their differences and the molecular basis of their functions.

Future investigations can take advantage of novel single cell and genome wide technologies. For example, single-cell technologies including scCrisprCAS9 perturbations³⁹, scRNAseq and scATACseq have already provided a wealth of information on the roles of TCF-1 in T cell specification⁹⁹, on T_REG cell heterogeneity and on tumor-infiltrating T cells^72,108–112. Similarly, integrating genome-wide transcription, epigenetic and chromosome conformation data is beginning to more accurately predict molecular functions of TCF-1. Addressing the molecular functions of TCF-1 will also require new model systems in which TCF-1 expression and/or protein levels can be temporally modulated in specific T cell developmental stages and lineages. Given that TCF-1 regulates multiple T cell properties, knowing how it functions in each situation will also provide the knowledge base to design context-specific therapeutic interventions in autoimmunity and cancer.