The multifaceted roles of TCF1 in innate and adaptive lymphocytes

Abstract

The immune system requires a complex network of specialized cell types to defend against a range of threats. The specific roles and destinies of these cell types are enforced by a constellation of gene regulatory programs, which are orchestrated through lineage-specifying transcription factors. T Cell Factor 1 (TCF1) is a central transcription factor in many of these programs, guiding the development and functionality of both adaptive and innate lymphoid cells. This review highlights recent insights into the function of TCF1 in a variety of lymphoid cell subsets and its potential for translational applications in immune disorders and cancer.

1. Introduction

The mammalian immune system is characterized by dozens of specialized cell types, which are required to provide both broad and specific defense against a veritable universe of threats. Innate lymphoid cells (ILCs) are tasked with broad and rapid protection, responding quickly to non-self or altered-self dangers, stemming the tide until an antigen-specific adaptive immune responses can develop to precisely eliminate the threat. The diversity of cell types and the range of responses in each depends on how external and internal cues coordinate programs for gene regulation, which are governed by the interplay between transcription factors and their revision of epigenomic landscapes (Hosokawa & Rothenberg, 2021; Korchagina et al., 2023).

Among the complex set of transcription factors, T Cell Factor 1 (TCF1), has emerged as a critical regulator of adaptive T cells and their innate counterparts, ILCs. TCF1, encoded by the Tcf7 gene, is critical for both the commitment and development of these lineages, especially in mature cells, where it regulates the balance between stem-like self-renewal versus effector functions. Here, we provide an overview of the key roles played by TCF1 in fate determination and responses of innate and adaptive lymphocytes, and briefly discuss the translational potential of harnessing TCF1 regulatory programs.

2. The TCF/LEF family and modes of gene regulation

The TCF/LEF family of transcription factors features four members: TCF1, LEF1, TCF7L1 (formerly known as TCF3), and TCF7L2 (formerly TCF4) (Cadigan & Waterman, 2012). Early in embryonic development, these factors exhibit a broad expression pattern and appear to cooperate in proper tissue development. In contrast, postnatally, TCF1 and LEF1 are mostly restricted to the hematopoietic lineage (Galceran et al., 1999; Oosterwegel et al., 1993).

TCF/LEF family genes feature multiple promoters that, together with alternative splicing, give rise to multiple isoforms, categorized as either long or short based on their composite functional domains (Fig. 1) (Arce et al., 2006). Full-length long isoforms contain an N-terminal β-catenin interacting domain, allowing them to bind with this co-activator and act as the nuclear effector of the Wnt signaling pathway (Nusse & Clevers, 2017). In the absence of stabilized β-catenin, TCF/LEF family members often act as repressors via interaction with the Groucho-related/TLE family of co-repressors (Brantjes et al., 2001). Short isoforms lacking the β-catenin interacting domain were initially thought to act as dominant negative inhibitors of the full-length isoforms. However, elegant studies by the Xue group showed that short isoforms have an intrinsic histone deacetylase (HDAC) activity, which can function to repress target genes (Xu et al., 2017). All TCF/LEF isoforms can bind DNA through their high mobility group (HMG) domain, with the “Wnt responsive element”, TYYCTTTG-ATSTT, as a consensus binding motif (Atcha et al., 2007; Oosterwegel et al., 1991; Wetering et al., 1992). In addition to this common HMG domain, some TCF1 and TCF7L2 isoforms feature a 30 amino acid cysteine clamp motif that further stabilizes DNA interactions, allowing enhanced binding to an extended set of recognition sites (Atcha et al., 2007).

Fig. 1. Common functional domains of TCF/LEF transcription factors.

TCF/LEF transcription factors feature common functional regions such as a β-catenin interaction domain, an HDAC domain, and a DNA-binding HMG domain. Alternative promoters and splicing yield two distinct classes: full-length long isoforms containing all domains and short isoforms lacking the β-catenin interaction domain.

In addition to their intrinsic HDAC activity and ability to regulate gene expression via their interaction with co-activators (β-catenin) or co-repressors (Groucho-related/TLE), TCF/LEF members have clearly been shown to regulate three-dimensional organization of the genome. Mechanistically, this is achieved partly by the intrinsic ability of HMG domains, which introduce sharp 130° bends in DNA (Giese et al., 1992; Grosschedl et al., 1994; Love et al., 1995), as well as indirectly via association with other chromatin architecture proteins such as CTCF and the cohesin-loading factor NIPBL (Shan et al., 2021, 2022; Wang et al., 2022).

3. TCF1 in T cell fate commitment and thymocyte development

TCF1 and LEF1 are critical regulators of the T cell developmental program in the thymus, coordinating gene expression changes at multiple stages (Zhao et al., 2022). Common lymphoid progenitor (CLP) cells entering the thymus from the bone marrow commit to a T cell fate by undergoing a stepwise process of maturation that can be characterized into distinct stages based on surface marker expression (Rothenberg, 2014). Immature thymocytes initially lack expression of both CD4 and CD8 and are classified into four general double negative (DN) stages based on their levels of CD25 and CD44. The earliest thymic progenitors are the CD25⁻CD44⁺ DN1 stage, progressing into CD25⁺CD44⁺ DN2 thymocytes as they become more committed to the T cell lineage. Fully committed T cells are identified as the CD25⁺CD44⁻ DN3 stage, where they begin rearranging the T cell receptor beta gene (Tcrb). Productive rearrangement of Tcrb enables signaling through the pre-TCR (Ptcra gene), which is necessary for rapid proliferation at the CD25⁻CD44⁻ DN4 stage, preceding the upregulation of CD4 and CD8 to form double positive (DP) thymocytes. Following selection steps verifying appropriate TCR reactivity, these DP cells mature into CD4⁺ or CD8⁺ single positive cells that exit the thymus to monitor antigens in the periphery.

Although bone marrow CLPs develop in Tcf7^−/− mice and can migrate to the thymus, they struggle to advance through the DN1 phase. This dramatically reduces cell counts throughout all stages of thymic development, culminating in an approximately 100-fold decrease in total thymocyte numbers (Germar et al., 2011; Schilham et al., 1998; Verbeek et al., 1995). Although inefficient, some TCF1-deficient thymocytes can complete T cell development, and Tcf7^−/− mice do retain a population of mature peripheral T cells. The leakiness in thymocyte development is largely due to compensatory functions of LEF1 in the TCF1-deficient cells. While thymocyte development is normal in LEF1-deficient mice, dual deficiency of TCF1 and LEF1 blocks this process before cells reach the DP stage (Okamura et al., 1998; Yu et al., 2012). While LEF1 can substitute for some functions of TCF1, they are not entirely interchangeable. TCF1 is expressed from the earliest DN1 stage of thymocyte development and restricts expression of LEF1, which appears only at the DN3 stage (Tiemessen et al., 2012). This limitation on LEF1 likely helps prevent malignant transformation during early T cell development, as Tcf7^−/− mice exhibit higher rates of such transformations, which can be mitigated by either conditionally deleting Tcf7 at later stages using Lck- or Cd4-Cre drivers or by additionally knocking out Lef1 (Wang et al., 2021; Yu et al., 2012). The other two TCF/LEF family members, TCF7L1 and TCF7L2, have not been shown to have any apparent role in T cell development (Zhao et al., 2020).

As CLPs migrate to the thymus, they encounter Delta-like ligands expressed on stromal cells (Koch et al., 2008), which stimulates Notch signaling, and promotes Tcf7 expression by acting at enhancer elements upstream of the Tcf7 gene (Fig. 2) (Germar et al., 2011; Sultana et al., 2010; Weber et al., 2011a). Initial Tcf7 activation also relies on RUNX transcription factors, as thymocytes lacking CBFβ, the essential RUNX binding partner, fail to induce Tcf7 or Gata3 expression, and are unable to proceed with T cell lineage specification (Guo et al., 2008). Similarly, ablation of both Runx1 and Runx3 from bone marrow progenitors blocked Tcf7 induction and thymocyte development at the earliest stages (Shin et al., 2021). Interestingly, a recent study has identified an intrinsically disordered domain in TCF1, termed L1, which is necessary for normal T cell development and for association of TCF1 with RUNX1, either through direct interaction or by allowing access to the same DNA binding sites (Goldman et al., 2023).

Fig. 2. TCF1 in early T cell commitment.

When CLPs enter the thymus, they encounter Delta-like (DLL) and Wnt ligands from thymic stromal cells, which activate the Notch and Wnt signaling pathways. Notch signaling triggers the expression of TCF1, a pioneering transcription factor that displaces nucleosomes to open chromatin at genes like Bcl11b and Gata3. TCF1 then collaborates with Notch to boost the expression of these genes and reinforce its own expression. As thymocytes progress to the DN3 stage, TCF1 activates genes that support survival while suppressing genes that are associated with other lineages, or could cause malignancy during this critical gene recombination and proliferation phase.

Following its initial activation, Tcf7 expression in T lineage cells can be maintained independent of Notch signaling (Ji et al., 2010). In precursor thymocytes, TCF1 functions as a pioneering transcription factor at genes like Gata3 and Bcl11b, targeting closed heterochromatic regions and rendering them accessible for other transcription factors (Johnson et al., 2018; Kueh et al., 2016; Weber et al., 2011b). These transcription factors then collaborate to steer developing thymocytes toward a T cell lineage and, importantly, to close off alternative myeloid, ILC, or B cell fates (Garcia-Perez et al., 2020; Hu et al., 2018; Li et al., 2013).

Once thymocytes commit to a T cell fate, TCF/LEF factors are necessary for efficient assembly of Tcrb in DN3 cells (Yu et al., 2012). While the exact mechanism remains unclear, TCF/LEF factors might be involved in generating or repairing DNA double-strand breaks associated with V(D)J recombination at Tcrb. In this context, TCF/LEF proteins have been shown to associate with DNA repair factors like DNA-PKcs and Ku80 (Shimomura et al., 2013), and bind in the genome at sites that commonly overlap those targeted by the V(D)J recombinase (Dose et al., 2014). Upon productive assembly of a Tcrb gene, thymocytes transition to the DN4 and DP stages, with Tcra gene rearrangement occurring in the latter. Once again, these cellular and molecular processes are influenced by TCF1 and LEF1, including their association with the Tcra enhancer (Giese et al., 1995; Okamura et al., 1998). TCF1 also sets the stage for the selection of successful TCRα rearrangements by decreasing the expression of Notch1 and Ptcra during the DN3 to DN4 transition, thus allowing only cells with a functional TCRα to persist (Yu et al., 2012; Yu & Xue, 2013). TCF1 is most highly expressed in DP thymocytes, where it cooperates with the E protein HEB to promote survival during successive waves of Tcra rearrangement (D’Cruz et al., 2010; Sharma et al., 2014). This function of TCF1 relies on co-binding and displacement of nucleosomes at over 50 % of HEB targets, and by protecting HEB from proteasomal degradation (Emmanuel et al., 2018). TCF1 also aids in DP cell survival by upregulating anti-apoptotic factors Bcl-XL and RORγt (Ioannidis et al., 2001; Wang et al., 2011; Xie et al., 2005).

In addition to the αβ-TCR lineage, TCF1 appears to regulate γδ-T cell development and fate in the thymus. While higher TCF1 expression favors the αβ-T cell fate, the alternative HMG protein Sox13 acts to partially counteract TCF1 function, thereby favoring a γδ-T cell fate (Melichar et al., 2007). Among committed γδ-T cells, TCF1 expression leads to the formation of interferon gamma (IFNγ) producing Tγδ1 cells by counteracting a Sox4, Sox13, and c-Maf centered transcriptional network that would otherwise support RORγt and IL17 expression. Conversely, reduced TCF1 expression allows this Sox4/13 and c-Maf transcriptional program to promote the development of IL17-producing Tγδ17 cells (Malhotra et al., 2013; Zuberbuehler et al., 2019). In vivo, reduced TCF1 expression can be mediated by γδ-TCR engagement, which upregulates the Id3 transcription factor, displacing E proteins from an intronic regulatory element in the Tcf7 gene body (Fahl et al., 2021).

4. TCF1 in CD8⁺ vs CD4⁺ T lineage commitment

Upon expression of a functional TCR, thymocytes must undergo positive and negative selection to recognize peptide-MHC complexes, while maintaining tolerance to self-antigens. A role for TCF1 in these selection processes remains inconclusive. In a classic model of negative selection, T cells that recognize the male-specific HY antigen are typically eliminated in males but not females. In Tcf7^−/− mice with a transgenic TCR recognizing HY antigen, there is an increase in HY-specific CD8⁺ T cells regardless of sex, suggesting a key role for TCF1 in negative selection (Kovalovsky et al., 2009).

After proceeding though thymic selection, T cells must commit to either the CD8⁺ or CD4⁺ single positive fates (Fig. 3). This choice is influenced by competing networks of transcription factors centered around the mutually antagonistic ThPOK and RUNX3 proteins (Egawa & Littman, 2008; He et al., 2008; Setoguchi et al., 2008). TCF1, LEF1, and Tox support ThPOK in committing to a CD4⁺ state (Steinke et al., 2014), while a network including Mazr, STATs, and RUNX factors silences the Cd4 gene to create a CD8⁺ state (McCaughtry et al., 2012; Park et al., 2010; Sakaguchi et al., 2010). Although TCF1 and LEF1 are not essential requirements for establishing the CD8⁺ single positive fate, they cooperate with RUNX3 to help silence Cd4. (Taniuchi, Osato, et al., 2002; Taniuchi, Sunshine, et al., 2002; Xing et al., 2016, 2018).

Fig. 3. Distinct functions of TCF1 in the differentiation of CD4⁺ and CD8⁺ single positive T cells.

Different transcription factor networks affect the outcome of DP thymocytes. The final fate hinges on the balance between the opposing ThPOK and RUNX3 transcription factors. TCF1 is necessary to activate ThPOK for the CD4⁺ lineage, but also contributes to RUNX3 suppression of the Cd4 gene in CD8⁺ cells.

5. TCF1 in CD8⁺ T cell activation, memory, and exhaustion

Once CD8⁺ T cells exit the thymus, they enter circulation to detect and destroy infected or malignant cells. The freshly exported CD8⁺ T cells express high levels of TCF1, which keeps them in a naïve state by repressing transcription factors that coordinate effector/cytotoxic programs, including BLIMP1, RUNX1, and ZEB2 (Chen et al., 2019; Xing et al., 2016). Once activated by cognate antigens and cytokines, CD8⁺ T cells clonally expand and rapidly suppress TCF1 expression, in a process which depends on both the TCR/PI3K and cytokine/STAT signaling pathways (Cannons et al., 2021; Danilo et al., 2018; Lin et al., 2015, 2016; Willinger et al., 2006). In addition, Tcf7 silencing is at least partially mediated by de novo DNA methylation (Danilo et al., 2018; Youngblood et al., 2017).

Following elimination of a tumor or an infection, most of the expanded CD8⁺ T cell pool dies, but a small fraction of TCF1⁺ T cells remain, establishing long-term memory. Notably, mature CD8⁺ T cells that survive thymocyte development in Tcf7^−/− mice exhibit minimal differences in their T effector (T^eff) response to a primary immune challenge, but they fail to efficiently form memory T cell populations, leading to significant defects upon secondary challenges. The deficiency in forming memory populations is attributed to their inability to activate TCF1 targets critical for a functional memory state, including homing markers like CCR7 and CD62L, as well as transcription factors such as Eomes. (Jeannet et al., 2010; Zhou et al., 2010). Similar results are seen when Tcf7 is conditionally deleted from only mature T cells (Danilo et al., 2018; Shan et al., 2017), and the block in memory T cell formation becomes nearly total when Tcf7 and Lef1 are deleted (Zhou & Xue, 2012).

Two general models have been proposed to explain the origin of long-term TCF1⁺ CD8⁺ memory cells. In one model, CD8⁺ memory cells form directly from TCF1⁺ naïve cells without proceeding through a TCF1⁻ T^eff stage. The underlying mechanism to support this model involves asymmetric cell division early in the clonal expansion phase, which can produce daughter cells that either repress or retain Tcf7 expression (Chang et al., 2007; Gräbnitz et al., 2023; Kakaradov et al., 2017; Lin et al., 2016; Pais Ferreira et al., 2020; Silva et al., 2023). This process results in a larger population of T^eff that combat the initial infection and a smaller group of memory precursor cells that retain some TCF1 expression, eventually forming long-lived memory T cells, which protect against future infections. In an alternative model, memory T cells develop from T^eff cells that undergo a de-differentiation process to reacquire TCF1 expression and become long-lived (Bannard et al., 2009; Youngblood et al., 2017). These models are not necessarily mutually exclusive. Indeed, a recent study tracking TCF1 expression in ex vivo differentiated CD8⁺ T cells using a Tcf7 reporter allele revealed that TCF1⁺ memory cells could emerge either directly from the naïve population or from T^eff cells through the reversal of TCF1 downregulation (Abadie et al., 2024). This flexibility in pathways may allow for the establishment of a scalable memory population proportional to the severity of the disease.

Unlike the response to acute challenges, chronic antigen stimulation leads to a state of T cell dysfunction known as exhaustion (Fig. 4) (Hsiung & Egawa, 2023; Im et al., 2016). During chronic antigen exposure, antigen-specific T cells remain, but they exhibit diminished effector functions, aiming to control the ongoing threat while preventing autoimmune damage (McLane et al., 2019). While it had classically been considered that chronic antigen stimulation hinders memory T cell development (Wherry et al., 2004); recent findings indicate that a minor stem-like subset of exhausted T cells (T^ex-stem) persists and generates terminally differentiated exhausted effector T cells (T^ex-eff) (Blank et al., 2019). T^ex-stem cells, characterized by high TCF1 expression, are most likely to endure chronic antigen stimulation in cases like cancer or persistent infection (Wu et al., 2016). These PD1⁺TCF1⁺ T^ex-stem populations are maintained in lymphoid tissues, including tertiary lymphoid structures at sites of chronic antigen stimulation, and can be revitalized through immune checkpoint blockade (ICB) therapy, enabling them to replenish the T^ex-eff cell pool and renew the assault on the persistent threat (Gill et al., 2023; Im et al., 2023).

Fig. 4. TCF1's role in CD8⁺ memory cell formation and chronic infection response.

Top: Upon activation, naïve CD8⁺ T cells become either effector cells (which control infections, then die after the infection is cleared) or memory precursor cells (which form long-lived memory cells). Effector T cells also have some capacity to reacquire TCF1 and become long-lived memory cells. Bottom: If the infection persists, ongoing antigen stimulation maintains stem-like exhausted cells, which generate exhausted effector cells that control the infection to an extent but with reduced functions to prevent autoimmune damage.

6. The TCF1-BLIMP1-BCL6 axis in Th1 vs Tfh differentiation

As with CD8⁺ T cells, naïve CD4⁺ T helper (Th) cells retain high TCF1 expression after leaving the thymus. Antigen-presenting cells activate naive CD4⁺ T cells by presenting antigens on MHC-II following infection. Once a CD4⁺ T cell identifies its specific antigen, it can differentiate into various subsets to best meet the threat. This specialization depends on the type of APC, co-stimulatory signals, and cytokine environment encountered during activation. The precise role of TCF1 in these processes varies, depending on the cues received during differentiation/polarization of CD4⁺ T cells (Luckheeram et al., 2012).

A well-studied role for TCF1 in CD4⁺ T cell specialization is in the divergence between Th1 and T follicular helper (Tfh) commitment. Depending on the activation conditions, naïve CD4⁺ T cells can develop into Th1 cells, which aid in eradicating infections by secreting IFNγ and related cytokines, or into Tfh cells, which assist B cells in germinal center reactions to generate robust and durable antibody responses. Initially, both Tfh and Th1 cells share a similar differentiation pathway, co-expressing T-bet and BCL6 (Nakayamada et al., 2011; Oestreich et al., 2012).

TCF1-deficient cells show greatly reduced Tfh cell formation and function, while its overexpression enhances Tfh production at the expense of the Th1 pathway, indicating the importance of TCF1 balancing Tfh–Th1 fate decisions (Choi et al., 2015, p. 2; Wu et al., 2015; Xu et al., 2015). Early in Th1 differentiation, elevated signaling through the high-affinity IL2 receptor upregulates BLIMP1 to limit Tcf7 expression. Conversely, during Tfh formation, higher TCF1 levels promote BCL6 expression, while also restraining the expression of the high-affinity IL2 receptor and BLIMP1 (Wu et al., 2015). Mechanistically, TCF1 binds the promoter region of BCL6, facilitating expression, while simultaneously occupying upstream regulatory elements in the Prdm1 locus to decrease BLIMP1 production (Xu et al., 2015). Surprisingly, the upregulation of BCL6 appears to be driven partially by association with the histone methyltransferase EZH2, as loss of this factor prevents the acquisition of an accessible chromatin landscape within the Bcl6 locus, curtailing its expression (Chen et al., 2020; Li et al., 2018). Although normally associated with establishing H3K27me3-based silencing of genes, a Ser21-phosphorylated form of EZH2 found in Tfh cells is associated with H3K27ac and an open chromatin state at TCF1 bound regions (Li et al., 2018). In addition to supporting BCL6 expression, TCF1 and LEF1 also are important for limiting expression of inhibitory markers such as CTLA4 and LAG3 in Tfh cells, ensuring that they can provide proper B-cell help (Li et al., 2021). In cases of chronic antigen exposure, a population of memory-like TCF1⁺BCL6^low progenitors continually gives rise to both TCF1⁻BCL6⁻ Th1 and TCF1⁺BCL6⁺ Tfh cells (Xia et al., 2022). Thus, a delicate balance between BLIMP1, TCF1, and BCL6 governs the fate and function of Th1 and Tfh cells (Fig. 5).

Fig. 5. TCF1 and BLIMP1 counter-regulate each other to determine Tfh vs Th1 fate.

Tfh and Th1 cells originate from a shared developmental pathway. Elevated TCF1 levels inhibit BLIMP1 while activating BCL6, steering the development towards a Tfh cell fate. Conversely, strong IL2 signaling elevates BLIMP1 expression, thereby suppressing TCF1 and BCL6, which leads to a Th1 cell fate.

7. TCF1 functions in Th2 cells

TCF1-deficient mice show a marked reduction in Th2 development, instead being diverted towards a Th1 state. TCF1 promotes Th2 identity by inhibiting Ifng production and by boosting Gata3 expression, leading to increased IL4 production (Yu et al., 2009). IL4 signaling can in turn reduce Tcf7 expression through STAT6, but this inhibition mainly decreases short isoforms, further enhancing the effect of the long β-catenin interacting isoforms (Fig. 6) (Maier et al., 2011). Conversely, LEF1 appears to negatively affect Th2 functions; when overexpressed in Th2 cells, it associates with GATA3 and prevents it from upregulating type 2 cytokine genes like Il4 (Hebenstreit et al., 2008; Hossain et al., 2008). Allergic diseases have been found to commonly feature a TCF7^hi population of multipotent progenitor cells that can give rise to effector Th2 cells that produce inflammatory cytokines (Kratchmarov et al., 2024).

Fig. 6. TCF1 reinforces GATA3 to establish the Th2 fate.

TCF1 enhances GATA3 expression in Th2 cells, which in turn activates Il4. LEF1 can potentially counteract GATA3, preventing it from activating Il4 transcription. Autocrine IL4 signaling reduces Tcf7 through STAT6 signaling, but this suppression is limited to the short isoforms that do not interact with β-catenin.

8. TCF1 in the Th17 subset

The role of TCF1 in Th17 cells is less understood than in other T helper subsets. In thymocyte development, TCF1 promotes RORγt expression during Tcra gene rearrangement to help maintain their survival through this extended process. However, TCF1 also silences RORγt and IL17 in newly produced CD8⁺ T cells, indicating a potential suppressive role on type 3 fates (Mielke et al., 2019; Wang et al., 2011). This dual function might be due to TCF1’s ability to act as an activator or repressor depending on the context of co-regulators with which it associates. While both Cd4-Cre and Cd4-Cre Ctnnb1^ex3 (expressing a constitutively active ß-catenin) T cells show TCF1 binding to the Rorc locus, only the Cd4-Cre Ctnnb1^ex3 cells showed increased locus accessibility and upregulated RORγt expression (Keerthivasan et al., 2014). This demonstrates that the effect of TCF1 binding at a gene can have disparate outcomes based on the regulatory partners with which it associates. Thus, TCF1’s role as a repressor or activator of type 3 genes may hinge on the relative abundance of co-activators and co-repressors.

TCF1 appears to negatively influence the production of type 3 cytokines, as blocking Wnt signaling during Th17 cell differentiation results in increased IL17 levels (Lee et al., 2012). Indeed, TCF1 can bind multiple regulatory elements in the Il17 locus, and type 3 polarization of Tcf7^−/− T cells results in approximately three times more IL17 production as compared to wild-type cells (Yu et al., 2011). This repression of IL17 relies on the intrinsic HDAC potential of TCF1, with TCF1-deficient T cells showing increased histone acetylation across the Il17 locus (Ma et al., 2011; Zhang et al., 2018). This increased propensity toward IL17 production occurs in a stage-specific manner, as it is seen in germline or early-stage conditional Tcf7 knockout mice (Vav1-Cre), but not when Tcf7 is removed at later stages of thymocyte development with Cd4-Cre drivers (Zhang et al., 2018).

It has been suggested that TCF1 expression in Th17 cells also helps maintain them in a more plastic state, allowing them to produce type 3 cytokines or IFNγ in different microenvironments (Muranski et al., 2011). This idea is supported in mouse models of autoimmune diseases, such as multiple sclerosis and inflammatory bowel diseases, where high TCF1 expression marks a progenitor-like Th17 population that can give rise to a pathogenic Th1-like cell population capable of producing both IL17 and IFNγ (Gaublomme et al., 2015; Karmaus et al., 2019; Schnell et al., 2021; Shin et al., 2018).

9. TCF1 in Treg function

TCF1 plays a complex role in the development and function of Tregs. It has been noted that Wnt signaling can increase TCF1 activity in Tregs, limiting their regulatory functions by curtailing FOXP3 transcription (van Loosdregt et al., 2013). Similarly, conditional expression of a constitutively active β-catenin using a Cd4-Cre driver resulted in abnormal cytokine production in Tregs, ultimately leading to colitis (Keerthivasan et al., 2014). Tregs also show lower TCF1 expression relative to conventional T cells, and TCF1 deficient mice have higher frequencies of Tregs in the thymus, seemingly because of a lowered threshold of TCR affinity for being diverted to a Treg fate (Barra et al., 2015). Likewise, LEF1 was found to be under-expressed in Tregs as compared to conventional T cells (Fu et al., 2012).

Together, these findings suggested that TCF1 negatively regulates Treg function. However, while the loss of Tcf7 or Lef1 alone does not appear to disrupt most Treg functions, their loss does impact the Tfr subset. Indeed, conditional deletion of both Tcf7 and Lef1 from Foxp3-Cre mice resulted in spontaneous autoimmune disease (Xing et al., 2019; Yang et al., 2019) The pathogenic phenotype appears to stem from the requirement for TCF1 to limit the expression of genes associated with inflammation, such as Prdm1 and Ifng, while also being required for the expression of genes for optimal Treg survival, such as Bcl6 and Ikzf4. Thus, in the context of autoimmunity, TCF1 and LEF1 play dual roles: maintaining the survival and lineage-commitment of Tregs, and limiting abnormal production of inflammatory cytokines.

10. ILC development

The innate counterparts of T cells, ILCs, share many similarities in transcriptional programs with their adaptive equivalents (Koues et al., 2016; Shih et al., 2016). TCF1 is pivotal in committing to an ILC fate. All ILCs originate from bone marrow CLPs, progressing along a path of increasingly lineage-restricted progenitor populations (Fig. 7) (Guia & Narni-Mancinelli, 2020). Using Tcf7-GFP reporter mice, a population called early ILC progenitors (EILPs) have been identified, arising from CLPs and distinguished by high TCF1, NFIL3, and TOX levels, but with lower levels of more specific ILC fate-associated markers like PLZF, Id2, and IL7 receptor (Yang et al., 2015). Single-cell RNA-seq confirmed that TCF1 expression increases during this EILP stage, helping to cement a lymphoid identity. These EILPs transition through a TCF1^low “specified” (sEILP) stage, where they retain some myeloid potential, then to a TCF1^hi “committed” (cEILP) stage more strictly enforcing an ILC fate (Harly et al., 2019). These developmental programs are mediated by TCF1 activation of lymphoid determining genes like Id2, Il7r, and Gata3, while it inhibits alternative fate genes such as Flt3, Spi1, and Irf8. In TCF1-deficient mice, ILC progenitor populations exhibit unusually high levels of these alternative fate-associated genes. (Seillet et al., 2016).

Fig. 7. An overview of ILC development.

CLPs in the bone marrow advance through a sequence of progenitor populations with gradually narrowing potential outcomes. Common developmental pathways shown by solid arrows, with less common pathways disgnated by dashed arrows. TCF1 plays a crucial role in this progression by promoting the production of factors linked to ILCs (IL7 receptor, NFIL3, Id2) and repressing those associated with the development of alternative myeloid fates (FLT3, PU.1, IRF8).

The cEILP population then progresses into the further restricted CHILP or ILCP populations, which can generate all helper ILC lineages (ILC1, ILC2, and ILC3) but have reduced natural killer (NK) cell potential (Klose et al., 2014). Notably, an upstream Tcf7 enhancer element targeted by Notch, TCF1, RUNX, and GATA3 is responsible for upregulating Tcf7 in both early T cell and ILC progenitors, but Notch binding to this enhancer is required for initiation of Tcf7 expression only in T cells (Harly et al., 2020). This suggests that while T cells and ILCs have many shared regulatory elements, their activities can vary due to differences in transcription factor expression between the two lineages. To varying extents, TCF1 expression is maintained in all mature ILC populations (Björklund et al., 2016; Yang et al., 2015), suggesting that, in addition to shaping ILC development, TCF1 also plays a role in mature ILC function.

While the branch point between progenitors for cytotoxic NK cells and helper type ILCs has historically been difficult to determine, recent studies have clarified this point (Ding et al., 2024; Liang et al., 2024; Rodriguez-Rodriguez et al., 2022). The EILP population can be further refined to identify early NK progenitor populations that can be distinguished by higher NFIL3, high Eomes expression, and lower PLZF that preferentially gives rise to cytotoxic NK cells. In contrast, lower NFIL3 and Eomes and higher PLZF expression marks progenitors that preferentially give rise to non-cytotoxic helper type ILCs (Ding et al., 2024; Liang et al., 2024).

11. TCF1 function in Type 1 ILCs

The founding member of ILCs, the type 1 ILC subset known as NK cells, are perhaps the most extensively researched concerning the role of TCF1. Studies using Tcf7-GFP reporter mice have shown that TCF1 is high in immature (CD27⁺CD11b⁻) NK cells, with levels decreasing in intermediately mature (CD27⁺CD11b⁺) NK cells and becoming very low in fully mature (CD27⁻CD11b⁺) NK cells (Jeevan-Raj et al., 2017). This decrease in TCF1 levels with maturation coincides with an increase in BLIMP1 expression (Kallies et al., 2011). A similar relationship is observed in humans between high TCF1 expression in the less-mature or helper type CD56^bright NK cells and low TCF1 in cytotoxic effector type CD56^dim NK population (Collins et al., 2019). While this may imply that CD56^bright NKs are precursors that give rise to CD56^dim NK cells, the exact developmental relationship between these two cell populations in humans remains unclear, with studies in non-human primates suggesting that these cell types may actually reflect two distinct lineages (Wu et al., 2014). Interestingly, CD56^bright NK cells transcriptionally resemble NK cells that develop from helper type ILC precursors in mice, whereas CD56^dim NK cells more closely resemble those that develop from the more cytotoxic-restrained early NK precursors, further hinting at the possibility that human CD56^bright and CD56^dim NK cells also develop from distinct pathways (Ding et al., 2024). Together, TCF1 appears to play a similar role in type 1 ILC development and function as it does in adaptive T cells, where a TCF1 and Bach2 centered transcriptional program supporting self-renewal dominates in the less mature or naïve cells (Collins et al., 2019; Li et al., 2022). The TCF1-centered stem-like program is then decommissioned with maturation and activation in favor of an Id2 and BLIMP1 or HOBIT centered network that supports the acquisition of effector markers such as granzyme B (Collins et al., 2019; Li et al., 2021, p. 201; Taggenbrock & van Gisbergen, 2023; Torcellan et al., 2024).

Despite a reduction in NK cell progenitor populations in Tcf7^−/− mice, maturation of NK cells still occurs (Held et al., 1999, 2003). However, NK cells that develop without TCF1 exhibit functional differences. Tcf7^−/− NK cells display significantly lower expression of the inhibitory receptor Ly49a. Additionally, TCF1 is crucial for controlling granzyme B production in NK cells. Although Tcf7^−/− NK cells produce higher levels of granzyme B, they show reduced cytotoxicity due to decreased survival and proliferation rates (Jeevan-Raj et al., 2017). The role of TCF1 in moderating effector functions in NK cells has been further validated with NK-specific Ncr1-Cre conditional knockouts of Tcf7 (Li et al., 2021). Indeed, TCF1 appears to be important for promoting the survival of mature NK cells, as its conditional deletion at various developmental stages, whether through Vav-Cre, Cd122-Cre, or Ncr1-Cre drivers, results in mature NK cells that are more susceptible to apoptosis (Liu et al., 2021).

TCF1 may also influence NK cell memory-like responses. In the murine cytomegalovirus (MCMV) model, memory NK cells show decreased Tcf7 expression and reduced accessibility at regulatory regions containing HMG motifs (Lau et al., 2018). The decrease in TCF1 in MCMV-specific memory NK cells seems to reflect that they are already poised for effector functions during secondary response, possibly explaining the relative independence from activation signals, such as IL18, in memory NK cells during secondary responses (Madera & Sun, 2015). Conversely, a mouse model of Zika virus infection resulted in a population of TCF1^hi NK cells that resembled memory T cells (Kujur et al., 2020). Likewise, chronic HIV infection can lead to an expansion of a TCF1^hi NK population, which had similarities to memory CD8⁺ T cells in chromatin and transcriptional profiles (Wang et al., 2020). This state can be replicated in vitro with IL12/15/18 treatment, but is blocked by a Wnt inhibitor or Tcf7 knockdown, indicating TCF1 is important for maintaining a memory-like NK state. These findings have important translational implications, especially with regards to using TCF1-dependent pathways in NK cell function for combating disease, Specifically, IL12/15/18 conditions are used to generate cytokine-induced memory NK cells that are being tested for cancer treatments (Terrén et al., 2022). Indeed, recent transcriptional and epigenetic profiling of NK cells stimulated with IL12/15/18 showed the production of distinct TCF1^hi and TCF1^low memory NK cell populations with unique functional characteristics (Foltz et al., 2024). In addition to cytokine treatment, NK cell detection of tumor or viral-derived DNA through the cGAS-STING pathway has been implicated in the expansion and maintenance of TCF1^hi NK cells (Lu et al., 2023).

Deciphering the role of TCF1 in non-NK ILC1s has traditionally been challenging due to their close resemblance to conventional NK cells. Research utilizing Eomes-reporter mice indicates that ILC1s and NK cells are indeed separate developmental entities, yet identifying markers that distinguish these populations without reporter mice remains contentious (Colonna, 2018). This issue is further complicated by the fact that NK cells, ILC2s, and ILC3s all can assume an ILC1-like state, making it difficult to categorize “true” ILC1s. Although the BLIMP1 homolog HOBIT has often been used as a marker of ILC1s, recent single-cell RNA sequencing work has clarified that ILC1 reliance on HOBIT is highly variable between different tissues. Many immature TCF1⁺ ILC1 populations lack expression of HOBIT, and some mature ILC1 populations, such as those found in the human intestinal epithelium, rely on BLIMP1 expression instead (Jaeger et al., 2024). It appears that much like for NK cells, liver ILC1s can still develop in Tcf7^−/− mice (Mielke et al., 2013). Similar to its function in limiting effector activities in NK cells, higher TCF1 expression in tissue-resident ILC1s located in the liver and salivary glands seems to characterize a less mature population, which develops into TCF1^low ILC1s. The latter produce high levels of effector molecules, such as IFNγ and granzyme B, following upregulation of the BLIMP1 or HOBIT (Friedrich et al., 2021; Jaeger et al., 2024; Yomogida et al., 2021).

Although the inverse relationship between TCF1 and BLIMP1/HOBIT expression has been well established in type 1 ILCs, the exact mechanisms for how these factors silence TCF1 expression upon activation have remained unclear. We have identified an intronic regulatory element in the Tcf7 locus that gains accessibility specifically during cell activation. This element binds BLIMP1 and maintains downregulation of Tcf7 during NK cell responses (unpublished observations).

12. ILC2 subsets depend on TCF1

Among ILC populations, the development of ILC2s seems to rely most heavily on TCF1. Mice deficient in Tcf7 show a near complete loss of ILC2s, and the few that do form are functionally impaired, unable to produce appropriate levels of IL5 or IL13 in response to stimuli such as papain-induced asthma or helminth infection (Mielke et al., 2013; Yang et al., 2013).

The development of ILC2 is influenced by Notch signaling, which can also be blocked in ILC progenitors when transduced with a dominantnegative MAML and restored when co-transduced with TCF1 (Wong et al., 2012; Yang et al., 2013). TCF1 promotes ILC2 development by upregulating GATA3, as seen in Th2 cells, but it also includes mechanisms independent of GATA3, such as the maintenance of IL7 and IL33 receptor expression.

Much like T cells, TCF1 regulates persistence and tissue localization in memory-like ILC2s. Following sensitization, a Tcf7^hi Tox^hi population of ILC2s establishes residency in the small intestine lamina propria during asthma remission. These memory-like ILC2s can return to the lung upon rechallenge, showing higher responses compared to naïve cells (Bao et al., 2024).

13. Roles for TCF1 in ILC3 and LTi cells

Type 3 ILCs in mice consist of three subsets: LTi-like CCR6⁺NKp46⁻ ILC3s that can secrete IL17 and aid in lymphoid organ formation, an intermediate CCR6⁻NKp46⁻ group, and a CCR6⁻NKp46⁺ subset that favors IL22 production (Colonna, 2018). There is a marked reduction in the total frequency of ILC3s in Tcf7^−/− mice, but TCF1 deficiency affects ILC3 subsets differently (Mielke et al., 2013; Yang et al., 2013). Although the development of Peyer’s patches is compromised in Tcf7^−/− mice, indicating a potential malfunction in LTis, these mice have normal lymph nodes, and LTi-like ILC3s are present in normal numbers (Mielke et al., 2013). Rather, the reduction in overall ILC3s seems to be due specifically to the loss of NKp46⁺ ILC3s. As was seen for germline Tcf7^−/− mice, conditional deletion of Tcf7 using Vav-Cre or Rorc-Cre drivers results in apparently normal LTi cell development, but loss of Peyer’s patches and tissue-specific disruptions in ILC3 function. In the small intestine lamina propria of Tcf7^f/f Vav-Cre mice, LTi cells show abnormally high IL17 production, while those residing in the few remaining Peyer’s patches of Tcf7^f/f Rorc-Cre mice show reduced GATA3 and lymphotoxin expression (Zheng et al., 2023). The latter finding potentially explains the specific defect in Peyer patches, while lymph node formation remains unaffected, as this phenotype is similar to that previously reported with reduced lymphotoxin production in Lta^+/−Ltb^+/− mice (Koni & Flavell, 1998; Mebius, 2003). Notch signaling has been found to play a significant role in the formation function of NKp46⁺ ILC3s, at least partially by upregulating TCF1 and T-bet expression. However, even without Notch signaling, a subset of NKp46⁺ ILC3s expressing TCF1 persists, implying that these cells can develop through both Notch-dependent and independent pathways (Chea et al., 2016).

As with T cells, TCF1 also seems to limit IL17 production in ILC3s, with all ILC3 subsets showing elevated IL17 levels in Tcf7^−/− mice (Mielke et al., 2013). Additionally, Tcf7^−/− ILC3s exhibit heightened IL17 and IL22 production in response to fungal ligands, supporting a TCF1 role in restraining type 3 cytokine production (Malhotra et al., 2013). Similarly, aberrant Wnt/β-catenin signaling in ILC3s, which would perturb TCF1-mediated downregulation of RORγt and create pro-inflammatory microenvironments, is associated with inflammatory bowel diseases and colorectal cancer (Hao et al., 2024).

14. Future challenges and opportunities

In this review, we highlight how TCF1 functions as a multifaceted regulator of immune cell fates, determining the balance between self-renewal in naïve or memory-like lymphocytes and effector functions in activated mature cells. Understanding the mechanisms of how TCF1 achieves these functions has led to numerous conceptual and translational insights. Perhaps most importantly, the discovery that high TCF1 expression marks a stem-like progenitor population that can be targeted with ICB to give rise to TCF1^low effector populations and reinvigorate responses against cancer (Kurtulus et al., 2019; Sade-Feldman et al., 2018; Siddiqui et al., 2019).

A similar observation that high TCF1 is associated with better responses to chimeric antigen receptor (CAR)-T cell therapy spurred hope that its enforced expression might help improve the durability of CAR-T cell treatments (Chen et al., 2021; Fraietta et al., 2018; Zheng et al., 2021). However, ectopic TCF1 expression in the context of human CAR-T cells has led to mixed results. While it has been reported that TCR-engineered T cells transduced to express TCF1 had improved survival (Zangari et al., 2022), overexpression of TCF1 alone is unable to enforce a memory-like phenotype in human CAR-T cells (Doan et al., 2024). Thus, much more remains to be learned about how TCF1 itself is regulated and how its expression status affects cell states.

One area that is critical for future studies is a deeper understanding of the factors and regulatory elements that maintain or silence Tcf7 expression to either sustain a self-renewing state or to initiate effector programs, respectively. Indeed, recent studies demonstrated that silencing TCF1 expression during T cell activation is reversible, allowing for flexibility in memory fate determination at multiple points during an immune response (Abadie et al., 2024). Understanding what signals drive early versus late TCF1 silencing and how this repression is either maintained or reversed could help tailor treatments such as ICB or CAR therapies more precisely for maximum effectiveness.