TCR signals fuel Treg cells.

Regulatory T (Treg) cells safeguard against autoimmunity and overshooting inflammation. During their development in the thymus, Treg cells are selected by stronger autoantigenic T cell receptor (TCR) signals than conventional T cells. These TCR signals are critically important for the initiation of two key lineage defining events, namely induction of the transcription factor Foxp3 and hypomethylation of a specific gene set. Recent studies shed light on the role of continuous (autoreactive) TCR signals for identity, homeostasis and functions of mature Treg cells.

Induced TCR ablation on mature Treg cells only minimally reduces Foxp3 expression [1, 2] and does not affect hypomethylation patterns [1]. Stable Foxp3 expression was also observed in Treg cells lacking the TCR signal transduction proteins Lck [3] or SLP-76 [4] and upon ablation of the co-stimulatory receptor CD28 [5]. Fittingly, mature Treg cells, expressing a TCR but deprived of peripheral autoantigenic stimulation due to lack of MHC II on hematopoietic cells, still express Foxp3 [6]. Therefore, once a core Treg cell identity has been established in the thymus, it is maintained independently of peripheral TCR signals.

In contrast, the Treg cell surface phenotype and their signature gene expression were strongly affected by the various means of inhibiting TCR signals [1–3, 6]. However, the peripheral Treg cell pool is heterogeneous in that it consists of naïve and various subsets of effector-like Treg cells. Induced TCR ablation showed that, at least under homeostatic conditions in healthy mice, effector-like Treg cells strictly depend on TCR signals for their generation and/or maintenance. Attempts to distinguish between naïve and effector-like Treg cells based on CD62L/CD44 [2] or CD25 [1] expression revealed that both require TCR signals to maintain their characteristic surface phenotype and gene expression, independently of their homeostasis. Interestingly, apart from reduced levels of TCR-activated transcription factors such as Egr2 and c-Rel and their respective target genes, loss of TCR signals strongly affects IRF4-controlled genes. The concept that DNA hypomethylation ensures the expression of key Treg cell genes is supported by the reduced, but still robust expression of CTLA-4, GITR and Eos of TCR-deficient Treg cells [1].

Both induced TCR and co-stimulatory CD28 ablation cause a decline in Treg cells in the absence of thymic T cell production, which goes hand in hand with completely abrogated [1] or reduced [5] homeostatic proliferation, respectively. This is not due to reduced responsiveness to homeostatic cytokines, such as IL-2. Interestingly, Treg cells deprived of MHC II contact proliferate well in response to anti-CD3 stimulation [6] and TCR-deficient Treg cells divide when stimulated with TCR-bypassing PMA/Ionomycin [2], indicating that these cells do not become anergic to proliferation-inducing signals. The protein levels of Bcl-2 and Bim are two-fold upregulated in Treg cells upon TCR loss, to the same levels as in naïve CD4 T cells [1]. However, none of the studies reported a significant difference in survival between TCR signaling impaired and normal Treg cells in vivo, and BrdU pulse-chase experiments suggest impaired turn-over/proliferation, but normal survival of Lck-deficient Treg cells [3]. Collectively, these experiments show that TCR signal-induced proliferation forms an essential homeostatic requirement for all Treg cell subsets.

Treg cells suppress immune responses through the release of inhibitory cytokines, competition for IL-2, access to and functional modulation of antigen presenting cells and direct cytotoxic killing. TCR-deficient Treg cells show reduced expression of inhibitory and cytotoxic molecules including IL-10 and Granzyme B. In addition, the expression of proteins through which Treg cells regulate the functions of antigen presenting cells, such as CTLA-4, NT5E/CD73, LFA-1 and NRP1, is significantly reduced [1, 2, 6]. TCR signaling impaired Treg cells fail to efficiently suppress T cell activation and proliferation in vitro [3-6]. Conversely, augmenting TCR signaling by ablating the DAG-metabolizing kinase DGKζ enhances the in vitro suppression capacity [4]. Importantly, Treg cells require constant TCR signals to sustain their suppressive abilities in vivo: MHC II contact-deprived Treg cells fail to control naïve T cell expansion upon co-transfer [6], TCR-deficient Treg cells cannot control effector T cell differentiation/proliferation and cytokine production in situ [2] and induced TCR ablation limits Treg cell-mediated control of colitis and EAE [1]. Therefore, TCR signals continuously arm Treg cells for suppression, in addition to their role in their homeostasis.

Expression height of Ly-6C, which is strongly elevated in Lck-deficient [3] and TCR-deficient [1] Treg cells, appears to differentiate Treg cells receiving strong TCR signals (Ly-6C−) from those that do not (Ly-6C+) [7]. Purified Ly-6C− and Ly-6C+ Treg cells differed dramatically in phenotype, gene expression and their ability to suppress in vitro and in vivo. Accordingly, effector Treg cells are exclusively Ly-6C−. However, on average half of the TCR-deficient Treg cells lacked Ly-6C (Vahl et al., unpublished), showing that not all Ly-6C− Treg cells result from continuous TCR triggering. Modern multiparameter flow cytometry should help to further unravel the dynamic complexities of Treg cell subsets.

In summary, we conclude that while TCR signals are dispensable for maintaining the core Treg cell identity, they are continuously fueling their homeostasis, effector differentiation and suppressive functions.