Defining Human Regulatory T Cells beyond FOXP3: The Need to Combine Phenotype with Function

Abstract

Regulatory T cells (Tregs) are essential to maintain immune homeostasis by promoting self-tolerance. Reduced Treg numbers or functionality can lead to a loss of tolerance, increasing the risk of developing autoimmune diseases. An overwhelming variety of human Tregs has been described, based on either specific phenotype, tissue compartment, or pathological condition, yet the bulk of the literature only addresses CD25-positive and CD127-negative cells, coined by naturally occurring Tregs (nTregs), most of which express the transcription factor Forkhead box protein 3 (FOXP3). While the discovery of FOXP3 was seminal to understanding the origin and biology of nTregs, there is evidence in humans that not all T cells expressing FOXP3 are regulatory, and that not all Tregs express FOXP3. Namely, the activation of human T cells induces the transient expression of FOXP3, irrespective of whether they are regulatory or inflammatory effectors, while some induced T cells that may be broadly defined as Tregs (e.g., Tr1 cells) typically lack demethylation and do not express FOXP3. Furthermore, it is unknown whether and how many nTregs exist without FOXP3 expression. Several other candidate regulatory molecules, such as GITR, Lag-3, GARP, GPA33, Helios, and Neuropilin, have been identified but subsequently discarded as Treg-specific markers. Multiparametric analyses have uncovered a plethora of Treg phenotypes, and neither single markers nor combinations thereof can define all and only Tregs. To date, only the functional capacity to inhibit immune responses defines a Treg and distinguishes Tregs from inflammatory T cells (Teffs) in humans. This review revisits current knowledge of the Treg universe with respect to their heterogeneity in phenotype and function. We propose that it is unavoidable to characterize human Tregs by their phenotype in combination with their function, since phenotype alone does not unambiguously define Tregs. There is an unmet need to align the expression of specific markers or combinations thereof with a particular suppressive function to coin functional Treg entities and categorize Treg diversity.

Keywords: FOXP3; Tregs; autoimmune disease; clones; function; human; type 1 diabetes.

Figure 1

Phenotypic characterization of different T cell subsets (nTregs, Tr1, Tr2, and activated effector T cells (Teff)).

Figure 2

Functional characterization of polyclonal and clonal Treg subsets. (A) Different mechanisms of immunosuppression by regulatory T cells (Tregs): 1. Interaction with antigen-presenting cells (dendritic cell, DC). 2. Secretion of inhibitory or immunosuppressive cytokines. 3. Metabolic disruption. 4. Direct contact with effector T cells (Teff). 5. Induction of apoptosis. (B) The revisited Treg universe, based on functional properties at the clonal level rather than on phenotype at the polyclonal level. There are Tregs with a single inhibitory function (e.g., killing, Functionality 2), while other Tregs may exert multiple immunoregulatory functionalities (some functions might be shared, as with Functionalities 1 and 3). CTLA4: cytotoxic T-lymphocyte associated protein 4, DR: death receptor, Gal-1: galectin 1, Gal-9: galectin 9, GzmB: granzyme B, HLA-II: human leukocyte antigen class II, ICOSL: inducible co-stimulator ligand, IDO: indoleamine 2,3 dioxygenase, Lag-3: lymphocyte-activation gene 3, PD-1: programmed cell death protein 1, PD-L1: programmed death-ligand 1, PFN: perforin, TIGIT: T cell immunoreceptor with immunoglobulin and ITIM domain, TIM3: T cell immunoglobulin and mucin domain 3, TRAIL: tumor necrosis factor-related apoptosis-inducing ligand. Figures were created with Biorender.com.