CD4⁺CD25⁺FOXP3⁺ regulatory T (Treg) cell subpopulation actively engaged in the induction of immune tolerance to self-antigens and in the control of immune responses (1, 2). By its nature the Treg population is heterogeneous with differences in the expression of transcription factor FOXP3, which is responsible for proliferation processes and immunosuppressive function of these cells. Tregs can be classified into thymic Treg (tTreg or nTreg), in vitro generated Treg (iTreg) and Treg induced on the periphery and outside the thymus – peripherally derived Treg (pTreg). In some conditions, such as autoimmune diseases or cancer, Treg cells respond to specific environmental signals and demonstrate plasticity and tissue-specific phenotypes, which help to implement context-dependent suppressive functions (3).

Different states of T-cell activation require the involvement of metabolic signal transduction pathways that are compatible with their function. Recent research indicates that the modulation of metabolism plays an important role in the control of some processes within Treg cells. The transition from lymphocytes of the CD4⁺ lineage to Treg cells is accompanied by active reprogramming of cellular metabolism. Intracellular metabolites and metabolic pathways directly participate in the regulation of FOXP3 expression as well as in the processes of transcription in Treg cells and in their functional plasticity (1, 4). Adaptation of Treg cellular metabolism is required to maintain a functional state in the tumor microenvironment (TME) or after tissue/organ transplantation (5). Such metabolic modifications allow these cells to mobilize all the resources necessary for rapid proliferation and/or full activation of the body’s defense mechanisms as a whole. Thus, cellular metabolism plays an essential part in the regulation of FOXP3 stability expression and Treg cell function.

Thus, the nature of metabolic processes in Treg cells is linked with the induction of a signal for the expression of transcription factor FOXP3. Apparently, this arrangement contributes to specific adaptability of Treg cells in various microenvironmental conditions. Optimization of Treg-based immunotherapies will provide the most accurate understanding of the proliferative, migratory and suppressive behavior of Treg. Recently, it has become obvious that the functional activity of immune cells is directly dependent on the possibilities of metabolism. Treg metabolism can be considered as an important part between the optimization of energy resources and the immune response mechanism, which may indicate the possibility of Treg plasticity between immune response and tolerance (6). This review presents recent research findings about the metabolism of Treg cells and its relation to their regulatory functions.

In the thymus, proliferation and development of Treg cells proceed according to the generally accepted mechanism of T-cell development: through the stages of positive and negative selection (7, 8). During the stage of positive selection, the binding of double positive (CD4+CD8+) TCR to a low-affinity peptide–MHC complex located on thymic antigen-presenting cells (APCs), which cooperate with cortical thymic epithelial cells, provides a survival signal for these thymocytes based on the strength of the interaction. Thus, if the interaction is strong the cell will survive to participate in the negative selection process. This interaction of TCR and coreceptor with MHC is necessary to create a population of T cells that can distinguish “self” from “non-self” antigens or cells (9). After positive selection, medullary thymic epithelial cells with dendritic cells produce chemokines CXCL12 and CCL25, which promote the migration of CD4⁺CD8⁺ thymocytes from the cortex to the medulla of the thymus (10). At the stage of negative selection CD4⁺CD8⁺ thymocytes interact through TCR with a high-affinity autopeptide presented by dendritic cells via class I and class II MHC complexes. Depending on the avidity of TCR to the antigen, thymocytes either undergo elimination or continue pathway of differentiation. As a result, populations of CD4⁺ T cells or CD8⁺ T cells emerge that have a specific repertoire of TCRs. In this review, we want to focus only on CD4+Treg and how their metabolism develops in general. Only those CD4⁺ T cells become Treg cells whose affinity for class II MHC is stronger than that naive CD4+ effector T cells but below the apoptosis induction threshold (11). The development of thymic regulatory T cells (natural Treg cells: nTreg) proceeds because of FOXP3 expression. The strength and duration of TCR engagement and CD28 costimulation are crucial for FOXP3 expression and for Treg phenotype induction (12). For instance, FOXP3 expression is observed predominantly at the negative selection stage and during the differentiation of CD4⁺ monopositive T cells into nTreg cells. Wherein, there are data on the presence of CD4+CD8+ thymocytes expressing FOXP3 in the thymus cortex in the absence of the TCR signal (13–15). Aside from the TCR signal, the costimulatory pathway mediated by the binding of CD28 of nTreg cells to CD80/86 located on APCs is also important for the modulation of FOXP3 expression. For example, Soskic et al. have demonstrated that positive costimulation of Treg cells results in FOXP3 expression and ensures their homeostasis (16). Thus, it can be deduced that the avidity of the interaction between thymocytes and APCs is a key determinant of Treg cell fate.

Metabolism takes part in the development of nTreg cells via a TGF-β–independent mechanism and notably, requires the stimulation of TCR and CD28, which are controlled by the PI3K–AKT–mTOR pathway (17, 18). However, Priyadharshini et al. have demonstrated that, depending on environmental conditions, TGF-β can reprograms Treg metabolism via blockade of PI3K–mediated mTOR signaling and glucose metabolism (19). Investigation of key regulators of metabolism-related transcription factors sheds additional light on the function of PI3K–AKT signal transduction in the development of Treg cells. There are four classes of PI3 kinases, but only class IA and class IB PI3Ks have been studied in detail in T cells. The typical object of research is the p110δ class IA catalytic subunit because this protein is predominantly expressed in leukocytes and directly participates in the TCR signal (20–23). p110δ is activated by receptor tyrosine kinases such as cytokine receptors CD25, CD122, CD127, and IFNGR as well as by TCR. p110δ-containing PI3K phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) yielding phosphatidylinositol-3,4,5-triphosphate (PIP3) on an inner cell membrane, thereby initiating the activation of downstream signaling factors such as PDK1 and its substrate AKT (21, 24). PDK1 selectively phosphorylates AKT on threonine-308, thus enabling AKT to phosphorylates and inhibits the transcription factor FOXO1, which serves as a positive regulator of Foxp3 expression. This cascade leads to the translocation of FOXO1 from the nucleus to the cytoplasm and to the loss of transcription-regulatory activity by this protein. Overexpression of AKT reduce the number of Treg cells and blocks the development of FOXP3⁺ Treg cells in the thymus (25, 26). Such negative effects are mainly explained by the fact that PDK1 interacts with many other molecules besides AKT, for example, with PKC isoforms (27). Nonetheless, studies focusing on negative regulators of PI3K–AKT signaling, such as the influence of a PH domain leucine-rich repeat-containing protein phosphatase (PHLPP) on PTEN (28), the relation between PI3K and AKT in the development of Treg cells in the thymus. Treg cells express larger amounts of PHLPP1 and PHLPP2 than normal T cells do and thus inhibit the PI3K enzyme, thereby promoting the development of Treg cells in the thymus (29, 30). For example, PHLPP1 knockout mice have a normal number of circulating Treg cells, which means that they may be more dependent on PHLPP2 or less dependent on PI3K–AKT signaling for Treg cell development (31).

The PI3K–AKT signaling cascade is connected with one of the key factors of metabolism: mammalian target of rapamycin complex 1 (mTORC1). mTORC1 is a key node for the integration of metabolic signals and promotes anabolic processes during cell proliferation and differentiation; subsequently, mTORC1 serves as a target of AKT (32, 33). At the positive selection stage, CD4⁺CD8⁺ T cells overexpress glucose transporter 1 (GLUT1). Presumably, glycolysis is required to a greater extent for the support of cell growth and for meeting proliferation-related needs of the cells. Multiple glycolytic enzymes are targets of AKT and the PI3K–AKT–mTOR pathway, as reported, promotes the transfer of GLUT1 to the cell surface (34, 35). Conversely, during the negative selection stage when CD4⁺ T cells differentiate toward Treg cells, T cells become more dependent on high intensity of lipid oxidation for their survival (36–38). Compared to CD4⁺ and CD8⁺ T cells, in the Treg population, GLUT1 is underexpressed, and blockade of PI3K or mTORC1 reduces glycolysis in T cells and promotes the differentiation into Treg cells in the thymus and in the periphery ( Figure 1 ) (39). For example, inhibition of mTORC1 causes a change in Treg cell metabolism towards increased lipogenesis and of the mevalonate-dependent pathway of cholesterol biosynthesis intended to maintain the proliferation and functioning of Treg cells (40, 41). In this context, excessive mTORC1 activation and unlimited glycolysis in Treg cells are characterized by low FOXP3 expression accompanied by a decline of the suppressive activity of Treg cells (42, 43).

According to the above-mentioned data, 5′AMP-activated protein kinase (AMPK) is responsible for reducing the expression of GLUT1 and mTORC1 (44). AMPK is a serine/threonine protein kinase consisting of a catalytic subunit (AMPKα) and two regulatory subunits (AMKPβ and AMPKγ), and its function is to respond to substantial energy consumption and to regulate the energy balance in the cell (45, 46). AMPK is activated when intracellular adenosine triphosphate (ATP) production diminishes and intracellular adenosine monophosphate (AMP) concentration goes up, as seen during nutrient deprivation or hypoxia (45). A recent research article revealed that AMPK is activated in T cells when they are stimulated (e.g. cytokines, TCR, or other factors) and plays an important part in T-cell metabolism (47). For example, Zhu et al. have shown that upon stimulation of TCR, AMPK activation triggers FOXP3 phosphorylation by inhibiting E3 ubiquitin ligases and proteasomes in vitro and in vivo (48). AMPK acts through phosphorylation of tuberin (TSC2), thereby launching a cascade of mTORC1-inhibitory reactions (49). TSC2 is a GTPase-activating protein (GAP) that participates in the regulation of energy flow during growth and stress and interacts with mTORC1 signaling through the GTP– bound Rheb (50). Phosphorylation of TSC2 on threonine-1227 and serine-1345 by AMPK activates TSC2, leading to the formation of a large amount of the GTP– bound Rheb, thereby shifting the balance toward the inhibition of activity of a substrate called Raptor in mTORC1 (49, 51). Inhibition Hypoxia-inducible factor 1 alpha (HIF-1α) a lead to the induction of the Treg phenotype in T cells at the negative selection stage ( Figure 1 ). Recent research suggests that Treg cells also utilize glucose. In particular, it was demonstrated that Treg cells have a large amount of a glycolytic enzyme called enolase, which is involved in FOXP3 splicing during the formation of pTreg cells at the periphery in vivo (52).

Treg cells show both high magnitude of fatty acid β-oxidation (FAO) and elevated expression of genes taking part in the regulation of FAO (53, 54). An important component of the regulation of fatty acid metabolism is carnitine palmitoyl transferase 1A (CPT1A). Treg cells harboring large amounts of activated AMPK along with characteristic FAO, probably due to a specific feature of AMPK, are able to modulate CPT1A activity and increase mitochondrial fatty acid import for β-oxidation. CPT1A is a protein found in the outer mitochondrial membrane and catalyzes the esterification of long-chain acyls by carnitine producing acylcarnitine; CPT1A is thought to be a regulator of the rate of long-chain fatty acid oxidation (55). On the other hand, studies showed that FOXP3 expression in Treg cells alters metabolic pathways for the oxidation of long-chain fatty acids, thus making Treg cells stable under some conditions (56, 57). FOXP3⁺ Treg cells are also known to take up long-chain fatty acids via CD36 as a receptor. Short- and medium-chain fatty acids passively diffuse through the cytoplasm and mitochondrial membranes to participate in FAO (58, 59). This means that some of the upregulated oxidative phosphorylation (OXPHOS) mediated by FOXP3 is linked with oxidation of exogenous fatty acids. These cells also simultaneously increase own expression of genes coding for complexes of the mitochondrial electron transport chain. Taken together, these data mean that the regulation of metabolic pathways is closely connected to T-cell development along the Treg lineage.

Depending on the immunosuppressive function of Treg cells, metabolism can be reprogrammed between glycolysis (60) and OXPHOS (61). HIF-1α can participate in regulation cytokine expression Treg. Groneberg et al. have demonstrated that HIF-1α induces IL-2 and Foxp3 expression in Treg, resulting in increased expression of IL-10 in Tregs (62). In the periphery, PD-L2 costimulation maintains Treg immunosuppressive activity by inducing CPT1A expression. Induction of CPT1A leads to the enhancement of OXPHOS process due to the stabilization of the TCA cycle, this promotes the active production of acetyl-CoA, which in its turn, leads to an increase in the expression of IL-10 pTreg (63). Resent research suggest that IL-35 plays an important role in regulating Treg metabolism. For example, inhibition of IL-35 expression leads to a change in the metabolic activity of Treg cells with an increased glycolysis and production of effector cytokines (64). In addition, the importance using of extracellular nutrients can influence Treg plasticity. For example, I. Matias et al, has shown that the addition of α-ketoglutarate led to a reprogramming of metabolism towards OXPHOS with storage lipids, which led to Treg differentiation with an inflammatory phenotype, manifested by increasing production of cytokines IFN-γ and granzyme B (65).

It can be concluded that metabolism has a significant role in the proliferation and development of Treg cells in the thymus and periphery. Moreover, this phenomenon manifests itself in the characteristic functioning of glycolysis, which promotes the flux of biomolecules in synthesis for the differentiation of CD4⁺ lymphocytes into Treg cells. Subsequently, the transition to the consumption of alternative energy sources according to the FAO mechanism is observed in the formed Treg. In this context, the strength of the signal from TCR and CD28 and the regulation of the PI3K–AKT–mTORC1 cascade are critical for FOXP3 expression and for the formation of Treg cells in general.

So far, investigation into such characteristics of Treg cells as adaptation to a microenvironment and to systemic changes remains a relevant and important scientific task. Treg cells can be considered a specific subclass of the CD4⁺ T-cell pool. Because they are physiologically present in almost all tissues and organs, while expressing a wide range of immunological and metabolic markers. Therefore, Treg cells are able to quickly respond to any immunological and metabolic signals under the conditions of tumor infiltration or after a tissue or organ transplant. Owing to Treg cells’ high sensitivity to external signals and their fundamental role in the switching between tolerance and immunity, metabolism can be regarded as one of the key mechanisms behind the regulation of Treg cells. In the following section, we describe the dynamics of metabolism in the regulation of Treg cells under conditions of tumors and transplantation.

The study of Tregs metabolism in tumors and transplantation is important due to negative effect of Tregs in immunotherapy. The interaction between tumor cells and Treg promotes their favorable cohabitation and population growth. During transplantation the conditions for the growth of Tregs are unfavorable, which is expressed by their small number in the environment and the occurrence of an immune response to the transplant ( Figure 2 ). In this regard, the study of Treg metabolism under the influence of tumors and transplantation will contribute to the development of new approaches in immunotherapy and transplantation.

The field of immunometabolism research and the regulation of Treg cells, in particular, is developing rapidly. Additional opportunities are arising for the manipulation of potentially useful features of Treg cell metabolism under various conditions. Nevertheless, there are still many unresolved issues. For example, the maintenance of a balance between metabolism and immunological function in Treg cells is a complex process of interrelations with a constant dependence on environmental conditions. Furthermore, investigators have yet not fully formulated the concept of switching of Treg cells from “classic” aerobic glycolysis to an alternative energy source in the form of OXPHOS via FAO. Recent advances in immunotherapy have focused on the checkpoint molecules of Tregs or the adoptive transfer of effector T cells. For instance, the metabolic potential of CAR-T cells and T effector cells could be enhanced to overcome the harmful effects of the TME and enhance their anti-tumor aptitude (97). In transplantation, induction of liver kinase B1 (LKB1) signaling may promote Treg survival and function (98, 99). Active research is underway regarding the regulation of metabolism during intercellular interactions in a microenvironment and regarding effects of metabolism stimulation on the suppressor function of Treg cells under various conditions, e.g., during tumor growth or after transplantation.

Technological advances, such as high-throughput RNA sequencing, are likely to provide complete information about the regulation of metabolic reprogramming of Treg cells in various situations. Present-day methods enable scientists to study genetic and pharmacological modulation of metabolic signaling pathways in Treg cells. This approach paves the way to research on the mechanisms of interaction of intracellular signals and lays the foundation for the development of new applications of Treg cells in therapy. Particularly, this knowledge will be needed in clinical practice to implement new modalities for the treatment of various diseases.

AB wrote the manuscript. JK prepared the figure. NS prepared the figure. HS planned and supervised the part of cancer study. SS planned and supervised the part of transplantation study. All authors contributed to the article and approved the submitted version.