The Redox-Metabolic Couple of T Lymphocytes: Potential Consequences for Hypertension

Abstract

Significance: T lymphocytes, as part of the adaptive immune system, possess the ability to activate and function in extreme cellular microenvironments, which requires these cells to remain highly malleable. One mechanism in which T lymphocytes achieve this adaptability is by responding to cues from both reactive oxygen and nitrogen species, as well as metabolic flux, which together fine-tune the functional fate of these adaptive immune cells.

Recent Advances: To date, examinations of the redox and metabolic effects on T lymphocytes have primarily investigated these biological processes as separate entities. Given that the redox and metabolic environments possess significant overlaps of pathways and molecular species, it is inevitable that perturbations in one environment affect the other. Recent consideration of this redox-metabolic couple has demonstrated the strong link and regulatory consequences of these two systems in T lymphocytes.

Critical Issues: The redox and metabolic control of T lymphocytes is essential to prevent dysregulated inflammation, which has been observed in cardiovascular diseases such as hypertension. The role of the adaptive immune system in hypertension has been extensively investigated, but the understanding of how the redox and metabolic environments control T lymphocytes in this disease remains unclear.

Future Directions: Herein, we provide a discussion of the redox and metabolic control of T lymphocytes as separate entities, as well as coupled to one another, to regulate adaptive immunity. While investigations examining this pair together in T lymphocytes are sparse, we speculate that T lymphocyte destiny is shaped by the redox-metabolic couple. In contrast, disrupting this duo may have inflammatory consequences such as hypertension.

Introduction

T lymphocytes comprise half of the adaptive immune system, and are critical for protection from invading pathogens and damage sustained to our bodies. These cells possess numerous subtypes with different inflammatory properties, but each plays a unique role in the host defense system.

Reactive oxygen species (ROS) and reactive nitrogen species (RNS) have been shown to shape differentiation into these various subtypes by regulating downstream signaling cascades, transcription factors, and acting as stress signals. In addition, the dynamic nature of T lymphocytes creates a great energy demand, which requires highly robust and plastic metabolic states. These shifts in metabolism create alterations in the redox environment, and conversely, redox alterations may dramatically affect cellular metabolism suggesting a tight connection between the two cellular processes. However, specific examinations into the redox-metabolic couple in T lymphocytes are infrequent, and the field remains in its infancy.

Herein, we briefly examine how ROS/RNS impact T lymphocytes, the metabolic reprogramming essential to these cells' activation and differentiation, the link between these two cellular processes in these critical adaptive immune cells, and how disruption of this couple may be an unexplored etiology of hypertension.

T Lymphocyte Biology

The immune system, composed of a diversity of cellular- and noncellular-based defenses, serves as the primary safeguard against foreign pathogens, microbes, and cells within the body. This comprehensive system is centralized to primary and secondary lymphoid organs, and utilizes the circulatory and lymph systems to travel throughout the body.

Cells of the immune system originate from hematopoietic stem cells within the bone marrow, but differentiate to eventually form myeloid and lymphoid progenitor stem cells. It is here where the innate (myeloid) and adaptive (lymphoid) immune systems and their respective cell subtypes diverge. Myeloid progenitors further develop into erythrocytes, megakaryocytes, phagocytes, and granulocytes. Lymphoid progenitors primarily differentiate into T and B lymphocytes, which constitute the adaptive immune system. Natural killer cells and three groups of innate lymphoid cells are also derived from lymphoid progenitors, however, despite their lymphoid lineage belong to the innate immune system (31).

After completion of positive and negative selection in the thymus, mature T lymphocytes are categorized into two major classes (1). The first major class is known as cluster of differentiation 4 (CD4⁺) or helper T lymphocytes. These cells possess minimal cytolytic capacity, and maintain a primary function of promoting the function and development of other cells of the immune system. Adult helper T lymphocytes, such as all adult lymphocytes, reside in a mature but naive state until activation. Upon activation, lineage-specific transcription factors along with the appropriate milieu of cytokines drive specific helper T lymphocyte subtype polarization such as T_H1, T_H2, T_H9, T_H17, T follicular helper cells (T_FH), or T regulatory cells (T_REG) (Fig. 1); all of which possess specific proinflammatory or anti-inflammatory properties that contribute to the adaptive immune response.

FIG. 1.

The various pathways of helper T lymphocyte polarization (differentiation). Naive CD4⁺ T lymphocyte cells differentiate into helper cell subtypes, including T_H1, T_H2, T_H9, T_H17, and T_FH, as well as a nonhelper cell subtype called T_REG cells. Polarization of naive T lymphocytes to a specific activation-dependent subtype requires antigen presentation, a specific subset of cytokines (shown above the arrows; these cytokines may be produced in both a paracrine and an autocrine manner from T lymphocytes as well as many other immune and nonimmune cell types), and the activation of certain intracellular signaling cascades (shown below the specific cell type; these reflect key signaling proteins that define specific cell populations). CD4⁺, cluster of differentiation 4; Foxp3, forkhead box P3; IFNγ, interferon gamma; IL-1β, interleukin 1β; IL-2, interleukin 2; IL-4, interleukin 4; IL-6, interleukin 6; IL-9, interleukin 9; IL-10, interleukin 10; IL-17, interleukin 17; IL-21, interleukin 21; IL-23, interleukin 23; IL-35, interleukin 35; RORγ, RAR-related orphan receptor gamma; T_FH, T follicular helper cells; TGFβ, transforming growth factor-beta; T_REG, T regulatory cells. Color images are available online.

The other major class of mature T lymphocytes is the CD8⁺ or cytotoxic T lymphocytes, which possess cytolytic capacity and protect the host through the killing of aberrant cells (e.g., cancer) or cells infected by intracellular pathogens. These cells, such as the helper T lymphocytes, also remain in a naive state until activated via their T lymphocyte receptor, which initiates an intracellular cascade that results in cell activation, expansion, and differentiation (45). Similar to the helper T lymphocytes, these cells have various forms of activated subtype polarization states such as T_C1, T_C2, and T_C17 (Fig. 2).

FIG. 2.

The various pathways of cytotoxic T lymphocyte polarization (differentiation). Naive CD8⁺ T lymphocytes differentiate into cytotoxic cell subtypes, including T_C1, T_C2, and T_C17. Polarization of naive T lymphocytes to a specific activation-dependent subtype requires antigen presentation, a specific subset of cytokines (shown above the arrows; these cytokines may be produced in both a paracrine and an autocrine manner from T lymphocytes as well as many other immune and nonimmune cell types), and the activation of certain intracellular signaling cascades (shown below the specific cell type; these reflect key signaling proteins that define specific cell populations). IL-12, interleukin 12; IL-13, interleukin 13; IL-15, interleukin 15; T_CM, central memory T lymphocytes; TNFα, tumor necrosis factor-alpha. Color images are available online.

The adaptive immune system works toward a targeted response to a specific antigen. This is achieved by T and B lymphocytes requiring gene rearrangement of their respective receptor segments. In doing so, it is estimated that more than 10⁸ unique antigen recognition sites are formed, allowing for a high probability of at least one receptor being able to recognize a foreign epitope (110). However, T lymphocytes are not able to respond to foreign antigens without aid from the innate immune system.

Specific phagocytes (i.e., dendritic cells [DCs] and macrophages) may also serve as antigen-presenting cells (APC), which pack antigenic molecules onto major histocompatibility complex proteins to present to T lymphocytes. The presentation of an antigen to T lymphocytes initiates the activation process, which leads to potentiated proliferation, differentiation, and functionality of these cells. This coordination between the innate and adaptive immune systems allows for significant checks and balances of the immune system so as not to lead to uncontrolled inflammation, which could cause pathological consequences.

Both activated helper and cytotoxic T lymphocytes reside in their effector states for only a short period of time, however, the long-term power of the adaptive immune system resides in its ability to form memory. After the initial activation and expansion, a contractile phase culls 90%–95% of T lymphocytes through apoptosis, while the remaining activated T lymphocytes form the memory pool (44, 166). Both helper and cytotoxic subtypes develop into memory cells, and are divided into two subtypes. These effector (T_EM) and central memory (T_CM) cells are characterized by their ability to proliferate as well as their effector functions and migration patterns (142). Memory cells benefit the host, as they do not rely as heavily upon costimulatory signals and their activation threshold is lower, allowing for a more rapid and effective response to a reoccurring pathogen (170).

Last, while T lymphocytes primarily reside in secondary lymphoid organs such as the spleen or lymph nodes for the majority of their life, these cells have the ability to infiltrate virtually every organ in the body upon activation. This requires an expansive ability to adapt to surroundings of low and high oxygen as well as hypervariable levels of metabolites. It is likely that these extreme microenvironments have shaped the ability of T lymphocytes to utilize various redox and metabolic signals in their function, but the exact specifics of processes in these adaptive immune cells remain elusive.

Redox Signaling in T Lymphocytes

ROS and RNS were long considered to be only harmful to the cell, but this view has become highly antiquated as the role of ROS/RNS in normal cellular function (i.e., redox signaling) is currently well accepted. However, when the regulation of these highly reactive species does become uncontrolled, irreversible damage may ensue leading to a shift toward harmful or even cellular death conditions termed oxidative stress. In contrast, excessive antioxidant or reductive capacity may swing the redox environment in the other direction leading to pathological consequences of reductive stress. Complex antioxidant systems have evolved to manage the balance among redox signaling, oxidative stress, and reductive stress, but the line between these cellular states is ill-defined (143).

Recent technology, primarily mass spectrometry, has significantly advanced our ability to investigate redox signaling post-translational modifications (e.g., reversible oxidation products, nitrosylation, and glutathionylation) or those associated with oxidative stress (e.g., irreversible carbonylation, oxidation, and nitration). However, these technologies come at a great deal of expense and expertise, limiting many investigators' ability to perform these intricate experiments. With this, it is often difficult for investigators to know if their experimental setup reflects physiological redox signaling, oxidative stress, or reductive stress, which may lead to conflicting results in the literature.

While the understanding of cellular redox signaling is still in its infancy, but it is evident that ROS/RNS play a physiological role in virtually every system within the body. This is particularly true within T lymphocytes, but due to the high plasticity and diversity of cell types within this class of adaptive immune cells, combined with a multitude of extreme microenvironments, the specifics of redox signaling in T lymphocytes are still be elucidated.

Exogenous pro-oxidants

Many studies have attempted to examine the role of ROS/RNS in T lymphocyte function by exogenous pro-oxidant supplementation, and this has been the primary method of examining redox effects in T lymphocytes to date. For example, hydrogen peroxide (H₂O₂) supplementation has been shown to have significant functional effects on T lymphocyte function. H₂O₂ concentrations, ranging from 10 nM to 50 μM, have been shown to decrease interferon gamma (IFNγ) production in activated T_H1 cells while increasing interleukin 4 (IL-4) production in activated T_H2 cells (40, 97). CD4⁺ T lymphocytes treated with 2,3-dimethoxy-1,4-naphthoquinone (DMNQ), an inducer of superoxide anion (O₂^•−) formation and subsequent H₂O₂ production, also promoted expression of a T_H2 cytokine profile (75). These effects of H₂O₂ on T lymphocytes appear to be due to the impairment of activator protein 1 (AP-1), nuclear factor of activated T lymphocytes (NFAT), and nuclear factor kappa-light-chain-enhancer of activated B (NFκB) signaling, which ultimately limits interleukin 2 (IL-2) production and polarization to the T_H1 subtype (36, 37, 81, 125, 151). Together, these data suggest that exogenous H₂O₂ appears to have a suppressive and anti-inflammatory effect on T lymphocytes.

However, extracellular H₂O₂ signaling has been shown to be essential in the homing of T lymphocytes to sites of inflammation, which would suggest a proinflammatory role for these ROS (115). Indeed, it has been suggested that exogenous H₂O₂ added to the site of a wound may have a greater healing effect due to the chemotactic and proinflammatory nature of these ROS as opposed to the direct killing effect of H₂O₂ on pathogens (115). These disparate results are likely due to differences concentration, timing, and many other variables in experimental setups between laboratories. Therefore, while exogenous supplementation allows for the simple examination of the effects of ROS on T lymphocytes, it is unclear if these experiments represent the in vivo physiological redox signaling responses of these immune cells.

NADPH oxidase family pro-oxidants

Like all cell types, sources of ROS/RNS are present in T lymphocytes, but studies examining how endogenously produced species affect T lymphocyte activation, differentiation, and proliferation have only begun in the last two decades. One primary source of intracellular and endogenous O₂^•− is the NADPH oxidase (NOX) family of enzymes. To date, T lymphocytes are known to express NOX2 (63, 149), dual oxidase 1/dual oxidase 2 (DUOX1/DUOX2) (80), and explicitly in humans NOX5 (6). NOX3 has not been identified in T lymphocytes to date (5, 124), and it remains unclear if NOX1 and NOX4 are specifically expressed in T lymphocytes (discussed later).

Early studies, utilizing mice lacking the NOX subunit p47phox, demonstrated that T lymphocytes initiate NOX-dependent ROS production soon after antigen presentation, and these ROS participate in redox-mediated signaling cascades essential for proper T lymphocyte activation (27, 63). Recent research has also shown that p47phox-deficient T lymphocytes are inhibited from polarizing to the T_H1 phenotype, but instead differentiate primarily to the proinflammatory T_H17 lineage (162). Work from this same group also demonstrated that the lack of T_H1 polarization in p47phox-deficient T lymphocytes was due to decreased expression of the interleukin 12 (IL-12) receptor and various proinflammatory chemokine ligands in these cells, which likely was due to diminished ROS production (123). Furthermore, p47phox-deficiency in mice leads to a general CD8⁺ dysfunction through diminished naive CD8⁺ T lymphocytes, an inability to activate T_C1 cells, and increased abundance of proapoptotic proteins in these CD8⁺ cells (28). While these data are suggestive of a direct role of either NOX1 or NOX2 (both may utilize p47phox) in T lymphocyte regulation, these studies are limited by the use of systemic p47phox knockout animals. Thus, it is difficult to conclude that these NOX isoforms directly regulate T lymphocytes, or if the effects are simply due to developmental or non-T lymphocyte effects of p47phox knockout.

Similar to p47phox knock-out, animal and human models globally deficient in NOX1, NOX4, NOX5, and DUOX1 demonstrate negative effects on the ability of T lymphocytes to properly activate and differentiate (56, 80, 98, 150, 159). Interestingly and in contrast, T lymphocytes lacking NOX2 demonstrated a pro-T_H1 phenotype with concurrent suppression of T_H2 differentiation, implying that while the ROS were critical for T lymphocyte activation, they played a suppressive role in the process (149). Together, these differential effects of ROS in T lymphocytes suggest that a “one size fits all” model may not be correct when considering redox shaping of immune function, but that likely the spatial and temporal nature of the ROS impacts outcomes. However, given the importance of redox signaling throughout all cell types and during development, it must be considered that at least some of the aforementioned effects on T lymphocytes in the various global knockout model systems are due to the systemic nature of the NOX subunit reduction throughout the body as opposed to direct effects due to loss within T lymphocytes.

While NOX-derived ROS appear to have differential effects in proinflammatory effector T lymphocytes, they also appear to play a role in anti-inflammatory T_REG cells. CD4⁺ T_REG lymphocytes lacking p47phox show a limited ability to suppress effector cells, suggesting NOX-derived ROS are essential for their regulatory functions (32). This holds true for CD8⁺ T_REG lymphocytes lacking p47phox, which also show an inability to suppress effector CD4⁺ T lymphocytes that could be rescued by the concurrent overexpression of NOX2 (172). Together, these data support a highly dynamic and complex regulatory role of NOX isoforms in T lymphocyte activation and function, but it is evident that more work is needed to identify the true nature and function of these reactive species in these cells.

Nitric oxide

In addition to ROS, RNS such as nitric oxide (NO) have also been shown to regulate T lymphocyte differentiation and function. Exogenous NO supplementation has demonstrated a proinflammatory response with increased T_H1 polarization via increased IL-12 receptor expression on these cells (114). Interestingly, exogenous NO also induces a specific subset of T_REG cells, and these cells appear to inhibit T_H17 cells without affecting T_H1 lymphocytes (112, 113).

Endogenous NO also has major effects on T lymphocyte function, and all three isoforms of nitric oxide synthase (NOS) have been found in T lymphocytes or T lymphocyte-derived cell lines (8, 57). For example, the use of pan-NOS inhibition strategies has demonstrated that reduced NO promoted T_H2 polarization, which suggests that NO may be important for T_H1 differentiation (19). Utilizing aminoguanidine to inhibit NOS activity in CD8⁺ T lymphocytes also showed decreased IFNγ mRNA expression and activation of these cells (83). T lymphocytes overexpressing nitric oxide synthase 2 (iNOS) show inhibition of T_REG cells, which would additionally support a proinflammatory environment (65). Furthermore, nitric oxide synthase 3 (eNOS) expression generally correlated with proinflammatory T lymphocyte responses, with knockout of this isoform being significantly limiting (57). Together, these studies on endogenous sources of NO show similar effects to exogenous NO supplementation in promoting a proinflammatory T lymphocyte phenotype.

Similar to other ROS effects on T lymphocytes, conflicting evidence on the effects of NO in T lymphocytes also exists in the literature. For example, T lymphocytes lacking iNOS counterintuitively demonstrated increased production of IFNγ to ovalbumin (168). These studies were performed in whole-body knockouts of iNOS, which again suggests the potential for off-target and non-T lymphocyte effects of iNOS loss. Further conflicting results were reported in a collagen-induced arthritis model where T lymphocyte proliferation and function were suppressed by myeloid-derived suppressor cell produced NO (23). These inconsistent findings highlight the complexity of redox signaling within T lymphocytes as well as between different types of immune cells and experimental designs, and support the importance of future investigation (including rigorous setups and controls) into the regulation of these oxidative species in the control of adaptive immunity.

Mitochondrial pro-oxidants

Another major source of ROS in T lymphocytes is the mitochondria, and several methods have been utilized in the attempts to examine this significant contributor to the redox environment in T lymphocytes (10).

One method that has been used is the perturbation of the electron transport chain of the mitochondria, and thus, alteration of ROS production from these sites. For example, using siRNA to Complex I of the mitochondria to limit O₂^•− from this site, it was shown that IL-2 and IL-4 production was significantly inhibited in T lymphocytes isolated from patients with atopic dermatitis (71). Using similar Complex I inhibition strategies, it was shown that mitochondria are a source of activation-induced ROS via PKCtheta translocation to the mitochondria (69). In addition, utilizing rotenone to inhibit Complex I-produced ROS led to impaired T lymphocyte proliferation, decreased cytokine production, and enhanced T_REG polarization with concurrent inhibition of T_H17 differentiation (122, 186). Complex III-derived ROS have also been demonstrated to be essential to both CD4⁺ and CD8⁺ T lymphocyte functions by utilizing a knockout mouse model of the Rieske iron/sulfur protein (145).

While these studies have elucidated several roles for mitochondrial-derived ROS in T lymphocyte function, they are limited by the fact they have affected mitochondrial metabolism in addition to ROS production, which may confound their interpretation.

To counteract this limitation, our group has examined the role of mitochondrial O₂^•− by the use of manganese superoxide dismutase (MnSOD) knockout mice. By eliminating the sole O₂^•− removal enzyme in the mitochondrial matrix, we have generated a system of chronically elevated steady-state mitochondrial O₂^•− without the direct perturbation of metabolism. Using this model, we first identified that uncontrolled mitochondrial O₂^•− led to significant impairment in T lymphocyte development, which ultimately caused a significant immunocompromised state (13). However, we identified that supplementation of mitochondrial O₂^•− scavengers early in life also led to impaired T lymphocyte development, which suggested the importance of mitochondrial redox signaling in adaptive immune system maturation (13). Furthermore, others have shown that MnSOD overexpression attenuates T lymphocyte activation-induced ROS production and downstream NFκB and AP-1 signaling (70). Together, these studies demonstrate a significant role for mitochondrial redox signaling in T lymphocyte function by directly manipulating ROS flux and sparing metabolic capacity.

Small-molecule mitochondrial antioxidants

Studies from our laboratory have elucidated that catecholamine-mediated cytokine profiles in T lymphocytes are partially mediated by mitochondrial O₂^•−, and the addition of MitoTEMPO, a mitochondrial-sensitive O₂^•− scavenger, could rescue these alterations (14). Other groups have also identified a role for mitochondrial ROS via the use of MitoTEMPO. It has been reported that scavenging elevated mitochondrial O₂^•− with MitoTEMPO promoted infiltration of cytotoxic T lymphocytes into tumors (152). MitoTEMPO treatment also has been shown to increase IFNγ and tumor necrosis factor-alpha (TNFα) production in cytotoxic T lymphocytes isolated from patients with a chronic hepatitis B infection (35). Furthermore, MitoTEMPO was shown to reverse the suppression of T lymphocyte function through attenuation of mitochondrial ROS produced by iron oxide nanoparticles (147). In addition, MitoTEMPO was able to reverse CD4⁺ T lymphocyte depletion in an in vivo model of nonalcohol fatty liver disease (92).

While mitochondrial ROS are essential for normal T lymphocyte activation and differentiation (previously discussed), these data suggest that reduction of mitochondrial ROS (or possibly enhanced reductive stress due to excess MitoTEMPO) elicits a proinflammatory phenotype in T lymphocytes. However, compounds such as MitoTEMPO have also been shown to be mitochondrial uncouplers (134), confounding the interpretation of many studies that utilize this pharmacological agent.

Endogenous antioxidants

To avoid the complication of mitochondrial uncoupling, other investigators have also utilized the approach of ROS perturbations through antioxidant gene modification to understand the role of the redox environment in T lymphocyte function. For example, overexpression of MnSOD in the Jurkat T lymphocyte cell line has been shown to enhance the downstream signal transduction molecules necessary for T lymphocyte activation (47); counterintuitive to what was previously discussed with MnSOD overexpression in T lymphocytes (70). However, it remains unknown if these effects were due to accelerated mitochondrial H₂O₂ formation, decreased mitochondrial O₂^•−, or depleted manganese bioavailability given the overabundance of MnSOD.

In addition, T lymphocytes deficient in glutathione peroxidase 1 (GPX1) have increased steady-state levels of H₂O₂, which appears to lead to elevated IL-2 production and enhanced proliferation ex vivo (178). GPX1 knockout also promotes T_H1 differentiation over T_H2, suggesting that elevated endogenous H₂O₂ is proinflammatory (178). This is another example of controversial evidence in the literature, as exogenous H₂O₂ has repeatedly been shown to attenuate T lymphocyte activation and T_H1 polarization as previously discussed. Catalase has also been shown to be upregulated in T lymphocyte cultures ex vivo, which appears to aid in their survival during activation-induced cell death (93). Interestingly, catalase-deficient mice show no defect in their T lymphocyte populations, suggesting limited importance of this enzyme in the resting state of these cells (55).

Other antioxidants such as peroxiredoxin-I and II (PrdxI and PrdxII), thioredoxin (TRX), and glutathione (GSH) also appear to impact T lymphocyte physiology. PrdxI has been shown to regulate the T_H2-driven airway inflammation in allergic asthma, as in vivo studies found increased IL-2 secretion in the lung homogenates of PrdxI-deficient mice (59). Furthermore, T lymphocytes from these animals showed increased proliferation in coculture experiments with DCs (59). PrdxII abundance is elevated in cytotoxic T lymphocytes after T cell receptor (TCR) stimulation (102), suggesting a potential functional role of this enzyme in T lymphocyte activation. Indeed, in in vivo models of acute viral and bacterial infections, PrdxII-deficient mice showed increased expansion of effector and memory cytotoxic T lymphocytes implying that PrdxII plays a suppressive regulatory role in these adaptive immune cells (102).

TRX supplementation can increase T lymphocyte proliferation and survival ex vivo (73, 148). T_REGS also have increased production of TRX, which appears to protect against oxidative damage during times of enhanced inflammation (109). Interestingly, TRX overexpression in T lymphocytes showed clustering of memory T lymphocytes and diminished effector function in tumor microenvironments, suggesting TRX's antioxidant properties may potentiate memory formation and an accelerated immune response (15).

Conversely, inhibition of TRX reductase (which would deplete reduced TRX) via low-dose auranofin appears to enhance T lymphocyte activation (169). This is counterintuitive with the fact that auranofin is utilized clinically to suppress T lymphocyte functions in rheumatoid arthritis. Additional studies have demonstrated that auranofin's ability to enhance T lymphocyte activation may be solely dependent on GSH levels (169), which have been reported to be low in several pathologic autoimmune conditions and can explain these confounding results (21, 101, 105).

Furthermore, depletion of GSH in T lymphocytes by the use of buthionine sulfoximine significantly impairs effector T lymphocyte activation as well as T_REG ability to suppress effector T lymphocytes (51, 169, 193). In addition, supplementation of reducing and pro-GSH molecules such as N-acetyl cysteine or β-mercaptoethanol also increases proliferation of T lymphocytes ex vivo (179). This may not be entirely an intracellular phenomenon, as this may be due to the ability of these small-molecule thiols to reduce cystine to cysteine in the extracellular environment, which is necessary for GSH synthesis in naive T lymphocytes (153).

Overall, it is clear that antioxidants act as regulators of oxidant levels and redox metabolism, and play an essential role in regulating T lymphocyte ROS signaling. However, the many conflicting findings suggest that these processes are highly dynamic and susceptible to variations in experimental design.

Nuclear factor erythroid 2-related factor 2

A master regulator of several antioxidant pathways is the nuclear factor erythroid 2-related factor 2 (Nrf2) transcription factor. Nrf2 is a constitutively expressed transcription factor that is sequestered in the cytoplasm by kelch-like ECH-associated protein 1 (Keap1) and constantly targeted for proteasomal degradation. Upon oxidation of specific cysteine residues on Keap1, Nrf2 is released and able to enter the nucleus where it upregulates expression of several antioxidant enzymes. This antioxidant response by Nrf2 alters the inflammatory phenotype of T lymphocytes lending further support to redox regulation in these cells. For example, in a model of induced acute kidney injury (AKI), T lymphocyte-specific overexpression of Nrf2 promoted T_REG polarization and an antioxidant phenotype, which significantly protected from both the histological and functional outcomes of AKI (117). In addition, treatment of T lymphocytes with tert-butylhydroquinone, a potent Nrf2 activator, promoted T_H2 lymphocyte skewing over T_H1 through increased IL-4 and interleukin 5 (IL-5) production, GATA3 transcription factor activation, and suppressed IFNγ production (135, 189).

In contrast, proinflammatory and pathological conditions are exacerbated in the absence of Nrf2. Constitutive Nrf2 knockout mice have shown elevated hepatic inflammation and damage in a concanavalin A-driven model of T lymphocyte-mediated acute liver injury (121). In addition, Nrf2 knockout mice have shown increased T lymphocyte-driven inflammation and exacerbated development of experimental autoimmune encephalomyelitis (EAE) (67). These proinflammatory events may be due to decreases in GSH levels that have been reported in these models (105), combined with the observation that Nrf2 deficiency enhances responses to proinflammatory stimuli in a GSH-dependent manner (106). Overall, the Nrf2 antioxidant pathway appears critical in the fine-tuning of the inflammatory response in the ability to control the level of oxidant species that may polarize T lymphocytes to either a pro- or anti-inflammatory phenotype.

Oxygen bioavailability

A final redox variable that affects virtually all T lymphocyte research described to date is oxygen concentration. T lymphocytes normally reside in specific organs and lymphoid tissues at low oxygen concentrations (2%–3%), but the majority of studies have performed their investigations ex vivo in ∼18% oxygen (when taking into account 5% carbon dioxide and 95% humidity) (130). This discrepancy has major impacts on T lymphocytes, as those grown in hyperoxic (18%) conditions show enhanced proliferation and cytokine production compared with those cells grown at physiological oxygen levels (2, 3). It is likely that differential oxygen concentrations alter the redox environment of T lymphocytes, which would ultimately change critical redox signaling pathways in these cells. Another possibility is that oxygen bioavailability affects cellular processes via hypoxia inducible factor (HIF) signaling. Indeed, stabilization and activation of HIF-1α under low-oxygen conditions have been shown to increase the secretion of IFNγ in T lymphocytes (139).

However, a different study showed the exact opposite effect in that HIF-1α expression and activation during low oxygen decreased IFNγ secretion and other functions of T lymphocytes (91). An additional study has demonstrated that HIF-1α (and hypoxia) may also be the switch between T_H17 and T_REG lymphocytes, furthering the complexity of oxygen biology in these cells (26). HIF signaling is well known to alter metabolism, which may be a major factor in determining the phenotype of T lymphocytes under differential oxygen environments. Interestingly, in addition to the relative abundance of oxygen, HIF signaling is highly responsive to cellular and mitochondrial ROS levels (16, 68). Thus, HIF signaling may be one link into how redox signaling couples with cellular metabolic processes in T lymphocytes; processes that have recently been shown to be essential to adaptive immune function.

A summary of the redox-regulated processes discussed in this section is found in Table 1.

Table 1.

Redox-Regulated Processes Involved in T-lymphocyte Physiology

ROS/RNS manipulation	Physiological effects	References
H2O2	Supplementation: ↓IFNγ in TH1 cells; ↑IL-4 in TH2 cells	(40, 97)
	Supplementation: impairment of AP-1, NFAT, NFκB signaling: ↓IL-2	(36, 37, 81, 125, 151)
	↑T lymphocyte homing to sites of inflammation	(115)
	DMNQ: ↑TH2 cytokine profiles	(75)
MnSOD	KO: ↓T lymphocyte development	(13)
	OE: ↑Increased T lymphocyte activation signal transduction	(47)
NOX2	KO: ↑TH1 polarization; ↓TH2 polarization	(63, 149)
	p47phox KO: ↓TH1 polarization; ↑TH17 polarization	(162)
DUOX1	KD: ↓TCR signaling and cytokine production	(80)
NO	Supplementation: ↑IL-12 receptor expression; ↑TH1 polarization	(114)
	Supplementation: ↑TREG polarization; ↓TH17 polarization	(112, 113)
NOS	Inhibition: ↑TH2 polarization	(19)
	eNOS KO: ↓Proinflammatory response	(57)
	iNOS KO: ↓TREG polarization; ↑IFNγ	(65, 168)
GSH	Depletion: ↓T lymphocyte activation	(51)
GPX1	KO: ↑TH1 polarization; ↑IL-2; ↑T lymphocyte proliferation	(178)
TRX	Supplementation: ↑T lymphocyte proliferation and survival	(73, 148)
NRF2	Activation: ↑TH2 polarization; ↓TH1 polarization	(135, 189)
	KO: ↑Proinflammatory response	(67, 105, 106, 121)
HIF1α	Stabilization: ↑IFNγ; ↑TH1 polarization	(139)
	Stabilization: ↓IFNγ; ↓TH1 polarization	(91)
	Stabilization: ↑TH17 polarization; ↓TREG polarization	(26)

AP-1, activator protein 1; DMNQ, 2,3-dimethoxy-1,4-naphthoquinone; DUOX1, dual oxidase 1; eNOS, nitric oxide synthase 3; GPX1, glutathione peroxidase 1; GSH, glutathione; H₂O₂, hydrogen peroxide; HIF, hypoxia inducible factor; IFNγ, interferon gamma; IL-2, interleukin 2; IL-4, interleukin 4; IL-12, interleukin 12; iNOS, nitric oxide synthase 2; KD, knockdown; KO, knockout; MnSOD, manganese superoxide dismutase; NFAT, nuclear factor of activated T lymphocytes; NFκB, nuclear factor kappa-light-chain-enhancer of activated B; NO, nitric oxide; NOS, nitric oxide synthase; NOX, NADPH oxidase; NRF2, nuclear factor erythroid 2-related factor 2; OE, overexpression; RNS, reactive nitrogen species; ROS, reactive oxygen species; TCR, T cell receptor; T_REG, T regulatory cells; TRX, thioredoxin.

Metabolic Control of T Lymphocytes

Naive T lymphocytes

Due to the plasticity of differentiation and extreme microenvironments T lymphocytes are exposed to, these cells possess dynamic and highly adaptive metabolic processes. T lymphocytes generate the majority of their energy through two main metabolic processes: oxidative phosphorylation (OXPHOS) and glycolysis. Naive T lymphocytes' energy requirements are vastly different than activated and differentiated cells, the latter requiring a remodeling of metabolism to maintain their functionality.

Naive T lymphocytes exist in a quiescent state, utilizing primarily amino acids (AA), lipids, and glucose through OXPHOS to fuel minimal adenosine triphosphate production (38). Naive T lymphocytes require extrinsic signals from cytokine receptors and TCR signaling to maintain their basal metabolic state and avoid cell death. Interleukin 7 (IL-7) promotes glucose uptake through regulation of glucose transporter 1 (Glut1) expression and is associated with maintaining naive T lymphocyte homeostasis (160, 177). Conditional deletion of the IL-7 receptor in naive T lymphocytes inhibits glycolysis, and thus promotes atrophy of these cells (64). Upon activation, naive T lymphocytes become effector T lymphocytes by initiating anabolic mechanisms to produce energy and metabolites to promote biosynthesis, growth, functionality, and differentiation (Fig. 3).

FIG. 3.

Metabolic reprogramming during the T lymphocyte life cycle. Naive T lymphocytes have a lower energetic demand and rely heavily upon OXPHOS. Activation and subsequent growth, proliferation, and differentiation will put a high energetic demand on what is now an effector T lymphocyte, which will rely on glycolysis and glutaminolysis for ATP production. During the contractile phase, a portion of effector T lymphocytes will be stored as memory T lymphocytes, which do not have a high energy demand but possess increased mitochondrial SRC. However, upon reactivation, this SRC allows for a quick utilization of both OXPHOS and FAO, leading to a rapid immunological recall to a reoccurring pathogen. ATP, adenosine triphosphate; FAO, fatty acid oxidation; Glut1, glucose transporter 1; Glut3, glucose transporter 3; OXPHOS, oxidative phosphorylation; SRC, spare respiratory capacity; T_EFF, effector T lymphocytes; T_M, memory T lymphocytes; T_N, naive T lymphocytes. Color images are available online.

Effector T lymphocytes

Once activated, T lymphocytes begin to rapidly proliferate and perform their effector functions, requiring a metabolic reprogramming. Glucose and glutamine metabolisms are increased with concurrent decreases in lipid metabolism during T lymphocyte activation (127). This enhanced glycolysis even in the presence of abundant oxygen is termed aerobic glycolysis, and was previously thought to be the only metabolic source of energy for effector T lymphocytes. While glycolysis remains the primary energy source of effector T lymphocytes, this view of glycolytic exclusivity is rapidly becoming outdated with recent research demonstrating a critical role for the mitochondria in effector function (145, 171).

Activated T lymphocytes, with an accelerated metabolism, do not remain active for the rest of their lives. After combating the pathogen or antigenic stimuli, a majority will die off and the remainder will shift toward a quiescent memory state that does not require an accelerated production of energy to sustain proliferation, differentiation, or effector function (12, 118).

Memory T lymphocytes

Memory T lymphocytes have a lower energy demand than effector lymphocytes, which requires an additional metabolic reprogramming to differentiate these cells to this memory state. Shifting back toward OXPHOS and fatty acid (FA) metabolism, memory T lymphocytes rely heavily on their mitochondria for energy demands. In fact, overexpression of glycolytic enzymes to promote and maintain glycolysis in CD8⁺ T lymphocytes inhibited the ability to form long-lived memory T cells (154). Interestingly, cytotoxic memory T lymphocytes rely less on free extracellular FA, and instead utilize extracellular glucose (converted to FA by lysosomal acid lipase) to fuel fatty acid oxidation (FAO) and OXPHOS (119).

Despite their low resting metabolic demand, memory T lymphocytes are engineered to recall and rapidly respond to a pathogen. Memory T lymphocytes have greater mitochondrial mass and spare respiratory capacity (SRC) than naive cells, believed to be mediated, in part, by interleukin 15 (IL-15) signaling (174). This allows them to respond more rapidly, secrete more cytokines, and promote a greater oxidative capacity compared with naive and primary effector T lymphocytes (176).

T lymphocyte subtypes

Not only is metabolism tailored to the transitioning of T lymphocytes from naive to effector, and lastly to memory states, but it is also unique to individual subtypes and promotes differentiation of one cell type over another. Not all effector T lymphocytes are created equal, and between the various subtypes of helper and cytotoxic T lymphocytes, there is a measurable difference in the bioenergetic demand between these cells that is linked to the energy requirements of their specific functions (140). For example, Glut1 expression in cytotoxic T lymphocytes is not critical for transitioning from naive to effector steady states, likely due to the increased redundancy of glucose transporters present. Indeed, in Glut1-deficient mice, cytotoxic T lymphocytes show elevated glucose transporter 3 (Glut3) and glucose transporter 6 (Glut6) expression (94). Interestingly, acutely stimulated naive CD4⁺ T lymphocytes show increased Glut1 expression and the subsequent enhanced glucose uptake to fuel glycolysis, which promotes increased proliferation and cytokine production (103).

While effector CD4⁺ T lymphocytes differentiated into T_H1, T_H2, and T_H17 subtypes also show increased Glut1 expression, activated T_REG cells show decreased Glut1 expression and promote lipid oxidation (103). Metabolically, T_REG cells differ greatly from helper T lymphocytes in their energy reliance sourced from FAO over glycolysis, and the subsequent activated transcription factors and metabolic checkpoints illustrate this observation. While the metabolic differences between the various subtypes of helper T lymphocytes are not extensively studied, there are some metabolic distinctions made between T_H1, T_H2, and T_H17 cell subtypes.

Metabolic checkpoints

Metabolic checkpoints function to promote specific helper and cytotoxic T lymphocyte subtype differentiation. One of these metabolic checkpoints is the inhibition of pyruvate dehydrogenase (PDH) by PDH kinase 1 (PDHK1), which is essential in promoting T_H17 over T_H1 differentiation, in an in vivo model of EAE (46). Additional checkpoints required for T_H2 polarization are the mammalian target of rapamycin (mTOR) complex 2 (mTORC2) and serum- and glucocorticoid-regulated kinase 1 (SGK1), which inhibit production of IFNγ, thus preventing T_H1 differentiation (53).

Both mTOR complex 1 (mTORC1) and mTORC2 are shown to be a master regulator of cytotoxic T lymphocytes (34, 190). While mTORC2-deficient mice showed enhanced memory formation and an accelerated recall response after antigen re-exposure in vivo, short-lived effector cytotoxic T lymphocytes functioned improperly due to their reliance mTORC2 expression (190).

mTOR activity is critical to metabolic signaling, and recent work has provided direct evidence of mTOR being a master regulator that controls T lymphocyte plasticity. mTOR acts to initiate the metabolic switch from OXPHOS to glycolysis, and in doing so regulates cell fate decisions of T lymphocytes. For example, through genetic modification of mTOR signaling, mice showed inhibited ability of T_H17 lymphocytes to transdifferentiate into T_H1-like cells (72).

Another metabolic sensor, HIF-1α, is also responsible for regulating the balance between T_H17 and T_REG differentiation. HIF-1α has been reported to be essential in controlling T_H17 differentiation over T_REG cells through promoting transcriptional activation of RAR-related orphan receptor gamma T (RORγt), as HIF-1α-deficient T lymphocytes failed to induce T_H17-dependent EAE (26). While early metabolic reprogramming in T_REG cells is shown to be controlled by HIF-1α through regulating forkhead box P3 (Foxp3) expression, the late stage appears to shift via signaling through the aryl hydrocarbon receptor (26, 99).

T_REG/T_H17 differentiation is also regulated by acetyl-CoA carboxylase 1 (ACC1), which is involved in FA synthesis and promotes T_H17 differentiation while suppressing T_REG differentiation. Pharmacological inhibition or T lymphocyte-specific deletion of ACC1 in vivo inhibits T_H17-induced autoimmune disease development (7).

Last, AMP-activated protein kinase (AMPK) has also been shown to promote differentiation into T_REG cells through phosphorylating ACC1 and inhibiting FA synthesis (52, 103, 104, 188). Interestingly, AMPK has a different regulatory role with cytotoxic T lymphocytes, as AMPK-deficient mice show impaired memory formation in an in vivo model of Listeria monocytogenes infection (138). Overall, the dynamic nature of metabolism between various subtypes of naive, effector, and memory T lymphocytes makes these cells highly dependent upon the bioavailability of metabolites, AA, and lipids to fuel their physiological processes.

Oxygen

One major factor regulating the bioavailability of these metabolic substrates is the local oxygen concentration. The oxygen concentration in the surrounding microenvironment is critical for physiological processes involved in T lymphocyte metabolism, and OXPHOS relies exclusively on available oxygen. Like many cells, T lymphocytes possess regulatory genes and oxygen sensors that allow them to adapt to changing oxygen concentrations. For example, the prolyl-hydroxylase (PHD) proteins that regulate HIF isoforms utilize molecular oxygen, which allows them to act as oxygen sensors (11). These sensors, in the presence of oxygen, work to degrade HIF isoforms via a hydroxylation/ubiquitination mechanism (62). Constitutive knockout of all three isoforms of the PHD enzymes (which would stabilize HIF) promoted increased CD4⁺ and CD8⁺ T lymphocyte trafficking and improved effector T lymphocyte function to pulmonary regions with cancer metastases (22). HIF-1α-deficient mice show impaired CD4⁺ and CD8⁺ T lymphocyte trafficking and effector function in the surrounding tumor microenvironment, further illustrating the importance of HIF signaling in immune regulation of tumor progression (126). These enhanced effector functions can likely be attributed to enhanced HIF activity, which would promote glycolytic activity.

Furthermore, PHD proteins may also inhibit differentiation of T lymphocytes, as aforementioned, HIF-1α is shown to promote both T_H1 and T_H17 differentiation while inhibiting T_REG polarization (26). Interestingly, HIF-1α expressed in macrophages inhibits T lymphocyte function in the surrounding tumor microenvironment (29), which is somewhat counterintuitive.

However, the tumor microenvironment is quite complex as it may be both nutrient and oxygen poor. Hypoglycemic conditions promote OXPHOS, but hypoxic conditions promote glycolysis, which puts a metabolic strain on T lymphocytes. T lymphocytes may compensate in these conditions by reprogramming their metabolism to rely heavily upon FAO, which preserves their effector functions in an oxygen- and nutrient-poor environment (154, 192). In summary, it is evident that oxygen and metabolites shape T lymphocyte activation and function. The close relationship between oxygen, metabolism, and ROS suggests that a tight couple exists between redox signaling and metabolism, but this area of research has only recently begun to be examined in the context of T lymphocytes.

The Redox-Metabolic Couple in T Lymphocytes

Based on the aforementioned studies, it is clear that both redox signaling and metabolism shape T lymphocyte effector functions, yet, the effects of these processes in T lymphocytes are often discussed as separate entities. Given the extensive list of molecules and pathways that both redox signaling and metabolism share, we propose that these processes are often coupled and may regulate T lymphocyte activity via one another. There are several pieces of literature that support this hypothesis showing that ROS affect metabolism and metabolism affects ROS, and the outcome is an altered T lymphocyte phenotype.

Metabolic reprogramming alters intracellular ROS levels; however, it has been shown that many metabolic control enzymes are also redox regulated. For example, NFAT activity has been shown to be potentiated by elevated mitochondrial ROS, which occurs during the increased metabolic demand of an activated T lymphocyte (145). This metabolic-redox activation of NFAT ultimately potentiates IL-2 production from T lymphocytes to promote proliferation and inflammation (145). Interestingly, NFAT also upregulates glucose transporters and enhances glycolysis, thus providing a feed-forward loop in perpetuating T lymphocyte activation (77, 165).

While harmful levels of intracellular ROS (or simply different forms as previously discussed) inhibit proliferation and activation and can even lead to cell death, physiological levels are critical for appropriate signaling to initiate metabolic reprogramming. These are not specific concentrations, but rather the beneficial to harmful effects of ROS exist in T lymphocytes along a spectrum dependent upon countless factors, including activation, differentiation, antioxidant status, as well as metabolic rate.

T lymphocytes have mechanisms in place to diminish the harmful levels of elevated ROS to promote proliferation and activation through metabolic switching. For instance, GSH-mediated ROS scavenging is one mechanism that appears critical in regulating the metabolic control of T lymphocytes. Glutamate cysteine ligase (GCL) is the rate-limiting step involved in GSH synthesis, and expression of this enzyme (thus increased GSH production) is upregulated during T lymphocyte activation (20). Inhibiting GSH synthesis, through deletion of subunits of GCL, not only increases intracellular ROS but it also inhibits T lymphocyte differentiation (85).

Furthermore, GSH has been shown to be essential for NFAT, mTOR, and c-Myc activation; all of which promote T lymphocyte activation, proliferation, and glycolytic metabolism (Fig. 4) (39, 96). While it appears counterintuitive that both GSH and mitochondrial ROS are needed for NFAT activation, this is a prime example of the complexity of redox signaling within T lymphocytes. It could be speculated that mitochondrial ROS activate a specific subset of transcription factors in close proximity to the mitochondria that are needed for NFAT expression, while GSH is able to either attenuate other sources of inhibitory ROS or possibly glutathionylate additional NFAT-inducing transcription factors in a ROS-independent manner. In any case, GSH is essential in maintaining redox homeostasis and supporting metabolic reprogramming to potentiate T lymphocyte activation and differentiation, and ablation of GCL or alterations in GSH pools inhibit the adaptive immune response.

FIG. 4.

ROS regulation involved in T lymphocyte metabolic reprogramming. T lymphocyte activation promotes increased ROS production, which has been shown to have both positive and negative effects on downstream metabolic and regulatory pathways. The conflicting effects of ROS in T lymphocytes can likely be attributed to differences in experimental design, which could affect the concentration, type, and subcellular localization of the ROS. Overall, ROS shape the metabolic environment of T lymphocytes, and both processes are critical to the normal activation, differentiation, and function of these immune cells. GSH, glutathione; GSSG, glutathione disulfide; mTOR, mammalian target of rapamycin; NFAT, nuclear factor of activated T lymphocytes; ROS, reactive oxygen species; TCR, T cell receptor. Color images are available online.

ROS are also essential for AMPK activation, mTOR activation, and perpetuation of aerobic glycolysis (54). It has been shown by the use of a porphyrin-based antioxidant that attenuation of ROS-enhanced AMPK signaling blocked mTOR signaling, and overall decreased glycolytic metabolism and the ability of T lymphocytes to properly activate and function (131). This is an example of the need of an appropriate balance of ROS levels in T lymphocytes, as utilizing a potent scavenger of intracellular ROS diminished concentrations to such an extent that metabolic reprogramming was inhibited.

The beneficial or harmful effects of ROS are reliant on maintaining homeostasis, and drifting far to either end of that balance can be detrimental to the cell. These examples demonstrate ways in which ROS may affect T lymphocyte function via affecting metabolic control enzymes. While this is one way in which redox signaling and metabolism are coupled to affect T lymphocyte function, another is through the potential for epigenetic modification.

Epigenetic regulation of gene expression is a broad term that encompasses DNA modifications, histone modifications, chromatin accessibility, and more. The array of modifications to these genetic structures exponentially increases the complexity of the epigenome, however, one unique theme among epigenetic modifications appears to be universal. With few (if any) exceptions, all epigenetic modifications found on DNA or histones are highly associated with specific metabolic processes and metabolites, which are all sensitive to redox processes (25). For example, the most prominent and well-studied epigenetic modification to the DNA is cytosine methylation, which occurs via the metabolite S-adenosyl-methionine (SAM). SAM is generated via the methionine cycle, and is also tightly regulated via the transsulfuration pathway that leads to GSH synthesis (191). Thus, alterations in the redox environment that affect GSH pools also lead to changes in SAM bioavailability, ultimately affecting the epigenetic control of DNA.

We have recently observed this phenomenon in MnSOD knockout T lymphocytes, which have elevated steady-state mitochondrial O₂^•− levels. In this work, we identified that elevated mitochondrial O₂^•− in T lymphocytes leads to a reduction in global DNA methylation, which was linked to a significant loss of SAM and GSH pools (107). We postulate that since mitochondrial O₂^•− unlikely affects nuclear epigenetic regulation directly due to its small diffusion radius, it instead works via perturbations in the cellular metabolite pools that affect these epigenetic processes.

Another example of redox- and metabolism-sensitive epigenetic enzymes would be the α-ketoglutarate-dependent family of hydroxylases. This class of enzymes is quite broad and includes the previously discussed PHDs that regulate HIF signaling, the ten-eleven translocation (TET) enzymes that regulate DNA (hydroxy)methylation, the Jumonji C (JmjC)-domain containing enzymes that regulate histone methylation, and several others. All of these enzymes require molecular oxygen, reduced iron, and α-ketoglutarate to function, and may be inhibited by elevated ROS, succinate, and fumarate levels; the latter being elevated during oxidative conditions (128). These metabolites and ROS inhibit many α-ketoglutarate-dependent hydroxylases, demonstrated to alter epigenetic modifications (82, 144, 183).

Interestingly, recent work has demonstrated the importance of the malate/aspartate shuttle of the mitochondria (which exchanges malate for α-ketoglutarate) in the proliferation and differentiation of T lymphocytes (4). While the authors did not evaluate redox directly in this article, they were able to show that the abundance of mitochondrial metabolites such as α-ketoglutarate directly altered nuclear epigenetic gene expression, which modified T lymphocyte polarization (4). As previously discussed, virtually all mitochondrial metabolites are affected by alterations in ROS; therefore, this demonstrates yet another example of ROS-mediated epigenetic control via cellular metabolite pools.

A final example of how redox and metabolism couple to affect T lymphocyte function is the class of epigenetic modifying enzymes known as the sirtuins (SIRT). SIRTs carry out an array of enzymatic reactions, including deacetylation, desuccinylation, and demalonylation, in an NAD⁺-dependent manner (184). Seven highly compartmentalized isoforms of SIRTs exist in mammals, and their cellular location dictates their primary function.

One highly studied function of SIRTs is the protection of the cell from excessive oxidative damage. For example, SIRT3 is exclusively found in the mitochondria, and has been shown to regulate the level of ROS in this organelle by controlling the acetylation status of MnSOD (132, 161). SIRT3-deficient mice show decreased SOD2 transcription and MnSOD activity in fibroblasts, and restoration of SIRT3 in these mice diminished MnSOD acetylation and restored MnSOD activity (161). Due to SIRT3's activity being highly dependent upon the metabolically derived substrate NAD⁺ (84), the regulatory control of mitochondrial ROS by this enzyme is tightly coupled with metabolism. For example, it could be postulated that a buildup of NAD⁺ in the mitochondria would signify a high flux of electrons through the electron transport chain, which would enhance the level of ROS produced in this organelle. The elevated NAD⁺ levels would increase SIRT3 activity, which in turn would deacetylate MnSOD potentiating its ability to remove superoxide from the mitochondria (Fig. 5).

FIG. 5.

SIRT3 regulation involved in the deacetylation of MnSOD to maintain mitochondrial ROS homeostasis. Increased oxidative metabolism initiates an elevated electron flux through the electron transport chain, which inevitably results in elevated mitochondrial ROS. While increased oxidative metabolism produces elevated mitochondrial ROS, NADH is being oxidized to an ever growing pool of NAD⁺. With elevated NAD⁺ concentrations, SIRT3 activity would increase and would deacetylate MnSOD and potentiate removal of superoxide from the mitochondria. MnSOD, manganese superoxide dismutase; SIRT, sirtuins. Color images are available online.

SIRT3 has also been shown to directly interact with several components of the electron transport chain, which could also serve as a regulatory checkpoint for electron flux and ROS generation (185). It remains to be seen if this pathway occurs in T lymphocytes, but this mechanism would create an efficient and elegant feedback to protect mitochondrial health during various states of metabolic flux.

While the SIRT enzymes have been extensively studied, less is known with regard to their specific function in T lymphocytes. The primary isoform examined in these adaptive immune cells is SIRT1, which has been shown to impact T lymphocyte activation, differentiation, and function. Interestingly, SIRT1 has been demonstrated to inhibit T lymphocyte activation through negative regulation of the NFκB transcription factor (17, 78, 79). SIRT1's immune suppression also impacts inflammatory cytokine production and altered differentiation of T lymphocytes. In an in vitro model of endotoxin tolerance, the blunting of NFκB signaling via SIRT1 has a direct impact on limiting TNFα and interleukin 1β (IL-1β) expression (88). Furthermore, through deacetylation of signal transducer and activator of transcription 3 (STAT3), SIRT1 activity was also shown to prevent proinflammatory T_H17 differentiation in both human and mice T lymphocytes (111). Interestingly, while SIRT1 is responsive to changes in the metabolic environment through sensing NAD⁺ levels, its activity is also able to shape the metabolic landscape. For example, by using SIRT1-deficient mice, it has been demonstrated that SIRT1 inhibits mTOR (87), which is responsible for downstream activation of HIF-1α as well as the metabolic switch from OXPHOS toward glycolysis as previously discussed.

Given the importance of metabolic switching during T lymphocyte activation and differentiation, this places SIRT1 as a critical upstream mediator of these processes, and the SIRT1/mTOR/HIF-1α axis is at a critical intersection between redox signaling and metabolic reprogramming for immune cell differentiation (129). However, little research has been performed examining the redox-metabolic-epigenetic linkage in T lymphocytes, especially with regard to the SIRT enzymes. This information will be a critical next step in understanding how these processes regulate normal T lymphocyte physiology, and also how when disrupted, may lead to altered inflammation with pathological consequences.

The Redox-Metabolic Couple of T Lymphocytes in Hypertension

Inflammatory processes are linked to the development of hypertension, and elevated blood pressure is a significant risk factor for the development of cardiovascular diseases and end-organ damage. It still remains unclear if proinflammatory T lymphocytes are the cause or effect of high blood pressure, but it is evident that these immune cells and their production of proinflammatory cytokines potentiate the hypertensive phenotype.

Effector and memory T lymphocytes are critical cell types that have been implicated in the development of hypertension, and are shown to exacerbate end-organ damage and cardiovascular diseases. Almost half a century ago, early reports demonstrated that rodents lacking T lymphocytes demonstrated a blunted pressor response to various hypertensive challenges (155–158). Additional work demonstrated that depletion of T lymphocytes using the pharmacological agent mycophenolate mofetil could attenuate various forms of experimental hypertension (133, 136). More recently, the use of genetically altered mice that lack T lymphocytes also showed diminished hypertensive responses (24, 50). Interestingly, adoptive transfer of T lymphocytes from various models of hypertensive rats was elegantly shown to make unchallenged rats hypertensive (120). Together, these studies illustrate strong evidence in favor of a T lymphocyte-mediated effect in the development and maintenance of hypertension.

The underlying cause of T lymphocyte activation in hypertension still remains elusive, but the pathophysiological changes to various organ systems in this disease may create an environment that promotes perturbation of the immune system. A hallmark of hypertension is increased sympathetic drive and norepinephrine (NE) outflow. Lymphoid organs such as the spleen and lymph nodes possess only sympathetic innervation, which enhances the contact of T lymphocytes to catecholamines. We have demonstrated that T lymphocytes exposed to elevated levels of NE show increased expression and secretion of interleukin 6 (IL-6) and interleukin 17a (IL-17a), which appear to be regulated, in part, by mitochondrial ROS and perturbations in mitochondrial metabolism (14, 108). Importantly, IL-6 and IL-17a are implicated in the pathogenesis of hypertension (9, 95), which demonstrates a link between neurotransmission, inflammation, redox, and metabolism in potentially regulating this disease.

Furthermore hypertension-induced elevations in NE and ROS are both shown to potentiate oxidative isoketal formation that accumulates in DCs. Formed from metabolic FAO, isoketals lead to modified protein structure and function through reactive oxidation, which leads to these molecules becoming antigenic to the immune system (76, 180, 182). Isoketal accumulation in DCs was shown to enhance their activation and further potentiate increased proinflammatory T lymphocyte activation, including IL-17a and IL-6 production (76, 180). These factors contribute toward the T_H17/T_REG imbalance that potentiates a proinflammatory phenotype and exacerbates hypertension development (Fig. 6) (30, 33, 43, 66, 74, 89, 90, 167, 173, 194).

FIG. 6.

A working model of immune-mediated contributions to hypertension development. Metabolic insults that lead to increased NE outflow from enhanced sympathoexcitation trigger both DCs and T lymphocytes to potentiate increased proinflammation. NE and increased ROS enhance isoketal accumulation in DCs, in turn initiating DC activation and production of proinflammatory IL-6, as well as promoting T lymphocyte activation and elevated expression of IL-6 and IL-17a. Furthermore, NE enhances IL-6 and IL-17a expression through additional mitochondrial redox mechanisms. Finally, elevated IL-6 expression works to inhibit T_REG cell differentiation, while potentiating T_H17 cell differentiation, to create a T_H17/T_REG cell proinflammatory imbalance that further exacerbates the hypertensive phenotype creating a feed-forward loop. DC, dendritic cell; IL-17a, interleukin 17a; MetS, metabolic syndrome; NE, norepinephrine. Color images are available online.

Complications associated with hypertension and the subsequent cardiovascular disease development are also highly associated with a strong memory T lymphocyte response. For example, children diagnosed with primary hypertension show increased proinflammatory memory T lymphocytes, and this memory population correlates with disease severity in terms of arterial stiffness and left ventricular hypertrophy (41). While elevated circulating populations of memory T lymphocytes may be harmful, it is more likely that their localized accumulation in tissue-specific areas is the root cause of exacerbated inflammation associated with hypertension development.

In humanized mice (i.e., the murine immune system is replaced by the human immune system), angiotensin II-induced hypertension potentiated increased memory T lymphocytes accumulating in the aorta and lymph nodes leading to vascular dysfunction, while pharmacological prevention with hydralazine and hydrochlorothiazide significantly attenuated memory T lymphocyte pooling in these tissues (60). Accumulation of memory T lymphocyte populations was also observed in the kidneys of L-NAME + high-salt diet-induced hypertensive mice; a site of increased sodium and water retention during times of inflammation. Mice deficient in CD70, a costimulatory molecule present on APC and required for memory T lymphocyte formation, were protected from hypertension development and renal damage (61). While it appears that memory T lymphocytes have a causal relationship with hypertension, the underlying source of these cells remains unknown.

One potential cause of increased memory T lymphocytes may be an altered proinflammatory cytokine milieu. Changes in circulating proinflammatory cytokines are a hallmark of cardiovascular diseases such as hypertension, and cytokines are shown to impact the mitochondrial metabolic processes to regulate T lymphocyte differentiation. IL-15 is elevated in the serum of individuals with hypertension, and is a critical cytokine involved in cytotoxic memory T lymphocyte formation (86). Exogenous IL-15 has been shown to regulate T lymphocyte mitochondrial metabolism through increased biogenesis, SRC, and expression of carnitine palmitoyl transferase 1a (CPT1a; the rate-limiting step to FAO), which potentiates memory T lymphocyte formation (174). In addition to cytokine changes, it is also possible that memory T lymphocytes are more susceptible to redox-metabolic changes given their long life span and exposure to an array of insults over time. These insults may inadvertently activate these proinflammatory cells, which could potentiate negative hypertensive sequelae. While healthy individuals likely possess fewer redox-metabolic fluxes over their lifetime, patients with metabolic disturbances observed in diseases such as metabolic syndrome (MetS) would be significantly predisposed to these challenges. This may be the reason MetS is often associated with and a precursor to hypertension.

MetS is known to promote a proinflammatory T lymphocyte response, which may further exacerbate the development of hypertension. MetS is defined by the presence of multiple metabolically associated cardiovascular disease risk factors such as obesity, glucose intolerance, diminished high-density lipoprotein (HDL) cholesterol, and hypertriglyceridemia (163). These risk factors are also confounded by vasculature-, tissue-, and cell-specific ROS imbalances. Inflammatory pathways are triggered by chronic hyperglycemic and hypertriglyceridemic conditions directly, as well as indirectly through oxidative stress (49). Excess free FAs and glucose have been shown to promote increased ROS production in various cell types, including smooth muscle, endothelial, hepatic, and myoblasts (58, 187). Obesity also can cause eNOS uncoupling leading to increased O₂^•−, decreased NO, and vascular dysfunction (42, 181). These alterations in the redox and metabolic homeostasis have been shown to increase production of proinflammatory cytokines such as IL-6 and TNFα (100). While these changes may independently increase the likelihood of hypertension development, they also enhance the risk through alterations in T lymphocyte activation, differentiation, and memory formation.

MetS is a complex and heterogeneous disease with several root metabolic dysfunctions, yet, we are unaware of any study that has directly examined the redox-metabolic couple in T lymphocytes in this disease. Given this, we would postulate that with elevated blood glucose levels often observed in this disease, the hyperglycemic state would promote proinflammatory T lymphocyte activation while inhibiting T_REG cell activation. T_REG cells are also likely suppressed by the marked decrease in HDL cholesterol in MetS, as T_REG cells rely on HDL cholesterol for cellular survival (137, 141). However, the numerous metabolic dysfunctions associated with MetS prove difficult to determine exactly how all T lymphocyte populations will be affected. As mentioned earlier, T lymphocyte reprogramming to inflammatory memory cells relies more heavily on FAO, and thus, we would hypothesize that the abundance of FA in MetS would promote a promemory inflammatory T lymphocyte environment.

While minimal work has been performed directly examining the redox-metabolic couple of T lymphocytes in MetS, the role of the immune system in this disease is well defined and these cells often serve as biological markers for the disease. Through correlation-based network analysis, patients diagnosed with obesity and MetS were found to have high associations with various populations of T lymphocytes (18). While these populations alone may not be specific indicators of MetS, they are highly suggestive of a definitive role of T lymphocytes in the disease. Indeed, in human immunodeficiency virus-infected patients, helper T lymphocyte populations become depleted, while cytotoxic T lymphocyte populations activate and expand, and the CD4⁺/CD8⁺ ratio is a significant biological marker for predicting the occurrence of these patients developing MetS (48). Furthermore, memory T lymphocyte populations positively correlated in patients with comorbid diagnosis for MetS and systemic lupus erythematosus (164).

Thus, given the aforementioned evidence, we put forth the hypothesis that metabolism, ROS, and proinflammatory cytokines are compounding factors, not individual entities, that drive an altered T lymphocyte phenotype in hypertension development. While metabolic dysfunction and elevated ROS may in fact alter T lymphocyte activation and differentiation independently, we postulate these two processes are likely tightly linked and function codependently to affect T lymphocyte (as well as other cell types) actions. Given the small effective radius of ROS, we further speculate that the widespread and lasting actions of redox are perpetuated through modulation of cellular metabolite pools, which inevitably affect cellular processes such as epigenetic gene regulation. Many of these processes have been described in hypertension and MetS to date, thus suggesting a feed-forward loop of cross-talking systems to further exacerbate pathological outcomes of these diseases.

Conclusions

Both redox control and metabolic control are undoubtedly necessary for T lymphocyte activation, differentiation, and function. The pathways underlying redox and metabolic systems incredibly overlap, and are likely tightly coordinated within these cells. However, the examination of these processes, separately or together, is still at the beginning stages of investigation. To date, minimal immunological redox and metabolic research has primarily focused on major subclasses of T lymphocytes, whereas newly discovered variants (e.g., T_H9 and T_FH) are yet to be examined for their redox or metabolic properties. The exposition of how these processes harmonize within T lymphocytes, among the various T lymphocyte subtypes, as well as with other cells of the immune system will be critical in the elucidation of novel approaches for targeting the redox and metabolic environments in inflammatory diseases such as hypertension.

Abbreviations Used

AA

amino acid

ACC1

acetyl-CoA carboxylase 1

AKI

acute kidney injury

AMPK

AMP-activated protein kinase

AP-1

activator protein 1

APC

antigen-presenting cells

ATP

adenosine triphosphate

CD4⁺

cluster of differentiation 4

c-Myc

master regulator of cell cycle entry and proliferative metabolism

CPT1a

carnitine palmitoyl transferase 1a

DC

dendritic cell

DMNQ

2,3-dimethoxy-1,4-naphthoquinone

DUOX1/DUOX2

dual oxidase 1/dual oxidase 2

EAE

experimental autoimmune encephalomyelitis

eNOS

nitric oxide synthase 3

FA

fatty acid

FAO

fatty acid oxidation

Foxp3

forkhead box P3

GCL

glutamate cysteine ligase

Glut1

glucose transporter 1

Glut3

glucose transporter 3

Glut6

glucose transporter 6

GPX1

glutathione peroxidase 1

GSH

glutathione

GSSG

glutathione disulfide

H₂O₂

hydrogen peroxide

HDL

high-density lipoprotein

HIF

hypoxia inducible factor

IFNγ

interferon gamma

IL-1β

interleukin 1β

IL-2

interleukin 2

IL-4

interleukin 4

IL-5

interleukin 5

IL-6

interleukin 6

IL-7

interleukin 7

IL-9

interleukin 9

IL-10

interleukin 10

IL-12

interleukin 12

IL-13

interleukin 13

IL-15

interleukin 15

IL-17a

interleukin 17a

IL-21

interleukin 21

IL-23

interleukin 23

IL-35

interleukin 35

iNOS

nitric oxide synthase 2

JmjC

Jumonji C

KD

knockdown

Keap1

kelch-like ECH-associated protein 1

KO

knockout

MetS

metabolic syndrome

MnSOD

manganese superoxide dismutase

mTOR

mammalian target of rapamycin

mTORC1

mTOR complex 1

mTORC2

mTOR complex 2

NE

norepinephrine

NFAT

nuclear factor of activated T lymphocytes

NFκB

nuclear factor kappa-light-chain-enhancer of activated B

NO

nitric oxide

NOS

nitric oxide synthase

NOX

NADPH oxidase

NRF2

nuclear factor erythroid 2-related factor 2

O₂^•−

superoxide anion

OE

overexpression

OXPHOS

oxidative phosphorylation

PDH

pyruvate dehydrogenase

PDHK1

PDH kinase 1

PHD

prolyl-hydroxylase

PrdxI

peroxiredoxin-I

PrdxII

peroxiredoxin-II

RNS

reactive nitrogen species

RORγt

RAR-related orphan receptor gamma T

ROS

reactive oxygen species

SAM

S-adenosyl-methionine

SGK1

serum and glucocorticoid-regulated kinase 1

SIRT

sirtuins

SRC

spare respiratory capacity

STAT3

signal transducer and activator of transcription 3

T_CM

central memory T lymphocytes

TCR

T cell receptor

T_EFF

effector T lymphocytes

T_EM

effector memory T lymphocytes

TET

ten-eleven translocation

T_FH

T follicular helper cells

TGFβ

transforming growth factor-beta

T_M

memory T lymphocytes

T_N

naive T lymphocytes

TNFa

tumor necrosis factor-alpha

T_REG

T regulatory cells

TRX

thioredoxin

Authors' Contributions

C.M.M. and A.J.C. contributed equally to the writing and preparation of the article.

Funding Information

This work was supported by National Institutes of Health R00HL123471 to A.J.C.