Regulatory T cells (Tregs) participate in the maintenance of tolerance to self-antigens and suppressive control of excessive immune responses to exogenous antigens. A lack or dysfunction of these cells is responsible for the pathogenesis and development of many autoimmune diseases. It is well known that CD4 Tregs play a major role in controlling immune responses and can be classified into two main populations: thymus-derived naturally occurring Tregs (nTregs) and induced Tregs (iTregs) generated from CD4+CD25− precursors in the peripheral lymphoid organs. The most extensively studied Tregs are the nTregs, which express the interleukin 2 (IL-2) receptor CD25 and the transcription factor Forkhead box P3 (Foxp3). On the other hand, iTregs contain multiple heterogeneous subsets, including interleukin (IL)-10-producing CD4 type I Tregs (Tr1 cells) and transforming growth factor -β-producing Th3 cells, and so on. However, the extent of the contribution of iTregs to immunoregulation in normal animals has been difficult to evaluate because of the lack of suitable cell surface markers. It has been found recently that IL-10-secreting iTregs can be delineated as CD4+CD25−Foxp3− T cells that characteristically express both the lymphocyte activation gene-3 (LAG3) and the early growth response gene-2 (EGR2). In this review, opinions about the roles of LAG3 and EGR2 in Tregs are presented.
Statistics from Altmetric.com
Regulatory T cells (Tregs) play crucial roles in preventing autoimmune diseases and maintaining immune homoeostasis. These Tregs predominantly comprise two groups: naturally occurring CD4+CD25- Tregs (nTregs) and induced Tregs (iTregs), generated in the thymus and periphery, respectively. The nTregs, which characteristically express the transcription factor Forkhead box P3 (Foxp3),1 have been intensively studied because their deficiency abrogates self-tolerance and causes autoimmune disease.2 Scurfy mice, which have a frame shift mutation in the Foxp3 gene, display extensive lymphoproliferation and severe inflammatory infiltration in some organs such as the lung, skin and liver.3 The autoimmune regulator (Aire) gene, which affects the central induction of tolerance by regulating the clonal deletion of self-reactive thymocytes, is responsible for autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED).4 Aire regulates the ectopic expression of a battery of peripheral-tissue antigens—for example, insulin, fat acid-binding protein and salivary protein-1.5 An additional defect in central tolerance induction in scurfy mice, which are generated by crossing mice that carry a null mutation in the Aire gene, did not noticeably extend the range of the affected sites, and many organs remained unaffected.6 This result suggests that additional important mechanisms other than central tolerance and the Foxp3 system are required to enforce immunological self-tolerance in the periphery.
Although Foxp3-independent iTregs have been a focus of active investigation, it is difficult to assess the in vivo physiological function of iTregs because of the lack of specific markers that can reliably differentiate them from the other T cell subsets. Recently, we reported the interleukin 10 (IL-10) secreting CD4+CD25−Foxp3− iTregs that express both the lymphocyte activation gene-3 (LAG3) and early growth response gene-2 (EGR2), and ectopic expression of Egr2 conferred suppressive function on naïve CD4 T cells.7 This review provides an overview of our knowledge about the molecular features of LAG3 and EGR2 in Tregs.
Molecular properties of LAG3
LAG3 (CD223), a type I membrane glycoprotein of the immunogloblin (Ig) superfamily, was first reported as a cell surface protein that is expressed on activated human natural killer cells and T cells.8 LAG3 expression has also been detected in many different cell types, such as plasmacytoid dendritic cells (DCs),9 B cells,10 natural killer T cells,11 γδ T cells,12 tumour-infiltrating lymphocytes,13 14 exhausted CD8+ T cells15 and Tregs.7 16
The gene coding for LAG3 protein lies adjacent to the gene coding for CD4 on human chromosome 12p13 and shares approximately 20% homology with the CD4 gene. These two molecules have four extracellular Ig-like domains with conserved structural motifs throughout D1 to D4 domains.8 Both human and mouse LAG3 bind to major histocompatibility complex (MHC) class II molecules with higher affinity than CD4.17 18 Although LAG3 is closely related to CD4, CD4 molecules are mainly expressed at the cell surface. In contrast, almost half of LAG3 molecules are retained intracellularly.19 LAG3 co-localises with Rab11b, which is a marker of the endosomal recycling compartment. This observation suggests that LAG3 is continuously recycled and/or rapidly translocated to the plasma membrane in response to antigenic stimulation.
In CD4 T cells, LAG3 dimerisation on the cell surface is required for the formation of stable MHC binding sites.20 After in vivo T cell activation in mice, significant amounts of soluble LAG3 (sLAG3) accumulated in their serum (∼200 ng/ml). Li et al21 reported that LAG3 was cleaved within the short C peptide located between the membrane proximal D4 domain and the transmembrane domain, resulting in the release of sLAG3. The cell surface expression of LAG3 is regulated by cleavage of the extracellular domains by two transmembrane metalloproteases, ADAM10 and ADAM17.22 ADAM10 is responsible for constitutive and activation-induced cleavage, while ADAM17 mediates protein kinase C-Θ-dependent cleavage. Additionally, ADAM10 siRNA suppressed T cell proliferation in an LAG3-dependent manner. Thus, the cell surface expression of LAG3 is strictly regulated by several mechanisms.
Role of LAG3 in CD4 T cells
LAG3 is well known as an activation-induced cell surface molecule. In naïve T cells, LAG3 expression is upregulated after antigen-specific activation.12 Although an initial analysis of LAG3 knockout mice did not show any defects in T cell function,23 a subsequent analysis demonstrated that LAG3 regulates both expansion of activated primary T cells and development of the memory T cell pool.24 The cross-linking of the T cell receptor (TCR) and LAG3 on activated CD4 T cells induced less calcium release than TCR stimulation alone.25 Consistent with these observations, it has been demonstrated that non-cleavable LAG3 mutant vector transduced-CD4 T cells exhibited a more potent inhibitory effect on their activation than wild-type LAG3 vector.22 Other reports have also suggested that mouse LAG3 negatively regulates CD4 T cell activation26,–,28 and that signalling is mediated by the KIEELE motif, which is conserved between humans and mice, in the cytoplasmic domain of LAG3.29 It was also proposed that the interaction of MHC class-II-bearing antigen presenting cells and CD4+LAG3+ T cells induces a T cell-intrinsic inhibitory signalling pathway.30 These observations clearly indicate that LAG3 is a ‘cell-intrinsic’ inhibitory molecule.
Expression of LAG3 identifies interleukin-10 (IL-10)-producing CD4+CD25−Foxp3− Tregs
LAG3 has recently been shown to be a new extrinsic and intrinsic inhibitory molecule that is required for the maximal regulatory function of CD4+CD25+Foxp3+ Tregs16 and controls effector T cell expansion and homoeostasis.28 Ectopic LAG3 expression also confers regulatory activity on naïve T cells.16 The LAG3 on CD4+CD25+ Tregs interacts with MHC class II molecules on dendritic cells (DCs), and the binding of LAG3 to the MHC class II molecules expressed by immature DCs induces ITAM-mediated inhibitory signalling, which suppresses DC maturation and immunostimulatory capacity.31 These observations indicate that LAG3 possesses ‘cell-intrinsic’ and also ‘cell-extrinsic’ regulatory activity. However, in naïve mice, LAG3 protein is hardly detected on the cell surface of CD4+CD25+Foxp3+ Tregs16 and is mainly restricted to a population of CD4+CD25−CD45RBlow memory T cells,7 which are assumed to include IL-10-secreting CD4+Foxp3− T cells, resembling type 1 Tregs (Tr1 cells).32 Although it is known that LAG3 is expressed in CD4+CD25− T cells, its role in CD4+CD25− T cells has remained elusive. In our analysis, approximately 2% of the CD4+CD25− T cell population in the spleen consisted of CD4+CD25−LAG3+ T cells.7 These CD4+CD25−LAG3+ T cells express higher levels of LAG3, IL-10 and Blimp-1 mRNA than other CD4 T cell subsets. The transcription factor Blimp-1 is required for IL-10 production by CD4 T cells33 and is also indispensable for the formation of IL-10-producing effector Tregs.34 Indeed, CD4+CD25−LAG3+ T cells secreted high amounts of IL-10 upon in vitro antigenic stimulation.7 In addition, CD4+CD25−LAG3+ T cells were hypoproliferative in response to anti-CD3 and anti-CD28 monoclonal antibody stimulation, and suppressed the in vivo development of colitis induced in RAG-1−/− recipients by the transfer of naïve T cells in an IL-10-dependent manner. These CD4+CD25−LAG3+ Tregs did not express Foxp3 protein, and scurfy mice still had functional CD4+CD25−LAG3+ Tregs, which expressed IL-10 mRNA and retained regulatory activity in vitro. Unlike CD4+CD25+Foxp3+ nTregs, high-affinity interactions with selecting peptide/MHC ligands expressed in the thymus did not induce the development of CD4+CD25−LAG3+ Tregs. These observations indicate that LAG3 is a phenotypic marker of IL-10-producing Foxp3-independent CD4 iTregs that plays a role in their regulatory activity in the normal immune system.
Anergy-associated EGR2 as a regulatory gene of CD4+CD25−LAG3+ Tregs
In order to address the molecular mechanisms responsible for the functions of CD4+CD25−LAG3+ Tregs, we compared the differential gene expression profiles of CD4+CD25−LAG3+ Tregs, CD4+CD25+ Tregs, CD4+CD25−LAG3− T cells and naïve CD4+CD25−CD45RBhigh T cells by gene array analysis. Interestingly, the transcription factor Egr2 genewas preferentially expressed in CD4+CD25−LAG3+ Tregs.7 Intracellular staining of Egr2 revealed a strong correlation between Egr2 and LAG3 expression in CD4 T cells (unpublished data).
Egr2 is a member of a family of Cys2His2-tpye zinc finger transcription factors, which consists of four members, Egr-1, -2, -3 and -4. Egr2 plays an essential role in hind-brain development and myelination of the peripheral nervous system, and the null mutation of Egr2 resulted in perinatal or neonatal death due to respiratory and/or feeding deficits.35 In CD4 T cells, it is reported that Egr2 is closely associated with anergy induction.36 37 Egr2 inhibits interferon -γ (IFN-γ) and IL-2 secretion by T cells and enhances the expression of the E3 ligase Cbl-b, which is critical for the regulation of T cell tolerance and anergy.37 Egr2 binds directly to the promoter of the cell cycle inhibitor p21cip1 in T cells.38 Consistent with these observations, Egr2-associated CD4+CD25−LAG3+ Tregs were fully anergic upon in vitro TCR stimulation, similarly to CD4+CD25+ Tregs. Intriguingly, in Egr2 conditional knockout mice, in which the Egr2 gene was deleted in CD2 T cells, the T cells did not show altered primary activation but became hyperproliferative in response to prolonged stimulation, leading to the development of a lupus-like syndrome.38 Although these observations indicate that Egr2 controls the self-tolerance of T cells, the extrinsic regulatory function of Egr2 has not been investigated. We therefore employed a retroviral gene transfer system. The forced expression of Egr2 converted naïve CD4 T cells into IL-10-secreting and LAG3-expressing Tregs.7 Moreover, these Egr2-transduced CD4 T cells exhibited antigen-specific immunosuppressive effects on the delayed-type hypersensitivity response. Thus, Egr2 confers suppressive activity on naïve CD4 T cells in vivo. These results suggest that Egr2, a ‘cell-extrinsic’ regulatory molecule, might be a critical regulator of IL-10-producing CD4+CD25−LAG3+ Tregs. Further analysis of Egr2 regulation and elucidation of the precise mechanisms by which Egr2 exerts its regulatory activity will provide important insights into autoimmunity and might result in new therapeutic modalities.
CD4+CD25−LAG3+ Tregs and other IL-10-producing CD4+FOXP3− Tregs
IL-10, an anti-inflammatory cytokine, can both directly and indirectly inhibit effector T cell responses during infection, autoimmunity and cancer.39,–,41 Published data have suggested major roles for IL-10-producing CD4+CD25−Foxp3– Tregs that are distinct from those of nTregs in the maintenance of the immune regulatory system,42 and Tr1 cells have also been a focus of active investigation. Tr1 cells are generated by repeated antigen stimulation in the presence of either excess IL-10 in vitro43 or by culture in the presence of vitamin D3 and dexamethasone.44 Interestingly, Tr1 cells obtained in vitro in the presence of vitamin D3 and dexamethasone also co-expressed high levels of LAG-3 and IL-10 (unpublished data). We have also confirmed that Egr2-transduced CD4 T cells expressed Egr2, Blimp-1 and IL-10 mRNA. These findings are in keeping with that of previous reports, which found that the phenotype of Blimp-1-expressing Tregs resembled that of IL-10-producing Tregs in humans and mice.34
However, it remains unclear whether Tr1 cells are naturally present in the normal immune system because Tr1 cells are mainly characterised by their unique pattern of cytokine production after in vitro stimulation. Tr1 cells produce high levels of IL-10, with or without transforming growth factor-β (TGF-β and IL-5; low levels of IFN-γ; and little or no IL-2 or IL-4.43 Their immunological properties, including their strong IL-10 production, peripheral development, anergic phenotype and lack of Foxp3 expression, suggest that CD4+CD25−LAG3+ Tregs are analogous to Tr1 cells. However, CD4+CD25−LAG3+ Tregs produce IL-10 and also IFN-γ but do not produce IL-5 or TGF-β.7 Although CD4+CD25−LAG3+ Tregs partially fulfil the criteria for Tr1 cells,32 43 45 further studies will be required to elucidate the inconsistency between these Tregs in more detail. Recently, some candidate cell surface markers for IL-10-producing CD4 Tregs have been reported. Ochi and colleagues reported that latency-associated peptide (LAP)-expressing CD4+CD25− Tregs produce both IL-10 and TGF-β.46 The CD4+CD25−LAG3+ Tregs hardly expressed LAP protein on their cell surface, indicating that they differ from CD4+CD25−LAP+ Tregs.7 Although it was also reported that a subpopulation of IL-10-producing CD4 Tregs express CD103 (integrin αEβ7),47 the CD4+CD25−LAG3+ Tregs did not express CD103 on their cell surface.7
Several IL-10-secreting T cell subpopulations with regulatory activities, such as Th1, Th2 or Th17 cells, have been identified.48,–,50 It is now evident that Tregs are composed of several phenotypically distinct subsets, and they appear to have plasticity.51 The Egr2-associated CD4+CD25−LAG3+ Tregs, which produce a large amount of IL-10, might be useful for investigating the induction mechanisms of IL-10-producing CD4 Tregs.
EGR2 and human diseases
Egr2 plays a critical role in the nervous system. Mutations in the Egr2 gene prevent Schwann cell development and peripheral nerve myelination in mice and lead to the development of demyelinating neuropathy.35 In humans, EGR2 mutations are associated with Charcot–Marie–Tooth disease type 1, Dejerine–Sottas syndrome and congenital hypomyelination neuropathy.52 Interestingly, several recent genome-wide association studies (GWAS) have found new genetic links between EGR2 and human autoimmune diseases. Two independent GWAS identified strong association signals which were located on chromosomes10q21 in Crohn's disease, the most common form of chronic inflammatory bowel disease.53 54 The associated intergenic region was flanked by EGR2, suggesting that this genetic variation regulates EGR2 expression. In line with these observations, Egr2-associated CD4+CD25−LAG3+ Tregs effectively prevented intestinal inflammation in a murine T cell transfer model of colitis.7
Furthermore, a candidate gene analysis indicated that polymorphisms in EGR2 influence systemic lupus erythematosus (SLE) susceptibility in humans.55 SLE is a systemic autoimmune disease characterised by autoantibody production and is associated with a wide range of clinical manifestations. Although the cause of SLE is unknown, overwhelming evidence points to a combination of underlying genetic susceptibility and environmental factors.56 Interestingly, T cell-specific Egr2 conditional knockout mice develop progressive lupus-like autoimmunity.38 We have found a human counterpart of murine CD4+CD25−LAG3+ Tregs in the tonsils (manuscript under preparation). The gene expression profile of human CD4+CD25−LAG3+ T cells showed a marked similarity to that of murine CD4+CD25−LAG3+ Tregs. Thus, it is conceivable that, like mouse Egr2, human EGR2 may be a key molecule for nerve systems and also for immune regulatory systems controlling both T cell and B cell responses.
Future prospects and conclusions
As mentioned above, CD4 Treg subsets can be classified into two main populations: thymus-derived CD4+CD25+Foxp3+ nTregs and iTregs, such as CD4+CD25−LAG3+ Tregs, which are generated in the periphery. However, the relative contribution of each Treg subset to the regulation of immune responses is poorly understood. A recent study by Huber and colleagues showed that pathogenic Th17 cells were controlled by both Foxp3+ and LAG3+Foxp3– CD4 Tregs in an IL-10-dependent manner.57 Accumulating evidence suggests that Th17 cells may play a significant role in the pathogenesis of multiple inflammatory and autoimmune disorders including SLE.58 Given that CD4+CD25−LAG3+ Tregs express high levels of Egr2, Blimp-1 and IL-10, CD4+CD25−LAG3+ Tregs might play a distinct role in regulating Th17 cells.
IL-10 has an antiapoptotic effect on B cells. IL-10 is also involved in B cell isotype switching, and plays a role in autoimmune diseases associated with B cell dysregulation such as SLE.59 60 However, nasal anti-CD3 administration induced IL-10-producing CD4+CD25−LAP+ Tregs which suppress lupus pathogenesis through the downregulation of IL-17+CD4+ICOS+CXCR5+ follicular helper T cells in an IL-10-dependent manner.61 These findings indicate that, in some settings, IL-10 can suppress the pathogenesis of lupus. The MRL/lpr mouse, possessing a single gene mutation in the Fas gene, is a prototypical model of human SLE.62 Deficits in the apoptosis-promoting Fas/FasL receptor–ligand pair of cell surface molecules in MRL/lpr mice are associated with hypergammaglobulinaemia, autoantibody production and a spectrum of autoimmune manifestations that resemble lupus. Interestingly, the adoptive transfer of CD4+CD25−LAG3+ Tregs from control MRL/+ mice suppressed the progression of nephritis and autoantibody production in MRL/lpr lupus-prone mice (manuscript under preparation). Inconsistent with the previous report, CD4+CD25+ Tregs from MRL/+ mice exhibited no significant therapeutic effect after being transferred to MRL/lpr mice.63 Thus, CD4+CD25−LAG3+ Tregs might be useful for treating autoantibody-mediated autoimmune diseases, including SLE.
Immunodysregulation, polyendocrinopathy, enteropathy, X linked (IPEX) syndrome is a rare disease caused by a mutation in the FOXP3 gene that results in the absence or a severe deficiency of CD4+CD25+ Tregs.64 Patients with the condition are characterised by autoimmune enteritis, type 1 diabetes mellitus, eczema, hypothyroidism, autoimmune haemolytic anaemia, membranous nephropathy and recurrent infections. Enteropathy, endocrinopathy and dermatitis, which are not common symptom in patients with SLE, are seen in most IPEX patients.65 Although a variety of autoantibodies to multiple organs were observed in IPEX patients, antinuclear antibodies are either absent or present at low titres, and anti-dsDNA antibodies are rarely detected. Glomerulonephritis, which is the most common form in patients with SLE, is rare in patients with IPEX syndrome.66 On the other hand, SLE is characterised by high serum titres of antinuclear antibodies and dsDNA antibodies.56 Thus, diagnosis of SLE has not been established in patients with IPEX syndrome.67 These facts suggest that Foxp3 might not have direct association with the physiopathology of lupus in humans. Indeed, analyses of the function and phenotypic properties of CD4+CD25+Foxp3+ Tregs in patients with SLE have led to conflicting results.68 The data for the role of CD4+CD25+Foxp3+ Tregs in the pathogeneses of lupus mouse models are also a matter of debate.63 69,–,71 On the other hand, mice in which the Egr2 gene was conditionally knocked out in T cells developed progressive lupus-like autoimmunity with antibodies to histone and dsDNA and were characterised by the accumulation of IL-17-producing CD4 T cells.38 These observations suggest that Egr2-associated CD4+CD25−LAG3+ Tregs are the major subset of circulating IL-10-producing Tregs and that their functions are complementary to those of CD4+CD25+Foxp3+ nTregs. Thus, fine tuning the balance of CD4+CD25+Foxp3+ nTregs and CD4+CD25−LAG3+ Tregs might provide new therapeutic methods for autoimmune diseases. Although additional studies are required to elucidate the relationship between CD4+CD25+Foxp3+ nTregs and CD4+CD25−LAG3+ iTregs, distinct markers of IL-10-producing CD4 Tregs, such as LAG3 and Egr2, will substantially contribute to resolving this issue.
Funding This work was supported by grants from the Japan Society for the Promotion of Science, Ministry of Health, Labor and Welfare, and the Ministry of Education, Culture, Sports, Science and Technology (MEXT) (in part by Global COE Program Chemical Biology of the Diseases, by the MEXT), Japan.
Competing interests None.
Provenance and peer review Commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.