Function and Therapeutic Intervention of Regulatory T Cells in Immune Regulation

Abstract

Although the detailed mechanisms of regulatory T cells (Tregs) in regulating immune responses have not been completely clarified yet, Tregs therapy on autoimmune diseases and organ transplantation is making robust progress, along with the gradually enhancing knowledge of the Tregs function. In this chapter, on the basis of summarizing the immunomodulatory functions of Tregs, we reviewed the latest scientific progress and status of our understanding, as well as the prospect of stimulation and expansion of Tregs in vivo and in vitro followed by adoptive transfer or autologous cell therapy in animal models and clinical trials, respectively. Moreover, we also assessed the current technological limitation and potential side effects of polyclonal and antigen-specific Tregs-based approaches and techniques, to promote the development of rescue, revive, or rejuvenate Tregs in the therapeutic intervention to treat autoimmune diseases and transplantation.

1. Introduction

Regulatory T cells (Tregs), as a subgroup of T cells with immunosuppressive function, were first reported in 1970s by Gershon and Kondo [1]. According to the developmental origin, Tregs can be broadly classified into two groups. Tregs that grow in the thymus are called natural (nTregs) or thymic (tTregs) Tregs, and that develop at the periphery by specific stimuli of conventional CD4⁺ T cells are termed peripheral Treg (pTregs). When Tregs are induced by specific factors, such as interleukin-2 (IL-2) and transforming growth factor (TGF)-β, in vitroare called induced Tregs (iTregs) [2]. At present, Tregs have emerged as a vital part in understanding the immune response to pathogens, controlling the development of allergies, transplantation, and autoimmune diseases, as well as in the application of treating tumors, since their “re-discovery” more than 20 years before [3]. However, the detailed mechanisms of Tregs in regulating both innate and adaptive immune responses are still not completely understood. In this chapter, we will review the latest scientific progress and status on our understanding and prospect of stimulation and expansion of Tregs in vivoand in vitrofollowed by adoptive transfer or autologous cell therapy in animal models and clinical trials. We will also assess the current technological limitation and potential side effects of polyclonal and antigen-specific Tregs-cell-based approaches and techniques.

2. Tregs function

2.1 Loss of Tregs and development of autoimmune diseases

Every manifestation stemming from Tregs paucity highlights a vital function of Tregs in preventing fatal autoimmune inflammation. The immunosuppressive function of Tregs is mainly dependent on continuous expression of the transcription factor forkhead box protein 3 (Foxp3), which is a critical regulator of CD4⁺CD25⁺ Tregs development and function. Loss function of Foxp3 results in a fatal autoimmune disease featuring all known types of inflammatory responses. Studies have demonstrated that the typical or fatal autoimmune responses that occurred in the Foxp3-mutant scurfy mice or Foxp3-null mice are related to the deficiency of CD4⁺CD25⁺ Tregs, but not to the cell-intrinsic dysfunction of CD4⁺CD25⁻ T cells. When being transferred into the neonatal Foxp3-deficient mice, Tregs can preferentially expand and control the development of autoimmune disease. Furthermore, ectopic expression of Foxp3 can confer suppressor function on peripheral CD4⁺CD25⁻ T cells [4]. Even in severely diseased mice, by reinstating Foxp3 protein expression and suppressor function in cells expressing a reversible Foxp3 null allele, the rescued Tregs normalized immune activation, quelled severe tissue inflammation, reversed fatal autoimmune disease, and provided long-term protection against them. It is indicated that Tregs are capable of resetting the immune homeostasis in broad-spectrum systemic inflammation and autoimmune diseases [5].

X-linked autoimmunity-allergic dysregulation syndrome (XLAAD), which has been renamed as Immunodysregulation, polyendocrinopathy, enteropathy, and X-linked syndrome (IPEX), was a rare inborn error of immune regulation and autoimmune lymphoproliferative illness in humans [6]. As one of the most well-known Mendelian disorders, IPEX is characterized by a loss of immunological tolerance caused by a lack of functioning Tregs and was discovered to be associated with the mutations in Foxp3 [7, 8]. In the absence of Tregs, activated CD4⁺ T cells instigate multi-organ damage resulting in type 1 diabetes (T1D), enteropathy, eczema, hypothyroidism, and other autoimmune disorders.

Moreover, studies on Tregs depletion by cytotoxic T-lymphocyte antigen (CTLA)-4 Ab (e.g., ipilimumab) in tumor patients have shown a strong correlation between the induction of tumor regression and autoimmunity [9]. Ipilimumab acts not only on effector T cells (Teffs) but also on Tregs because the latter ones in both mouse systems and humans can be directly targeted by ipilimumab due to the constitutive expression of CTLA-4 on their cell surface [10, 11]. Except for a decreased frequency of circulating CD25⁺CD4⁺ Tregs can be observed upon ipilimumab, CTLA-4 blockade renders Teffs resistant to the inhibitory activity of Tregs, rather than modulating the immunosuppressive effects of Tregs on T cells and NK cells [12]. It is indicated that loss of Tregs has a close relationship with the development of autoimmune diseases from another perspective.

2.2 Immune tolerance and prevention of autoimmunity

Our body’s immune system has evolved to perform self-tolerance to resist the autoimmune reactions directed against our own cells via sophisticated mechanisms. On the T cell level, self-tolerance is executed by deletion of T cells with self-reactive T cell receptor (TCR) in the thymus (central tolerance) or maintained by specialized cells, including Tregs, outside of the thymus (peripheral tolerance). The importance of Tregs for the maintenance of immune tolerance has also been illustrated to have a close relationship with the expression of the Foxp3 gene, both in humans and mice [13, 14]. Foxp3, together with other transcription factors and coactivators/corepressors, represses the transcription of IL-2 in Tregs, rendering them highly dependent on exogenous IL-2 (mainly produced by activated non-Tregs) for their maintenance and function. Tregs constitutively express the high-affinity IL-2 receptor (α chain), which serves as a sink for IL-2 that controls the expansion of Teffs. The development of autoimmune/inflammatory disease can be promoted if disrupting this IL-2-mediated feedback loop at any step. Further, manipulation of this feedback loop is instrumental in tuning the intensity of Tregs-mediated suppression, hence the strength of a variety of immune responses [15]. Foxp3 also activates the genes encoding Tregs-associated molecules, including CD25, CTLA-4, and Glucocorticoid induces tumor necrosis factor receptor (GITR) and confers suppressive activity to Tregs, which directly suppress non-Tregs or modulate the function of antigen-presenting cells (APCs) to activate non-Tregs [16].

2.3 Mechanisms of immune suppression

The Tregs-mediated immune suppression may be related to three mechanisms, including secretion of immunosuppressive cytokines [17], cell-contact-dependent suppression [18], and functional modification or killing of APCs [19]. More than one mechanism may operate for controlling the particular immune response in a synergistic and sequential manner.

IL-10 and TGF-β may act as the main immunosuppressive cytokines contributing to control the autoimmune disorders or inflammatory diseases secreted by Tregs [17]. IL-10 indirectly prevents antigen-specific T cell activation, which is associated with downregulation of the antigen presentation and accessory cell functions of monocytes, macrophages, and dendritic cells (DCs), as well as inhibits T-cell expansion by directly inhibiting IL-2 production by these cells. The pivotal function of TGF-β is to maintain tolerance via the regulation of lymphocyte proliferation, differentiation, and survival. TGF-β can block the proliferation of T lymphocytes by suppressing the expression of IL-2 (via Smad3 signaling pathway), cyclins (including cyclin D2 and cyclin E), cyclin-dependent kinase (CDK)-4, and c-myc. TGF-β also can inhibit the differentiation of Th1 and Th2 cells by blocking the T-bet/STAT4 and GATA-3/NFAT signaling transduction pathway, and down-regulating the differentiation of cytotoxic T lymphocyte (CTL) via regulating the expression of c-myc and T-bet [20]. TGF-β can also induce the expression of Foxp3 and the generation of Tregs. In addition, nTregs can also predominantly produce IL-35, a new member of the IL-12 family, to perform the suppressive function [21]. IL-35 is a novel Epstein-Barr-virus-induced gene (Ebi) 3-IL-12α heterodimeric cytokine, and Ebi3, which encodes IL-27 β, is a downstream target of Foxp3. Ebi3^−/− and IL12α^−/− Tregs have significantly reduced regulatory activity in vitroand fail to control homeostatic proliferation and cure inflammatory bowel disease (IBD) in vivo.

Antigen-activated Tregs, which are highly mobile, are swiftly recruited to APCs (especially DCs), upon being stimulated by the specific antigen. The recruitment of Tregs to APCs is in chemokines or adhesion molecules depended on manner. Once the Tregs aggregate around the APCs, they will outcompete antigen-specific naïve T cells regarding interaction with DCs, mainly because of the high expression of adhesion molecules on Tregs, such as lymphocyte function-associated antigen (LFA)-1 [22].

Tregs can modulate the function of APCs. Activated Tregs promote the downregulation of CD80 and CD86 on APCs or stimulate DCs to form the enzyme indoleamine 2, 3-dioxygenase both by a CTLA-4-dependent mechanism [23, 24]. Indoleamine 2, 3-dioxygenase is capable of catabolizing the essential amino acid tryptophan to kynurenines, which are toxic to T cells. Alternatively, Tregs can induce the apoptosis of responder T cells or APCs by secreting granzyme/perforin or immunosuppressive cytokines (such as IL-10), or through the delivery of a negative signal (possible including intracellular cyclic AMP) to inactivate the responder T cells [19]. The upregulation of intracellular cyclic AMP will lead to the inhibition of T cell proliferation and IL-2 production, as well as the generation of pericellular adenosine catalyzed by CD39 and CD73 by Tregs.

3. Tregs and autoimmune and autoinflammatory diseases

3.1 Tregs and type 1 diabetes

Type 1 diabetes (T1D) is a typical kind of autoimmune disease affecting millions of people worldwide with a steadily rising incidence, and islet infiltrating self-reactive T cells mediated β-cell destruction is considered to be primary pathogenesis of this disease. The initiation of the autoimmune process is related to the recognition of self-antigens by the autoreactive subsets of CD4⁺ T-helper lymphocytes, which can preferentially produce the Th1 cytokine spectrum after activation. The presence of autoreactive CD8⁺ cytotoxic T lymphocytes is necessary for the further development of T1D as well. It has been demonstrated that CD4⁺CD25⁺FoxP3⁺ Tregs also play an indispensable role in the development of T1D by preventing destructive autoimmunity [25]. Although the application of immunosuppressive reagents is one of the available therapies, it can have severe side effects. Optimal immune-based therapies for T1D should restore self-tolerance without inducing chronic immunosuppression. Thus, efforts to repair or replace Tregs in T1D probably can reverse autoimmune response and protect the remaining insulin-producing β cells. There is a large body of evidence to suggest that Foxp3⁺ Tregs function is altered in patients with T1D, though the overall frequency of Foxp3⁺ Tregs may be unaltered in these individuals [26, 27]. Data from the non-obese diabetic (NOD) mouse model of autoimmune diabetes and human with T1D suggest that increasing resistance of Teffs to Tregs regulation may be the primary cause for reduced suppression, and it can be explained by the inability of Teffs to provide an environment conducive to Tregs fitness and function, including the reduced IL-2 production or downregulation of the IL-2 signaling pathway by Teffs [28, 29].

Apart from their canonical function of immune suppression, it is now well accepted that Tregs can likewise be induced in the periphery in an antigen-specific manner and take residence in tissues to play important roles in maintaining tissue homeostasis. So, antigen-specific induction of disease-relevant Tregs will offer the opportunity to treat or prevent the T1D for a long-standing goal. It has been demonstrated that in the peripheral blood of children who are at risk to develop T1D, the proportion of insulin-specific Tregs reduced during the onset of islet autoimmunity, while the higher reduction was related to a rapid progression to clinically overt T1D [30]. This finding suggested that inducing these insulin-specific Tregs may delay the progression to clinically symptomatic T1D. Nevertheless, very little is known about pancreas residing Tregs, and all studies conducted so far on these tissue-specific Tregs focused solely on NOD mice with ongoing insulitis. A recent study found that a combinatorial regimen involving the anti-CD3, cyclophosphamide (CyP), and IAC (IL-2/JES6–1) antibody complex can promote the engraftment of antigen-specific donor Tregs through ablating host conditioning and control islet autoimmunity without long-term immunosuppression [31].

3.2 Tregs and rheumatoid arthritis

Rheumatoid arthritis (RA) is one kind of common systemic inflammatory autoimmune disease, and its typical clinical symptoms are musculoskeletal pain, joint swelling, and stiffness, which can seriously damage body function and reduce the quality of life of patients. Patients with RA are more likely to develop osteoporosis, infection, cardiovascular diseases, respiratory diseases, cancer, and other diseases than the general population. Similar to other autoimmune diseases, Tregs also play a vital role in the pathogenetic process in RA. When the number and/or function of Tregs are decreased or inhibited, autoantigen or ligand death receptors (DRs) related immune cascade can be amplified, and the levels of various cytokines, such as IL-2, will be rapidly increased, leading to the activation of macrophages in the synovium of bones and joints to produce many inflammatory cytokines including IL-1, IL-6, and IL-8 [32, 33]. These inflammatory reactions destroy articular cartilage and eventually lead to joint deformities.

However, contradictory results on the number (increased [34], unchanged [35], or decreased [36]) and functional characteristics (enhancement [32] or attenuation [33]) of Tregs in the peripheral blood of patients with RA have been reported in different studies, and this discrepancy can be explained by the ongoing difficulties in the recognition of Tregs. In most studies, the high-level expression of Foxp3, CD25, and low-level expression of CD127 (the α-chain of the IL-7 receptor) are used to define Tregs, and the CD3⁺CD4⁺CD25^highCD127^low phenotype is most commonly isolated from Tregs population. However, Foxp3 requires intracellular staining and the expression levels in Tregs in the resting state and activated state are different, and conventional T cells (Tconvs) also express a low levels of Foxp3 and CD25 upon TCR stimulation and low levels of CD127 [37]. Thus, some other supplementary cell surface markers, such as CD62 ligand, integrin Ea (CD103), GITR (TNFRSF18), CTLA-4 (CD152), CD45RO, and neuropilin, have been also used to identify Tregs in clinical practice [38]. CD45RA and CD45RO can be used to distinguish immature Tregs (CD45RA⁺Foxp3^low) from activated memory Tregs (CD45RA⁻Foxp3^high) cells [39]. A more stringent method to define Tregs has revealed the number of Tregs decreased in peripheral blood and increased in synovial fluid by performing a meta-analysis [40].

However, although the Tregs isolated from RA patients can show normal inhibitory activity in vitro, they function abnormally when circulating in the synovial fluid, which is caused by the overexpression of IL-6 induced inflammatory environment [41, 42]. Teffs in this inflammatory environment are resistant to Tregs-mediated repression, and the sensitivity of APCs to the inhibition of Tregs also decreased [43, 44]. Moreover, the arthritic synovial fibroblasts can promote the transformation of CD25^lowFoxp3⁺CD4⁺ T cells into Th17 cells in the oxygen deficiency synovial microenvironment, and the latter one shows a stronger ability to induce osteoclast production [45]. The process of transformation is closely related to the activation of the hypoxia-inducible factor-1a (HIF-1a) pathway.

3.3 Tregs and autoimmune hepatitis

Autoimmune hepatitis (AIH) is a severe hepatopathy that occurs globally in all ethnicities and affects children and adults of all ages. It is with a female predominance and characterized by hypergammaglobulinemia, interface hepatitis on histology, and seropositivity for disease-defining autoantibodies. In AIH, the autoimmune reaction resulting in liver injury initiates with the presentation of liver autoantigen by APCs to an uncommitted T lymphocyte. Following antigen encounter, Th0 becomes activated and differentiates into Th1, Th2, and Th17. Th1 cells secrete interferon (IFN)-γ and IL-2, which can lead to the activation of macrophages and upregulation of major histocompatibility complex (MHC) class I and II by hepatocytes [46]. Th2 lymphocytes secrete IL-4 and IL-10, which can promote the B cell activation and maturation into plasma cells. Plasma cells then produce autoantibodies and mediate cell cytotoxicity in turn [47]. Activation of Th17 cells, which can secrete IL-17 proinflammatory cytokines, has been associated with the induction of pro-fibrotic events [48]. The autoimmune attack will continue perpetrating and favoring the progression of tissue damage if these events are not opposed by effective immunoregulation.

It has been demonstrated that the impairment of Tregs plays an important role in the initiation and progression of AIH. A numerical and functional defect in CD4⁺CD25^+/highFoxp3⁺ cells was reported in patients with AIH compared with the healthy subjects [49]. Before immunosuppressive treatment is instituted, Tregs isolated from AIH patients are also impaired in their ability to expand, and unable to regulate CD4⁺ and CD8⁺ T cells proliferation and modify CD4⁺ and CD8⁺ T cells cytokine profile as in the case of healthy controls [50, 51].

This deficiency of Tregs in AIH patients might be linked to increased expression of the cell surface marker CD127 [52] and defects in the expression of the ectonucleotidase CD39 [53]. CD127 is also known as the α-chain of the IL-7 receptor (IL-7Rα), while CD39⁺ Tregs decrease in frequency in AIH patients leading to the failure to control the production of IL-17 by Th17. So, Tregs in AIH subjects are more prone to acquire features of effectors than their counterparts when exposed to a proinflammatory challenge, which suggests the defective immunoregulation of Tregs in AIH might have some relationship with the increased conversion of Tregs into effector lymphocytes [53]. A recent study confirmed that impaired CD39 levels derive from alterations of aryl hydrocarbon receptor (AhR) signaling [54]. AhR is a mediator of toxin responses and adaptive immunity. Upon binding to endogenous or exogenous ligands, AhR undergoes activation, which will bring about the upregulation of CD39.

However, aberrantly high levels of aryl hydrocarbon receptor repressor and estrogen receptor alpha (Erα) can be detected in AIH. AhR binds Erα with higher affinity than aryl hydrocarbon receptor nuclear translocator (ARNT), the classical AhR binding partner. These non-conventional binding give rise to impaired CD39 upregulation.

Impaired Tregs function in AIH is also linked to defective levels of Galectin-9 (Gal-9). Gal-9, a member of the galectin family, is one kind of β galactoside binding protein expressed on Tregs. It can bind to the mucin domain 3 (Tim-3) on CD4⁺CD25⁻ Teffs. Upon Gal-9 binding to Tim-3, apoptosis in CD4⁺CD25⁻ Teffs will be induced. Thus, reduced expression of Gal-9 in Tregs in AIH contributes to the less suppressing ability of Tregs and rendering CD4⁺CD25⁻ Teffs less prone to the control of Tregs [55].

In addition, defective Tregs function in AIH is linked with reduced ability to produce IL-10 as well. It is resulting from poor response to IL-2 as reflected by impaired ability to upregulate the phosphor signal transducer and activator of transcription 5 (pSTAT-5) [56].

3.4 Tregs and inflammatory bowel disease

Inflammatory bowel disease (IBD) is a chronic, inflammatory, and autoimmune disorder. The types of IBD include ulcerative colitis (UC) and Crohn’s disease. The etiology of IBD is possibly linked to the dissonance of the host immune system, genetic variability as well as an environmental factor, and the pathogenesis of this disorder has not been fully elucidated [57]. In recent years, it has been found that the abnormal intestinal mucosal immune system plays a crucial role in the occurrence, development, and prognosis of IBD, involving the imbalance in Th17 and Tregs [58]. The differentiation of Th17 cells goes through three stages—initiated by IL-6 and TGF-β, expanded by IL-21, and IL-23 maintains the stable maturation of Th17 cells during the later stage of differentiation [59]. Except for protecting the intestinal mucosa via keeping the balance of the immune microenvironment, Th17 cells also can exacerbate the intestinal inflammatory response through secreting proinflammatory cytokines, such as IL-17. Compared with the healthy controls, Th17 cells infiltrate the intestinal mucosa of IBD patients and the amount of IL-17 increases [60]. Tregs and Th17 cells are related through differentiation and share a common signal pathway mediated by TGF-β. In the UC mouse model, Th17 cells in the peripheral blood of mice increased, yet Tregs decreased [61]. Therefore, Tregs deficiency may be the central link in the pathogenesis of IBD and the regulation of Th17/Tregs balance is prospective to be a new target for the treatment of IBD. The immunological factors affecting the Th17/Tregs balance in IBD consist of both TCR and costimulatory signals and cytokines. IL-2 inducible T cell kinase (ITK), a critical regulator of intracellular signaling downstream of the TCR, positively regulates the differentiation of Th17 and negatively regulates the differentiation of Tregs [62]. The T cell costimulatory molecule OX40 and its cognate ligand OX40L collectively play an essential role in keeping the growth of Th17 and Tregs, that is, activation of OX40 enhanced Th17 function while blocking OX40L decreased Tregs proliferation [63].

4. Tregs and transplantation

While organ transplantation is one of the greatest achievements in modern medicine, rejection is still the major barrier to successful transplantation. The immune response to an allograft is an ongoing dialog between the innate and adaptive immune systems. One of the reasons that transplantation induces such a dynamic immune response is the high precursor frequency of T cells capable of responding to mismatched MHC molecules. Although immunosuppression regimens are effectively able to control the acute rejection and decrease graft loss in the first year after transplantation, it is difficult to get a durable effect on long-term graft survival with these modern regimens, owing to a combination of drug toxicities, the emergence of chronic alloimmune responses and the serious complications, such as chronic infections or malignancies. Studies on experimental transplant models have suggested a role for Tregs in protecting allografts by suppressing both autoimmune and alloimmune responses [64, 65]. Further, Tregs-based therapies do not require harsh conditioning and have a risk of graft-versus-host disease.

4.1 Tregs and solid organ transplantation

The first step in the adaptive immune response to a transplant in a solid organ transplantation recipient is T-cell recognition of alloantigen or allorecognition. Graft-specific Tconvs, which are capable of direct recognition of alloantigen, are present at a very high frequency so that they can respond to the transplant without first clonally expanding in lymph nodes. When graft-specific Tconvs are recruited to the graft, they will lead to inflammation and tissue damage. Increasing graft-specific Tregs combined with the reduction of graft-specific Tconvs allow the former one to dominate in the graft and prevent recruitment and activation of the later one. Moreover, once entering the draining lymph node, inflammatory APCs can activate more graft-specific Tconvs, while tolerogenic APCs are able to expand graft-specific Tregs and prevent the expansion of graft-specific Tconvs to maintain tolerance [66].

Moreover, Tregs with direct alloantigen specificity, which are also present at high frequency, play important role in the induction of tolerance, whereas Tregs with indirect alloantigen specificity are important for the maintenance of tolerance [67]. Tregs control transplant rejection by first migrating to the organ to prevent graft damage and then retreating to draining lymph nodes to maintain tolerance [65]. During an active alloimmune response, Tregs with both direct and indirect specificities expand and infiltrate the organs, but the homeostatic function of Tregs is insufficient to prevent rejection from occurring due to the potency of alloimmune responses until the organs have suffered substantial damage [67]. That is because Teffs arrive at the graft site first and expand in number before the arrival of Tregs so that the grafts are dominated by Teffs [64], and at the peak of alloimmune responses, a high antigen load, vigorous co-stimulation, and high concentrations of cytokines, such as IL-1 and IL-6, override Tregs suppression so that effective immune functions can be carried out to induce rejection quickly. Thus, prevention of rejection and establishment of tolerance by Tregs require attenuation of Teffs responses and inflammation control.

4.2 Tregs and allogeneic hematopoietic cell and bone marrow transplantation

Currently, allogeneic hematopoietic cells transplantation (HCT) or bone marrow transplantation (BMT) in humans is widely used in the treatment of tumors of the hematopoietic and immune systems, including leukemia, lymphoma, and myeloma. However, they are usually complicated by serious and potentially lethal side effects, such as immunodeficiency and graft-versus-host disease (GVHD). GVHD represents a dysregulated immune response and has been assessed across both major and minor histocompatibility barriers, and the pace of these reactions is much more accelerated across major histocompatibility barriers. The onset and course of GVHD depend on the degree of major and minor MHC disparity and the T-cell dose. It has been demonstrated by using animal models that T cells rapidly migrate to nodal sites, spleen, and mesenteric lymph nodes and begin to dramatically expand by 3–4 days following adoptive transfer, and within 7–10 days, they infiltrate the major sites of GVHD pathophysiology, such as lymph nodes, spleen, gastric intestinal tract (GI tract), liver, and skin [68]. Depletion of CD4⁺CD25⁺ T cells from the donor graft accelerated the GVHD course and increased lethality, which provided evidence for the role of Tregs in mediating GVHD [69]. Tregs also expand dramatically upon adoptive transfer and traffic to nodal sites to promote immune reconstitution and suppress GVHD across both major and minor histocompatibility barriers, while interestingly allowing for the maintenance of graft-versus-tumor (GVT) responses [70, 71, 72]. Tregs proliferate in the same way as Tconvs with similar kinetics and tend to fade out over time. Upon the adoptive transfer, the dramatic expansion of Tconvs can be detected, whereas when the same numbers of Tregs were adopted along with the Tconvs, this dramatic proliferation of Tconvs is significantly reduced, yet the homing and activation of Tconvs are not impacted [73]. It is indicated that the adopted Tconvs are still able to be activated and home to specific sites within the body, yet this drastic T-cell expansion required for GVHD is diminished. Thus, clinical strategies to enhance the function of Tregs hold great promise to improve outcomes following allogeneic HCT and BMT.

5. Tregs therapy

The ability of Tregs to maintain self-tolerance means they are critical for the control and prevention of autoimmune diseases. Currently, a large body of data in the literature has provided evidence on the possible Tregs therapy for various immune-mediated diseases. Restoring immune homeostasis and tolerance through the promotion, activation, or delivery of Tregs has emerged as a focus for therapies aimed at curing or controlling autoimmune diseases. A variety of Tregs-based therapies are being explored in the treatment and prevention of autoimmune diseases, such as ex vivo-expanded polyclonal Tregs or Tregs transduced with an autoantigen-specific TCR, chimeric antigen receptor (CAR). In addition, some other non-cell-based therapies related to Tregs, including low-dose IL-2 and heat shock protein (HSP), may also be beneficial.

5.1 Polyclonal Tregs therapy

Polyclonal Tregs therapy uses autologous ex vivo-expanded Tregs to restore tolerance and is considered a next-generation cellular therapy for several autoimmune diseases and inflammatory immune disorders. Tregs isolated from peripheral blood are stimulated and expanded in vitroby using anti-CD3/CD28 antibody-coated beads and high dose IL-2, or anti-CD28 super agonists [74, 75]. The first preclinical proof of concept for use of polyclonal Tregs was demonstrated in 1995 that CD4⁺CD25⁺ T cells could be used to transfer tolerance in athymic nude mice by suppressing self-reactive lymphocytes [76]. Then, expanded Tregs with a polyclonal specificity are reportedly more efficient in suppressive function and have demonstrated potential in various preclinical models of GVHD [77], solid organ transplantation [78], and autoimmune diseases [79]. Several clinical trials have been carried out to examine the safety and feasibility of polyclonal Tregs for T1D [80], transplantation [81], and GVHD [82], and the use of polyclonal Tregs in these diseases have shown significant therapeutic potential. For example, one robust clinical trial on T1D with polyclonal Tregs demonstrated that the expanded autologous Tregs retained their T cell receptor diversity and owned enhanced functional activity. Fourteen adult subjects with T1D received ex vivo-expanded autologous CD4⁺CD127^low/−CD25⁺ polyclonal Tregs (0.05 × 10⁸ to 26 × 10⁸ cells). A transient increase in Tregs, which retained a broad CD4⁺Foxp3⁺CD25^hiCD127^low phenotype long-term, was detected in recipients. There were no infusion reactions or cell therapy-related high-grade adverse events [80]. Besides, some other clinical trials for polyclonal Tregs therapy in autoimmune hepatitis (NCT02704338), Crohn’s disease (NCT03185000), Pemphigus (NCT03239470), and Alzheimer’s disease (NCT03865017) are also under investigation. Positive results from these clinical trials have allayed the concerns that polyclonal Tregs therapy would promote generalized immune suppression, leading to an increased risk of infection and cancer, which has not been found.

However, the therapeutic effect of this clinical trial correlated with increased Tregs post-infusion, and only persisted for a short time. The subsequent trials confirmed the limited persistence of expanded Tregs even after a second infusion [83], and obtaining sufficient cell numbers can be challenging in many disease scenarios [84], although polyclonal Tregs therapy is generally considered safe and efficacious. Perhaps the use of other Tregs-promoting therapies in combination with polyclonal Tregs therapy would prolong the suppressive effect and increase the number of Tregs with improved patient outcomes.

5.1.1 Low-dose IL-2 in combination with polyclonal Tregs therapy

As mentioned earlier, it has been widely accepted that IL-2 plays a critical role mainly in Tregs fitness and homeostasis, thus low-dose IL-2 therapy alone has the effect of expanding in vivoTregs. Co-administration of polyclonal Tregs and low-dose IL-2 has been considered as an additional strategy to restore the defective Treg pool and is expected to boost Tregs number and function after administration. Recent work has reported the possibility to expand Tregs using low-dose IL-2 in vivo[85]. In this phase I-II clinical trial on 46 individuals with mild to moderate forms of various autoimmune diseases, including RA, ankylosing spondylitis, systemic lupus erythematosus, psoriasis, Bechet’s disease, granulomatosis with polyangiitis, Takayasu’s disease, IBD, AIH and sclerosing cholangitis, all the patients received low-dose IL-2 (1 million IU/day) for 5 days, followed by fortnightly injections for 6 months. Low-dose IL-2 can be well tolerated whatever the disease and the concomitant treatments, and specific Tregs expansion and activation can be detected in all patients, without effector T cell activation. The increase in Tregs percentage was mainly evident on day 8 and was then contained thereafter, despite the levels remaining slightly higher compared to baseline. However, specific data on AIH patients (n = 2) were not presented, which indicated the effects of low-dose IL-2 in AIH patients remain unclear. In another study on two AIH patients with persistent disease activity [86], low-dose IL-2 was administered at 1 million IU for 5 days monthly for a total of 6 months. The proportion of circulating Tregs increased in both cases with a peak observed on day 9 and returning to baseline levels on day 28. This suggests that the effect of low-dose IL-2 on Tregs frequency is transient. Given the small number of cases enrolled, further studies in larger numbers of subjects should be performed to assess the efficacy as well as the long-term effects of this treatment on Tregs, particularly on their suppressive function, expansion, and plasticity.

Furthermore, an advantage of low-dose IL-2 therapy is that recombinant human IL-2 is already available as a therapeutic drug called Aldesleukin or Proleukin for the treatment of malignant melanoma and renal cell carcinoma in the clinic [87].

5.1.2 HSP in integrating Tregs expansion and activation

HSPs are highly conserved proteins present in all kingdoms of organisms, and expressed under stress conditions to protect the cells from injuries. They are classified into five families according to their molecular weight, including HSPH, HSPC, HSPA, HSPD, and DNAJ. Intracellular HSPs play an essential role in physiological processes, involving of folding of nascent and stress-accumulated protein-substrate assembly and preventing the aggregation of these proteins, transporting across membranes and degrading other proteins. While extracellular or receptor-bound HSPs mediate immunological functions and immunomodulatory activity, including the induction, proliferation, suppressive function, and cytokine production of Tregs [88]. In patients with juvenile idiopathic arthritis (JIA), DNAJ was found to improve the suppressive function of Tregs in culture and stimulate T cells for the production of IL-10, and high serum levels of DNAJ correspond with a milder course of the disease, indicating epitopes derived from human DNAJ can induce differentiation and/or stimulate cell proliferation of Tregs [89]. Acting as co-stimulators of human Tregs, HSPD can enhance the suppression and proliferation of Tregs via binding of Toll-like receptor (TLR) 2 on the Tregs surface to inhibit target T cell proliferation, IFN-γ and tumor necrosis factor (TNF)-α secretion, as well as upregulate the expression of IL-10 [90]. HSPD can enhance the differentiation of cord blood mononuclear cell (CBMC) into CD4⁺IL-10⁺Foxp3⁺ Tregs as well [91]. HSPA can stimulate the suppressive activity of Tregs, increase the production of IL-10, and downregulate the production of inflammatory cytokines via the TLR4-signaling pathway, which may be important for Foxp3 induction [92]. Animal studies have shown that oral, nasal, intraperitoneal, or intradermal administration of HSPA significantly inhibits the development of the autoimmune arthritic model, which suggested that suppression of autoimmune response in experimental animals was mediated by increased expansion of Tregs specific for HSPA, and the secretion of anti-inflammatory IL-10 [93, 94, 95]. Moreover, HSPC can promote Tregs-dependent suppression as well [96]. HSP gp96, the endoplasmic reticulum form of HSPC, is required for Tregs maintenance and function, as loss of GP96 resulted in instability of the Tregs lineage and impairment of suppressive functions in vivo. In the absence of HSP gp96, Tregs are unable to maintain Foxp3 expression levels and can lead to systemic accumulation of IFN-γ-producing and IL-17-producing T cells, because HSP gp96 is an essential chaperone for the cell-surface protein glycoprotein A repetitions predominant (GARP). GAPR is a docking receptor for latent membrane-associated TGF-β (mLTGF-β). The HSP gp96-deficient Tregs prevent the expression of mLTGF-β and resulted in inefficient production of active TGF-β [97]. Meanwhile, immunization of HSP gp96 can increase Tregs frequency, expansion, and suppressive function, which shows obvious therapeutic effects in a Lyn^−/− mouse model of systemic lupus erythematosus and myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE) [98].

5.1.3 Foxp3-transduced T cells

Tregs constitutively express the transcription factor Foxp3, which is critical for their immunosuppressive function. Several studies have provided evidence that ectopic expression of Foxp3 can confer a suppressive phenotype to naïve or memory CD4⁺ T cells, so it is probably a way to circumvent the requirement of a large number of polyclonal Tregs for therapy [99]. Lentiviral delivery of the Foxp3 gene into IPEX patient-derived CD4⁺ T cells can acquire the characteristic features, such as decreased proliferation, hyporesponsiveness, reduced cytokine release, and suppressive activity, which are able to mirror the Tregs population from healthy donors, and these induced Tregs were demonstrated to be stable in inflammatory conditions not only in vitrobut also in vivoin a xenograft mouse model of GVHD [100]. Several other studies have also shown the efficiency of Foxp3-transduced Tregs in combating autoimmune diseases, such as allergy [101] and collagen-induced arthritis [102]. Recently, the CRISPR/Cas9 system has been applied to the domain for stable and high-level expression of Foxp3 in Tconvs, and these edited Tregs-like cells were able to suppress the immune response in a xeno-GVHD mouse model [103]. These studies demonstrate the applicability of gene correction in the treatment of autoimmune diseases.

Cell permeable form of Foxp3 is another approach to enforce Tregs differentiation. This protein form can link to the protein transduction domain (PTD) from the HIV transactivator of transcription and allow Foxp3 to be delivered to the cytoplasm and nucleus, which has been shown to induce a Tregs phenotype in both human and mouse T cells [104]. However, a major limitation of this approach is the high cost for human patients.

5.2 Antigen-specific Tregs

While the initially limited success of polyclonal Tregs is encouraging, the amounts of cells needed for infusions are quite large and the risk of nonspecific immunosuppression should be considered. Tregs developed in the thymus (i.e., nTregs) harbor a TCR repertoire that is skewed toward self-antigens, while Tregs induced in the periphery in an antigen-specific manner (i.e., pTregs) can be characterized with a TCR repertoire different from their nTregs counterparts [105]. So, it is a good strategy to induce disease-relevant antigen-specific Tregs with the goal to interfere with the unwanted immune reactions in allergies and autoimmunity and to restore the self-tolerance, and it has been verified through considerate research in humans or mice [106, 107]. Compared with the polyclonal Tregs therapy, growing evidence from animal models indicates that antigen-specific Tregs may be more efficient in controlling pathological immune responses in a disease-specific manner. It is possible because infused Tregs migrating toward the tissues of cognate antigen exposure will lead to more effective and localized control of inflammation, along with risk reduction of broad immunosuppression and its related adverse events [108, 109]. Moreover, the enhanced migration ability of antigen-specific Tregs to target tissues can probably lead to a lower administration number of Tregs than the polyclonal approach, and facilitate the obtainment of Tregs via standard in vitroexpansion protocols.

5.2.1 TCR-Tregs therapy

Tregs therapy can be enhanced by the introduction of an autoantigen-specific TCR (TCR-Tregs), which have the ability to redirect their response toward the desired autoantigen specificity. Tregs can be ex vivo-transduced to express a high-affinity and autoantigen-specific TCR by way of retroviral or lentiviral transduction and subsequently expanded to treat a specific autoimmune disease. For instance, a low number of autoantigen-specific Tregs were needed to sufficiently prevent or even reverse T1D in a NOD mice model, which was engineered to express a diabetogenic TCR [108]. Another group showed that as few as 2000 antigen-specific Tregs were all that was required to prevent T1D in mice [110]. Preclinical studies in mouse models have also shown that TCR-engineered Tregs are more effective in suppressing the Teffs responses against specific antigens in autoimmune diseases, such as colitis, multiple sclerosis, and arthritis [111, 112, 113], even in tolerance induction to MHC mismatched heart grafts [114].

Compared to polyclonal Tregs, fewer antigen-specific Tregs may be needed to alleviate autoimmune disease; however, the challenge of the identification of an appropriate, high-affinity, autoantigen-specific TCR for transduction onto Tregs still remains, due to some autoimmune diseases being with poorly defined dominant epitopes. It is hard to isolate and identify antigen-specific Tregs due to both the great diversity in TCRs and very low count of them naturally circulating in the peripheral blood. The majority of antigen-specific Tregs were generated using TCRs isolated from Tconvs, which would influence the stability, avidity, and migration to specific parts of the engineered Tregs, for the reason that the intrinsic affinity and specificity of TCRs isolated from Tregs are distinct from Tconvs. Moreover, there are some other limitations of this approach, such as the requirement for MHC restriction and the risk of mispairing with endogenous TCR.

Single-cell sequencing is required for TCR identification since each T cell clone expresses a different TCR sequence from the others, and the successful sequencing of both the α and β chain TCR is required to successfully identify one TCR [115]. In a recent study, single-cell TCR analyses of islet Tregs revealed their specificity for insulin and other islet derived antigen, and these antigen-specific Tregs were reported to be efficient in protecting NOD mice from diabetes [116].

5.2.2 CAR-Tregs therapy

Although Tregs engineered with TCRs (TCR-Tregs) seem to be promising, they are still MHC-restricted and their modular application in individual patients is constrained. Engineer with genes encoding chimeric antigen receptors (CARs), which typically consist of a single-chain variable fragment (scFv) for binding to a monoclonal antibody, an extracellular hinge, a transmembrane region, and intracellular signaling domains, is an MHC-independent strategy of generating antigen specificity for Tregs [117]. In animal models, CAR-Tregs have shown great potential for treating different diseases, especially allograft rejection and various autoimmune diseases.

HLA-A mismatching is often associated with poor outcomes after transplantation, so, HLA-A is a potential target antigen to generate antigen-specific Tregs for inducing transplantation tolerance. One kind of HLA-A2-specific CAR (A2-CAR) Tregs was created in a peptide-independent manner, and not only can maintain high expression of canonical Tregs markers, including Foxp3, CD25, Helios, CTLA-4, and a high degree of demethylation of the Treg-specific demethylated region (TSDR) of the FOXP3 locus but also can enable stronger antigen-specific activation than did an endogenous TCR [118]. Further, CAR-stimulated Tregs had a higher surface expression of CTLA-4, latency-associated peptide (LAP), and the inactive precursor of TGF-β than TCR-Tregs. Unlike TCR-Tregs, CARs could also stimulate IL-2-independent Tregs proliferation in the short term [118]. Thus, CAR-Tregs may be superior to TCR-Tregs.

CAR Tregs isolated from transgenic BALB/c mice with a CAR specific for 2,4,6-trinitrophenol (TNP), an antigen commonly used in a mouse model of colitis, were reported to capable of suppressing the proliferation of Teffs in vitroeven in the absence of B7-CD28 co-stimulation, and the mortality rate of TNP-CAR-tg mice significantly decreased in comparison with WT mice [119]. In situ fluorescent micro endoscopic evaluation verified that TNP-CAR Tregs localized to the inflamed colonic mucosa. Thereafter, a novel protocol that enabled efficient and reproducible retroviral transduction and expansion of murine nTregs was developed in a non-transgenic mouse model, bringing about a highly enriched population of TNP-specific Tregs. The TNP-CAR Tregs show suppressive capabilities to Teffs both in vitroand in vivo, and TNP-CAR Tregs-mediated suppression in vitrowas partially dependent on cell-cell contact but not on IL-10 or TGF-β1 [120]. Moreover, based on the previously engineered Tregs to express a TCR specific for a myelin basic protein (MBP) peptide, which can suppress the proliferation of MBP-reactive Teffs and ameliorated MOG-induced EAE, the approach by creating human Tregs expressing functional single-chain CAR (scFv CAR), targeting either MBP or MOG was extended. These scFv CAR-transduced Tregs retained Foxp3 and Helios after long-term expansion in vitro. Importantly, these engineered CAR-Tregs were able to suppress autoimmune pathology in EAE, demonstrating that these Tregs have the potential to be used as a cellular therapy for multiple sclerosis (MS) patients [121].

5.3 Taking Tregs into medicine

5.3.1 Comparison table showing different approaches, techniques, and stages among studies

Different approaches that involve boosting Tregs have been tested in several disease settings so for. Polyclonal Tregs and antigen-specific Tregs therapy have demonstrated their efficacy in immunotherapy in various clinical trials or preclinical models (Table 1).

Approach	Technique	Indication	Stage of study	Study ID or references
Polyclonal Tregs therapy	Autologous polyclonally expanded Tregs	T1D	Clinical trials phase 1 (completed)	NCT01210664
Polyclonal Tregs therapy	Ex-vivoexpanded donor regulatory T cells	GVHD	Clinical trials phase I (active)	NCT01795573
Polyclonal Tregs therapy	Autologous polyclonally expanded Tregs	Kidney transplant	Clinical trials phase I/II (Active)	NCT02129881
Polyclonal Tregs therapy	Donor alloantigen reactive Tregs	Liver transplant	Clinical trials phase I (recruiting)	NCT02188719
Polyclonal Tregs therapy	Autologous polyclonal expanded nTregs	AIH	Clinical trials phase I/II (unknown)	NCT02704338
Antigen-specific Tregs therapy	CD4^+CD25^+ T cells from TCR-transgenic BDC2.5 mice expanded in vitrowith BDC peptide and NOD DCs	T1D	Preclinical studies (NOD model)	[122]
Antigen-specific Tregs therapy	CD4^+ T cells transduced with Foxp3 and a TCR of a CIA-associated T cell clone	RA	Preclinical studies (DBA1 mice)	[123]
Antigen-specific Tregs therapy	CD4^+CD25^+ T cells from TCR-transgenic Tg4 mice expanded in vitrowith anti-CD3/CD28 beads	MS	Preclinical studies (B10.Pl mice)	[124]
Antigen-specific Tregs therapy	CAR-engineered CD4^+CD25^+ Tregs specific for CEA	Colitis	Preclinical studies (CEABAC mice)	[111]
Antigen-specific Tregs therapy	CAR-engineered human CD4^+CD25^+ Tregs specific for HLA-A2	Skin transplantation	Preclinical studies (CEABAC mice)	[125]
Antigen-specific Tregs therapy	TGF-β-induced iTregs generated from CD4^+ T cells of TxA23 mice	Autoimmune gastritis	Preclinical studies (BALB/c mice)	[126]
Antigen-specific Tregs therapy	TGF-β-induced OVA-specific iTregs generated from CD4^+ T cells of OT-II mice	GVHD	Preclinical studies (C57Bl/6 mice)	[127]

Table 1.

Clinical trials or preclinical models with polyclonal Tregs or antigen-specific Tregs in different diseases.

5.3.2 Challenge and bottleneck of Tregs therapy

To sum up, Tregs are crucial in maintaining tolerance. Hence, Tregs immunotherapy is an attractive therapeutic option in autoimmune diseases and organ transplantations. However, there are still many challenges and bottlenecks in implementing Tregs therapy.

At first, the cellular variability of Tregs is wide. It is important to characterize the phenotype and suppressor function of each subtype of Tregs present in the periphery or the thymus. The success of Tregs therapy depends initially on the isolation and characterization of cells, while current research does not use a universally applicable standard for Tregs identification. This gap in identification leads to conflicting and doubtful research results. Meanwhile, one of the drawbacks of this cell therapy is the time delay to administer Tregs from taking peripheral blood to obtaining sufficient numbers of cells, and antigen-specific Tregs technology may presumably need administration of lower Tregs numbers than polyclonal approachesSecondly, to improve the efficacy of Tregs immunotherapy, it is necessary that Tregs can migrate, survive, and function in the specific target tissue. The plasticity of polyclonal or CAR-Tregs in an inflamed microenvironment is still an unknown factor. The inflamed microenvironment enriched with pro-inflammatory cytokines can either lead to a reduction in the potency of Tregs or resistance of Teffs to Tregs suppression, or even converting Tregs into pathogenic Teffs. There are also questions to be addressed regarding the long-term proliferative potential and survival of polyclonal or antigen-specific Tregs in the tissue microenvironment, which is enriched with cytokines, metabolites, low oxygen levels, and microbial peptides.

Thirdly, the application of CAR-Tregs is an exciting option in both transplantation and autoimmune diseases, when the antigen is known. Nevertheless, before CAR-Tregs can be put into practice in the clinic, there are still obstacles required to be overcome, because antibodies specific for self- or alloantigen must be characterized to construct antigen-specific CAR-Tregs. For the reason that autoimmune diseases always have a large autoantigenic repertoire of T or B cells, or spreading epitope, it will be not adequate to focus Tregs therapy on one specific epitope for an autoantigen. Although CAR-Tregs own a greater affinity to the cognate antigen than TCR-Tregs, the former requires the target cells to have at least 100 target autoantigens for successful recognition and Tregs stimulation. Moreover, it has not been confirmed yet whether CAR-Tregs would also lead to adverse reactions, such as cytokine storm and neuronal cytotoxicity, as the treatments with anti-tumor CAR-T cells.

Besides, it is still a hard nut to crack to access the localization of infused Tregs to the exact target site, and exhaustion of Tregs may limit their efficacy in immunosuppression. Meanwhile, the choice of immunosuppression in patients with Tregs therapy is crucial, for example, rapamycin has been shown to enhance Tregs frequency. To achieve efficacious and successful Tregs therapy, it is necessary to continue on immunosuppression that is favorable to Tregs survival and proliferation.

Therefore, more work is required to administer Tregs therapy effectively and safely to restore tolerance in transplantations and autoimmune diseases.

6. Prospect (perils and promises)

Tregs have proved to be a major breakthrough as an exciting immunotherapy option in the last two decades. Early phase clinical trials demonstrated safety, feasibility, and early efficacy with Tregs therapy in both autoimmune diseases and organ transplantation. The development of antigen-specific Tregs and CAR-Tregs would lead to exciting new frontiers in the cell therapy field as these cells are more efficacious and lesser numbers are required due to their target tissue homing affinity. It is crucial to obtain tissue biopsies following Tregs infusion to access the localization of infused cells. Optimizing the manufacturing processes and culture media will support infused Tregs survival in future clinical trials. In addition, improving our understanding on the patient’s omics profile with new technology will also allow us to put the personalized Tregs immunotherapy into effect.

In a word, although challenges still remain, the prospect of Tregs immunotherapy is exciting if the cell therapy community can maintain the collaboration closely. The immunosuppression-free period for patients with autoimmune disease and transplantation is in front of us.