Engineered antigen-specific regulatory T cells for autoimmune skin conditions

Abstract

Regulatory T cells (Tregs) are a subset of T cells responsible for the regulation of immune responses, thereby maintaining immune homeostasis and providing immune tolerance to both self and non-self-antigens. An increasing number of studies revealed Treg numbers and functions in a variety of autoimmune diseases. Treg deficiency can cause the development of several autoimmune skin diseases including vitiligo, alopecia areata, pemphigoid and pemphigus, psoriasis, and systemic sclerosis. Many clinical trials have been performed for autoimmune conditions using polyclonal Tregs, but efficiency can be significantly improved using antigen-specific Tregs engineered using T cell receptor (TCR) or chimeric antigen receptor (CAR) constructs. In this review, we systematically reviewed altered frequencies, impaired functions, and phenotypic features of Tregs in autoimmune skin conditions. We also summarized new advances in TCR and CAR based antigen-specific Tregs tested both in animal models and in clinics. The advantages and limitations of each approach were carefully discussed emphasizing possible clinical relevance to patients with autoimmune skin diseases. Moreover, we have reviewed potential approaches for engineering antigen-specific Tregs, and strategies for overcoming possible hurdles in clinical applications. Thereby, antigen-specific Tregs can be infused using autologous adoptive cell transfer to restore Treg numbers and to provide local immune tolerance for autoimmune skin disorders.

1. The role of regulatory T cells (Tregs) in preventing autoimmunity

Tregs are a subpopulation of T cells, which play a key role in the regulation of the immune response. By controlling the immune response to self and foreign antigens, Tregs can prevent the development of autoimmune disorders [1]. Tregs are generated from immature CD4 single positive cells receiving T cell receptor (TCR) signals of intermediate strength to escape clonal deletion in the thymus and are committed to differentiate into thymic or natural Tregs (tTregs or nTregs) [2]. Alternatively, naïve CD4⁺ T cells can differentiate into inducible Tregs (iTregs) when exposed to transforming growth factor β (TGF-β) and interleukin (IL-2) in the periphery. This classical subset of iTregs stably expresses the transcription factor forkhead box protein 3 (FOXP3). Type 1 T regulatory cells (Tr1) is another most extensively studied subset of iTregs. Tr1 cells are induced by interleukin (IL-10) and are FoxP3-negative. These two distinct populations of iTregs produce IL-10 and TGF-β to elicit their suppressive activities [3,4].

iTregs maintain peripheral tolerance, modulate effector T cell responses in several autoimmune diseases and prevent allograft rejection [5]. The possibility of generating in vitro expanded antigen-specific iTregs has encouraged their clinical use in autoimmunity and Graft vs. Host Disease (GvHD) [6]. Circulating Tregs are considered heterogeneous, mainly due to their plasticity, and acquite features specific to the type of immune response that they control. Under inflammatory conditions and in autoimmune diseases, Tregs can lose FOXP3 expression and convert into effector type producing Th1, Th2, or Th17 cytokines, which impairs immune homeostasis and contributes to the progression and pathogenesis of the disease [5,7]. Therefore, highly stable FOXP3 expression in vitro is associated with a low risk for interleukin (IL-17) production in vivo under inflammatory conditions [7]. The suppression mechanisms of Tregs can be direct, by cell-to-cell contact, or indirect, through a bystander effect [8,9]. Examples of bystander mechanisms include the secretion of cytokines such as IL-10, TGF-β and IL-35 and the production of granzyme and perforin, enzymes leading to apoptosis in target cells [10]. Moreover, high expression of CD25 receptors enables Tregs to take up more IL-2 and “starve” the surrounding cells of this cytokine [11]. In the absence of Tregs, as in patients with FOXP3 mutations, autoreactive T cells can persistently attack healthy cells [8,12]. Limited availability of Tregs was observed in several autoimmune diseases such as rheumatoid arthritis (RA), alopecia areata, multiple necrosis, and vitiligo [13].

2. Overview of autoimmune skin conditions and principle immune mechanisms

Autoimmune disease is a condition arising from an abnormal immune response to self-antigens that can affect any part of the body, including the skin [14]. Autoimmune skin disorders include alopecia areata, pemphigoid and pemphigus, vitiligo, psoriasis, and systemic sclerosis (Fig. 1) [15]. Upon progression, autoimmune skin disorders severely affect the quality of life and can become life-threatening [16–18]. Both genetic and environmental factors might be involved in their etiology [19,20]. Greater knowledge of the genetic underpinnings will allow early recognition of persons predisposed to these diseases, whereas identification of environmental triggers and possible autoantigens can uncover opportunities for preventing autoimmunity [21].

Fig. 1.

Autoimmune etiology of autoimmune skin diseases. (a) Illustration of immune surveillance in healthy skin. Adoptive transfer of antigen-specific Tregs to control (b) autoimmune depigmentation in vitiligo, (c) autoimmune blistering of the skin in pemphigus vulgaris, (d) autoimmune psoriatic patches in psoriasis, (e) autoimmune hair loss in alopecia areata, (f) autoimmune fibrosis in systemic sclerosis.

2.1. Vitiligo

The above conditions affect the skin and can all potentially benefit from a greater Treg activity on site. Here we provide a comparison to benefits achievable for progressive depigmentation in vitiligo. Vitiligo is a skin condition caused by the loss of melanocytes [22]. Vitiligo has been identified as a T cell mediated autoimmune disease (Fig. 1b) [23]. Several studies support an increased abundance of CD8⁺ T cells reactive to tyrosinase, melanoma antigen recognized by T cells (MART-1) and other melanosomal proteins, as shown by HLA multimer staining of circulating lymphocytes from patients [24]. Another name for MART-1 is MelanA; in contrast to TYRP1, MART-1 and other melanosomal proteins induce cytotoxic T-cells, but not humoral responses [25]. These melanosomal proteins might thus be considered autoantigens [23,26]. Indeed, cytotoxic responses to these and other melanocyte antigens can eliminate melanocytes and cause depigmentation [23]. Recent studies suggest the role of exosomes derived under pathological conditions can impact Tregs and Th17 cells [27] and CD8⁺ balance [28], suggesting a contribution for exosomes in the underlying mechanism of melanocyte destruction in vitiligo [29]. In vitiligo, inducible heat shock protein 70 (HSP70i) plays a central role in the autoimmune response by activating dendritic cells (DCs). These activated DCs further initiate autoimmune reactivity towards melanocytes [30,31].

The percentage of immunosuppressive Tregs in vitiligo skin is greatly reduced compared to healthy skin, suggesting that autoreactive cytotoxic T cells remain uninhibited, contributing to depigmentation in vitiligo [32]. Increased Treg infiltration was directly associated with repigmentation in a spontaneous mouse model of the disease, which is suggestive of a role for Tregs in preventing autoimmunity towards melanocytes [33]. Indeed, reduced numbers of skin Tregs with impaired function have been described for vitiligo patients [32,34–36]. Contested by some, decreased numbers of circulating Tregs have been reported in vitiligo patients as well [35], particularly in patients with active disease [34,35,37]. Reduced Treg numbers were found in unaffected and perilesional skin, but this population may ultimately repopulate lesional skin of vitiligo patients, and restore immune homeostasis [32,38–41].

TGF-β can polarize naïve T cells to a Tregs phenotype, and induce FOXP3 expression in iTregs [42], while also supporting Treg function [43]. Importantly, the levels of TGF-β are reduced in serum and skin of vitiligo patients, potentially affecting Treg suppressive activity [44,45]. TGF-β titers, as well as IL-10 levels, are selectively decreased in active, but not stable vitiligo [46,47]. In vitiligo, reduced Treg infiltration to the skin has been explained by the limited expression of CCL22 in vitiligo patient skin, which otherwise favors Treg trafficking to a CCL2-rich environment [32]. A broad knowledge that exists about target antigens selectively expressed by melanocytes targeted in vitiligo can open the door to antigen-specific Treg therapy for this disease.

2.2. Pemphigus and pemphigoid

Autoimmune bullous diseases are classified as pemphigus and pemphigoid [48]. In pemphigus, autoantibodies target desmosomal proteins, leading to disruption of epidermal adhesion and subsequent intraepidermal blistering. In pemphigoid, autoantibodies target structural proteins of the dermo-epidermal junction, causing tense, subepidermal blistering. The most common types of pemphigus and pemphigoid diseases are pemphigus vulgaris (PV), which is mediated by circulating autoantibodies directed to desmoglein 3 (Dsg3) on the keratinocyte cell surface, and bullous pemphigoid (BP) caused by autoantibodies to the structural protein BP180 at the epidermal basement membrane zone (Fig. 1c) [49].

The contribution of Tregs to BP remains controversial. Recent studies showed a significant reduction in circulating Treg frequency [50,51] and decreased numbers of FOXP3⁺ and IL-10⁺ cells were observed in BP skin, while no differences were found in TGF-β cells when compared to psoriasis, atopic dermatitis and control skin. The inflammatory milieu may dictate decreased homing of Tregs in BP-affected skin as an unintended consequence, rather than a cause of disease [52]. Circulating CD4⁺CD25^high Tregs express a normal level of cutaneous lymphocyte antigen (CLA), and their inhibitory capacity was not affected in BP [53]. In fact, Treg-depletion in the DEREG mouse model of BP induced excessive inflammation and blistering, increased neutrophil infiltrates and expression of Th1 and Th2 cytokines and chemokines, suggesting a role of Tregs in BP pathogenesis [54].

In PV patients, a greatly reduced number of Tregs are found [55]. Expression of FOXP3 is downregulated among circulating patient Tregs of PV patients, suggesting reduced suppressive function [56]. Despite abundant infiltration of Tregs into PV skin [52], the link between Treg dysfunction and the progression of PV has become more evident from studies employing the Treg-depleted PV mouse model.

Increasing or decreasing the number of Tregs, by adoptive transfer or antibody-mediated depletion, impacts autoreactive antibody production and disease development in PV mice [57,58]. Meanwhile, Dsg3-specific Tregs can be developed from Dsg3^−/− mice and adoptively transferred into Dsg3-sufficient mice to effectively suppress anti-Dsg3 antibody production in the PV mouse model [57]. Type 1 regulatory cells (Tr1 cells) are responsible for maintaining immune tolerance against self and non-self-antigens [59]. Dsg3-reactive Tr1 cells that inhibit proliferation of Dsg3-responsive autoreactive Th cells via secretion of IL-10 and TGF-β are more frequent in peripheral blood of healthy donors expressing PV-predisposing HLA than in actual PV patients [60]. Thus, autoantigen-specific Tregs can serve to suppress Dsg3-specific immune responses and associated disease in PV.

Efforts have been made to induce tolerance in PV by administering Dsg3 amino acids 186–204, which specifically bind PV-associated HLA-DR alleles, but this did not affect autoantibody titers against Dsg3 [57], possibly due to an absence of co-stimulation required to generate antigen-specific Tregs in the trial.

2.3. Psoriasis

Psoriasis is a chronic inflammatory skin disease mediated by both innate and adaptive immune responses. The disease is characterized by incomplete, defective basal keratinocyte differentiation induced by an inflammatory cascade in the dermis involving dendritic cells, macrophages, mast cells, and T cells (Fig. 1d) [61]. Parallels have been drawn between the etiopathomechanism of psoriasis and vitiligo [62–64], which has contributed to a rich history of therapeutics repurposed from psoriasis to vitiligo, including light therapies [65–67]. However, the skin microenvironment in active disease is quite different in either case. Nevertheless, both psoriasis and vitiligo can benefit from immune tempering measures, and functional skin Tregs can be of therapeutic benefit in either case. Several autoantigens derived from keratinocytes have been identified to explain the dominant T cell component observed in psoriatic skin. This includes LL37 cathelicidin/nucleic acid complexes, a newly recognized group of lipid antigens [68]. These may trigger initial activation of T cells, particularly IL-17-producing Th17 cells, T helper (Th)1 and Th22 cells [68].

A Treg-related autoimmune etiology of psoriasis has been reported [69]. Despite contradictory data concerning the frequency of Tregs in peripheral blood of psoriasis patients, a profound difference in Treg functional activity has been observed. Sugiyama H. et al. showed that circulating CD4⁺CD25^high cytotoxic T-lymphocyte antigen 4-positive (CTLA-4⁺) FOXP3^high Tregs of patients with chronic plaque type psoriasis display an anergic phenotype and exhibit decreased potential to suppress T cell proliferation in vitro [70–72]. In patients with moderate and severe psoriasis, tempered suppressive activity was accompanied by increased type 1 cytokines, STAT3 phosphorylation and proinflammatory cytokines (IL-6, IL-21 and IL-23) [73]. A pathogenic switch of patient Tregs into IL-17-producing RORγt⁺ Th17 was observed, with CD4⁺FOXP3^+/−IL-17⁺ cells infiltrating the lesional dermis of patients with severe disease [74]. Interestingly, the proportion of Treg among CD4⁺ T cells was elevated in non-lesional skin of patients with chronic plaque psoriasis [75], whereas decreased Treg infiltration was associated with acute but not chronic skin lesions, suggesting that a transient deficiency in Tregs supports the initiation of new lesions to exacerbate psoriasis [71].

A therapeutic role for Tregs in restoring self-tolerance has been demonstrated in mouse models of psoriasis. Tregs suppressed infiltration of lesional skin by GM-CSF⁺ T cells, resulting in disease regression, while Treg depletion exacerbated imiquimod-induced disease [76]. Such depletion of FoxP3⁺ Tregs allowed for type I IFN production by mononuclear phagocytes and accumulation of CD8⁺ T cells [77]. In human TNF-α-induced psoriatic arthritis, Tregs suppressed macrophage infiltration and the development of psoriasis [78]. In the PL/J murine model, Treg activity reduced psoriatic skin inflammation, and CD18 expression proved essential for the Treg-DC crosstalk that induces proliferation and TGF-β-dependent function of alloantigen-specific Tregs [79]. In conclusion, there is a definitive and unique role for Tregs in psoriasis pathogenesis, and great potential for Treg-based therapeutics in this keratinocyte-directed disease.

2.4. Alopecia areata

Alopecia areata (AA) is one of the most common autoimmune diseases that affects hair follicles. The autoimmunity is site-specific and is mediated by CD8⁺ T cells activated by aberrant epithelial MHC-I and MHC-II expression in the bulb of the hair follicle (Fig. 1e) [80,81]. The striking similarities between the etiology of vitiligo and AA further extend to the antigens targeted in either disease, as both melanocyte and keratinocyte-derived antigens in hair follicles may be targeted in alopecia [82]. Potential AA-associated antigens include keratinocyte-derived trichohyalin 1 (TCHH-1), trichohyalin 2 (TCHH-2) and melanocyte-derived tyrosinase (TYR), tyrosinase-related protein (TYRP2) [80,82].

Aberrant Treg function has been linked to AA pathology. Decreased frequencies of Tregs have been observed in the skin, in draining lymph nodes and also in the spleen of C3H/HeJ mice with chronic and transplanted AA. Low infiltration of Tregs in the AA lesions in mice was accompanied by low CD40 expression [83]. Transfer of Tregs obtained from skin draining lymph nodes of mice with a normal pelage to AA-affected mice prevented systemic AA development and site-specific hair loss in the latter [84].

Related data were obtained in humans. Low levels of CD4⁺FOXP3⁺ and CD25⁺FOXP3⁺ Tregs were found in the scalp lesions of patients with AA [85]. Compared to healthy controls, among CD4⁺CD25⁺FOXP3⁺, CD39-expressing Tregs were also reduced in the hair follicles of patients with active AA. However, no significant reduction in the percentage of circulating CD25⁺FOXP3⁺ Tregs or their memory subset CD25⁺FOXP3⁺CD45RO⁺ were found in AA patients with active disease [86]. At the same time, subcutaneous injections of low-doses of IL-2 to AA patients, resistant to conventional treatments, induced the recruitment of CD4⁺CD25⁺FOXP3⁺ Tregs into lesional skin and successful hair regrowth [87].

Single-nucleotide polymorphisms have been reported in chromosome regions encoding CD25, the ikaros family member Eos (IKZF4), CTLA-4, and FOXP3 in AA patients [88]. Data related to Treg-related cytokine abundance in AA are conflicting, with a decrease in serum TGF-β levels reported by some [89], and an increase by others when compared to healthy controls [90]. Surprisingly, no other cytokine differences (including IL-10) have been detected or reported in AA [89], but the impaired suppressive activity of circulating CD4⁺CD25⁺T CD127^low/− Tregs of AA patients was found in vitro [91]. Despite the fact that the mechanisms that render Treg dysfunction in AA patients and their recruitment defects to a site of AA are currently unknown, the data account for the crucial role of Tregs in the pathogenesis of AA and suggest the potential of their therapeutic utilization.

2.5. Systemic sclerosis

Systemic sclerosis (SSc) is a complex systemic autoimmune disease characterized by thickened and sclerotic skin, which also affects organs such as the lungs, the gastrointestinal tract, the heart and the kidneys [92]. SSc is the most severe connective tissue disease, associated with high mortality. The pathology of SSc involves three pathways, namely: (1) vasculopathy, (2) fibrosis caused by fibroblast dysfunction generating excessive accumulation of matrix components, and (3) aberrant immune activation leading to the production of autoantibodies and cell-mediated autoimmunity (Fig. 1f) [93].

Treg cells from SSc patients can inhibit T-cell proliferation and IL-2, IL-4, IL-5, IL-10, IFN-γ and TNF-α production, thus suggesting the presence of functional Tregs in SSc [94]. However, decreased frequencies of Tregs are found in peripheral blood of untreated SSc patients and SSc lesions [94–102]. The frequency of circulating Tregs is inversely correlated with disease activity, age of onset, staging and complexity in SSc [103–105]. The reduced abundance of circulating Tregs was associated with a marked increase in Th17 cells [106], and not surprisingly, decreased IL-10 and TGF-β serum levels. Comparable findings were reported for SSc lesional skin [102], though a reduction in Tregs was not accompanied by increased cutaneous Th17 [107]. Instead, in diffuse SSC, IL-4 and IL-13 producing Tregs are significantly increased to drive the differentiation and proliferation of fibroblasts and excess fibrosis [108].

Despite limited knowledge of the mechanism by which Tregs participate in tissue fibrosis and SSc, Treg augmentation could be a viable therapy for the disease. For example, in a clinical trial of chronic GvHD-cases that closely resemble that of scleroderma, patients were treated with low-dose IL-2. This treatment can boost Tregs, and it led to a decrease in skin fibrosis in some patients [109]. In SSc patients undergoing autologous hematopoietic stem cell transplantation, improved Treg function correlated with reduced skin fibrosis [110]. Also, in a bleomycin-induced pulmonary fibrosis mouse model, adoptive transfer of Tregs significantly reduced pulmonary fibrosis [111]. However, only late depletion of Tregs led to increased fibrosis [112], suggesting that Tregs therapy might be effective only in the late stages of SSc.

Some opposing data have been recently published. Though TGF-β is clearly involved in Treg suppressive functions, it is also strongly implicated with SSc pathogenesis, and targeting TGF-β-regulated gene expression with neutralizing antibody improved SSc clinical symptoms [113]. It is thus possible that TGF-β-regulated gene expression primarily mediates clinical manifestations of SSc. Further investigation of the therapeutic potential of Treg in SSc is needed. Treg activity may be enhanced by antigen specificity, potentially targeting overexpressed matrix components found in (proximity to) hyperactivated fibroblasts.

3. Treg-based methods tested to suppress autoimmune skin conditions

Therapeutics to enhance Tregs function offer a potential tool for the treatment of autoimmune diseases, and can serve to replace the chronic use of immunosuppressants that can be associated with undesirable side effects. Clinical trials have been initiated using polyclonal Tregs for autoimmune skin diseases; a phase I non-randomized, open-label clinical study was conducted for active cutaneous lupus. Patients received a single dose of 10⁸ autologous, polyclonal Tregs (NCT02428309), but the study was terminated due to limited recruitment feasibility. In an ongoing Treg-based, phase I clinical trial for PV and pemphigus foliaceus, patients have been administered a single dose of 1–2.5×10⁸ autologous Tregs, and the results of this study are pending (NCT03239470). Numerous preclinical studies addressing the treatment of autoimmune skin disorders hail the benefits of using antigen-specific Tregs expressing either a transgenic TCR or a CAR, and await translation to a clinical trial using genetically modified Tregs [6,114–118].

As an alternative to autologous Treg transfer, therapeutic agents can be used to support Treg function. Drug-based treatment is usually considered more cost-effective and may come at a lower risk for adverse events compared to adoptive transfer of live cells. A limited set of clinical trials of this kind is being conducted. Recent studies report that low-dose IL-2 has been associated with the promotion and expansion of Tregs in patients with autoimmune conditions [119,120]. This includes a large-scale, ongoing clinical trial to test low-dose IL-2 for the treatment of multiple autoimmune disorders, including skin conditions such as psoriasis, systemic sclerosis and Systemic Lupus Erythematosus (SLE) (NCT01988506) [120,121]. A recent review also discussed the potential of using low-dose IL-2 for RA [122]. In animal studies, rapamycin promoted Treg development via protein kinase B with subsequent mTOR inhibition [123], and provided benefit by reducing the effector T cell: Treg ratio [124–127]. However, in clinical trials of rapamycin, no significant benefit has been reported to date [124–127].

Supporting Treg skin homing can also be considered for the treatment of autoimmune skin diseases. By cutaneous CCL22 overexpression in a mouse model of vitiligo, Treg numbers were restored and continued treatment was able to suppress vitiligo development [128]. Delivery of CCL22 might be achieved by local needle-free jet injection of DNA to attract Treg to the injection site [129]. The same approach can support anti-tumor responses to melanoma, by redirecting Tregs away from the tumor and towards the skin [130].

Supporting microbial diversity or supplying particular microbes can also serve as preventive or therapeutic measures to drive Treg activity and alleviating autoimmune responses of the skin. This approach is gaining increased attention [131]. In one study, oral administration of Bifidobacterium infantis was associated with increased IL-10 secretion and FOXP3 expression in patient blood [132], and in an increase in Tregs in a mouse model of S. typhimurium disease [133]. We found that neomycin treatment can alter the microbiome of vitiligo-prone mice to significantly delay depigmentation and promote the infiltration of Tregs to the skin [134].

4. Why include antigen specificity?

In early-phase clinical trials, polyclonal Tregs have been used to prevent autoimmunity after allogeneic hematopoietic stem cell transplantation (HSCT) and in patients with type-1 diabetes, demonstrated treatment safety and efficacy [135]. However, the inadvertent suppression of immune responses to infection or malignancies can form an important consideration when applying polyclonal Tregs [136]. Moreover, generating polyclonal Tregs in numbers sufficient for clinical use can be challenging. These factors might be overcome by including antigen specificity to adoptively transferred cells. Animal studies performed for diabetes, central nervous system demyelinating autoimmune diseases, and allograft transplantation demonstrated that antigen-specific Tregs can provide superior protection from autoimmunity over polyclonal Tregs [116,137–144]. Thus, using antigen-specific Tregs might overcome the two primary challenges with polyclonal Treg therapy, namely production of potent Tregs with relevant specificity to provide superior protection from autoimmune activity.

5. TCRs and CARs: A functional comparison

Antigen-specific Tregs can be generated using TCR or CAR constructs. Engineered Tregs with specific TCR can be generated at high efficiency using retroviral or lentiviral constructs, but the use of TCR-Tregs is limited by its MHC-restriction requirement, which impacts the number of patients that will benefit from this therapy [145]. On the other hand, TCR-Tregs can respond to antigen regardless of the intended cellular expression site of the molecule. Tregs transduced with genes encoding a CAR will overcome this MHC-dependence. The latter constructs are instead composed of a single-chain variable fragment, scFv - the binding portion of a monoclonal antibody, followed by an extracellular hinge region, a transmembrane region, and intracellular TCR signaling domains. In CAR-T cell therapy, optimizing and selecting the correct CAR affinity and intracellular signaling domains is crucial for the resulting therapeutic activity and cellular persistence of the resulting T cells [146]. Since CARs are constructed using antibody variable regions, they hold a higher affinity to their cognate antigen compared to TCRs [147]. A limitation to the use of CAR Tregs is that a suitable antigen must be identified that is readily accessible by antibodies, generally on the surface of the targeted cell [148]. Moreover, high-affinity CAR stimulation, at least in CD8 T cells, can ultimately result in reduced activity and loss of specificity of host T cells [115]. Therefore, the optimal affinity range for the domains responsible for the specificity of the CARs, and the benefit of high affinity interaction form an area of active investigation.

6. Currently available TCRs and CARs of potential use for autoimmune skin diseases

Current cancer immunotherapies can involve using TCR and CAR-transduced T cells, with CD19-reactive CAR T cells now approved for the treatment of B cell lymphoma [149]. The outcomes suggest that similar approaches could hold promise for the treatment of autoimmune diseases [150,151]. Ex vivo-expanded autologous polyclonal CD4+CD25^high CD127^low/− regulatory T cells have been in clinical trials for Type 1 Diabetes and GvHD [152]. These Tregs were sorted based on surface markers to purify Tregs from peripheral blood, and sorted Tregs were cultured with clinical-grade activation beads in the presence of IL-2 [153]. Prior to infusion, FOXP3 expression and viability were established [152,154]. The use of CAR constructs to generate alloantigen-specific Tregs has shown promise in preclinical studies for organ transplantation, to prevent GvHD [6]. In one example, Tregs expressing an HLA-A2–specific CAR (A2-CAR) maintained a stable phenotype, with superior suppressive activity to polyclonal Tregs in vitro and prevention of xenogeneic GvHD caused by HLA-A2⁺ T cells in mice [6]. Another recent study reports the use of CAR Tregs targeting ganglioside D3 (GD3) antigen in a vitiligo prone mouse model. The results show enhanced protection from depigmentation when compared to untransduced Tregs [155]. The scheme of adoptive Treg therapy is shown in Fig. 2.

Fig. 2.

Scheme for Adoptive Treg transfer. (a) Leukapheresis of lymphocytes from peripheral blood of the patient. (b) Polarization of CD4⁺ T cells in to FOXP3⁺ Tregs in the presence of TGF-β. (c) Viral transduction of FOXP3⁺ Tregs with either TCR or CAR construct. (d) Tregs transduced with TCR and CAR construct. (e) Ex vivo expansion of Tregs to generate sufficient numbers for adoptive Treg transfer. (f) Sorting for Tregs using either magnetic or FACS based methods. (g) Infusion of either TCR or CAR Tregs into patients.

In another example, a chimeric autoantibody receptor (CAAR) specific for pathogenic antibody-producing B cells has been developed for the treatment of PV [156]. This CAAR was generated by fusing an autoantigen (Dsg3) to a CD137/CD3ζ signaling domain. Preclinical studies in a mouse model of PV showed that these Dsg3-CAAR T cells were functional against human anti-Dsg3 B cells and prevented blistering without measurable off-target toxicity. Thus, CAAR T-reg therapy might hold potential for deleterious PV by eliminating self-reactive B cells, and opens new avenues for the use of CAR Tregs in other autoimmune skin conditions [156,157].

7. Opportunities for antigen-specificity in vitiligo

Melanocytes express several currently known melanosomal proteins such as TRP-1 and TRP-2 [158]. Tyrosinase meanwhile, is a target antigen that has been targeted by TCR transgenic Tregs in a mouse model of melanoma. Brusko et al. tested human tyrosinase specific TCR (TyrTCR) transduced antigen-specific Tregs on a tumor model, and results showed the effective suppression of antigen-specific T cells to facilitate tumor growth [159]. Therefore, these TyrTCR Tregs might likewise suppress effector T cells in vitiligo and serve as a potential treatment for the disease. The concept is then, that tyrosinase can serve as a target for antigen-specific Tregs to suppress autoreactive T cells in vitiligo without eliciting non-specific immunosuppression [160]. Especially with the local deficiency of peripheral Tregs in vitiligo in mind, adoptive transfer of antigen-specific Tregs responsive to melanocytes might prevent general immunosuppression and bring superior specificity to the treatment.

8. Setbacks: how to generate Tregs in bulk and maintain a Treg profile

Adoptive Treg transfer has been administered in clinics for GvHD, transplantation, and autoimmune diseases [152,161]. However, ex vivo expansion and maintenance can become laborious and require manual handling, posing a risk for contamination. Isolation of Tregs for a clinical purpose can be performed using the “CliniMACS Prodigy®” fully automated system that meets GMP standards [162]. Peripheral blood can serve as a source of Tregs, with purification performed by magnetic enrichment in the sterile CliniMACS system. The isolation protocol includes the depletion of CD8⁺ and CD19⁺ lymphocytes followed by enrichment of CD25⁺ cells with a purity of FOXP3⁺ estimated at >90% and with yield up to 2.12 × 10⁹ cells [162]. Besides, flow cytometry-based sorting was also frequently used in clinical trials based on CD4 and CD25, and lack of CD127 surface marker expression to isolate Tregs of high purity (>99%) [152,163,164]. Among sorted Tregs, the CD4⁺CD25⁺CD127^lowCD45RA⁺ Tregs subset is most stable for FOXP3 expression, demonstrating resistance to Th17 effector conversion [165].

However, Tregs make up only ~3% of total T cells in peripheral blood, creating a challenge to generating sufficient numbers for infusion. Protocols for the expansion of Tregs are critical to achieving large numbers for clinical administration [166–168]. Polyclonal Tregs were 300-fold expanded in 5 weeks using activator beads and high dose IL-2 [169]. Some suggest adding rapamycin during expansion [164,170]. This immunosuppressant inhibits the mechanistic target of rapamycin (mTOR) protein kinase, which can inhibit effector T cell proliferation and favors FOXP3 upregulation [171], conferring higher stability and suppressive ability to the expanded Tregs [170]. Following a Treg expansion protocol using TGF-β together with rapamycin and IL-2, the expanded Tregs demonstrated higher suppression compared to the use of rapamycin alone [172]. Others developed a protocol to generate antigen-specific Tregs from co-culture with activated allogeneic B-cells or donor-derived DCs in the presence of IL-2 in organ transplantations. These antigen-specific Tregs also showed superior suppressive ability in vitro and in vivo compared to polyclonally expanded Tregs [153].

9. A reality check- Will antigen specific Tregs become available for autoimmune diseases?

In contrast to conventional immunosuppressive drugs, biologics, alkylating agents, and antimetabolites, Tregs can provide greater specificity with complex therapeutic benefits, and restore immune tolerance in various autoimmune diseases [173]. The application of polyclonal Tregs for autoimmune diseases has already have been attempted in clinical trials [152]. Overall, this and other phase I clinical studies provided answers to the isolation and expansion that surrounded Treg immunotherapy [152,163,174,175]. Yet the efficacy of polyclonal Treg transfer is not self-evident to date. Meanwhile, adoptive transfer of islet-specific Tregs outperformed polyclonal Tregs for blocking type 1 diabetes progression compared to polyclonal Tregs [173,176,177]. Unfortunately, diabetes is generally detected in patients when pancreas destruction is near complete, but other conditions may be more amenable to Treg based treatment in a clinical setting. Preclinical studies likewise suggest a superior efficacy of antigen-specific Tregs in transplantation procedures [144,[178], [179], [180]]. In different applications, antigen-specific CD19 CAR-T cells have been approved for clinical use by the Food and Drug administration (FDA) for large B-cell lymphoma and B-cell acute lymphoblastic leukemia [149]. It can be expected that antigen-specific CAR Tregs will meet with less side effects than CAR-T cell therapies in use for cancers [181]. Moreover, clinical protocols exist to counter cytokine release syndrome (CRS) and neurotoxicity after CAR Tregs infusion [152,174,175]. Thus, there is potential for antigen-specific Tregs to be prepared for the treatment of autoimmune skin diseases, which is expected to provide high efficacy while preventing non-specific immunosuppression.

A limitation to intravenous injection of antigen-specific Tregs might be that these much-needed immunosuppressive cells display a paucity at the desired site [177]. Should systemically applied Tregs not respond as required, the local injection might be needed, or the CCR4 Treg homing receptor ligand CCL22, can be introduced where Tregs are needed to attract systemically applied Tregs [128]. This leaves autoimmune diseases of the skin especially suited for adoptive treatment by antigen-specific Tregs when relevant antigens can be identified. To date, only alloantigen-reactive Tregs are currently being tested to prevent rejection after organ transplantation in clinical trials [177]. One of the other challenges of adoptive transfer is the cost and scalability of the technique. This has prompted the concept of developing off-the-shelf “Universal CAR Tregs” suitable for all patients [182].

Indeed, adoptive Treg treatment is expected to be most efficacious during active disease. In human patients however, progressive disease periods are interspersed with periods of inactivity, providing melanocyte stem cells with an opportunity to differentiate and repopulate the depigmented lesions For the intermittent treatment of vitiligo, it will be beneficial to store autologous CAR Tregs for later use [183]. By virtue of their antigen specificity, these Tregs might provide better safety profiles and decrease the risk of generalized immunosuppression.

10. Conclusion

The future of Treg-based therapy holds the potential for autoimmune skin diseases. Polyclonal Treg infusion was demonstrated to be safe and well-tolerated in patients. The results from currently ongoing clinical trials bring important insights regarding the Treg dose, efficacy and possible side effects. The safety of patients and well-understood protocols for managing possible side effects is crucial for the therapy. It is expected that the efficacy of Treg therapy will be enhanced if engineered as antigen-specific cells according to the several pre-clinical data. In vivo tracking will allow research groups to better understand the real potential of antigen-specific Tregs. The best practices for Treg-based therapy might be using antigen-specific Tregs accompanied by either IL-2 or rapamycin.