Regulatory T cell induction strategies and applications in the treatment of immune and non-immune diseases

Abstract

As a key regulator of the immune system, regulatory T cells (Treg cells) suppress excessive immune responses and maintain self-tolerance through various immunosuppressive mechanisms, both contact-dependent and contact-independent. They have demonstrated significant therapeutic potential in immune-related diseases and some non-immune diseases. In this paper, we systematically review the induction strategies of Treg cells, covering various aspects including antigenic stimulation, cytokine modulation, metabolic pathways, epigenetic regulation, drug induction, and engineered technology. Although each strategy has its own advantages, no single approach can comprehensively address the complex issues surrounding Treg cell stability, target specificity, and large-scale production. Therefore, we propose that integrating multiple induction strategies during the Treg cell induction process is crucial for designing Treg cell therapies with superior functionality and broader applications, thereby overcoming the current limitations of Treg cell therapies. In addition, we introduce their therapeutic applications in immune diseases (such as autoimmune disease, organ transplantation, allergic asthma, and cancer) and non-immune diseases (such as insulin resistance, tissue repair, fibrotic diseases, and osteoporosis), and recent advancements. We also discuss current research bottlenecks of Treg cell-based therapeutic strategies and provide an outlook on their future development, aiming to deepen the understanding of Treg cells and promote the advancement of Treg cell-based cellular therapeutic strategies.

Introduction

The fundamental mission of the immune system is to recognize and eliminate non-self-antigens, while maintaining a harmless state of dormancy toward self-components. The inherent contradiction of this dual task dictates that the immune system must possess both a powerful effector system and a finely tuned regulatory system. The effector system is typically composed of CD8⁺ cytotoxic T cells, CD4⁺ helper T cells, and innate lymphoid cells, working in concert to fend off invading pathogenic microorganisms and monitor tumor cells. The regulatory system, centered on regulatory T cells (Treg cells), is responsible for applying the brakes to excessive immune responses, thereby preventing the immune response from damaging normal tissues. Treg cells are a specialized subpopulation of T cells that are critical in downregulating immune responses, suppressing autoimmune diseases, and maintaining immune homeostasis.¹ In 1969, Nishizuka et al.² found that the development of organ-specific autoimmune diseases was observed when the mice were thymectomized at 3 days of age, but similar diseases did not occur in 7-day-old mice following thymus removal. This finding indicated that 3–7 days after birth is a critical period for the development of Treg cells in the thymus of mice. The removal of the thymus on postnatal day 3 induced an overproliferation of autoreactive CD4⁺ T cells, and by day 7, the body had already produced sufficient Treg cells to be able to exercise immune tolerance to self-antigens, which was the first evidence for the existence of Treg cells. In recent years, as research has continued to deepen, Treg cells have been recognized as the key cells for maintaining peripheral immune tolerance.

According to the classification of developmental origin, Treg cells are classified as thymic Treg cells (tTreg) and induced Treg cells (iTreg). tTreg cells are known to develop and mature in the thymus³ and have been demonstrated to play a pivotal role in the suppression of autoimmunity.⁴ Conventional T cells (Tconv cells) can also be converted to Treg cells in response to environmental signals that promote the expression of the forkhead box protein 3 (Foxp3) gene and immunosuppressive functions, resulting in a population of Treg cells called iTreg cells. When this occurs peripherally in vivo, these cells are referred to as peripheral-derived Treg cells (pTregs),^5,6 which can enhance the body’s tolerance to food and commensal microorganisms.⁴

Abnormal Treg cell numbers or functional inactivation show a significant association with the onset and progression of a variety of diseases. For example, in classic autoimmune diseases such as systemic lupus erythematosus and multiple sclerosis, Treg cells in peripheral blood or at local lesions often exhibit a trend toward a reduced relative proportion or impaired suppressive function, allowing effector T cells to escape restraint and attack the body’s own tissues. Similarly, in the context of solid organ transplantation and hematopoietic stem cell transplantation, insufficient host Treg activity is also a major driver of uncontrolled transplant rejection and graft-versus-host disease (GVHD). These findings have given rise to a therapeutic concept with enormous potential for clinical translation. This involves using artificial intervention to selectively expand or induce the differentiation of Treg cells under specific spatiotemporal conditions, thereby restoring immune homeostasis.

In this paper, we systematically review the strategies of Treg cell induction, including antigen stimulation, cytokine-based regulation, metabolic pathways, engineering techniques, and drug induction. Furthermore, a single induction strategy often struggles to balance induction efficiency, phenotypic stability, and safety; therefore, in this paper, we explore how to integrate these strategies to overcome the limitations of existing Treg cell therapies. We also summarize the main applications of Treg cells in immunotherapy and non-immunotherapy in recent years. Finally, we discuss the difficulties and challenges faced by Treg cells in the process of disease treatment and propose some possible strategies to address these challenges. We hope to provide some theoretical references for research work in related fields and to promote the further development of Treg cell-based disease treatment strategies. (Fig. 1)

Fig. 1.

Treg cell induction strategies and applications of Treg cells. TGF-β transforming growth factor β, IL-2 interleukin 2, IL-10 interleukin 10, IL-35 interleukin 35, CAR chimeric antigen receptor, APC antigen-presenting cells, Treg regulatory T cell, Foxp3 forkhead box protein 3, CTLA-4 cytotoxic T lymphocyte-associated antigen-4, TCR T cell receptor, LAG-3 lymphocyte activation gene-3. The graphic is created with BioRender.com

Immunosuppressive function of Treg cells

It is currently believed that Treg cells employ immunomodulatory functions through both contact-dependent and contact-independent inhibition mechanisms (Fig. 2). Treg cells can suppress the immune response by downregulating T cell co-stimulatory signals such as CD80/CD86 on the surface of antigen-presenting cells (APCs) through high expression of cytotoxic T lymphocyte-associated antigen-4 (CTLA-4).⁷ At the same time, attenuation of CD80/CD86 signaling can increase programmed cell death ligand 1 (PD-L1) on APCs, thereby inhibiting the function of effector T cells expressing programmed death receptor 1 (PD-1),⁸ providing a dual inhibitory effect on T cell immune responses. Lymphocyte activation gene-3 (LAG-3) is considered to be one of the biomarkers of Treg cells.⁹ In the presence of effector cells, LAG-3 expression on Treg cells is significantly enhanced, and LAG-3 can inhibit dendritic cells (DCs) maturation and activation by binding to major histocompatibility complex (MHC)-II-like molecules on the surface of DCs, leading to a decrease in their antigen presentation capacity.¹⁰ In addition, antigen-specific Treg cells can form strong interactions with DCs, and the strong binding force leads to the removal of the antigenic peptide major histocompatibility complex (pMHC) from the DC surface,¹¹ further reducing the antigen presentation capacity of DCs and suppressing the immune response. Besides, in the tumor microenvironment, Treg cells can induce effector T cell death in a granzyme B- or perforin-dependent manner, mediating immune downregulation.¹² Additionally, Treg cells have been shown to possess the capacity to elicit immunosuppressive effects through contact-independent mechanisms. Treg cells suppress immune responses by secreting the immunosuppressive cytokines interleukin 10 (IL-10),¹³ interleukin 35 (IL-35),¹⁴ and transforming growth factor β1 (TGF-β1). CD39/CD73, located on the surface of Treg cells, can synergistically hydrolyze extracellular adenosine triphosphate (ATP) to adenosine, thereby inhibiting the activity of effector cells via the adenosine A2A receptor (A2AR).^15,16 This process has been demonstrated to reduce the secretion of proinflammatory factors. Treg cells express high-affinity interleukin 2 (IL-2) receptors on their surface, which allows them to inhibit CD8⁺ T cell responses by competitively depleting IL-2.¹⁷ The conventional mechanism by which Treg cells exert immunosuppression is often considered non-specific and is a possible mechanism by which polyclonal Treg cells modulate Tconv cells at steady state. Under conditions of infection, infection-induced pathogen-associated epitope mimicry¹⁸ and tissue damage may result in the release of large quantities of self-antigens, thereby driving autoimmunity, whereas in practice, infection-induced autoimmune disease is uncommon, implying that Treg cells can selectively modulate the Tconv response to self-antigenic peptides without affecting the Tconv cell response to pathogen-derived exogenous antigens (Fig. 2).¹⁹ Researchers find that antigen-specific Treg cells are intrinsically poised to expand earlier than Tconv cells expressing the same T cell receptor (TCR) during infection. Although antigen-specific Treg cells are not able to prevent the antigen recognition and initial activation of Tconv cells, they can aggregate around Tconv cells with the same specificity, inhibiting their proliferation and differentiation through competition for self-antigen pMHC complexes and IL-2, thereby ensuring the clearance of pathogens and maintaining self-tolerance.¹⁹ Current evidence suggests that many tissue-resident Treg populations (including those in VAT,^20,21 skeletal muscle,^22,23 and GATA binding protein 3 (Gata3)⁺ colonic Tregs²¹) originate from tTreg cells, so we speculate that, compared to Tconv cells, these Treg cells have a higher affinity for self-antigens in infected tissues, thereby acquiring stronger proliferative capacity to maintain tolerance to self-antigens. In contrast, Tconv cells have a stronger binding affinity for foreign antigens, successfully activating the immune response against infection. Additionally, some studies based on mouse Treg cells have shown that the immune suppression mechanisms of Treg cells induced by different methods may vary. Studies have shown that tTreg cells do not always mediate immune tolerance by inhibiting the proliferation and activation of Tconv cells. For example, polyclonal tTreg cells can suppress Tconv cell migration to lymph nodes by downregulating their chemokine receptors (such as C-X-C chemokine receptor type 4 (CXCR4) and Sphingosine-1-phosphate 1 (S1P1)), reducing the recruitment of Tconv cells to target organs.²⁴ However, iTreg cells primarily suppress the antigen-presenting function of DCs through IL-10-dependent pathways, thereby inhibiting the T cell immune response.²⁴

Fig. 2.

Mechanisms by which Treg cells exert immunosuppressive functions. a Regulatory T cells (Treg cells) express high levels of cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), which has been shown to downregulate the surface expression of CD80/86 on antigen-presenting cells (APCs), thereby inducing immune tolerance. LAG-3 can bind to MHC-II molecules on the surface of APCs, resulting in the inhibition of APC maturation and activation and reducing their antigen-presenting capacity. b The attenuation of CD80/86 signaling can increase programmed death ligand 1 (PD-L1) expression on APCs, inhibiting conventional T cells (Tconv cells), which express programmed death receptor 1 (PD-1). c Treg cells kill effector T cells through granzyme B or perforin-mediated cytolysis. d Treg cells can competitively deplete interleukin 2 (IL-2) to suppress effector T cell activity. e Treg cells secrete inhibitory cytokines (IL-10, IL-35, TGF-β) to exert immunosuppressive functions. f CD39 and CD73, located on the surface of Treg cells, catalyze the hydrolysis of extracellular adenosine triphosphate (ATP) to adenosine, which subsequently binds to A2AR, thereby inhibiting effector T cell function. g Treg cells secrete cyclic adenosine monophosphate (cAMP) to induce metabolic suppression of effector T cells and inhibit the proliferation and activation of effector T cells. h Upon infection with pathogens expressing self-antigen peptides, self-peptide-specific Treg cells can suppress the proliferation and differentiation of Tconv cells with the same antigen specificity through competition for self-peptide-MHC complexes and IL-2, thereby preventing them from infiltrating target organs. This selective inhibition prevents the onset of autoimmune diseases while preserving the response of Tconv cells to foreign antigens. Mechanisms (a–c) are contact-dependent inhibition, and (d–g) are contact-independent inhibition. IL-2R interleukin 2 receptor, ADP adenosine diphosphate, ADO adenosine, A2AR adenosine A2A receptor, cAMP cyclic adenosine monophosphate, pMHC peptide major histocompatibility complex, TNF-α tumor necrosis factor-α. The graphic is created with BioRender.com

Phenotypes of Treg cells

In 1982, Shimon Sakaguchi and his colleagues conducted a seminal study that explored the cellular mechanism of the autoimmune response after thymectomy. They revealed for the first time that ovarian destruction in a model of autoimmune ovarian inflammation induced by thymectomy in mice 2–4 days after birth was mediated by Lyt-1 (CD5)⁺ T cells.²⁵ Since then, numerous studies have aimed to provide a more comprehensive characterization of these T cells, which can mediate immunosuppression by utilizing reliable cell-specific markers. In 1995, Shimon Sakaguchi was the first to identify a subpopulation of CD25⁺ CD4⁺ T cells, characterized by high expression of the alpha chain of the IL-2 receptor (CD25). These cells were able to maintain autoimmune tolerance by downregulating the body’s immune response to its antigens in a non-specific manner, marking the first identification of a subpopulation of T cells with an immunomodulatory function.²⁶

Forkhead box protein 3 (Foxp3)

Foxp3 is a member of the forkhead transcription factor family.²⁷ A clinically serious genetic disorder of X chromosome abnormality, immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome is caused by mutations in the Foxp3 gene.²⁸ In 2003, Shohei Hori discovered that the transcription factors encoded by the Foxp3 gene were expressed in mouse naturally occurring Treg cells.²⁹ However, these factors were found to be genetically defective in humans and mice suffering from autoimmune and inflammatory syndromes. Concurrently, it was determined that transfer of the Foxp3 gene by retroviruses could phenotypically convert naïve T cells to Treg cells. This revealed the important role of Foxp3 in the development of Treg cells. Foxp3 plays an important role in Treg cell development. In the same year, a series of studies further confirmed that Foxp3 was a key regulatory gene for the immunosuppressive function of Treg cells.^30,31 Furthermore, Foxp3 is not only required for the initial differentiation of Treg cells,²⁹ but also for the maintenance of the phenotypic characteristics and the suppressive function of mature Treg cells.³²

The mRNA of Foxp3 in both humans and mice is regulated by five elements, including the promoter and conserved non-coding sequences 0–3 (CNS0-3) (Fig. 3).^33,34 IL-2 signaling has been demonstrated to be essential for the development of Treg cells within the thymus.³⁵ During the process of Foxp3 induction, IL-2- signal transducer and activator of transcription (STAT) 5 signaling is observed to converge on enhancer CNS0, which facilitates the conversion of IL-2-dependent CD25⁺ Foxp3⁻ precursor cells to Treg cells within the thymus.³⁶ Furthermore, it has been shown that CNS0 and CNS3 can initiate and stabilize Foxp3 gene expression through enhancer-promoter and enhancer-enhancer interactions, which play a key control role in the development, maintenance, and functioning of Treg cells.³⁷ Foxp3 can achieve gene expression and function through the process of TGF-β-mediated mothers against DPP homolog (Smad) signaling.³⁸ Upon binding of TGF-β to its receptor, Smad family member 2 (Smad2) and Smad3 are activated and form a complex with Smad4, which translocates to the nucleus and binds to the CNS1 region, thereby initiating Foxp3 transcription.³⁹

Fig. 3.

Foxp3 loci and Foxp3 protein structural domains. a The primary functions of CNS0, CNS1, CNS2, and CNS3 in Foxp3, and the key transcription factors for each regulatory element. b Foxp3 protein structural domain, the primary functions of each domain, and the key interacting proteins. CNS conserved non-coding sequence, STAT5 signal transducer and activator of transcription 5, Smad mothers against DPP homolog, NFAT nuclear factors of activated T cells, Foxo forkhead box O, CREB CAMP-response element binding protein, RUNX1 runt-related transcription factor 1, CpG cytosine-phosphate-guanine, HIF-1α hypoxia-inducible factor 1α, RORγt retinoic acid receptor-related orphan receptor gamma t, HDAC histone deacetylase. The graphic is created with BioRender.com

Common phenotypes of Treg cells residing in different tissues

Treg cells have been identified in both lymphoid and non-lymphoid tissues of the body. In non-lymphoid tissues, Treg cells are typically resident.⁴⁰ Tissue-resident Treg cells have been shown to possess tissue-specific functions that contribute to the maintenance of tissue homeostasis.⁴¹ Treg cells residing in different tissues exhibit different phenotypes (Table 1).

Table 1.

Phenotypes and functions of tissue-resident Treg cells in different tissues

Tissue-resident Treg cells	Characteristic cell markers	Main biological functions	References
Intestinal Treg cells	Helios	Focus more on tissue repair	42,465
	RORγ (Helios− and Neuropilin-1−)	Focus more on modulating the immune response (particularly by suppressing Th1/Th17 cells), which plays a key protective role in intestinal inflammation
Tumor Treg cells	Blimp-1	Participate in tumor immune escape	466
Brain-resident Treg cells	CD69, PD-1, KLRG1, CD103, Neuropeptide Y, Osteopontin	Suppress neuroinflammation	41
VAT Treg cells	CCR1, CCR2, CCR9, CCL6, ITGAV, MIP-2, Cxcl10, Ppar-γ	Anti-inflammatory and maintains metabolic homeostasis	47,48
Skin Treg cells	Skin-specific chemokine receptors (including CCR2, CCR6, CCR8, CCR10, CXCR4, CXCR6), RORα, Gata3	Inhibit skin inflammation, maintain homeostasis, and promote repair of damaged skin tissues	467–469
Treg cells in skeletal muscle	CD103, KLRG1, TIM-3	Promote tissue repair and regeneration	41,467

ROR retinoic acid-related orphan receptor, T-bet T-box expressed in T cell, IRF4 interferon regulatory factor 4, Gata3 GATA binding protein 3, Blimp-1 B lymphocyte induced maturation protein 1, PD-1 programmed death 1, KLRG1 killer cell lectin-like receptor G1, CCR C-C motif chemokine receptor, CCL6 C-C motif chemokine ligand 6, ITGAV integrin αV subunit, MIP-2 macrophage inflammatory protein 2, Cxcl10 C-X-C motif chemokine ligand 10, Ppar-γ peroxisome proliferator-activated receptor γ, CXCR C-X-C chemokine receptor, TIM-3 T cell immunoglobulin domain and mucin domain-3

Studies have shown that there are two main subsets of Treg cells in intestinal tissue: Helios⁺ Treg cells and RORγ⁺ Treg cells (Helios⁻ and Neuropilin-1⁻).⁴² Among these, the transcription factor Helios, a member of the IKAROS family of zinc finger proteins, is expressed in a large subset of Foxp3⁺ Treg cells.⁴³ Neuropilin-1 is a membrane protein that is highly and widely expressed in Foxp3⁺ Treg cells in both mice and humans.⁴⁴ Retinoic acid-related orphan receptor γ (RORγ) can be divided into RORγ1 and RORγ2 (RORγt). RORγt is the major transcription factor of T helper cell 17 (Th17). GATA-3 (widely recognized as the “master regulator” of Th2 cell differentiation) positive Treg cells are a subset of Helios⁺ Treg cells in the gut. In the gut, the immune regulatory function of Gata3⁺ Helios⁺ Treg cells is strictly dependent on Foxp3 and is not regulated by gut microbiota, whereas RORγ⁺ Treg cells are regulated by the microbiota and their function is only partially dependent on Foxp3.⁴² In addition to the above cellular markers, Treg cells in the intestine have been observed to express T-box expressed in T cells (T-bet, a key transcription factor for Th1 cells)⁴⁵ and interferon regulatory factor 4 (IRF4).⁴⁶ These factors have been demonstrated to play a pivotal role in the inhibition of the Th1/Th2 response, thereby maintaining intestinal tolerance to microbial and food antigens.

In the context of visceral adipose tissue (VAT), tissue-resident Treg cells exhibit a distinctive chemokine/chemokine receptor profile, with chemokine receptors chemokine (C-C motif) receptor 1 (CCR1), CCR2, CCR9, chemokine C-C motif chemokine ligand 6 (CCL6), integrin αV subunit (ITGAV), macrophage inflammatory protein 2 (MIP-2), and C-X-C motif chemokine ligand 10 (Cxcl10) highly expressed and the expression of CCL5 and CXCR3 decreased.⁴⁷ In addition, peroxisome proliferator-activated receptor γ (Ppar-γ) is highly expressed on adipose tissue Treg cells,⁴⁸ and it is a major factor in the accumulation of Treg cell numbers in adipose tissue.⁴⁹ It is widely accepted that the majority of Treg cells in lymph nodes (LNs) are circulating, yet 10–20% of Treg cells are believed to resemble resident cells, remaining in their respective LNs for extended periods ranging from weeks to months. The majority of these Treg cells exhibit the characteristic human tissue-resident memory T cell (TRM) phenotype.⁴⁰ Interestingly, a small proportion of Treg cells in humans were found to be able to secrete proinflammatory cytokines such as IL-17 and IFN-γ,^50,51 but when co-cultured with Tconvs, these Treg cells were still able to function as immunosuppressive cells in vitro and exhibited classic Treg cell features such as demethylation of the Foxp3 gene.^51,52 CD161 has been proposed as a marker for these Foxp3⁺ Treg cells.⁵¹ CD161⁺ Treg cells are highly suppressive⁵³ and can significantly inhibit the function of CD8⁺ T cells.⁵⁴ Therefore, CD161⁺ Treg cells may play an important role in maintaining the balance between tolerance and immunity, and their exact function in the human immune system and specific mechanisms remain to be investigated.

However, the phenotype and function of Treg cells in these different tissues are not unchanged; rather, they are profoundly shaped by the local microenvironment. Under different disease conditions, Treg cells may undergo significant phenotypic changes. This plasticity may serve as a mechanism for maintaining tissue homeostasis or act as a driving force in the onset and progression of disease. For example, in the skin of psoriasis patients, tissue-resident Treg cells exhibit high expression of spermidine/spermine N1-acetyltransferase (SSAT, encoded by the SAT1 gene). SAT1hi Treg cells exhibit impaired suppressive function and acquire a Th17-like phenotype, expressing high levels of IL-17A, CXCR6, and PD-1, while still maintaining Foxp3 expression, presenting a “fragile” rather than a fully transdifferentiated state.⁵⁵

Although different Treg cell subpopulations exhibit a great deal of phenotypic heterogeneity, it is generally accepted that CD4⁺ CD25⁺ Foxp3⁺ is the core phenotype of Treg cells.⁵⁶ However, a small amount of transient Foxp3 expression can also be present in CD4⁺ Tconv cells,^57–59 and high levels of Foxp3 expression can be induced in the presence of TGF-β,⁶⁰ especially among humans. A single-cell study analyzing Foxp3 defects in humans and mice showed that in the case of loss of function of the Foxp3 gene mutation, other features of Treg cell genes can still be expressed in the absence of functional Foxp3.⁶¹ In both humans and mice, T regulatory type 1 (Tr1) cells are a subset of Treg cells that do not constitutively express Foxp3.⁶² They are characterized by the expression of the surface markers CD49b and LAG-3.⁶³ Upon activation, Tr1 cells secrete IL-10 in a Foxp3-independent manner,⁶⁴ thereby regulating immune responses. They play a crucial role in suppressing autoimmune diseases⁶⁵ and tumor immune escape.⁶⁶ This suggests the presence of regulatory networks that are not dependent on Foxp3. Most crucially, Foxp3 expression levels do not correlate with inhibitory function. Foxp3⁺ T cells may lack immunosuppressive functions even when Foxp3 expression is high and stable.⁶⁰ CD25 is expressed on antigen-activated effector T cells,⁶⁷ and CD25⁻ T cells can also be Foxp3⁺ Treg cells with immunosuppressive functions.^68,69 These results suggest that higher levels of both Foxp3 and CD25 expression cannot be used as a criterion to accurately determine whether a T cell is a Treg cell or not.⁷⁰ In recent years, studies reporting the immunoregulatory role of certain CD8⁺ T cells in various diseases have demonstrated the existence of the CD8⁺ Treg cell subset,⁷⁰ which we describe in detail below. Interleukin 7 receptor subunit alpha (CD127) is one of the biomarkers of Treg cells,⁷¹ and its low expression on Treg cells can help us to distinguish Treg cells from Tconvs to some extent.^72–74 Helios is often associated with a more stable phenotype and stronger immunosuppressive function.^75–77 In mice, Helios is generally recognized as a specific marker for tTreg cells.²⁴ Studies have demonstrated that during intense inflammatory responses, the Helios protein maintains Treg cell survival and stability by activating the IL-2Rα-STAT5 pathway.⁷⁸ In addition, CTLA-4 (CD152),^8,79–81 inducible synergistic co-stimulation molecules (ICOS, CD278),^82–84 CD39,^85,86 tumor necrosis factor receptor type 2 (TNFR2, CD120b),⁸⁷ LAG-3^9,88 and recombinant glycoprotein A33 (GPA33)⁸⁹ are also widely recognized as reliable surface markers for Treg cells. However, many markers are not specific for Treg cells, such as TNFR2, LAG-3, Helios, and GPA33, which are also expressed on effector T (Teff) cells.^90–92 It is difficult to accurately define Treg cells using either a single biomarker or a combination of markers. Therefore, when characterizing Treg cells, the phenotype of the cell should be considered in conjunction with its immunosuppressive function, which is the best criterion for defining Treg cells.

CD8⁺ regulatory T cells

CD4⁺ Treg cells have been extensively studied, but knowledge about CD8⁺ Treg cells remains limited. As early as 1978, Harvey Cantor et al. discovered a CD8⁺ T cell subset capable of suppressing CD4⁺ T cells to inhibit immune responses.⁹³ Jiang et al. observed in EAE mouse models that depletion of CD8⁺ T cells in mice recovering from primary EAE infection significantly reduced their resistance to secondary EAE induction.⁹⁴ This phenomenon further demonstrated that specific CD8⁺ T cell subsets participate in regulating autoreactivity. Subsequent studies revealed that these CD8⁺ T cells in humans and mice can suppress pathogenic T cells in both autoimmune diseases and infectious diseases.⁹⁵ These immunoregulatory CD8⁺ T cells are CD8⁺ Treg cells. CD8⁺ Treg cells exhibit heterogeneity in phenotype, generation, and homeostatic effects, with different CD8⁺ Treg phenotypes potentially sharing specific molecules or signals associated with their generation and function.⁷⁰ Depending on whether their immunosuppressive function depends on Foxp3, CD8⁺ Treg cells can be classified into Foxp3⁺ CD8⁺ Treg cells and Foxp3^- CD8⁺ Treg cells. Similar to Foxp3⁺ CD4⁺ Treg cells, Foxp3⁺ CD8⁺ Treg cells also acquire their immunosuppressive function through Foxp3-dependent mechanisms. Under steady-state conditions, in untreated mice and healthy humans, Foxp3⁺ CD8⁺ Treg cells are typically extremely rare, accounting for 0.1% and 0.4% of peripheral blood T cells, respectively.⁹⁶ Studies have shown that Foxp3⁺ CD8⁺ Treg cells share many immunosuppressive mechanisms with Foxp3⁺ CD4⁺ Treg cells, including directly interacting with DCs, suppressing the expression of co-stimulatory molecules such as CD80/86 on DCs,⁹⁷ and secreting anti-inflammatory cytokines such as TGF-β and IL-10.⁹⁸ Interestingly, in addition to their immunosuppressive function, Foxp3⁺ CD8⁺ Treg cells still possess cytotoxic activity similar to that of CD8 cytotoxic T cells, which is mediated primarily by granzyme B and FasL.⁷⁰ However, it remains unclear whether Foxp3⁺ CD8⁺ Treg cells exhibit conserved cytotoxicity in all biological contexts.⁷⁰ Ly49⁺ CD8⁺ Treg cells are among the most typical non-Foxp3 CD8⁺ Treg cells. Ly49⁺ CD8⁺ Treg cells express a set of inhibitory Ly49 receptors (Ly49A, Ly49G2, Ly49C/I, and Ly49F) but do not express the activating Ly49 receptors (Ly49D and Ly49H).^70,99 Ly49⁺ CD8⁺ Treg cells can recognize CD4⁺ T cells expressing the Qa-1/peptide complex via TCR.^100,101 Subsequently, they eliminate self-antigen-specific CD4⁺ T cells through perforin-mediated cytotoxic effects.¹⁰² For Ly49⁺ CD8⁺ Treg cells, studies have revealed that the Helios transcription factor enhances cellular responsiveness to IL-2 by binding to and activating the STAT5b/IL-2Rα pathway. This mechanism is crucial for establishing and maintaining their phenotype and immunoregulatory activity.⁷⁸ Helios deficiency results in the loss of Ly49⁺ CD8⁺ Treg cell immunosuppressive function. In human tonsil-derived immune organoids, Chen et al. investigated the regulatory roles of Foxp3-expressing CD4⁺ Treg cells and killer cell immunoglobulin-like receptor-expressing CD8⁺ Treg cells (corresponding to Ly49-expressing CD8⁺ Treg cells in mice^95,103) in autoimmune responses. CD4⁺ Treg cells primarily function as inducers of humoral immune tolerance by controlling autoreactive B cells to prevent autoantibody accumulation. In contrast, CD8⁺ Treg cells primarily act as maintainers of cellular immune tolerance by suppressing the expansion and activation of autoreactive T cells through granzyme B-mediated toxicity.¹⁰⁴ This suggests that, in maintaining immune self-tolerance, CD4⁺ and CD8⁺ Treg cells perform distinct and potentially complementary functions.

Although the role of other Treg cell subsets, represented by CD8⁺ Treg cells, in immune tolerance has garnered increasing attention, significant controversy persists regarding the stable surface markers, differentiation pathways, and precise functions of CD8⁺ Treg cells in vivo. The translation of these findings into clinical practice remains in the preliminary exploratory phase. CD4⁺ Treg cells are the most extensively studied Treg population, with the clearest molecular mechanisms and functional characteristics. Therefore, in this review, we focus primarily on CD4⁺ Treg cells, aiming to systematically summarize the current progress and integration logic of their induction strategies. We hope to provide a more instructive theoretical framework for clinical translation in this field.

Strategies for Treg cell induction and integration approaches

Conventional immunosuppressive therapies (e.g., the utilization of immunosuppressive drugs such as glucocorticoids¹⁰⁵) have been demonstrated to effectively suppress the body’s inflammatory response. However, the non-specific nature of these therapeutic strategies may contribute to the occurrence of adverse effects, including an increased risk of infection¹⁰⁶ and metabolic disorders.¹⁰⁷ In contrast, cellular therapeutic strategies that induce Treg cells provide a more specific therapeutic approach for restoring immune homeostasis by directly modulating cell populations with immunomodulatory functions. tTreg cells develop progressively in the thymus. In addition, Treg cells can be induced through different biological pathways. In this section, we first discuss the differentiation of tTreg cells, because the mechanisms underlying their development provide a crucial theoretical foundation for understanding the principles of Treg cell induction. Then we summarize the current common strategies for Treg cell induction and integrated approaches combining these strategies. Currently, Treg cell induction strategies focus on antigenic stimulation, cytokine modulation, metabolic pathways, reagent and drug induction, epigenetic regulation, and engineered technology (chimeric antigen receptor). The overarching objective of these strategies is to augment the functional Treg cell population through a multifaceted array of pathways, as opposed to a uniform suppression of the immune system. However, the clinical translation of these strategies faces a series of interrelated fundamental challenges, including the persistence and maintenance of the suppressive function of induced Treg cells in vivo, targeting specificity, and difficulties in large-scale production. A single induction strategy often struggles to comprehensively overcome these obstacles. Therefore, integrating multiple induction strategies may enable the design of Treg cell therapies with superior functionality and broader applicability. (Fig. 4)

Fig. 4.

Treg cell induction strategies. Different induction strategies address distinct key challenges in the clinical translation of Treg cell therapies. Only through the efficient integration of these strategies can precisely targeted, stably maintained, and safely produced Treg cells be achieved. The arrows in the figure show which category of challenges each induction strategy focuses on resolving. HDAC histone deacetylase, HAT histone acetyltransferase, PI3K phosphatidylinositol 3-kinase, Akt protein kinase B‌, TSC tuberous sclerosis complex, AMPK adenosine 5′-monophosphate (AMP)-activated protein kinase, Rheb ras homolog enriched in brain, mTORC mammalian target of rapamycin. The graphic is created with BioRender.com

Antigen stimulation and chimeric antigen receptor (CAR) technology confer targeting capabilities to Treg cells, thereby reducing the risks of infection and tumor development associated with non-specific immunosuppression. Cytokine regulation, epigenetic modulation, and metabolic reprogramming enhance Treg cell stability and functional persistence. Additionally, CAR technology facilitates the large-scale generation of Treg cells. Mikami et al. successfully induced effector T cells, including both naïve Tconv cells and differentiated Th1/2/17 cells, to generate functionally stable antigen-specific Treg cells in vitro through the efficient integration of multiple induction approaches.¹⁰⁸ It was demonstrated that in the presence of antigen stimulation (simulated by binding anti-CD3 antibody to CD3), IL-2 and TGF-β cytokines, and the enhancement of STAT5 signaling pathways with CDK8/19 inhibitors (Senexin A), Foxp3 expression was efficiently induced, and effector T cell differentiation was suppressed. Concurrently, the deprivation of CD28 co-stimulatory signals during the induction phase promoted DNA demethylation in the Foxp3 CNS2 region, thereby establishing a stable epigenetic landscape. After the initial stimulation phase, researchers introduced a resting culture phase to enable the cells to recover from the intense activation process. During this period, the cells were only given IL-2 in order to consolidate epigenetic modifications and prevent activation-induced cell death. After a four-day resting culture, the second stimulation phase began. In this phase, researchers repeated the stimulation components from the first phase and introduced an anti-FasL antibody into the culture system to prevent activation-induced cell death. This induction protocol demonstrated exceptionally high and stable efficiency in mice, producing iTreg cells with stable Foxp3 expression and potent regulatory capabilities in vivo and in vitro. However, this protocol exhibited low (80%) and unstable efficiency in human Tconv cells. Therefore, they optimized this induction strategy by targeting metabolic pathways or using drugs. They simultaneously employed retinoic acid (RA) to enhance conversion efficiency and Foxp3 expression and utilized vitamin C, a cofactor for TET demethylases, to promote demethylation of the Foxp3 CNS2 region. The optimized protocol also demonstrated excellent iTreg cell induction effects in human Tconv cells. Similarly, in humans and mouse models of pemphigus vulgaris (a skin autoimmune disorder mediated by anti-desmoglein 3 (Dsg3) autoantibodies), researchers used CDK8/19 inhibitors, IL-2, TGF-β, all-trans retinoic acid, and ascorbic acid (a TET demethylase cofactor) to convert Dsg3-specific, pathogenic, autoreactive CD4⁺ T cells into stable, antigen-specific Treg cells in vitro, thus achieving a disease-specific treatment.¹⁰⁹

Differentiation of thymic Treg cells

In the thymus, some thymocytes undergo a series of signaling stimuli, which results in their differentiation and subsequent maturation into mature tTreg cells. This process can be subdivided into two distinct phases (Fig. 5).¹¹⁰ The first phase is TCR-dependent, and in this phase, some thymocytes expressing TCRs with a moderate affinity for self-antigenic peptides can recognize APC-presented pMHC-II and differentiate into CD4⁺ CD25⁺ Foxp3⁻ Treg cells. The second phase is TCR-independent, and in this phase, Treg cell precursors express Foxp3 in response to cytokines (mainly IL-2). Overall, the development of tTreg cells is contingent on three signals: IL-2 signaling, TCR-specific recognition of pMHC-II, and CD80/CD86 co-stimulatory molecules.¹

Fig. 5.

The process of differentiation of tTreg cells in the thymus. The maturation of thymocytes into mature tTreg cells in the thymus is characterized by two distinct phases: a TCR-dependent phase and a TCR-independent phase. In the TCR-dependent phase, antigen-presenting cells (APCs), such as mTECs, CD8α⁺ DCs, and SIRPα⁺ DCs in the thymus, present autoantigens to provide TCR signals to thymocytes. Concurrently, they provide CD80/CD86 co-stimulatory signals, which induce thymocytes to differentiate into Foxp3- Treg cell precursors. During the TCR-independent phase, the secretion of interleukin 2 (IL-2) by Tconvs and of interleukin-15 (IL-15) by B cells and monocytes/macrophages can promote the development of Treg cell precursors into mature Treg cells, which possess immunosuppressive functions. Aire autoimmune regulator, mTEC medullary thymic epithelial cell, TCR T cell receptor, SIRPα signal-regulatory protein α, Tconv conventional T cell, PLCγ1 phospholipase Cγ1, DAG diacylglycerol, IP₃ inositol triphosphate, CaN calcineurin, NFAT nuclear factors of activated T cells, PKCθ protein kinase C-θ, CARMA1 coactivator-associated arginine methyltransferase 1, Bcl-10 B cell leukemia/lymphoma 10, MALT1 mucosa-associated lymphoid tissue lymphoma translocation protein 1, IKK inhibitor of kappa B kinase, IκB inhibitor of NF-κB, NF-κB nuclear factor kappa B. The graphic is created with BioRender.com

Previous studies have demonstrated that the thymic tolerance mechanism that the organism relies on tTreg cells to establish is driven by TCR signaling.¹¹¹ In the thymus, thymocytes that have undergone TCR rearrangements are subjected to a series of positive selection and negative selection processes in the thymic cortex and medulla, depending on their TCR reactivity. Positive selection enables immature CD4⁺ CD8⁺ double-positive (DP) T cells to acquire MHC restriction and differentiate into single-positive (SP) T cells. In the subsequent negative selection process, the majority of SP T cells (autoreactive T cells) capable of recognizing high-affinity self-peptide-MHC complexes undergo apoptosis. A small proportion of SP T cells have TCRs with a moderately increased affinity for certain self-antigens. These cells are known as thymic Treg cell precursors. When the TCRs of these precursor cells receive signals within a specific intensity range, a unique series of intracellular events is triggered, resulting in their differentiation into tTreg cells. Autoimmune regulator (Aire) is expressed in medullary thymic epithelial cells (mTECs), which induces the expression of tissue-specific antigens in mTECs and promotes tTreg cell differentiation by presenting them to developing Treg cells.¹¹² Furthermore, CD8α⁺ DCs and bone marrow-derived DCs with high expression of signal-regulatory protein alpha (SIRPa) have been shown to play a significant role in presenting self-antigens, thereby promoting the development of regulatory T (Treg) cells.^113,114 The pMHC complexes presented by these APCs bind to TCRs on thymocytes and promote hydrolysis of PtdIns(4,5)P₂ by phospholipase Cγ1 (PLCγ1) to produce diacylglycerol (DAG) and inositol triphosphate (IP₃).¹¹⁵ DAG activates protein kinase C-θ (PKCθ), which can activate a protein complex consisting of coactivator-associated arginine methyltransferase 1 (CARMA1), B cell leukemia/lymphoma 10 (Bcl-10), and mucosa-associated lymphoid tissue lymphoma translocation protein 1 (MALT1). This complex activates inhibitor of kappa B kinase (IKK), induces the nuclear translocation of NF-κB family transcription factors, and then promotes the expression of Foxp3 in tTreg cells.^116,117 IP3 has been shown to induce calcium ion release from the endoplasmic reticulum, and calcium signaling can promote the dephosphorylation of nuclear factors of activated T cells (NFAT) and induce its nuclear translocation through the activation of calcineurin. NFAT has been demonstrated to bind to the Foxp3 enhancer region, thereby maintaining enhancer activity, and to synergize with other transcription factors, such as Smad3, to induce Foxp3 expression.¹¹⁸ The potency of TCR signaling can exert a significant influence on the in vivo fate of CD4⁺ T cells in the thymus, including their selection for differentiation into tTreg cells.¹¹⁹ A study based on a mouse model found that pathogenic autoimmune Tconv cells were generated when the intensity of TCR signaling was reduced to a certain level and affected the production of Treg cells in the thymus, resulting in the development of autoimmune diseases or inflammation.¹²⁰ In addition, transient and discontinuous TCR stimulation was required to induce stable Foxp3 expression. Dimethylation and trimethylation of H3K4 near the transcription start site (TSS) and 5′ non-coding region (UTR) of the Foxp3 gene are closely associated with Foxp3 induction potential. However, these modifications are lost under continuous TCR signaling stimulation, thereby inhibiting Foxp3 expression.¹²¹ Thus, the duration of TCR signaling stimulation affects the probability of Foxp3 induction and is also involved in the regulation of Treg cell fate in vivo.

CD80/CD86 are expressed by APCs such as mTECs and DCs in the thymus,¹ and the co-stimulatory signaling they mediate plays an integral role in tTreg cell generation. B7-CD28 co-stimulatory signaling is required for the induction of endogenous tissue-restricted antigen (TRA)-specific tTreg cell generation.¹²² CD28 binds to CD80/86, and the C-terminal proline motif in the tail of the cytoplasmic segment of CD28 can bind to the lymphocyte-specific protein tyrosine kinase (Lck), which activates NF-κB and promotes tTreg cytogenesis.¹²³ The absence of B7-CD28 co-stimulatory signaling has been demonstrated to result in the accumulation of functional autoreactive CD4⁺ T cells in the periphery, a process that has been shown to cause autoimmune diseases.¹²²

In humans and mice, the predominant production of IL-2 in the thymus is attributable to mature αβ and γδ thymocytes.¹²⁴ It was found that IL-2Rβ knockout mice were unable to produce tTreg cells.¹²⁵ However, thymus-specific transgenic expression of IL-2Rβ in mice was able to re-establish the normal development of tTreg cells and effectively prevent lethal autoimmunity.¹²⁵ This suggests that IL-2 is essential in tTreg cell development. Furthermore, in the phase of human and murine tTreg generation, the T cell co-stimulatory molecule tumor necrosis factor receptor superfamily member 4 (TNFRSF4, also known as OX40) is able to induce tTreg cell maturation and proliferation through the protein kinase B‌ (AKT)-mammalian target of rapamycin (mTOR) signaling pathway in an IL-2-dependent manner.¹²⁶ During tTreg cell development, IL-15 can also play a similar role to IL-2 by inducing the differentiation of tTreg precursors into Foxp3⁺ T cells and promoting tTreg stabilization and proliferation.^1,124 However, it is worth noting that the roles of IL-2 and IL-15 in the development of tTreg cells are non-redundant.¹²⁷ In the absence of IL-2 or IL-15 signaling, the development of Treg cells will be partially impaired.¹²⁸

Mature tTreg cells express specific homing receptors and chemokine receptors, such as CCR4¹²⁹ and CCR7, enabling them to respond to peripheral signals, enter the bloodstream and secondary lymphoid organs, perform immune regulatory functions, and maintain peripheral immune tolerance. Studies have revealed that Treg cells in the periphery circulate back to the thymus via a chemokine-dependent mechanism. This forms a negative feedback loop within the thymus that suppresses the development of new Treg cells.¹³⁰ Thiault et al. used a Rag-GFP transgenic mouse model, in which GFP labels newly developed cells in the thymus that gradually fade as the cells leave the organ, to discover that the proportion of newly developed GFP⁺ Treg cells in the thymus decreased progressively with increasing age, while the proportion of mature Treg cells that lack GFP expression increased significantly (from 10% to approximately 90%). Furthermore, principal component analysis revealed that the GFP⁻ Treg cells in the thymus were more similar to the GFP^- Treg cells in the spleen than to the newly developed GFP⁺ Treg cells within the thymus itself. This suggests that the growing population of GFP⁻ Treg cells in the thymus does not consist of cells that have never left the organ and remain long-term residents, but rather comprises Treg cells that have undergone peripheral circulation and then returned to the thymus. In humans, peripheral Treg cells can also recirculate to the thymus, exhibiting an activated and differentiated phenotype (high expression of the co-stimulatory molecule ICOS and the transcription factor T-bet).¹³⁰ These recirculating cells typically express CCR6¹³¹ and CXCR3,¹³⁰ which play a crucial role in Treg cell recirculation back to the thymus, but do not express CCR7.¹³¹ CCR7 plays a crucial role in the migration of tTreg cells from the cortex to the medulla, but it is not essential for Treg cell production or migration. Conversely, the absence of CCR7 enhances the thymus’s receptivity to Treg cell thymic homing.¹³¹ CCR7 restricts the peripheral recirculation of Treg cells back to the thymus, thereby maintaining the equilibrium of the thymic Treg cell pool. Under physiological conditions, recirculating Treg cells suppress the development of newly generated Treg cells within the thymus by competitively consuming IL-2, thereby forming a negative feedback loop.¹³⁰ Under inflammatory conditions, these recirculating Treg cells maintain the stability of the thymic immune microenvironment, thereby protecting the development of newly generated tTreg cells. These recirculating Treg cells, which exert protective effects under inflammatory conditions, express IL-1R2.¹³² IL-1R2 is a decoy receptor for the inflammatory cytokine IL-1.¹³³ It can bind to IL-1β without triggering subsequent signal transduction and exert an anti-inflammatory effect. Under inflammatory conditions, elevated levels of IL-1β in the thymus activate mTECs and t-DCs. This leads to the production of inflammatory cytokines, such as interferon-γ (IFN-γ), IL-1 and IL-6, which suppress tTreg cell development.^132,134,135 However, recirculating IL-1R2⁺ Treg cells entering the thymus bind to and neutralize IL-1β, thereby maintaining stability in Treg cell development. Additionally, when the thymus is damaged, tTreg cells are actively and extensively recruited to the injured thymus. These Treg cells activate signaling pathways such as ERK by secreting amphiregulin (AREG), which acts on epidermal growth factor receptors (EGFR) expressed on thymic epithelial cells and fibroblasts. This promotes the proliferation of stromal cells and restores the thymic cellular development environment, thereby driving thymic regeneration.¹³⁶ Further research has revealed that there is a similar population of CD39⁺ ICOS⁺ tTreg cells in humans, which also have the potential to promote thymic regeneration.¹³⁶

Induction by antigenic stimulation

TCR signaling plays a key role in the induction and functional realization of Treg cells.¹³⁷ The differentiation of tTreg cells is contingent on TCR recognition of its pMHC complex, whereas the differentiation of peripheral pTreg cells is dependent on responses to non-self-antigens, such as food, other allergens in the environment, and commensal microbes.^119,138,139 Furthermore, antigen-specific iTreg cells can be obtained by co-culturing known antigens, naïve T cells, and DCs in the presence of TGF-β. In comparison with extensive immunosuppression, antigen-specific Treg cell-mediated immune tolerance is targeted and has fewer side effects. The suppression of self-reactive cells without the occurrence of systemic immunosuppression is considered to be an ideal treatment for autoimmune diseases. This objective may be achieved by the induction of antigen-specific Treg cells. A study was conducted to explore the potential for treatments to be found for type 1 diabetes (T1D), in which researchers prepared a nanoparticle (NP) co-delivering the aryl hydrocarbon receptor (AhR) ligand 2-(1′H-indole-3′-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE) and the β-cell antigen proinsulin.¹⁴⁰ The NP was administered to 8-week-old non-obese diabetic (NOD) mice, which resulted in the induction of a tolerogenic DC phenotype and the promotion of Treg cell production in vivo. Similarly, Christina Krienke et al. developed a liposomal formulation (mRNA-LPX) loaded with mRNAs encoding experimental autoimmune encephalomyelitis (EAE)-associated autoantigens and substituted 1-methylpseudouridine (m1Ψ) for uridine (U) to eliminate the inflammatory properties of the single-stranded mRNAs.¹⁴¹ This preparation was able to deliver m1Ψ mRNA to splenic CD11c APCs, inducing the generation of antigen-specific Treg cells. These results suggest that nano-delivery technology has great potential for inducing antigen-specific Treg cell generation. However, in the case of many autoimmune diseases, the target antigens remain elusive and may differ from patient to patient, which is one of the challenges that needs to be overcome to utilize antigen-specific Treg cells for the treatment of related diseases.

The conventional treatment of food allergies is predicated on the principle of strict allergen avoidance or the administration of antiallergic medications such as epinephrine and antihistamines in the event of accidental exposure. These treatment strategies are not curative and can result in a limited quality of life for patients.¹⁴² In contrast, the utilization of oral immunotherapy (OIT) to induce desensitization or achieve oral tolerance has the potential to be a safer and more efficacious treatment modality. Oral tolerance is employed to denote the ingestion of a known antigen by the oral route, with the objective of inducing immune tolerance to that antigen. During OIT, treatment typically begins with a very low dose (far below the levels encountered in the daily diet), which is then gradually increased. Low doses may preferentially activate Treg cells rather than Tconv cells.¹⁴³ This method has been proven to be an effective means of controlling food allergies. In contrast to thymus-derived tTreg cells, intestinal Treg cells possess a peripheral TCR pool that is responsive to the commensal microbiota, as well as to food antigens.¹⁴⁴ The vast majority of small intestinal Treg cells are induced by food antigens, and these pTreg cells are effective in suppressing potentially strong immune responses to ingested protein antigens in the organism.¹⁴⁵ A study of peanut oral immunotherapy (OIT) in 43 peanut-allergic patients found that oral administration of peanut induced the production of peanut-specific Treg cells (ai-Treg) and increased the transcriptional level of Foxp3 in ai-Treg.¹⁴⁶ In comparison with conventional anti-allergy medications, OIT has been shown to elicit antigen-specific immune tolerance through a gradual escalation of allergen dosage. This approach aims to attain sustained non-reactivity to allergens, representing a novel approach for radical therapies in the management of food allergies.¹⁴⁷

However, antigen stimulation alone is far from sufficient during Treg cell induction. TCR signaling merely initiates the primary program of T cell activation, while the functional maturation and phenotypic stability of Treg cells are highly dependent on the precise regulation of a series of co-stimulatory signals and the cytokine environment. Therefore, antigen stimulation strategies partially address the issue of targeting, but they often fail to resolve the critical challenge of achieving both efficacy and durability in treatment.

Cytokine modulation

During OIT-mediated immune tolerance, desensitized mast cells (MCs) promote Treg cell expansion via an IL-2-dependent pathway.¹⁴⁸ In humans, researchers found that allergen-specific Th2 cells were also capable of releasing IL-2 to induce sustained Treg cell production.¹⁴⁹ These experimental results suggest that the cytokine IL-2 is essential in the induction of Treg cells by antigenic stimulation. Furthermore, other cytokines, including TGF-β, IL-10, and IL-35, also play pivotal roles in the induction of Treg cells.

Interleukin 2 (IL-2)

The critical role of IL-2 in tTreg cell induction, as well as the finding that neutralizing IL-2 inhibits the proliferation of pTreg cells,¹⁵⁰ implies that IL-2 is crucial in inducing the generation of pTreg cells and iTreg cells. The high-affinity IL-2 receptor (IL-2R) is a trimer composed of three subunits, CD25 (IL-2Rα), CD122 (IL-2Rβ), and CD132 (γc).¹⁵¹ The binding of IL-2 and IL-2R has been demonstrated to promote the binding of IL-2Rβ to Janus kinase 1 (JAK1) and the binding of γc to JAK3, activating JAK1/3,¹⁵² which in turn induces the recruitment and phosphorylation of STAT5 in human T lymphocytes.¹⁵³ This is pivotal to the biological function of IL-2 during Treg cell development.¹⁵⁴ Recent studies have revealed that in the presence of IL-2, the anti-CD25 antibody PC61 is able to promote Gαi/o-dependent integrin activation through the Rap1-RIAM/Lpd-talin1 signaling pathway. This process facilitates the migration of Treg cells to the site of inflammation or injury and increases the secretion of the inhibitory cytokines IL-10 and TGF-β, thereby enhancing the immune-suppressive function of Treg cells (Fig. 6).¹⁵⁵ Notably, PC61 antibodies have traditionally been used as tools for depleting Treg cells. In vivo, intact PC61 antibodies (containing the Fc fragment) may bind to FcγRIII⁺ phagocytes (such as macrophages) via their Fc fragment, thereby mediating antibody-dependent cellular phagocytosis to effectively clear CD25-expressing Treg cells.¹⁵⁶ However, in the study by Sun et al., they used a PC61 mutant (mPC61) without Fc function, which effectively avoided Fc-mediated clearance. Meanwhile, the Fab fragment of mPC61 could still bind to CD25, thereby activating downstream signaling pathways and enhancing the immunosuppressive function of Treg cells.¹⁵⁵ Thus, these two observations are not contradictory; instead, they fully reflect the dual mechanism of action of the PC61 antibody. IL-2 is a cell growth factor that is critical for T cell proliferation, as well as for the production and mediation of immune responses by effector and memory cells.¹⁵⁷ The use of high doses of IL-2 in stimulating antitumor immune responses has been demonstrated.¹⁵⁸ In both human and mouse Treg cells, high expression of IL-2Rα significantly increases the sensitivity of IL-2 signaling,¹⁵⁹ which renders Treg cells more sensitive to low levels of IL-2.¹⁶⁰ In conditions of low-dose IL-2 stimulation, Treg cells are preferentially activated,¹⁶⁰ resulting in the competitive depletion of IL-2 and thus limiting Tconv development.¹⁷ Therefore, the use of low-dose IL-2 to induce Treg cells for the treatment of autoimmune diseases and transplant rejection has a good application prospect. More than a decade ago, it was reported that the administration of low-dose IL-2 could promote the survival of Treg cells and prevent the development of diabetes in mice.¹⁶¹ Currently, relevant therapeutic regimens have been applied to the treatment of various diseases and have entered clinical trials.¹²⁵

Fig. 6.

IL-2 induces Treg cell differentiation and promotes Treg cell immunosuppression. It is widely accepted that the binding of interleukin 2 (IL-2) to its receptor instigates three significant signaling pathways: the mitogen-activated protein kinase (MAPK) signaling pathway, phosphoinositide 3-kinase (PI3K)-protein kinase B (AKT) signaling pathway, and Janus kinase (JAK)-signal transducer and activator of transcription (STAT) signaling pathway.⁴⁹⁵ In particular, when IL-2 binds to interleukin 2 receptor α (IL-2Rα) (CD25), it promotes the activation of JAK1/3, which further promotes the phosphorylation of three key tyrosine residues in the tail of the cytoplasmic segment of IL-2Rβ,⁴⁹⁵ in which the phosphorylation of Y392 and Y510 (Y395 and Y498 in mice) induces STAT5 recruitment and tyrosine phosphorylation,⁴⁹⁶ followed by STAT5 dimerization and translocation to the nucleus to activate Foxp3 gene expression. Furthermore, under conditions of IL-2 binding to CD25, the anti-CD25 antibody PC61 can induce regulatory T cell (Treg cell) integrin activation via Gαi/o-mediated phospholipase C (PLC) activation, which promotes the immunosuppressive function of Treg cells. P phosphoric acid, mTOR mammalian target of rapamycin, Ras rat sarcoma, Raf rapidly accelerated fibrosarcoma, MEK mitogen-activated extracellular signal-regulated kinase, ERK extracellular regulated protein kinases, GPCRs G protein-coupled receptors, PKC protein kinase C, RAP1 ras-proximate-1, RIAM RAP1-interacting adapter molecule. The graphic is created with BioRender.com

However, the concept of “low dose” may be different for different diseases and individual patients. The half-life of IL-2 in human blood is notably short. Early studies found that following intravenous administration of recombinant IL-2 in humans, it was rapidly cleared from the plasma, with an initial rapid-phase half-life of 12.9 min, followed by a slower phase with a half-life of 85 min out to 4 h after the bolus.¹⁶² Following a single intravenous injection of recombinant IL-2 in mice, it also exhibited rapid clearance, with a terminal half-life of 16.8 min.¹⁶³ The dosage, frequency, and duration of IL-2 administration exhibited significant variability across diverse studies. The potential impact of specific treatment regimens on the phenotypes of Treg cells obtained from the induction of IL-2 remains a subject of further investigation. The effect of specific treatment regimens on the induced Treg cell phenotype also needs to be further investigated.

Off-target effects represent a significant hindrance to the efficacy of IL-2 therapies,¹⁶⁴ and in order to overcome this limitation, researchers have developed strategies such as anti-IL-2 antibodies^159,165,166 and IL-2 variants (e.g., IL-2 muteins¹⁶⁷) to improve the selectivity of Treg cells induced by IL-2. For instance, JES6-1 is a monoclonal antibody that binds to mouse IL-2. Upon binding to IL-2, JES6-1 alters its conformation,¹⁵⁹ thereby blocking its interaction with IL-2Rβ and IL-2Rγ while preserving binding to IL-2Rα. This mechanism reduces the risk of Tconv cell activation induced by IL-2 and enables the selective expansion of Treg cells.¹⁶⁸ Similarly, F5111 is a fully human anti-IL-2 antibody that promotes specific binding of IL-2 to CD25 high cells. Since Treg cells tend to express high levels of CD25, whereas naïve CD4⁺ and CD8⁺ T cells, as well as natural killer (NK) cells, do not express CD25, the IL-2/F5111 complex can selectively activate Treg cells.¹⁵⁹ To enhance the affinity of the antibody for IL-2, researchers optimized F5111 through affinity maturation to yield the high-affinity variant, F5111.2. The F5111.2/hIL-2 complex retains the ability to selectively activate Treg cells. Furthermore, under low-dose IL-2 conditions, F5111.2/hIL-2 demonstrates superior efficacy to F5111/hIL-2 in inhibiting STAT5 signaling in Teff cells.¹⁵⁹ However, in the IL-2/F5111.2 complex, IL-2 remains bound to F5111 in a non-covalent manner. This poses a risk of dissociation that could counteract the Treg cell-biased properties of F5111.2. Furthermore, using a single drug rather than a mixture of components facilitates the streamlining of the clinical approval process. Investigational new drug-enabling nonhuman primate studies of cytokine/antibody complexes require species-matched cytokines, but the antibody may not cross-react with nonhuman primate cytokines.¹⁶⁹ Therefore, to overcome the limitations of the IL-2/antibody complex, researchers further developed F5111 immunocytokine (F5111 IC), which was a single-chain human IL-2/F5111 immunocytokine factor. F5111 IC was a single-chain human IL-2/F5111 antibody fusion protein.¹⁶⁹ In comparison with the IL-2/F5111 complex,¹⁵⁹ the F5111 IC fusion protein demonstrated higher stability and exhibited stronger Treg cell-biased activity. Engineered IL-2 muteins preferentially stimulate Treg cell activation while reducing off-target effects. CD122 and CD132 are IL-2 receptors expressed on NK cells and Tconv cells. Thus, one approach to engineering IL-2 is to reduce its affinity for CD122 and CD132 in order to enhance its selectivity for Treg cells.^170,171 Khoryati et al. introduced two amino acid substitutions into murine IL-2 that reduce CD122 binding (N103R and V106D, equivalent to the N88 and V91 positions of human IL-2) to create a CD25-biased IL-2 mutein.¹⁶⁷ To extend the half-life of the IL-2 mutein in vivo, they fused the IL-2 muteins to an immunoglobulin G2a Fc fragment mutated to reduce FcR binding and effector function, thereby creating Fc.IL-2 muteins.¹⁷² They found that Fc.IL-2 mutein could enhance the expression of IL-10 and CTLA-4 in Treg cells, thereby suppressing dendritic cells and promoting T cell anergy. This approach exhibited superior therapeutic efficacy in NOD mouse models.¹⁷² Similarly, Efe et al. reduced the binding affinity of IL-2 to CD122/CD132 through the introduction of the H16L mutation, while maintaining its strong affinity for CD25. This enabled the selective expansion of Treg cells and demonstrated good immunomodulatory effects and transplant protection in multiple preclinical transplantation models.¹⁷³

Transforming growth factor-β

In addition to IL-2, TGF-β is another key cytokine in Treg cell-induced development. There are three highly homologous isoforms of TGF-β, i.e., TGF-β1, TGF-β2, and TGF-β3. TGF-β1 is essential in regulating the immune function of the body and is preferentially expressed in immune cells (Fig. 7).^174,175 The number of pTreg cells in TGF-β1 expression-deficient mice is significantly lower than that in normal mice, while the expression of Foxp3 is also reduced.¹⁷⁶ Additionally, a study demonstrated that the knockout of the endogenous transforming growth factor beta receptor 2 (TGFBR2) in chimeric antigen receptor (CAR) T cells using CRISPR/Cas9 technology led to a significant reduction in the induction of Treg cell production.¹⁷⁷ These results suggest that TGF-β is deeply involved in pTreg cell development. TGF-β has been shown to induce Treg cell maturation through a signaling pathway that is mediated by the activation of Smad transcription factor phosphorylation,^39,178 which has already been discussed in the previous sections. Interestingly, TGF-β-mediated Foxp3-CNS1 is essential for the generation of iTreg cells, but is not indispensable for the development of tTreg cells in the thymus. In a study of mice with defective TGF-β1 expression, it was found that tTreg cells in the thymus of mice developed normally,¹⁷⁶ presumably due to the ability of CNS3 to compensate for the function of CNS1 in tTreg cells.¹⁷⁹ In the thymus, the primary role of TGF-β is to confer a survival advantage on tTreg cell precursors,^180,181 rather than to promote Foxp3 expression through Smad binding to Foxp3-CNS1.¹¹⁹ Furthermore, the Arkadia/ Sloan-Kettering Institute (SKI)/ Ski novel (SnoN) signaling pathway has been demonstrated to play a pivotal role in the induction of Foxp3 expression by TGF-β.¹⁷⁵ Recent studies revealed that TGF-β1 can induce mitochondrial fusion. Specifically, TGF-β1 activates Smad2 and Smad3, promotes peroxisome proliferator-activated receptor γ coactivator 1-α (PGC-1α)-mediated mitochondrial fusion, and then inhibits the expression of hypoxia-inducible factor 1α (HIF-1α), which promotes metabolic reprogramming from glycolysis to fatty acid oxidation, inducing Treg cell generation.¹⁸² Ivan Koprivica et al. designed particles loaded with all-trans retinoic acid (ATRA) as well as TGF-β (TGF-β MPs), which were able to efficiently induce the proliferation of Treg cells in the pancreatic draining lymph nodes (PLN) when administered orally.¹⁸³ TGF-β stimulation alone is often insufficient to drive the development of human Treg cells.⁶⁰ Early studies revealed that, in the presence of TGF-β, retinoic acid (RA) (the active metabolite of vitamin A) can effectively induce the generation of functionally suppressive iTreg cells¹⁸⁴ and inhibit the production of IL-6-mediated proinflammatory Th17 cells.¹⁸⁵ Studies using model antigens to immunize mice for EAE treatment found that RA acted as an adjuvant to induce antigen-specific Tr1 cells, with the addition of IL-2 significantly enhancing this effect.⁶⁴ These studies have provided some ideas for the use of TGF-β to induce Treg cell proliferation to maintain immune homeostasis. Within the TGF-β superfamily, TGF-β1 is the predominant cytokine that fosters the growth and development of Treg cells. Furthermore, another subfamily member, activin A, has been shown to activate the Smad signaling pathway through a synergistic interaction with TGF-β1, which induces transcription of the Foxp3 gene and facilitates the development and maturation of Treg cells.^175,186 However, the mechanism of the synergy between activin A and TGF-β1 remains to be elucidated. Further research is required to ascertain whether other members of the TGF-β superfamily contribute to the development of Treg cells.

Fig. 7.

TGF-β signaling plays a key role in the differentiation of Treg cells. a TGF-β binds to TGF-βR, which activates the CNS1 region in the Foxp3 gene locus through activating Smad proteins, promoting Foxp3 expression and Treg cell generation. b Mitochondrial fusion is induced by TGF-β through the Smad2/3/PGC-1α signaling pathway, and it is a switch that governs metabolic reprogramming during Treg cell differentiation. TGF-βR transforming growth factor β receptor, IL-10R interleukin 10 receptor, Smad mothers against DPP homolog, SKI sloan-kettering institute, SnoN SKI novel, PGC-1α peroxisome proliferator-activated receptor γ coactivator 1α, Mfn1/2 mitofusin 1/2, Opa1 optic atrophy protein 1, HIF-1α hypoxia-inducible factor 1α, HK2 hexokinase 2, CPT1 carnitine palmitoyltransferase 1. The graphic is created with BioRender.com

Role played by other cytokines in the induction of Treg cells

Interleukin-10 (IL-10) is a pivotal anti-inflammatory cytokine that suppresses the activation of immune cells and the production of inflammatory factors.¹⁸⁷ Previous studies have demonstrated that IL-10 is a significant mediator of the immunosuppressive function of Treg cells.^13,188 Recent studies have indicated that the absence of innate immune IL-10Rβ signaling can compromise the production and function of wild-type Treg cells in vivo,¹⁸⁹ suggesting a potential impact of IL-10 on the generation and development of Treg cells in humans and mice. A study based on samples from clinical gastric carcinoma (GC) patients found that in the GC tumor microenvironment, the microRNA-192-5p/RB1 axis had been shown to regulate the secretion of the anti-inflammatory cytokine IL-10 through the modulation of nuclear factor-κBp65 (NF-κBp65), inducing the differentiation of Treg cells.¹⁹⁰ A diet with a high magnesium content has been shown to increase the number of Treg cells in the spleen of mice, thereby alleviating rheumatoid arthritis (RA). However, the protective effect of this diet was found to disappear, and Treg cell proliferation was impaired after IL-10 knockout in mice.¹⁹¹ This suggests that the induced proliferation of Treg cells in a high magnesium diet is IL-10-dependent. A study of Treg cell production in chronic helminthic infections found that viable cyst products prepared from the helminth Taenia solium were able to stimulate the production of prostaglandin E2 (PGE2) and IL-10 by human peripheral blood mononuclear cells (PBMCs) and thus induced Treg cell production.¹⁹² These findings suggest that IL-10 plays an important role in inducing Treg cell production. In addition, in humans, IL-10 inhibits the PI3K/Akt signaling pathway and enhances the function of forkhead box O (Foxo) 1,¹⁹³ a key regulator of Treg cell function,¹⁹⁴ which is critical in Treg cell proliferation, development, and function.^195–197

In addition, IL-35 has been demonstrated to stimulate the generation of Treg cells. IL-35 is an inhibitory cytokine that is predominantly produced by Treg cells and is essential for the optimal inhibitory activity of these cells.^14,198 Treatment of human or animal naïve T cells with IL-35 induces a subpopulation of Treg cells, termed iTR35. The immunosuppressive effects of iTR35 are mediated by IL-35 rather than through IL-10 or TGF-β. More specifically, iTR35 cells do not express Foxp3 but have strong stability and immunosuppressive capacity in vivo.¹⁴ Binding of IL-35 to IL-35R (IL-12Rβ2⁺ gp130) induces the phosphorylation of STAT1 and STAT4 to form a STAT1:STAT4 heterodimer, which binds specifically to multiple sites in the promoters of the genes encoding the IL-12 subunits p35 and Ebi3, and induces iTR35 production as well as IL-35 secretion.¹⁹⁹ Recent studies have demonstrated the capacity of IL-35 to stimulate the proliferation of Foxp3⁺ Treg cells. For instance, in an apolipoprotein E (ApoE) knockout mouse model of atherosclerosis, IL-35 has been observed to induce the generation of CD4⁺ Foxp3⁺ Treg cells, which subsequently hindered the progression of atherosclerosis.²⁰⁰

In summary, cytokines are crucial signals that drive the differentiation, expansion, and short-term functional maintenance of Treg cells. However, strategies based on the induction of cytokines cannot confer antigen specificity to Treg cells and may induce broad immunosuppression. Combining these strategies with antigen-stimulus-based induction strategies may mitigate these limitations.²⁰¹ Furthermore, phenotypes and functions that are maintained by exogenous cytokines are often unstable and reversible. Induced Treg cells readily undergo phenotypic loss and functional decline when exposed to proinflammatory microenvironments.^202,203 Additionally, cytokine induction strategies, particularly IL-2-based approaches, carry the risk of off-target effects. To overcome these challenges, researchers can employ a synergistic approach combining cytokine-induced Treg cell differentiation with epigenetic regulation strategies to stabilize the expression of key genes such as Foxp3, integrated with metabolic reprogramming to optimize cellular energy utilization. It may mitigate the potential toxicity of cytokines such as IL-2 and enhance the long-term phenotypic stability and functional persistence of induced Treg cells in vivo.

Metabolic pathways

Mammalian target of rapamycin (mTOR) is an evolutionarily conserved serine/threonine kinase, and it is capable of sensing and integrating a variety of environmental signals, including nutrients and growth factors. mTOR is found in two multiprotein complexes: mTORC1 (sensitive to the immunosuppressant rapamycin) and mTORC2 (somewhat resistant to rapamycin).²⁰⁴ The mTOR signaling pathway is a central regulator of cell growth factors and metabolism, playing an important role in metabolic reprogramming, the induction of proliferation, the stability of immunosuppressive function, and the functioning of Treg cells.²⁰⁵ Increased mTOR activity enhances HIF-1α-mediated glycolytic metabolism²⁰⁶ and downregulates Foxp3 expression.²⁰⁷ TSC complex subunit 1 (TSC1) is a negative regulator upstream of mTORC1. In the presence of an inflammatory environment, TSC1-deficient Treg cells have been observed to undergo a loss of Foxp3 expression, subsequently transforming into Tconv cells. These cells then secrete proinflammatory cytokines, including IL-17 and IL-1β, which exhibit a reduced capacity for immunosuppression in a model of colitis. However, the knockdown of the mTORC1 downstream target protein S6 kinase 1 (S6K1) was able to reverse this phenotypic transition.²⁰⁸ Furthermore, the deletion of phosphatase and tensin homolog (PTEN), a pivotal negative regulator of the PI3K signaling pathway, has been shown to result in elevated activation of mTORC2, leading to the instability of Treg cells.²⁰⁹ Conversely, the inhibition of the mTOR signaling pathway promotes the activation of Foxo1 and Foxo3a transcription factors, thereby inducing Foxp3 expression.^210,211 Prior to TCR stimulation, transient inhibition of mTOR by the mTOR inhibitor rapamycin effectively promotes TCR-induced proliferation of Treg cells in humans and mice.²¹² Progesterone (P4) has also been shown to induce the generation of iTreg cells by negatively regulating the mTOR signaling pathway. The iTreg cells induced by P4 were more effective than those induced by IL-2 and TGF-β1 in inhibiting the development of EAE.²¹³ Despite the extensive recognition of robust mTOR signaling as a pivotal negative regulatory pathway for Treg cell differentiation and proliferation, it does not imply that Treg cell homeostasis and functionality are not contingent on this signaling.^214,215 Although studies have shown that transient inhibition of the mTOR signaling pathway prior to TCR stimulation promotes the induction and proliferation of human and mouse Treg cells, the maintenance of their proliferation often requires intact mTOR activity.²¹² This may be because Treg cells require a low metabolic rate to enter the cell cycle and initiate proliferation. Therefore, inhibiting the mTOR signaling pathway before activation, thereby reducing cellular metabolic activity, facilitates the initiation of their proliferation program. Conversely, sustained proliferation demands elevated metabolic rates, as cells require substantial nutrients and growth factors. Activation of the mTOR signaling pathway enhances cellular uptake and utilization of these nutrients, thereby regulating anabolic metabolic processes.²¹² Activation of Raptor, a key constituent protein of the mTORC1 complex in Treg cells, can promote cellular cholesterol and lipid metabolism. The mevalonate pathway is a key metabolic pathway for cholesterol and isoprenoid lipid synthesis. Research has revealed that it is crucial for Treg cell proliferation and the upregulation of the inhibitory molecules CTLA-4 and ICOS. When Raptor is completely knocked out, the inhibitory activity of Treg cells is severely disrupted.²¹⁶ Therefore, suitable mTOR activity is necessary to maintain the stability and optimal inhibitory function of iTreg cells.

Liver kinase B1 (LKB1) is a recognized tumor suppressor,²¹⁷ and it can regulate a range of metabolic processes to maintain the stability and function of Treg cells and induce their proliferation. Research has found that, in Treg cells lacking LKB1, a significant proportion of cells lose Foxp3 expression and transform into exTreg cells that exhibit effector T cell-like functions.²¹⁸ The essential amino acid tryptophan (Trp) is metabolized by tryptophan 2,3-dioxygenase (TDO2) and indoleamine 2,3-dioxygenase 1 (IDO1) for the catabolism to kynurenine (Kyn),²¹⁹ which activates AhR.²²⁰ AhR can induce the generation of Treg cells through multiple pathways, depending on the specific ligand type and environment. AhR has been shown to enhance the transcriptional activity of S-Phase Kinase-Associated Protein 2 (Skp2), thereby promoting the ubiquitination of serine/threonine kinase liver kinase B1 K63. This enhances LKB1-mediated fatty acid oxidation (FAO), thus inducing Treg cell production.²²¹ Furthermore, LKB1 regulates the oxidative phosphorylation (OXPHOS) process in Treg cells in a microtubule affinity-regulating kinase (MARK)- and salt-inducible kinase (SIK)-dependent manner,²²² and the supply of energy for OXPHOS is required for the generation of human and mouse effector Treg cells with immune-suppressing functions.²²³ In the lipid metabolic pathway, the deletion of LKB1 has been observed to increase CD36 expression and a concomitant decrease in the expression of the ATP-binding cassette (ABC) transporter.²²⁴ CD36 is required for cellular uptake of cholesterol and is essential for the regulation of intracellular cholesterol levels.²²⁵ ABC A1-type and G1-type (ABCA1/ABCG1) are responsible for cholesterol efflux.²²⁶ A disturbance in the expression of CD36 and ABC will result in the abnormal accumulation of intracellular cholesterol levels and the inhibition of the mevalonate pathway, thereby disrupting the stability and function of Treg cells and promoting their conversion to an inflammatory phenotype.²²⁴ Furthermore, LKB1 promotes the production of geranylgeranyl pyrophosphate (GGPP) by regulating the activity of the mevalonate pathway. Studies have revealed that GGPP can enhance IL-2-induced STAT5 phosphorylation, thereby promoting the maintenance of Treg cell function.²²⁴

In addition to the intracellular core energy sensing pathways described above, metabolic signaling molecules in the microenvironment (such as gut microbial metabolites and lactic acid) also have a profound impact on the induction and functional maintenance of Treg cells.

Research has shown that Trp derivatives produced by the gut microbiota through the metabolism of Trp (such as indole-3-acetic acid and indole-3-aldehyd²²⁷) can induce the generation of intestinal Treg cells by activating AhR. The effects of these microbial metabolites on Treg cells and the body’s immune homeostasis are not limited to the gut. They can also cross the intestinal barrier and influence the immune microenvironment of multiple organs via the circulation. For example, there is an intestinal-placental immune axis in pregnant mice. RORγt⁺ FoxP3⁺ Treg cells are believed to migrate from the gut to the MFI during pregnancy and exert immunosuppressive functions. The absence or dysregulation of the gut microbiota leads to a reduction in RORγt⁺ FoxP3⁺ Treg cells at the maternal-fetal interface (MFI), disrupts the homeostasis of IFN-γ and IL-17 responses at the MFI, and results in fetal resorption.²²⁸ In germ-free mice, oral administration of indole-3-carbinol (I3C) or colonization of the gut with Lactobacillus murinus, which can metabolize Trp, restored levels of RORγt⁺ Treg cells in the uterus, effectively reducing fetal resorption.²²⁸ This suggests that the gut microbiota can activate AhR through Trp metabolites, inducing RORγt⁺ Treg cells to migrate from the gut to the maternal-fetal interface, thereby helping to maintain maternal-fetal immune tolerance. Furthermore, in a mouse model of hepatitis, indole-3-carbaldehyde (ICA), a Trp metabolite derived from the gut microbiota, can act on T cells in the liver via the bloodstream. By binding to the AhR within T cells, it induces the nuclear translocation of AhR and upregulates the expression of the Pik3ip1 gene. Pik3ip1 is a negative regulator of the PI3K signaling pathway. Its upregulation inhibits the PI3K/Akt/mTOR signaling pathway, inducing Treg cell generation and thereby alleviating liver damage.²²⁹

In recent years, lactic acid has been shown to be an important metabolic signaling molecule that can actively regulate immune cell function. In the tumor microenvironment, lactic acid exerts a particularly significant influence on tumor-infiltrating Treg cells (TI-Tregs). Tumor cells often preferentially utilize glucose for glycolysis to generate energy (Warburg effect), resulting in a tumor microenvironment characterized by high lactic acid and low glucose levels. Under such metabolic conditions, the function of Teff cells is impaired, whereas TI-Treg cells exhibit unique metabolic adaptability. In a high-lactic acid environment, TI-Treg cells highly express lactic acid dehydrogenase (LDH, encoded by the Ldha gene) and the monocarboxylate transporter MCT1 (encoded by the Slc16a1 gene). TI-Treg cells take up lactic acid via MCT1. Subsequently, lactic acid is converted to pyruvate by LDH, enters the tricarboxylic acid cycle, and promotes oxidative phosphorylation,²³⁰ yielding citrate and malate.²³¹ TI-Treg cells are capable of incorporating carbon derived from lactic acid into phosphoenolpyruvate (PEP), which is generated by phosphoenolpyruvate carboxykinase (PEPCK) when malate leaves the mitochondria and is converted to oxaloacetate. This process supports the maintenance of Treg cell proliferation in low-glucose environments.²³¹ A study based on samples from clinical cancer patients and mouse models of cancer found that PEP increased cytoplasmic Ca²⁺ concentrations and promoted the translocation of nuclear factor of activated T cells 1 (NFAT1) into the nucleus, thereby enhancing the transcription of PD-1 and facilitating the immunosuppressive function of TI-Treg cells.²³² Furthermore, studies have shown that high glucose concentrations impair the immunosuppressive function of Treg cells, whereas lactic acid pretreatment effectively mitigates the harmful effects of high glucose concentrations, thereby helping to maintain Treg cell function.²³¹ Lactic acid not only supports the function of Treg cells in the tumor microenvironment but also plays a potential role in the in vitro expansion of Treg cells. Karoliina Tuomela et al. found that, during the in vitro expansion of human Treg cells, the addition of lactic acid 3 days after stimulation of the initial Treg cells significantly improved Treg purity, survival, and immunosuppressive function.²³³

In summary, the precise regulation of metabolic signaling molecules and metabolic networks not only provides energy for Treg cells but also plays a crucial role in determining their functional persistence and environmental adaptability. However, although extracellular signals and metabolic regulation provided by cytokines and metabolic signals can initiate Treg cell differentiation and maintain Treg cell function to a certain extent, they do not regulate Foxp3 gene expression in a permanent manner. Intervention at the epigenetic level could potentially overcome the core bottleneck of phenotypic loss experienced by Treg cells during sustained expansion.

Epigenetic regulation

The stability of Treg cell phenotype and function is contingent on the stability of Foxp3 and its coordinated expression with other Treg cell signature genes. Strategies such as antigen stimulation, cytokine activation, and metabolic pathways induce Treg cells through signal initiation, lineage differentiation, and energy supply mechanisms. However, many induced iTreg cells frequently exhibit suboptimal stability within the in vivo environment, resulting in the loss of the Foxp3 phenotype and subsequent conversion to effector T cells.²³⁴ This process leads to the loss of their immunosuppressive function, and it is closely associated with the dysregulation of the epigenetic regulation of the Foxp3 gene. The use of epigenetic regulation to regulate the transcription of genes that characterize Treg cells is an important way to maintain their stability.

The CNS2 element in the Foxp3 gene is enriched with CpG gene sequences, and this segment is termed the Treg-specific demethylated region (TSDR). This region can undergo specific demethylation during the development and maturation of tTreg cells,⁵² which increases chromatin accessibility and protein-binding capacity.²³⁵ Furthermore, transcription factors, such as Ets-1,²³⁶ are able to bind specifically to the demethylated TSDR, thus stabilizing and highly repressing Foxp3 expression in effector Treg cells. In tTreg cells, the CpG site undergoes complete demethylation, and this phenomenon has also been observed in pTreg cells. However, the region is rarely demethylated in iTreg cells, which is the primary reason why Foxp3 expression in iTreg cells is highly unstable and why iTreg cells are susceptible to loss of immunosuppressive function.^237–239 In addition to affecting the stability of Treg cells through metabolic pathways, LKB1 deficiency leads to elevated levels of STAT4 phosphorylation. Activated STAT4 binds to the CNS2 region of the Foxp3 gene, recruiting the DNA methyltransferase Dnmt1 and promoting methylation in this region. This impacts the expression stability of Foxp3 in Treg cells.²¹⁸ The binding of the ten-eleven translocation (Tet) family enzyme Tet1 to Tet2 and Foxp3 is promoted by TGF-β-activated Smad3 and IL-2-activated STAT5.²⁴⁰ Tet family enzymes are able to convert 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC),^241,242 which is a key link in demethylation. The enhancement of the catalytic activity of Tet family enzymes using hydrogen sulfide,²⁴⁰ vitamin C,^243,244 and ascorbic acid (ASC)²³⁵ has been demonstrated to promote TSDR demethylation and achieve stable Foxp3 expression. Furthermore, it has been shown that the provision of strong TCR stimulation enhances the binding of Tet2 to CNS2 and thus maintains the expression of Foxp3 in iTreg cells.²⁴⁵ It has been demonstrated that STAT6 is able to bind to the Foxp3 silencer, thereby inhibiting TGF-β-induced Foxp3 expression.²⁴⁶ Furthermore, the degree of demethylation of the TSDR region in iTreg cells induced in the absence of the STAT6 signaling pathway is enhanced.²⁴⁷ Based on these findings, the use of CRISPR-TET1 technology to mediate TSDR demethylation can further enhance the phenotypic stability and immunosuppressive ability of iTreg cells.²³⁹ In addition, activation of Foxp3 transcription using CRISPR/dCas9 gene activation technology has been found to promote demethylation of CNS2, which also enhances the stability and repressive function of Treg cells.²⁴⁸

Acetylation of histone tails at the Foxp3 promoter has been demonstrated to be essential for the initiation of Foxp3 transcription,²³⁵ and this process is modulated by histone acetyltransferases (HATs) and histone deacetylases (HDACs). The former mediates the acetylation of histone lysine residues²⁴⁹ and consequently enhances chromatin accessibility, while the latter can reverse this process.²⁵⁰ HDAC inhibitors, such as tricosactin-A (TSA), have been shown to stimulate the development of Treg cells.²⁵¹ Histone acetylation has been demonstrated to enhance the induction efficiency, expression level, and expression stability of Foxp3 during the early stages of iTreg cell development,²³⁵ thus facilitating Treg cells to perform their optimal immunosuppressive functions. cAMP-response element binding protein (CREB)-binding protein (CBP) and p300 (E1A binding protein P300 (Ep300) or lysine acetyltransferase 3A (KAT3B)) are common HATs that can regulate the expression of the Foxp3 gene by targeting histone acetylation.²⁵² CBP and p300 exhibit significant sequence similarity and functional overlap, thus typically being regarded as one integral CBP/p300.²⁵³ Research has found that, in human Tregs, CBP/p300 primarily promotes Foxp3 expression by opening the chromatin structure through the acetylation of histone 3 lysine 27 (H3K27) within the Foxp3 gene.²⁵² Experiments in mice have also demonstrated that the deletion or pharmacological inhibition of p300 leads to a reduction in histone acetylation levels in the Foxp3 promoter region, thereby increasing apoptosis and impairing the inhibitory function of Treg cells.²⁵⁴ As there is some functional redundancy between CBP and p300, the loss of either CBP or p300 alone results in mice developing minimal autoimmune diseases, whereas the combined loss of CBP and p300 leads to fatal autoimmune diseases by 3–4 weeks of age.²⁵⁵ Furthermore, in addition to regulating Foxp3 gene expression via epigenetic modifications, HAT is also capable of maintaining the stability of the Foxp3 protein by promoting the acetylation of Lys-31, Lys-262, and Lys-267 sites on the Foxp3 protein, thereby regulating the immunosuppressive function of Treg cells.²⁵⁶ Therefore, intervening in HDACs and HATs may be an effective strategy to regulate the induction and immunosuppressive functions of Treg cells. Additionally, retinoic acid mediates the binding of retinoic acid receptors (RARs) and retinoid X receptors (RXRs) to the major binding site within the Foxp3 gene CNS1 and to minor sites within its promoter. This increases histone acetylation in the Smad3 binding site region, enhancing the binding of phosphorylated Smad3 (pSmad3) and promoting the expression and maintenance of Foxp3.²⁵⁷ Recently, studies have confirmed that metabolites produced by the gut microbiota can modulate the immunosuppressive function of Treg cells through epigenetic regulatory mechanisms. Studies have shown that the gut microbiota primarily regulates the differentiation and function of intestinal Treg cells through metabolites such as short-chain fatty acids (SCFAs), bile acids (BAs), and tryptophan derivatives.²⁵⁸ Among these, Trp derivatives primarily induce Treg cell generation by activating AhR, which is discussed in detail in “Metabolic pathways.” Short-chain fatty acids are the primary metabolites produced when the gut microbiota breaks down dietary fiber and other indigestible carbohydrates. Propionate, one of the SCFAs, can inhibit HDAC from functioning as a histone deacetylator.²⁵⁹ In addition, butyrate has been found to induce acetylation of histone H3 in the Foxp3 promoter and enhancer CNS1 region^259,260 via a pathway that is dependent on free fatty acid receptor 2 (FFAR2),²⁶¹ thereby inducing Treg cells to establish intestinal immune homeostasis. Bile acids are primarily synthesized in the liver. Then approximately 95% of them are reabsorbed in the ileum, while the remainder is converted into secondary bile acids by gut microbiota through the enzymatic activity of bacterial bile salt hydrolases (BSH) and hydroxysteroid dehydrogenases (HSDHs).²⁶² By screening a library of bile acid metabolites, Hang et al. identified two bile acid metabolites (3-oxoLCA and isoalloLCA) produced by gut bacteria that maintain intestinal immune balance by regulating Th17/Treg cells.²⁶³ Specifically, 3-oxoLCA inhibits the differentiation of proinflammatory Th17 cells by binding to and suppressing the activity of the Th17 transcription factor RORγt. IsoalloLCA significantly increases the levels of mitochondrial reactive oxygen species (mtROS) in Treg cells, promotes H3K27 acetylation at the Foxp3 gene enhancer CNS3, and thereby facilitates Treg cell differentiation and the exertion of their immunosuppressive functions.²⁶³ Further research has found that bacteria in the gut, such as Bacteroidetes, can metabolize 3-oxoLCA into isoalloLCA, thereby promoting the production of mtROS. This, in turn, activates nuclear receptor subfamily 4 group A member 1 (NR4A1), facilitates the binding of NR4A1 to the Foxp3 promoter, and ultimately stimulates Foxp3 transcription in a CNS3-dependent manner.²⁶⁴ In fecal samples from patients with inflammatory bowel disease, significantly reduced levels of isoalloLCA and the abundance of genes associated with its synthesis were observed, further highlighting its potentially important role in intestinal immune homeostasis.²⁶⁴

Engineered technology (CAR-Treg)

The difficulty of achieving large-scale production remains a major challenge to the clinical application of Treg cells. However, the development of chimeric antigen receptor (CAR) technology offers a potential breakthrough for this issue. Furthermore, CAR-engineered Treg cells are a cutting-edge approach to enriching and expanding antigen-specific Treg cells. CAR-T cell therapy was initially developed for use in cancer treatment. This therapy involves genetically engineering T cells to express CARs capable of recognizing specific tumor antigens, thereby endowing the T cells with the ability to recognize specific antigens on tumor cells independently of MHC presentation. This activates T cell-mediated cytotoxicity, enabling the precise elimination of tumor cells. It represents a breakthrough approach for treating diseases such as hematologic malignancies.²⁶⁵ CAR is an engineered synthetic receptor that typically comprises an antigen-binding domain of a specific antibody, a hinge region, a transmembrane domain, and one or more intracellular signaling domains.²⁶⁶ These four components integrate all the essential signals required for T cell activation. This enables the cells to bind independently to target cells and become activated, thus avoiding dependence on APC stimulation.²⁶⁷ Among these components, the antigen-binding domain of the specific antibody confers antigen specificity to CAR-T cells. In recent years, CAR technology has also been used to alter the specificity of Treg cells, thereby increasing local immune tolerance to particular antigens,²⁶⁸ which can effectively reduce the risk of infections and tumors caused by non-specific immunosuppression. The production of CAR-Treg cells involves several key steps. Firstly, CD4⁺ CD25⁺ polyclonal Treg cells are sorted from PBMCs using magnetic-activated-cell-sorting (MACS) technology.²⁶⁷ Furthermore, compared with MACS, fluorescence-activated cell sorting (FACS)²⁶⁹ can effectively improve sorting purity by applying additional selective markers (such as CD127low) and employing density-defined cut-offs for sorting, but this technology places greater demands on the equipment. Secondly, CAR gene modification is used to obtain CAR-Treg cells. This process is typically achieved through methods such as gene transfection mediated by viral vectors (e.g., lentiviruses, retroviruses), non-viral approaches (e.g., electroporation for the delivery of mRNA or plasmid DNA), and gene editing techniques (e.g., the site-specific insertion of CAR sequences into the T cell receptor alpha constant region (TRAC) using CRISPR-Cas9 technology).²⁶⁹ Finally, CAR-Treg cells are expanded in vitro to generate sufficient numbers of cells to support effective treatments.²⁷⁰ In order to obtain higher purity CAR-Treg cells, the mTOR inhibitor rapamycin can be added to selectively deplete Tconv cells.²⁶⁷ Katherine G MacDonald et al. designed a CAR (A2-CAR) targeting the human leukocyte antigen (HLA) class I molecule, A2 (HLA-A2). Subsequently, they isolated CD25^hi CD45RA⁺ cells from peripheral blood and subjected them to CAR gene modification. CAR-Treg cells targeting HLA-A2 were successfully obtained, and the stability of CAR-Treg cells in vitro and in vivo was demonstrated for the first time. Furthermore, the efficacy of CAR-Treg cells in preventing transplant rejection as well as GVHD was demonstrated in mouse models.²⁷¹ In addition to their vital role in transplant tolerance, CAR-Treg cells have shown great promise as a therapy in various disease models, including inflammatory bowel disease,²⁷² multiple sclerosis,²⁷³ and autoimmune disorders such as type 1 diabetes.²⁷⁴

Despite its significant clinical potential, CAR-Treg cell therapy still faces numerous critical challenges in terms of practical preparation and translational application. Treg cells constitute an extremely small proportion of PBMCs, and Treg cells isolated via MACS often exhibit low purity, potentially leading to contamination with Tconv cells in the final product. Consequently, obtaining sufficient quantities of sufficiently pure starting cells represents the primary challenge in CAR-Treg cell preparation. Some researchers have modified CD4⁺ or CD3⁺ T cells with CARs, simultaneously introducing Foxp3 cDNA to differentiate these cells into CAR-Treg cells.^275,276 While this strategy partially overcomes the limitation of insufficient peripheral blood Treg levels, it requires the coordinated expression of multiple genes, placing greater demands on vector design and production processes. How to efficiently and safely integrate CAR into primary Treg cells also poses a significant challenge. Treg cells exhibit a relatively inactive cell cycle, limited endocytic capacity, and high functional sensitivity, all of which make them challenging to transfect. Conventional methods, such as retroviral or lentiviral transduction strategies, pose significant safety risks. In recent years, non-viral transfection approaches, including nanoneedle-based electroporation²⁷⁷ and non-viral vector delivery of CAR mRNA, have emerged as promising alternatives due to their simplicity and improved immunological safety. Furthermore, CAR-Treg cells infused into recipients tend to be depleted and persist poorly. CD28 co-stimulatory signaling is considered crucial for Treg cell development and expansion,¹²² and the expression of 4-1BB is one of the defining features of activated human Treg cells.²⁷⁸ Both of these co-stimulatory molecules are widely used in CAR design for Tconv cells. Consequently, past studies typically integrated either CD28 or 4-1BB into CARs for Treg cells. Research has shown that the CD28 co-stimulatory domain is more effective at maintaining the inhibitory function of human CAR-Treg cells, whereas 4-1BB may impair their immunosuppressive capacity.²⁷⁹ Therefore, selecting CD28 as the co-stimulatory domain for CAR-Treg cells may be a superior option compared to 4-1BB. However, similar to CAR-Tconv cells, CAR expression in human Treg cells is often accompanied by tonic signaling. Studies have shown that 4-1BB CAR-Treg cells and CD28 CAR-Treg cells may induce strong persistent background signaling, leading to excessive activation of signaling pathways such as the Akt/mTOR pathway. This results in reduced stability of the Treg cell lineage and diminished in vivo suppressive capacity.^267,280,281 To address this challenge, researchers discovered that briefly administering vitamin C (a cofactor for Tet enzyme) and mTOR inhibitors such as rapamycin during the in vitro expansion phase could partially restore the inhibitory function of human CAR-Treg cells in vivo while prolonging their survival time.²⁸¹ Combining CAR-Treg cells with low-dose IL-2 may also enhance the survival and persistence of human Treg cells.²⁸² In terms of design, exploring co-stimulatory molecules that are more compatible with Treg cells could be a potential strategy for overcoming this challenge, as they would sustain CAR-Treg expansion and stability. In addition, the appropriate design of the binding affinity between the CAR and the antigen is also a crucial factor in maintaining the immunosuppressive function of CAR-Treg cells. Studies have shown that, compared with TCR/CD28-activated Treg cells, human CAR-Treg cells with high antigen affinity exhibit greater cytotoxicity and reduced immunosuppressive capacity, and may even promote the secretion of proinflammatory cytokines. Conversely, reducing the antigen-binding affinity of the CAR can decrease the production of inflammatory cytokines, restore some of its suppressive function, and make human CAR-Treg cells more similar to naturally TCR/CD28-activated Treg cells.²⁸³ CAR-Treg depletion further raises concerns regarding repeated dosing and potential immunogenicity. Clinical studies indicate that patients may develop humoral and cellular immune responses against CAR components in autologous CAR-T cells or donor-specific antigens in allogeneic CAR-T cells.^284,285 Similarly, repeated administration of CAR-Treg cells can induce CAR-specific antibodies in the recipient, compromising their persistence and efficacy.²⁸⁶

Insufficient levels of Treg cells in the peripheral blood are also a significant factor that limits the large-scale production of CAR-Treg cells. Quality control testing for CAR-Treg products is not standardized,²⁶⁷ and there is still no consistent and effective method to predict the in vivo immune regulatory function of CAR-Treg cells.²⁸⁷ Furthermore, the high cost of preparation, lengthy production cycles, and complex manufacturing processes are major constraints on the large-scale production and clinical translation of CAR-Treg cells. Despite the ongoing challenges in scaling CAR-Treg production, these cells demonstrate high controllability, consistency, and automation potential. Once their preparation processes have been standardized and optimized, CAR-Tregs could provide a replicable technical pathway for large-scale Treg production, offering a solution to the clinical challenge of scaling Treg cell manufacturing.

Drugs for inducing and enhancing Treg cells

In the preceding sections, we systematically review the core biological principles and strategies for inducing functional Treg cells and mention various drugs and reagents that play crucial roles in Treg cell differentiation, proliferation, and stabilization. Table 2 lists additional drugs that are currently reported to effectively induce and enhance Treg cells. Many of these drugs induce Treg cell differentiation, proliferation, and maintenance of function through the aforementioned strategies, while others induce Treg cells via other mechanisms or by combining multiple strategies.

Table 2.

Reagents and drugs to induce Treg cells

Reagents/drugs	Primary targets/pathways	Mechanism	Strategy module	(Disease) modeling	Reference
Prostaglandin E2 (PGE2)	PEG2 receptor	It binds to PEG2 receptors, prostaglandin E receptor (EP) 2 and EP4, and acts synergistically with IL-10 to induce Treg cell production.	Cytokine modulation	Neurocysticercosis (NCC), inflammatory epilepsy	192
Protein inhibitor of activated STAT3 (PIAS3)	STAT3 signaling pathway	It regulates Th17/Treg cell imbalance by inhibiting STAT3-related pathways and upregulates Treg cells.	Cytokine modulation	Spondyloarthritis (SPA)	470
JAK inhibitors (e.g. 2-[(3-Carbamoyl-2-thienyl) amino]-2-oxoethyl (2,6-dichlorophenyl) acetate)	STAT3 signaling pathway	It promotes iTreg cell differentiation by inhibiting STAT3-related pathways.	Cytokine modulation	Rheumatoid arthritis (RA)	471
C3aR/C5aR antagonists (e.g., anti-C3a and anti-C5a monoclonal antibodies)	C3aR/C5aR	It inhibits the PI3K-AKT-mTOR signaling pathway and upregulates the protein kinase A (PKA) signaling pathway, while promoting the production of endogenous TGF-β1 and inducing human iTreg cells with potent immunosuppressive functions.	Cytokine modulation	EAE	472
Neomangiferin	ERK/NF-κB/RORγt inflammatory signaling axis	It inhibits ERK phosphorylation, thereby suppressing NF-κB activation and RORγt expression and function. This inhibits the production of proinflammatory cytokines such as IL-6, IL-17, and TNF-α, while promoting IL-10 production and Treg cell development.	Modulate the cytokine environment	Colitis	473
Sorafenib	VEGFR/AKT/Foxo1 signaling pathway	Sorafenib can activate Foxo1 through the VEGFR/AKT/Foxo1 signaling pathway, which induces Treg cells.	Metabolic pathways	/	474
Ginkgo biloba extract (EGb)	HIF-1α/HK2 signaling pathway	It promotes Treg differentiation by inhibiting the HIF-1α/HK2 pathway.	Metabolic pathways	Ischemic stroke	475
Norisoboldine (NOR)	Core target: AhR; Core signaling pathway: NAD/SIRT1/SUV39H1/H3K9me3 signaling pathway	NOR is a natural AhR agonist that inhibits the formation of HIF-1α/ARNT protein complex in CD4+ T cells under hypoxic conditions, decreases the expression of HK2 and Glut1, and inhibits glycolysis, thereby inducing Treg cell production.	Metabolic pathways	Ulcerative Colitis (UC)	476
Dimethyl Fumarate (DMF)	NRF2/SLC7A11/GSH signaling pathway	Treg proliferation depends on the induction of cystine/glutamate reverse transport solute carrier family 7 member 11 (SLC7A11), and DMF is able to restore impaired SLC7A11 induction in relapsing-remitting multiple sclerosis (RRMS) patients.	Metabolic pathways	RRMS	477
Mesenchymal stem cell (MSC)	/	MSC-mediated transfer of mitochondria (MitoT) induces Treg differentiation.	Metabolic pathways	GVHD	478
G9a inhibitor (e.g., UNC0642)	G9a	G9a inhibition alters the transcriptional regulation of genes involved in lipid biosynthesis in T cells, resulting in increased intracellular cholesterol content and upregulation of lipid pathways, which induces Treg cell differentiation and development through OXPHOS and enhanced lipid membrane composition.	Metabolic pathways	Intestinal inflammation	479
Statins (such as atorvastatin)	HMG-CoA reductase	Statins can competitively inhibit HMG-CoA reductase, thereby suppressing the synthesis of isoprenoid intermediates, such as farnesyl pyrophosphate or geranylgeranyl pyrophosphate. This process affects the activation of the small GTPases, including Ras and Rho-GTPases. Two key substrates of GTPase, PI3K-Akt-mTOR121 and ERK signaling pathways,480 play a crucial role in regulating Foxp3 expression in activated naïve T cells.	Metabolic pathways	Rheumatoid arthritis	362
Melatonin	ROS/TXNIP/HIF-1α	Melatonin downregulates proinflammatory metabolic states by inhibiting the ROS/TXNIP/HIF-1α signaling axis, thereby promoting Treg cell differentiation.	Metabolic pathways	Autoimmune uveitis	481
Sirtuin-1 (Sirt1) inhibitors (e.g., EX 527)	Sirt1	It induces iTreg cell differentiation by inhibiting Sirt1. In addition, Sirt1 deficiency in iTreg cells increases Foxp3 stability.	Epigenetic regulation	GVHD	482
CDK8/19 inhibitor (such as AS2863619 (4-[1(2-methyl-1H-benzimidazol-5-yl)-1H-imidazo[4,5-c]pyridin-2-yl]-1,2,5-oxadiazol-3-amine dihydrochloride))	CDK8/19	Inhibiting CDK8/19 kinase activity, it relieves the inhibitory phosphorylation of the key transcription factor STAT5, thereby enhancing STAT5 activity and stability. This activates the expression of Treg cell-specific genes, including Foxp3.	Epigenetic regulation (by modulating upstream signaling pathways to influence gene expression patterns)	Pemphigus vulgaris; allergic contact dermatitis; delayed-type hypersensitivity reactions, etc.	109,483
Berberine	AhR	It promotes Treg cell differentiation through AhR-induced Foxp3 activation and upregulation of cytochrome P450, family 1, subfamily A, polypeptide 1.	Others	RA	484
CTLA-4-Ig	CD28	Blocking CD28-mediated co-stimulation during TGF-β-dependent Treg cell induction can induce CD101+ Treg cells. Furthermore, in mice sensitized via non-oral routes, in vivo blockade of CD28 signaling using CTLA-4-Ig promotes the differentiation of antigen-specific T cells into CD101+ Treg cells, thereby enhancing oral tolerance.	Others (Engineering protein regulation to modulate co-stimulatory signals)	Food allergy	485
miPEP31	miRNA-31	Encoded by the non-coding RNA (ncRNA) pri-miRNA-31, it functions as a transcriptional repressor to inhibit the expression of miRNA-31, a negative regulator of Treg cells.	Others	EAE	486
ShenFuShanYuRou decoction	STAT1, Gbp5	The bioactive components therein induce Foxp3 expression in a STAT1- and GBP5-dependent manner.	Others (Regulation of mitochondrial metabolism and cell fate via signaling nodes)	Hemorrhagic shock injury	487
Dleu2-17aa	Core target: Smad3; Core signaling pathway: TGF-β/Smad signaling pathway	A 17-amino acid micropeptide, encoded by the sORF in the lncRNA Dleu2, promotes iTreg cell generation by interacting with Smad3 and enhancing its binding to the Foxp3-CNS1 region.	Others (Enhancing Smad3 transcriptional activity to empower TGF-β-induced cytokine strategies)	EAE	38

VEGFR vascular endothelial growth factor receptor, AKT protein kinase B, Foxo forkhead box O, IL-10 interleukin 10, TGF-β transforming growth factor β, GDM gestational diabetes mellitus, sORF short open reading frame, lncRNA long-chain non-coding RNA, Smad mothers against DPP homolog, Foxp3 forkhead box protein 3, CNS conserved non-coding sequences, ncRNA non-coding RNA, pri-miRNA precursor miRNA, EAE experimental autoimmune encephalomyelitis, STAT signal transducers and activators of transcription, GBP guanylate binding protein, HIF-1α hypoxia-inducible factor 1 alpha, HK2 hexokinase 2, AhR aryl hydrocarbon receptor, ARNT aryl hydrocarbon receptor nuclear translocator, Glut1 glucose transporter type 1, GVHD graft-versus-host disease, PBMCs peripheral blood mononuclear cells, Th17 T helper cell 17, JAK Janus kinase, PI3K phosphoinositide 3-kinase, mTOR mammalian target of rapamycin, OXPHOS oxidative phosphorylation, ERK Extracellular regulated protein kinases, NF-κB Nuclear factor kappa-light-chain-enhancer of activated B cells, RORγt Retinoic acid receptor-related orphan receptor γ-t, NAD Nicotinamide adenine dinucleotide, SIRT1 Silent information regulator 1, SUV39H1 Suppressor of variegation 3–9 homolog 1, NRF2 Nuclear factor erythroid 2-related factor 2, SLC7A11 Solute carrier family 7 member 11, GSH Glutathione, ROS Reactive oxygen species, TXNIP Thioredoxin interacting protein

Treg cells in immunotherapy: applications and challenges

As a core regulator of immune homeostasis, depletion or defective function of Treg cells can lead to autoimmune diseases and other immune-related diseases (Fig. 8).^288,289 For these diseases, increasing the number of Treg cells and inducing their optimal immunosuppressive function by various means is an effective strategy for treatment. In tumor or infectious diseases, over-infiltration or hyperfunction of Treg cells has been demonstrated to inhibit the body’s antitumor and anti-infection immunity. Therefore, it is necessary to selectively deplete Treg cells or induce their dysfunction to facilitate the immune system’s robust antitumor and anti-infection functions.^290,291 (Table 3) In the preceding discussion, we have elaborated on integrated strategies combining multiple Treg cell induction methods. However, when applying these strategies to specific disease treatments, the concept of “integration” must be elevated from a technical level to the level of clinical issues. Given that different diseases exhibit distinct pathological mechanisms and therapeutic targets, we should select the most appropriate induction strategies and combinations based on the specific disease context.

Fig. 8.

Applications of Treg cells in immune diseases and non-immune diseases. In immune diseases, Treg cells can be used to treat autoimmune diseases (e.g., systemic lupus erythematosus, type I diabetes, multiple sclerosis, etc.), organ transplantation (including graft-versus-host disease (GVHD) and host-versus-graft rejection (HVGR)), allergic diseases (e.g., allergic asthma, allergic rhinitis, etc.), immunopathological impairment in infectious diseases, tumors, and other diseases. In non-immune diseases, Treg cells are used in metabolic diseases (e.g., insulin resistance, non-alcoholic fatty liver disease), tissue injury (e.g., intestinal mucosal repair, bone tissue repair), fibrotic diseases (e.g., pulmonary fibrosis, renal fibrosis, liver fibrosis), cardiovascular diseases, and osteoporosis. HIV human immunodeficiency virus, HPV human papillomavirus. The graphic is created with BioRender.com

Table 3.

Treg cells in immunotherapy: applications and challenges

Application	Main application strategies	Advantages/Mechanism of action	Main challenges	Examples
Treatment of autoimmune diseases	• Adoptive transfer of autologous/allogeneic Treg cells expanded in vitro • Inducing the expansion/enhancement of Treg cell numbers/functions in vivo using strategies such as low-dose IL-2 therapy	• Restoring immune tolerance • Inhibiting pathogenic autoreactive T/B cells • Reducing tissue inflammation and damage	• The efficiency of in vivo amplification is low. • The quality of Treg cells obtained by in vitro expansion is poor. • It is difficult to achieve targeted delivery to specific organizations/antigens. • Excessive immunosuppression may increase the risk of tumors and infections.	• Systemic lupus erythematosus308 • Type 1 diabetes mellitus318 • Multiple sclerosis329
Transplant tolerance	• Adoptive transfer of autologous Treg cells • Low-dose IL-2 and other therapies induce/expand Treg cells in recipients • Combination therapy with immunosuppressive agents	• Inhibiting alloreactive T cells • Promoting long-term survival of transplanted organs • Reducing or avoiding side effects caused by long-term use of systemic immunosuppressants.	• It is difficult to induce donor-specific Treg cells. • There are issues with the survival, homing, and functional maintenance of transplanted cells in the body.	• Liver transplant383 • Kidney transplant488 • Lung transplant489 • GVHD after hematopoietic stem cell transplantation490
Treatment of allergic diseases	• Inducing/enhancing allergen-specific Treg cell production/function • Oral immunotherapy (OIT) for food allergies	• Restoring the balance between Treg and Th17 cells • Inhibiting mast cell and eosinophil activation • Inducing immune tolerance to allergens	• Different allergens present complexities. • There are significant individual differences in therapeutic efficacy.	• Allergic asthma390 • Food allergy146
Tumor immunotherapy	• Depleting tumor-infiltrating Treg cells; • Specifically inducing Treg cell dysfunction	Treg cells are a key mechanism in tumor immune escape.	• How to achieve targeted elimination/suppression of Treg cells in tumor tissues while avoiding immune-related adverse events requires further investigation. • Treg stability induced by the tumor microenvironment is relatively strong.	394

IL-2 interleukin 2, GVHD graft-versus-host disease

Autoimmune disease

Autoimmune diseases are a group of disorders that arise from the abnormal activation of the immune system, resulting in the body mistakenly attacking its own healthy tissues. The specific pathogenesis is complex and multifactorial, involving genetic susceptibility,^292,293 environmental factors (e.g., infections,^294,295 etc.), imbalance of immune regulation,^296,297 and resulting in local or systemic inflammatory responses mediated by autoantibodies or autoreactive T cells.²⁹⁸ Treg cells play a crucial role in maintaining good autoimmune tolerance and thus avoiding the development of autoimmune diseases.¹⁹ The traditional treatment of autoimmune diseases usually relies on the use of hormones and immunosuppressants. However, these strategies have been observed to demonstrate limited efficacy and associated risks. In contrast, the therapeutic potential of Treg cell therapy has been demonstrated in the context of autoimmune diseases such as systemic lupus erythematosus, type I diabetes, and multiple sclerosis. This approach has better specificity and is expected to achieve precise targeting of inflammation sites to help correct immune dysregulation and restore the body’s immune balance.

Systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is a disease in which the immune system attacks healthy cells and tissues throughout the body²⁹⁹ and is characterized by the presence of autoantibodies directed against nuclear antigens, immune complex deposition, and chronic inflammation of typical target organs such as skin, joints, and kidneys.³⁰⁰ In patients diagnosed with SLE, there is an increase in the cell numbers of effector/memory T cells³⁰¹ and Th17 populations³⁰² and a decrease in the number of Treg cells, as well as impaired Treg cell-mediated immunosuppression.³⁰³ Furthermore, the degree of Treg cell reduction and functional downregulation correlates positively with the severity of the disease.³⁰⁴ Overexpression of bactericidal permeability-increasing protein (BPI) in T cell-derived exosomes or peripheral blood T cells upregulates the expression of the ZFP36 ring finger protein-like 2 (ZFP36L2) gene,³⁰⁵ which in turn downregulates Helios transcription to inhibit Treg cell differentiation.³⁰⁶ Consequently, the expression of BPI may serve as a biomarker and a pathogenic factor of human SLE.³⁰⁵ Therefore, targeting BPI to modulate Treg cells in SLE patients may represent a potential therapeutic approach for SLE. Furthermore, defects in Treg cells in SLE patients are closely related to IL-2 deficiency. In vitro experiments have demonstrated that the absence of IL-2 secretion from CD4⁺ T cells is responsible for the defective expression of CD25 in SLE Treg cells, which can be reversed by administering low-dose IL-2 stimulation. Simultaneously, low-dose IL-2 selectively corrects Treg cell defects, favoring the treatment of human SLE.³⁰⁷ In recent years, the strategy of using low-dose IL-2 to induce Treg cell activation for the treatment of SLE has entered clinical trials (such as NCT02955615).^308–310 A recent single-arm clinical trial named Charact-IL-2 (NCT03312335) similarly found that low-dose IL-2 therapy significantly reduced disease activity scores in SLE patients by expanding Treg cells and decreasing their dependence on the glucocorticoid prednisone.³⁰⁹ In this study, researchers also discovered that low-dose IL-2 successfully induced the differentiation of CD38 HLA-DR double-negative (DN) Treg cells into three distinct subsets with different functions in SLE patients: CD38⁺ Treg cells (with intestinal homing properties), HLA-DR⁺ Treg cells (with skin homing properties), and CD38⁺ HLA-DR⁺ Treg cells (with inflammatory site homing properties). This finding provides a theoretical basis for the further development of tissue-specific Treg cell immunotherapies.³⁰⁹ However, clinical data revealed that for patients with refractory SLE, after four consecutive low-dose IL-2 treatment cycles (daily subcutaneous injections of recombinant human IL-2 (aldesleukin) at single doses of 1.5 or 3.0 million international units (only the second cycle) on five consecutive days separated by washout periods of 9–16 days), at the 6th week, the patient condition worsened: the Systemic lupus erythematosus disease activity index (SLEDAI) score increased to 10 while the number of Treg cells decreased.³¹¹ The combination of low-dose IL-2 with rapamycin (low dose of IL-2 (100 WIU, 3–5 d/month, subcutaneous injection) and rapamycin (0.5 mg, once every other day, oral)) could lead to an increase of Treg cells till the 24th week, proving more beneficial for disease treatment (ChiCTR-IPR-16009451),³¹² thereby partially overcoming the limitations of monotherapy.

Type 1 diabetes mellitus (T1DM)

Type 1 diabetes mellitus (T1DM) is a serious chronic disease caused by immune system-mediated destruction of pancreatic β-cells,³¹³ and defects in Treg cell induction and function have been suggested to be an important factor in the development of T1DM.^314,315 It was found that miR142-3p was elevated in human and NOD mouse pancreatic islet autoimmunity, inhibited Tet2, and destabilized Treg cells by upregulating the methylation level of CNS2 in Foxp3, which resulted in defective immunosuppressive function of Treg cells and facilitated the activation of autoimmunity as well as the development of T1DM.³¹⁶ In pediatric patients, the capacity of insulin-specific Treg cells to impede the progression from islet autoimmunity (pre-T1DM) to clinical T1DM underscores the pivotal role of Treg cells in the management of islet autoimmunity.³¹⁷ The findings of numerous phase 1 clinical trials (e.g., NCT01210664 and NCT02772679) employing autologous polyclonal expanded Treg (expTreg) cells for the treatment of T1DM have evidenced the feasibility and safety of expTreg cell therapy for T1DM.^318–320 However, a recent phase 2 clinical trial of 110 children and adolescents with new-onset T1DM (NCT02691247) found that single-dose expTreg cell therapy was ineffective in halting the progression of T1DM compared to the placebo group. The study concluded that the quality of the Treg cells, rather than the dose, was an important factor influencing the efficacy of treatment.³²¹ The ineffectiveness of autologous polyclonal Treg cell therapy may be attributable to several factors, including its lack of tissue specificity, the instability of the Treg cell phenotype, characterized by suboptimal Treg cell quality, and an inadequate supply of IL-2 signaling support. A phase 1 clinical study (NCT02772679) investigated the combination of expTreg cells with low-dose IL-2 for the treatment of T1DM. The results indicated that the combination therapy not only increased the number of endogenous Treg cells but also activated NK cells, CD8⁺ T cells, and mucosal-associated invariant T cells.³¹⁹ This suggests that the likelihood of realizing a clinical benefit from this combined treatment approach is low. Gene I Uenishi et al. transformed autologous CD4⁺ T cells into antigen-specific Treg cells stably expressing Foxp3 by gene editing, and introduced the T cell receptor targeting the islet antigen glucose-6-phosphatase catalytic subunit 1 (IGRP) (IGRP305-TCR) and a chemically inducible signaling complex (CISC) that provides IL-2 signaling support in response to rapamycin. This Treg cell therapy (GNTI-122) showed promising therapeutic effects in a mouse model of diabetes.³²² Simultaneously, the activation of CISC mediated by rapamycin in this strategy enables IL-2 signaling to activate GNTI-122 cells more specifically, thereby reducing the potential risks associated with IL-2 off-target effects in clinical trials.

Multiple sclerosis

Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system characterized by oligodendrocyte loss, axonal damage, focal demyelination, and astrocyte proliferation.³²³ In addition, Treg cells are dysfunctional in MS patients,^324,325 which are unable to effectively inhibit local infiltration and destruction of myelin sheaths by Tconv cells reactive to myelin-associated autoantigens. The proliferative potential of Treg cells in MS patients is significantly lower than that in the healthy population,^324,326 which may be related to impaired IL-2-IL-2R-STAT5 signaling and abnormal activation of the mTOR pathway in Treg cells in patients.³²⁶ Reduced expression of 44 and 47 kDa Foxp3 transcripts, the two Foxp3 transcripts most relevant to Treg cell function and homeostasis, was observed in MS patients, resulting in aberrant Treg cell proliferation and function.^326,327 In addition, the primate-specific short isoform (PRDM1-S) of positive regulatory domain zinc finger protein 1 (PRDM1) is significantly upregulated in memory Treg cells from MS patients and induces serum and glucocorticoid-induced kinase 1 (SGK1) expression by directly binding to the promoter and enhancer regions of SGK1 genes. SGK1 subsequently destabilizes Foxp3 by inhibiting the phosphorylation of Foxo1, thereby disrupting the immunosuppressive function of Treg cells. At the same time, inflammatory T cells (e.g., Th17 cells) are redirected toward pathogenic Tconv cells, leading to the onset and progression of autoimmune diseases.³²⁸ Therefore, restoring the proliferative potential of Treg cells and correcting their dysfunction is the key to utilizing Treg cells in the treatment of MS. Kelly R Rhodes et al. developed a biodegradable microparticle containing rapamycin for the treatment of MS.²⁰¹ They functionalized the surface of the microparticles with F5111 ICs and myelin autoantigenic peptide/MHC-II-like molecular complexes, in which the antigenic peptide/MHC-II-like molecular complex could help the microparticles to target antigen-specific T cells and then selectively activate antigen-specific Treg cells through the synergistic effect of F5111 ICs. At the same time, the particles were also able to further enhance the function of Treg cells through the localized slow release of rapamycin, which effectively prevented the onset of EAE (an animal model of MS) in mice and ameliorated the severity of EAE.²⁰¹ This fully demonstrates the advantages of integrating multiple Treg cell induction strategies. Wang et al. designed an immunoswitch nanomodulator (aT-IL2C NPs), which utilized a polymeric carrier (Pep-PLG-GEE) consisting of poly (L-glutamic acid) (PLG), glycine ethyl ester (GEE), and Fc-III-4C peptide. High-affinity interactions between the Fc-III-4C peptide in Pep-PLG-GEE and the Fc fragment of the anti-T cell immune receptor with Ig and ITIM domains (TIGIT) monoclonal antibody (aTIGIT) and the IL-2-specific monoclonal antibody (JES6-1) bound Pep-PLG-GEE, aTIGIT, and JES6-1.³²⁹ T cell immune receptor with TIGIT is a co-inhibitory receptor that is highly expressed specifically on Treg cells, with relatively low expression on CD4⁺ and CD8⁺ Tconv cells.³³⁰ Therefore, aT-IL2C NPs were able to specifically recognize Treg cells through aTIGIT and deliver the IL-2/JES6-1 complex. Since Treg cells specifically expressed high-affinity IL-2R,³³¹ the IL-2/JES6-1 complex released IL-2 to promote the activation of Treg cells, thus realizing active targeting of Treg cells and selective proliferation, which showed significant therapeutic effects in the EAE model.³²⁹ A phase I clinical study (EudraCT: 2014-004320-22) demonstrated the feasibility of using Treg cells to treat MS. The study also revealed that the intrathecal administration of freshly isolated Treg cells yielded superior therapeutic outcomes compared to the intravenous administration of ex vivo expanded Treg cells. This finding suggests that local immunomodulation may be more effective than systemic treatment for neuroinflammation.³³² Alemtuzumab is believed to alleviate relapsing-remitting multiple sclerosis (RRMS) by inducing lymphopenia followed by immune system reconstitution.³³³ Despite the risk of secondary autoimmune complications, alemtuzumab has been approved by the Food and Drug Administration (FDA) for the treatment of RRMS due to its demonstrated sustained clinical efficacy. Clinical studies (NCT00548405 and NCT00930553) revealed that following alemtuzumab treatment, immune reconstitution in RRMS patients favored regulatory subsets. Following treatment, the proportion of CD4⁺ Treg cells increased significantly, while the effector/regulatory T cell (Teff/Treg) ratio decreased.³³⁴ Alemtuzumab, a humanized monoclonal antibody targeting CD52, mediates rapid lymphocyte depletion by targeting CD52. Following depletion, serum IL-7 levels increase significantly,³³⁵ and the depleted T cell pool is predominantly composed of memory T cells, particularly CD4⁺ CD25 high T cells.³³⁶ During subsequent immune reconstitution, Treg cells expand preferentially in the circulation and persist over time.³³⁵ The proportion of these expanded Treg cells among patients’ lymphocytes increased, and their ability to suppress myelin-specific T cell responses was also significantly enhanced.³³⁷ Moreover, the mRNA expression of anti-inflammatory cytokines remained consistently upregulated, while the expression of multiple proinflammatory cytokines and chemokines was suppressed.³³⁷ The shift from effector to regulatory immunity triggered by alemtuzumab is considered a key factor in achieving long-term clinical efficacy against MS. This therapeutic strategy also provides a new insight into Treg cell induction. By extensively depleting existing lymphocyte pools, the body’s intrinsic homeostatic mechanisms are subsequently harnessed during immune reconstitution to guide newly generated immune cells toward a regulatory phenotype, achieving Treg cell proliferation. More importantly, this strategy not only increases Treg cells but also achieves preferential expansion of B cells and NK cells with regulatory phenotypes.³³⁴ Compared with other Treg induction strategies, the approach of paradoxically increasing Treg cells after immune depletion may yield longer-lasting therapeutic effects,³³⁷ as it fundamentally alters the composition of immune cell lineages. However, this approach carries risks of infection or secondary autoimmunity.³³⁸ Therefore, precise regulation of the immune reconstitution process following immune depletion is crucial.

Rheumatoid arthritis

Rheumatoid arthritis (RA) is one of the most prevalent chronic inflammatory diseases,³³⁹ characterized by polyarticular swelling and pain, typically affecting the small joints in the hands and feet.³⁴⁰ In severe cases, it can result in permanent joint damage and disability. In addition, RA is a systemic disease associated with multiple coexisting conditions and extra-articular manifestations,³⁴⁰ affecting various organs throughout the body, including the heart, skin, eyes, and lungs.³⁴¹ The development of RA is strongly associated with genetic factors,³⁴² environmental factors (e.g., smoking³⁴³), infections,³⁴⁴ and other factors. The role of Treg cells in the development of RA is not fully understood. Several past studies of Treg cell counts in the peripheral blood of RA patients have produced conflicting results, including increased,^345,346 unchanged,^347,348 and decreased^349–351 Treg cell counts. However, a growing number of studies have demonstrated an increased frequency of Tregs in the synovial fluid of patients with RA.^346,348,352 Interestingly, these Treg cells in patients’ synovial fluid can exhibit normal immunosuppressive activity at the in vitro level.^34,353 However, under RA inflammatory conditions, Teff cells demonstrate some resistance to Treg cell-mediated immunosuppression, thereby counteracting some of the inhibitory activity of Treg cells.^346,354 In addition, the immunosuppressive function of Treg cells in the peripheral blood of RA patients has been shown to be disrupted, promoting the onset and progression of RA.^355,356 This is specifically manifested by downregulation of CTLA-4 expression, dysregulation of the TCR signaling pathway,³⁵⁶ and reduced expression of T cell immunoglobulin domain and mucin domain-3 (Tim-3), which is critical for the optimal inhibitory function of Treg cells,³⁵⁷ and Treg cells may even be transformed into pathogenic IL-17⁺ Foxp3⁺ T cells under the mediation of IL-6 derived from synovial fibroblasts.³⁵⁸ An ideal therapeutic strategy for RA is to promote the restoration of Treg cell numbers and function and to induce self-tolerance before significant tissue damage occurs.³⁵⁴ In addition, Treg cells have been found to inhibit osteoclast-mediated bone destruction, regulate bone homeostasis,³⁵⁹ and alleviate joint damage in RA. The feasibility of using low-dose IL-2 (NCT01988506)^360,361 and atorvastatin³⁶² to restore the number and quality of Treg cells in RA patients has been demonstrated in several studies. Additionally, combining Treg-based therapies with methotrexate (MTX), a first-line RA medication, has demonstrated favorable efficacy and safety in RA treatment, significantly improving clinical and immunological markers (NCT 02467504).³⁶¹ Recent studies (ChiCTR-INR-16009546) have investigated combining low-dose IL-2 with IL-6 receptor antagonists (e.g., tocilizumab) to treat RA. The results indicate that while the low-dose IL-2 monotherapy group showed a significant increase in Treg cells, the levels of Th1, Th2, and Th17 cells also increased. In contrast, the combination therapy group showed a moderate increase in Treg cells alongside a downward trend in Th1, Th2, and Th17 cells, with a more pronounced decrease in the Th17/Treg ratio. This demonstrates a safer and more effective therapeutic outcome, making this strategy particularly suitable for patients with immune disorders mediated by high levels of Th2 and Th17 cells.³⁶³ It also provides a more effective therapeutic approach for treating patients with autoimmune diseases characterized by highly active proinflammatory cells. Specifically, combining drugs with different mechanisms of action can induce Treg cells and suppress Teff cells simultaneously, thereby restoring immune homeostasis more comprehensively.

Organ transplantation

Organ transplantation is a critical intervention for end-organ failure. However, transplant rejection causes dysfunction and shortens the survival of allogeneic grafts.³⁶⁴ GVHD and host-versus-graft rejection (HVGR) are the two basic types of allogeneic transplant rejection. HVGR is mainly manifested by the recipient’s immune system rejecting the graft.³⁶⁵ Hematopoietic cell transplantation (HCT) is a curative therapy for patients with hematological malignancies and bone marrow disorders.³⁶⁶ GVHD is a major serious complication of allogeneic hematopoietic cell transplantation,³⁶⁵ which occurs when immunocompetent T cells in the donated tissue (the graft) recognize the recipient (the host) as foreign.³⁶⁷ Transplantation tolerance is a state in which the body’s immune system permanently accepts allogeneic grafts in the presence of cessation of immunosuppressive drugs.³⁶⁸ Treg cells have been shown to play a key role in the induction of transplantation tolerance³⁶⁹ as well as the reconstitution of transplantation tolerance after acute rejection of organ transplantation.³⁷⁰ Treg cell depletion increases the risk of allograft rejection in humans and mouse models.^371,372 At present, clinical studies of Treg cells in transplant rejection have mainly focused on two strategies: autologous Treg cell transfusion therapy and low-dose IL-2 therapy. Early clinical studies have shown that in patients suffering from chronic graft-versus-host disease (cGVHD), low-dose IL-2 therapy significantly upregulated the number of Treg cells in the body.³⁷³ Furthermore, the earlier the treatment is initiated, the more effective the treatment becomes.³⁷³ In the prevention of acute GVHD, a phase 2 clinical trial (NCT01660607) found that combined treatment with Treg cells and tacrolimus was superior to Treg cell therapy alone.³⁷⁴ Recent findings from this clinical trial further indicate that, for patients undergoing allogeneic hematopoietic cell transplantation, administering a dose of 2–3 × 10⁶/kg of donor CD34⁺ hematopoietic stem cells and donor Treg cells 2–3 days prior to the infusion of 3 × 10⁶/kg of donor Tconv cells effectively reduces the incidence of acute and chronic GVHD and improves patient quality of life. Furthermore, this therapy enhanced patient responsiveness to first-line corticosteroids even when subjects developed GVHD and effectively avoided the need for potent second-line immunosuppressants such as ruxolitinib.³⁶⁶ In addition to demonstrating promising clinical efficacy in the prevention and treatment of GVHD, Treg cell therapy also shows significant potential in the management of HVGR. The infusion of recipient Treg cells to promote immune tolerance following solid organ transplantation has been shown in several clinical studies to be a safe and feasible approach, particularly in kidney (NCT02371434)^375,376 and liver (NCT02166177) transplantation.³⁷⁷ Multiple preclinical studies have demonstrated that inducing Treg cell generation using IL-2 promotes immune tolerance after solid organ transplantation and improves allograft survival rates. These studies have focused on mouse heart transplantation,³⁷⁸ mouse corneal transplantation,³⁷⁹ mouse skin transplantation,³⁸⁰ mouse islet transplantation,³⁸¹ and mouse lung transplantation.³⁸² However, in clinical practice, the role of low-dose IL-2 therapy in solid organ transplantation remains poorly understood. A clinical trial in stable liver transplant recipients 2–6 years post-transplantation (NCT02949492) demonstrated that low-dose IL-2 significantly and consistently increased circulating Treg cells in participants, however, there was no increase in the number of donor-specific Treg cells or promotion of targeted transport of circulating Treg cells to the transplanted liver, and low-dose IL-2 therapy also had off-target effects that triggered an IFN-γ-mediated inflammatory response in the liver.³⁸³ Ultimately, all liver transplant recipients who attempted to discontinue the immunosuppressive agent tacrolimus while receiving low-dose IL-2 therapy developed signs of immune rejection, leading to the premature termination of the trial. This study provides a critical cautionary note for the clinical translation of Treg cell therapy. In the future, efforts should focus on developing more targeted Treg cell expansion methods by integrating various Treg induction strategies (such as antigen stimulation and CAR technology), as well as enhancing Treg antigen specificity and tissue homing capacity.

Allergic asthma

Allergic asthma is a chronic inflammatory airway disease. Within the respiratory system, Treg cells have been observed to attenuate the activity of inflammatory cells, including Th2 cells, eosinophils, and basophils. This regulatory function is achieved through the secretion of various mediators, such as IL-10, IL-35, and TGF-β, which are implicated in the induction of tolerance responses, thereby mitigating the sensitization response triggered by allergen exposure.³⁸⁴ In patients diagnosed with asthma, the balance among Treg, Th2, and Th17 immune cells is disrupted.³⁸⁵ Allergens and environmental pollutants have been shown in mouse models to induce Treg cells to express Notch4, which in turn interferes with the stability and immunoregulatory function of Treg cells by activating the downstream Wnt and Hippo pathways, leading to the differentiation of Treg cells to Tconv cells, such as Th17 and Th2 cells, and exacerbating airway inflammation.³⁸⁶ Analysis of peripheral blood Treg cells in asthma patients also revealed that Notch4 expression levels correlate positively with asthma severity, and that Treg cells expressing high levels of Notch4 exhibit impaired immunosuppressive function.³⁸⁶ Additionally, in humans, lysophosphatidylglycerol 18:0 (LPG 18:0), a biomarker of asthma, has been found to induce mitochondrial dysfunction as well as impair Treg cell differentiation and function by affecting Foxp3 acetylation through the NAD⁺/SIRT1 pathway in Treg cells.³⁸⁷ Therefore, inducing Treg cells and maintaining their normal function is an effective strategy for the treatment of allergic asthma. Flagellin B (FlaB) is a ligand for Toll-like receptor 5 (TLR5), with the capacity to induce regulatory dendritic cell (rDC) phenotypes in a TLR5-dependent manner.³⁸⁸ In a study conducted on mouse bronchial lymph nodes, it was observed that CD11c⁺ rDCs obtained through the administration of a high dose of FlaB exhibited elevated levels of CD80 and MHC-II-like molecules, accompanied by reduced CD86 expression.³⁸⁹ Additionally, these cells demonstrated a heightened capacity to secrete IL-10. Subsequently, in the presence of IL-10, rDCs were able to induce and enhance the function of Treg cells, thereby effectively ameliorating asthma symptoms.³⁸⁹ Recently, some scholars constructed a lipid nanoparticle (LNP) vaccine platform co-loaded with antigenic mRNA and celastrol.³⁹⁰ Celastrol has potent anti-inflammatory and immunosuppressive activities³⁹¹ and can inhibit the maturation of DCs induced by the inflammatory components of the LNP³⁹² and maintain DCs in a tolerant and inactive state. In a mouse model of allergic asthma, this formulation was able to selectively target splenic DCs, inducing the generation of tolerogenic DCs and antigen-specific Treg cells and promoting the migration of Treg cells to the lungs, which effectively alleviated the symptoms of allergic asthma.³⁹⁰ This provides a new integrated approach for inducing Treg cells in the treatment of asthma and other immune disorders.

Cancer

Tumor cells originate from the body’s tissues. However, due to the expression of tumor-specific mutated genes, the immune system is able to recognize them as non-self cells and destroy them.³⁹³ However, tumor cells can evade recognition and attack by the body’s immune system through a variety of mechanisms, thus promoting tumor progression. Among these cells, Treg cells are abundant in tumor tissues,³⁹⁴ and they promote tumor immune escape by establishing immune tolerance against tumor cells.^393,394 This is a major obstacle in antitumor immunotherapy strategies and an important factor contributing to the difficulty of tumor treatment and poor prognosis. Consequently, depletion of tumor-infiltrating Treg cells or specific induction of Treg cell dysfunction is regarded as an effective strategy in cancer immunotherapy.^394,395 Current research primarily focuses on utilizing anti-CTLA-4 antibodies to deplete intratumoral Treg cells,³⁹⁴ blocking TIGIT to reduce Treg cell numbers and suppress their function in tumors,^394,396 and inhibiting chemokine receptors on tumor-associated Treg cells such as CCR4³⁹⁷ and CCR8³⁹⁸ to reduce Treg infiltration in tumor tissues, and targeting CD25 to restore the Treg/Tconv balance within tumors.³⁹⁴

A clinical trial (EudraCT no. 2014-004388-20) applying a phosphatidylinositol 3-kinase δ (PI3Kδ) inhibitor for the treatment of human solid tumors found that the PI3Kδ inhibitor, AMG319, was able to reduce the level of Foxp3 transcripts, decrease the number of tumor-infiltrating Treg cells, and enhance the cytotoxicity of tumor-infiltrating CD8⁺ T cells, inhibiting tumor growth.³⁹⁹ However, during the course of the trial, 12/21 patients experienced immune-related adverse events (irAEs) manifested by symptoms such as rash, diarrhea, and elevated aminotransferases, which were thought to be related to the reduction of Treg cells as well as their abnormal function.³⁹⁹ Therefore, when developing cancer therapeutic strategies targeting Treg cells, in addition to ensuring therapeutic efficacy, how to avoid the occurrence of irAEs is also a key issue to consider.

It is also noteworthy that not all tumor types respond to Treg depletion strategies. In a spontaneous pancreatic cancer mouse model, studies revealed that Treg depletion failed to effectively alleviate immune suppression and instead accelerated tumor progression.⁴⁰⁰ Tregs serve as a key source of TGF-β. Their depletion reprogrammed the fibroblast population, resulting in the loss of tumor-restraining, smooth muscle actin-expressing fibroblasts. Depletion of Tregs also led to increased secretion of the chemokine ligands CCL3, CCL6, and CCL8 by fibroblasts and epithelial cells. These subsequently recruited large numbers of immunosuppressive myeloid cells via CCR1, inducing compensatory immunosuppression. Furthermore, depletion of Treg cells released more pathological CD4⁺ T cells, intensifying Th2-mediated protumor responses.⁴⁰⁰ Treg cell function may exhibit duality in tumors, potentially limiting excessive tumor growth under specific conditions by regulating stromal cells and suppressing harmful inflammation. Moreover, the tumor immune microenvironment constitutes a modifiable network. Downregulating Treg cells may trigger compensatory upregulation of other immunosuppressive mechanisms. Therefore, successful therapy lies not merely in eliminating a specific cell type, but also in reprogramming the tumor immune microenvironment from an inhibitory state to one conducive to potent antitumor immunity.

Emerging applications of Treg cells in non-immune diseases

In the past, Treg cells have been widely studied for their central role in maintaining immune tolerance and suppressing excessive inflammatory responses. However, with the increasing understanding of Treg cells, an increasing number of studies have revealed that Treg cells also play important roles in metabolism-related diseases, tissue repair, and fibrotic diseases, broadening the scope of application of Treg cells and providing a new paradigm of immune intervention for the treatment of these non-immune diseases (Fig. 8).

Insulin resistance

Tissue-resident Treg cells, found in metabolic tissues such as adipose tissue, are essential for the proper functioning of these tissues and play a significant role in the regulation of systemic metabolism.⁴⁰¹ Normal mouse abdominal adipose tissue is characterized by a high concentration of Treg cells, which possess unique phenotypes and exhibit extremely elevated IL-10 transcript levels. These cells are capable of secreting substantial amounts of IL-10, thereby inhibiting the expression of inflammatory mediators in adipocytes and alleviating insulin resistance.²⁰ In contrast, in an obese mouse model, obesity-induced soluble growth stimulation expressed gene 2 (sST2) attenuates IL-33 signaling and disrupts the homeostasis of Treg cells and group 2 innate lymphoid cells (ILC2s) in adipose tissue.⁴⁰² The imbalance between Tregs and ILC2s exacerbates adipose tissue inflammation and fibrosis as well as promotes insulin resistance.⁴⁰² An increase in Th17 cells and a decrease in Treg cells in mouse adipose tissue were observed after knockdown of the Rab4b gene in mouse T cells, which also exacerbated insulin resistance in obese mice.⁴⁰³ These findings suggest that depletion of Treg cells in adipose tissue contributes to an imbalance in glucose metabolism and impairs insulin sensitivity. A study based on mice found that Treg cells in epididymal visceral adipose tissue (eVAT) were able to inhibit the differentiation of stromal adipocyte precursors through the secretion of oncostatin-M (OSM), which maintained metabolic homeostasis and insulin sensitivity.⁴⁰⁴ And a regulatory axis is seemingly conserved in humans.⁴⁰⁴ In addition, in mice and humans, hematopoietic prostaglandin D synthase (HPGDS) in Treg cells inhibits the proliferative activation of Tconv cells and prevents local inflammation and insulin resistance.⁴⁰⁵ Treg cells are expected to be a reliable therapeutic target for the treatment of metabolic diseases such as insulin resistance. Interestingly, the pathophysiological processes of obesity and age-related insulin resistance differ significantly. The accumulation of adipose tissue-resident Treg cells during aging exacerbates the decline in adipose metabolic function and insulin resistance,⁴⁸ demonstrating that a degree of inflammation may be beneficial for both adipose tissue remodeling and maintenance of metabolic function.⁴⁰⁶ Failure to maintain an optimal immune status in aging adipose tissue may directly contribute to metabolic disorders such as age-related insulin resistance and diabetes.⁴⁸

Tissue repair

In recent years, an increasing number of studies have shown that Treg cells play an important role in tissue repair and regeneration. Research has found that the local administration of exogenous Treg cells promoted wound healing in bone, muscle, and skin tissues in mice. Mechanistically, Treg cells exert their regenerative effects by regulating monocytes and macrophages in damaged tissues through both direct and indirect mechanisms and by promoting their transition to an anti-inflammatory and pro-healing state via factors such as interleukins.⁴⁰⁷ Treg cells have been shown to protect the ecological niche of hematopoietic stem and progenitor cells (HSPCs) from the immune system,⁴⁰⁸ while interacting with innate immune cells such as neutrophils and macrophages to control inflammation and promote tissue healing after tissue injury.⁴⁰⁹ In addition, Treg cells can secrete repair mediators that affect tissue-resident non-immune cells, such as epithelial cells, thereby mediating tissue repair and regeneration.⁴¹⁰ For example, all-trans retinoic acid (ATRA) can directly induce CD161 expression in Treg cells.⁵³ Under TCR signaling, CD161⁺ Treg cells are able to produce soluble factors in a BTB domain and CNC homolog 2 (BACH2)-dependent manner, which can promote wound healing in the intestinal mucosa.⁵³ Independent of the CD161⁺ Treg cell population, the CCR8⁺ Treg cell subpopulation in human blood may be a circulating or precursor form of tissue Treg cells and may be involved in tissue repair processes.⁴¹¹ By comparing human and mouse Treg cells, researchers identified a conserved chromatin repair signature closely associated with the basic leucine zipper ATF-like transcription factor (BATF), which could bind to the upstream enhancer of the CCR8 gene and regulate CCR8 expression.⁴¹¹ BATF is a key factor in the differentiation of tisTregST2 precursors.⁴¹² TisTregST2 is a class of tissue-resident Treg cells expressing growth stimulation expressed gene 2 (ST2) and killer cell lectin-like receptor G1 (KLRG1).⁴¹³ TisTregST2 is capable of expressing tissue regeneration-associated cytokines, such as AREG,⁴¹⁴ and plays an important role in regeneration as well as homeostasis maintenance of non-lymphoid tissues such as VAT⁴¹⁵ and colon.⁴¹⁶ CCR8 is a good marker for recognizing tisTregST2 in mice and humans.⁴¹¹ In fracture healing, BATF⁺ CCR8⁺ Treg cells support skeletal stem cell (SSC) accumulation and osteogenic differentiation by secreting progranulin (PGRN) to promote bone repair.⁴¹⁷ In summary, BATF⁺ CCR8⁺ Treg cells play a critical role in tissue repair.

Fibrotic diseases

Fibrosis is an extensive deposition of fibrous connective tissue, characterized by the accumulation of extracellular matrix (ECM) components such as collagen.⁴¹⁸ The role of Treg cells in fibrotic diseases is currently controversial and may be relevant to specific types of disease models. In acute myocardial infarction (MI) in rats caused by ligation of the anterior descending branch of the left coronary artery, the transfer of exogenous Treg cells has been found to alleviate cardiac fibrosis.^419,420 Mechanistically, these cells can rapidly reach the area of cardiac infarction and regulate monocytes/macrophages by secreting IL-10 to mediate cardiac tissue repair.⁴²⁰ In addition, Treg cells were transplanted into mice injected with angiotensin II-induced hypertension, and it was observed that Treg cells ameliorated cardiac hypertrophy, cardiac fibrosis, and alleviated cardiac injury in mice.⁴²¹ However, in mouse models of ischemic cardiomyopathy and heart failure, Tregs in the heart exhibit a proinflammatory phenotype characterized by high expression of tumor necrosis factor alpha (TNF-α), interferon gamma (IFN-γ), and tumor necrosis factor receptor 1 (TNFR1), driving cardiac fibrosis through immune activation and tissue damage.⁴²² Treg cells in the lung have been found in mice to secrete platelet-derived growth factor (PDGF)-B under noninflammatory conditions, which stimulates fibroblasts and promotes the onset and progression of pulmonary fibrosis.⁴²³ Treg cells are also thought to play an important role in the development of irradiation-induced pulmonary fibrosis following thoracic radiotherapy.^424,425 Depletion of Treg cells has been shown to delay this process by reducing fibroblast accumulation.⁴²⁵ However, in a silica-induced silicosis mouse model, exogenous α-lipoic acid (ALA)⁴²⁶ and Baicalin⁴²⁷ were able to successfully alleviate the symptoms of pulmonary fibrosis by activating Treg cells. In cases of pulmonary fibrosis caused by the autoimmune disease systemic sclerosis, the transfer of Treg cells and low-dose IL-2 therapy have also been shown to significantly alleviate parenchymal fibrosis.⁴²⁸ In a mouse model of chronic liver injury and fibrosis induced by CCl₄ injection, it was demonstrated that intrahepatic Treg cells attenuate tissue fibrosis by inhibiting the activation and proliferation of profibrotic immune cells (e.g., Th2 cells) in the liver.⁴²⁹ However, in non-alcoholic fatty liver disease (NAFLD), AREG produced by Treg cells activates the profibrotic transcriptional program in hepatic stellate cells via epidermal growth factor receptor (EGFR) signaling, which promotes the progression of liver fibrosis.⁴³⁰ The profibrotic or antifibrotic effects of Treg cells in chronic tissue injury need to be further explored. In recent years, the interaction between Treg cells and non-immune cells (e.g., fibroblasts and epithelial cells) has been investigated, revealing its broad potential application in fibrotic diseases.

Osteoporosis

Osteoporosis is a systemic metabolic bone disease characterized by deterioration of bone microarchitecture, decreased bone mineral density and strength, and increased risk of fragility fractures.⁴³¹ The etiology of osteoporosis is frequently the result of a combination of factors acting over an extended period, including metabolic disorders, malnutrition, endocrine diseases, and decreased estrogen levels. Furthermore, the skeletal system is closely associated with the immune system. Research based on mice found that Th17 cells secreted proinflammatory cytokines, such as IL-6 and TNF-α. This enhanced the expression of receptor activator of NF-κB ligand (RANKL) on osteoblasts and fibroblasts. RANKL then binds to its receptor, leading to increased osteoclast formation and bone resorption. IL-17 and TNF-α enhanced the expression of RANKL on osteoblasts and fibroblasts. RANKL binding to its receptor (RANK) promoted osteoclast differentiation and bone resorption, playing a central role in osteoporosis progression.⁴³² Treg cells can inhibit the differentiation of osteoclast precursor cells by binding to B7-1/B7-2 on their surface through the high expression of CTLA-4 on their own surface.³⁵⁹ Treg cells also suppress osteoclast differentiation and activity by secreting inhibitory cytokines such as IL-10 and TGF-β, thereby reducing bone resorption.⁴³³ Additionally, Treg cells can promote the differentiation of SSCs into osteoblasts by secreting factors such as PGRN, which enhances bone formation.⁴¹⁷ It is evident that the dynamic equilibrium between Th17 and Treg cells is crucial for maintaining bone health and reducing the risk of osteoporosis. In postmenopausal women, estrogen deficiency leads to intestinal barrier damage⁴³⁴ and dysbiosis, causing intestinal immune imbalance and disruption of the Th17/Treg cell equilibrium. The resulting systemic inflammation is a major cause of osteoporosis.^435–437 In the ovariectomy (OVX) animal model, an increase in Th17 cells and a decrease in Treg cells were observed.⁴³⁷ In this model, researchers found that administering Lactobacillus rhamnosus GG (LGG) to OVX rats helped to maintain gut health, improve immune imbalance in the intestine, and regulate the bone microenvironment via the circulatory system. LGG treatment was found to downregulate the expression of RORγt, a key transcription factor of Th17 cells in bone marrow, while simultaneously upregulating FOXP3. This restored immune balance in the skeleton, effectively improving symptoms of osteoporosis.⁴³⁷ The probiotic Lactobacillus acidophilus has also been found to suppress bone loss in OVX mice by upregulating Treg cells through its immunomodulatory properties.⁴³⁸ Yang et al. designed T cell-depleting nanoparticles (TDNs) that release the chemokine monocyte chemotactic protein 1 (MCP-1) rapidly in vivo, thereby recruiting activated T cells to their vicinity. Subsequently, T cell apoptosis is induced via the FasL-Fas pathway, which is mediated by Fas-related apoptosis ligand (FasL) bound to the surface of the nanoparticles. This enhances the osteogenic and differentiation capabilities of bone marrow mesenchymal stem cells (BMMSCs). Concurrently, apoptotic T cells release apoptotic extracellular vesicles (ApoEVs). These ApoEVs then induce macrophages to shift from M1 to M2 polarization. The anti-inflammatory factors secreted by M2 cells, such as TGF-β and IL-10, promote Treg cell differentiation. This restores the proportion and function of Treg cells in OVX mice and further attenuates the osteogenic deficiency of BMMSCs and the osteopenia phenotype.⁴³⁵ In summary, Treg cells play a vital role in immunoregulation and bone protection during osteoporosis. Currently, most Treg-based immunotherapies for osteoporosis are still in the preclinical research stages. As our understanding of the relationship between immunity and bone health improves, we will ultimately be able to precisely reprogram the bone immune microenvironment. This advancement will shift the therapeutic window for osteoporosis to the immune prevention stage, enabling drug intervention at an early stage of bone loss.

Summary and outlook

Due to their unique immunomodulatory function, Treg cells are becoming a research hotspot in the field of medicine, and many new technologies and strategies have accelerated the rapid development of Treg cell-based immunotherapy strategies. However, the molecular mechanisms of Treg cell differentiation are not fully understood, especially the dynamic regulatory network under complex pathological conditions. In the opening section of the chapter on strategies for Treg cell induction, we summarized the three main challenges in Treg cell induction: the persistence and maintenance of suppressive function of induced Treg cells in vivo, targeting specificity, and difficulties in large-scale production. In this section, we will provide a more comprehensive discussion of the challenges encountered during Treg cell induction and clinical application.

Challenges encountered during Treg cell induction

The safety and specificity of Treg cell induction strategies remain to be optimized, and cytokine-dependent induction strategies may induce off-target effects.³⁸³ Particularly in strategies utilizing low-dose IL-2 to induce Treg cells, due to the narrow therapeutic window, IL-2 may also enhance the activity of Teff cells⁴³⁹ and NK cells,⁴⁴⁰ resulting in limited therapeutic efficacy and potentially triggering severe adverse reactions.

Long-term stabilization of induced iTreg cells is also a challenge to overcome. Incomplete demethylation of the iTreg Foxp3 TSDR⁴⁴¹ and the absence of signals essential for long-term survival and functional maintenance in vivo (such as IL-2) are key factors contributing to the instability of induced Treg cells. The proinflammatory cytokine IL-6 can recruit Dnmt in a STAT3 signaling-dependent manner, thereby enhancing Foxp3 CpG methylation, reducing Foxp3 expression, and impacting Treg cell stability.⁴⁴² However, recent studies have also found that, when exposed to IL-6 and TNF-α, the proliferation of human Treg cells stimulated by anti-CD3 and anti-CD28 (aCD3/28) beads or CD28 superagonists was significantly enhanced, and the proliferated human Treg cells retained stable Treg lineage characteristics and normal immunosuppressive function. Mechanistically, TNF-α promotes the proliferation of human Treg cells via TNFR through autocrine/paracrine mechanisms, and IL-6 may enhance this effect by upregulating TNFR2 expression.⁴⁴³ This appears to contradict the widely held view that IL-6 inhibits Treg proliferation and activation. Nevertheless, these findings offer new strategic insights for expanding functional Treg cells in inflammatory environments for the treatment of diseases. Other proinflammatory cytokines, such as IL-23 and IL-1, have also been shown to convert Treg cells into exTreg cells.⁴⁴¹ Additionally, the inflammatory microenvironment of the organism may compromise Foxp3 stability,⁴⁴⁴ driving them toward exTregs that exhibit effector T cell-like functions, thereby promoting the activation of autoreactive T cells and other innate immune cells.^445,446 Freuchet et al. were the first to characterize human exTreg cells as CD4⁺ CD8⁻ CD56⁺ CD16⁺ cells that share TCR CDR3 sequences with Treg cells. Unlike Treg cells, these exTreg cells do not exert immunosuppressive functions, but instead exhibit inflammatory and cytotoxic properties.⁴⁴⁷ Studies have found that in both mouse and human Treg cells, under inflammatory conditions, endoplasmic reticulum stress-induced mitochondrial dysfunction within cells affects the stability of Treg cells, promoting their conversion into exTreg cells.^448,449 The p24 trafficking protein 4 (TMED4) is an endoplasmic reticulum stress-responsive protein that regulates Treg stability and suppressive function by sustaining inositol-requiring enzyme 1 α (IRE1α)-dependent regulation of cellular reactive oxygen species (ROS) and nuclear factor erythroid derived 2-like 2 (NRF2)-related antioxidant responses. When TMED4 is absent, HMG-CoA reductase degradation 1 (HRD1)-mediated ubiquitination and degradation of IRE1α increase, leading to a significant reduction in its protein levels and activity. This, in turn, impairs the activity of XBP1s, a downstream transcription factor of IRE1α, resulting in mitochondrial dysfunction and diminished antioxidant capacity. Consequently, excessive intracellular ROS accumulation promotes the degradation of Foxp3 protein in Treg cells. Simultaneously, impaired mitochondrial oxidative phosphorylation forces cells to switch to glycolysis, accompanied by mTOR signaling activation, further destabilizing Treg cells.⁴⁴⁵ Another study indicated that an excessive increase in IRE1α protein levels, caused by HRD1 deficiency, led to the abnormal activation of the p38 MAPK signaling pathway downstream. This resulted in decreased and unstable Foxp3 expression.⁴⁴⁶ This indicates that TMED4 and HRD1 form an antagonistic relationship, jointly regulating IRE1α activity within an optimal range to maintain Treg cell homeostasis. Under inflammatory conditions, this equilibrium may be disrupted, leading to Treg cell dysfunction. This further promotes the onset and progression of disease. Therefore, the functional stability of therapeutic Treg cell products is crucial to their efficacy and safety. Efficiently integrating multiple induction strategies to enhance the stability of iTreg cells under inflammatory conditions is a key area for our continued in-depth research.

Currently, Foxp3 is widely recognized as the decisive transcription factor for the Treg cell lineage and serves as the ultimate target for the vast majority of induction strategies. However, as described in the “Common phenotypes of Treg cells residing in different tissues” section, the absence of Foxp3 expression does not necessarily imply a lack of immunoregulatory function, nor does high Foxp3 expression always equate to potent immunoregulatory capabilities.⁶⁰ Therefore, when inducing Treg cells, we should not solely focus on whether Foxp3 positivity is increased or not. Instead, we should prioritize the immunoregulatory function and long-term stability of the cells. This further requires us to learn how to integrate multiple induction strategies to ensure that the induced Foxp3-positive cells are truly immunoregulatory agents with therapeutic potential.

Implications of differences between human and mouse Treg cells for translating induction strategies

The fundamental mechanisms underlying the various Treg induction strategies discussed in this article are largely derived from mouse models. However, successfully translating these findings into human clinical applications is not a straightforward process. There are significant differences between human and mouse Tregs in terms of development, phenotype, plasticity, and functional stability.⁴⁵⁰ A critical evaluation of these differences is essential to ensure that induction strategies possess genuine clinical potential.

In mice, neuropilin-1 (NRP1)^451,452 and Helios^24,453 are recognized as specific markers for pTreg cells and thymus-derived tTreg cells, respectively. However, in humans, no effective methods have yet been developed to accurately distinguish Treg cell subsets of different origins (e.g., tTreg cells versus pTreg cells).^24,34,454 Treg cells exhibit high heterogeneity, with distinct functional and tissue-specific differences among subsets. Precisely differentiating and regulating specific human Treg cell subsets to adapt to different diseases represents a future research direction. There are also differences in the regulation of Foxp3 expression between mouse and human Treg cells. The Foxp3 gene in mice encodes only one protein. However, the Foxp3 gene in humans encodes two major isoforms through alternative splicing: a longer isoform (FOXP3 FL), which contains all the coding exons, and a shorter isoform lacking the amino acids encoded by exon 2 (FOXP3 ΔE2).^455,456 Research has revealed that patients who express only the FOXP3 ΔE2 subtype failed to maintain self-tolerance, resulting in the development of IPEX syndrome. This demonstrates that the exon 2 region of the human Foxp3 gene plays an indispensable role in sustaining normal Treg function.⁴⁵⁶ This implies that, when applying Treg cell therapy in clinical settings, it is crucial not only to induce stable Foxp3 expression but also to ensure that the induced form is the functionally dominant FOXP3 FL isoform. Furthermore, in mice, Foxp3 is generally considered to be expressed exclusively on Treg cells. However, Foxp3 can be induced in human naïve CD4⁺ Foxp3⁻ T cells by activation.⁴⁵⁷ However, these Foxp3⁺ T cells may lack immunosuppressive function⁴⁵⁷ and readily lose Foxp3 expression,¹⁷⁸ and they may produce proinflammatory cytokines upon stimulation.⁶⁰ Previous studies have found that, in humans, all dividing effector T cells were capable of expressing Foxp3. However, this expression was transient, and the expression levels were significantly lower than those in Treg cells with immunosuppressive functions.⁵⁹ Additionally, Table 4 further shows the differences in key characteristics between human and mouse Treg cells, as well as the implications of these differences for induction strategies.

Table 4.

Comparison of key characteristics between human and mouse Treg cells

Comparison dimensions	Mouse Treg cells	Human Treg cells	Implications for the clinical translation of induction strategies
Foxp3 gene and protein	The Foxp3 gene encodes a single protein product and is a reliable marker of Treg lineage and function.	Alternative splicing generates the major isoforms: Foxp3 FL (full-length, typically associated with stronger function) and FOXP3-ΔE2 (missing exon 2)456	Stricter evaluation criteria need to be established: In humans, it is necessary to clarify whether the induced form is the functionally advantageous isomer, Foxp3 FL.
		Even high expression of Foxp3 is insufficient to define cells as Tregs. Foxp3+ cells obtained by stimulating TCR with anti-CD358 or inducing with TGF-β60 may lack immunosuppressive function.	In humans, Foxp3 expression is not a reliable marker for Tregs, as their functional acquisition mechanisms are more complex than those of mice. When preparing Tregs for adoptive transfer therapy in vitro, product quality cannot be determined based solely on Foxp3 positivity rates; rigorous functional validation must also be conducted. Furthermore, changes in Foxp3+ T cell counts in patients undergoing clinical trials should be interpreted with caution with regard to their clinical significance.
Foxp3 stability	Mouse Treg cells exhibit greater stability than human Treg cells.	Human Treg cells exhibit unstable Foxp3 expression during long-term culture.457	We need to design culture conditions suitable for expanding functional Foxp3+ Treg cells and integrate multiple induction strategies to maintain the long-term functional stability of human Treg cells.
Key cytokine response	The induction and functional maintenance of mouse Treg cells depend on IL-2.	The induction of human Treg cells similarly depends on IL-2, but the optimal concentration and signaling threshold may differ.	Dose optimization studies are required to determine the optimal induction dose at human levels.
	CD4+ CD25− Foxp3− mouse cells, when activated in vitro by TCR stimulation in the presence of TGF-β, can induce Foxp3 expression. The resulting Foxp3+ Treg cells exhibit potent immunosuppressive functions.	Fxop3+ T cells induced by TCR stimulation and TGF-β lack immunoregulatory activity and may even produce proinflammatory cytokines such as IFN-γ.60	Directly applying successful mouse-based protocols to humans may produce cells with similar phenotypes but with functional deficiencies. Therefore, clinically, it is necessary to combine multiple induction strategies to produce iTreg cells that are fully functional.
Metabolic preference	Mouse Treg cells primarily utilize fatty acid oxidation,491,492 with relatively low levels of glycolysis. The activation of TLR1 and TLR2 signaling in mouse Treg cells increases Treg glycolysis and proliferation, while reducing their suppressive capacity.493	Human Treg cells exhibit greater metabolic flexibility and, in the tumor microenvironment, predominantly rely on glycolysis.494 The glucose competition they mediate is a key mechanism that induces senescence in Tconv cells. TLR8 signaling suppresses glycolysis, thereby reversing the immunosuppressive function of Treg cells.494 Notably, inhibiting the mTORC1–HIF-1α pathway reverses the suppressive function of mature human Treg cells.494 However, during the Treg cell induction phase, moderate inhibition of glycolytic activity may favor lineage commitment. For example, compared to mouse cells, reducing glucose and glutamine concentrations in an in vitro culture system when inducing human Tconv cells into iTreg cells significantly increases the proportion and stability of Foxp3+ cells.108	Strategies targeting fatty acid oxidation demonstrate significant efficacy in mice, but may require integration with glucose metabolism regulation in human cells. In humans, reprogramming the metabolism of mature Treg cells by modulating TLR8 signaling and glucose metabolism processes could serve as a novel immunotherapeutic approach for treating autoimmune diseases and tumors.

TCR T cell receptor, TGF-β Transforming growth factor β, IL-2 Interleukin 2, IFN-γ Interferon-γ, mTORC1 Mechanistic target of rapamycin complex 1, TLR Toll-like receptors

Challenges encountered in the Treg cell clinical application

Previously, many studies have focused on the use of Treg cells in immunotherapy, and related therapeutic strategies have moved into the clinical research phase. As our understanding of the function of Treg cells has improved, they have become an increasingly popular treatment for non-immune diseases. While Treg cells have demonstrated promising therapeutic outcomes in various disease models, significant challenges remain in their clinical application. The systemic expansion of Treg cells can lead to systemic non-specific immunosuppression, increasing the risk of infection and tumor development. Therefore, achieving a balance between immunosuppression and immune monitoring in Treg cell-based therapeutic strategies is crucial for success. In addition, expanding Treg cells in vitro is costly and time-consuming, and there is no standardized quality control system for cell products. Therefore, large-scale production is difficult. Currently, there is a lack of universally applicable and reliable biomarkers for monitoring treatment efficacy in clinical practice.³⁸⁷ At the same time, the effectiveness of Treg cell therapy may be significantly impacted by differences in patients’ genetic backgrounds, degrees of disease progression, and immune statuses. Particularly in therapeutic strategies involving the in vitro expansion and subsequent reinfusion of autologous Treg cells, the use of autologous Treg cells faces challenges related to the uncontrollable nature of the initial material. Research revealed that during the in vitro culture of a patient’s autologous Treg cells, variability in the total yield of Treg cells significantly correlated with patient disease status and the number of Treg cells in the patient’s peripheral blood.⁴⁵⁸ This requires the development of tailored induction regimens for each patient. Furthermore, samples obtained from patients during past clinical trials proved difficult to amplify to the minimum cell count required for therapeutic infusion,⁴⁵⁹ presenting one of the primary bottlenecks in personalized medicine. Furthermore, the current regulatory framework is unable to accommodate the personalized medicine model of “one patient, one drug”, and the different requirements imposed by various indications further increase the cost and complexity of the approval process for Treg cell products. These are urgent challenges that must be addressed in the clinical translation of Treg cell therapy.

Optimization strategies for these challenges

Although there are many challenges in Treg cell induction and large-scale clinical applications, researchers are actively exploring optimized solutions as science and technology continue to advance, and our understanding of Treg cell differentiation and function continues to deepen. In order to mitigate the off-target risks associated with the IL-2 cytokine induction strategy, researchers have developed anti-cytokine antibodies and cytokine muteins to facilitate targeted proliferation and selective activation. These advances have significantly enhanced the safety and efficacy of the cytokine induction strategy.^159,165,169 With the innovation of single-cell sequencing, multi-omics analysis, gene editing, and other technologies, it has become possible to accurately target and regulate the differentiation and function of Treg cells. Aleksandar Obradovic et al. identified TI-Tregs as master regulators (MRs), which were mechanistic determinants of the transcriptional status of TI-Tregs, by emerging means such as multi-omics analysis, machine learning, and CRISPR-Cas9. They also used high-throughput drug screening technology to screen gemcitabine as a specific inhibitor of highly MR-active TI-Treg cells (e.g., Helios⁺ CD103⁺ subpopulation), which specifically reduced the number of TI-Treg cells in tumor tissues without significant effects on pTreg cells.⁴⁶⁰ The development and maintenance of the immunosuppressive function of Treg cells largely depend on stable and coordinated Foxp3 expression and that of other Treg cell signature genes.³⁴ Most genes determining Treg cell lineage and stability are directly dependent on Foxp3.⁴⁶¹ Therefore, it is possible to maintain Treg cell stability through epigenetic regulation of the Foxp3 locus and the regulation of the three-dimensional spatial structure of the Treg cell genome.³⁴ Novel delivery strategies and engineered Treg cells have also brought new ideas for Treg cell therapy; for example, an increasing number of studies are focusing on the use of nano-delivery systems to selectively regulate Treg cells to achieve precise regulation of the immune system,^462–464 while the development of CAR technology provides a new direction for the maintenance of a balance between immunosuppression and immunosurveillance based on Treg cells. In addition, the proposed individualized therapeutic strategy from “generic” to “customized” and the establishment of a standardized production system will accelerate the clinical translation of Treg cell therapy. Above all, integrating multiple induction strategies in an organic way to obtain phenotypically and functionally stable Treg cells is essential for overcoming the bottlenecks in the clinical application of Treg cell therapy. Future research should focus on exploring the synergistic effects, temporal logic, and modular application of these strategies further. At the same time, species specificity should be incorporated as a core parameter in the design of integrated strategies. Validating induction protocols using humanized mouse models or organoids will mitigate the risks of clinical translation, thereby establishing a more robust framework for Treg cell therapy. (Fig. 9)

Fig. 9.

Challenges encountered during Treg cell induction, clinical applications, and strategies to address them. The middle figure summarizes the common strategies currently used to induce Treg cells, as well as the important applications of Treg cell therapy. The left side lists the current challenges encountered during Treg cell induction, along with possible optimization strategies. The right side lists the current challenges encountered during the clinical application of Treg cell therapy, along with possible optimization strategies. The graphic is created with BioRender.com

In the future, Treg cell therapy is expected to shift from single-disease intervention to systemic immune remodeling. Despite these challenges, its enormous potential in disease treatment will be explored one by one, and a leap from basic research to clinical application will be realized.

Author contributions

Conceptualization: Yingying Shi. Literature collection: Yaqi Chen. Writing- Original Draft: Yaqi Chen. Writing- Review & Editing: Yingying Shi, Jian You, and Yaqi Chen. Visualization: Yaqi Chen. Supervision: Yingying Shi and Jian You. Funding Acquisition: Jian You and Yingying Shi. All authors have read and approved the article.

Competing interests

The authors declare no competing interests.