|Home | About | Journals | Submit | Contact Us | Français|
Naturally occurring regulatory T cells (Treg) are T cells with pivotal immunomodulatory capabilities. Treg cells express CD4, high levels of the IL-2 receptor CD25, and the Treg-specific transcription factor Foxp3, that was previously shown to be sufficient to confer immunosuppressive ability upon introduction into non-regulatory T cells (1, 2). Natural Treg develop in the thymus and are subsequently exported in the periphery. Induced Treg develop in the periphery from naive CD4+ T cells following T cell receptor (TCR) triggering in the presence of TGFβ signaling (3). Treg were first described in murine models for human disease following the observation that depletion of CD25+ cells lead to many different autoimmune conditions, while reconstitution with CD4+CD25+ cells restored tolerance in the recipient animals (4). It is now apparent that Treg cells are involved in the control of many adaptive immune processes, either directed against self or against pathogenic agents (5). Indeed, mutations affecting the expression and/or function of Foxp3 result in a number of autoimmune dysfunctions in both human patients (6) and mice (7).
More recently, Treg have been used to prevent graft rejection in different transplantation settings, and have shown promising results in murine models for solid organ as well as hematopoietic cell transplantation (HCT) (8–17). Allogeneic HCT is an effective treatment for many hematopoietic diseases such as leukemia, lymphoma and myelodysplastic syndromes. Donor derived bone marrow cells or G-CSF mobilized peripheral blood stem cells are infused in the patient following a chemotherapy and/or radiation based conditioning regimen, which is aimed to both target malignant cells and to create a temporary state of immune suppression in order to minimize graft rejection. The donor-derived cells are enriched with hematopoietic progenitors, and also contain a percentage of mature T cells. The latter serve multiple beneficial purposes, such as sustaining the engraftment process, replenishing the adaptive immune system to protect against infection, and recognizing and eliminating residual malignant host cells (named graft versus tumor effect, GVT).
A major complication of HCT is graft-versus-host disease (GVHD), where alloreactive donor-derived T cells infiltrate and injure target organs such as the liver, gut and skin. While immunomodulation is necessary to control the adverse GVHD reaction, an effective immune response is required for successful tumor eradication. The traditional approach to GVHD prevention and/or treatment is immunosuppressive therapy, which predisposes the transplanted patients to high risk of infections, impairs the T cell mediated antitumor response, and may interfere with the physiological growth of younger patients. Thus, it is of crucial importance to uncover alternative therapeutic strategies to be coupled to allogeneic HCT for improved control of GVHD and maintenance of GVT.
Recently, our and other laboratories have shown that freshly isolated natural Treg as well as in vitro expanded Treg cells can prevent lethal GVHD induced by donor-derived conventional T cells (CD4+ and CD8+ T cells; Tcon) following allogeneic bone marrow transplantation (BMT), in multiple models across both major and minor histocompatibility barriers (11–17). Furthermore, when cotransferred into recipient mice with established leukemia or lymphoma, Treg cells were shown to suppress Tcon cell proliferation and prevent lethal GVHD, while preserving GVT activity (11, 16, 17).
In vivo bioluminescent imaging (18) has been utilized in order to visualize the patterns of homing and proliferation of luciferase+ Treg and Tcon in vivo. These studies demonstrated that both Treg (19) and Tcon (20) cells must first traffic to nodal sites to become activated, proliferate and exploit their biological function. Notably, GVHD suppression is restricted to the Treg fraction expressing the lymphoid homing molecule CD62L+ (l-selectin), while CD62L− Treg largely lacked protective capacity (21, 22). CD62L expression was associated with accumulation of Treg cells in the spleen and lymph nodes of the transplanted mice, suggesting that Treg-mediated suppression of Tcon proliferation occurs at the priming sites of alloreactive Tcon. However, Treg seem to also exert their suppressive action in situ by infiltrating GVHD target organs, since Treg lacking the chemokine receptor CCR5 (whose ligands are expressed at the sites of inflammation during ongoing GVHD) were shown to be far less protective than wild type Treg (23).
Previous work has demonstrated the critical role of recipient APCs for GVHD induction. Antigen presenting cells (APCs) are critical for Treg activation and suppressive activity, as Treg cells need to be of donor origin in order to confer survival benefit, while host-derived Treg lack protective potential (13) in a major mismatch model. Indeed, several in vitro studies have unveiled the importance of APC-Treg interaction for Treg function (24–26). Mechanistically, natural Treg-mediated protection was shown to be partly dependent on IL-10 production in a major mismatch model (13). Additionally, another important effector molecule for Treg activity has been identified as the TNF-R family member CD30 (27). CD30 upregulation occurs following allogenic stimulation of Treg in vitro and in vivo (27), and following T cell activation in general. Notably, CD30 knockout Treg cells were shown to be far less potent in inhibiting GVHD compared to their wild type counterparts (27). The ligand for CD30 (CD30L) is expressed on thymic epithelial cells (TECs), APCs, activated T cells, neutrophils, eosinophils, and resting B cells. Early blockade of the CD30/CD30L axis following HCT did not affect Treg trafficking to lymphoid organs, but impaired their expansion and was associated with decreased Tcon apoptosis, suggesting that CD30 signaling on Treg is important for proliferation and/or survival, and that the impact of Treg cells on Tcon-cell expansion following BMT is at least in part mediated by induction of allogeneic T cell apoptosis. Late CD30L blockade did not ameliorate GVHD, indicating that GVHD suppression requires an intervention during the early phases of Tcon activation occurring at the priming sites. CD101 can serve as an additional useful marker for the purification of murine Treg cells with the highest suppressive potential for GVHD suppression in vivo (28), although the specific function and the identity of the CD101 ligand are currently unknown.
Treg represent only a small fraction of peripheral CD4+ T cells (5–10 %), therefore, the potential clinical use of regulatory T cells is limited by their low number. To achieve higher numbers of Treg cells that may be required for clinical applications, several different expansion strategies have been employed.
One strategy takes advantage of the fact that IL-2 is a critical growth factor for Treg, yet they do not produce this cytokine (29). Treg survival depends entirely on the exogenous addition of IL-2 (30, 31). Also important for Treg expansion and survival is signaling through CD3 and CD28 receptors (32, 33). The combination of high dose IL-2 and anti-CD3/CD28 coated microbeads is one of the most common approaches to expand Treg ex vivo, resulting in a pool of Treg that maintain a diversified repertoire (polyclonal Treg) (14, 34–37). This protocol was used by different research groups to expand both murine and human Treg with an expansion efficiency from ten to hundreds fold (34, 35, 38, 39). Expanded Treg can be used in vivo to reduce GVHD where CD4+CD25+ cells expanded ex vivo in the presence of immobilized anti-CD3 antibodies and 100U/ml IL-2, when administered in an animal model of GVHD resulting in an increase in the median survival of mice from 10 days in control mice to 72 days (14).
The purity of the starting Treg pool is of high concern during the expansion protocol, as even a small amount of contaminating effector cells can be expanded under these culturing conditions and, eventually, outgrows the Treg population (40). Several studies have shown the addition to the culture of rapamycin, a well-known immunosuppressant, suppresses effector T cell proliferation conferring a selective advantage on Treg proliferation (41–43). Other studies have suggested that by starting the expansion with the CD45RA+ fraction of CD4+CD25+ cells or the CD62L+ fraction, a homogeneous Treg population can be obtained (44, 45). In addition, Shevach’s group identified a unique set of cell surface markers (CD121a/CD121b and LAP) expressed only on Foxp3+ Treg that can be used to separate the expanded Foxp3+ Treg cells from the non-Foxp3+ cells (46).
Antigen-specific Treg expansion using peptide pulsed DC (47) or tetramer sorting (48) is a more technically challenging strategy than the polyclonal expansion, but presumably will allow for a decreased number of Treg needed for therapy. Bluestone’s group was able to expand an islet peptide-mimic reactive Treg from a polyclonal population of NOD Treg by replacing the anti-CD3 mAb on coated beads with recombinant MHC class II presenting an islet peptide-mimic (49). Alternatively, antigen-specific Treg capable of recognizing allogeneic antigens can be obtained by culturing purified Treg in the presence of recipient type APC and IL-2 (15) (50, 51). Using this strategy, Trenado et al. were able to generate Treg that were more efficient in providing protection from GVHD than the polyclonal expanded Treg (52).
Recent studies have indicated a direct correlation between the DNA methylation/demethylation of the CpG motif within the 5’ untranslated region of the foxp3 promoter and the stability of Foxp3 expression (53). Thus, natural Treg are completely demethylated at this evolutionarily conserved region called “Treg -specific demethylated region” (TSDR). In contrast, Treg generated ex vivo from naïve CD4+CD25− cells by TCR stimulation in the presence TGF-β and IL-2 are only partially demethylated at the TSDR region suggesting a transient Foxp3 expression and, subsequently, a loss of Foxp3 expression following TGF-β withdrawal (53). Based on these results, Choi et al. designed a novel Treg expansion strategy in which hypomethylating agents, decitabine and azacitidine, were used to generate functional Foxp3+ Treg from CD4+CD25− T cells (54). Decitabine treatment of D4+CD25− T cells stimulated with anti-CD3/CD28 beads and low dose IL-2 induced Foxp3 expression in 60–70% of cells and Foxp3 expression was maintained even 7 days after the decitabine treatment (54).
The molecular basis of Treg cell suppressive activity is still unclear. A better understanding of Treg biology will allow the optimization of innovative pre-clinical protocols for the enhancement of Treg immunomodulatory activity in vivo, while preserving the Tconv-mediated GVT effect following allogeneic HCT. The ultimate goal is to translate the insights obtained from these pre-clinical studies into novel therapeutic strategies, with the idea of minimizing the use of immunosuppressive drugs following HCT. Indeed, the cotransfer of Treg cells into lethally-irradiated mice, recipient of T cell-depleted bone marrow and Tconv cells, resulted in enhanced immune reconstitution and increased survival following challenge with viral pathogens, without interfering with the Tconv-mediated tumor clearance. These pre-clinical models have been adapted to ongoing clinical trials assessing the efficacy of human Treg in the clinic (55, 56) (Negrin, RS unpublished results). Brunstein et al. recently reported the results of the first clinical trial with umbilical cord blood (UCB)-derived in vitro expanded Treg in patients receiving a nonmyeloablative double UCB transplant (55). Although Treg infusion (up to 30 × 105/kg on days +1 and +15) mildly delayed but did not significantly ameliorate GVHD compared to historical controls, Treg immunotherapy proved to be safe, increased the number of circulating CD4+Foxp3+ cells, and facilitated mixed chimerism. It remains possible that prolonged in vitro expansion partly affected the suppressive potential of the cultured Treg. Thus, it will be important to identify the optimal culture conditions as wells as the maximum tolerated dose of in vitro expanded Treg in human patients following transplantation, in order to plan the future studies that will allow a thorough evaluation of the true impact of Treg engraftment and GVHD suppression.
Furthermore, the results of a phase I/II clinical trial evaluating the impact of freshly-isolated Treg cells, on GVHD prevention and immunological reconstitution was recently reported by Martelli’s group in Perugia, Italy (56). Following a conditioning regiment including total body irradiation and chemotherapy, 22 patients received CD4+CD25+ freshly isolated donor Treg three days before administration of highly purified CD34+ cells together with mature Tconv, at a Treg:Tconv ratio of 1:1.5. Notably, no post-transplant GVHD prophylaxis was used in this study. Although some patients succumbed to transplant-related opportunistic infections (6/22), most patients engrafted (20/22) and displayed long lasting full donor-type chimerism. Strikingly, Treg immunotherapy not only improved immune recovery, but most patient did not present any manifestation of GVHD following Tconv administration (17/20), while 2 patients developed mild grade I cutaneous GVHD, and only 1 patient developed grade III GVHD. Thus, Treg immunotherapy rendered the administration of high dose of Tconv tolerated for the first time reported in the setting of haploidentical transplantation, providing sustained protection from GVHD and improved immune reconstitution. Another approach has been to expand Treg in vivo utilizing differences between conventional and regulatory T cell biology. The first such attempt combined CD4+ cell donor lymphocyte infusions with low dose IL-2, which resulted in a preferential expansion of Treg in patients (57). This strategy and others may result in the selective proliferation of Treg in certain clinical situations, for example, chronic GVHD.
If successful, further understanding of Treg biology may be extended to additional clinical applications, such as tolerance induction in the context of severe autoimmune diseases and/or solid organ transplantation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.