Regulatory T (Treg) cells are being used to treat autoimmunity and prevent organ rejection; however, Treg cell-based therapies have been hampered by the technical limitation in obtaining a high number of functional Treg cells. In this study, we show how to generate functional Treg cells from induced pluripotent stem (iPS) cells and to determine the potential role of such cells for Treg cell-based immunotherapy against autoimmunity in a therapeutic setting. Ligation of a Notch ligand and transduction of the gene Foxp3 induce iPS cells to differentiate into Treg cells. Expression of Foxp3 and coculture on Notch ligand-expressing stromal cells augment expression of CD3, TCR, CD4, CD25, and CTLA-4 on iPS cell-differentiated Treg cells, which are able to secrete TGF-β and IL-10 both in vivo and in vitro. Importantly, adoptive transfer of iPS cell-derived Treg cells expressing large amounts of Foxp3 and Bcl-xL significantly suppresses host immune responses and reduces arthritis development within murine models. These data suggest that Notch signaling and Foxp3 regulate the development and function of Treg cells derived from iPS cells. Our results provide a novel approach for generating potentially therapeutic Treg cells for the treatment of autoimmune diseases.

Regulatory T (Treg) cells, formerly named as suppressor T cells, are one subtype of T cells, which are crucial for the maintenance of immunological tolerance. Treg cells can shut down cell-mediated immunity toward the end of an immune reaction and can also suppress autoreactive immune cells. There are two major classes of CD4+ Treg cells, including the naturally occurring Treg (nTreg) cells and the induced or adaptive Treg cells. nTreg cells are also known as CD4+CD25+Foxp3+ Treg cells, which arise in the thymus, and induced Treg cells are known as Tr1 or Th3 cells, which arise from CD4+CD25+Foxp3 precursor cells (1). Several proteins have been determined as possible surface markers for Treg cells, including CD25, CD45RB, CD62L, CD103, CD95 (Fas), MHC class II, CD127, neuropilin-1, lymphocyte activation gene-3, CTLA-4, glucocorticoid-induced TNF receptor, and Foxp3 (24). The latest research suggests that Treg cells are defined by expression of Foxp3, but a clear defining marker for Treg cells remains to be identified.

Despite the fact that the mechanisms underlying Treg cell development are not fully understood, advances have been made regarding Treg cell-based therapies. Adoptive cell transfer (ACT) of Treg cells has been shown to be a promising treatment for autoimmune diseases in the experimental setting, for example, rheumatoid arthritis, systemic lupus erythematosus, and type 1 diabetes (57). Recently, it has become desirable to isolate a larger Treg cell subset for exogenous expansion and adoptive transfer, because the intrinsic resistance of Treg cells to exogenous expansion and a high number of Treg cells are required to perform the ACT-based immunotherapy. Unfortunately, because of the lack of specific surface markers, purification of a more comprehensive Treg cell subset raises the contamination by the nonregulatory effector T cell population. Regardless of an increasing number of protocols for isolating subsets of Treg cells, no approach to date has confirmed the capacity to isolate the entire Treg cell population with 100% specificity. Prominently, in autoimmune-prone mice or humans, contamination of therapeutic Treg cells by effector T cells can contain pathologic autoreactive immune cells. These immune cells may have potential to aggravate autoimmune diseases when expanded and transferred back into the donor. Therefore, ACT with a large number of Treg cells is often not feasible owing to difficulties in generating these cells from patients.

In this study, we show a novel approach to generate a high number of functional Treg cells from induced pluripotent stem (iPS) cells for the treatments of autoimmune diseases. Genetic modification of iPS cells with the Foxp3 gene and stimulation with in vitro Notch ligand direct iPS cell differentiation into Treg cells, which are able to produce suppressive cytokines and inhibit other immune cell activities. Additionally, adoptive transfer of iPS cell-derived Treg cells suppresses arthritis development in murine models. Importantly, adoptive transfer of allogeneic Treg cells from iPS cells in mismatched MHC suppresses the recipients’ autoimmunity. Our data demonstrate that Notch signaling and Foxp3 regulate the development and function of Treg cells derived from iPS cells, and we identify a novel approach for generating potentially therapeutic Treg cells by reprogramming iPS cells.

The mouse iPS-MEF-Ng-20D-17 cell line, which was induced from mouse embryonic fibroblasts by retroviral transfection of Oct3/4, Sox2, Klf4, and c-Myc (8), was obtained from Dr. Shinya Yamanaka (Institute for Frontier Medical Sciences, Kyoto University, Kyoto, Japan) through the RIKEN Cell Bank (Ibaraki, Japan). OP9-DL1 cells were obtained from Dr. J.C. Zuniga Pflucker (University of Toronto, Toronto, ON, Canada) and expressed murine MHC class II protein I-Ab by gene transduction (9). C57BL/6, B6.129S7-Rag1tm1Mom (Rag1-deficient), and DAB/1 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All experiments were done in compliance with the regulations of The Pennsylvania State University College of Medicine Animal Care Committee in accordance with guidelines by the Association for the Assessment and Accreditation of Laboratory Animal Care.

All iPS cells were maintained on feeder layers of irradiated SNL76/7 cells in six-well culture plates (Nunc) as previously described, and were passaged every 3 d (10). Gene-transduced iPS cells were washed once in OP9-DL1 medium before plating onto subconfluent OP9-DL1 monolayers for Treg lineage differentiation in the presence of 5 ng/ml murine Flt3L and 1 ng/ml murine IL-7 (PeproTech, Rocky Hill, NJ).

cDNA for Foxp3 or Foxp3 with Bcl-xL was subcloned into the murine bicistronic retroviral expression vector MiDR, and cloning was confirmed by PCR amplification and gene sequencing. Retroviral transduction was performed as described previously (11, 12). Expression of DsRed was determined by fluorescent microscopy as well as flow cytometry gating on GFP+ cells. DsRed+GFP+ cells were purified by cell sorting using a MoFlo high-performance cell sorter (DakoCytomation, Fort Collins, CO).

PE- or allophycocyanin-conjugated anti-mouse IL-2 (JES6-5H4) and IFN-γ (XMG1.2) were obtained from BD Pharmingen (San Diego, CA). PE-, PE/Cy7-, or allophycocyanin-conjugated anti-mouse TCRβ, CD3, CD25, CD127 CTLA-4, IL-10, and LAP (TGF-β1) and allophycocyanin/Cy7- or PerCP-conjuaged anti-mouse CD69 were obtained from BioLegend (San Diego, CA). FITC- or PE-conjugated anti-mouse CD8 (6A242) was obtained from Santa Cruz Biotechnology (Santa Cruz, CA).

Thymectomy was performed on anesthetized mice 4–12 wk of age as previously described (13). Control sham-thymectomized mice underwent the entire procedure except the final removal of the thymus. Total thymectomy was confirmed for all of the thymectomized mice at the time of sacrifice by inspection of the thorax.

Adult thymus lobes were dissected and cultured on sponge-supported filter membranes (Gel Foam surgical sponges; Amersham Pharmacia, Piscataway, NJ) at an interphase as described previously (14). Adult thymic organ cultures were treated with 1.1 mM 2-deoxyguanosine (Sigma-Aldrich) and reconstituted with gene-transduced iPS cells and incubated as previously described (15).

DsRed+ Treg progenitors (3 × 106) from iPS cells in PBS were injected i.v. into 4-wk-old Rag1-deficient, C57BL/6, or DBA/1 mice. After 4–6 wk, Treg cell development in the lymph nodes, spleen, and peripheral blood was determined by flow cytometry.

Gene-transduced iPS cells were cocultured with OP9-DL1 cells for various days, and the expression of CD3, TCRβ, CD4, CD25, CD127, and CTLA-4 was analyzed by flow cytometry after gating on DsRed+ cells or other markers such as CD3 or TCR. iPS cell-derived Treg cells from the culture were stimulated with plate-coated anti-CD3 plus soluble anti-CD28 mAbs, intracellular TGF-β, and IL-10 as well as IL-2 and IFN-γ were analyzed by flow cytometry after gating on live CD4+CD25+ cells.

Male C57BL/6 or DBA/1 mice (>4 mo of age) were injected at the base of the tail with 0.1 ml emulsion containing 100 μg chicken or bovine type II collagen (CII) (Chondrex, Redmond, WA) in CFA (Chondrex) using a 1-ml glass tuberculin syringe with a 26-guage needle. Mice were assessed for clinical arthritis in the paws as described before (11).

Cytokines were measured by ELISA as before (16). Proliferation was measured in triplicate cultures by incorporation of [3H]thymidine (1 μCi/well; ICN Pharmaceuticals) during the last 12 h culture.

Live cells were lysed in ice-cold RIPA lysis buffer for 30 min. Insoluble material was removed and lysates were used for immunoblot; protein content was determined using the Bio-Rad protein assay kit (Bio-Rad). Equal amounts of protein (30–50 μg) were loaded onto 4–12% NuPAGE Bis-Tris precasting gels (SDS-PAGE), transferred onto polyvinylidene difluoride membranes (Invitrogen), and immunoblotted. All immunoblots were developed with the ECL immunodetection system (Amersham Pharmacia Biotech).

Mice were sacrificed on day 60 after the CII challenge. Hind foot paws were amputated, fixed in 10% formalin, and decalcified in Formical-4 (Decal Chemical, Tallman, NY). The tissues were embedded in paraffin, sectioned at 4 μm, and stained with H&E or Safranin O-fast green as described before (11). H&E staining for bone erosions was scored using a semiquantitative scoring system from 0 to 4 (0, no erosions; 4, extended erosions and destruction of bone). Safranin O staining for the loss of proteoglycans was scored with a semiquantitative scoring system (0–3), where 0 represents no loss of proteoglycans and 3 indicates complete loss of staining for proteoglycans.

Immunofluorescent staining was performed on the sections after deparaffinization and rehydration using xylene and ethanol. Endogenous peroxidase activity was blocked in 3% hydrogen peroxide for 3 min after Ag retrieval. Slides were blocked for nonspecific binding in 3% BSA in PBS at room temperature in a moist chamber for 60 min. FITC-labeled anti-mouse CD4 (BD Pharmingen) and PE anti-mouse Foxp3 (BioLegend) or FITC-labeled anti-mouse CD25 (BD Pharmingen) and PE anti-mouse Foxp3 were diluted (1:2) in blocking solution and applied on the section and incubated for 2 h at room temperature in moist chamber. Sides were washed five times in PBS for 5 min. The stained slides were counterstained for nuclear staining with antifade reagent with DAPI present in mount media (Enzo Life Sciences) and then stored in the dark at 4°C until analysis under fluorescent microscope. The same staining procedure with relevant isotype Ab (mouse) was performed as negative control.

Blood samples were collected from the orbital sinus or by heart puncture on day 60 after primary immunization with CII. Total IgG was measured using the Easy-Titer IgG assay kit (Pierce, Rockford, IL) in accordance with the recommendations of the manufacturer. The levels of anti-CII IgG in these sera were measured by ELISA as described previously (11).

One-way ANOVA was used for the statistical analysis between groups and significance was set at 5%. All statistics were calculated using GraphPad Prism (GraphPad Software, San Diego, CA).

The development of natural CD4+CD25+Foxp3+ Treg cells occurs within the thymus and includes a series of positive and negative selection processes that are, in large part, still poorly understood. To determine whether iPS cells have the ability to differentiate into Treg cells, we used retrovirus-mediated transduction of Foxp3 into iPS cells followed by coculture in adult thymic organ cultures. We used the mouse stem cell virus-based retroviral vector pMig to generate Foxp3-transduced iPS cells. Because iPS cells express GFP as an endogenous marker, we replaced GFP of pMig with DsRed as a new marker for monitoring gene integration and named the resulting vector MiDR. Upon gene transduction, DsRed expression was visualized by fluorescent microscopy (Fig. 1A), and DsRed+GFP+ cells were sorted (Fig. 1B). We confirmed Foxp3 expression in the sorted cells by Western blot and flow cytometry (Fig. 1C). We cocultured the Foxp3 gene-transduced iPS cells on feeder layers of irradiated SNL76/7 cells as described previously (10), and we generated a large number of Foxp3+ iPS cells (DsRed+GFP+).

The adult thymic organ culture system has been used to study T cell development ex vivo (17). Coculture of Foxp3-transduced iPS cells in adult thymic organ cultures for 2 wk resulted in the generation of CD4+CD25+Foxp3+ Treg cells (Fig. 1D). Moreover, the cells had the ability to produce suppressive cytokines (Fig. 1E), suggesting that the differentiated cells were functionally mature Treg cells. In a separate set of experiments, Foxp3-transduced iPS cells were cocultured on OP9-DL1 cells for 7 d and adoptively transferred into thymectomized Rag1−/− mice that had no mature T cells or B cells. Six weeks later, we also observed CD4+CD25+Foxp3+ Treg cells in the lymph nodes and spleen (Fig. 1F), and the cells had the ability to produce TGF-β and IL-10 (Fig. 1G). Thus, Foxp3-tranduced iPS cells are capable of differentiating into functional Treg cells both in vitro and in vivo.

Notch signaling is essential for the initial commitment to the T cell lineage and surface expression of CD25 (18). Previously, we showed T lineage differentiation from iPS cells by coculturing with OP-DL1 cells, which are OP9 cells expressing the Notch ligand Delta-like 1 (19). We hypothesize that iPS cells that are transduced with the Foxp3 gene and stimulated with the Notch ligand are capable of differentiating into functional Treg cells. We cocultured the Foxp3-transduced iPS cells on OP9-DL1 stromal cells in the presence of murine recombinant Flt3L and IL-7. After 7 d in culture, the Foxp3 gene-transduced iPS cells differentiated and became the lymphocyte-like cells on day 14, which were associated with nonadherent multicellular clusters. Interestingly, on day 30, the lymphocyte-like cells spread fully across the culture dish with an ∼1000-fold increase in cell number (Fig. 2A). These observations were described previously (10). These data also agree with studies performed with embryonic stem (ES) cells and hematopoietic stem cells (HSCs) in the in vitro coculture system (19, 20).

The iPS cell-differentiated cells were analyzed for the cell surface markers of Treg cells. On day 14 of in vitro coculture, CD3 and TCR, two markers of T lymphocytes, were significantly expressed as analyzed by flow cytometry. The CD3+TCRβ+ populations expressed CD4 and CD8, which correlated with previous studies (21). Most of the CD3+TCRβ+CD4+ cells (>75 ± 6%) expressed CD25 and CTLA-4 but not CD127 (Fig. 2B). We also determined that Foxp3 expression on the iPS cell-derived T lymphocytes persisted even after long-term in vitro stimulation with the Notch ligand as detected by intracellular staining analyzed by flow cytometry (Fig. 2B) and Western blot (Fig. 2C). Additionally, similar to our previous observation that Foxp3 and Bcl-xL can cooperatively promote Treg cell development (11), the Foxp3 expression in the Foxp3 and Bcl-xL–transduced group was much higher compared with the Foxp3-tranduced group (Supplemental Fig. 1). Collectively, these data suggest that CD4+CD25+Foxp3+ Treg cells are generated from iPS cells after Foxp3 gene transduction and in vitro stimulation with the Notch ligand.

Although the molecular mechanisms by which Treg cells exert regulatory activity have not been fully characterized, the immunosuppressive cytokines TGF-β and IL-10 have been implicated in Treg cell activity (22, 23). To determine the functional status of iPS cell-derived Treg cells, we examined whether these cells had the capacity to secrete these suppressive cytokines following stimulation. On day 45 of in vitro coculture, cells were isolated and stimulated with immobilized anti-CD3/anti-CD28 mAbs, and the expression of IL-2, IFN-γ, TGF-β, and IL-10 was assessed. Intracellular TGF-β and IL-10 were detected in ∼36% of stimulated CD4+CD25+ cells (Fig. 3A), but the inflammatory cytokines IL-2 and INF-γ were not detected (Fig. 3B), indicating that iPS cell-derived Treg cells have suppressive activities.

To investigate the functional status of iPS cell-derived Treg cells beyond their ability to secrete TGF-β and IL-10, we performed an in vitro suppressive assay. The iPS cell-derived Treg cells or nTreg cells (effector cells) were cocultured with naive CD4+ T cells (target cells) from C57BL/6 mice (Treg cells/T cells, 1:1) in the presence of anti-CD3/anti-CD28 mAbs for 2 d. Supernatants from target cells stimulated with iPS cell-derived Treg cells or nTreg cells showed a significant decrease in the amounts of IL-2 and IFN-γ as compared with target cells alone (Fig. 3C). In a separate set of experiments, effector cells significantly suppressed the proliferation of target cells after CD3/CD28 stimulation (Fig. 3D). Importantly, there was no significant difference at suppression in cytokine secretion and proliferation between iPS cell-derived Treg cells and nTreg cells (Fig. 3C, 3D). Taken together, these results show that in vitro-derived Treg cells from iPS cells are functionally mature.

Treg cells have been shown to play a critical role in controlling autoimmune diseases, and previously we demonstrated that adoptive transfer of Treg cells derived from naive T cells transduced with genes encoding Foxp3 and Bcl-xL suppressed arthritis development (11). Therefore, we investigated whether iPS cell-derived Treg cells could be used to control autoimmune diseases.

First, we tested whether adoptive transfer of iPS cell-derived Treg cells had the ability to suppress arthritis development. Foxp3 or Foxp3 plus Bcl-xL–transduced iPS cells were cocultured on OP9-DL1 cells for 7 d, and DsRed+ cells were sorted and adoptively transferred into male C57BL/6 mice. Fifteen days prior to adoptive transfer (day 0), collagen-induced arthritis (CIA) was induced in the recipient C57BL/6 mice by one intradermal immunization at two sites in the base and slightly above the tail with chicken CII in CFA as described previously (24). Arthritis incidence (Fig. 4A) and clinical score (Fig. 4B) were assessed in the paws. Mice receiving Foxp3 plus Bcl-xL–transduced Treg cells had a decreased incidence of arthritis as compared with mice receiving Treg cells transduced with Foxp3 alone (15 versus 32% on day 60; Fig. 4A). Most importantly, mice receiving Foxp3 plus Bcl-xL–transduced Treg cells had a lower clinical score than did those receiving Treg cells transduced with Foxp3 alone (a score of 0.6 versus 1.3 on day 60; Fig. 4B). However, adoptive transfer of Foxp3- or Foxp3 plus Bcl-xL–transduced Treg cells reduced arthritis incidence and clinical score.

We also examined whether disease severity correlated with Ab production and histological observations of the joints. We found that mice receiving iPS cell-derived Treg cells significantly decreased anti-CII IgG2a in the serum (Fig. 4C). These results correlate with our previous study that anti-CII IgG2a is the major component of anti-CII IgG on day 60 (11). The preventive effects of Treg cell transfer on CIA were also confirmed by histological analyses. In the control mice, there were significant signs of arthritic pathology as indicted by destruction of cartilage with fibrosis and leukocyte infiltration around bones and in the joint space on day 60 (Fig. 4D, 4E). In contrast, ACT of iPS cell-derived Treg cells resulted in a marked suppression of these symptoms characteristic of CIA. Additionally, there were only a few foci of leukocyte infiltration and less destruction of bones and joints in mice that received iPS cell-derived Treg cells (Fig. 4F, 4G).

Moreover, we observed a large number of infiltrating Treg cells in the joints of mice receiving iPS cell-derived Treg cells by immunofluorecent histology (Fig. 5). In these experiments, we adoptively transferred one group of mice with nTreg cells (2.5 × 106/mouse) as positive control (25). We obtained similar results from the group of mice with Foxp3-transduced Treg cells. Collectively, these findings show that adoptive transfer of iPS cell-derived Treg cells can effectively control the development of arthritis.

Adoptive transfer of allogeneic Treg cells with mismatched MHC has been shown to suppress autoimmunity (2628). To determine the potential application of allogeneic Treg cells derived from iPS cells in adoptive immunotherapy in physiological settings, we used the animal protocol wherein murine arthritis was induced in DBA/1 mice by bovine CII (11). In the first set of experiments, we tested whether adoptive transfer of allogeneic iPS cell-derived Treg cells had the ability to suppress arthritis development. By using the scoring system as described previously (11), we found that adoptive transfers of allogeneic iPS cell-derived Treg cells significantly reduced arthritis incidence and clinical score (p < 0.05). Similarly, as described in MHC-matched recipients (Fig. 4), mice receiving Foxp3 plus Bcl-xL–transduced Treg cells had decreased incidences and lower clinical scores of arthritis as compared with mice receiving Treg cells transduced with Foxp3 alone (Fig. 6A, 6B). Adoptive transfer of Foxp3- or Foxp3 plus Bcl-xL–transduced Treg cells reduced arthritis incidence and clinical score.

In the second set of experiments, we determined the antipathogen immune responses and performed immunohistological analysis. Mice receiving allogeneic iPS cell-derived Treg cells dramatically decreased anti-CII IgG2a Ab in the serum (Fig. 6C) and largely reduced signs of arthritic pathology around bones and in the joint space on day 60 (Fig. 6D–G). Importantly, we visualized a large number of infiltrating Treg cells in mouse joints in mice receiving allogeneic iPS cell-derived Treg cells (Fig. 7). Taken together, these results suggest that allogeneic iPS cell-derived Treg cells effectively suppress autoreactive immune cells and control the development of autoimmunity.

In this study, we show the generation of functional Treg cells by reprogramming iPS cells. Additionally, we demonstrate that transduction of the Foxp3 gene and stimulation with the Notch ligand DL1 are sufficient to induce functional Treg cells from iPS cells, suggesting that Notch signaling and Foxp3 regulate the development and function of Treg cells derived from iPS cells. Importantly, we develop a more efficient approach for generating potentially therapeutic Treg cells.

Adoptive transfer of Treg cells has garnered wide attention, but their effective use is limited by the need for multiple ex vivo manipulations and infusions that are complex and expensive. Thus, the generation of Treg cells by reprogramming stem cells for ACT-based immunotherapy has been proposed in the treatments of autoimmune diseases. The use of iPS cells can bypass ethical and feasibility issues related to the use of ES cells and HSCs. For example, the approach to obtain ES cells from patients is not possible. Although the use of HSCs for therapeutic purposes has been widely applied clinically, especially in HSC transplantations (29), HSCs are similar to other adult stem cells, which can proliferate only for a limited number of cycles. Therefore, the response of HSCs to differentiation signals declines with each cycle. Thus, compared with ES cells and iPS cells, HSCs have reduced differentiation and proliferative capacities, making them difficult to expand in cell culture (30). Identification of approaches that successfully expand HSCs in vivo or in vitro would indeed be a huge boost to all present and future medical uses of HSCs. iPS cell technology continues to progress rapidly, and clinically applicable iPS cells can be generated from patients with noninvasive medical procedures. Collectively, iPS cells have a greater potential to be used in ACT-based immunotherapy for autoimmune diseases compared with ES cells and HSCs. Our study significantly facilitates this application.

Although we could generate mature nTreg cells in vitro after long-term coculture of Foxp3-transduced iPS cells with OP9-DL1 cells, many differentiated Treg cells died after 1 mo coculture, suggesting that other survival signals may be required. It has been shown that IL-2 plays a crucial role in the homeostasis and normal function of Treg cells, and TGF-β promotes Treg cell development by providing survival signals and constraining negative selection through increasing Bcl-2 and decreasing Bim expression (31). Therefore, supplementation of IL-2 and TGF-β may aid the survival of iPS cell-derived Treg cells. Alternatively, we cocultured Foxp3-transduced iPS cells with OP9-DL1 cells for a week and then adoptively transferred the iPS cell-derived Treg progenitors into thymectomized Rag1−/− or recipient mice. This approach resulted in the generation of a large number of functional nTreg cells in vivo, suggesting that the appropriate survival signals required for maintenance of Treg cells are readily available in vivo.

The development of αβ T cells in the thymus is a well-ordered procedure. The most immature thymocyte population (CD4CD8) is referred to as double-negative (DN) cells. DN precursors are subdivided into sequential developmental subsets based on expression of CD44 and CD25 as follows: DN1 (CD44+CD25), DN2 (CD44+CD25+), DN3 (CD44CD25+), and DN4 (CD44CD25). Rearrangement of the TCRβ locus catalyzed by RAG1 and RAG2 is initiated as cells transit from the DN2 to the DN3 stage. Only DN3 cells that have generated a functional TCRβ-chain that can pair with the invariant pre–Tα- and the CD3 signaling chains to form a pre-TCR are selected for further differentiation. This event, called β-selection, represents the first checkpoint during T cell development. Pre-TCR formation signals proliferation, termination of TCRβ locus rearrangement, and differentiation of DN thymocytes to the CD4+CD8+ double-positive stage (32). Our in vitro stimulation with the Notch ligand drove Foxp3-transduced iPS cells to pass through the β-selection checkpoint in 1–2 wk and became pre-T cells (CD3+TCRVβ+, CD25CD44, CD4+CD8+). Additional 2 wk stimulation allowed pre-T cells to transit into CD4+ Treg cells (CD3+TCRVβ+, CD4+CD8, CD25+CD127CTLA-4+Foxp3+). Additionally, we also observed a small population of CD8+ T cells (CD3+TCRVβ+, CD4CD8+, CTLA-4+Foxp3+) that expressed suppressive cytokines (e.g., IL-10, TGF-β) and had suppressive function.

The generation of iPS cell-derived Treg cells without Foxp3 gene transduction is an attractive approach for ACT-based immunotherapy. It has been reported that retinoic acid enhances Foxp3+ Treg cell conversion by inhibiting the secretion of cytokines that interfere with conversion (33, 34). It also has been shown that TGF-β is critical to the thymic development of nTreg cells (35). Importantly, CD4+CD25+FOXP3+ Treg cells had been generated with IL-2, TGF-β, and retinoic acid (22, 36). We tested to improve our approach to generate Treg cells from iPS cells by culturing iPS cells with Notch ligands (i.e., DL1 and DL4) and TGF-β plus retinoic acid. Although we found that Foxp3+ Treg cells differentiated from this system, the number of Treg cells derived from iPS cells was much lower than the approach of Foxp3 gene transduction as described in this study.

Ag-specific Treg cells are the crucial mediators of peripheral tolerance, resulting in the suppression of autoimmune response, and it is known that effector/memory-like Ag-specific Treg cells are the optimal population for ACT-based immunotherapy. These Ag-specific Treg cells express high levels of E/P-selectin–binding ligands, multiple adhesion molecules (e.g., the integrin αEβ7), as well as receptors for inflammatory chemokines (e.g., CCR4, CCR6, and CXCR3) (4, 37), allowing efficient migration into inflamed sites. The ACT-based immunotherapy in the present study does not cure established arthritis, which may be explained by the absence of specificity that directs the movement of Treg cells to the inflamed paws. Thus, programming of Ag-specific Treg cells from iPS cells may promote immunosurveillance for established autoimmunity.

In summary, we provide evidence that Notch signaling and Foxp3 cooperatively direct iPS cells to differentiate into functional Treg cells. Adoptive transfer of iPS cell-derived Treg cells is able to overcome strain limitation and suppress autoimmunity. Thus, targeting iPS cell-derived Treg cells may be a therapeutic option in autoimmune diseases.

We thank Dr. S. Yamanaka (Institute for Frontier Medical Sciences, Kyoto University) for providing the iPS-MEF-Ng-20D-17 cell line, Dr. J. Zuniga-Pflucker (University of Toronto) for the supporting OP9-DL1 cell line, Dr. P. Wang (University of Southern California, Los Angeles, CA) for providing the construct of Mig–I-Ab, and Dr. A. Rudensky (Sloan-Kettering Institute, New York, NY) for providing the construct of Mig–Foxp3. We also thank Drs. Michael Croft (La Jolla Institute for Allergy and Immunology, La Jolla, CA), Shao-cong Sun (The University of Texas MD Anderson Cancer Center, Houston, TX), and Todd D. Schell (Penn State University) for critical reviews of the manuscript.

This work was supported in part by National Cancer Institute Grant K18CA151798 and by the Barsumian Trust (to J.S.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

ACT

adoptive cell transfer

CIA

collagen-induced arthritis

CII

type II collagen

DN

double-negative

ES

embryonic stem

HSC

hematopoietic stem cell

iPS

induced pluripotent stem

nTreg

naturally occurring T

Treg

regulatory T.

1
Bollyky
P. L.
,
Wu
R. P.
,
Falk
B. A.
,
Lord
J. D.
,
Long
S. A.
,
Preisinger
A.
,
Teng
B.
,
Holt
G. E.
,
Standifer
N. E.
,
Braun
K. R.
, et al
.
2011
.
ECM components guide IL-10 producing regulatory T-cell (TR1) induction from effector memory T-cell precursors.
Proc. Natl. Acad. Sci. USA
108
:
7938
7943
.
2
Fritzsching
B.
,
Oberle
N.
,
Eberhardt
N.
,
Quick
S.
,
Haas
J.
,
Wildemann
B.
,
Krammer
P. H.
,
Suri-Payer
E.
.
2005
.
In contrast to effector T cells, CD4+CD25+FoxP3+ regulatory T cells are highly susceptible to CD95 ligand- but not to TCR-mediated cell death.
J. Immunol.
175
:
32
36
.
3
Li
R.
,
Perez
N.
,
Karumuthil-Melethil
S.
,
Prabhakar
B. S.
,
Holterman
M. J.
,
Vasu
C.
.
2007
.
Enhanced engagement of CTLA-4 induces antigen-specific CD4+CD25+Foxp3+ and CD4+CD25 TGF-β1+ adaptive regulatory T cells.
J. Immunol.
179
:
5191
5203
.
4
Van
V. Q.
,
Darwiche
J.
,
Raymond
M.
,
Lesage
S.
,
Bouguermouh
S.
,
Rubio
M.
,
Sarfati
M.
.
2008
.
Cutting edge: CD47 controls the in vivo proliferation and homeostasis of peripheral CD4+ CD25+ Foxp3+ regulatory T cells that express CD103.
J. Immunol.
181
:
5204
5208
.
5
Wright
G. P.
,
Notley
C. A.
,
Xue
S. A.
,
Bendle
G. M.
,
Holler
A.
,
Schumacher
T. N.
,
Ehrenstein
M. R.
,
Stauss
H. J.
.
2009
.
Adoptive therapy with redirected primary regulatory T cells results in antigen-specific suppression of arthritis.
Proc. Natl. Acad. Sci. USA
106
:
19078
19083
.
6
Scalapino
K. J.
,
Daikh
D. I.
.
2009
.
Suppression of glomerulonephritis in NZB/NZW lupus prone mice by adoptive transfer of ex vivo expanded regulatory T cells.
PLoS ONE
4
:
e6031
.
7
Bluestone
J. A.
,
Tang
Q.
.
2004
.
Therapeutic vaccination using CD4+CD25+ antigen-specific regulatory T cells.
Proc. Natl. Acad. Sci. USA
101
(
Suppl. 2
):
14622
14626
.
8
Okita
K.
,
Ichisaka
T.
,
Yamanaka
S.
.
2007
.
Generation of germline-competent induced pluripotent stem cells.
Nature
448
:
313
317
.
9
Dai
B.
,
Wang
P.
.
2009
.
In vitro differentiation of adult bone marrow progenitors into antigen-specific CD4 helper T cells using engineered stromal cells expressing a notch ligand and a major histocompatibility complex class II protein.
Stem Cells Dev.
18
:
235
245
.
10
Lei
F.
,
Haque
R.
,
Weiler
L.
,
Vrana
K. E.
,
Song
J.
.
2009
.
T lineage differentiation from induced pluripotent stem cells.
Cell. Immunol.
260
:
1
5
.
11
Haque
R.
,
Lei
F.
,
Xiong
X.
,
Wu
Y.
,
Song
J.
.
2010
.
FoxP3 and Bcl-xL cooperatively promote regulatory T cell persistence and prevention of arthritis development.
Arthritis Res. Ther.
12
:
R66
.
12
Lei
F.
,
Zhao
B.
,
Haque
R.
,
Xiong
X.
,
Budgeon
L.
,
Christensen
N. D.
,
Wu
Y.
,
Song
J.
.
2011
.
In vivo programming of tumor antigen-specific T lymphocytes from pluripotent stem cells to promote cancer immunosurveillance.
Cancer Res.
71
:
4742
4747
.
13
Terszowski
G.
,
Müller
S. M.
,
Bleul
C. C.
,
Blum
C.
,
Schirmbeck
R.
,
Reimann
J.
,
Pasquier
L. D.
,
Amagai
T.
,
Boehm
T.
,
Rodewald
H. R.
.
2006
.
Evidence for a functional second thymus in mice.
Science
312
:
284
287
.
14
Whalen
B. J.
,
Weiser
P.
,
Marounek
J.
,
Rossini
A. A.
,
Mordes
J. P.
,
Greiner
D. L.
.
1999
.
Recapitulation of normal and abnormal BioBreeding rat T cell development in adult thymus organ culture.
J. Immunol.
162
:
4003
4012
.
15
Ramsdell
F.
,
Zuniga-Pflucker
J. C.
,
Takahama
Y.
.
2006
.
In vitro systems for the study of T cell development: fetal thymus organ culture and OP9-DL1 cell coculture.
Curr. Protoc. Immunol.
71
:
3.18.1
3.18.18
.
16
Song
J.
,
So
T.
,
Cheng
M.
,
Tang
X.
,
Croft
M.
.
2005
.
Sustained survivin expression from OX40 costimulatory signals drives T cell clonal expansion.
Immunity
22
:
621
631
.
17
Whalen
B. J.
,
Marounek
J.
,
Mordes
J. P.
,
Rossini
A. A.
,
Greiner
D. L.
.
2003
.
Type 1 cytokines polarize thymocytes during T cell development in adult thymus organ cultures.
J. Autoimmun.
20
:
27
42
.
18
Garbe
A. I.
,
Krueger
A.
,
Gounari
F.
,
Zúñiga-Pflücker
J. C.
,
von Boehmer
H.
.
2006
.
Differential synergy of Notch and T cell receptor signaling determines αβ versus γδ lineage fate.
J. Exp. Med.
203
:
1579
1590
.
19
Schmitt
T. M.
,
de Pooter
R. F.
,
Gronski
M. A.
,
Cho
S. K.
,
Ohashi
P. S.
,
Zúñiga-Pflücker
J. C.
.
2004
.
Induction of T cell development and establishment of T cell competence from embryonic stem cells differentiated in vitro.
Nat. Immunol.
5
:
410
417
.
20
Zhao
Y.
,
Parkhurst
M. R.
,
Zheng
Z.
,
Cohen
C. J.
,
Riley
J. P.
,
Gattinoni
L.
,
Restifo
N. P.
,
Rosenberg
S. A.
,
Morgan
R. A.
.
2007
.
Extrathymic generation of tumor-specific T cells from genetically engineered human hematopoietic stem cells via Notch signaling.
Cancer Res.
67
:
2425
2429
.
21
Van Coppernolle
S.
,
Verstichel
G.
,
Timmermans
F.
,
Velghe
I.
,
Vermijlen
D.
,
De Smedt
M.
,
Leclercq
G.
,
Plum
J.
,
Taghon
T.
,
Vandekerckhove
B.
,
Kerre
T.
.
2009
.
Functionally mature CD4 and CD8 TCRαβ cells are generated in OP9-DL1 cultures from human CD34+ hematopoietic cells.
J. Immunol.
183
:
4859
4870
.
22
Maynard
C. L.
,
Hatton
R. D.
,
Helms
W. S.
,
Oliver
J. R.
,
Stephensen
C. B.
,
Weaver
C. T.
.
2009
.
Contrasting roles for all-trans retinoic acid in TGF-β-mediated induction of Foxp3 and Il10 genes in developing regulatory T cells.
J. Exp. Med.
206
:
343
357
.
23
Collison
L. W.
,
Pillai
M. R.
,
Chaturvedi
V.
,
Vignali
D. A.
.
2009
.
Regulatory T cell suppression is potentiated by target T cells in a cell contact, IL-35- and IL-10-dependent manner.
J. Immunol.
182
:
6121
6128
.
24
Inglis
J. J.
,
Simelyte
E.
,
McCann
F. E.
,
Criado
G.
,
Williams
R. O.
.
2008
.
Protocol for the induction of arthritis in C57BL/6 mice.
Nat. Protoc.
3
:
612
618
.
25
Zhou
X.
,
Kong
N.
,
Wang
J.
,
Fan
H.
,
Zou
H.
,
Horwitz
D.
,
Brand
D.
,
Liu
Z.
,
Zheng
S. G.
.
2010
.
Cutting edge: all-trans retinoic acid sustains the stability and function of natural regulatory T cells in an inflammatory milieu.
J. Immunol.
185
:
2675
2679
.
26
Adeegbe
D.
,
Bayer
A. L.
,
Levy
R. B.
,
Malek
T. R.
.
2006
.
Cutting edge: allogeneic CD4+CD25+Foxp3+ T regulatory cells suppress autoimmunity while establishing transplantation tolerance.
J. Immunol.
176
:
7149
7153
.
27
Demirkiran
A.
,
Bosma
B. M.
,
Kok
A.
,
Baan
C. C.
,
Metselaar
H. J.
,
Ijzermans
J. N.
,
Tilanus
H. W.
,
Kwekkeboom
J.
,
van der Laan
L. J.
.
2007
.
Allosuppressive donor CD4+CD25+ regulatory T cells detach from the graft and circulate in recipients after liver transplantation.
J. Immunol.
178
:
6066
6072
.
28
Adeegbe
D.
,
Levy
R. B.
,
Malek
T. R.
.
2010
.
Allogeneic T regulatory cell-mediated transplantation tolerance in adoptive therapy depends on dominant peripheral suppression and central tolerance.
Blood
115
:
1932
1940
.
29
Peerani
R.
,
Zandstra
P. W.
.
2010
.
Enabling stem cell therapies through synthetic stem cell-niche engineering.
J. Clin. Invest.
120
:
60
70
.
30
Himburg
H. A.
,
Muramoto
G. G.
,
Daher
P.
,
Meadows
S. K.
,
Russell
J. L.
,
Doan
P.
,
Chi
J. T.
,
Salter
A. B.
,
Lento
W. E.
,
Reya
T.
, et al
.
2010
.
Pleiotrophin regulates the expansion and regeneration of hematopoietic stem cells.
Nat. Med.
16
:
475
482
.
31
Ouyang
W.
,
Beckett
O.
,
Ma
Q.
,
Li
M. O.
.
2010
.
Transforming growth factor-β signaling curbs thymic negative selection promoting regulatory T cell development.
Immunity
32
:
642
653
.
32
Tanigaki
K.
,
Honjo
T.
.
2007
.
Regulation of lymphocyte development by Notch signaling.
Nat. Immunol.
8
:
451
456
.
33
Mucida
D.
,
Park
Y.
,
Kim
G.
,
Turovskaya
O.
,
Scott
I.
,
Kronenberg
M.
,
Cheroutre
H.
.
2007
.
Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid.
Science
317
:
256
260
.
34
Nolting
J.
,
Daniel
C.
,
Reuter
S.
,
Stuelten
C.
,
Li
P.
,
Sucov
H.
,
Kim
B. G.
,
Letterio
J. J.
,
Kretschmer
K.
,
Kim
H. J.
,
von Boehmer
H.
.
2009
.
Retinoic acid can enhance conversion of naive into regulatory T cells independently of secreted cytokines.
J. Exp. Med.
206
:
2131
2139
.
35
Liu
Y.
,
Zhang
P.
,
Li
J.
,
Kulkarni
A. B.
,
Perruche
S.
,
Chen
W.
.
2008
.
A critical function for TGF-β signaling in the development of natural CD4+CD25+Foxp3+ regulatory T cells.
Nat. Immunol.
9
:
632
640
.
36
Lu
L.
,
Zhou
X.
,
Wang
J.
,
Zheng
S. G.
,
Horwitz
D. A.
.
2010
.
Characterization of protective human CD4CD25 FOXP3 regulatory T cells generated with IL-2, TGF-β and retinoic acid.
PLoS ONE
5
:
e15150
.
37
Siewert
C.
,
Lauer
U.
,
Cording
S.
,
Bopp
T.
,
Schmitt
E.
,
Hamann
A.
,
Huehn
J.
.
2008
.
Experience-driven development: effector/memory-like αE+Foxp3+ regulatory T cells originate from both naive T cells and naturally occurring naive-like regulatory T cells.
J. Immunol.
180
:
146
155
.

The authors have no financial conflicts of interest.