Department of Molecular Medicine and Surgery, Division of Cardiothoracic Surgery and Anesthesiology, Karolinska Institute, Karolinska University Hospital, Stockholm, Sweden
Department of Molecular Medicine and Surgery, Division of Cardiothoracic Surgery and Anaesthesiology, Karolinska University Hospital, S-171 76, Stockholm, Sweden. Telephone: +46-8-51770000; Fax: +46-8-51773437
In order to study the ability of costimulation blockade to induce tolerance to human embryonic stem cells (HESC), severe combined immunodeficient (SCID), and immunocompetent C57BL/6 mice treated with costimulation blockade received intratesticular and intramyocardial HESC transplants. All SCID mice with intratesticular HESC transplants developed teratoma. When SCID mice were transplanted intramyocardially, only two of five mice developed teratoma-like tumors. C57BL/6 mice transplanted intratesticularly and treated with costimulation blockade all developed teratoma and were surrounded by CD4+CD25+Foxp3+ T-cells, while isotype control treated recipients rejected their grafts. Most C57BL/6 mice transplanted intramyocardially and treated with costimulation blockade demonstrated lymphocytic infiltrates 1 month after transplantation, whereas one maintained its graft. Isolation of regulatory T-cells from intramyocardial transplanted recipients treated with costimulation blockade demonstrated specificity toward undifferentiated HESC and down-regulated naive T-cell activation toward HESC. These results demonstrate that costimulation blockade is sufficiently robust to induce tolerance to HESC in the immune-privileged environment of the testis. HESC specific regulatory T-cells developed to HESC transplanted to the heart and the success of transplantation was similar to that seen in SCID mice.
Disclosure of potential conflicts of interest is found at the end of this article.
Author contributions: K.-H.G.: conception and design, administrative support, provision of study material, collection and assembly of data, data analysis and interpretation, manuscript writing and final approval of manuscript; R.G.: conception and design, administrative support, collection and assembly of data and manuscript writing; M.K.-B. and C.D.: conception and design, collection and assembly of data, data analysis and interpretation, and manuscript writing; A.A. and A.M.-B.: collection and assembly of data, data analysis and interpretation; G.D. and C.S.: conception and design, administrative support, manuscript writing, final approval of manuscript and financial support; A.-M.S.: provision of study material; H.E.: financial support and final approval of manuscript; O.H.: administrative support, provision of study material, manuscript writing and final approval of manuscript; M.C.: conception and design, data analysis and interpretation, financial support, manuscript writing and final approval of manuscript.
Human embryonic stem cells (HESC) are pluripotent stem cells with an inborn capacity for self-renewal and differentiation [1, 2]. These characteristics make HESC an attractive source of cells for cardiac repair, but obstacles still remain to be overcome before these cells can be used for cardiomyoplasty in a clinical setting. The conversion of undifferentiated HESC into cardiomyocytes is an inefficient process, where 15–25% of HESC differentiate into cardiomyocytes in vitro, depending on which protocol is used [3, 4]. There is a severe risk of teratoma development as demonstrated by studies conducted with mouse ESC injected into syngeneic mice . Mouse and human ESC share several genomic and proteonomic features . However if the results described by Kolossov and co-workers working with mouse ESC can be extrapolated to HESC is still unresolved.
Apart from these obstacles, another major challenge will be to overcome immune rejection of the implanted cells. This can be achieved by manipulating both the HESC and the recipient. Originally immuno-surgery was used in the derivation process, fetal calf serum (FCS) was a constituent of the culture medium and HESC were cultured on mouse fibroblast feeder cells [1, 2]. HESC were exposed to animal constituents in every step, which constitutes both a risk of transmitting infections and the acquisition of xenogeneic antigens which might augment immune rejection . To minimize the involvement of animal products in the derivation and culture process, multiple steps have been implemented. Mechanical isolation of the inner cell mass , culturing on post-natal human skin fibroblast feeder cells , and utilizing a culturing medium with serum replacement instead of FCS [10, 11] will reduce the exposure of HESC to animal products.
The immunogenicity of HESC is relatively uncharted territory. In recent articles published by Li et al.  and Drukker et al. , in vitro and in vivo data was presented demonstrating that HESC possess immune-privileged properties and thus can be transplanted over the xenogeneic  and allogeneic barriers . Their conclusions contrast to our findings when working with a similar cell line where HESC induce a robust immune response in both allogeneic and xenogeneic settings . According to our data, the use of HESC-derived cells for cardiomyoplasty would require chronic immunosuppression, which substantially increases the likelihood of developing opportunistic infections , cancer , and morbidity secondary to chronic drug toxicity.
Two possible approaches to overcome the immunological barriers without using immunosuppression are to employ either somatic cell nuclear transfer (SCNT) or induced human pluripotent stem cells. In the process of SCNT, the nucleus of the oocyte is removed and replaced by the genetic material from a somatic cell of the graft recipient. The oocyte cytoplasm enables the donor genome from a somatic cell to support embryonic development by a process called reprogramming . The embryo is cultured to the blastocyst stage from which the embryonic stem cell lines are derived. The HESC lines produced express the recipients transplantation antigens, except for the antigens derived from the mitochondria of the embryo donor. Successful production of embryonic stem cell lines through SCNT from both mice [18, 19] and cattle  have been reported. In humans, early blastocysts have been produced through SCNT , but the embryo failed to establish an HESC line. The alternative strategy is to transfect the human somatic cell with up to four different transcription factors and thereby reprogramming the somatic cells into induced pluripotent stem cells (iPS) [22, , –25]. Although human iPS cells are similar to HESC they are not identical. There is a difference in the gene expression between iPS cells and HESC . Furthermore, it is still too early to speculate whether iPS cells have the same potential as HESC to generate the different cell types in the human organism. All of the groups that created iPS used viral vectors to deliver the transcription factors into the human somatic cells. Analysis of the reprogrammed iPS cells showed that these cells have on average 20 or more retroviral integration sites [24, 25]. This is not acceptable from a clinical point of view because of the high-risk of retroviral insertional mutagenesis. It is, therefore, too early to say if iPS or SCNT can replace the allogeneic HESC and their progeny in regenerative medicine.
Another attractive option to circumvent immune rejection and chronic immunosuppression is to induce immunological tolerance toward the transplanted HESC by using costimulation blockade.
CTLA4Ig is a fusion protein consisting of the extra cellular domain of CTLA4 and a human IgG1. CTLA4Ig binds to the B7 family of costimulatory molecules expressed on dendritic cells during activation. By binding these molecules, reacting T-cells do not receive the necessary secondary signals to initiate T-cell activation, which leads to anergy and apoptosis of reacting T-cells [26, –28]. Anti-CD40L binds to CD40L, a molecule expressed on T-cells which is a central regulatory signal in the immune system when initiating an immune response. Inhibition of this signal leads to prevention of T-cell memory development and effector cell activation, B-cell differentiation and isotype switching, as well as macrophage activation . Anti-LFA-1 blocks the adhesion molecule LFA-1, which is ubiquitously expressed on nearly every cell in the immune system. It has a pivotal role in leukocyte function and the formation of the immunological synapses between T-cells and dendritic cells and T-cells and their targets . By administering these three substances at the time of transplantation, tolerance can be induced to allogeneic and xenogeneic tissues and organs . This state of tolerance seems to be mediated partially by regulatory T-cells that inhibit naive T-cells directed toward the grafted cells.
In an attempt to further clarify these issues, the present study was designed to test the tumorigenicity and immunogenicity of HESC transplanted into both the myocardium and testis of immunocompetent and immunodeficient mice. Furthermore, we hypothesized that the successful strategy to induce tolerance toward grafts transplanted over the xenogeneic barrier using costimulation blockade might be applicable to HESC. Therefore we tested whether or not a short course treatment with anti-CD40L/anti-LFA-1 and CTLA4Ig could induce long-term acceptance of HESC implanted into mice and whether or not there was a difference in engraftment between testis and the myocardium.
Materials and Methods
Derivation and Culture of Human ESC
HESC line HS401 was derived and cultured at the Fertility Unit, Department of Gynecology, Karolinska University Hospital [9, 11]. The HESC line was derived from supernumerary blastocysts donated by couples undergoing fertility treatment. The HESC were cultured on post-natal human skin fibroblast feeder cells in serum replacement medium . The colonies were split mechanically without the use of enzymes and replated onto new human fibroblast feeder cells every 5–7 days. The HS401 expressed markers typical of undifferentiated HESC, namely alkaline phosphatase, stage-specific embryonic antigen (SSEA-4), tumor-related antigen (TRA 1–60, TRA 1–81), and octamer-binding transcription factor-4 (OCT-4), and formed teratoma when injected into severe combined immunodeficient (SCID) mice. For the implantation experiments, the HS401 colonies were divided manually with metal needles on a stereo-microscope (Nikon ×1500 [Nikon, Inc., Melville, NY, http://www.nikonusa.com]) and loaded into syringes (200.000 HESC in 15 μl culture medium).
Animals and Reagents
Eight-week-old male C57BL/6 mice (B and K Universal AB, Sollentuna, Sweden, http://www.bku.com) along with eight week old male SCID/Beige (C.B.-17/GbmsTac-SCID-bgDFN7; M and B, Taconic, Denmark, http://www.taconic.com) remains were used. The Animal Care Committee of Karolinska University Hospital approved all procedures.
The active costimulation blocking reagents used were anti-LFA-1 (clone M17/5.2), anti-CD40L (clone MR1), CTLA4 Ig, and their respective isotype control antibodies: rat IgG2b (clone 9A2), hamster IgG1 and human IgG1, (Bioexpress, West Lebanon, NH, http://www.bioexpress.com). The C57BL/6 mice treated with active costimulation blockade (costimulation blockade treated group), received 0.5 mg anti-CD40L, 0.2 mg anti-LFA-1, and 0.5 mg CTLA4-Ig given intraperitoneally (ip) every other day for 8 days. The control mice received corresponding amount of the isotype control reagents (isotype control treated group).
Anesthesia and Post-Operative Care
The mice were anesthetized with a mixture of Midazolam (5 mg/kg; F. Hoffmann-La Roche Ltd., Basel, Switzerland, http://www.roche.com), Medetomidine (0.1 mg/kg; Orion Corporation, Espoo, Finland, http://www.orion.fi), Fentanyl (0.3 mg/kg; B. Braun Medical AG, Seesatz, Switzerland, http://www.bbraun.ch) and given ip. In the group where HESC were injected into the myocardium, the mice were tracheotomized, and the ventilation was maintained using a Zoovent ventilator, model CWC600AP, (B and K Universal). The ventilation was not supported when the cells were injected into the testis of the mice. The anesthesia was reversed by an ip injection of Flumazenil (0.1 mg/kg; F. Hoffmann-La Roche Ltd) and Atipamezol hydrochloride (5 mg/kg; Orion Corp). For post-operative analgesia, Buprenorphin hydrochloride (Schering-Plough Corp., Kenilworth, U.K., http://www.sgp.com) was given ip used.
Cell Transplantation and Induction of Immunological Tolerance
Through an incision in the distal part of the abdominal wall, through the linea alba, the right testis was exposed and 200.000 HESC were injected under the testis capsule of SCID mice (n = 6) and C57BL/6 mice, which were randomized to receive either costimulation blockade (n = 6) or isotype control reagents (n = 6). The mice were euthanized after 8 weeks, and the testis was harvested and analyzed histochemically.
The testis is described to be an immune-privileged organ . Although the HESC survive in a testis from an immunocompetent mouse treated with costimulation blockade, the result might be different if the HESC are transplanted to the heart. To further study this hypothesis, we injected another set of mice with the same amount of HESC through a left thoracotomy , into the myocardium of SCID mice (n = 5) and C57BL/6 mice, which were again randomized to costimulation blockade (n = 8) or isotype control treatment (n = 8). Two mice from both the costimulation blockade treated and the isotype control treated groups were euthanized after 4 weeks, whereas the remaining mice were euthanized after 8 weeks (i.e., SCID n = 5, C57BL/6 costimulation blockade treated n = 6; C57BL/6 isotype control treated n = 6).
Because of failure to induce teratoma and the appearance of cellular infiltrates at 1 month, a second set of transplants was performed. Costimulation blockade or their isotype controls, both at the time of transplantation and 3 weeks later (n = 5 in each group), were injected. We named these groups costimulation blockade x2 and isotype control x2, respectively. The mice were euthanized after 8 weeks.
Detection of Transplanted Cells and Histological Evaluation of the Immune Response
The hearts and testes from the euthanized mice were freeze-sectioned into 5 μm thick sections, and the site of injection was identified by hematoxylin and eosin staining. In case of teratoma formation, these sections were further evaluated searching for structures representing the three germ layers. The HESC and their differentiated progeny were traced using fluorescence in situ hybridization technique (FISH), which has previously been described . In brief, fluorescence inmarked DNA-probes hybridize to total human DNA of viable cells, labeling the whole nucleus red. The sections with surviving HESC derived cells in the heart were further stained for the cardiomyocyte markers α-actinin (clone EA-53; Sigma-Aldrich Corp, St. Louis, http://www.sigmaaldrich.com) and Desmin (clone DE-R-11; Dako Cytomation, Glostrup, Denmark, http://www.dakocytomation.com), together with TRA 1–60 (clone TRA-1–60, Chemicon International Inc., Temecula, CA, http://www.chemicon.com) which is a marker for undifferentiated HESC. After fixation with 4% formaldehyde and blocking with 5% rabbit serum (X0902; Dako Cytomation) for 30 minutes, the slides were incubated with the respective primary antibody over night in a humidified chamber. The sections were then incubated with fluorescence-labeled rabbit anti-mouse immunoglobulins/ FITC (Dako Cytomation) and visualized in the fluorescence microscope (Olympus BX60; Olympus Optical Ltd, Tokyo, http://www.olympus-global.com). As positive controls we used human heart (Desmin and α-actinin) and HESC in the myocardium of SCID/beige mice, which were euthanized and freeze-sectioned directly after injection (TRA 1–60).
The inflammatory response was evaluated by hematoxylin and eosin staining together with immunohistochemical analysis. After fixation with 4% formaldehyde (Histolab Products Ltd, Gothenburg, Sweden, http://www.histolab.se) and blocking with 5% rabbit serum (code No. X0902; Dako Cytomation) and 5% mouse serum (code no. X0910; Dako Cytomation) in TBS (stock solution 10x concentration: 87.66 g NaCl, 60.55 g Tris diluted to 1,000 ml in distilled water, pH: 7.4). The sections were subsequently incubated over night in a humidified chamber with the primary antibodies. These included rat anti-mouse CD3 (clone KT3), rat anti-mouse CD4 (clone YTS191.1), rat anti-mouse CD8 (clone KT15), rat anti-mouse CD11b (clone 5C6) (Serotec, Oslo, Norway, http://www.ab-direct.com) and rat anti-mouse Foxp3 (clone FJK-16s; eBioscience, San Diego, http://www.ebioscience.com). The slides were then incubated with the secondary antibodies rabbit anti-rat-IgG (FITC code No. F0234; Dako Cytomation) for 2 hours in a humidified chamber and mounted with an anti-fading reagent containing 4,6-diamidino-2-phenylindole (DAPI) before visualization in the fluorescence microscope (Olympus BX60). Spleens from C57BL/6 were used as positive controls, whereas hearts from SCID/beige mice served as negative controls for the primary antibodies.
Mixed Leukocyte Reaction
Splenocytes from naive C57BL/6 mice, together with C57BL/6 mice from the costimulation blockade treated and the isotype control groups, were used for this experiment. The spleens were homogenized and the CD4+ T-cells were separated using a CD4+ T-cell Isolation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany, http://www.miltenyibiotec.com) as previously described . The purity of the CD4+ T-cells was more than 95%, as estimated by fluorescence-activated cell sorter analysis (FACS). CD4+CD25+ T-cells from C57BL/6 mice of the costimulation blockade and isotype control groups were separated according to the manufacturer's protocol using a CD25 MicroBead Kit (Miltenyi Biotec) and the purity was estimated to 70% by FACS analysis.
The dendritic cells (DC) were prepared from the bone marrow of the femurs of syngenic C57BL/6 mice according to the protocol previously described .
The DC (1 × 106) from C57BL/6 mice were co-cultured with irradiated (15Gy) HESC (3 × 105) and human fibroblasts (HFib) (3 × 105), respectively, in culture medium [Roswell Park Memorial Institute (RPMI)-1640 supplemented with 5% heat inactivated fetal calf serum, 2 mmol/l l-glutamine, 100 μg/ml streptomycin, 100 IU/ml penicillin and 5 × 10−5 mol/l 2-mercaptoethanol] in humidified air containing 5% CO2 at 37°C for 24 h. The DC were then irradiated (15Gy). The mixed leukocyte reactions (MLR) was performed exposing naive CD4+ T-cells (5 × 104 cells per well) to syngenic DC (2 × 104 cells per well) alone or syngenic DC co-cultured with HESC or HFib (2 × 104 cells per well). To study if the regulatory T-cells from mice treated with costimulation blockade or its isotype control reagents could specifically down-modulate the immune response induced by HESC, the previously separated CD4+CD25+ T-cells (5 × 104 cells per well) were added to the wells. Five wells per culture were set up in 96-well round-bottomed microtiter plates (Corning Life Sciences, Acton, MA, http://www.corning.com/lifesciences) and incubated in humidified air containing 5% CO2 at 37°C. Each culture was labeled with 2 μCi [3H] thymidine (Amersham, Buckinghamshire, U.K., http://www.amersham.com) for 20 hours prior to harvest. The proliferation of the responders was measured at day 6 as counts per minute (cpm) and each experiment was repeated three times. Two mice from each group were used for each MLR.
Data are presented as mean ± SD. For statistical analysis of MLR data, Wilcoxon Signed Rank Test was used as a nonparametric test for two related samples. A p < .05 was considered significant.
HESC Transplanted into the Testis
The HESC injected into the testis of SCID mice (n = 6) and costimulation blockade treated C57BL/6 mice (n = 6) developed into teratoma in all animals when studied histologically at two months after transplantation. This was in contrast to the findings in the isotype control treated C57BL/6 mice (n = 6), where no surviving HESC was found. The teratoma did not differ between the SCID mice and the costimulation blockade treated C57BL/6 mice. They formed exophytic cystic tumors, where histological analysis revealed neuroepithelium, bronchoepithelium, and bone, hence typical histological features represent the three germ layers (Fig. 1A–C). The cells in the teratoma were of human origin as demonstrated by FISH staining (Fig. 1D), and the normal testis tissue was compressed toward the capsule. There was no sign of any acute CD4+ T-cell-mediated inflammatory response in the teratoma of the costimulation blockade treated C57BL/6 mice, and the only CD4+ T-cells present were mouse CD4+CD25+Foxp3+ regulatory T-cells (Fig. 2A). These cells were clustered at the capsule of the testis close to the interface of normal tissue and teratoma from where they seemed to migrate into the peripheral parts of the teratoma. No CD4+CD25+Foxp3+ regulatory T-cells were seen in the central parts of the teratoma. In the isotype control treated C57BL/6 mice no FISH positive cells could be identified after 2 months. There were no signs of an ongoing inflammatory response and only a small scar could be seen at the site of the injection.
HESC Transplanted into the Heart
In SCID mice, HESC engrafted in the heart in two of five mice formed expansive cystic tumors. These were mainly composed of different epithelial structures and fibrous stroma (Fig. 1E, F). Although their growth characteristics were similar to teratoma, they do not fulfill the criteria of a teratoma because cells from all three germ layers were not found. We have thus termed these tumors teratoma-like. The cells in the teratoma-like tumors did not stain positive for Desmin or α-actinin, indicating that the HESC were not cardiomyocytes and thereby were not influenced in their differentiation by the surrounding myocardium. In the other three SCID hearts the injection site was clearly present but there was no surviving HESC. This is in contrast to the testis where the HESC developed into teratoma in all animals tested (n = 6). In the costimulation blockade treated and isotype control treated C57BL/6 mice we could not find any surviving HESC after 1 (n = 2 in each group) and 2 months (n = 6 in each group), but the histological features differed between the two groups. In the isotype control treated group there was only a scar left in the myocardium, with no sign of an inflammatory response. In the costimulation blockade treated group, the hearts showed signs of lymphocytic infiltrates at 1 month after transplantation (Fig. 2B–F), which was also seen in one of the hearts after 2 months (Fig. 2G, H). The site of implantation of HESC was infiltrated by activated macrophages (CD11b+) and the T-cell response was characterized by both CD4+ and CD8+ cells. These histological finding were similar to the acute immune response seen during the first days after transplantation of HESC to untreated immunocompetent mice hearts . With costimulation blockade this immune response was delayed. In the other five hearts removed at 2 months a scar with no FISH positive cells was found at the site of injection (Fig. 2I).
In an attempt to induce tolerance toward these HESC-derived progeny, we injected another group of immunocompetent mice with a second round of treatment with costimulation blockade 3 weeks after transplantation (costimulation blockade x2 treated, n = 5) or with isotype controls (isotype control x2 treated, n = 5). As expected, no surviving HESC was found in the isotype control x2 treated group after two months, and the only histological finding was a scar at the site of injection. In one of the hearts (n = 1/5) in the costimulation blockade x2 treated group we found surviving HESC-derived cells after 2 months (Fig. 3). These cells were encapsulated in the myocardium and they had not developed into any teratoma-like structure. Furthermore, the HESC-derived cells stained negative for TRA 1–60 and the cardiomyocyte markers Desmin and α-actinin, indicating that they were no longer undifferentiated HESC or myocytes. In the tissue surrounding the engrafted HESC-derived cells there was no sign of inflammatory response nor could we identify a significant amount of regulatory T-cells in the tissue surrounding the engrafted HESC-derived cells.
The Functional Impact of Regulatory T-Cells in Inducing Tolerance Toward HESC
MLR were designed to mimic the clinical setting of HESC transplantation where the host DC present antigens from processed HESC to host CD4+ T-cells. Naive CD4+ T-cells from C57BL/6 mice were cultured together with syngenic DC, previously co-cultured with either HESC or with HFib. CD4+CD25+ T-cells were separated from the spleens of syngenic C57BL/6 mice previously transplanted with HESC into the heart and sacrificed after 2 months. The influence that the CD4+CD25+ T-cells separated from the costimulation blockade or isotype control groups had on naive CD4+ T-cell proliferation was measured. As shown in Figure 4, the CD4+CD25+ T-cells separated from the costimulation blockade group significantly downregulated naive CD4+ T-cell proliferation with a mean of 22% (p < .05), whereas CD4+CD25+ T-cells from the isotype control group did not inhibit proliferation of naive T-cells. Furthermore, the downregulatory effect of the immune response mediated by CD4+CD25+ T-cells was specific toward undifferentiated HESC, because they did not inhibit the proliferation of naive T-cells toward HFib.
The results presented herein clearly demonstrate that HESC are immunogenic and are rejected when transplanted over the xenogeneic barrier into the testis or the myocardium of immunocompetent C57BL/6 mice. The environment of the testis is considered an immune-privileged area , nonetheless, HESC were rejected when transplanted to the testis. The HESC line used has to a minimal degree been exposed to animal components during the derivation and culturing process. Altogether this means that the immune response seen in these experiments was induced by the HESC and not by contaminating animal products. These findings further emphasize that HESC are not immune-privileged and confirms our previous findings .
In order to circumvent immune rejection without using chronic immunosuppression, we attempted to induce a state of immunological tolerance toward the engrafted cells. Ever since Medawar′s pioneering experiments , the “Holy Grail” of the transplant community has been to induce tolerance to transplanted tissues and organs. The use of costimulation blockade has been heralded as a great leap toward achieving this goal [31, 35, , , –39]. The strategy that has attracted the greatest attention targets the costimulatory molecules, CD40L, and B7 molecules, expressed on T-cells and DC under inflammatory circumstances. By blocking these pathways with anti-CD40L and CTLA4Ig (a fusion protein of a high affinity receptor for B7 and the Fc component of human IgG1) given during the first week after transplantation, indefinite acceptance of vascularized mouse and rat cardiac transplants in mice can be achieved [40, 41]. Double costimulation blockade (anti-CD40L/CTLA4Ig) has a varied outcome depending on the type of tissue transplanted and the strain of the recipient mouse . By simultaneously blocking LFA-1, CD40L-independent immune responses can be inhibited, which has been shown to be advantageous in the transplantation of discordant xenogeneic cellular transplants . This triple costimulation blockade (anti-CD40L/CTLA4Ig/anti-LFA-1) given during the first 8 days after transplantation induced long-term acceptance and improved pig islet graft function in diabetic mice when compared with double costimulation blockade (anti-CD40L/CTLA4Ig) . Using the same triple costimulation blockade regime, HESC engrafted in the testes of all immunocompetent mice forming exophytic growing teratoma at 2 months after transplantation, with similar growth characteristics as in the SCID mice. There was no sign of CD4+ T-cell-mediated immune rejection. The development of CD4+CD25+ Foxp3+ T-cells indicate that costimulation blockade had induced immunological tolerance toward the HESC-derived cells. These results are in contrast to those obtained when HESC were injected into the myocardium of C57BL/6 mice treated with the same short course of costimulation blockade. The HESC induced an acute CD4+ T-cell-mediated immune response one month after injection, causing rejection of the implanted HESC. Nonetheless, CD4+CD25+ T-cells specific to undifferentiated HESC with the capacity to downregulate naive T-cell activation developed in these recipients. This could imply that the host was tolerant of undifferentiated HESC but, as the cells differentiate, their gene expression changes , and thereby new antigens, which the host has not been tolerated toward, are expressed. These new antigens could thereby activate other clones of T-cells which were not present at the time or place of tolerization and, subsequently, induce rejection.
In order to study the possibility to induce tolerance toward the in vivo differentiating HESC, we repeated the costimulation blockade treatment three weeks after transplantation. In one of five mice, the HESC-derived cells were found in the myocardium after 2 months, but did not develop into teratoma, since cells representative of all three germ layers were not present. Instead, the engrafted cells were encapsulated in the myocardium, and there was no sign of inflammatory response.
The difference in survival of HESC-derived cells between the testis and the heart of the costimulation blockade treated mice may also be due to nonimmunological factors. Injection of HESC into the hearts of SCID mice only induced tumor growth in two of five mice, while teratoma formations were seen in the testes of all SCID mice. The difference in tumor development between the heart and testis thus seems to be due to factors pertaining to engraftment in the myocardium, similar to the findings when nonpurified mESC-derived cardiomyocytes were injected into the heart .
CD4+CD25+Foxp3+ T-cells were seen clustered at the interface between normal tissue and teratoma in the testes of the costimulation blockade treated mice. These findings support the hypothesis that costimulation blockade induces peripheral tolerance mediated by regulatory T-cells. In contrast, no regulatory T-cells were seen around the engrafted HESC-derived cells in the myocardium, which is in accordance to the findings when pig islets were transplanted into mice using the same costimulation blockade protocol . To further evaluate the role of regulatory T-cells in maintaining peripheral tolerance to implanted HESC, we performed MLRs. In the experiments described in this report, 95% pure populations of naive CD4+ T-cells were isolated and exposed to syngenic DC co-cultured with HESC or HFib, in order to imitate the clinical situation as closely as possible. To study if the regulatory T-cells from mice treated with costimulation blockade or isotype control reagents could specifically down-modulate the immune response induced by HESC, regulatory T-cells from the different groups were added to the ongoing indirect immune response. Regulatory T-cells isolated from the costimulation blockade group down-modulated the immune response induced by HESC significantly by 22%, whereas regulatory T-cells from the isotype control group did not downregulate the immune response at all. The effects mediated by the regulatory-T-cells also seemed to be specific toward HESC, since the immune response induced by HFib was not affected. These data indicate that, even though the grafts were not maintained in the heart, the recipient nonetheless was tolerized to undifferentiated HESC.
The tumorigenicity of mESC in the heart of syngeneic mice has been well-characterized , but the tendency of HESC to form teratoma in different compartments is still a relatively unexplored area of research. To study the tumorigenicity of HESC, we injected undifferentiated HESC into both the myocardium and the testes of SCID mice. In the testis, teratoma formations were seen in every mouse, while in the myocardium teratoma-like formations were seen in two of five mice, 2 months after transplantation. This indicates that HESC are as tumorigenic as mESC and from this standpoint, it seems logical that pure HESC-derived cardiomyocyte populations should be used for cardiomyoplasty.
An interesting observation is that the engrafted HESC-derived cells in the myocardium of costimulation blockade treated mice did not develop into teratoma. This may be due to the fact that repeated treatment with costimulation blockade has induced tolerance to one type of HESC-derived cell, whereas other cells with another gene expression are rejected. From this point of view, costimulation blockade might reduce the tumorigenicity of HESC by empowering the immune system with the capacity to remove cells that are not of the same phenotype as the cells to which tolerance toward is desired.
Altogether, these findings indicate that it is possible to induce immunological tolerance toward undifferentiated HESC transplanted into the testis and to induce regulatory T-cells to undifferentiated HESC when transplanted into the heart. The sertoli cells of the testis express Fas ligand (FasL, CD95) which induces apoptosis of reactive T-cells toward intratesticular grafts . This capacity, when combined with costimulation blockade, tips the balance from rejection to the induction of regulatory T-cells and tolerance induction . In the heart, HESC differentiate, changing their antigen expression and thereby they are no longer protected from rejection by the regulatory T-cells induced at the time of tolerization. By using highly purified HESC-derived cardiomyocytes already differentiated into the desired phenotype, it might be possible to induce peripheral tolerance with costimulation blockade and reduce the risk of teratoma formation. Until these goals have been reached, clinical trials with HESC-derived cells should not be performed, because they are both immunogenic and tumorigenic.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.